Surg Endosc (1997) 11: 625631

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic partial fundoplication vs laparoscopic Nissen-Rosetti fundoplication

Short-term results of 231 cases
D. D. Coster, W. H. Bower, V. T. Wilson, R. T. Brebrick, G. L. Richardson
Grinnell Regional Medical Center, 200 Fourth Avenue, Grinnell, IA 50112, USA, and The Grinnell Institute for Minimally Invasive Surgery, 122 Fourth Avenue, Grinnell, IA 50112, USA Received: 12 December 1995/Accepted: 12 August 1996

Abstract Background: Since 1992, all patients at our institution who have met standard accepted criteria for surgical intervention for complicated gastroesophageal reflux disease have been entered into a prospective sequential clinical study to evaluate outcomes of the laparoscopic approach to the NissenRosetti procedure and a modified Toupet procedure. Methods: A standardized workup with upper GI series, esophagography, and endoscopy was used in all patients. Manometry, pH testing, and other special tests were used selectively. A measuring technique was used to determine wrap size without the use of dilators. The short gastric vessels were left intact in all patients. A cosurgeon approach was used, with technical factors described herein. Results: Some 226 of 231 cases were completed laparoscopically (98%)125 patients in the Nissen-Rosetti group and 101 in the partial fundoplication group. There were no clinical failures in either group. The partial fundoplication group performed better than the Nissen-Rosetti group in all categories of comparison. Return to normal eating habits was much earlier in the partial wrap group (p < 0.0001). Postop distal esophageal sphincter pressures in the two groups were equal at 15 mmHg. Eight patients suffered significant dysphagia requiring endoscopy and dilatation, all in the Nissen-Rosetti group (p < 0.01). Minor complications occurred in 12% of the total group. There was a total surgical revision rate of 3%. There were no gastric or esophageal perforations. Average operative time was 30 min. Average hospital stay was 1.4 days. Hospital charges for the laparoscopic approach averaged $6,000 dollars compared to $12,000 for the open approach. Conclusion: Laparoscopic partial fundoplication is as effective as laparoscopic Nissen-Rosetti fundoplication, with a higher satisfaction rate and fewer side effects. Measuring

for wrap and hiatus size eliminates the need for and risk of using stiff dilators. By utilizing cosurgeons and currently available technology, cost, operative time, hospital time, and complications can be reduced to a finite minimum. Key words: Partial fundoplication Nissen-Rosetti fundoplication Toupet procedure

Over the past 4 years laparoscopic techniques have been newly applied to the treatment of complicated gastroesophageal reflux disease (GERD) in the United States. A variety of procedures including the Nissen fundoplication, the Rosetti modification of the Nissen fundoplication, the Toupet partial fundoplication, the Hill procedure, and the Belsey Mark 4 fundoplication have all been successfully performed via videoscopic technology with good results [12, 6, 7, 9, 20, 21]. The vast majority of the procedures have been performed for type I sliding hiatal hernia with one or more of the following surgical indications: chronic symptomatic reflux with objective evidence of esophagitis refractory to maximal medical management, grade 4 esophagitis with stricture, Barretts metaplasia, esophageal ulceration with hemorrhage, and chronic aspiration with resultant pneumonia, asthma, or acute airway obstruction. We have previously reported on the outcomes and cost analysis of our first 52 Nissen-Rosetti fundoplications done or attempted laparoscopically [2]. Emphasis on technical factors that enhance the ease of the procedure was made and the safety and cost-effectiveness of the procedure was confirmed. Following is an update and results analysis of 125 laparoscopic Nissen-Rosetti fundoplications as well as the description and results analysis of a modified laparoscopic Toupet partial fundoplication technique used in 101 cases. Materials and methods

Correspondence to: D. D. Coster, Surgical Associates, 122 Fourth Avenue, Grinnell, IA 50112, USA

In October 1992 we embarked on a prospective sequential clinical study to evaluate the outcome and effectiveness of the laparoscopic approach to

626

Figs. 1 and 2. Sharp dissection of the peritoneum just above the fundus from the left crus all the way over to the first short gastric vessel, followed by excessive blunt dissection behind the fundus. Fig. 3. The size of the hiatal opening is then measured using the end of the USSC roticulating bowel grasper. Fig. 4. A sling maneuver is then performed, passing the fundus back and forth to be certain it has no twists. Figs. 5 and 6. The first stitch incorporates the esophagus at the right gastroesophageal junction, the fundus, and the pre-aortic fascia and muscle below the junction of the crural leaves with the first stitch. The esophagus, fundus, and right crus at the level of the crural reapproximation are incorporated with the second, thus fixing in place at least a 2-cm intraabdominal length of esophagus. Fig. 7. A third suture incorporates the right anterior esophagus and the midportion of the wrap, bringing the right side of the partial fundoplication into its final position.

antireflux surgery. All patients who have been considered to be surgical candidates based on the standard accepted criteria for surgical intervention for complicated gastroesophageal reflux disease and who have been medically competent to withstand surgery have been entered into the study. All of the fundoplication procedures have been done by the same five surgeons with a standardized cosurgeon approach. The medical/diagnostic workup of patients entering the study consisted of a history and physical examination, chest X-ray, EKG, complete laboratory profile, upper GI series, fluoroesophagography, upper abdominal ultrasound, and esophagogastroduodenoscopy with or without biopsy. Esophageal manometry and 24-h pH testing were reserved for cases with an atypical presentation, lack of esophagitis on endoscopy, or symptoms or findings suggestive of esophageal motility disturbances based on history or on any one of the evaluative tests. During the 3 years of the study, the referral pattern has changed so that gastroenterologists, internists, and surgeons have referred patients as well as family physicians. Those patients with thorough evaluations done elsewhere were not retested unless symptoms had changed significantly. The

majority of diagnostic upper endoscopies, esophageal dilatations, and other endoscopic therapeutic procedures were done by the surgeons involved in the study. The technical aspects of our approach to laparoscopic antireflux surgery are as follows. A cosurgeon approach is used in all cases, reducing operative time and improving safety. Each experienced surgeon operates from his side of the table or assists the opposite surgeon, depending on who can best do each part of the procedure from their position at the table. A six-trocar technique allows for placement of a right lateral port for a fixedposition liver retractor, a right upper quadrant port for the cosurgeons use, an upper midline port and a left upper quadrant port for the surgeons use, and a left lateral port for a Babcock retractor for manipulation of the stomach and gastroesophageal junction. A 45 lens is a necessity so that a downward view of the operative field can be obtained through the supraumbilical port. A nasogastric tube is placed for gastric decompression, though occasionally placement must wait until the hiatal hernia has been reduced in order to get the tube to traverse the gastroesophageal junction. Complete dissection of the hiatus with nothing in the esophagus is perfectly

627

acceptable. Starting the dissection by opening the lesser sac near the caudate lobe of the liver immediately exposes the right crus as a landmark, facilitating the rest of the dissection. The space between the right crus and esophagus is then opened, followed by the complete dissection of the right crus posteriorly until it is seen joining the left crus, actually dissecting the majority of the left crus and creating as much of the posterior esophageal window as possible from the right side of the esophagus. Once the anterior esophageal peritoneal covering is opened, and the fundus and angle of His is taken down, the window behind the esophagus is already completely open and little further dissection is required. The most important technical point of the entire operation is to completely take down the fundus off of the left hemidiaphragm, as the fundus is what makes up the wrap, not the body. This requires sharp dissection of the peritoneum just above the fundus from the left crus all the way over to the first short gastric vessel, followed by extensive blunt dissection behind the fundus until it is laying there floppy (Figs. 1 and 2). The window behind the esophagus will then be huge, and any remaining attachments can easily be removed using a blunt dissector applied from the patients right side. It is not necessary to take down any short gastric vessels as long

as this technique is used. There is always ample fundus to use for the wrap using this dissection approach. The crura are always approximated behind the esophagus, taking a large bite of muscle and peritoneum. One stitch is usually all that is necessary, and the strength of that stitch is in the peritoneal lining. The size of the hiatal opening is then measured using the end of the USSC roticulating bowel grasper. It should be 3 to 3.5 cm from front to back, large enough to accommodate a 60 Fr. bougie, but eliminating the actual need to use one (Fig. 3). Once the size of the opening is confirmed, the roticulating grasper can then be passed behing the esophagus to grasp the fundus 5 cm lateral to the gastroesophageal junction as it is passed by the other surgeon. A sling maneuver is then performed, passing the fundus back and forth to be certain it is slack and has no twists. The sling is measured to 6 cm in length for a partial fundoplication, and to 8 to 9 cm for a complete fundoplication. This measurement is based on the formula {3.14 diameter} for determining the circumference of a circle, in this case, an imaginary esophagus with an imaginary 60 Fr. bougie in it that has a total diameter of 2.5 to 3 cm (Fig. 4). Measuring eliminates the actual need for

628 Table 1. Patient dataa Number Men Women Esophagitis, grade 13 Grade 4 esophagitis with stricture Sliding hiatal hernia Barretts metaplasia Respiratory complications Gallstones Epiphrenic diverticulum Hypertensive esophagus Decreased esophageal motility Pre-op esophageal manometry Pre-op pH testing Combined bile/acid reflux
a

Table 2. Total group resultsa Percent 44 56 77 23 88 12 5 8 1 1 4 27 2 1 Cases attempted Cases completed laparoscopically Operative time, first 49 cases (average) Operative time, last 177 cases (average) Hospital stay, first 50 cases (average) Hospital stay, last 177 cases (average) Hospital charges (ave) laparoscopic Hospital charges (ave) open Minor complication rate (3 blood transfusions, 2 trocar site hernias, 12 ileus, 10 subcutaneous emphysemas, 2 pleural effusions, 1 gastritis) Revisions (5 open, 2 laparoscopic)
a

100 126 174 52 198 26 12 19 2 2 10 62 3 2

231 226 (98%) 1 hour 25 minutes 2.3 days 1.4 days $6,000 $12,000 10.9% 7 (3%)

226 patients, ages 8 to 84, underwent a laparoscopic antireflux procedure. Important pre-op findings are listed. Manometry and pH testing were used selectively for cases with an uncertain or questionable diagnosis based on history, physical findings, UGI series, fluoroesophagography, endoscopy, or biopsy.

bougie placement and the potential for tissue trauma or esophageal or gastric perforation. Suture placement for the Nissen or Nissen-Rosetti is well known, two stitches commonly being used to fasten the right and left sides of the fundal sling to the anterior esophagus at the gastroesophageal junction and 2 cm above it, being careful to avoid the anterior vagus nerve. The Endostitch (USSC) instrument is used to do all of the sewing, as it can significantly reduce operative time. Many different suture placements for the partial fundoplication have been described. We use a five-stitch technique. The first stitch incorporates the esophagus at the right gastroesophageal junction, the fundus, and the pre-aortic fascia and muscle below the junction of the crural leaves with the first stitch. The esophagus, fundus, and right crus at the level of the crural reapproximation are incorporated with the second, thus fixing in place at least a 2-cm intraabdominal length of esophagus (Figs. 5 and 6). The final two sutures fasten the anterior superior fundus on the left to the left anterior gastroesophaeal junction and to the esophagus 2 cm above that point (Fig. 7). The completed partial wrap should leave at least a 1-cm area of the anterior esophagus bare. This will allow proper relaxation of the wrap and esophagus with swallowing and always create an adequate highpressure zone. A third suture incorporates the right anterior esophagus and the midportion of the wrap, bringing the right side of the partial fundoplication into its final position (Fig. 7). Upon completion of the procedure, the skin is closed with clips, the port sites are injected with Marcaine, and the NG tube is pulled. Oral Toradol and PCA Demerol are used if the patient has any significant pain. A single dose of IV Zofran and Ancef is given preoperatively for control of postop nausea and infection prophylaxis, respectively. Diet is advanced from full liquids to regular as quickly as the patient can tolerate it, generally within a few hours of surgery. The patient is usually discharged within 24 h; up to 10% may go home the day of surgery. Skin clips are replaced with benzoin and Steristrips at discharge. No carbonated beverages are allowed for at least a few days. Very cold drinks are discouraged, as they cause esophageal spasm in the immediate postop period.

Three open revisions were due to the development of a paraesophageal hernia or symptomatic migration of the wrap into the chest. All occurred in the first seven cases of the series who did not have the crura reapproximated. All subsequent patients had crural reapproximations, with no such further complications. One revision was done due to persistent pain. One revision was done due to adhesion formation 6 months postop, resulting in dysphagia unresponsive to dilatation. One revision was done for persistent dysphagia. One revision was done for a technically imperfect wrap. (The only revision necessary in the partial group.) The laparoscopic approach to revision was not difficult, as few adhesions were encountered. Six of the seven revisions were done in the Nissen-Rosetti group. Three were converted from a Nissen-Rosetti to a partial fundoplication. Four of the seven revision cases were done for potentially avoidable technical factors. A 1% revision rate can be expected for technically perfect procedures.

square test for determining probability values for certain comparative data sets within the two subgroups of patients.

Results The overall group results are summarized in Table 2. Major reductions in operative time and length of hospital stay were noted during the course of the study. Laparoscopic costs were dramatically less than open costs. Complication rates and revision rates were low. There were no gastric or esophageal perforations. All but five cases (2%) were completed laparoscopically. Reasons for conversion to an open procedure included extensive adhesions from previous surgery in two, fragile tissues unable to tolerate the trauma of the laparoscopic instruments in one, a bowel injury while placing a trocar using an open technique in one with multiple previous surgeries and dense adhesions, and bleeding from an aberrant left hepatic artery in one. Major operations done concurrently with the antireflux procedure included cholecystectomy in 19, umbilical or inguinal hernia repair in four, highly selective vagotomy in one, epiphrenic diverticulectomy in one, and cricopharyngeal myotomy in one for a symptomatic Zenkers diverticulum. A total of 11 (5%) postop endoscopies were done for prolonged dysphagia after the laparoscopic Nissen-Rosetti fundoplication or for follow-up of Barretts metaplasia, with none demonstrating any visual evidence of esophagitis. Those with Barretts metaplasia all had a decrease in the amount and severity of inflammation and no progression of disease, but there was no regression of the abnormal epithelium.

Patient data
Patient data for the total group is summarized in Table 1. The patients were referred by 19 family physicians, four internists, six surgeons, and two gastroenterologists. A total of 37 communities and five states are represented by the group. All patients were seen at 1 and 6 weeks postoperatively, at least. Further follow-up was done by a standardized questionnaire that was sent out to all participants in the study in late 1995, 1 month to 3 years after initial surgery. Data analysis was done using the Mann-Whitney test and the chi-

629 Table 3. Objective comparative group resultsa Nissen-Rosetti 1. Number of patients 2. Median follow-up 3. Dysphagia requiring dilatation 4. Preop LES pressure (62 patients) 5. Postop LES pressure 6. Measured length of LES pressure zone 7. Recurrent postop stricture requiring dilatation (both within 3 weeks of surgery)
a

Partial fundoplication 101 12 months 0 (0%)

125 30 months 8 (6%) (p < 0.01) 2.7 mmHg (average both groups) 15 mmHg (average) (15 patients) 2.03.0 cm 2 (1.6%)

15 mmHg (average) (10 patients) 2.03.0 cm (both groups) 0

Most patients refused repeat postop studies, as they felt clinically well; 25 agreed to postop esophageal manometry.

All patients with respiratory symptoms thought to be related to esophageal reflux had either complete resolution of all symptoms or a dramatic improvement with decreased bronchodilator requirements. Two who were on prednisone for refractory asthma were taken off of that drug within 2 months of surgery. All reported decreased shortness of breath and improved exercise tolerance. Measurable objective comparisons between the NissenRosetti group and the partial fundoplication group are summarized in Table 3. Subjective comparisons between the two groups based on the responses to the standardized questionnaire are summarized in Table 4.

Discussion Laparoscopic antireflux surgery has rapidly evolved as the treatment of choice for complicated gastroesophageal reflux disease in patients who can withstand a general anesthetic. All of the standard procedures have been applied laparoscopically with excellent results and minimal complication rates. It would appear that antireflux surgery is a nearly perfect type of operation for the laparoscope, as it is a matter of only dissection and repositioning of organs into their proper locations, with minor modifications. The exposure with modern equipment is superb, and advancing technology has made dissecting and suturing an easier task. Unfortunately, it has been our experience that many patients are not referred when it first becomes evident that medical management is not working, resulting in a considerable delay (sometimes years) before definitive surgical treatment can be undertaken. The usual explanation for the delay is that the patients condition is not yet bad enough, or the expected side effects of the surgery are worse than the disease itself (a clear misconception based on the long, tight wrap of old). This untimeliness of intervention can result in permanent esophageal fibrosis, injury, and dysfunction that surgery cannot alleviate, emphasizing the importance of recognizing early on those patients that require surgical treatment [16]. Patients with combined acid/bile reflux do not respond well to treatment with H2 blockers or proton-pump antagonists, so prolonged treatment with those agents in the face of this problem is con-

traindicated [8]. Those patients with respiratory complications of GERD should be operated upon early, as no amount of acid-reducing or propulsive medication is going to solve their respiratory problems. Severe lung damage can occur over the years, some of which may not be reversible. Progression to dysphagia and stricture in spite of maximal medical therapy is a clear indication for surgery. Repeated stricture dilatation in this situation is treating a complication of a disease process rather than treating the disease itself and is inappropriate. Persistent symptoms in spite of maximal medical management, continued objective evidence of esophagitis in spite of maximal medical management, esophageal ulcer formation, hemorrhage, and Barretts epithelium are all indications for surgery. A shortened esophagus is not a contraindication to a laparoscopic approach to surgical repair. On the contrary, visualization is superb in the lower mediastinum with a 45 lens, and dissection and lengthening of the retracted esophagus is relatively straightforward. The main fibers holding the retracted esophagus in the mediastinum are the dense posterior phrenoesophageal ligaments, which must be transected posterior to the esophagus, thus releasing the organ to its usual length. The anterior and lateral esophageal attachments, though important to release in the lower mediastinum, are not the main retractile elements holding the esophagus in an abnormal location. We have yet to see a true shortened esophagus, although we have operated on many with that preoperative diagnosis. We question the existence of such an entity; retracted esophagus better describes the condition and reflects the fact that the esophagus is in fact of normal length, not shortened. Barretts metaplasia is presumed to be the end result of inadequately treated GERD and significantly increases the risk of adenocarcinoma of the esophagus. This risk appears to be highest in patients with combined acid and bile reflux. The number of cases of Barretts metaplasia in our series is discouraging and indicates a failure of adequate medical treatment or delay in referral in every case. These patients are not only exposed to an increased risk of cancer, but also must undergo semiannual endoscopy with biopsy for surveillance, an expensive and anxiety-provoking situation for the patient. A number of these people will go on to develop carcinoma and will require esophagectomy. Clearly there

630 Table 4. Subjective comparative surveya Nissen-Rosetti 1. Number of respondents 2. Symptomatic control (1 no improvement, 10 cured) 3. Return to completely normal eating habits 4. Side effects of gassiness, spasm, or bloating (temporary) 5. Inability to burp 6. Need for prescription meds for esophageal symptoms 7. Return to work 8. Return to daily routine
a

Partial fundoplication 62 (62%) 9.8 (average) (98% 8, 9, or 10) 13 days 62% 0 0 10 days 9 days

82 (66%) 9.2 (average) (90% 8, 9, or 10) 83 days (p < 0.0001) 86% 12 (15%) (p < 0.001) 2 (Propulsid) (1.6%) 17 days (p 0.22) 11 days

Results of a standardized questionnaire sent to all patients are listed. The overall satisfaction rate and general performance of those in the partial fundoplication group were superior.

are ample reasons to promote a curative surgical approach before the development of such complications. Is there any advantage to performing a 360 (Nissen) wrap? The side effects of the Nissen wrap include flatulence, early satiety, mild weight loss, nausea, odynophagia, dysphagia, bloating, and difficulty belching and vomiting. The majority of these side effects soften with time, are tolerable to the patient, and are a reasonable trade-off for their usual symptoms of GERD. However, the surgeon must be prepared to help the patient deal with these side effects, which may be long term for many [17]. Cure rates for partial fundoplication procedures are identical to the Nissen procedure, objective findings of postoperative pH testing and esophageal manometry are virtually identical, and the side effects are definitely fewer [1012, 18]. Our findings clearly support the findings of others who have documented equal cure rates and objective measurements of wrap function as well as increased patient satisfaction due to decreased side effects for the partial fundoplication compared to the Nissen fundoplication. The ability of those undergoing the partial wrap to resume intake of regular food immediately after surgery is impressive in our experience. It virtually eliminates the risk of significant dysphagia and the need for prolonged diet modification. A partial fundoplication can be done on any patient with complicated GERD and is the preferred wrap for those with scleroderma or a hypomotile esophagus. We fully expect to see the partial fundoplication become the procedure of choice for the surgical treatment of complicated GERD due to its ease of application, well-documented curative outcome, and few side effects. There is no circumstance when a complete wrap would be superior to a partial one, except in the case of an intraoperative distal esophageal perforation, where a complete wrap may be necessary to reinforce the repair of the perforation, something that should be an extremely rare event using a measuring technique to properly fashion the fundoplication. In general, patients that present with a classic history for GERD and endoscopic and radiographic findings that confirm the problem always have an incompetent lower esophageal sphincter mechanism on manometry and have abnormal 24-h pH testing [19]. Therefore, these straightforward cases do not require manometry and 24-h pH testing for

diagnostic workup (63% of patients in this series) [5]. Those cases with an atypical presentation suggestive of esophageal motility problems, gastric emptying abnormalities, medication-induced disease, or primary duodenogastric bile reflux must all have esophageal manometry and may need a combination of tests including 24-h pH testing, gastric emptying studies, and biliary scans. Respiratory complaints are the most difficult to evaluate, so a high index of suspicion is necessary for those presenting with a combination of asthma and GERD. The patient with complicated GERD is extremely pleased with the outcome of laparoscopic antireflux surgery [2, 13, 20]. Reduced pain, very short hospitalization, lower cost, and quick return to normal activities and work make this approach very attractive, although patient satisfaction appears to be more closely related to relief of symptoms than to the technical approach to the procedure [13]. In addition, the ability to eliminate all reflux-related medications is a big plus from a convenience and cost standpoint. The cost savings over 10 years for a surgical approach to complicated GERD has been estimated to be $1,500, and over 20 years to be $10,000, taking into consideration cost variations across the country as well as surgical and medical complications and reoperation [4]. In summary, surgery has been conclusively demonstrated to be significantly more effective than medical management of complicated gastroesophageal reflux disease as far as symptom improvement and objective signs of esophagitis are concerned, with excellent long-term results [14, 15]. It is also the only method of reestablishing the normal reflux barrier of the lower esophageal sphincter mechanism, allowing the return of normal esophageal and gastric physiology. Restoration of function requires the restoration of normal anatomy, meaning elimination of the hiatal hernia if present and creation of a fixed intraabdominal esophageal component. A laparoscopic approach to curative treatment of complicated GERD has the added benefits of significantly reducing cost, recovery time, morbidity, and time lost from work. A partial fundoplication has equally good results compared to the Nissen fundoplication, with fewer side effects and better patient satisfaction. Patients should be referred for curative treatment before they have developed irreversible esophageal damage. By utilizing two surgeons

631

and currently available technology, operative time, hospital time, and complications can be reduced to a finite minimum. The economic benefits to the patient and employers as well as third-party payers are significant as far as total cost savings and minimizing lost work hours go. Laparoscopic antireflux surgery for complicated GERD may well prove to be one of the most important developments in the field of laparoscopic general surgery because of the number of patients who can benefit from it and the overall expected long-term improvement in their health as well as the positive overall impact on economic costs.
Acknowledgment. Special thanks to Tom Moore, Ph.D., for statistical analysis work. Special thanks to Carlos Ferguson for supporting art work.

References
1. Aye RW, Hill LA, Kraemer JM, Snopkowski P (1994) Early results with the laparoscopic hill repair. Am J Surg 167: 542546 2. Coster DD, Bower WH, Wilson VT, Butler DA, Locker SC, Brebrick RT (1995) Laproscopic Nissen fundoplicationa curative, safe, and cost-effective procedure for complicated gastroesophageal reflux disease. Surg Laparosc Endosc 5(2): 111117 3. Cuschieri A (1993) Laparoscopic antireflux surgery and repair of hiatal hernia. World J Surg 17(1): 4045 4. Deloitte & Touche Management Consulting (1995) Economic impact of laparoscopic Nissen fundoplication, executive summary 5. De Meester TR, Wang CI, Wernly JA, Pellegrini CA, Little AG, Klementschitsch P, Bermudez G, Johnson LF, Skinner DB (1980) Technique, indications, and clinical use of 24 hour esophageal pH monitoring. J Thorac Cardiovasc Surg 79: 656670 6. Hill LD, Kraemer SJM, Aye RW, Kozarek RA, et al (1994) Laproscopic hill repair. Contemp Surg 44(1): 1320 7. Hinder RA, Filipi CJ, Wetscher G, Neary P, et al (1994) Laparoscopic Nissen fundoplication is an effective treatment for gastroesophageal reflux disease. Ann Surg 220(4): 472483

8. Hinder RA, Filipi CJ (1995) The laparoscopic management of gastroesophageal reflux disease. Adv Surg 28: 4158 9. Kraemer SJM, Aye R, Kozarek RA, Hill LD (1994) Laparoscopic hill repair. Gastrointest Endosc 40(2, Part 1): 155159 10. Lundell L, Abrahamsson H, Ruth M, Snadberg N, Olbe LC (1991) Lower esophageal sphincter characteristics and esophageal acid exposure following partial or 360 degree fundoplication: results of a prospective, randomized, clinical study. World J Surg 15: 115121 11. McKernan JB (1994) Laparoscopic repair of gastroesophageal reflux disease/Toupet partial fundoplication versus Nissen fundoplication. Surg Endosc 8: 851856 12. Mosnier H, Leport J, Aubert A, Kianmanesh R, et al (1995) A 270 degree laparoscopic posterior fundoplasty in the treatment of gastroesophageal reflux. J Am Coll Surg 181: 220224 13. Rattner DW, Brooks DC (1995) Patient satisfaction following laparoscopic and open antireflux surgery. Arch Surg 130: 289294 14. Shirazi SS, Schulze K, Soper RT (1987) Long-term follow-up for treatment of complicated chronic reflux esophagitis. Arch Surg 122: 548551 15. Spechler SJ (1992) Comparison of medical and surgical therapy for complicated gastroesophageal reflux disease in veterans. New Engl J Med 326(12): 786792 16. Stein HJ, Eypasch EP, De Meester TR, Smyrk TC, Attwood SE (1990) Circadian esophageal motor function in patients with gastroesophageal reflux disease. Surgery 108: 769777 17. Swanstrom L, Wayne R (1994) Spectrum of gastrointestinal symptoms after laparoscopic fundoplication. Am J Surg 167: 538541 18. Thor KBA, Silander T (1989) A long-term randomized prospective trial of the Nissen procedure versus a modified Toupet technique. Ann Surg 719724 19. Waring JP, Hunter JG, Oddsdottir M, Wo J, Katz E (1995) The preoperative evaluation of patients considered for laparoscopic antireflux surgery. Am J Gastroenterol 90(1): 3538 20. Weerts JM, Dallemagne B, Hamoir E, Demarche M, et al (1993) Laparoscopic Nissen fundoplication: detailed analysis of 132 patients. Surg Laparosc Endosc 3(5): 359364 21. Yang HK, Del Guercio LRM, Steichen FM (1995) Thoracoscopic Belsey-Mark IV fundoplication. Surg Rounds 277291

Case reports
Surg Endosc (1997) 11: 668670

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic repair of a diaphragmatic hernia through the foramen of Morgagni

M. Orita,1 M. Okino,1 K. Yamashita,1 N. Morita,2 K. Esato2
1 2

Department of Surgery, Onoda City Hospital, 1863-1 Higashitakadomari, Onoda, Yamaguchi, 756 Japan First Department of Surgery, Yamaguchi University School of Medicine, 1144 Kogushi, Ube, Yamaguchi, 756 Japan

Received: 3 April 1996/Accepted: 3 May 1996

Abstract. A 78-year-old woman is described who presented with a diaphragmatic hernia through the foramen of Morgagni. A definitive diagnosis was confirmed by a sagittal view on magnetic resonance imaging prior to surgery. The hernia was repaired laparoscopically under an abdominal wall lifting technique without pneumoperitoneum, and her symptoms completely resolved postoperatively with no evidence of recurrence. The laparoscopic repair was considered a suitable and safe procedure for the treatment of a Morgagni hernia. Key words: Morgagni hernia Abdominal wall lifting technique without pneumoperitoneum Laparoscopy Omentum

Diaphragmatic hernias through the foramen of Morgagni are rare, and often the preoperative diagnosis is difficult. The standard surgical procedure has required a laparotomy or a thoracotomy for symptomatic patients. We report a case of a Morgagni hernia which was diagnosed prior to surgery and repaired laparoscopically, and our patient was the first case repaired under an abdominal wall lifting technique without pneumoperitoneum. Case report
A 78-year-old woman, complaining of epigastral discomfort and tenderness, presented in our clinic in February 1995. She was 139 cm in height and weighed 56 kg. She had a history of progressive weight gain of 19 kg over the previous year secondary to poor control of her hypothyroidism. Findings from all routine laboratory studies were normal; however, pulmonary function tests showed a restrictive pattern. PA chest radiographs demonstrated an abnormal shadow with a clear border present at the right cardiophrenic angle; the lateral projection determined an anterior location of the shadow. Chest computed tomography

(CT) exhibited a homogeneous, solid mass (6 11 cm) with smooth margins and a region of fat density in the right anterior mediastinum (Fig. 1A). A mediastinal lipoma was the suspected diagnosis at this time. The sagittal view of a magnetic resonance imaging (MRI), however, showed continuous fatty tissue anterior to the liver through an anteromedial portion of diaphragm directly behind the xiphoid (Fig. 1B). Thus, an omental herniation through the foramen of Morgagni was considered to be the diagnosis. Laparoscopic repair was performed on February 27, 1995. Under general anesthesia, an abdominal wall lifting technique was employed without pneumoperitoneum [4, 6]. Two Kirschner wires (1.2 mm in diameter) were tunneled subcutaneously, one transversely just above the umbilicus, and another transversely just below the xiphoid process. Both wires were attached to lifting handles, which were raised upright, and then the abdominal wall was lifted. A laparoscope was introduced through the umbilicus. There appeared to be an oval-shaped defect (2 3.5 cm) in the anterior aspect of the diaphragm, containing most of the omentum (Fig. 2A). The omentum slid synchronously with her respirations. Two additional ports (15 mm in diameter) then were placed on both sides of the costal margins, through which the hernia contents were gently pulled down into the peritoneal cavity with grasping forceps (Fig. 2B). There were no adhesions between the omentum and the hernial sac. The hernial space could be approached and visualized more easily with laparoscopy than by laparotomy or thoracotomy (Fig. 2C). The hernia defect was closed with five 2-0 silk interrupted sutures, using an intra-abdominal suturing technique (Figs. 2D, 3A,B). The sac was not removed. A thin piece of Prolene mesh (3.5 5 cm, Ethicon, Inc. Somerville, NJ, U.S.A.) was placed on the closed hernia ring and fixed to the diaphragm with hernia stapler (Auto Suture, ENDO UNIVERSAL 65, United States Surgical Corporation, Norwalk, CT, U.S.A.) in order to reinforce the repaired site (Fig. 3C,D). A Penrose drain was placed in the subphrenic space to decompress the air. The patient recovered nicely and has had no further symptoms or evidence of recurrence following surgery.

Discussion Hernias which occur in the retroxiphoid region are called hernias of the foramen of Morgagni. These patients are frequently obese adults. Our particular patient had rapid weight gain due to the poor control of her hypothyroidism. The differential diagnosis included a pleuropericardial cyst, pleural mesothelioma, pericardial fat pad, mediastinal lipoma, tumor or cyst of the diaphragm, thymoma, and ante-

Correspondence to: M. Orita

669

Fig. 1. A Chest computed tomography demonstrating a large, retrosternal, homogeneous mass with smooth margins and a fat density. B A sagittal view on magnetic resonance imaging showing continuous fatty tissue anterior to the liver through an anteromedial portion of the diaphragm directly behind the xiphoid.

Fig. 2. A A laparoscopic view of the Morgagni hernia containing omentum extending from the transverse colon. B The omentum was pulled back down into the peritoneal cavity easily with grasping forceps. C An oval-shaped defect (2 3.5 cm) in the anterior aspect of the diaphragm. D The hernia defect was closed with an intra-abdominal suturing technique; the sac itself was not removed.

Fig. 3. A,B The threads were ligated intra-abdominally. C,D A small piece of Prolene mesh was placed on the closed hernia ring and fixed to the diaphragm with hernia stapler in order to reinforce the repaired site.

670 Table 1. Laparoscopic repair in cases of Morgagni hernias Author Kuster et al. 1992 [5] Rau et al. 1994 [8] Newman et al. 1995 [7] Smith and Ghani 1995 [10] Orita et al. 1996 Age 67 42 57 22 70 60 78 Gender Female Male Female Female Female Female Female Diagnosis Laparoscopic Preoperative Laparoscopic Incidental Incidental Incidental Preoperative Side of the lesion Right Right Right Right Right Right ? Right Size of the defect ? 6 cm ? ? 10 15 cm 2 3.5 cm 2 3.5 cm Contents Omentum, colon Omentum Omentum, colon Liver ? Omentum, colon Omentum Removal of the sac Not removed Removed Removed Removed ? Removed ? Not removed Not removed Mesh placement Not placed Placed Placed Not placed Not placed Not placed Placed

rior chest wall tumor [2]. Often the diagnosis is a difficult one to make preoperatively, especially when the sac contains only omentum. In our case the sagittal view of an MRI scan was very useful for making this distinction [11]. Operative repair is recommended in symptomatic cases or for suspected strangulation [3]. Since this hernia occurs more frequently in obese or elderly patients [9], the laparoscopic approach seems more suitable than the abdominal or transthoracic approach, especially if the preoperative diagnosis can be established. There were six case reports of Morgagni hernias repaired laparoscopically before our case [5, 7, 8, 10] (Table 1). They completed the repair under a pneumoperitoneum without intraoperative complications. Pneumoperitoneum with Morgagni hernias, however, may lead to respiratory or circulatory complications [1]. Furthermore, the intraperitoneal suturing technique is facilitated by employing an abdominal wall lifting technique without pneumoperitoneum. We, therefore, conclude that the laparoscopic repair of a Morgagni hernia using an abdominal wall lifting technique instead of pneumoperitoneum is technically easy, safe, and a less invasive approach to surgical treatment.

References
1. Chin EF, Duchesne ER (1955) The parasternal defect. Thorax 10: 214219 2. Comer TP, Clagett OT (1966) Surgical treatment of hernia of the foramen of Morgagni. J Thorac Cardiovasc Surg 52: 461468 3. Fisher L, ODonnell CJ (1990) A complication of a Morgagni hernia. Australas Radiol 34: 8688 4. Hashimoto D, Nayeem SA, Kajiwara S, Hoshino T (1993) Laparoscopic cholecystectomy: an approach without pneumoperitoneum. Surg Endosc 7: 5456 5. Kuster GGR, Kline LE, Garzo G (1992) Diaphragmatic hernia through the foramen of Morgagni: laparoscopic repair case report. J Laparoendosc Surg 2: 93100 6. Nagai H, Kondo Y, Yasuda T, Kasahara K, Kanazawa K (1993) An abdominal wall-lift method of laparoscopic cholecystectomy without peritoneal insufflation. Surg Laparosc Endosc 3: 175179 7. Newman L, Eubanks S, Bridges WM, Lucas G (1995) Laparoscopic diagnosis and treatment of Morgagni hernia. Surg Laparosc Endosc 5: 2731 8. Rau HG, Schardey HM, Lange V (1994) Laparoscopic repair of a Morgagni hernia. Surg Endosc 8: 14391442 9. Saha SP, Mayo P, Long GA (1982) Surgical treatment of anterior diaphragmatic hernia. South Med J 75: 280281 10. Smith J, Ghani A (1995) Morgagni hernia: incidental repair during laparoscopic cholecystectomy. J Laparoendosc Surg 5: 123125 11. Yeager BA, Guglielmi GE, Schiebler ML, Gefter WB, Kressel HY (1987) Magnetic resonance imaging of Morgagni hernia. Gastrointest Radiol 12: 296298

Letters to the editor

Surg Endosc (1997) 11: 696

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The totally extraperitoneal laparoscopic hernia repair

We read with interest the paper by Vanclooster and colleagues [11] and commend their contribution to this procedure. However, we would offer three comments: First, totally extraperitoneal laparoscopic hernia repair was developed by Dulucq [13] in 1989/90 and by McKernan [7] not a great deal later. While it seems legitimate for others to publish their own technical variations, developments, and outcomes, we strongly support recent reminders [6] that journal editors and their peer referees owe the reader a duty of diligence: they should insist that authors exercise proper scholarship by giving credit where it is due. Otherwise the uninformed reader may assume originality and the informed may infer plagiarism, where the author intended neither. Second, the mesh configuration suggested by Vanclooster et al. was presented by one of us several years ago [4, 10]. However, the concept of amputating the inferior and lateral corner so the mesh fits better on the iliac vessels and the psoas muscle is flawed, since it has subsequently been reported [12] that recurrences may occur dorsal/ inferior to this inferolateral corner. The most extensive possible coverage of the psoas muscle belly is therefore appropriate. Third, the need for mesh fixation remains debatable. However, to fix the cranial border to prevent early migration or slipping is illogical: In our joint experience of over 1,000 cases and, to our knowledge in all reports in the world literature, recurrences pass uniformly caudal to the inferior border of the prosthesis. Fixation of the inferior medial part of the mesh to Astley Coopers ligament alone [5] may not offend against the original tension-free notion of Stoppa [8, 9] nor interfere with the mechanics of prosthesis retention. To fix the superior border to points that move relative to one another within a musculofascial structure contravenes both principles. Finally, on a minor point, if the structure annotated as D in Fig. 1 is the testicular vascular bundle, where is the vas deferens? Despite these comments we congratulate the authors on their low complication rate. References
1. Dulucq J-L (1991) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritone ale en retrope ritone oscopie. Cah Chir 79: 1516 2. Dulucq J-L (1992) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritoneal en pre -pe ritoneoscopie. Chirurgie 118(12): 8385 3. Dulucq J-L (1992) The treatment of inguinal hernias by implantation of mesh through retroperitoneoscopy. Postgrad Gen Surg 4: 173174 4. Fiennes AGTW, Taylor RS (1994) Learning laparoscopic hernia repair. In: Inguinal hernia, advances or controversies? Arregui M, Nagan R (eds) Radcliffe, Oxford, pp 475482 5. Fiennes AGTW, Himpens J, Dulucq J-L (1994) Preperitoneoscopic groin hernioplasty, current synthesis. Surg Endosc 8(8): 989 6. Horton R, Smith R (1996) Time to redefine authorship (editorial). Br Med J 312: 723 7. McKernan JB (1993) Laparoscopic extraperitoneal repair of inguinofemoral herniation. Endosc Surg Allied Tech 1(4): 198203 8. Stoppa R, Petit J, Abourachid H (1973) Proce de original de plastie des hernies de laine. Linterposition sans fixation dune prothe ` se en tulle de Dacron par voie me diane pre pe ritone ale. Chirurgie 99: 119 9. Stoppa RE, Rives JL, Warlaumont CR (1984) The use of Dacron in the repair of hernias of the groin. Surg Clin North Am 64: 269285 10. Taylor RS, Fiennes AGTW (1992) A tension free modification of the Dulucq preperitoneal laparoscopic hernioplasty. Min Inv Ther 1(Suppl 1): 101 11. Vanclooster P, Meersmann AL, de Gheldere CA, Van de Ven CK (1996) The totally extraperitoneal laparoscopic hernia repair. Surg Endosc 10: 332335 12. Vivekanadan S, Fiennes AGTW (1995) Totally extraperitoneal groin hernioplasty: mechanism of delayed indirect recurrence. Min Inv Ther 4(Suppl 1): 55

A. Fiennes
Department of Surgery St Georges Hospital Medical School Cranmer Terrace London, SW17 ORE, United Kingdom

J. Himpens
Department of Digestive Surgery University Hospital Ste Pierre Rue Haute 201 B-1000 Brussels, Belgium

Surg Endosc (1997) 11: 697

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank you for the opportunity to answer the letter of Mr. Fiennes. First of all, we certainly do acknowledge that Mr. Dulucq and Mr. Mc Kernan were the pioneers of the extraperitoneal laparoscopic hernia repair. Honor to whom honor is due. We did not intend to pretend to be the pioneers of this technique. We wished only to describe the technique the way we perform it, to describe our own findings, and to give a fair report of our preliminary results. The reason for cutting the inferolateral corner is not just that we think it fits better on the iliopsoas but also because we are afraid to cause damage to the nerves running on it by dissecting unnecessarily high on the muscle. Since the mesh measures 15 15 cm, we do not think we compromise the strength of the repair by merely removing a small piece of its inferolateral corner. We think that dissecting very high on the muscle just to position the whole inferolateral corner of the mesh flat on the muscle is unnecessary and dangerous. We do agree totally that fixation of the mesh is unnecessary provided the mesh is large enough, which is obviously the case when using a 15 15 cm mesh. In fact, we have not fixed the mesh since January 1996. We also agree that the vas deferens is not clearly seen on Fig. 1. We chose this shot because of the clearly visible large direct defect.

C. de Gheldere
Heilig Hart Ziekenhuis Kolveniersvest 20B-2500 LIER Belgium

P. Vanclooster
Bouwelsesteenweg 6 2560 Nijlen Belgium

Surg Endosc (1997) 11: 698

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The influence of pneumoperitoneum on the peritoneal implantation of free intraperitoneal cancer cells
Recently Hubens et al. published an interesting article entitled The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells [2]. They reported on the possible implantation of cancer cells at trocar wounds at the moment of deflation as cells are forced through these wounds by the pressure gradient created by the pneumoperitoneum during laparoscopic surgery for malignant disease. We would like to point out that this chimney effect, as originally described by us, can occur during the entire laparoscopic procedure and not only at the moment of deflation, as leakage of CO2 alongside trocars during surgery is impossible to prevent with the existing trocars [3]. Consequently, deflation of the pneumoperitoneum by letting CO2 escape through one of the trocars before pulling these trocars out of the abdomen will not prevent the occurrence of entrapment of cancer cells in the trocar wounds. We fully agree with the authors on the possible advantage of gasless laparoscopy as this could prevent the chimney effect. In our experimental work we found significantly less tumor growth at the port sites following gasless laparoscopic surgery for colon cancer in the rat as compared to laparoscopic surgery using a CO2 pneumoperitoneum. This technique seems promising to treat malignant disease laparoscopically.
Correspondence to: G. Kazemier

References
1. Bouvy ND, Marquet RL, Jeekel J, Bonjer HJ (1996) Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Surg Endosc 10: 551 2. Hubens G, Pauwels M, Hubens A, Vermeulen P, Van Marck E, Eyskens E (1996) The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells. Surg Endosc 10: 809812 3. Kazemier G, Bonjer HJ, Berends FJ, Lange JF (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 11411142

G. Kazemier1 F. J. Berends1 N. D. Bouvy1 J. F. Lange2 H. J. Bonjer1

1

Department of Surgery University Hospital Rotterdam-Dijkzigt Dr Molewaterplein 40 3015 GD, Rotterdam The Netherlands 2 Department of Surgery St. Clara Hospital Rotterdam The Netherlands

Surg Endosc (1997) 11: 699

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank Dr. Kazemier et al. for their kind remarks and fully agree with them that gas leakage can occur during the entire procedure with subsequent implantation of tumor cells at the trocar sites. At the moment we are conducting further experimental studies on the possible effects of gas leakage on tumor cell implantation and the chimney effect, as they have called it. Results will be ready for publication soon. G. Hubens
Department of Surgery University Hospital University of Antwerp Wilrijkstraat 10 2650 Edegem Belgium

Surg Endosc (1997) 11: 645649

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic cholecystectomy using abdominal wall retraction

Hemodynamics and gas exchange, a comparison with conventional pneumoperitoneum
D. W. Meijer,1,2 B. P. M. Rademaker,3 S. Schlooz,3 W. A. Bemelman,4,5 L. T. de Wit,4 J. J. G. Bannenberg,6 T. Stijnen,2 D. F. Gouma4
1

Working Group Development Surgical Technology, Surgical Division, Academic Medical Centre, IWO-gebouw 1 etage k151, University of Amsterdam, Amsterdam, The Netherlands 2 Department of Epidemiology and Biostatistics NIHES, Erasmus University Rotterdam, Rotterdam, The Netherlands 3 Department of Anesthesiology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands 4 Department of Surgery, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands 5 Department of Surgery, Academic Hospital Leiden, University of Leiden, Leiden, The Netherlands 6 Department of Experimental Surgery, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands Received: 28 May 1996/Accepted: 14 October 1996

Abstract Background: Disadvantages related to CO2 pneumoperitoneum have led to development of the abdominal wall retractor (AWR), a device designed to facilitate laparoscopic surgery without conventional pneumoperitoneum (15 mmHg CO2). We investigated the effects of the AWR on hemodynamics and gas exchange in humans. We also investigated whether the use of an AWR imposed extra technical difficulties for the surgeon. A pilot study revealed that cholecystectomy without low-pressure pneumoperitoneum was technically impossible. Methods: A prospective randomized controlled trial: Twenty patients undergoing laparoscopic cholecystectomy were randomly allocated into group 1: AWR with lowpressure pneumoperitoneum (5 mmHg), or group 2: conventional pneumoperitoneum (15 mmHg). Results: Surgery using the AWR lasted longer, 72 16 min (mean SD) vs 50 18 min compared with standard laparoscopic cholecystectomy. There were no differences between the groups with respect to hemodynamic parameters, although a small reduction of the cardiac output was observed using conventional pneumoperitoneum (from 3.9 0.7 to 3.2 1.1 l/min) and an increase during AWR (from 4.2 0.9 to 5.2 1.5 l/min). Peak inspiratory pressures were significantly higher during conventional pneumoperitoneum compared to AWR. A slight decrease in pH accompanied by an increase in CO2 developed during pneumoperitoneum and during the use of the AWR. In both groups arterial PO2 decreased.

Conclusions: The results indicate that the view was impaired during use of the AWR and therefore its use was difficult and time-consuming. Possible advantages of this devices effects on hemodynamics and ventilatory parameters could not be confirmed in this study. Key words: Abdominal wall retraction Abdominal wall retractor Pneumoperitoneum

Correspondence to: D. W. Meijer, Department of Surgical Research, IWO gebouw I-151, Academic Medical Centre, Meibergdreef g, 1105AZ Amsterdam, The Netherlands

Carbon dioxide (CO2) pneumoperitoneum of 15 mmHg intraabdominal pressure is generally used for laparoscopic surgery. Side effects of a pneumoperitoneum such as cardiovascular depression and respiratory acidosis have been described and may be potentially dangerous in patients with underlying diseases [12]. In addition, CO2 embolism is a feared, although rare, complication of laparoscopic surgery with pneumoperitoneum, with potentially fatal outcome [7]. These disadvantages have led to development of alternative strategies. The abdominal wall retractor (AWR) is a new device designed to create a good view during laparoscopic surgery without the use of a pneumoperitoneum [1, 3, 5, 6, 10, 14, 16]. Recently the feasibility of using the AWR for laparoscopic surgery in pigs has been analyzed [11]. The use of the abdominal wall retractor was associated with fewer hemodynamic side effects and disturbances of gas exchange. However, the effectiveness in humans has not been analyzed. Although laparoscopic cholecystectomy has generally been performed by experienced surgeons, in our institution a pilot human study with the AWR showed that laparoscopic cholecystectomy without pneumoperitoneum was extremely difficult. It was not always possible to achieve ad-

646

equate exposure of the triangle of Calot, which is essential for safe dissection of Calots triangle. However, the addition of a low-pressure pneumoperitoneum enabled the surgeon to perform the procedure. Therefore, it was decided to add 5 mmHg pneumoperitoneum while using the AWR. The purpose of this study was to assess the safety and efficacy of the AWR in a prospective, randomized controlled clinical trial, comparing the use of AWR combined with low-pressure pneumoperitoneum with the CO2 pneumoperitoneum, with particular interest in hemodynamics and gas exchange during laparoscopic cholecystectomy. We were also interested in whether the use of an AWR posed extra technical difficulties for the surgeon. Patients and methods
Twenty patients with ASA classification 1 or 2 undergoing elective laparoscopic cholecystectomy for uncomplicated symptomatic gallstone disease gave informed consent to participate in the study. The patients were randomly allocated into one of the two groups. Patients in group 1 underwent laparoscopic cholecystectomy by abdominal wall retraction with a low-pressure pneumoperitoneum of 5 mmHg. Patients in group 2 underwent standard laparoscopic cholecystectomy with CO2 pneumoperitoneum of 15 mmHg. The study protocol was approved by the Hospital Ethical Committee.

Table 1. Patient characteristics Abdominal wall retraction + low-pressure pneumoperitoneum (5 mmHg) 8/1 2250 68 14 163 9 ns ns ns ns

Pneumoperitoneum (15 mmHg) Sex F/M Age (years) Weight (kg) Height 7/2 3052 65 11 168 7

Table 2. Operation data and complications Abdominal wall retraction + low-pressure pneumoperitoneum (5 mmHg) 1 72 16* 0

Pneumoperitoneum (15 mmHg) Conversion to open surgery Duration operation (min) Postoperative complication 1 50 18 0

* p < 0.05 compared with conventional pneumoperitoneum. tronic Laparoflator 26012, Storz, Tuttlingen, Germany). Intraabdominal pressure was controlled from the manometer on the insufflator. Pneumoperitoneum was achieved by inserting a Veress needle subumbilically. Two additional 10/11-mm and one 5-mm trocars are inserted after establishment of the pneumoperitoneum. The trocars are placed similar to the trocars used with the AWR.

Anesthesia
Premedication consisted of lorazepam 1 mg given orally approximately 1 h before induction of anesthesia. A peripheral intravenous infusion of NaCl 0.9% was administered at a rate of 6 ml kg1 h1. Electrocardiogram and pulse oximetry were continuously monitored during the procedure. Anesthesia was induced with thiopental 35 mg/kg1, followed by atracurium 0.5 mg/kg to facilitate endotracheal intubation and fentanyl 5-kg/kg. Anesthesia was maintained with isoflurane 1.15% (end-tidal concentration). Additional doses of atracurium were given to maintain one or two responses to train-of-four stimulation. During the operation additional doses of fentanyl were given when signs of insufficient analgesia were present, as indicated by a rise in pulse rate or a blood pressure greater than 20% of preinduction values. After endotracheal intubation the lungs were ventilated with a mixture of oxygen in air (FiO2 0.5). Total minute ventilation was adjusted until an end-tidal CO2 value between 30 and 40 mmHg was achieved (Dra ger, Cicero, Germany). After induction of anesthesia a 20-gauge catheter was inserted in the left radial artery for blood pressure measurements, cardiac output measurements, and blood gas sampling.

Measurements
Measurements were performed at 1, 5, 10, 15, 30, 45 and 60 min after starting the pneumoperitoneum or introducing the abdominal wall retraction. Control measurements were made 5 min after ceasing the pneumoperitoneum or the abdominal wall retraction. In the final analysis, measurements at 45 and 60 min were not included because in seven patients using conventional pneumoperitoneum the procedure was finished within 45 min. The following hemodynamic variables were measured: heart rate, blood pressure, and cardiac output; arterial blood pressure was recorded using disposable transducers (Baxter, TX). Cardiac output was computed continuously from the radial artery pressure, as described by Wesseling et al. [15]. The following ventilatory parameters were measured; end-tidal CO2 (EtCO2), arterial pH, arterial PCO2, arterial PO2, and peak inspiratory pressure. End-tidal CO2 was measured with an infrared mainstream transducer (Hewlett Packard, Saronno, Italy). Blood-gas samples were analyzed by a routine method (ABL 4, Radiometer A/S, Copenhagen, Denmark).

Surgical technique

Laparoscopic cholecystectomy with AWR. For retraction of the abdominal wall an AWR with 10-cm wings was used as described by Smith et al. [14] (Laparolift TM, Origin Med Systems, Inc. Menlo Park, CA). Low-pressure pneumoperitoneum (5 mmHg) was added to the lifting procedure in all patients. The position of the surgeon is between the legs of the patient. First a 10/11-mm trocar is inserted through the umbilicus using an open technique. Second, the fan is introduced through a right subcostal split incision under direct laparoscopic vision to prevent slipping omental fat between the legs of the fan and the abdominal wall. The fan is lifted with the abdominal wall retractor (AWR) up to a pressure of 1012 on the indicator of the fan. Two additional trocars are inserted, a 5-mm and a 10/11-mm trocar in the right lower and left abdomen, respectively. A 5-mmHg pneumoperitoneum is applied to achieve adequate exposure of Calots triangle.

Statistical analysis
Results are expressed as mean SD. Data were analyzed with two-way ANOVA for repeated measures between and in between the groups. When indicated, differences between means within the groups were analyzed using paired t-test means and unpaired t-tests for differences between means between the groups. Patient characteristics and operation time were analyzed with the Mann-Whitney U test. p values of <0.05 were considered statistically significant.

Results Patient characteristics are presented in Table 1. There were no significant differences between the study groups. Operations performed using the AWR with supplemental low-

Laparoscopic cholecystectomy with pneumoperitoneum. Abdominal insufflation with CO2 was obtained with a pressure-controlled insufflator (Elec-

647

Fig. 1. Hemodynamic parameters. Data are mean SD. *p < 0.05: () pneumoperitoneum, () abdominal wall retractor compared with baseline.

pressure pneumoperitoneum lasted significantly longer compared to conventional surgery (Table 2). In two patients (one in each group) the operation could not be completed due to technical difficulties for which the operation was converted (Table 2). Hemodynamic and ventilatory data are presented in Figs. 1 and 2. There were no significant differences between the groups with respect to baseline and changes in heart rate and systolic blood pressure. Transient increases in diastolic blood pressure during abdominal wall retraction did not result in significant differences with pressures measured during conventional pneumoperitoneum. Cardiac output remained unchanged during conventional pneumoperitoneum and showed a significant increase during abdominal wall retraction with low-pressure pneumoperitoneum, whereas at the start of insufflation a significant reduction of the cardiac output was observed with conventional pneumoperitoneum. Peak inspiratory pressures increased significantly during laparoscopy using conventional pneumoperitoneum. In contrast, peak inspiratory pressures remained unchanged during abdominal wall retraction with supplemental low-pressure pneumoperitoneum. Differences between both groups were significant in this respect. A decrease in arterial pH accompanied by an increase in CO2 occurred during both conventional and low-pressure pneumoperitoneum with abdominal wall retraction although these changes appeared earlier during conventional pneumoperitoneum. During both techniques blood gas analysis showed a decrease in partial arterial oxygen pressure, although values at which hemoglobin oxygen saturation may become impaired were not reached. Discussion The pilot study indicated that laparoscopic cholecystectomy using the AWR without pneumoperitoneum was technically difficult. The major problem during laparoscopy without pneumoperitoneum was a view obscured by bowel movement in front of the camera. Clear exposure of Calots tri-

angle, essential for safe dissection of the cystic duct, was impossible. Adding a positive intraabdominal CO2 pressure of 5 mmHg was enough to solve this problem. Although this technique was feasible as shown in the present study, the operation lasted longer when compared with the procedure using conventional pneumoperitoneum. The results also indicate that the use of the AWR combined with low-pressure pneumoperitoneum leads to similar hemodynamic and gas-exchange changes as compared with conventional pneumoperitoneum. These results are in contrast with other studies, which suggest that laparoscopic cholecystectomy can be performed with AWR without pneumoperitoneum [3, 6, 14]. This contradiction may be explained by differences in abdominal wall retraction methods, such as wiring of the subcutaneous tissues [6]. This may result in a better view as compared with the view using the abdominal wall retractor. However, these techniques involve difficult and lengthy assembly and require extra stab wounds, which makes them unpopular with most surgeons. On the other hand, some authors used the same abdominal wall retractor as in this study [3, 14]. Smith used the device without pneumoperitoneum successfully in 81% of the laparoscopic cholecystectomies. The results may be due to extensive training. The results of this study also contrast with our previous study in which the hemodynamic effects of abdominal wall retraction were assessed in pigs [11]. The V-shaped chest of a pig as compared to the more flat human chest may have enabled a clear vision in this particular experimental model. The mean duration of laparoscopic cholecystectomy using the AWR was longer as compared with conventional pneumoperitoneum. Smith et al. did not report the average operation duration [14]. It is unlikely that lack of experience negatively influenced our results. The patients were operated upon by two surgeons with extensive experience in laparoscopic surgery. Furthermore, there was no difference in operation time between the first and last procedure while using the abdominal wall retractor; there was also no difference in operation time between the two surgeons.

648

Fig. 2. Ventilatory and gas exchange parameters: () abdominal wall retractor; () pneumoperitoneum. Data are mean SD. *p < 0.05 and p < 0.001 compared with baseline. p < 0.05 compared with pneumoperitoneum.

Hemodynamic data indicate that blood pressure and heart rate are affected similarly by both methods (ANOVA). Although cardiac output increased during abdominal wall retraction compared with a small decrease during conventional pneumoperitoneum, differences in cardiac output between the two techniques were not significant. These findings are not in accordance with a previous study in pigs which indicated that laparoscopy using abdominal wall retraction results in less cardiovascular depression compared to conventional pneumoperitoneum [11]. Others also showed that in pigs, positive and expiratory pressure (PEEP) affected hemodynamics less during AWR than during conventional pneumoperitoneum [16]. The finding that adding 5 mmHg of pneumoperitoneum results in hemodynamic changes similar to those of higher intraabdominal pressures may indicate that these changes are not caused by increased intraabdominal pressure. It has been suggested that the hemodynamic changes during laparoscopy are at least partly due to the pharmacological effects of the absorbed CO2. Our findings support this contention. One may say that cardiac output was measured by a new noninvasive method, as described before [15]. Using this device, the cardiac output measurements were computed continuously from the radial artery pressure. The cardiac output changes observed during conventional pneumoperitoneum are similar to those reported in the literature using established cardiac output measurement techniques. It is possible that the computer model does not measure the absolute values of cardiac output; however, it reliably tracks relative changes of cardiac output, which is sufficient for this study. Respiratory acidosis develops using conventional pneumoperitoneum as shown by the increase of the arterial PCO2 and pH. The use of AWR was also associated with the gradual development of respiratory acidosis. This finding is in agreement with those of others who showed that the increase of PCO2 during laparoscopy is not linearly related to the intraabdominal pressure of CO2 pneumoperitoneum [8]. It is suggested that recruitment of peritoneal absorption area is an important factor to determine the rate of CO2 absorption from the peritoneal cavity [8, 9]. It is conceivable that recruitment of more gas-exchange area during abdominal wall retraction may result in an increase of PCO2 similar to the increase observed at higher intraabdominal pressures. Alveolar dead space ventilation is also an important contributor to respiratory acidosis during laparoscopy [8]. The alveolar dead space was not measured in this study. However, it seems possible that alveolar dead space ventilation increases to the same extent during both methods. The decrease in pH during conventional pneumoperitoneum seems larger, compared with the use of the abdominal wall retractor, although statistical significance was not reached. It is possible that with a longer operation time this difference might reach statistical significance. However, the majority of the laparoscopic cholecystectomies using conventional pneumoperitoneum lasted on average 50 min. Consequently we were unable to complete a full set of hemodynamic and gas-exchange values after 30 min in all cases. It seems reasonable to assume that the abdominal wall retractor might be of value with respect to acidbase equilibrium during operations of longer duration, such as bowel surgery.

649

Arterial oxygenation shows a gradual reduction of the arterial PO2 during the laparoscopic procedures, without significant differences between both methods. Increased intrapulmonary shunt or decreased ventilation perfusion ratio as a result of atelectases secondary to the cranial movement of the diaphragm, may be the underlying mechanism of this phenomenon. There is only one case report that describes severe hypoxemia in a patient with sickle cell anaemia undergoing laparoscopic cholecystectomy [4]. Because arterial PO2 values remained much higher than values at which hemoglobin oxygen desaturation occurs, it remains doubtful whether the decrease in arterial PO2 is of any clinical relevance. During laparoscopy, using conventional pneumoperitoneum the cranial shift of the diaphragm is associated with diminished intrathoracic volume. When mechanical ventilation with fixed tidal volumes is applied, increased airway pressures will be generated. Indeed, in our study peak airway pressures increased during laparoscopy using conventional pneumoperitoneum. This finding is in agreement with those of others who reported increased peak and plateau airway pressures secondary to reduced compliance during laparoscopy [2]. In contrast, during abdominal wall retraction peak airway pressures did not increase. High airway pressures may have damaging effects on lungs with bullous emphysema that are prone to the development of barotrauma [13]. In theory, the use of the abdominal wall retractor might be advantageous in these patients because its use is not associated with increased airway pressures. Conclusion In conclusion, the results of this study are disappointing with respect to the use of the abdominal wall retractor. In the first place, the use of this device is difficult and does not permit laparoscopic cholecystectomy entirely without pneumoperitoneum. This takes away some of the suggested advantages of using the abdominal wall retractor, such as low costs [6], and the possibility of using conventional instruments [14]. Second, the beneficial effects suggested by several experimental studies with respect to hemodynamics and gas exchange could not be confirmed in this human study. Considering the results, AWR should not be used during laparoscopic cholecystectomy. AWR might be valuable for lower abdominal surgery, although this has to be evaluated by further study.

References
1. Banting S, Shimi S, Vander VG, Cuschieri A (1993) Abdominal wall lift. Low-pressure pneumoperitoneum laparoscopic surgery. Surg Endosc 7: 5759 2. Bardoczky GI, Engelman E, Levarlet M, Simon P (1993) Ventilatory effects of pneumoperitoneum monitored with continuous spirometry. Anaesthesia 48: 309311 3. Chin AK, Eaton J, Tsoi EK, Smith RS, Fry WR, Henderson VJ, McColl MB, Moll FH, Organ CJ (1994) Gasless laparoscopy using a planar lifting technique. J Am Coll Surg 178: 401403 4. Cunningham AJ, Schlanger M (1992) Intraoperative hypoxemia complicating laparoscopic cholecystectomy in a patient with sickle hemoglobinopathy. Anesth Analg 75: 838843 5. Edelman DS (1994) Alternative laparoscopic technique for cholecystectomy during pregnancy. Surg Endosc 8: 794796 6. Hashimoto D, Nayeem SA, Kajiwara S, Hoshino T (1993) Abdominal wall lifting with subcutaneous wiring: an experience of 50 cases of laparoscopic cholecystectomy without pneumoperitoneum. Surg Today 23: 786790 7. Lantz PE, Smith JD (1994) Fatal carbon dioxide embolism complicating attempted laparoscopic cholecystectomycase report and literature review [review]. J Forensic Sci 39: 14681480 8. Lister DR, Rudston-Brown B, Wariner B, Mc Ewen J, Chan M, Walley KR (1994) Carbon dioxide absorption is not linearly related to intraperitoneal carbon dioxide insufflation pressure in pigs. Anesthesiology 80: 129136 9. Mullet CE, Viale JP, Sagnard PE, Miellet CC, Ruynat LG, Counioux HC, Motin JP, Boulez JP (1993) Pulmonary CO2 elimination during urgical procedures using intra- or extraperitoneal CO2 insufflation. Anesth Analg 76: 622626 10. Newman LL, 3d, Luke JP, Ruben DM, Eubanks S (1993) Laparoscopic herniorrhaphy without pneumoperitoneum. Surg Laparosc Endosc 3: 213215 11. Rademaker BMP, Meyer DW, Bannenberg JJG, Klopper PJ, Kalkman CJ (1995) Laparoscopy without pneumoperitoneum. Effects of abdominal wall retraction versus carbon dioxide insufflation on hemodynamics and gas exchange in pigs. Surg Endosc 1995;9:197201 12. Safran DB, Orlando R (1994) Physiologic effects of pneumoperitoneum [review]. Am J Surg 167: 281286 13. Slutsky AS (1993) Mechanical ventilation. Chest 104: 18331859 14. Smith RS, Fry WR, Tsoi EK, Henderson VJ, Hirvela ER, Koehler RH, Brams DM, Morabito DJ, Peskin GW (1993) Gasless laparoscopy and conventional instruments. The next phase of minimally invasive surgery. Arch Surg 128: 11021107 15. Wesseling KH, Jansen JRC, Settels JJ, Schreuder J (1993) Computation of aortic flow from pressure in humans using a nonlinear, three element model. J Appl Physiol 74: 25662573 16. Woolley DS, Puglisi RN, Bilgrami S, Quinn JV, Slotman GJ (1995) Comparison of the hemodynamic effects of gasless abdominal distention and CO2 pneumoperitoneum during incremental positive endexpiratory pressure. J Surg Res 58: 7580

Editorial
Surg Endosc (1997) 11: 613614

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Complete and partial laparoscopic fundoplication for gastroesophageal reflux disease

Fundoplication for gastroesophageal reflux disease (GERD), first reported in 1990, represents a significant leap forward in the development of effective therapy in this common condition. The paper by Dr. Coster and colleagues in this edition of Surgical Endoscopy makes a strong case for laparoscopic antireflux surgery early in the course of GERD, and the authors support their thesis by reporting the short-term results of 231 patients following complete or partial fundoplication. Their paper demonstrates that experienced laparoscopic surgeons in a private practice setting using two co-surgeons can perform fundoplication rapidly, safely, economically, and with good short-term outcome. Several features of the report are likely to give rise to controversy, and it is worthwhile to assess their policy and results against the background of other workers in the same specialty. These may conveniently be grouped under three headings: patient selection, operative strategy, and outcome assessment. There is no argument that reduction of unnecessary preoperative investigations is desirable to limit both costs and patient discomfort. These authors noted a change in the referral pattern as the reputation of the operation spread, and thus they were referred patients from different sources with different degrees of workup. This phenomenon is likely to become more widespread and it should encourage surgeons to become experts in understanding the disease process and to be able to take charge of the preoperative workup. Surgeons who do their own endoscopy, as Dr. Coster and his colleagues do, have a different perspective than internists, since the endoscopic view is automatically related to the anatomy the surgeon sees every day in the operating room. The anatomic location and reducibility of the gastroesophageal junction and the function and position of the crura all impact the surgical decision-making process. What about the necessity of physiologic investigation of the esophagusmanometry and 24-h pH monitoring? Are these uncomfortable tests really necessary for patient care, or are they simply icing on the cake, niceties devised by academic surgeons to confirm in numeric form what everyone knows anyway? The development of 24-h esophageal pH monitoring has certainly advanced our understanding of the pathophysiology of GERD, but is it necessary to confirm excessive esophageal acid exposure in every patient before recommending surgery? Dr. Coster and colleagues seem to suggest that the presence of endoscopic esophagitis is sufficient. We disagree. Nonreflux causes of esophagitis, especially pill-induced, may be present in up to 10% of patients, and the visual and histologic characteristics do not distinguish these causes. However, when the patient has a combination of a significant hiatal hernia typical symptoms, and endoscopic esophagitis, the proportion of patients with negative 24-h pH monitoring is very small, and the test may arguably be omitted in this circumstance. The problem is that a rising indication for the procedure is drug dependency in patients reported to have had esophagitis in the past, but who are now healed. In such patients we believe that confirmation of the diagnosis prior to surgical intervention with 24-h pH monitoring is a necessity. Preoperative appreciation of esophageal contractility is an important factor in deciding on the operative approach. Dr. Costers group believes that preoperative motility tests may be limited to patients with suggestive symptoms, though there is abundant evidence that this is not so [1]. Although it is unlikely to be subjected to a randomized controlled trial, the detection of motility abnormalities on preoperative manometry is widely regarded as important in choosing a partial fundoplication rather than a total one to avoid the problem with postoperative dysphagia. The basis for this choice comes from the disastrous results which follow fundoplication in unrecognized achalasia and other hypoperistaltic disorders. Further, an unpublished study carried out by one of us (T.R.D.) demonstrated in an in vitro model using the excised dog esophagus that a total fundoplication could support a column of water of 1520 cm but a partial fundoplication exerted virtually no outflow resistance. The authors of the present study bypass the problem by advocating partial fundoplication as a routine. The key issue in the complete vs partial fundoplication controversy is durability. Neither this present paper nor any other from the laparoscopic era can yet answer the question because the necessary follow-up is not available. But years of careful follow-up of the most carefully studied partial fundoplication, the Belsey Mark IV operation, indicates steady deterioration with time, in contrast to the Nissen procedure, where accumulated experience confirms long-term durability. We would counsel caution in adopting partial fundoplication as the primary surgical treatment for reflux disease. If these authors wish to make a further valuable contribution, they will carefully follow up their patients over the next 10

Correspondence to: T. R. DeMeester

614

years. This is painstaking and labor-intensive work, but such an important lesson can be obtained in no other way. Until then, their confident assertion that there is no circumstance where a complete wrap would be superior to a partial one except in the case of an intraoperative distal perforation is premature. This same caveat should also be applied to the authors concept of esophageal shortening. Positive-pressure insufflation leads to diaphragmatic elevation and makes it possible to bring below the diaphragm an esophagus which at open surgery would remain retracted in the mediastinum. If true shortening is present, the fundoplication will be constructed under tension and there will a tendency for future migration into the chest and/or breakdown. Once again, careful long-term follow-up of such patients will be needed to demonstrate the durability of the fundoplication constructed under those circumstances. Many of the feared complications of the Nissen procedure stem from older studies in which a long or tight fundoplication was performed, and they have largely ceased to be a significant problem in contemporary practice [5]. In addition, careful symptomatic review has shown that even with medical treatment patients frequently experience symptoms such as flatulence and bloating [3]. Several operative details will be of immediate interest to surgeons tackling laparoscopic antireflux operations. The authors are correct to emphasize the importance of the retro fundic dissection in mobilizing the fundus and cardia from the left crus of the diaphragm. This step is much more deliberate than was common in open fundoplication, and the use of the 45 lens adds significantly to the ease and safety of the periesophageal dissection. The use of the USSC automatic suture placement device is certainly time saving, but it may not allow the precision by which experienced surgeons feel the depth of the bites taken of crura, esophagus, or stomach. Does this paper add to the controversy about the necessity to take down the short gastric arteries prior to constructing the fundoplication? Two important recent studies given conflicting messages in this connection. The Adelaide group recently reported the results of a small randomized trial comparing patients with short gastric division vs those without. No different in symptomatic outcome occurred [4]. On the other hand, Hunter et al. recently published a comparative study in which the incidence of postoperative dysphagia was higher in the Nissen-Rosetti patients than in either those with Nissen procedures where the short gastrics were divided, or in those with Toupet procedures [2]. Further, Hunter et al. also demonstrated the torsion effect produced by the Nissen-Rosetti procedure.

The current study seems rather to support Hunter, in that their Nissen-Rosetti patients had a higher incidence of dysphagia and a longer time to resumption of an unrestricted diet than the Toupet patients. Rather than resort to an operation of uncertain durability such as the Toupet, why not compare the Nissen-Rosetti with the true Nissen as currently performed, where the fundus is mobilized and the fundus is plicated without torsion around the lower esophagus? A properly done laparoscopic Nissen fundoplication gives good symptomatic results with a low incidence of dysphagia and other unpleasant side effects, and since it anatomically corresponds to the open Nissen, we can have the same confidence in its long-term outcome. In conclusion, although the authors have given us a good description of two antireflux procedures which they can perform safely and with extraordinary rapidity, we continue to believe that a correctly constructed, nontwisted, 360 Nissen fundoplication in correctly selected patients is the best long-term solution for most patients with GERD. We now accept that GERD requires life-long medical therapy, and if the patient opts for surgical therapy, it, too, must last a lifetime! References
1. Costantini M, Crookes PF, Bremner RB, Hoeft SF, Ehsan A, Peters JH, Bremner CG, DeMeester TR (1993) The value of physiological assessment of foregut symptoms in a surgical practice. Surgery 114: 780787 2. Hunter JG, Swanstrom L, Waring JP (1996) Dysphagia after laparoscopic antireflux surgery: the impact of operative technique. Ann Surg 224: 5157 3. Spechler SJ, the VA Gastroesophageal Reflux Study Group (1992) A prospective trial of medical and surgical therapies for gastroesophageal reflux disease. N Engl J Med 326: 786792 4. Watson DI, Pike GK, Mathew K, Bairgie RJ, Devitt PG, Britten Jones R, Jamieson GG (1996) Prospective double blind randomized trial of laparoscopic Nissen fundoplication with division and without division of the short gastric vessels. In: Peracchia A (ed) Recent advance in diseases of the esophagus. Monduzzi, Milan 5. Woodward ER, Thomas HF, McAlhany JC (1971) Comparison of crural repair and Nissen fundoplication in the treatment of esophageal hiatus hernia with peptic esophagitis. Ann Surg 173: 782792

P. F. Crookes T. R. DeMeester
University of Southern California School of Medicine Department of Surgery 1510 San Pablo Street Suite 514 Los Angeles, CA 90033-4612 USA

Surg Endosc (1997) 11: 655657

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Complications of pediatric laparoscopic surgery

C. Esposito, G. Ascione, V. Garipoli, G. De Bernardo, G. Esposito
Department of Pediatrics, Division of Pediatric Surgery, Federico II University of Naples, Via Pansini 5, 80131 Napoli, Italy Received: 5 July 1996/Accepted: 19 November 1996

Abstract Background: Surgical complications of laparoscopy most often occur during Veress needle or primary trocar placement. Veress needle punctures are insignificant and require no further treatment, whereas trocar-induced vascular injuries can be catastrophic. The frequency of vascular and viscus injuries is difficult to calculate because several complications are not reported in the literature. Methods: During a 10-year-period (19841995), at the Division of Pediatric Surgery at Federico II University of Naples, 430 laparoscopic procedures were performed in 395 children with a mean age of 5 years. The incidence of complications related to laparoscopy was 1.8% with eight complications, one of which was rather severe. The complications included one abdominal wall hematoma, two perforations of abdominal viscus (stomach, ovary), one umbilical scar complication, one postoperative hydrocele, one subcutaneous emphysema, and one pneumothorax during a Nissen procedure. The only severe complication occurred in a young girl with neurologic problems and a kyphoscoliosis operated on via laparoscopy for a gastroesophageal reflux. She suffered injuries of both right common iliac vessels and several intestinal perforations due to blind introduction of the first umbilical trocar. Results: In this case rapid conversion, complex vascular reconstruction, and multiple intestinal sutures were performed. The Nissen fundoplication with pyloroplasty was performed traditionally and the patient left the hospital free of symptoms after 20 days. The other seven complications were resolved without any problem intra- or postoperatively. Conclusions: The authors believe that the open approach with a blunt trocar is most important in helping to avoid complications in pediatric laparoscopy. Key words: Laparoscopy Complications Open approach

The study of laparoscopic complications raises the problem of how to evaluate this new surgical technique, which has been developing in pediatric surgery only in the last 56 years. Although there are reports in the literature on the risks related to diagnostic laparoscopy that date back to 1950, and although important studies have been published on both incidence and severity of complications deriving from the use of laparoscopy in adults for digestive or gynecological surgery, the data on pediatric ages are scanty [3, 5, 9]. Peters in his case series of 5,400 laparoscopic interventions reports a complication rate of 5.38%, with a Veress needle complication rate of 2.6% compared to the open complication rate of 1.2% [15]. The aim of this study is to examine, on the basis of the experiences of these authors, the real incidence of laparoscopic complications in children and how to prevent them.

Patients and methods

At the Division of Pediatric Surgery of the Federico II University of Naples 430 laparoscopic investigations were performed over a period of 10 years extending from November 1984 to June 1995. The patients ages ranged from 8 days to 16 years, with a mean age of 5 years. There was a 1.8% complication rate, equal to eight complications, one of which was rather severe (Table 1). In five cases, the complications occurred intraoperatively, whereas three were postoperative complications. The complications were: two cases of endoabdominal organ perforation with the Veress needle (a large ovarian cyst and a case of extremely dilated stomach), a wound infection involving the umbilical scar, a small hematoma in the abdominal wall, a moderate-degree subcutaneous emphysema that was resolved intraoperatively, a left hydrocele that manifested itself a few months after surgery for varicocele, and a pneumothorax during a Nissens procedure that was resolved with a simple puncture at the end of intervention. Also the left hydrocele, consequent to varicocelectomy, was treated successfully by simple puncture. The gastric perforation was treated conservatively according to Taylors technique, with a nasogastric intubation for 4 postoperative days. The only severe complication occurred in a 5-year-old girl suffering from gastroesophageal reflux who was also affected by a severe neurological disorder and kyphoscoliosis. The patient underwent laparoscopy for antireflux fundoplication according to Nissen. After inducing a pneumoperitoneum with the Veress needle and introducing a 10-mm disposable trocar in the umbilical scar, a severe hemoretroperitoneum appeared evident through the optics of the laparoscope. An immediate conversion revealed lesions of the right common iliac artery, lesions of the homologous vein, and the presence of multiple intestinal perforations. After reimplant-

Correspondence to: C. Esposito

656 Table 1. Laparoscopic complications in our experience 1.8% (8 cases) 2 1 1 1 1 1 1 Endoabdominal organ perforations Hydrocele Pneumothorax Subcutaneous emphysema Small hematoma of the abdominal wall Vascular injury Umbilical scar infection

ing the iliac artery on the aorta and suturing the lesion on the homologous vein and intestinal perforations, the antireflux gastroplasty, according to Nissens technique, was completed using conventional surgery. The child was released from hospital 20 days later without any gastroesophageal reflux or consequences related to the laparoscopic intervention. Since this episode, we have begun to utilize the open laparoscopy in every case, to avoid complications related to the closed introduction of the first trocar.

Discussion Surgical complications due to laparoscopy can be classified according to cause or in relation to the moment when they occur [4, 6]. Based on the cause, we can either speak of nonspecific or specific complications: the former are related to procedures that are common to all laparoscopies independently of the indication, such as the creation of a pneumoperitoneum; the latter, instead, depend on specific interventions done by the surgeon. Moreover, complications can either be intrasurgical or postsurgical, depending on whether they occur during or after the intervention. While we know that the incidence of complications in adult laparoscopy is about 4%, with a 0.03% mortality rate, in pediatric laparoscopy no such data are available in the literature [11, 12]. Surgical complications are generally related to the introduction of either the Veress needle or the first trocar. Veress needle punctures are generally safe and require no further treatment, whereas lesions caused by a trocar can have catastrophic consequences [1, 4, 10]. Bleeding caused by lesions of the abdominal wall vessels generally clot spontaneously; a ligation may be necessary only if the lesion involves the epigastric vessels. Lesions of the endoabdominal organs are generally eventless if they are promptly identified and treated, whereas endo- and retroperitoneal vessel lesions generally require immediate conversion. The endoabdominal vessels most frequently involved are the aorta, the inferior vena cava, and the iliac vessels, due to the fact that they are located right below the umbilicus, only 23 cm away [13, 14, 16]. In case of a vessel lesion, the laparoscopic surgeon must be able to perform an emergency conversion, doing a temporary hemostasis first and a definite hemostasis after having a global balance of the lesion. The solution to vessel complications can be more or less simple; in more severe cases it is possible to use of venous grafts or prosthesis. Vessel lesions treated immediately have a good prognosis; a delay in treatment can entail severe complications [7]. There are also complications related to insufflation: Even if performed correctly, a pneumoperitoneum can later on expand itself beyond the endoabdominal cavity. The subcutaneous or scrotal diffusion is trivial and uneventful, whereas thoracic involvement, causing either a pneumotho-

rax or a pneumomediastinum, like our case of pneumothorax, occurred in the case of Nissens fundoplication. Generally a simple puncture is sufficient to treat the pneumothorax, which does not occur again after deflation of the abdominal cavity [3]. The so-called specific complications are related to the type of intervention to be performed: They can be due to dissection or coagulation maneuvers, to the extraction of organs or structures that have been dissected, or to neoplastic or infectious contamination [12]. No such complications occurred in our series, mainly due to our use of bipolar coagulation to induce hemostasis, which seems safer and less dangerous compared to monopolar coagulation [6, 8]. As to the removal of dissected organs, the main problem remains related to the size of the organ, as it might be necessary to enlarge the trocar opening or to create minilaparotomies to extract the organ. In the case of a splenectomy, it is important to remember that an intraperitoneal spleen rupture can cause splenosis [8]. Postsurgical complications are theoretically reduced in laparoscopy; however, the most common ones are infections, small hematomas, epiploic evisceration or, even rarer, occlusions due to adhesions. In our series we have two examples of this kind of complication, an umbilical scar infection and a small hematoma in the abdominal wall, but they were resolved without any problem in the first postoperative days, respectively, by administration of antibiotics for 3 days in the first case and a local compressive medication in the second one. Another complication that we noticed in the postoperative period was a hydrocele in a patient operated on with an Ivanissevich procedure for a left varicocele. This problem depends, as in traditional surgery, on clipping together the spermatic veins and the small lymphatic vessels also present in the spermatic cord. This complication could be resolved, as in our case, with a simple puncture of the hydrocele. It is important to take into account all contraindications to laparoscopic surgery. There are, basically, clotting problems, hypovolemic shock, large tumors, and advanced intestinal occlusion; moreover, on the basis of our experience, all vertebral abnormalities in which there is reduced space between the abdominal wall and the retroperitoneal vessels also constitute a contraindication. As several authors have reported, the incidence of complications due to the introduction of either the Veress needle or the first trocar does not seem to be related to the surgeons expertise but rather to the number of procedures performed. This type of problem has been recently resolved by the open laparoscopy technique, where the introduction or the first trocar is eye-guided [2, 9]. The other complications can be certainly avoided by the surgeons and the teams aptness, together with a proper compliance to the indications for surgery. We believe that in pediatric laparoscopy as in all the other fields of surgery, proper knowledge and careful attention to all the details are the most important factors in avoiding complications. References
1. Apelgren KN, Scheeres DE (1994) Aorta injury, a catastrophic complication of laparoscopic cholecystectomy. Surg Endosc 8: 689691

657 2. Begin FG (1993) Cre ation du pneumope ritoine sous contro le visuel. J Coeliochir 5: 1820 3. Berci G (1994) Complications of laparoscopic surgery. Surg Endosc 8: 165168 4. Bloom DA, Erhlich RM (1993) Omental evisceration through small laparoscopy port sites. J Endourol 7: 3132 5. Capelouto CC, Kavoussi LR (1993) Complications of laparoscopic surgery. Urology 42: 212 6. Delarue A, Guys JM, LouisBorrione C, Simeoni J, Esposito C (1994) Pediatric endoscopic surgery: pride and prejudice. Eur J Pediatr Surg 4: 323326 7. Duckett JW (1994) Editorial: pediatric laparoscopy: prudence please. J Urol 151: 742743 8. El Ghoneimi A, Valla JS, Limonne B, Montupet P, Chavrier Y, Grinda A (1994) Laparoscopic appendectomy in children: report of 1379 cases. J Pediatr Surg 29: 786789 9. Hasson H (1971) Modified instrument and method for laparoscopy. Am J Obstet Gynecol 110: 886887 10. Juricic M, Bossavy JP, Izard I, Cuq P, Vaysse P, Juskiewensky S (1994) Laparoscopy appendectomy: case report of vascular injury in two children. Eur J Pediatr Surg 4: 327328 11. Niebuhl H, Nahrestedt U, Ruckert K, Hollmann S (1993) Laparoscopic surgery: mistakes and risks when the method is introduced. Surg Endosc 7: 412415 12. Nord HJ (1992) Complications of laparoscopy. Endoscopy 24: 693 700 13. Oshinsky G, Smith AD (1992) Laparoscopic needle and trocar: an overview of design and complications. J Laparoendosc Surg 2: 117 125 14. Oza KN, ODonnell N, Fisher JB (1992) Aortic laceration: a rare complication of laparoscopy. J Laparoendosc Surg 2: 235237 15. Peters CA (1995) Complications in pediatric urological laparoscopy: results of a survey. J Urol 155: 10701073 16. Waldschmidt J, Schier F (1991) Laparoscopical surgery in neonates and infants. Eur J Pediatr Surg 1: 145150

Surg Endosc (1997) 11: 663667

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Long-term effects of repeated injection sclerotherapy on esophageal motility and mucosa

B. Isaksson, F. Johnsson, B. Jeppsson
Department of Surgery, Lund University Hospital, S-221 85 Lund, Sweden Received: 22 April 1996/Accepted: 20 August 1996

Abstract Background: Endoscopic sclerotherapy (ST), widely used as treatment of bleeding esophageal varices, might cause motility disturbances of the esophagus as well as mucosal damage. We performed this study to evaluate the long-term effects of repeated sclerotherapy on esophageal motility and mucosa. Methods: Ten patients with liver cirrhosis and bleeding esophageal varices treated with repeated ST were evaluated after the last ST, median 52 months, by esophageal manometry and gastroscopy where forceps biopsies were taken. Results: We found a significant difference in the distal esophageal sphincter intraabdominal length. The distal esophageal sphincter pressure was somewhat lower in the ST group although the difference did not reach statistical significance. There was infiltration of neutrophil leukocytes in biopsies from four patients and normal findings in the rest. Conclusions: Long-term follow-up evaluation showed statistically longer distal esophageal intraabdominal length in the ST group. No mucosal alterations were found at the histopathological investigation. Key words: Sclerotherapy Esophageal manometry Histopathological evaluation

may require dilatation. Besides this, there may also be motility disturbances of the esophagus [4, 7, 10, 14]. The incidence and severity of these disturbances vary in different reports. We therefore designed this study in order to evaluate the long-term effects of repeated sclerotherapy on esophageal motility and relate them to the number of sclerotherapy sessions and volume of injected sclerosants. We also tried to evaluate any histological alterations in the mucosa and submucosa in the same group of patients. Materials and methods Patients
Ten patients, seven men and three women, from 50 to 77 years of age, median age 6112 years, who had been treated with repeated endoscopic sclerotherapy of esophageal varices were compared to a control group of ten healthy persons 38 to 78 years old with a median age of 42 years. In the control group there were six men and four women. Initially the patients in the ST group all had bleeding esophageal varices. The varices were due to portal hypertension secondary to liver cirrhosis, in most cases induced by alcohol abuse. Two patients represented Childs class A and the other eight Childs class B.

Methods Endoscopic injection sclerotherapy (ST) is a wellestablished method in the treatment of bleeding esophageal varices. It has proven to be effective in the treatment of both acute bleeding episodes as well as electively for eradication of varices [1, 6, 15, 16]. The most common complaints after injection sclerotherapy are fever, retrosternal discomfort, and transient dysphagia; they resolve within 2448 h. Esophageal ulcerations are common and usually asymptomatic. They are more likely to occur after large volume and frequent injections of sclerosants. Esophageal strictures
Correspondence to: B. Isaksson

Sclerotherapy. The sclerotherapy treatment was administered by a flexible standard gastroscope, Olympus GIF-K, with an injection needle catheter through which 10 mg/ml polidocanol (Aethoxysclerol) (Chemische Fabrik Kreussler & Co, GmbH, D-6200, Wiesbaden-Briebrick, Germany) was given in bolus injections of 0.51 ml submucosally and paravariceally. The injections were given at the gastroesophageal junction and 57 cm proximally. The maximal amount given at each treatment session was 30 ml polidocanol. The first sclerotherapy session was performed acutely due to bleeding esophageal varices and the following were elective in order to eradicate the varices.

Esophageal manometry. This investigation was performed using a stationary pull-through technique. A three-lumen polyvinyl catheter, with the

664 Table 1. Results of esophageal manometry DESP (mmHg) ST Control group Median Minimum Maximum Median Minimum Maximum 11.35 7.00 16.30 14.70 11.70 18.00 n.s. DESTL (mm) 41.50 27.00 57.00 43.00 23.00 50.00 n.s. DESIAL (mm) 28.50 10.00 40.00 18.50 10.00 33.00 p < 0.03321 AMPLDISES (mmHg) 21.75group 0 30.00 22.50 12.00 36.00 n.s.

distal side holes at 5-cm intervals and differing 120 in radial orientation, was introduced nasally and positioned with all the side holes in the stomach. Each lumen was connected to an Arndorfer hydrocapillary pump and perfused with water at a rate of 0.5 ml/min. With the patient in supine position the catheter was pulled back in 1-cm increments. The patients performed dry swallows at each centimeter. From the manometry tracing we calculated the pressure, total length, and abdominal length of the lower esophageal sphincter (LES). The pressure in the LES was measured endexpiratory at the respiratory inversion point relative to intragastric pressure. The total length of the LES was defined as the total length of the highpressure zone at the gastro-esophageal junction. The abdominal length of the LES was defined as the length below the respiratory inversion point. We also calculated the amplitude of the esophageal contraction wave at dry swallows in the distal 5 cm of the esophagus. The mean value from the five swallows and the three channels was used. pH measurements were not performed.

the esophageal varices were eradicated; three patients had varices of grade I and one patient had varices of grade II. No esophagitis or stenosis of the esophagus were found; neither could any varices of the fundus be seen in any patient. There were no clinical signs of esophagitis.

Age The median age of the ST group was higher, 61.5 years, compared to the control group, where the median age was 42. This difference is statistically significant, p 0.031.

Esophageal manometry
Biopsies. After the manometry the patients were investigated by gastroscopy and 12 forceps biopsies (Olympus FB-25 K) were taken at three levels from the gastroesophageal junction at 0, 2, and 4 cm above for histopathological examination.

Histopathological evaluation. The biopsies were treated and analyzed according to clinical routines. After the first examination, which was made by different pathologists, all specimens were reexamined by one consultant pathologist.

Distal esophageal sphincter pressure (DESP). The ST group showed a lower value than the control group although it was not statistically significant (one-sided p < 0.03156 two-sided p < 0.06312). The median value for the ST group was 11.35 mmHg compared to 14.70 for the control group (Table 1). Distal esophageal sphincter total length (DESTL). No significant difference between the groups could be observed (Table 1).

Follow-up. The esophageal manometry was performed around 4 years after the last ST session. Gastroscopy with forceps biopsies was performed after manometry at the same visit.

Statistical analysis. The data were stored and calculated using a WinSTAT statistical program. A Mann-Whitney test and Fishers exact test were used to discriminate differences between the groups. A probability of more than 95% (p < 0.05) for rejection of the null hypothesis was preset to determine statistical significance.

Distal esophageal sphincter intraabdominal length (DESIAL). We found a significant difference in that the sclerotherapy group showed a significantly longer intraabdominal length of the sphincter compared to the control group (27.6 mm vs 18.5 mm), p < 0.03321 (Table 1).

Results Amplitude of the distal esophageal sphincter (AMPLDISES). No statistically significant differences could be seen between the groups (Table 1). Peristalsis of the corpus esophagi. The esophageal peristalsis was normal in seven patients in the ST group. Three patients had dysmotility in the distal esophagus. Of these, one patient had aperistalsis and occasionally retroperistaltic contraction waves and two patients had abnormal configuration at the peristaltic waves. All patients in the control group had normal peristalsis of the esophageal body. Cal-

Sclerotherapy The patients in the ST group had a median number of ST sessions of four (minimum two and maximum 11). The total amount of sclerosants given was median 78 ml (minimum 38 and maximum 163 ml). The median time between the last sclerotherapy treatment and the esophageal manometry and the follow-up gastroscopy was 52.5 months with a range from 4 to 113 months. The follow-up gastroscopy showed that in six patients

665

Fig. 1. Sclerotherapy group. Slight chronic inflammation with lymphocytes in the lamina propria. Hematoxylineosin (400).

Fig. 2. Sclerotherapy group. Junctional zone cardiaesophagus. Heavy lymphocytic infiltration. Hematoxylineosin (250).

culated by Fishers exact test no significant difference could be noted between the groups.

Histopathological findings Signs of inflammation with infiltration of neutrophil leukocytes as in esophagitis were seen in the biopsies from four patients; the other six patients had all-normal squamous epithelium. In no specimen were the biopsies deep enough to show any material from the muscle layer. This made it impossible to judge whether there was any fibrous reaction in the stroma (Figs. 13).

Discussion Although first reported in 1939 by Crafoord and Freckner [2], who used a rigid endoscope for injection treatment of bleeding esophageal varices, endoscopic injection sclerotherapy of esophageal varices did not gain popularity until the sixties when the flexible endoscopes became more commonly available. There are two ways to administer the sclerosing agent. In Europe, except in Great Britain, paravariceal injection sclerotherapy is the most widely used technique. The most common technique in the US and in the UK is the intravariceal. The aim of the paravariceal or the submucosal

technique is to thicken the overlying mucosa; the intravariceal approach is aimed to induce variceal thrombosis [13]. In our study we used the paravariceal submucosal technique of injecting the sclerosing agent, 1% polidocanol, in bolus injections of 0.51 ml at a maximum of 30 ml at every treatment session. The median total amount of sclerosing agent was 78 ml, the median number of treatment sessions was four. No patient in our study had any clinical signs of esophagitis. This is in agreement with Sauerbruch and colleagues [9] who used 1% polidocanol in their treatment group. They did not record the administration rank but the mean number of ST sessions was ten and the mean amount of sclerosant agent was 133 ml. The length from ST to study was a mean of 21 months. Compared to their control group, consisting of cirrhotic patients with untreated varices, they found no statistically significant differences in esophageal pH and mean of reflux episodes. The long-term complications of sclerotherapy treatment may include disturbances of normal function of the esophagus. Both decreased distal esophageal sphincter pressure and abnormal peristaltic waves in the esophagus have been reported [4, 8, 10, 11]. We found in our study a lowered distal esophageal sphincter pressure in the ST group compared to our control group although the difference did not reach a statistically significant level. The median value for the treatment group was 11.35 mmHg compared to 14.70 mmHg for the control group. Other investigators report significant reduction of the

666

Fig. 3. Sclerotherapy group. Junctional zone. Neutrophil leucocytes in the epithelium and some lymphocytes and plasma cells in the lamina propria. Hematoxylineosin (225).

lower esophageal sphincter pressure after sclerotherapy. Ogle et al. [7] found that their sclerotherapy group had a maximum sphincter pressure of 12.7 mmHg. The value for their cirrhotic patients untreated varices was 17.6 mmHg and 20.9 mmHg for their patients with cirrhosis without varices. In this study the patients were treated with intravariceal injections of ethyl amine. They were all studied much earlier after ST than our patients. Uribarrena and colleagues [14] report (in their study where they treated the varices by intravariceal injections of 5% ethanolamine oleate) a LESP (DESP) of 17.52 mmHg for the sclerotherapy group, 20.26 mmHg for their group of cirrhotic patients with untreated varices and 22.86 mmHg for their control group of healthy persons. Their patients were studied on the average 12.3 months after ST. Sauerbruch et al. [8] also report lower tonus of the LES in their sclerotherapy group, 10.7 mmHg vs 13.4 mmHg for their healthy volunteers. In this study both intravariceal and paravariceal injections of polidocanol 1% were given. Monometry was performed mean 9.6 months after bleeding. Other investigators do not find any significantly lowered pressures of the distal esophageal sphincter. So derlund and Thor [12] used the intravariceal technique and injected 1% polidocanol and studied the patients five months after ST. The same conclusions about the unaltered LES pressure were drawn by Grande and colleagues [4] and by Larson et al. [5]. The group of Grande used ethyl amine injected into the varices and the group headed by Larson used 5% sodium

morrhuate injected intravariceally. These studies were performed after 1 month and 6 months, respectively, but the total volume of sclerosant agent and number of ST sessions were not recorded. The lowered distal esophagus sphincter pressure could be a result of the known fibrous reaction [3, 10] after sclerotherapy. In our study we found that the intraabdominal length of the distal sphincter was significantly longer in the ST group compared to the control group. However, there was no significant difference in the total length of the distal esophageal sphincter, which in our study had a median length for the sclerotherapy group of 41.7 mm and 43.0 for the control group. Ogle et al. [7] report an increased LES length. They found that the length of the LES in cirrhotic patients with varices treated by sclerotherapy was 47.2 mm; in patients with cirrhosis and untreated varices the length was 42.8 mm; and in the patient group with cirrhosis without varices the LES length was 33.5 mm. It is not likely that the length can be prolonged after sclerotherapy. This finding might represent an artefact in that the actual length is not altered but seems so measured by manometry due to the known fibrous reaction, which might create a stiffness of the esophageal wall [3, 10]. All these studies were performed within the first months after sclerotherapy. According to our findings, these alterations remain permanent. Regarding the peristalsis of the corpus esophagi, we found three patients with dysmotility in the distal esophagus. The patient with aperistalsis and occasionally retroperistaltic contractive waves was treated at five ST sessions and a total of 97 ml polidocanol, which is normal for the ST group. Manometry was performed 76 months after the last ST session. Of the two patients with abnormal configuration of the peristaltic waves one was treated at seven ST sessions and with a total of 161 ml sclerosant agent. This is more than the median for the group but not the largest amount of polidocanol in the group, 163 ml, which was given at five ST sessions to a patient who did not show any disturbances of the peristalsis of the corpus esophagi. Manometry was performed 12 months after the last ST session in this second patient with dysmotility. The third patient with abnormal configuration at the peristaltic waves only got 38 ml polidocanol at two ST sessions. In this case manometry was performed 81 months after the last ST treatment. We could not find any connection between the amount of injected sclerosing agent and the disturbances of peristalsis of the esophagi. There are two histopathological and morphological studies by Evans et al. [3] and by Soehendra et al. [10] where the alterations of the esophagus wall after sclerotherapy are investigated. In the study of Evans the intravariceal technique was used; in the Soehendra work a combined para- and intravariceal method was used. The 1st week after the initial injection an acute reaction with intramural inflammation and ulceration occurs. This is later replaced by formation of fibrosis at the site of the injection. The mucosa is first invaded by fibrous tissue; the reaction goes further transmurally and in some cases the muscular layer of the esophagus wall is reached. After the second sclerosing session, often after about a week, more

667

intensive necrosis takes place. Inflammation reaching submucosa and the connective tissue of the adventitia occurs and these layers are markedly more affected than the tunica muscularis. If the fibrous reaction involves the muscular layer stricture formation may take place. When the necrosis has healed, after about a month after the last injection, isolated scars of the esophageal inner wall can be identified. When the cicatrization is marked, the esophageal wall often appears rigid and manifests reduced motility. Over time mild strictures often normalize. The histological findings are extensive fibrotic tissue in the submucosa; in some cases the muscle layer is involved. In areas where depressed scars are noticed, the structure of the esophageal wall is not destroyed despite considerable fibrosis. To our knowledge no other investigators have reported histopathological findings from a clinical study in combination with manometry investigation. In our study we used a standard technique to take forceps biopsies from the lower esophagus. We did not manage to show any fibrous reaction in any of our specimens, since they all consisted of the squamous layer of the esophageal wall. However, in both the above-mentioned studies the histologic examinations were performed within 1 or 2 months after sclerotherapy, and no information is available on the long-term effect in the mucosa-submucosa after submucosal paravariceal injections. The results of this study show that this injection technique does not seem to cause any histologic alterations even after a long time.
Acknowledgment. Inga Ha gerstrand M.D., Ph.D., Department of Clinical Pathology, Lund University Hospital, Lund, Sweden performed the histopathological examinations.

3. Evans DMD, Jones DB, Clearly BK, Smith PM (1982) Oesophageal varices treated by sclerotherapy: a histopathological study. Gut 23: 615620 4. Grande L, Planas R, Lacima G, Boix J, Ros E, Esteve M, Morillas R, Gasull MA (1991) Sequential esophageal motility studies after endoscopic injection sclerotherapy: a prospective investigation. Am J Gastroenterol 86(1): 3640 5. Larson GM, Vandertoll DJ, Netcher DT, Polk HC (1984) Esophageal motility: effects of injection sclerotherapy. Surgery 96: 703710 6. OConnor KW, Lehman G, Yune H, Brunelle R, Christiansen P, Hast J, Compton M, McHenry R, Klatte E, Cockerill E, Holden R, Becker G, Kopecky K, Hawes R, Pound D, Rex D, Lui A, Snodgrass P, Weddle R, Crabb D, Lumeng L (1989) Comparison of three nonsurgical treatments for bleeding esophageal varices. Gastroenterology 96: 899906 7. Ogle SJ, Kirk CJC, Bailey RJ, Johnson AG, Williams R, Murray-Lyon IM (1978) Oesophageal function in cirrhotic patients undergoing injection sclerotherapy for oesophageal varices. Digestion 18: 178185 8. Sauerbruch T, Wirsching R, Leisner B, Weinzierl M, Phahler M, Paumgartner G (1982) Esophageal function after sclerotherapy for bleeding varices. Scand J Gastroenterol 17: 745751 9. Sauerbruch T, Wirsching R, Holl J, Grobl J, Weinzierl M (1986) Effects of repeated injection sclerotherapy on acid gastroesophageal reflux. Gastrointest Endosc 32: 8183 10. Soehendra N, de Heer K, Kempeneers I, Frommelt L (1983) Morphological alterations of the esophagus after endoscopic sclerotherapy of varices. Endoscopy 15: 291296 11. Spence RA, Smith JA, Isacs S, Terblance J (1990) Disturbed oesophageal motility after eradication of varices by chronic sclerotherapy. South Afr Med J 77: 138140 12. So derlund C, Thor K (1985) Oesophageal function after sclerotherapy for bleeding varices. Acta Chir Scand Preprint 524: 15 13. Terblanche J, Burroughs AK, Hobbs KEF (1987) Controversies in the management of bleeding esophageal varices. N Engl J Med 320: 1393 1398, 14691475 14. Uribarrena R, Jimenez J, Fortun MT, Jiminez C, Guerra A, Borda F (1990) Esophageal motility disorders in cirrhotics treated by sclerosing the varices. Rev Esp Enferm Dig 78(1): 15 15. Vestaby D, Hayes PC, Gimson AES, Polson RJ, Williams R (1989) Controlled clinical trial of injection sclerotherapy for active variceal bleeding. Hepatology 9: 274277 16. Warren WD, Millikan WJ Jr (1990) The relative role of sclerotherapy vs. surgical procedures in portal hypertension. Adv Surg 23: 119

References
1. Burnett DA, Rikkers LF (1990) Nonoperative emergency treatment of variceal hemorrhage. Surg Clin North Am 70(2): 291306 2. Crafoord C, Freckner P (1939) New surgical treatment of varicous veins in the oesophagus. Acta Otolaryngol 27: 422429

Surg Endosc (1997) 11: 622624

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

A murine model of laparoscopic-assisted intervention

J. D. F. Allendorf, M. Bessler, R. L. Whelan
Department of Surgery, Atchley Pavilion 522, Columbia University, 161 Fort Washington Avenue, New York, NY 10032, USA Received: 19 June 1996/Accepted: 2 November 1996

Abstract Background: In order to better investigate the effects of laparoscopic surgery, it is necessary to establish reliable, reproducible, and economical animal models of laparoscopic intervention. Here we describe a mouse model of laparoscopic-assisted colon resection. Methods: After successful induction of anesthesia the mouse is placed in Trendelenburg position and the peritoneal cavity is insufflated with carbon dioxide gas through an angiocatheter placed in the right upper quadrant. A 4-mm rigid scope with camera attachment is then inserted through a midline port created just caudal to the xiphoid. A second port is then created in the right lower quadrant to allow introduction of laparoscopic forceps into the peritoneal cavity. The cecum, which extends 1.5 cm beyond the ileocecal valve, is grasped with forceps and exteriorized through the operative port. Extracorporeally, the cecum is ligated and resected before the cecal stump is returned to the peritoneal cavity. The abdominal wall defects are then stapled closed. Results: This simple model can be mastered by individuals with very limited surgical experience. This laparoscopic model has been used successfully in our laboratory in a number of experiments with an intraoperative complication rate of 3.2% (3/94), which was similar to the open surgery group rate of 2.1% (2/95, p 0.99 by chi square). We observed no postoperative leaks in either group. The only postoperative death occurred in the open resection group due to dehiscence of the laparotomy wound. Conclusions: We propose that this model may be useful for comparing the effects of open to laparoscopic surgery. Key words: Mouse Murine Model Small animal Laparoscopy Laparoscopic surgery Bowel resection cecectomy

In a remarkably brief period of time laparoscopic techniques have been introduced for a number of general surgery proCorrespondence to: R. L. Whelan

cedures including cholecystectomy, gastric fundoplication, splenectomy, and bowel resection. The purported and often cited advantages of laparoscopic methods include less pain, earlier return of bowel function, and shorter hospitalizations [7]. The rapid clinical introduction of these minimal access methods precluded basic science studies to determine the precise physiologic differences between open and laparoscopic procedures. Recently, investigators have begun to study the effects of laparoscopic surgery vis-a ` -vis open surgery in a more rigorous manner. Utilizing animal models, our laboratory has been investigating the physiologic and immunologic consequences of laparoscopic-assisted bowel resection for the past 4 years. Initially, we used a porcine model of laparoscopic segmental sigmoid resection to evaluate several hemodynamic and physiologic parameters [5, 6]. While large-animal models are advantageous because the surgery more closely approximates that done in humans and because normal-sized instruments can be used, such models suffer from several shortcomings. These models are very expensive, especially for survival experiments; therefore the size of the study must be limited to a small number of animals. Furthermore, physiologic, immunologic, and oncologic assays often require species-specific reagents. Unfortunately, there are limited assays and reagents available for large-animal models. Finally, few cultured tumor cell lines have been developed for large animals, which makes it difficult to study in vivo tumor behavior. It was for these reasons that our laboratory endeavored to develop a mouse model for our laparoscopic studies. While the comparison of sham laparotomy to pneumoperitoneum is quite feasible in mice [1], the main drawback of the mouse model is that the small size makes it difficult to devise reasonable laparoscopic procedures that can be performed quickly and safely. Performing an actual bowel resection or other intervention poses a number of problems. It would be technically very challenging to perform bowel anastomoses in mice. The model of laparoscopic-assisted resection presented here permits a bowel resection yet does not require an anastomosis. As the effects of laparoscopic surgery continue to be investigated, it is necessary to establish reliable, reproduc-

623

Fig. 1. Normal anatomy of the mouse colon, ileum, and cecum, demonstrating how the cecum can be amputated without disrupting bowel continuity.

ible, and economical animal models of laparoscopic intervention. For this reason we established and report on a mouse model of laparoscopic bowel resection. Materials and methods Animals
Five- to 6-week-old female C3H/He mice (Charles River Laboratories, Wilmington, MA) weighing approximately 3035 g were acclimated to a light cycle and temperature-controlled environment for no less than 48 h prior to surgery. Mice were housed five to a cage in the Columbia University Institute of Comparative Medicine and were offered standard rodent chow and water ad libitum. All studies were performed under protocols approved by the Columbia University Institutional Animal Care and Use Committee.

Fig. 2. The mouse instrumented for laparoscopic surgery.

Equipment
This model requires a laparoscopic cart including a monitor, camera attachment, light source, insufflator with compressed CO2, and 4-mmdiameter 0 rigid scope. Instruments required include two pairs of fine dissecting forceps, dissecting scissors, 2-mm-diameter laparoscopic forceps, a small angiocatheter, and a large-bore needle.

quadrant to allow introduction of the laparoscopic forceps into the peritoneal cavity. As above, an 18-gauge needle is used to create a defect which is spread with forceps to a diameter of 2 mm. Again, no trocar is needed (Fig. 2). With the pneumoperitoneum established the abdomen is laparoscopically explored for the terminal ileum and cecum, which are usually found in the left lower quadrant. The cecum is grasped at its end with the forceps (Fig. 3) and carefully exteriorized through the 2-mm wound in the right lower quadrant. There is no mesentery, nor are there ligamentous attachments tethering the cecum; however, an artery and vein course longitudinally along the cecum to its end. Extracorporeally, the cecum and its vascular supply are ligated together with a single 4-0 silk ligature (Fig. 4). The cecum is then resected and irrigated with normal saline before the stump is gently returned to the peritoneal cavity. The abdominal wall defects are then stapled closed and the animal is allowed to recover from anesthesia.

Laparoscopic cecectomy
The normal mouse anatomy of the ileocecal region is shown in Fig. 1. The mouse cecum is proportionally much longer than in humans. The ileum inserts high into the right colon, creating a 1.5-cm blind-ended cecum. The procedure described below involves the amputation of this structure, which constitutes a bowel resection, and yet does not require that an anastomosis be performed. After successful induction of anesthesia with ketamine (50 mg/kg) and xylazine (5 mg/kg), the abdomen of a 56-week-old mouse is painted with Betadine and alcohol. The animal is placed in Trendelenburg position and the peritoneal cavity is insufflated with carbon dioxide gas at a pressure of 35 mmHg through a 25-gauge angiocatheter inserted into the right upper quadrant. An 18-gauge needle is then used to create a midline defect in the abdominal wall just caudal to the xiphoid. This defect is spread to a diameter of 4 mm using forceps. The 4-mm rigid scope with camera attachment is then inserted through this wound. No trocar is used. The elasticity of the abdominal wall fascia and musculature provides an airtight seal around the scope. A second port is then created in the right lower

Results To date our laboratory has done 94 laparoscopic-assisted cecal resections using this model which have been paired with 95 open cecal resections. Early in our experience, three mice (3.2%) undergoing laparoscopic-assisted cecal resection were sacrificed intraoperatively after inadvertent injury to bowel or a vascular structure. This was similar to the intraoperative complication rate in the open resection group (2.1%, p 0.99 by chi square). There were no postoperative leaks from the cecal stump noted in any of the animals. Furthermore, there were no deaths due to peritonitis noted in either group. The only postoperative death was in the open resection group secondary to evisceration due to dehiscence of the laparotomy wound.

624

Fig. 3. A laparoscopic view of the pelvis showing the cecum grasped at its blind end.

mastered by individuals with very limited surgical experience after several procedures. The procedure can be completed in less than 10 min in experienced hands. In our investigations, the length of surgery has been standardized to 20 min. Murine models have several advantages over largeanimal models for the scientific investigation of laparoscopy. Syngeneic species are genetically equivalent to other members of the same strain and are therefore ideal for immunologic studies. Many immunologic assays are well established in murine species. Tumor models are also well established in murine species. Genetic manipulation is most practical in mice. Finally, the short amount of time required to complete the procedure and the low cost per animal will allow investigators to include a larger number of animals in their studies. Using this mouse model of cecal resection, our laboratory has tested differences in tumor growth and establishment after open vs laparoscopic intervention [2]. We have also developed a similar model in rats which we have used to test for postoperative immunological differences after open or laparoscopic-assisted bowel resection [3, 4]. Our experience with 40 rats has shown an intraoperative complication rate of 2.5%, which was the same as that seen after the open resection. Recently, excellent 2-mm laparoscopes as well as improved 2-mm grasping laparoscopic forceps and dissecting scissors have been developed by a number of manufacturers and are commercially available. Such instruments will allow investigators to further limit the size of incisions necessary to perform laparoscopic procedures on small animals and should, therefore, improve these minimal access models.
Acknowledgment. This research was made possible by generous support from the Ethicon division of Johnson and Johnson Inc. and the Association of Women Surgeons.

References
1. Allendorf JDF, Bessler M, Kayton ML, Oesterling SD, Treat MR, Nowygrod R, Whelan RL (1995) Increased tumor establishment and growth after laparotomy vs laparoscopy in a murine model. Arch Surg 130: 649653 2. Allendorf JDF, Bessler M, Whelan RL, Laird D, Horvath K, Kayton ML, Nowygrod R, Treat MR (1996) Tumors grow larger after open vs laparoscopic bowel resection in a murine model. Surg Endosc 10: 182 3. Allendorf JDF, Bessler M, Whelan RL, Trokel M, Laird DA, Terry MB, Treat MR (1996) Postoperative immune function varies inversely with degree of surgical trauma in a murine model. Surg Endosc 11:427430 4. Allendorf JDF, Bessler M, Whelan RL, Trokel M, Laird DA, Terry MB, Treat MR (1996) Better preservation of immune function after laparoscopic-assisted versus open bowel resection in a murine model. Dis Colon Rectum (submitted) 5. Bessler M, Whelan RL, Halverson A et al. (1994) Is immune function better preserved after laparoscopic versus open colon resection? Surg Endosc 8: 881 6. Bessler M, Whelan RL, Halverson A, Allendorf JDF, Nowygrod R, Treat MR (1996) Controlled trial of laparoscopic-assisted vs open colon resection in a porcine model. Surg Endosc 10:732736 7. The Southern Surgeons Club (1991) A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 325: 15171518

Fig. 4. The cecum is ligated and resected extracorporeally.

Discussion The mouse model of laparoscopic cecal resection has proven to be a reliable, economical, and reproducible model of laparoscopic bowel resection. As stated previously, this model allows a bowel resection to be performed yet does not require an anastomosis. A bowel anastomosis in a mouse would be technically very challenging to perform. As described, the cecal resection can be carried out quickly with low morbidity and mortality. Experience with nearly 100 mice has resulted in an intraoperative complication rate of 3.2%, no postoperative deaths, and no leaks. This model has proven to be easy to master. This simple model can be

Surg Endosc (1997) 11: 618621

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The influence of laparotomy and laparoscopy on tumor growth in a rat model

C. A. Jacobi,1 J. Ordemann,1 B. Bo hm,1 H. U. Zieren,1 C. Liebenthal,2 H. D. Volk,2 J. M. Mu ller1
1 2

Department of Surgery, Humboldt-University of Berlin, Charite , Schumann Strae 20/21, 10098 Berlin, Germany Institute of Medical Immunology, Humboldt-University of Berlin, Charite , Schumann Strae 20/21, 10098 Berlin, Germany

Received: 7 November 1996/Accepted: 3 December 1996

Abstract Background: The effects of laparotomy and laparoscopy with different gases on subcutaneous and intraperitoneal tumor growth have not been evaluated yet. Methods: Tumor growth of colon adenocarcinoma DHD/ K12/TRb was measured in rats after laparotomy, laparoscopy with CO2 or air, and in control group. Cell kinetics were determined after incubation with carbon dioxide or air in vitro and tumor growth was measured subcutaneously and intraperitoneally after surgery in vivo. Results: In vitro, tumor cell growth increased significantly after incubation with air and CO2. In vivo, intraperitoneal tumor weight was increased after laparotomy (1,203 780 mg) and after laparoscopy with air (1,085 891 mg) and with CO2 (718 690 mg) compared to control group (521 221 mg) (p < 0.05). Subcutaneous tumor growth was promoted after laparotomy (71 35 mg) and even more after laparoscopy with air (82 45 mg) and CO2 (99 55 mg) compared to control group (36 33 mg). Conclusions: Insufflation of air and CO2 promote tumor growth in vitro. In vivo, intraperitoneal tumor growth seems to be promoted primarily by intraperitoneal air and subcutaneous tumor growth by CO2. Key words: Laparotomy Laparoscopy Tumor growth Air Carbon dioxide

tumors. The mechanism of this phenomen is still hypothetical [2, 8, 14]. In contrast to these clinical observations, Allendorf et al. have demonstrated that subcutaneous tumors were more easily established and grew more aggressively after laparotomy than after insufflation with CO2 in a mice model [1]. It was suggested that the difference in tumor growth is due to postoperative immune advantage in the laparoscopic group compared to open surgery. Since immunologic changes have not been evaluated in this study the hypothesis remains theoretical and has not been confirmed yet. Another important factor in tumor growth, which has not been evaluated until now, may be the influence of the gas used to establish pneumoperitoneum. Whether carbon dioxide, commonly utilized in laparoscopic surgery, causes subcutaneus port site recurrences has to be evaluated. The influence of air on tumor cell growth has also not been investigated. Therefore, we investigated the influence of different gases commonly used in laparotomy and laparoscopy on tumor cell growth in vitro, ex vivo, and in vivo.

Methods
In the first experiment the influence of carbon dioxide and air on tumor cells was investigated in vitro in order to evaluate the influence of different gases on the cell lines without immunological interaction of the host. Air was chosen to duplicate the situation during laparotomy. The hypothesis of the study was that there is a difference in tumor growth between the different gases. Colonic carcinoma cell line DHD/K12/TRb cells (ECACC) were cultured in Dulbeccos MEM (Biochrom, Germany) and Hams F10 medium (Biochrom, Germany) 1:1 supplemented with 10% fetal bovine serum (Gibco BRL, Germany), 2 mmol/l Gluthamin (Biochrom, Germany), and penicillin-streptomycin (Gibco, Germany) 1,000 IU/ml. Cells (5 106) were suspended in 10 ml of culture medium. They were incubated either with pure carbon dioxide or air or underwent no further manipulation in the control group. The gases were insufflated into the culture flasks (75 cm2/250 ml) through a sterile filter and drawn out through a control valve. Pressure during incubation was 0.5 (01.0) mmHg in each experiment. Gas analyses of the atmosphere in the flasks detected absence of oxygen during the incubation with carbon dioxide. Partial pressure of oxygen was 20.8 0.7 kPa after air incubation and 20.5 0.4

Laparoscopic techniques are still discussed controversially for treatment of malignancies because port site recurrences and peritoneal tumor seeding after laparoscopic cholecystectomy, colectomy, and gastric resection have been reported [3, 4, 6, 7, 11, 12, 15]. Apparently tumor seeding and development of subcutaneous metastases appear more often after laparoscopic than after open resections of malignant

Correspondence to: C. A. Jacobi

619

Fig. 1. In vitro tumor cell growth after insufflation with carbon dioxide (n 15) and air (n 15) in comparison to the control group (n 15) (mean and standard deviation, * p < 0.05: carbon dioxide vs control group, p < 0.05: air vs control group).

Fig. 2. In vitro tumor cell growth after insufflation in vivo with carbon dioxide (n 15) and air (n 15) in comparison to the control group (n 15)(mean and standard deviation, * p < 0.05: carbon dioxide vs control group, p < 0.05: air vs control group).

kPa in the control group. Before intervention, oxygen pressure in the medium was 24.8 1.7 kPa and carbon dioxide pressure was 2.3 0.3 kPa in all groups. After intervention, oxygen pressure changed to 6.7 0.7 kPa (carbon dioxide), 25.5 1.9 kPa (air), and 23.3 1.8 kPa (control) whereas carbon dioxide pressure was 79.3 3.3 kPa (carbon dioxide), 1.4 0.2 kPa (air), and 2.7 0.6 kPa (control) in the medium. After 3 h of incubation, cells were washed twice, resuspended, and cultured in 15 wells (1.88 cm2) with a concentration of 104 viable tumor cells/well. Three wells of each group were washed twice with PBS (Charite , Berlin) after 24 h and incubated with 300 l 0.05 trypsin/0.02% EDTA (Biochrom, Germany) for 30 min at 37C to remove adherent cells. Viable and dead cells were counted after staining with acridine-orange by fluorescence microscopy for 4 days. Experiments were repeated five times so that 15 wells in each group could be evaluated every 24 h. In the second experiment, the effect of pneumoperitoneum in rats with carbon dioxide or air on intraperitoneal tumor cells was evaluated. Tumor cells were injected into the peritoneal cavity of the animals, insufflated for 3 h, and removed and cultured ex vivo. These experiments were performed to evaluate the influence of gas insufflation in vivo (pneumoperitoneum) and to minimize the influence of the immunologic response of the tumor host. The hypothesis of this experiment was that there is a difference in tumor growth after in vivo insufflation with air and carbon dioxide. Fifteen rats were randomized into three groups and all received intraperitoneal injection of 5 106 cancer cells. Pneumoperitoneum with 8 mmHg was performed in two groups with insufflation of pure carbon dioxide (n 5) or air (n 5). The other five rats underwent the same 3 h of anesthesia and were used as control group. Gas analyses of the intraabdominal atmosphere detected again absence of oxygen at the end of insufflation with carbon dioxide and a partial oxygen pressure of 20.5 0.6 kPa after air insufflation. After 3 h, rats underwent laparotomy and intraperitoneal lavage with 15 ml of medium to harvest the cells. The total amount of viable tumor cells in the lavage was determined before centrifugation of the cell suspension. Cells were again cultured in 15 wells (1.88 cm2) with a concentration of 104 cells/well. Kinetics of cell growth were determined for 4 days as in the first experiment. In the animal model, the influence of laparotomy and laparoscopy with carbon dioxide or air was investigated after intraperitoneal and subcutaneous application of tumor cells. Tumor growth was measured 5 weeks after application. The hypothesis of this experiment was that tumor growth is increased after laparotomy compared to laparoscopy and that there is also a difference in tumor growth between the different gases used to establish pneumoperitoneum. Intraperitoneal and subcutaneous injection was performed to evaluate the effects on different tumor sites. One hundred male inbred BD IX rats 2 months old (Iffa-Credo, LArbresle, France) were acclimated to a climate- and light-cyclecontrolled environment for at least 7 days prior to investigations. The animals were allowed standard laboratory food and water ad libitum. All studies were performed under protocols approved by the local committees of Animal Use and Care. Tumor cells 1 104 were introduced intraperitoneally and subcutaneously in all animals. The rats were randomized into four different groups. Laparotomy was accomplished in the first group (n 25) and the abdomen was closed again after 30 min. In the second and third groups, lapa-

roscopy was performed with insufflation of either pure carbon dioxide (n 25) or air (n 25) at a pressure of 8 mmHg over 30 min. The abdominal wall was incised three times with introduction of 4.5-mm trocars to evaluate the development of tumor growth at these incisions. The control group (n 25) underwent tumor cell injection and incisions of the abdominal wall with introduction of 4.5-mm trocars. The animals were sacrificed 5 weeks later, tumors were excised intraperitoneally and subcutaneously, and the weights were assessed on a balance. Data are given as mean and standard deviation. Data between groups were compared using a Kruskal Wallis test for continuous data and the Fishers exact test for categorical data, if appropriate; p values less than 0.05 were considered significant.

Results

First experiment All tumor cells grew for 4 days (Fig. 1). Insufflation with carbon dioxide and air led to increased tumor cell growth in vitro compared to the control group with significant difference on day 4 (p < 0.05). Tumor growth was promoted even more after incubation with air.

Second experiment Cells were removed from the peritoneal cavity and grown over 4 days (Fig. 2). Tumor growth was stimulated even more in this ex vivo experiment and the number of viable tumor cells counted on the 4th day was four times higher in all three groups compared to the results of the first experiment. Tumor cell growth in vitro increased after in vivo insufflation of the cells with carbon dioxide compared to the control group. The number of viable cells was more than three times higher in the carbon dioxide group than in the control group (p < 0.05) after 48 and 72 h. Cell kinetics of the air group also showed a strong promotion of tumor cell growth, reaching significant difference compared to the control group only on day 4 because of an extremely high standard deviation in this group. After incubation, cell growth showed no significant difference between air and carbon dioxide insufflation.

620

Fig. 3. Weight of tumors taken from the subcutaneous site 5 weeks after cell application and insufflation (n 25 in each group) (mean and standard deviation, as * p < 0.05: carbon dioxide vs control group).

Fig. 4. Weight of intraperitoneal tumors 5 weeks after cell application and insufflation in the rats (n 25 in each group) (mean and standard deviation, * p < 0.05: laparotomy and laparoscopy with air vs control group).

Third experiment In vivo, subcutaneous tumor growth was promoted after laparotomy (71 35 mg) and even more after laparoscopy with air (82 45 mg) and CO2 (99 55 mg) compared to control group (36 33 mg) (p < 0.05) (Fig. 3). Intraperitoneal tumor weight also differed (p 0.007) between the four groups (Fig. 4). Tumor weights were 1,203 780 mg in the laparotomy group, 1,085 891 mg after laparoscopy with air, 718 690 mg in the CO2 group, and 521 221 mg in the control group. Frequency of tumor growth was increased (p < 0.001) at the abdominal trocar sites in carbon dioxide and air groups compared to the control group (Table 1). These tumors developed most often subcutaneously (95%). After laparotomy, seven rats developed tumor nodes at the laparotomy incision, ten rats had tumor at the abdominal incisions after insufflation with air, 20 rats had tumors after carbon dioxide, and only two rats developed tumors in the control group. Tumor growth at more than one incision was found in five rats of the air group, in ten rats of the carbon dioxide group, and did not occur at all in the control group.

Table 1. Incidence (%) of tumor at the abdominal incisions 5 weeks after the laparoscopy with carbon dioxide (n 25) or air (n 25) or without pneumoperitoneum (n 25) Carbon dioxide (n 25) 24% 24% 40% 12% Air (n 25) 52% 24% 16% 8% Control (n 25) 92% 8% p p p p 0.0001 0.2 0.0003 0.1

Tumor growth None At 1 incision At 2 incisions At 3 incisions

Discussion Metastases in the abdominal wall appear more often after laparoscopic surgery than after open resection for malignant tumor. Reports of early port site recurrence after laparoscopic procedures have reduced enthusiasm about laparoscopic techniques for malignant disease. Currently, an accepted explanation of what may cause these abdominal wall metastases does not exist [3, 4, 6, 7, 11]. In the first experiment, the direct influence of carbon dioxide and air on cultured cells was investigated in vitro. Cell growth of the adenocarcinoma cell line was significantly stimulated after insufflation with both gases compared to control group. The number of viable cells was even higher after air incubation than after carbon dioxide. It is hypothesized that the gases act on the cells themselves, but the mechanisms leading to promotion of the tumor cell growth in vitro either by carbon dioxide or air are not clear yet. Therefore, further studies are needed to elucidate the pathophysiological mechanisms of the observations.

In the second experiment, the effect of pneumoperitoneum with carbon dioxide or air on intraperitoneal tumor cells was evaluated in rats. Again tumor cell growth increased in the carbon dioxide and air groups compared with the control group. Additionally, the number of counted cells after insufflation in this experiment was higher than in the first experiment, which may be caused by peritoneal macrophages producing interleukin-1 (IL-1) and lead to promotion of cell growth [5]. The animal model used is well established and it is known that rat colonic carcinoma DHD/K12/TRb grows progressively when injected subcutaneously [13] or intraperitoneally [10] in syngeneic animals. In this study, 104 tumor cells were administered subcutaneously and intraperitoneally because the number of cells is sufficient to establish tumor growth in all animals in preliminary studies. Intraperitoneal tumor growth differed significantly between the four groups in vivo. Laparotomy markedly increased intraperitoneal tumor growth compared to laparoscopy with carbon dioxide and in the control group. Interestingly, intraperitoneal tumor weight was similar after laparoscopy with air insufflation and laparotomy. Thus, tumor growth seems to be promoted by intraperitoneal air and less by an additional trauma in the laparotomy group. In comparison to laparotomy and laparoscopy with air, laparoscopy with carbon dioxide led only to a minor increase of intraperitoneal tumor growth. This important difference has not been seen in the experiments in vitro, which indicates that reactions of the tumor host do modify tumor growth in vivo. In contrast to the abdominal site, tumor growth at the

621

subcutaneous site was not significantly different between the treated groups. Subcutaneous tumor weight was mostly pronounced in the laparoscopic group with carbon dioxide. Carbon dioxide seems to act more on the subcutaneous site than intraperitoneally, which may be explained by the high solubility of carbon dioxide in subcutaneous tissue. The results indicate that carbon dioxide actually is an important factor promoting tumor growth in small trocar incisions. Most of the reported abdominal wall metastases after laparoscopic surgery did not occur in the abdominal wall but in subcutaneous tissue close to the trocar incisions [3, 6, 9], which is in accordance with our experiments. Almost all tumor nodes at the abdominal incision were located in the subcutaneous tissue. If this can be confirmed in future studies, it may be that early port site recurrences are not as disastrous as they appear today because local excision may be a curative treatment. In contrast to our results, Allendorf et al. have demonstrated that tumors were more easily established subcutaneously and grew more aggressively after laparotomy than after insufflation with carbon dioxide [1]. In their study a mammary carcinoma cell line was used without intraperitoneal application of the cells. Pneumoperitoneum was performed over 30 min with an intraabdominal pressure of only 46 mmHg, which might have caused lower levels of carbon dioxide in the subcutaneous tissue. They also found a higher incidence of tumor in the insufflation group compared to the control group. The authors suggested that there is a postoperative immune advantage in laparoscopy leading to minor subcutaneous tumor growth compared to laparotomy. This hypothesis is still unproven. It remains unclear whether immunologic suppression after laparotomy is related to surgical trauma or other factors like intraperitoneal air. Many questions remain unanswered because the pathomechanisms of these phenomena are not known yet. If the results can to be confirmed in other studies they may change an attitude regarding treatment of cancer patients with questionable peritoneal carcinomatosis. They should rather undergo laparoscopy in order to evaluate the intraperitoneal tumor status than open laparotomy because laparotomy may harm patients much more than laparoscopy and may reduce survival time. Furthermore, alternative gases should be evaluated in experimental studies to avoid the effects of carbon dioxide on subcutaneous port site metastases. In conclusion, the results of the study indicate that the difference in intraperitoneal tumor growth between laparotomy and laparoscopy is mainly caused by intraabdominal

air and probably not to an additional trauma in the laparotomy group. However, the hypothesis that air might influence the immune system of the tumor host and increase immune suppression after laparotomy more than after laparoscopy remains theoretical. Furthermore, the data suggest that insufflation of carbon dioxide may be an important factor which promotes subcutaneous metastases.

References
1. Allendorf JDF, Bessler M, Kayton ML, Oesterling SD, Treat MR, Nowygrod R, Whelan RL (1995) Increased tumor establishment and growth after laparotomy vs laparoscopy in a murine model. Arch Surg 130: 649653 2. Bessler M, Whelan RL, Halverson A, Treat MR, Nowygrod R (1994) Is immune function better preserved after laparoscopic versus open colon resection? Surg Endosc 8: 881883 3. Cava A, Roman J, Gonzales Quintela A, Martin F, Aramburo P (1990) Subcutaneous metastasis following laparoscopy in gastric adenocarcinoma. Eur J Surg Oncol 16: 6367 4. Clair DG, Lautz DB, Brooks DC (1993) Rapid development of umbilical metastases after laparoscopic cholecystectomy for unsuspected gallbladder carcinoma. Surgery 113: 355358 5. Dinarello CA (1994) The biological properties of interleukin-1. Eur Cytokine Netw 5(6): 517531 6. Drouard F, Delamarre J, Capron J (1991) Cutaneous seeding of gallbladder cancer after laparoscopic cholecystectomy. N Engl J Med 325: 1316 7. Fusco MA, Paluzi MW (1993) Abdominal wall recurrence after laparoscopic-assisted colectomy for adenocarcinoma of the colon. Dis Colon Rectum 36: 858861 8. Horgan PG, Fitzpatrick M, Couse NF, Gorey TF, Fitzpatrick JM (1992) Laparoscopy is less immunotraumatic than laparotomy. Minimally Invasive Ther 1: 241244 9. Jacobi CA, Keller H, Mo nig S, Said S (1995) Implantation metastasis of unsuspected gallbladder carcinoma after laparoscopy. Surg Endosc 9: 351352 10. Lagadec P, Jeannin JF, Reisser D, Pelletier H, Olsson O (1987) Treatment with endotoxins of peritoneal carcinomatosis induced by colon tumor cells in rats. Invasion Metastasis 7: 8395 11. ORourke N, Price PM, Kelly S, Sikora K (1993) Tumor inoculation during laparoscopy. Lancet 342: 368 12. Pezet D, Fondrinier E, Rotman N, Guy L, Lemesle P, Lointier P, Chipponi J (1992) Parietal seeding of carcinoma of the gallbladder after laparoscopic cholecystectomy. Br J Surg 79: 845 13. Reisser D, Fady C, Lagadec P, Martin F (1991) Influence of the injection site on the tumorigeniticity of a cloned colon tumor cell line in the rat. Bull Cancer 78: 249252 14. Torkel MJ, Bessler M, Treat MR, Whelan RL, Nowygrod R (1994) Preservation of immune response after laparoscopy. Surg Endosc 8: 13851388 15. Wexner SD, Cohen SM (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 295298

EndoScope: world literature reviews

Surg Endosc (1997) 11: 679683

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Original articles from a wide range of international surgical journals are selected by our editors and presented here as a structured summary and critical review. EndoScope serves as a quick and comprehensive survey of the expansive endoscopic literature from all the corners of the globe.

Long term results of endoscopic stenting and surgical drainage for biliary stricture due to chronic pancreatitis
Smits, ME, et al Br J Surg (1996) 83:764768 A retrospective evaluation of endoscopic stenting in 58 patients with benign biliary stricture due to chronic pancreatitis is reported with the intention of determining whether endoscopic therapy alone can replace surgical drainage. Persistent jaundice in 46 and cholestasis in 12 patients are noted (cholestasis being defined as increased alkaline phosphatase levels thrice the normal value for >1 month). ERCP is performed in all patients with placement of 10-Fr stents. Follow-up is approximately 39 months thereafter, with stents being exchanged electively or when patients had recurrent symptoms (jaundice, cholangitis, pain). Success is defined as clinical improvement with stricture regression and permanent removal of the stent. Median follow-up is 49 months. All patients had relief of jaundice and cholestasis within 2 weeks of stent insertion. The overall rate of complication following ERCP is 9% (two bleedings, one pancreatitis, one cholecystitis and liver abscess, one fever of unknown origin, and zero mortality). Patients had a median of three stents over a median of 14 months. Stent-related complication rate is 64%, which includes clogging, migration, and erosion necessitating replacement stents. With 4.1-year median follow-up, 28% of the study group show regression of stricture with permanent removal of stent in about a median of a 10-month period; 72% had persistent biliary stricture: 45% requiring continued stenting and 27% undergoing surgery. Indications for surgery are frequent stent-related complications in nine patients, duodenal stenosis in two, and pancreatic surgery for pain in five with chronic pancreatitis. Procedure type ranges from simple sphincteroplasty to Whipple resection with the majority receiving various biliary bypasses. As such, the resultant postop morbidity of 38% ranged from wound infection to biliary leakage. The authors conclude that both endoscopic stenting and surgery are effective treatments for biliary stricture in patients with chronic pancreatitis; of particular note is the lower complication rate associated with endoscopy. They do concede, however, that late stent-related complications remain a major limitation. Overall, this paper addresses an important issue, the management of biliary stricture in the setting of chronic

pancreatitis. As expected, endoscopic management remains a safe and effective means for treating the early phase of biliary obstruction, and the authors emphasize this point very well. The patients receiving surgical treatment, on the other hand, remain a distinct group standing apart from the endoscopy responders. The 16 patients requiring surgical therapy all underwent previous repeated stenting followed by failurea major indicator probably reflecting the severity of chronic pancreatitis. Although early stenting is associated with a lower complication rate, the direct nonstratified comparison to surgical postop morbidity rate remains equivocal. The authors do identify a distinct role for endoscopic therapy in biliary stricture due to chronic pancreatitis.

Endoscopic perforation in unsedated patients undergoing endoscopy

Hedenbro, JL, Ekelund, M Br J Surg (1996) 83:845846 The complication rate of esophagogastroduodenoscopy in 13,639 cases is presented from data collected since 1990 in a university hospital referral center. Most procedures are performed without sedation (97.9%) using standard flexible endoscopes. Only two cases report perforation that required operative repair. The authors have shown that unsedated endoscopy with throat analgesia is safe and associated with low perforation rate (0.015%). No significant cardiopulmonary events are noted and the perforation is well below the usual 0.03 0.05% rate. An additional interesting point is that oxygen saturation monitoring and oxygen supplementation are not routinely employed although the cost-effectiveness of this step is probably negligible.

Relations between oesophageal acid exposure and healing of oesophagitis with omeprazole in patients with severe reflux oesophagitis
Holloway, RH, et al Gut (1996) 38:649654 The effect of omeprazole and severe ulcerative esophagitis due to acid reflux is examined. Eight-week treatment of

680

omeprazole at 20 mg per day in 60 patients with grade 3 or 4 ulcerative esophagitis is followed by an additional 8-week course of omeprazole at 40 mg daily for those unhealed with initial therapy. Endoscopy and 24-h esophageal pH monitoring are performed before treatment and at the end of each treatment phase. Only patients with severe erosive peptic esophagitis are entered, and those with reflux disease marked by columnar metaplasia or benign stricture without erosion are excluded. Of the 60 entered for the study, 55 patients eventually underwent the omeprazole trials. Thirty percent failed to heal in the initial trial, and they all exhibited greater 24-h acid exposure than the group that healed. In the second phase of the trial, 47% of the unhealed also failed the doubled dosage trial for additional 8 weeks. Once again, the failed group exhibited greater acid exposure than the healed subjects. The authors conclude that patients with severe ulcerative esophagitis who are refractory to omeprazole have greater esophageal acid exposure while receiving treatment than responding patients. They attribute this to reduced responsiveness to acid suppression. An interesting point one gathers from this study is the significant number of patients who may still respond to increased omeprazole dosage after the initial trial failure. Although one-third failed the initial dose, nearly half of the unhealed patients responded with increased omeprazole administration. The next question is, How about an additional 8-week trial for the ones who failed to heal?

disease group, only seven completed the study. Perhaps more numbers in this group will reinforce the data presented. However, these results do point to the role in gastric inflammation of H. pylori in peptic ulcer disease.

Randomized, prospective, single-blind comparison of laparoscopic versus small-incision cholecystectomy

Majeed, AW, et al Lancet (1996) 347:989994 A prospective randomized comparison between elective laparoscopic and small-incision cholecystectomy in 200 patients involving four surgeons was conducted at Sheffield, UK. All subjects had symptomatic gallstones confirmed by ultrasonography and were randomized for elective cholecytectomy by either method. The four participating surgeons had performed a minimum of 40 laparoscopic cholecytectomies each as either the principle operator or assistant and no comment was made about their small-incision experience. The small-incision method requires a high transverse subxiphoid incision over the junction of the cystic duct and common bile duct. The length of the incision is tailored to individual cases (median of 7 cm). Standard instruments are used for the fundus-last dissection, and finally, as in laparoscopy, all cases are accompanied by intraoperative cholangiogram. The authors report a longer time required for laparoscopic cholecytectomies when compared to the smallincision procedure (median 65 min vs 40 min, respectively, and including cholangiogram time). They found no difference between the groups for hospital stay, time back to work, and time to full activity. Of the 205 meeting the criteria for elective cholecytectomy between 1992 to 1995, only five cases are eliminated because of refusal for randomization, hepatic metastasis, or cirrhosis. Although patient selection was randomized, the four surgeons participating in the study were still able to affect the outcome of the studya design flaw that is always an inherent problem in this type of study. The longer time reported for the laparoscopic procedure may not be a significant finding as it can certainly be shortened. Finally, the complication reported is notable for one bile duct injury only in the laparoscopic group, which was fortunately identified after an intraoperative cholangiogram. An interesting observation is that the overall conversion rate in the laparoscopic group turned out to be 20%, a figure that is a little too preferential for the open technique. Certainly, the authors have shown that for the surgeons well adapted to the small-incision technique, postoperative recovery results can be achieved that are similar to those with the laparoscopic method. However, it is impossible to remove the bias for or against the procedure of the surgeons in the study, and therefore true randomization remains questionable.

Eradication of Helicobacter pylori in patients with duodenal ulcer lowers basal and peak acid outputs to gastrin releasing peptide and pentagastrin
Harris, AW, et al Gut (1996) 38:663667 The relationship between Helicobacter pylori and gastric acid hypersecretion is explored in patients with duodenal ulcer disease. Ten healthy volunteers with normal endoscopy and negative H. pylori and ten positive H. pylori patients with documented duodenal ulcer were followed for 6 months after H. pylori eradication with omeprazole and amoxycillin therapy. H. pylori status was determined by histology, culture, and by the 13C-urea breath test. Basal acid output (BAO) and peak acid outputs (PAO) following gastrin-releasing peptide (GRP) and pentagastrin stimulation were recorded before and 6 months after H. pylori eradication. Basal and stimulated acid output levels are elevated more in the H. pylori group than the negative controls. Six months after H. pylori eradication, basal and stimulated acid outputs are all below the initial starting levels. The results show H. pylori involvement in hypersecretion and duodenal ulcer disease by revealing lower basal and peak acid outputs to gastrin-releasing peptide and pentagastrin following H. pylori eradication. Of the ten patients selected initially in the

681

Acute phase is the only significantly reduced component of the injury response after laparoscopic cholecystectomy

Targarona, EM, et al World J Surg (1996) 20:528534 This study investigates the response to injury induced by laparoscopic vs open cholecystectomy by evaluating the neuroendocrine, acute phase, and metabolic responses. Twenty-five patients with noncomplicated gallstones are studied in prospective but not randomized fashion (12 in laparoscopic vs 13 in open cholecystectomy groups). As expected, the requirement for postoperative analgesia and hospital stay are significantly less in the laparoscopic group. Both procedures result in increased levels of serum glucose, lactate, white blood cell count, prolactin, ACTH, cortisol, interleukin-6, C-reactive protein, and arterial PCO2 while the levels of total proteins, albumin, prealbumin, free fatty acids, hemoglobin, and arterial pH are decreased. No appreciable changes in the levels of growth hormone, insulin, or glucagon are noted. The notable differences between the two are significantly decreased levels of interleukin-6, C-reactive protein, and prealbumin in the laparoscopic group when compared to the open group. The authors conclude that the less intense acute-phase response may be attributable to the reduced size of the wounds in laparoscopy. The invited commentary by Haglund and Rastad follows the article and outlines well some of the problems in the study, such as lack of randomized selection groups, too small a number of subjects to make an adequate conclusion, and the question of precision of the IL-6 bioassay involved. In addition, while some of the reported parameters in this study substantiate previous reports, others contradict. Perhaps this confusion could be clarified by focusing on the relevance of these biochemical-parameter measurements in injury response and on the validity of using them to generalize about acute-phase response in laparoscopy. This is an interesting study overall with notable measurements, but perhaps the title and the conclusion could be less general, as the paper still does not show a direct link between wound size and the various inflammatory responses reported.

identified the 100 cases. Of these, 26 are included with failed ERCP retrieval of common bile duct (CBD) stones. The CBD stones are then retrieved laparoscopically using a combination of choledochoscope and an intracorporeal electrohydraulic or pulsed-dye laser lithotriptor. The approach is either by trancystic duct CBD (n 63) or direct CBD (n 33) exploration via choledochotomy, with only four cases converted to open procedure. Biliary drain placement as well as completion cholangiogram is performed for all cases. Retained CBD stones are identified in five patients, of which two are retrieved by ERCP and three by lithotripsy. The same surgeon is the principal operator in the entire series. Finally, the biliary drains are removed 5 weeks postop after obtaining follow-up cholangiogram. Major morbidity reported includes bile leakage from the cystic stump due to slipped clips (n 2) and accidental removal of biliary drain (n 1), all of which are treated with nasobiliary decompression. One mortality involves cardiogenic shock on postop day 3 in an 81-year-old who had undergone failed ERCP clearance of large CBD stones and developed cholangitis prior to laparoscopy. This article clearly demonstrates the relative safety and feasibility of laparoscopic retrieval of CBD stones as a onestage procedure in conjunction with laparoscopic cholecystectomy. Although there is no control group in the trial, the advantages of a minimally invasive procedure apply in this scenario as well. The issue of cost will be the major hurdle; the degree of expertise required for this procedure to gain wide acceptance is also an issue. As in many laparoscopic reports, the data from this series reflect experience of a single surgeon. It would be interesting to examine pooled data from multicenter trials to evaluate reasonable success and morbidity rates. What the best treatment of CBD stone is is a question still unanswered, and further data comparing the ERCP retrieval vs the laparoscopic method will no doubt be available in the near future.

Laparoscopic treatment of nonparasitic liver cysts: adequate selection of patients and surgical technique
Gigot, JF, et al World J Surg (1996) 20:556561 Laparoscopic fenestration of nonparasitic liver cysts is performed in 26 patients. Seventeen had symptomatic solitary liver cyst while nine suffered from polycystic liver disease. The major symptom reported is abdominal pain and eight (31%) of these patients had prior treatment failures with conventional methods. Laparoscopic fenestration includes wide cystic wall deroofing of as many cysts as possible with hemostasis achieved by argon-beam coagulation. A peritoneal drain is placed in 23 patients, and all excised cysts are confirmed by histologic exam. The mean duration of follow-up with routine CT scan is 9 months. Recurrence is defined as reappearance of cysts on CT scan or symptoms. The feasibility of laparoscopy is 92%, with only two cases requiring laparotomy because of inaccessibility of the

Laparoscopic treatment of gallbladder and common bile duct stones: a prospective study
Lezoche, E, et al World J Surg (1996) 20:535542 Single-stage laparoscopic treatment of gallstones and ductal stones in 100 patients is reported. A total of 950 consecutive patients admitted for elective laparoscopic cholecytectomy were submitted to intraoperative cholangiogram, which

682

cysts. Although the authors claim no major morbidity or mortality, one case of intraoperative cardiac arrest during insufflation is noted. The mean postop stay is 4.6 days. The two patients who underwent laparotomy stayed for 8 and 12 days, respectively. Clinical recurrence of symptoms is seen in six cases (23%) but only two are attributable to recurrence of liver cyst. Recurrence documented on CT scan is seen in ten patients (38%) with only one undergoing reoperation by open fenestration. The application of laparoscopy in the treatment of liver cysts is an appropriate method available and any study that contributes to this database is always valuable. Because of the rather noticeable recurrences, the authors advocate severe restriction of surgical indications and careful selection of patients. They go so far as to recommend preoperative ultrasound or CT-guided aspiration of these cysts to test for relief of symptoms, and the laparoscopic approach be reserved only if cyst-related symptoms recur with the cyst recurrence. Although laparoscopy has been shown to be feasible and relatively safe, the follow-up recurrences are discouraging and certainly do not measure up to conventional treatment, which will ultimately decide the future of any new application. As advocated, perhaps there is a role for laparoscopy in cases of isolated, easily accessible, liver cysts.

carcinoma. There are five false-positive laparotomies in the series, and these should be included in the morbidity category. Perhaps laparoscopy should be considered in such cases. At any rate, the lack of yield for carcinoma by sigmoidoscopy is notable, with the rate of perforation at 0.5%. Barium enema, on the other hand, reveals the 10% from the same group of patients that required further workup with colonoscopy with 0% complication. However, despite barium contrast study and colonoscopy, five patients had to undergo negative laparotomies because of equivocal findings. Overall, this study is a valuable contribution and will no doubt help us to monitor the current policy involved in massive screening and workup for the fight against colorectal carcinoma.

Cost-effectiveness management of complicated choledocholithiasis: laparoscopic transcystic duct exploration or endoscopic sphincterotomy
Liberman, MA, et al J Am Coll Surg (1996) 182:488494 This is a valuable study that examines the results and actual costs of common bile duct stone therapy in the context of minimally invasive surgery. In a retrospective analysis over a 4-year period, all patients undergoing laparoscopic cholecystectomy (LC) plus endoscopic sphincterotomy (ES) are compared to those undergoing LC plus laparoscopic transcystic common bile duct exploration (LTCBDE). All procedures were performed at the same institution by different surgeons with varying levels of experience. Of the 76 patients with choledocholithiasis, 59 patients underwent LC plus LTCBDE (group 1) and 17 patients underwent LC plus ES (group 2). A subset of 21 patients in group 1 who had urgent LC plus LTCBDE (group 3) because of cholecystitis or cholangitis, with or without pancreatitis, were also examined separately. In terms of hospital stay, group 1 or even group 3 is associated with significantly shorter average period (6.1 and 6.9 days, respectively) than group 2 (12.4 days, p < 0.001). The average cost in group 1 is $13,151, and $14,732 with professional reimbursement, and this includes the seven complications (12%) and one death (1.7%) reported. The cost in group 2 averages about $18,712, or $21,125 with professional reimbursement, while there are seven (41%) complications reported, all related to ES, with no deaths, and this is factored into the cost. When group 3 is examined, the urgency of LC plus LTCBDE does not increase the average cost ($13,564, or $15,150 with reimbursement) or the morbidity rate (two complications or 10%). Thus, the paper concludes that LC plus LTCBDE for CBDS, whether urgently or electively, has decreased morbidity rates, length of hospital stay, and costs when compared to LC plus ES. The relatively high rate of morbidity reported in group 2 probably reflects the degree of illness of these patients at presentation as well as the small number involved in the retrospective study. As the authors report, the usual reported ES complication rate is 8.2% with 1.3% mortality. Clearly, the 41% morbidity rate reported is too high, and this will no

Endoscopic and surgical complications of work-up in screening for colorectal cancer

Kewenter, J, Brevinge, H Dis Colon Rectum (1996) 39(6):676680 The purpose of this extensive study is to evaluate the complication rate of diagnostic and therapeutic procedures involved in massive screening (68,306) and workup for colorectal neoplasm. Flexible sigmoidoscopy (FS), up to 60 cm, is performed in 2,108 patients who were Hemoccult positive on initial screening. Only 39 had to be repeated due to poor bowel prep; 554 polypectomies are performed in 413 (20%) patients with three reported cases (0.5% polypectomies) of sigmoid colon perforation requiring repair. Only benign adenomas are noted on biopsy. A total of 1,987 patients (94%) of the 2,108, comply with barium enema without complication; 190 (10%) of these patients then undergo colonoscopies for abnormal barium studies. Polyps are removed in 113 patients during colonoscopy, of which three (0.03%) sustain complications (one hemorrhage, two perforations). Laparotomies with appropriate procedures are performed on 104 patients (carcinoma 79, adenoma 13, diverticular disease four, no tumor five, inflammatory bowel disease two, abdominal aneurysm one). Significant complications developed in 14 (13%) patients, and although the authors report no deaths, one eventually died 2.5 months postop. A total of ten surgeons participated in this trial for examining and operating the patients involved. Out of 2,108 patients undergoing sigmoidoscopy, 104 (0.05%) eventually undergo laparotomy with 13% morbidity and 0.01% mortality; 79 (0.04%) are diagnosed with

683

doubt lead to higher costs and hospital stay. The authors do attempt to adjust for this by creating group 3 (21 out of 59 patients whose urgent operative indications are similar to group 2). Interestingly, the cost in group 3 is still less when compared to group 2; however, the high complication rate in group 2 accounts for this difference. Perhaps more numbers in this study and a prospective trial will elucidate the true complication rate of ES.

Intravenous infusion cholangiography for investigation of the bile duct: a direct comparison with endoscopic retrograde cholangiopancreatography

Bloom, IT, et al Br J Surg (1996) 83:755757 The newer technique of slow intravenous infusion cholangiography (IIC) using meglumine iotraxate is assessed using (ERCP) endoscopic retrograde cholangiopancreatogram as a gold standard in patients considered to be at high risk for bile duct stones. A total of 111 consecutive patients with cholelithiasis documented by ultrasound with the following criteria are selected: (1) history of jaundice but not on admission, (2) acute pancreatitis, (3) dilated bile duct 6 mm or bile duct stones seen on ultrasound, and (4) abnormal serum bilirubin or alkaline phosphatase at presentation. IIC

is performed in all patients using 2-h IV infusion of meglumine iotroxate (Biliscopin; Schering-UK) followed by abdominal radiographs and then ERCP within 24 h of the study. Out of 111 consecutive patients, three (2.7%) had incomplete IIC and eight (7.2%) had unsuccessful ERCP, leaving 100 in the study group. No complication is reported for IIC while three develop pancreatitis following ERCP; 81 patients had normal studies while 16 patients were detected for bile duct stones by both procedures. IIC had two falsenegatives and one false-positive for bile duct stones when compared to ERCP (IIC sensitivity 89%, specificity 99%). Ultrasound was neither sensitive nor specific with five false-negatives and 24 false-positives. Reactions to contrast media had been the major resistance to use of IIC for bile duct evaluation. With the use of newer contrast, in this case meglumine iotroxate, an irresistible alternative to ERCP is available. The reported complication rate for IIC is 1%, while for ERCP an association of 5% or higher can be expected with global mortality rate of 0.08%. At Stockport, UK, IIC has 0% morbidity, with an ERCP rate of 3%, while the cost of ERCP when compared to IIC is tenfold. The authors advocate IIC evaluation preoperatively for patients in a high-risk category for bile duct stones followed by ERCP if necessary, thus reserving ERCP for a therapeutic rather than a diagnostic role. This is a most welcome contribution for evaluation of common bile duct stone in the elective setting and no doubt will help to popularize IIC use in the future.

Reviewers for this issue: F. Chae, J. Sackier

Surg Endosc (1997) 11: 693695

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

A three-trocar technique for limited laparoscopic renal surgery

G. S. Ferzli, J. B. Hurwitz, H. Usal, A. A. Massaad
Department of Laparoendoscopic Surgery, Staten Island University Hospital, 78 Cromwell Avenue, Staten Island, NY 10304, USA Received: 29 March 1996/Accepted: 1 July 1996

Abstract. Laparoscopic renal surgery usually involves the use of five or six trocars. This report concerns the authors technique for performing such surgery through only three trocars. Semilateral patient positioning, along with additional table rotation, is utilized to facilitate visceral rotation and optimize exposure of the kidney. Four laparoscopic renal procedures were performed: one renal cyst decortication and three upper pole partial nephrectomies with ureterectomies for duplications of the collecting system. Mean operative time was 148 min with no conversions; there were no intra- or postoperative complications. All patients tolerated a liquid diet on postoperative day 1, and the median hospital stay was 2 days. In selected cases laparoscopic renal surgery may be approached safety through three trocars. Key words: Laparoscopy Renal surgery Nephroureterectomy

ments of the left colon along the white line of Toldt from the splenic flexure to the iliac vessels. The splenocolic ligament is divided, and the operating table is then rotated an additional 30 to place the patient in a 75 lateral decubitis position. This significantly aids in mobilizing the left colon medially. Adequate exposure of Gerotas fascia and the ureter is also achieved.

Case reports
Four laparoscopic renal resections were performed at this institution by a single surgeon (G.S.F.) between June and September 1995. None of these patients had undergone prior abdominal surgery. The first patient underwent partial excision of the left renal cyst while the other three patients had left upper pole nephroureterectomies.

Case 1
A 40-year-old female presented with recurrent flank and back pain. Preoperative workup with CT of the abdomen, renal and bladder sonography, and urodynamic studies revealed a left renal midpole inhomogeneous cyst, with evidence of intracystic hemorrhage. Aspiration of the cyst was negative for malignancy. Utilizing the technique described above, the cyst was accessed by incising Gerotas fascia and retracting the perirenal fat. Ultrasonically activated coagulating shears (LCS, Ultracision Inc., Smithfield, RI) were then used to unroof the cyst circumferentially. The contents were evacuated without spillage, the septae were divided, and multiple biopsies were taken. These all proved negative for malignancy. Operative time was 120 min, with 300 cc estimated blood loss.

Since 1990 a number of authors have documented the feasibility of a laparoscopic approach to the kidney for the treatment of a wide variety of urological disorders. All these studies, however, utilized five or six trocars. This is a review of the authors experience with four laparoscopic renal procedures performed with only 3 ports.

Methods
A Foley catheter and a nasogastric tube are inserted in each case following the administration of general anesthesia. The patient is placed in a 45 semilateral right decubitus position and pneumoperitoneum is established with a Veress needle. A 30 angled scope is placed through a 10-mm supraumbilical trocar. Two additional 5-mm trocars are placed under direct vision, one subcostally in the midclavicular line, and the other just above the iliac crest in the anterior axillary line (Fig. 1). After a routine exploratory laparoscopy, the dissection is started by incising the lateral attach-

Case 2
A 23-year-old female presented with recurrent urinary tract infections. Intravenous pyelography, voiding cystourethrography, cystoscopy with retrograde pyelography, renal sonography, and abdominal CT revealed a duplex system on the left, with a nonfunctioning upper pole and an ectopic ureter. Intraoperatively, once the midureter was exposed, the 5-mm laparoscope was repositioned from the supraumbilical to the subcostal trocar. The dilated ureter was divided after its origin from the atrophic upper pole had been verified. The distal end was dissected as far down as possible, ligated with an absorbable Endoloop, and removed. During this dissection, care was taken to avoid injury to the normal ureter. The proximal end of the

Correspondence to: G. S. Ferzli

694 Table 1. Case 1 2 3 4 Procedure Unroofing of cyst Upper pole nephroureterectomy Upper pole nephroureterectomy Upper pole nephroureterectomy Age 40 23 18 24 years years months months Operative time (min) 120 173 180 120 Blood loss 300 200 10 10

Discussion There have been many applications of laparoscopy to renal surgery since 1990 [1, 3, 6, 8, 9], but all laparoscopic transperitoneal neurectomy techniques reported to date have required five or six trocars [3, 4, 6]. We have been able to achieve adequate exposure using only three trocars by utilizing the semilateral decubitus position described above. The rationale behind using fewer ports is not cosmesis, but rather an interest in reducing early and late complications of unnecessary incisions; these include vascular and bowel injuries, as well as trocar-site hernias [2, 5, 7]. In this series of four patients, the average operative time (158 min) and hospital stay (3 days) compared quite favorably with studies in which five or six trocars were used. The operative times in those series were 240 and 498 min, and the stays were 6 and 5 days, respectively [4, 6]. Rotating the operating table 30, in addition to the initial 45 patient position, allows gravity to retract abdominal viscera, avoiding the need for additional trocar retractors. Intestinal manipulation with laparoscopic instruments is also minimized, which may eliminate the risk of the iatrogenic enterotomies, which have been reported as complications of this procedure [6]. The use of an angled scope is a necessity, especially for safe dissection and manipulation of the hilar structures. The exposure is further improved by placing the laparoscope through the supraumbilical port during the nephrectomy and switching it to the subcostal port during the ureterectomy. Other authors have used a four-trocar technique in a retroperitoneal approach [7]. While this may avoid the necessity of intraperitoneal adhesiolysis, it does not provide exposure of the lower segment of the ureter. Open retroperitoneal approaches often utilize two incisions. We feel that the transperitoneal technique we have adopted offers the best blend of flexible exposure of the kidney and the full length of the ureter, together with minimal morbidity. Limited laparoscopic renal surgery via three trocars is feasible, and is safe in selected cases. The use of fewer ports is a simplification of current approaches; whether or not it offers significant advantages over more conventional techniques remains to be determined.

Fig. 1. Positioning of three trocars for renal surgery.

ureter was passed behind the gonadal and renal vessels cephalad. The laparoscope was returned to its supraumbilical position, and 30 of Trendelenburg were added to patient positioning; this permitted the spleen to retract medially, allowing optimum exposure of the renal hilum and superior pole. The upper pole vessels were ligated and divided between clips, after which the parenchyma was excised with the LCS along the line of demarcation. Specimens were retrieved in a bag and removed through slightly enlarged trocar sites, which were closed in two layers. Operative time was 173 min, with 200 cc estimated blood loss.

Case 3
Case 3 was an 18-month-old baby girl with recurrent urinary tract infections. Renal sonography, voiding cystourethrography, and nuclear renal scan revealed a left duplex system with a nonfunctioning upper pole and an ectopic ureter. The approach was similar to case 2, with an operative time of 180 min and an estimated blood loss of 10 cc.

Case 4
Case 4 was a 24-month-old baby girl with recurrent urinary tract infections. Renal sonography, voiding cystourethrography, cystoscopy with retrograde pyelography, and nuclear renal scan revealed a left duplex system with a nonfunctioning upper pole and an ectopic ureter. Operative technique was once again similar to case 2. Operative time and blood loss were 120 min and 10 cc, respectively. Clinical data from the above cases is summarized in Table 1. There was no morbidity or mortality. No intraabdominal drains were used. Hospital stays ranged from 1 to 3 days. Postoperative pain was minimal and patients required analgesics only for the first 24 h. All were given meperidine intramuscularly: 75 mg for the adults (five doses for one patient and six for the other), and 5 mg once for the pediatric patients. All were started on liquid diet on postoperative day 1 and were tolerating regular diet by the day of discharge. On follow-up ranging from 3 to 6 months, all were doing well and had normal BUN and creatinine levels, as well as negative urine cultures.

References
1. Clayman RV, Kavoussi LR, Soper NJ, Dierks SM, Meretyk S, Darcy MD, Roomer FD, Pingleton ED, Thomas PG, Long SR (1991) Laparoscopic nephrectomy: initial case report. J Urol 146: 278282 2. Hanney RM, Alle KM, Cregan PC (1995) Major vascular injury and laparoscopy. Aust N Z J Surg 65: 533535 3. Jordan GH, Winslow BH (1993) Laparoendoscopic upper pole partial nephrectomy with ureterectomy. J Urol 150: 940943 4. McDougall EM, Clayman RV, Elashry O (1995) Laparoscopic nephro-

695 ureterectomy for upper tract transitional cell cancer: the Washington University experience. J Urol 154: 975979 5. Patterson M, Walters D, Browder W (1993) Postoperative bowel obstruction following laparoscopic surgery. Am Surg 59: 656657 6. Rassweiler JJ, Henkel TO, Potempa DM, Coptcoat M, Alken P (1993) The technique of transperitoneal laparoscopic nephrectomy, adrenalectomy, nephroureterectomy. Eur Urol 23: 425430 7. Rassweiler JJ, Henkel TO, Stoch C, Greschner M, Becker P, Preminger GM, Schulman CC, Frede T, Alken P (1994) Retroperitoneal laparoscopic nephrectomy and other procedures in the upper retroperitoneum using a balloon dissection technique. Eur Urol 25: 229236 8. Stoller ML, Irby PB III, Carrol PR, Osman M (1992) Laparoscopic renal cyst resection. J Endourol 6: S56 9. Suzuki K, Ihara H, Kurita Y, Kageyama S, Ueda D, Usiyama T, Ohtawara Y, Kawabe K (1993) Laparoscopic nephrectomy for atrophic kidney associated with ectopic ureter in a child. Eur Urol 23: 463465

Letters to the editor

Surg Endosc (1997) 11: 696

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The totally extraperitoneal laparoscopic hernia repair

We read with interest the paper by Vanclooster and colleagues [11] and commend their contribution to this procedure. However, we would offer three comments: First, totally extraperitoneal laparoscopic hernia repair was developed by Dulucq [13] in 1989/90 and by McKernan [7] not a great deal later. While it seems legitimate for others to publish their own technical variations, developments, and outcomes, we strongly support recent reminders [6] that journal editors and their peer referees owe the reader a duty of diligence: they should insist that authors exercise proper scholarship by giving credit where it is due. Otherwise the uninformed reader may assume originality and the informed may infer plagiarism, where the author intended neither. Second, the mesh configuration suggested by Vanclooster et al. was presented by one of us several years ago [4, 10]. However, the concept of amputating the inferior and lateral corner so the mesh fits better on the iliac vessels and the psoas muscle is flawed, since it has subsequently been reported [12] that recurrences may occur dorsal/ inferior to this inferolateral corner. The most extensive possible coverage of the psoas muscle belly is therefore appropriate. Third, the need for mesh fixation remains debatable. However, to fix the cranial border to prevent early migration or slipping is illogical: In our joint experience of over 1,000 cases and, to our knowledge in all reports in the world literature, recurrences pass uniformly caudal to the inferior border of the prosthesis. Fixation of the inferior medial part of the mesh to Astley Coopers ligament alone [5] may not offend against the original tension-free notion of Stoppa [8, 9] nor interfere with the mechanics of prosthesis retention. To fix the superior border to points that move relative to one another within a musculofascial structure contravenes both principles. Finally, on a minor point, if the structure annotated as D in Fig. 1 is the testicular vascular bundle, where is the vas deferens? Despite these comments we congratulate the authors on their low complication rate. References
1. Dulucq J-L (1991) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritone ale en retrope ritone oscopie. Cah Chir 79: 1516 2. Dulucq J-L (1992) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritoneal en pre -pe ritoneoscopie. Chirurgie 118(12): 8385 3. Dulucq J-L (1992) The treatment of inguinal hernias by implantation of mesh through retroperitoneoscopy. Postgrad Gen Surg 4: 173174 4. Fiennes AGTW, Taylor RS (1994) Learning laparoscopic hernia repair. In: Inguinal hernia, advances or controversies? Arregui M, Nagan R (eds) Radcliffe, Oxford, pp 475482 5. Fiennes AGTW, Himpens J, Dulucq J-L (1994) Preperitoneoscopic groin hernioplasty, current synthesis. Surg Endosc 8(8): 989 6. Horton R, Smith R (1996) Time to redefine authorship (editorial). Br Med J 312: 723 7. McKernan JB (1993) Laparoscopic extraperitoneal repair of inguinofemoral herniation. Endosc Surg Allied Tech 1(4): 198203 8. Stoppa R, Petit J, Abourachid H (1973) Proce de original de plastie des hernies de laine. Linterposition sans fixation dune prothe ` se en tulle de Dacron par voie me diane pre pe ritone ale. Chirurgie 99: 119 9. Stoppa RE, Rives JL, Warlaumont CR (1984) The use of Dacron in the repair of hernias of the groin. Surg Clin North Am 64: 269285 10. Taylor RS, Fiennes AGTW (1992) A tension free modification of the Dulucq preperitoneal laparoscopic hernioplasty. Min Inv Ther 1(Suppl 1): 101 11. Vanclooster P, Meersmann AL, de Gheldere CA, Van de Ven CK (1996) The totally extraperitoneal laparoscopic hernia repair. Surg Endosc 10: 332335 12. Vivekanadan S, Fiennes AGTW (1995) Totally extraperitoneal groin hernioplasty: mechanism of delayed indirect recurrence. Min Inv Ther 4(Suppl 1): 55

A. Fiennes
Department of Surgery St Georges Hospital Medical School Cranmer Terrace London, SW17 ORE, United Kingdom

J. Himpens
Department of Digestive Surgery University Hospital Ste Pierre Rue Haute 201 B-1000 Brussels, Belgium

Surg Endosc (1997) 11: 697

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank you for the opportunity to answer the letter of Mr. Fiennes. First of all, we certainly do acknowledge that Mr. Dulucq and Mr. Mc Kernan were the pioneers of the extraperitoneal laparoscopic hernia repair. Honor to whom honor is due. We did not intend to pretend to be the pioneers of this technique. We wished only to describe the technique the way we perform it, to describe our own findings, and to give a fair report of our preliminary results. The reason for cutting the inferolateral corner is not just that we think it fits better on the iliopsoas but also because we are afraid to cause damage to the nerves running on it by dissecting unnecessarily high on the muscle. Since the mesh measures 15 15 cm, we do not think we compromise the strength of the repair by merely removing a small piece of its inferolateral corner. We think that dissecting very high on the muscle just to position the whole inferolateral corner of the mesh flat on the muscle is unnecessary and dangerous. We do agree totally that fixation of the mesh is unnecessary provided the mesh is large enough, which is obviously the case when using a 15 15 cm mesh. In fact, we have not fixed the mesh since January 1996. We also agree that the vas deferens is not clearly seen on Fig. 1. We chose this shot because of the clearly visible large direct defect.

C. de Gheldere
Heilig Hart Ziekenhuis Kolveniersvest 20B-2500 LIER Belgium

P. Vanclooster
Bouwelsesteenweg 6 2560 Nijlen Belgium

Surg Endosc (1997) 11: 698

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The influence of pneumoperitoneum on the peritoneal implantation of free intraperitoneal cancer cells
Recently Hubens et al. published an interesting article entitled The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells [2]. They reported on the possible implantation of cancer cells at trocar wounds at the moment of deflation as cells are forced through these wounds by the pressure gradient created by the pneumoperitoneum during laparoscopic surgery for malignant disease. We would like to point out that this chimney effect, as originally described by us, can occur during the entire laparoscopic procedure and not only at the moment of deflation, as leakage of CO2 alongside trocars during surgery is impossible to prevent with the existing trocars [3]. Consequently, deflation of the pneumoperitoneum by letting CO2 escape through one of the trocars before pulling these trocars out of the abdomen will not prevent the occurrence of entrapment of cancer cells in the trocar wounds. We fully agree with the authors on the possible advantage of gasless laparoscopy as this could prevent the chimney effect. In our experimental work we found significantly less tumor growth at the port sites following gasless laparoscopic surgery for colon cancer in the rat as compared to laparoscopic surgery using a CO2 pneumoperitoneum. This technique seems promising to treat malignant disease laparoscopically.
Correspondence to: G. Kazemier

References
1. Bouvy ND, Marquet RL, Jeekel J, Bonjer HJ (1996) Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Surg Endosc 10: 551 2. Hubens G, Pauwels M, Hubens A, Vermeulen P, Van Marck E, Eyskens E (1996) The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells. Surg Endosc 10: 809812 3. Kazemier G, Bonjer HJ, Berends FJ, Lange JF (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 11411142

G. Kazemier1 F. J. Berends1 N. D. Bouvy1 J. F. Lange2 H. J. Bonjer1

1

Department of Surgery University Hospital Rotterdam-Dijkzigt Dr Molewaterplein 40 3015 GD, Rotterdam The Netherlands 2 Department of Surgery St. Clara Hospital Rotterdam The Netherlands

Surg Endosc (1997) 11: 699

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank Dr. Kazemier et al. for their kind remarks and fully agree with them that gas leakage can occur during the entire procedure with subsequent implantation of tumor cells at the trocar sites. At the moment we are conducting further experimental studies on the possible effects of gas leakage on tumor cell implantation and the chimney effect, as they have called it. Results will be ready for publication soon. G. Hubens
Department of Surgery University Hospital University of Antwerp Wilrijkstraat 10 2650 Edegem Belgium

Technique
Surg Endosc (1997) 11: 684686

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic-assisted abdominoperineal resection in the prone position

An alternative technique
K. Koerner, S. Datena, L. Erwin
Emanuel Hospital, 2801 N. Gantenbein Avenue, Portland, OR 97227, USA Received: 23 October 1995/Accepted: 5 August 1996

Abstract. With the introduction of laparoscopic-assisted abdominoperineal resection (LAPR), the traditional LloydDavies position with the Mayo two-team combined approach is being adapted. The Lloyd-Davies position allows two teams of surgeons to work simultaneously, minimizing operating time. The conditions required for laparoscopy restrict a simultaneous procedure. Since LAPR is typically performed as a two-stage procedure, we introduce an alternative position which facilitates the perineal dissection. We review the results and technique of LAPR in the prone position in three patients who were suitable candidates for this procedure. Three patients underwent LAPR. No operative or postoperative complications were encountered and the procedures were in keeping with oncologic principles of resection. Total anesthesia times were less than 3.5 h for these initial patients. No hemodynamic problems were encountered due to the choice of patient positioning. The prone jackknife position greatly increases visualization of deep structures, reduces blood loss, enhances dissection, and reduces the technical demands of the laparoscopic portion of the procedure. Key words: Laparoscopic abdominoperineal resection Prone jackknife position Alternative technique

for a patient with benign disease. LAPR in keeping with oncologic surgical principles was addressed by Descanini et al. [3]. They performed LAPR on a series of 11 fresh cadavers and subsequently performed autopsies to determine the extent of the resection. They found that their recommended technique of LAPR with proximal ligation of the inferior mesenteric artery, wide clearance of pelvic side walls, and complete removal of the mesorectum could be performed for oncologic procedures. In each of the above cases, the patients have been positioned in either the perineolithotomy position or position was not discussed. The Lloyd-Davies position was originally recommended based on the two-team approach with one team conducting the abdominal resection while the other carried out the perineal dissection. With the introduction of LAPR this approach is being reconsidered. The O.R. conditions required for laparoscopy restrict a simultaneous direct procedure. In our experience, LAPR is more readily conducted as a twostage procedure. We therefore suggest an alternative position for the perineal dissection in order to facilitate dissection. We have found that the prone jackknife position greatly increases visualization of deep structures, reduces blood loss, and enhances dissection.

Case reports Goals of minimally invasive surgery include reducing operative time and minimizing the technical demands made upon the surgeon. In performing traditional abdominoperineal resection (APR) with the patient in perineolithotomy position, the majority of difficult dissection occurs through the abdomen for access to the lateral pedicles along the rectal wall. Identification of the ureters, especially the left, deep in the pelvis, can also be problematic. Sackier [11] reported a technique for laparoscopic APR (LAPR) in 1992
Patient 1 is a 40-year-old black male with a 1-year history of increasing rectal bleeding, pain, and decreasing caliber of stool. Digital exam revealed a mass 2.5 cm from the anal verge. Rigid sigmoidoscopy performed to 20 cm revealed similar findings with approximately two-thirds obstruction of the lumen. Biopsy confirmed moderately differentiated adenocarcinoma. CT of the abdomen and pelvis visualized a mass lesion within the rectum without evidence of perirectal invasion, extension, pelvic adenopathy, or metastases. The patient underwent LAPR with estimated blood loss less than 75 cc. O.R. time was 205 min. G.I. continuity returned on the 4th postoperative day. His recovery has been uneventful. Patient 2 is a 42-year-old black male who had a mass 1.5 cm above the dentate line. Rigid sigmoidoscopy performed to 20 cm revealed similar findings. CT revealed no evidence of perirectal invasion, extension, pelvic adenopathy, or metastases. He underwent LAPR with estimated blood loss of 50 cc. The specimen contained 12 nodes none of which were positive.

Correspondence to: S. Datena

685 O.R. time was 195 min. G.I. function returned on postoperative day 3. His recovery has been uneventful. Patient 3 is a 58-year-old white male who had a fixed mass 2 cm above the dentate line. This patient received preoperative irradiation with a reduction in the size of the mass. LAPR was performed with a total estimated blood loss of 50 cc. No positive nodes were found in the specimen. The patient was discharged home on postoperative day 3. He received postop chemotherapy and irradiation. At 6 months out there was no evidence of recurrence.

Technique The patient is initially placed in the prone position with a slight degree of jackknife and with the head down to optimize hemodynamics. The perineum is incised and dissection commences. Using gentle traction the rectum can be manipulated easily to facilitate dissection. Significant vessels are identified and ligated under direct vision. There is less tissue staining with blood as compared to lithotomy position, which facilitates more careful dissection. With gentle traction on the rectum, the difficult plane between the prostate and anterior rectum can be carefully dissected. Downward traction facilitates an excellent view of the plane of dissection outside the presacral fascia, thereby avoiding the presacral venous plexus. The dissection is carried out up to and slightly beyond the peritoneal reflection. With the specimen freed to this level, it is then placed into a thick specimen bag and tied with umbilical tape, which is sutured into place to prevent slipping. The bag and its contents are then tucked up into the pelvis. A single drain can be placed into the pelvis and brought out through a separate stab incision. The perineum is closed in layers and dressed. The patient is then turned to the supine position for the laparoscopic portion of the procedure. Since the majority of the difficult dissection has already been performed, a minimum of access ports will be required. We used gasless laparoscopic technique. Our standard approach for all colon resections is to use a maximum of four ports. A 12-mm port is placed at the umbilicus, followed by a second 12-mm port through the previously marked colostomy site. Two 5-mm ports or one 5-mm and one 10-mm port are placed on the right side of the abdomen for dissection and traction on the colon. This minimum number of ports reduces the morbidity for the patient postoperatively. Placing the largest port through the planned ostomy site is also common practice to reduce morbidity. By using two 12-mm ports the operator has the option to use stapling devices through one of two ports should one provide a more advantageous angle of attack over the other. If it has not already been done, the peritoneal reflection is incised and the mesentery of the rectosigmoid is taken down. The inferior mesenteric artery is ligated high, leaving a 1520-mm stump. Although this has not been shown to improve survival in patients with nodal involvement [11], it does aid removal of the rectosigmoid lymphatic drainage for lymph node sampling [3]. The inferior mesenteric vein is also ligated. The pelvis is reperitonealized using hernia staplers. Vicryl mesh screen can also be placed in the pelvis with the hernia stapler if the patient is to receive postoperative irradiation. A pretied Roeder loop is passed around a small corner of the distal end of the sigmoid colon. This is brought up to the planned ostomy site. The umbilical tape attached to the

specimen bag is then brought out through the ostomy site port. The abdomen is then fully examined for hemostasis before preparing the colostomy site. The bagged specimen is then delivered through the ostomy site and sent immediately to pathology to determine whether an adequate mesenteric specimen has been obtained. The colonic segment for the ostomy is brought up with the aid of the pretied Roeder loop. The abdomen is then reexamined in order to verify the colon has not been torqued in any manner prior to maturing the colostomy. The colostomy is then matured. All ports are removed and wounds closed.

Discussion The significantly improved visualization obtained in the prone jackknife position appears to contribute to decreased blood loss. Total blood loss in each of our cases was 75 cc or less (excluding blood contained within the specimen). This reduction in operative blood loss over conventional technique offers an advantage to patients with rectal cancer. Postoperative infection rates have been shown to be increased in colorectal cancer patients who have received blood transfusions [4]. Additionally, continued debate occurs as to the relationship of blood transfusion to colorectal cancer recurrence [1, 10, 12, 13, 14]. Currently, it appears that a noncausal relationship exists. In any event, most would agree that limiting operative blood loss is preferred. Hatada et al. [7] examined hemodynamics in the prone jackknife position in 19 patients and found no significant adverse effect. In addition, they report an approximate 50% decrease in mean blood loss over lithotomy position based on their previous work. We elected to begin with the perineal portion of the procedure. Corman recommends that the abdominal portion should be performed initially [2]. The principles behind initially performing the abdominal portion center around minimizing morbidity and ease of dissection. Of primary concern is the need to perform adequate resection. Our preoperative evaluation in each patient established that the tumor was of suitable size, location, and mobility to allow for reverse order. Appropriate workup and clinical judgment will determine which patients are suitable for this technique. We believe this sequence permits ease of dissection. Total operating time was less than 3.5 h in each case. The procedure can just as easily be performed, with slight modification, by completing the laparoscopic portion initially. The laparoscopic portion would then be carried out using techniques such as that described by Descanini [3]. The abdominal specimen is then placed in a specimen bag and buried in the pelvic cavity in the classic Miles method. The patient would then be turned to the prone jackknife position for the remainder of the procedure. Laparoscopic APR can be performed in keeping with oncologic principles [3]. Our specimens had clear margins and all had negative lymph nodes with adequate sampling. Abdominal wall recurrence at a trocar site has been reported [5, 8]. In one case, the trocar site where the recurrence occurred was used only for dissection and did not come in contact with the specimen. In a review by Hughes et al., tumor recurrence in abdominal wall scar tissue following open large-bowel cancer surgery was examined [8]. They

686

suggested that while involvement of the scar tissue was uncommon, in those cases where it did occur it was more likely to be a manifestation of incurable systemic malignant disease as opposed to an isolated occurrence. Regardless of the technique, the prognosis appears poor if abdominal scar recurrence is found [8]. To date, no definitive rate of abdominal wall recurrence following laparoscopic largebowel cancer surgery has been established. Nduka et al. in 1994 were able to identify 18 reported cases of cannula-site recurrence following laparoscopic oncologic procedures [9]. All of these cases involved adenocarcinoma. If upon further review the recurrence rate is significantly higher with laparoscopy, then its use will need to be modified or curtailed as indicated. With appropriate postoperative follow-up these rare recurrences should be identified and treated. Geis et al. [6] considered LAPR to be a moderately difficult procedure according to their scale, in part because of the necessity to mobilize deep in the pelvis. Performing the majority of the difficult dissection with the patient in the prone position decreases the technical proficiency required to complete the laparoscopic portion of the surgery. A surgeon with moderate experience in advanced laparoscopic technique should be able to easily perform LAPR with this method. Conclusion Laparoscopic-assisted abdominoperineal resection is becoming a common technique in the treatment of benign and malignant tumors. While LAPR is not appropriate in all cases, preoperative evaluation and clinical judgment will determine which procedure should be attempted. Physical examination, endoscopy, biopsy, CT, and transrectal ultrasound can be used in evaluating the patient and planning the appropriate procedure. Our experience has found that blood

loss is considerably decreased and dissection is greatly facilitated by performing the perineal portion of LAPR with the patient in the prone jackknife position. References
1. Busch ORC, Hop WCJ, Marquet RL, Jeekel J (1994) Blood transfusions and local tumor recurrence in colorectal cancer: evidence of a noncausal relationship. Ann Surg 220: 791797 2. Corman M (1993) Colon and rectal surgery. 3rd ed. J. B. Lippincott, Philadelphia 3. Descanini WITH, Milsom JW, Bo hm B, Fazio VW (1994) Laparoscopic oncologic abdominoperineal resection. Dis Colon Rectum 37: 552558 4. Ford CD, VanMoorleghem G, Menlove RL (1993) Blood transfusions and postoperative wound infection. Surgery 113: 603607 5. Fusco MA, Paluzzi MW (1993) Abdominal wall recurrence after laparoscopic-assisted colectomy for adenocarcinoma of the colon: report of a case. Dis Colon Rectum 36: 858861 6. Geis WP, Colletta AV, Verdeja JC, Plasencia G, Ojogho O, Jacobs M (1994) Sequential psychomotor skills development in laparoscopic colon surgery. Arch Surg 129: 206212 7. Hatada T, Kusunoki M, Sakiyama T, Sakanoue Y, Yamamura T, Okutani R, Kono K, Ishida H, Utsunomiya J (1991) Hemodynamics in the prone jackknife position during surgery. Am J Surg 162: 5558 8. Hughes ESR, McDermott FT, Polglase AL, Johnson WR (1983) Tumor recurrence in the abdominal wall scar tissue after large-bowel cancer surgery. Dis Colon Rectum 26: 571572 9. Nduka CC, Monson JRT, Menzies-Gow N, Darzi A (1994) Abdominal wall metastases following laparoscopy. Br J Surg 81: 648652 10. Rehman SU (1993) Laparoscopic abdominoperineal resection of the rectum [letter; comment]. Br J Surg 80: 1080 11. Sackier JM, Berci G, Hiatt JR, Hartunian S (1992) Laparoscopic abdominoperineal resection of the rectum. Br J Surg 79: 12071208 12. Sibbering DM, Locker AP, Hardcastle JD, Chir M, Armitage NC (1994) Blood transfusion and survival in colorectal cancer. Dis Colon Rectum 37: 358363 13. Tang R, Wang JY, Chien CRC, Chen JS, Lin SE, Fan HA (1993) The association between perioperative blood transfusion and survival of patients with colorectal cancer. Cancer 72: 341348 14. Tartter PI (1992) The association of perioperative blood transfusion with colorectal cancer recurrence. Ann Surg 216: 633638

Surg Endosc (1997) 11: 636638

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic cholecystectomy in the elderly

F. M. Tagle,1 J. Lavergne,1 J. S. Barkin,1 S. W. Unger2
1 2

Division of Gastroenterology, Mt. Sinai Medical Center, University of Miami, School of Medicine, 4300 Alton Road, Miami Beach, FL 33140, USA Department of Surgery, Mt. Sinai Medical Center, University of Miami, School of Medicine, 4300 Alton Road, Miami Beach, FL 33140, USA

Received: 17 September 1996/Accepted: 14 October 1996

Abstract Background: Advanced age with its concomitant comorbid conditions may be associated with increased postoperative laparoscsopic cholecystectomy (LC) complications and more frequent conversion to open cholecystectomy (OC). The purpose of this study was to evaluate the outcome of LC in patients age 65 and older. Methods: Ninety consecutive patients were studied age 65 and older, of whom 39 (43%) were males and 51 (57%) were females, mean age 74 years (range 6598), with 20 patients (22%) 80. Indications for surgery included biliary colic 55 (61%), acute cholecystitis 22 (24%), pancreatitis 10 (11%), and cholangitis 3 (4%). Seventeen patients (19%) had preoperative ERCP, 12 of which were normal; five had sphincterotomy with stone extraction. Comorbid conditions included hypertension (44%), CAD (17%), cardiac arrhythmias (18), CHF (9%), and COPD (7%). Results: Operative timemean 1 h 51 min SD 43 min. Conversion to OCthree patients (3%). Length of stay mean 5 days (range 126). Mortalitytwo patients (2%) >80 years old, one patient with septicemia and multiorgan failure whose comorbid diseases included CAD, C.F., COPPED, and elevated BP, one patient with MI postsurgery, morbid diseases included DM and CAD. Complicationsfive patients (5%): bile leak from cystic duct stump (one), postsurgery MI (two), incarcerated incisional hernia (one), septicemia (one). Conclusion: Morbidity rates for LC in the elderly population are not different from that reported for patients less than 65 years of age. (5% vs 6%, Fried et al., Surg Clin North Am 1994;74 [2]: 375387). Our 2% mortality rate is statistically different from previously reported in a series of patients of all ages (0.6%, Fried et al.). The 3% rate of conversion to OC in this older population is not significantly different from the patients in Fried et al. series (4%). Key words: Elderly Conversion Comorbid condition

Laparoscopic cholecystectomy (LC) is currently the procedure of choice for managing gallstone disease [2, 3]. Overall, its mortality rate is the same as that for open cholecystectomy but its morbidity is less [2, 3, 4, 6]. Advanced age with its concomitant comorbid conditions has been felt to be associated with increased postoperative complications [6], as well as more frequent conversion to open cholecystectomy (OC). With the increasing mean age of our population, it is important to know the morbidity and mortality of patients >65 years of age who undergo LC. Therefore, the purpose of this study was to evaluate the outcome of LC in patients age 65 and older. Patients and methods
We conducted a retrospective study evaluating the medical records of 90 consecutive patients age 65 or older who underwent LC from January 1990 to June 1994 at Mount Sinai Medical Center, Miami, FL. This represents 24% of the total number of patients who had LC during that period. The following parameters were evaluated: age, sex, indication for surgery, comorbid disease states, operative time, conversion to OC, complications, performance of ERCP, length of hospital stay, and mortality.

Results Ninety patients with a mean age of 74.3 7 years (range 65 to 98 years) were evaluated. Twenty (22%) patients were age 80 or older. Thirty-nine (43%) were males and 51 (57%) were females. The indications for surgery included biliary colic in 55 (61%) patients, acute cholecystitis in 22 (24%), acute pancreatitis in ten (11%), and cholangitis in three (4%). Comorbid conditions included hypertension in 40 of 90 patients (44%), coronary artery disease in 15 of 90 (17%), cardiac arrhythmias in 14 of 90 (18%), congestive heart failure in eight of 90 (9%), and COPD in six of 90 (7%). Two patients had cirrhosis incidentally discovered intraoperatively and one patient had a cerebral meningioma (Table 1). The mean duration of surgery was 111 min 43 min (range 50 to 255 min). Conversion to OC was required in three of 90 patients

Correspondence to: J. S. Barkin

637 Table 1. Comorbid diseases No. patients 6579 years Hypertension Coronary artery disease Cardiac arrhythmias Congestive heart failure COPD Cirrhosis Cerebral meningioma Diabetes 31 14 8 3 4 2 1 3 No. patients 80 years 9 1 6 5 2 4 Total no. patients (N 90) 40 (44%) 15 (17%) 14 (18%) 8 (9%) 6 (7%) 2 (2%) 1 (1%) 7 (8%) No. patients (%) No. deaths (%) Table 3. Mortality in patients undergoing LC divided by age groups 6579 years old 70 (78) 0 (0) 80 years old 20 (22) 2 (10)

Table 2. Postoperative complications of laparoscopic cholecystectomy No. patients Directly related to the surgical procedure Bile leak Incisional hernias Medical Myocardial infarction Sepsis 1 1 2 1

(3%) due to their distorted anatomy. Postoperative complications occurred in five patients (Table 2). Two of the five complications were related directly to the surgical procedure itself, including one patient with a postoperative cystic duct leak which was treated successfully with an ERCP and sphincterotomy on postoperative day 2. One patient had an incarcerated umbilical hernia at the incision site requiring decompressive surgery. Medical complications that were not directly related to the procedure itself included two patients with postoperative myocardial infarction (MI). The mean postoperative hospital stay was 3 days (range 126). Mortality occurred in two of 90 patients (2%), both of whom were older than 80 years (Table 3). One patient had a postoperative MI complicated by fatal arrhythmias. The second patient with comorbid diseases of congestive heart failure and COPD was admitted with cholangitis and died of sepsis and multiorgan failure on postoperative day 21. ERCP was performed preoperatively in 17 patients; 12 had normal studies and five had common bile duct stones that required sphincterotomy and stone extraction. Three patients had postoperative ERCP, two of them for suspected retained common bile duct stones (normal studies) and one for the treatment of a bile leak from the cystic duct stump. Discussion The incidence of complications with OC increases with age [2]. The reported complication rates in patients >65 years undergoing open cholecystectomy range between 20% and 30% [4, 6]. Currently, LC has replaced open cholecystectomy as the preferred approach for the treatment of symptomatic cholelithiasis due to its lower morbidity, shorter length of hospital stay, and lower cost compared to open cholecystectomy [12]. However, the morbidity and mortality of patients >65 years of age who undergo LC have been only scantly reported.

The majority of our patients underwent LC for symptoms of biliary colic (55%). This is similar to that reported by Fried et al. [2], who also found a higher incidence of acute cholecystitis and pancreatitis in older patients as an indication for surgery compared to those patients below age 65. Patients >65 years would be expected to have an increased incidence of comorbidity. We found the expected high incidence, with approximately half of our patients having CAD, 20% with cardiac arrhythmias and 10% with CHF. The comorbid conditions that were present in our study population are comparable to those reported by other authors evaluating LC in the elderly [7, 11, 12] in which hypertension and CAD are the most commonly encountered comorbid disease. We found a 24% incidence of acute cholecystitis, and 11% of our patients had acute pancreatitis. This falls between the reported range in patients > age 65 who have rates of 8% to 32% of acute cholecystitis incidence and is also in accordance with Fried, who had 22% for acute cholecystitis and 11% for acute pancreatitis. Our mean operative time of 111 min falls within the range reported in a large compilation of series by Godacz of between 54 and 138 min. It is difficult to compare the incidence of postoperative complications between series due to the lack of standardization of reporting as well as the retrospective nature of our study. Nevertheless, our 5% complication rate in patients >65 years is comparable to the 5% to 10% published in other series [2, 12]. Other series reported an overall complication rate of 24% in patients >65 years old undergoing LC [8]. However, 9% were directly related to the surgical procedure itself, including biliary tract injury, wound infection, operative bleeding, and postprocedure cholangitis. We found no wound infections in our series. Fried et al. [2] found that wound complications were the most common problem overall and were found more frequently in the elderly (1.3%) than in younger patients. The reported incidence of biliary tract injury ranges from 0.3 to 1% [3, 5, 8, 10]. LC is associated with a two- to tenfold increase in bile duct injury when compared with conventional cholecystectomy, probably because of inadequate dissection, inherent limitations to the technique, and/ or the learning curve associated with a new technology [1]. None of our patients had bile duct or vessel injuries; however, the surgeon was highly experienced. Only one of our patients had a postoperative bile leak from the cystic duct stump requiring ERCP with sphincterotomy and stent placement. ERCP was performed prior to surgery in 17 of our patients with clinical and biochemical suspicion of choledocholithiasis, only five of whom had CBD stones that were extracted endoscopically. The issue of ERCP and its timing related to LC have been a subject of intense debate in the past few years. Only 5% of patients considered low risk for common bile duct stones actually have stones on intraoperative cholangiogram [9]. Our data agrees that pre-LC

638

ERCP should be reserved only for patients with signs and symptoms suggestive of common bile duct pathology. Conversion to OC was performed in 3% of our patients. This rate is lower than in previously reported series [2, 3, 10, 12]. Fried et al. [2] found a global conversion rate of 5.4% and, interestingly, reported that the frequency of conversion to OC was significantly higher in the patients older than age 65 (10.4% vs 4.1%). They felt that this reflected the increased likelihood of complicated gallstone disease, difficulty in resolving the anatomy in older patients, and increased frequency of adhesions. Our lower rate probably reflects the experience and technical expertise of our surgical colleagues. The overall mortality rates after LC regardless of age range from zero to 0.9% in U.S. series [2, 3] and from 0.1 to 0.2% in European series [10]. In Fried series [2], the two deaths occurred in patients age 65 and older but less than age 80. Two deaths of our 90 patients occurred in the patients older than age 80, both of whom had severe comorbid disease. Our overall mortality of 3% is statistically higher than previously reported, which seems to be due to the inclusion of patients >80 years who had extensive comorbid disease. Interestingly, though, our subgroup of patients ages 6579 had significant comorbidity as well (Table 2) and their mortality was 0%. Conversely, considering only the 20 patients age 80 and older, the mortality in that group is 10% (Table 3). Our data suggest that patients >80 years who have multiple comorbid conditions have a worse prognosis than those below 80 years. In conclusion, age up to 80 years, even in patients with extensive comorbid disease, does not seem to be a poor

prognostic factor. These patients should not be excluded from surgical intervention. However, >80 years with extensive comorbid disease seems to be a negative prognostic factor in patients undergoing LC. References
1. Davidoff AM, Pappas TN, Murray EA, Hilleren DJ, Johnson RD, Baker ME, Newman GE, Cotton PB, Meyers WC (1992) Mechanisms of major biliary injury during laparoscopic cholecystectomy. Ann Surg 215: 196202 2. Fried GM, Clas D, Meakins JL (1994) Minimally invasive surgery in the elderly patient. Surg Clin North Am 74(2): 375386 3. Gadacz TR (1990) US experience with laparoscopic cholecystectomy. Am J Surg 165: 450454 4. Ibach JR Jr, Hyme HA, Erb WH (1968) Cholecystectomy in the aged. Surg Gynecol Obstet 126: 523528 5. Kozarek RA, Gannan R, Baerg R, Wagonfeld J, Ball T (1992) Bile leak after laparoscopic cholecystectomy. Diagnostic and therapeutic applications of ERCP. Arch Int Med 152: 10401042 6. Margiotta SJ, Horvitz JR, Willis IH, Wallack MK (1988) Cholecystectomy in the elderly. Am J Surg 156: 509512 7. Massie MT, Massie LB, Marrangoni AG, DAmico FJ, Sell HW (1993) Advantages of LC in the elderly and in patients with high ASA classifications. J Laparoendosc Surg 3(5): 467475 8. Nennex RP, Impenato PJ, Alcorn CM (1993) Complications of LC in a geriatric population group. NY State J Med 92(12): 518520 9. Neoptolemos JP, Shaw DS, Carr-Locke DL (1989) A multi variate analysis of preoperative risk factors in patients with common bile duct stones: implications for treatment. Ann Surg 209: 157161 10. Perissat K (1993) Laparoscopic cholecystectomy. The European experience. Am J Surg 165: 444449 11. Saxe A, Lavson J, Phillips E (1993) Laparoscopic cholecystectomy in patients aged 65 or older. J Laparoendosc Surg 3(3): 215219 12. The Southern Surgeons Club (1991) A prospective analysis of 1518 laparoscopic cholecystectomies. N Engl J Med 324: 10731078

Surg Endosc (1997) 11: 676678

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Palliative gastrojejunostomy
A minimally invasive approach
R. S. Chung,1 P. Li2
1 2

Department of Surgery, Meridia Huron Hospital, 13951 Terrace Road, Cleveland OH 44112, USA Hillcrest Hospital, 6780 Mayfield Road, Mayfield Heights, OH 44124, USA

Received: 28 June 1996/Accepted: 26 July 1996

Abstract. Palliative bypass for neoplastic gastric outlet obstruction should be minimally invasive. We designed a laparoscopically assisted approach that appears to meet the need. The proximal jejunum is exteriorized by laparoscopy via an epigastric trocar-site incision. An EEA anvil is installed in the exteriorized jejunum, which is returned to the abdomen. Through this mini-incision, the anterior wall of the stomach is opened for insertion of the EEA stapler, which penetrates the posterior gastric wall. When snapped to the anvil and fired, an antecolic gastrojejunostomy is created. No mortality or anastomotic leak occurred in two cases. The operation and recovery appeared to be faster than historic controls. This operation is minimally invasive and expeditious, ideal for patients requiring palliative bypass. Key words: Minimally invasive Palliative gastrojejunostomy Gastric outlet obstruction

(retrieval loop) is threaded into the hole in the shaft of the anvil (Fig. 2A), which, together with the jejunum, is returned into the abdomen, leaving only the retrieval loop on the exterior (Fig. 2B). This allows a clear exposure of the gastric antrum. Between stay sutures, the anterior gastric wall is opened with a small incision for introduction of the EEA stapler (Fig. 3A), the sharp tip of which penetrates the posterior wall. The stomach in this location has been cleared of the omentum and gastroepiploic arcade for a short distance to accommodate the jejunum. Aided by the retrieval loop, the anvil is handguided and snapped to the stapler. The correct approximation of the stomach and jejunum is checked visually (Fig. 3B). Firing of the stapler effects an antecolic, retrogastric gastrojejunostomy. The anastomosis is inspected

Palliative bypass for inoperable malignant gastric outlet obstruction should be expeditious and minimally invasive, in view of limited life expectancy. The following technique, using a mini-incision, was first conceived for a patient who developed gastric outlet obstruction 2 years after a pylorussparing Whipple procedure.

Surgical technique
Under general anesthesia, laparoscopy is first performed through a subumbilical port. A 10-mm port is inserted at the epigastrium and a laparoscopic bowel clamp is used to secure the proximal jejunum loop suitable for the anastomosis. By cutting down on the cannula (Fig. 1), this port site is converted into a 4-cm incision through which the jejunum is exteriorized. The anvil of a 30-mm EEA stapler (U.S. Surgical) is inserted into the bowel and secured by a pursestring suture (Fig. 2A). A strong suture

Correspondence to: S. Chung

Fig. 1. Locations of the ports used; conversion of the epigastric port site into a 4-cm incision. Note that the incision may be horizontal, vertical, or oblique depending on anatomical factors such as adhesions, size of stomach, etc.

677

Fig. 2. A The proximal jejunum exteriorized via this incision, and the anvil of the EEA stapler secured with a pursestring suture. Note the retrieval loop threaded through the hole at the tip of the anvil. B The jejunum has been returned to the abdomen to facilitate exposure of the stomach, leaving only the retrieval loop on the exterior.

Fig. 3. A The EEA stapler inserted into the stomach via an anterior gastrostomy. B The gastrojejunostomy staple line is inspected prior to removal of the stapler. The entire circumference can be viewed by moving the retractors.

internally for hemostasis, and the potency of the efferent and afferent limbs is verified by digital exploration. The gastrostomy is then closed with conventional suturing. A final laparoscopic view of the completed operation is obtained by reinflation of the abdomen after wound closure.

Case report
A 62-year-old black female presented with postcibal epigastric fullness, vomiting and weight loss of 1 month duration. Two years prior to presen-

tation, she was treated for a well-differentiated pancreatic carcinoma, staged pathologically T2N1M0, by a pylorus-sparing Whipple operation, followed by chemotherapy and irradiation. She had remained well until the onset of current symptoms. Physical examination showed emaciation, moderate amount of ascites, a palpable liver, and a positive succession sign. CT scan confirmed a large stomach and ascites but showed no liver metastasis. Ascites tap was negative for malignant cells. Endoscopy showed a patent end-to-side anastomosis of the duodenal cuff to the jejunum, but marked narrowing just distal to it. The afferent jejunum, which bears the bile duct and pancreatic anastomoses, was patent. Endoscopic biopsies of the stenosed areas failed to reveal neoplastic involvement.

678 The patient underwent stapled gastrojejunostomy by the above technique. The operation took 35 min and she was discharged on day 4. Symptomatic improvement was maintained and weight gain was recorded at follow-up 1 month after operation.

We have since used this method in another patient, suffering from inoperable distal gastric cancer, with equal success. This simple procedure has little or no learning curve and deserves to be used and confirmed by others.
Acknowledgment. This work was supported by departmental research funds.

Discussion Laparoscopic gastrojejunostomy has been reported for a similar indication using a totally intracorporeal technique for the anastomosis [1, 2]. While our technique does not claim to be laparoscopic, it also requires no special laparoscopic expertise, since the main operation is done in the open wound. The laparoscope facilitates a general examination and identification of the jejunum loop, enabling a mini-incision to be used. The operation is also much more expeditious than a totally laparoscopic one, which takes longer and requires four to five ports, probably equivalent to the mini-incision in length and discomfort.

References
1. Kum CK, Tap CH, Goh PM (1995) Palliation of advanced gastric cancer by laparoscopic gastrojejunostomy. Singapore Med J 36: 228 229 2. Rangraj MS, Mehta M, Zale G, Maffucci L, Herz B (1994) Laparoscopic gastrojejunostomy: a case presentation. J Laparoendosc Surg 4: 8187

Surg Endosc (1997) 11: 639642

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic Douglasectomy in the treatment of painful uterine retroversion

P. von Theobald, P. Barjot, G. Levy
Department of Gynecology, CHRU Caen, Avenue Georges Cle menceau, 14033 Caen Cedex, France Received: 22 April 1996/Accepted: 15 July 1996

Abstract Background: One of the etiologies of pelvic pain in women, often unrecognized, is the Masters-Allen syndrome, which was described in 1955 as the universal joint cervix syndrome. It has the following three elements: (1) etiology: obstetrics-related trauma; (2) clinical findings: uterine retroversion with hypermobile cervix following elongation or desinsertion of the uterosacral ligaments; (3) anatomy: visualization of a tearing of the posterior serosa and subperitoneal fascia of the ligamentum latum. Methods: Forty-one laparoscopic Douglasectomies with uterosacral ligamentopexy were performed in the department of Gynecology at the University Hospital of Caen during the period between 1990 and 1995 in patients with painful retroverted uterus. The patient selection was made thanks to the pessary test. The surgical endoscopic procedure, identical to the operation first promoted by Jamain and Letessier in 1976 by laparotomy, is described. Results: Total pain relief was experienced by 31 patients (75%) and partial relief by five patients (5%). Two main complications occurred, requiring one laparotomy (bleeding from a pelvic varicose vein with a concomitantly occurring breakdown of the washing-aspiration system) and one second laparoscopy at day 15 (one case of hematoma below the peritonization revealed by pain). Twenty-three women became pregnant again, and had normal deliveries except for two cesareans, with no recurrence of pain. Douglasectomy is compared to alternative techniques in the literature. Other indications for Douglasectomy are discussed. Conclusion: Douglasectomy is the only definitive procedure for restoring normal anatomy of the pelvic floor in case of painful uterine retroversion occurring in a setting of Masters-Allen syndrome. Additionally, it provides for pathological analysis of the excised peritoneum. The results of this procedure are excellent when the indication is correctly set, particularly as concerns positive pessary testing.

Key words: Laparoscopic Douglasectomy Uterine retroversion Pelvic floor

In 1976, Jamain, who had first promoted Douglasectomy, presented his 12-year experience with the surgical technique on 265 cases in which the main indication was painful retroversion and hypermobility of the uterus [7]. In fact, peritoneal resection of the Douglas pouch, associated with tightening of uterosacral ligaments and peritonization to exclude the Douglas pouch, is one of the best methods to use to definitively correct retrodeviation and excessive mobility of the body of uterus and suppress the neurovascular compression responsible for characteristic pelvic pain. The Douglasectomy avoids the occurrence of peritoneal cysts related to the peritoneal liquid trapped below the partitionment of the Douglas pouch created by the uterosacral ligamentopexy. Treatment of chronic pelvic pain is a much-discussed topic. Multiple etiologies have been incriminated, namely psychosomatic causes, and clinical and laboratory tests are often of little help. Diagnostic laparoscopy has been useful in diagnosing endometriosis (22% of cases), pelvic adhesions (27.4% of cases), pelvic varicose veins (11.4%), ovarian cysts (6.5%), ovarian dystrophy (3.2%), and subserous myoma (2.1%). Typical Masters-Allen syndrome is a finding in only 3.8% of cases in this series of 184 laparoscopic examinations reported by Priou [9] and the pelvis appears normal in 28.8% of the female patients. Masters-Allen syndrome, which was described in 1955 as the universal joint cervix syndrome, has the following three elements [1, 3, 8, 14]:

Correspondence to: P. von Theobald

Etiology: obstetrics-related trauma Clinical findings: uterine retroversion with hypermobile cervix following elongation or desinsertion of the uterosacral ligaments

640

Anatomy: visualization of a tearing of the posterior serosa and subperitoneal fascia of the ligamentum latum

Materials and methods

All female patients included in these series were multiparous; the average age was 32 years (27 to 41 years). All suffered from typical pelvic pain which had been investigated and treated on many occasions. A diagnosis of Masters-Allen syndrome was made and inclusion in the series was decided on the basis of the following criteria: 1. Obstetric anamnesis showing a past history of dystocia: large babies, instrumental extraction, breech presentation. Frequently, the uterus was retrodisplaced prior to pregnancy and the beginning of the disorder could be traced to some obstetrical episode. 2. Typical pelvic pain: It appears and exacerbates during prolonged standing, is aggravated by fatigue, and disappears in the recumbent position. It is invariably associated with dyspareunia. Dysmenorrhea is absent and the pain does not fluctuate during the menstrual cycle. 3. Clinical findings: An important degree of uterine retroversion/flexion Pain provoked by mobilization of the uterine fundus: the acute pain described by the patient can be elicted Abnormal mobility of the cervix in relation to the uterine body: The cervix can be mobilized in all directions with the finger, as if it was articulated on a ball joint. 4. Pessary test: vaginal insertion of a pessary of the appropriate vaginaadapted size, positioned above the levatores ani muscles, provides for correcting the uterine retroversion and hypermobility. When positive, it suppresses almost immediately the painful symptomatology and constitutes a major argument in favor of Masters-Allen syndrome. When negative, it will cause no change. The pessary is left in place for about 10 days. The pessary is then removed and the pain reappears immediately, unchanged. The relief induced by the pessary is often so marked, when the test is positive, that one female out of two requests that it be left in place until the operation. 5. Diagnostic laparoscopy: Most of all, this presents the advantage of excluding other etiologies of pelvic pain, such as endometriosis, adhesions, or salpingitis [9]. However, it should be noted that many patients presenting Masters-Allen syndrome have undergone multiple laparoscopic examinations, which have resulted in the wrong diagnosis of chronic salpingitis or even endometriosis because of the inflammatory appearance of the pelvic peritoneum. Laparoscopy cannot help visualize true ligament disinsertion lesions located at the subperitoneal level [12, 14]. It will reveal nonspecific signs, such as minor serous effusion, varicocele, marked retrodeviation of the uterus with a very deep Douglas pouch reminiscent of enterocele, and typical, sometimes recent, but often long-standing, fibrous and retractile tears of the ligamentum latum. Soutouls sign should be tested for to confirm uterine hypermobility: 180 rotation of the uterine body around the saggital axis is easily produced [13].

Fig. 1. Posterior view of the uterus. Drawing of the peritoneal incision.

Fig. 2. Douglasectomy.

(Fig. 3) is carried out using three or four separate stitches with absorbable decimal 4 suture (Endostitch greatly facilitates the procedure), or by mechanical endoscopic stapling. Peritonization (Fig. 4) is carried out transversely by stapling or by continuous absorbable thread. The mobility of the ureters is checked at the end of the procedure.

Results Forty-one laparoscopic Douglasectomies were performed in the Department of Gynecology at the University Hospital of Caen during the period between 1990 and 1995 (Fig. 5). The average duration of the procedure is 75 min (45 to 130) and, out of the 41 operations conducted to date, the only complications have been:

Operating procedure
This is identical to the reference procedure by laparotomy. The uterus is catheterized in order to be able to anteverse the uterine body. Three trocars are put in place: one umbilical trocar for the optical device, and two suprapubic trocars of 10- or 12-mm diameter, fitted with 5-mm reducers, allowing for the passage of the needed dissecting instruments (grip forceps, scissors, washing/aspiration material) and suturing instruments (laparoscopic needle holder, Endostitch, or endoscopic staplers). Sometimes it is necessary to insert a fourth trocar laterally to the umbilicus on the assistant surgeons side in order to insert a laparoscopic spreader to push the intestinal loops back. Dissection of the Douglas peritoneum (Figs. 1 and 2) is carried out much more conveniently than by laparotomy; it is aided by aquadissection, image magnification, the length of the instruments and of the optical device, and by barohemostasis which prevents bleeding from any source other than pelvic varicose veins, the latter being ideally stopped by means of titanium clips, if needed. The incision of the peritoneum is performed just above the uterosacral ligaments after optical localization of the ureters. The anterior limit is the torus uteri and the posterior limit is the anterior wall of the rectum. Ligamentopexy of the uterosacral ligaments

One case of bleeding from a pelvic varicose vein with a concomitantly occurring breakdown of the washingaspiration system, forcing the operators to terminate by conventional surgery One case of subperitonization hematoma revealed by pain at day 8, requiring repeat laparoscopy at day 15 One case of postoperative endometritis treated with antibiotics

Surgical results were positive otherwise, although often quite painful for the first 48 h; hospital release was on day

641

Fig. 3. Uterosacral ligamentopexy. Fig. 5. Results of the 41 procedures.

Fig. 4. Peritonization.

4 on average (range 2 to 8). Patients were seen again 2 months postoperatively, and then yearly, with an average follow-up period of 2.9 years (6 months to 5 years). To date, we have not lost sight of any of the cases. Thirty-one females experienced complete pain relief (75%) and five more (12%) had their symptoms greatly improved. Five patients (12%) experienced no relief after the procedure. None of the successfully operated patients had a recurrence of the pelvic pain and dyspareunia. Twenty-three women became pregnant again and had a normal delivery (except for two cesareans performed for miscellaneous indications), with no recurrence of pain. Four out of the ten who did not improve at all or were only partially improved showed undiagnosed microscopic endometriosis of the excised Douglas pouch peritoneum. Discussion Chronic pelvic pain in women is difficult to manage and diagnostic investigations are seldom decisive. Once the

most common etiologies have been excluded, such as infection and endometriosis, many practitioners attribute the pain to psychological causes. Two-thirds of the female patients in our series had two or more laparoscopic examinations carried out before our intervention, most often by different operators. These investigations frequently resulted in the description of catarrhal salpingitis or soft endometriosis, or even normal pelvic findings. But antibiotic and antiinflammatory treatments remained consistently without effect. We think that the diagnosis of Masters-Allen syndrome is too often missed. It does not seem to us justified to classify all these women as hypochondriacs. In fact, our results prove us right: When the indication is properly set, 87% of the women experience pain relief, which no operation-related placebo effect can, in itself, explain. When symptoms are present, Masters-Allen syndrome is characterized by static disorders of the pelvis, and Douglasectomy with uterosacral ligamentopexy is the only definitive procedure likely to restore normal anatomy. Laparoscopy has considerably facilitated this procedure thanks to image magnification and barohemostasis, and has reduced surgical sequelae for the patients. The parietal complications related to all laparotomies are avoided: hematoma, abscess, evisceration, cosmetic sequelae. Some laparoscopic techniques exist as an alternative to Douglasectomy:

Ligamentopexy of the round ligaments [4, 5] This used to be conducted by laparotomy as Doleriss procedure and was dropped because of the very high rate of recurrences within 2 years postoperatively. Manhe ` s has revived it through the use of laparoscopy. Laparoscopic ligamentopexy is an easy and rapid procedure consisting in a retightening of the round ligaments by grasping those extraperitoneally under laparoscopic control and fixing them by plicature to the aponeurosis of the rectus abdominis muscle. The uterus is thus put back into the intermediate

642

position, while the degree of anteversion is controlled endoscopically. In our opinion, this procedure presents several disadvantages:

lapse, sometimes responsible for digestive obstruction by way of incarceration, and of positional pain [11]. Conclusion Douglasectomy is the only definitive procedure available to restore normal anatomy of the pelvic floor in the case of painful uterine retroversion occurring in a setting of Masters-Allen syndrome. Additionally, it provides for pathological analysis of the excised peritoneum. The results of this procedure are excellent when the indication is correctly set, particularly as concerns positive pessary testing. The relief obtained by pessary insertion is spectacular, so much so that many female patients request that it be left in place until the surgical procedure. The same relief is regained after Douglasectomy and tightening of the uterosacral ligaments have been carried out. References
1. Allen WM (1971) Chronic pelvic congestion and pelvic pain. Am J Obstet Gynecol 109: 198202 2. Ameline A, Hughier J (1957) La suspension poste rieure au disque lombo sacre . Gynecol Obstet 45: 5694 3. Atkinson SM Jr (1970) Syndrome de luniversal joint cervix (Allen et Masters). Laceration traumatique des supports ute rins. Obstet Gynecol 36: 510514 4. Bruhat MA, Dubuisson JB, Pouly JL (1989) Ligamentopexie ante rieure des ronds par voie coelioscopique dans la cure des retroversions ute rines. E.M.C., (Paris, France), Techniques Chirurgicales, Urologie Gynecologie, 41515: 3233 5. Durand A, Abeille JP (1979) La cure de retrode viation percoelioscopique: 6 ans apre ` s. Gyne cologie 30: 275278 6. Good MC, Copas PR, Droody MC (1992) Uterine prolapse after laparoscopic uterosacral transection. A case report. J Reprod Med 37: 995996 7. Jamain B (1976) La douglassectomie en gyne cologie. Concours Me d 98: 36873691 8. Masters WM, Allen WM (1955) Traumatic laceration of uterine support. Am J Obstet Gynecol 70: 500513 9. Priou G, Arvis P, Rind A, Fraisse E, Grall JY (1984) Etude de lapport diagnostique de la coelioscopie dans le bilan des algies pelviennes chroniques. J Gynecol Obstet Biol Reprod 13: 395402 10. Querleu D (1994) Cure de prolapsus par ventrofixation: re e dition coeliochirurgicale dune mauvaise ope ration. Rev Fr Gyne col Obste t 89: 164 11. Querleu D, Delahousse G, Decocq J, Parmentier D (1988) La place actuelle de la douglassectomie. Communication aux 4 Journe es Annuelles de la Socie te Internationale Francophone de Chirurgie Pelvienne, Roubaix, 2526 mars 1988 12. Robert HG, Palmer R, Boury-Heyler C, Cohen J (1974) Les anomalies de position. In: Pre cis de gyne cologie. Masson, Paris, pp 471477 13. Soutoul JH (1973) Essai de classification des le sions ligamentaires dans le syndrome de de sinsertion ute rine (Masters et Allen) gra ce a ` une exploration dynamique percoelioscopique et ope ratoire de la mobilite ute rine. Gyne cologie 24: 339344 14. Soutoul JH (1974) Syndrome de Masters et Allen avec de sinsertion ute rine. Bilan fonctionnel des le sions ligamentaires et indications ope ratoires. J Gynecol Obstet Biol Reprod 3: 1320 15. Wood C, Maher P, Hill D (1993) Laparoscopic removal of endometriosis in the pouch of Douglas. Aust N Z J Obstet Gynaecol 33: 259299

Recurrences are not less frequent than before. A poor surgical laparotomic technique is not made good even when touched by the grace of laparoscopy. In fact, the round ligaments are quickly loosened and stretched, and the treatment of the retroversion is transient. Ligamentopexy of the round ligaments tears the uterus forwards and increases the risk of enterocele.

Ventrofixation [10] This technique, which had long been dropped by all operators, consists of fixing the anterior side of the deperitonized uterus to the anterior wall of the abdomen. This procedure has also been revived by laparoscopy. This method seems to us even more worthy of criticism than ligamentopexy of the round ligaments, since it opens the Douglas pouch much more yet and hyperanteverses and immobilizes the body of the uterus. Enterocele is sure to occur. Section of the preacral nerve [6] This procedure was first used by Cotte in 1925, primarily in the treatment of severe algomenorrhea, and has regained popularity among Anglo-Saxon authors with the advent of laparoscopy. The effect on pain seems quite satisfactory, although no serious study has been published. The disadvantage here is the lack of repair of the static disorder of the pelvis. Enterocele-type complications have begun to be seen. Insertion of a pessary This can only be a temporary cure of the pelvic pain and dyspareunia in these young women, because it often involves vaginal infection and restrains one from regular intercourse. Douglasectomy is indicated in pathologies other than Masters-Allen syndrome [11]:

Endometriosis of the Douglas pouch: Peritoneal resection provides for removal of all lesions and constitutes, in this location, an alternative to laser vaporization [15]. Treatment of prolapse by the abdominal route and, more particularly, by laparoscopy: Some authors use Douglasectomy as a complementary procedure. It permits one to obtain a perfect exposure of the posterior side of the cervix and vagina, in the case of posterior prosthesis, and to correct or to prevent the occurrence of enteroceles [2]. Isolated enterocele, associated or not with genital pro-

Surg Endosc (1997) 11: 687692

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Balloon dissection of the space of Bogros via the femoral canal for total extraperitoneal laparoscopic herniorrhaphy
R. C. Read,1,2 R. A. de la Torre,3 J. S. Scott3
1

General Thoracic Surgery, Surgical Service (112 LR), John L. McClellan Memorial Veterans Hospital, 4300 West 7th Street, Little Rock, AR 72205, USA 2 University of Arkansas for Medical Sciences, Little Rock, AR, USA 3 Suite 214, 600 Medical Drive, Wentzville, MO 63385, USA Received: 3 April 1996/Accepted: 15 September 1996

Abstract. To obviate the need for general anesthesia or dissection of the rectus sheath, we have transferred laparoscopic herniorrhaphy back to the groin by first dissecting the suprainguinal parietoperitoneal space of Bogros via the femoral canal. Cadaver dissections demonstrated that the preperitoneal plane could be routinely fingered and distended with a digitally placed balloon introduced through a 1-cm incision immediately below the inguinal ligament. A 10-mm femoral laparoscopic port was then inserted and pressurized, allowing two standard 5-mm ports to be introduced from above, through the lower quadrant, under vision. The procedure was then carried out in the usual way, the mesh being inserted from below. Ten patients (two women), 2373 years old, selected because general anesthesia was inadvisable, underwent uncomplicated prosthetic repair of unilateral (eight) or bilateral (two) inguinal defects. Half of the peritoneal sacs were pushed up and out of the inguinal canal; 18 months later there were no recurrences (inguinal or femoral). Preliminary experience with this new technique is promising. It may prove applicable to retroperitoneal exposure of the distal aorta and iliac vessels, allowing laparoscopic bypass for Leriche syndrome. Key words: Total extraperitoneal inguinal laparoscopic herniorrhaphy Femoral canal Space of Bogros

inflates the communicating lateral extraperitoneal spaces of Bogros [12] thereby providing a view of the posterior surface of the groin(s). To allow the routine use of spinal, epidermal, regional block, or local infiltration anesthesia we have attempted to return the entire operation to the groin. Instead of distending the suprainguinal parietoperitoneal space of Bogros [12] indirectly by first inflating the midline suprapubic space of Retzius we have explored the possibility of reversing the sequence. The space of Bogros has been dissected first with a balloon introduced digitally through the femoral canal. Since, in this new approach to a predominantly unilateral affliction, the rectus sheath does not need to be dissected, laparoscopic herniorrhaphy could potentially be practicable in patients who have a history of previous lower midline abdominal surgery or in the 10% of individuals whose posterior rectus sheath extends anomalously [9] to the pubes, thereby preventing safe trocar insertion into the space of Retzius.

Anatomy The space of Bogros In 1823 the French surgeon Jean Bogros (17861825) presented his M.D. thesis to the University of Paris. In it he noted that the peritoneal cavity did not reach all the way to the groin but left a parietoperitoneal space filled with fatty cellular tissue (Fig. 1). Since the distal external iliac artery coursed posteriorly therein (The peritoneum extending from the iliac portion of the anterior abdominal wall to the iliac fossa leaves in front a space 13.515.5 mm wide where the external iliac artery terminates), he recommended that ligation of this vessel for aneurysm be performed within this extraperitoneal space. He approached it by a groin incision,

A major limitation of laparoscopic repair is the need for general anesthesia. It has been required, even with the total extraperitoneal technique, not only to provide relaxation of the abdominal musculature but also for periumbilical anesthesia. This allows the space of Retzius [14] to be approached via the rectus sheath. Distension of the former
Correspondence to: R. C. Read

688

transecting the roof and floor of the inguinal canal, 60 years before Bassini described the same exposure for his classic repair of inguinal herniation! The advantage of Bogross operation over the then-standard approach for proximal extraperitoneal ligation, described in 1790 by John Hunters pupil, Abernethy [1], was that the latters incision was located just medial to the anterior superior iliac spine. It entailed a considerable risk of death from peritonitis, if the closely applied underlying peritoneum was divided (30 years before Lister introduced antisepsis!). Bogross original anatomical observations were confirmed almost a century later by a compatriot, Rouviere [15]:
The peritoneum which lines the deep aspect of the abdominal inguinal wall is reflected from the abdominal wall towards the iliac fossa creating a fold of peritoneum in the shape of a gutter concave above and behind. This fold of peritoneum is such that from the abdominal wall to the iliac fossa the outer layer of the peritoneum is in contact with the soft tissues of the iliac fossa, some 11.5 cm above the inguinal ligament. The peritoneum thus demarcates, with a dihedral angle formed by the fascia transversalis and the iliac fossa inferiorly, a triangular prismatic interval filled with preperitoneal adipose tissue called the space of Bogros.

Later, Testut and Latarget [17] stressed the continuity between the parietoperitoneal spaces of Bogros and Retzius, the kidney, and the spermatic cord. Bogros also observed that the inferior epigastric vessels arose in the floor of his space and then coursed inferiorly, overlying their parent external iliac vessels. They turned anteriorly to supply the rectus muscle within its sheath. Mackay [7] (1889) later pointed out that at the bend the epigastric vessels pierce the transversalis fascia to enter the abdominal wall. They, thus, do not course in the extraperitoneal fat as is classically taught. The vein drains from a suprainguinal plexus recently detailed by Bendavid [3]. This vasculature is not located, as he described, within the space of Bogros but rather between Coopers two laminae of the transversalis fascia [13], the deepest layer of the anterior abdominal wall. The inferior epigastric artery and vein after they enter the parietes lie external to the internal spermatic fascia derived from the transversalis fascial lining of the space of Bogros (posterior lamina) (Fig. 1). This preperitoneal funnel passes through one of the two outlets of the space of Bogros, the secondary internal abdominal ring of Fowler [13], into the internal abdominal ring in the thicker external lamina of transversalis fascia, thereby forming the internal spermatic fascia. The ureter courses down to the pelvis in the posteriomedial portion of the space of Bogros from which it may herniate into the inguinal canal. This anatomical relationship was repeatedly observed by Henry [4] during an extensive experience with the extraperitoneal approach for the removal of bilateral ureteric stones in Egyptian patients infested with the Bilharzia parasite. Consequently, when a young woman presented with bilateral femoral herniation, he rediscovered Cheatles midline, extraperitoneal, posterior approach to the groin. This operation evolved into the giant prosthetic reinforcement of the visceral sac (GPRVS) of Stoppa [16], which has formed the basis for laparoscopic, total extraperitoneal prosthetic repair. A unilateral modifi-

Fig. 1. Sagittal diagram of the femoral canal (F) passing up under the inguinal ligament and canal (I) over the pectineus muscle, ligament of Astley Cooper (E), and pubic ramus into the space of Bogros (D) beneath the peritoneum (A). Note posterior lamina transversalis fascia (B), epigastric vessels (C), anterior femoral sheath (H), posterior sheath (iliac fascia), and pectineus fascia (G).

cation of the Cheatle-Henry procedure was originally described by McEvedy in 1950 [8]. Prosthetic repairs through this approach include the unilateral GPRVS of Wantz [19]. This sequence of events prompted us to investigate whether the surgical anatomy was such that, just as the open midline Cheatle-Henry posterior preperitoneal approach gave rise to McEvedys unilateral variant, extraperitoneal balloon dissection for laparoscopic herniorrhaphy could likewise be transferred from the umbilicus (where it began with transperitoneal procedures) to the groin for total extraperitoneal repair. Femoral canal The external iliac vessels as they pass into the thigh take with them a sheath of transversalis and iliac fasciae lining the extraperitoneal space of Bogros (Fig. 1). Since the medial portion of the femoral sheath forms a canal we decided to use it as a point of entry into the suprainguinal preperitoneal space which could then be distended to allow laparoscopic herniorrhaphy of the groin. The abdominal opening of the conical 1.53.0-cm-long femoral canal is bounded by an elastic fascial ring enclosed within a rigid aperture. The former is composed anteromedially by Henles ligament and the iliopubic tract, which separates the overlying

689

Fig. 2. One-centimeter skin incision deepened over the medial portion of the femoral sheath immediately below the inguinal ligament and above the inguinal skin crease (dashed line) and the entry of the saphenous vein (cribriform plate).

Fig. 3. Fingering of the space of Bogros through the femoral canal and ring.

inguinal canal. Laterally, a septum borders the femoral vein. The rigid component comprises, posteriorly the superior pubic ramus overlain by Coopers ligament, medially Gimbernats ligament, and anteriorly the inguinal ligament. The femoral ring (927 mm in diameter) is separated from the space of Bogros and the femoral fossa of the parietal peritoneum by a flimsy cup-shaped septum, concave superiorly, derived from the posterior lamina transversalis fascia as it inserts into Coopers ligament internal to the inferior epigastric vessels and their branches. Femoral herniae exit the canal through an orifice in the fascia lata (foramen ovale) covered by the cribriform plate of Scarpas fascia where the saphenous vein enters the femoral sheath. They become incarcerated or strangulated either here or higher up at the femoral ring. This textbook description of the anatomy is open to argument. Some consider the femoral canal to be a potential space, its presence only becoming manifest with herniation. Measurements taken after death provide information regarding what obtains, predominantly, in the elderly, debilitated or diseased with impaired musculature, aponeuroses, and fasciae. The presence or absence of clot in the external iliac vein, after embalming, also affects data pertaining to the size of the femoral ring. Methods
We first studied retrograde intubation of the femoral canal in cadavers obtained from the dissecting room or at autopsy. Initially, entry was attempted at the foramen ovale. However this proved to be unsatisfactory for two reasons. First, in the absence of femoral herniation, the femoral canal did not consistently extend this far into the thigh. Secondly, superficial branches of the saphenous vein interfered and had to be ligated. We therefore moved proximally to the inguinal skin fold, immediately below the inguinal ligament, just proximal to the inguinal skin sulcus (Holdens line), the surface landmark separating the abdomen from the thigh. A 1-cm transverse skin incision was made medial to the palpated femoral artery

and vein. The underlying femoral sheath was cut to unroof the canal (Fig. 2). The index finger was then inserted and passed upward beneath the inguinal ligament and the transversalis fascial floor of the inguinal canal to Coopers ligament overlying the superior pubic ramus. The flimsy transversalis fascial femoral septum separating the canal from the space of Bogros offered no resistance, the tip of the finger falling into the suprainguinal extraperitoneal space (Fig. 3). It could be swept around medially behind the rectus abdominis tendon, inferiorly over and behind the pubic ramus toward the obturator foramen, anteriorly to the space of Retzius, and laterally up behind the secondary internal abdominal ring of Fowler and the inner (preperitoneal) spermatic cord to the anterior abdominal wall (Fig. 1). The external iliac vein is separated from the femoral canal by a fibrous septum. It is important that the index finger as it passes upward beneath the inguinal ligament run parallel to it and not wander out of the sheath, making a false passage into the inguinal canal, anteriorly, or Gimbernats ligament, medially. We distended the space of Bogros with a 750-ml saline-filled balloon. It was originally mounted on a standard straight, rigid mandrel designed for infraumbilical insufflation. However, in one instance the overlying peritoneum was perforated by the blunt tip. We therefore decided to simply place the balloons digitally. After removing the balloon, a standard 10-mm laparoscopic port was inserted through the femoral incision and pressurized to 810 mmHg with CO2. A camera was introduced from below and two 5-mm ports were inserted with trocars through the distended lower quadrant of the anterior abdominal wall under direct vision. Laparoscopic repair was then carried out from above in the usual manner, the 11 15 cm2 polypropylene prosthesis being introduced from below through the larger femoral port.

Clinical experience
Early in 1995, eight men and two women, whose ages ranged from 23 to 73 years, underwent balloon dissection of the space of Bogros via the femoral canal by the technique described with uncomplicated laparoscopic repair of primary and recurrent, unilateral (eight) or bilateral (two) inguinal, with indirect (eight) or direct (four) defects. These patients were selected out of our main experience with infraumbilical dissection because either their preference or physical condition made general anesthesia inadvisable. Some had a history of previous lower midline abdominal surgery. After preoperative assessment and workup had been conducted on an outpatient basis, patients were scheduled and arrived for 1-day surgery early in the morning, having fasted overnight. They received prophylactic

690 antibiotics. They were premedicated and asked to void. They were then placed supine on the operating room table and the skin over the groin was prepared. After infiltration with a local anestheticalong with intravenous sedationthe femoral skin incision was made and the space of Bogros was fingered. A 750-ml nonelastomeric balloon (Spacemaker Balloon Dissector, GSI) was inserted and inflated. Repair was accomplished using the technique developed in the cadaver.

Results The femoral canal was patent in all, allowing the space of Bogros to be consistently fingered (Fig. 2). There was no bleeding either before or after insertion and inflation of the balloons, which consistently stretched and everted the body wall in the groin extending into the midline. The appearance of the extraperitoneal space was similar to that seen with standard infraumbilical laparoscopy. However, there was better dissection of the preperitoneal cord structures posterolaterally and superiorly. The transversalis fascial lining of the space of Bogros was disrupted only at the femoral ring, remaining internal to the rectus muscles and epigastric vessels in the space of Retzius. Despite unilateral inflation, bilateral dissection, and, if necessary, repair, were easily performed since the space of Retzius inflated secondarily. Interestingly, reduction of processi vaginales from within the inguinal canal was complete in half of those with indirect defects. This facilitated dissection and separation of the elements of the spermatic cord from each other and the sac at the internal abdominal ring prior to parietalization. Follow-up of at least 6 months has shown no recurrences (inguinal or femoral). All patients were able to go home after outpatient surgery on oral analgesics. The femoral wounds healed well without the late bruising and tenderness often seen in the usual periumbilical location. None has developed phlebitis. Discussion The presence of a femoral canal, potential or manifest, allows the vessels supplying blood to and from the leg to expand in response to gravity or increased flow; unfortunately it can also let bowel escape from the abdominal cavity. Nevertheless, this channel continues to serve as a path through the abdominal wall, allowing surgeons to relieve visceral entrapment at the femoral ring. As Monroe [10] has stated, The low approach to femoral herniation is almost as old as surgery itself. Before the days of anesthesia operation was undertaken only for strangulation or obstruction the concern of the surgeon being to release the constricting band without harming the content or causing injury to blood vessels or bladder. The first elective repair was that of Socin (1879), ligating and removing the sac from below. Later, to avoid recurrence, a number of surgeons, notably including Bassini (1893), added suturing of Pouparts ligament, Gimbernats, the iliopubic tract, or the transversus abdominis muscle and tendon down to either the inguinal ligament, the ligament of Astley Cooper, or the pectineus muscle and fascia. Tension posed a problem and therefore others plugged the femoral canal with an agate marble, saphenous vein, or prosthetic mesh. Almost all of these procedures are now performed under local anesthesia [6]. In 1989, Bendavid [2] reported a recurrence rate of 6.1%

in 248 patients who had undergone, 58 years previously, fascial repair of primary femoral herniation, 91% from below. The rate in 211 recurrent cases was 22%. He therefore recommended that all large or recurrent defects be closed with his 8-cm polypropylene mesh umbrella inserted under local anesthesia up to the femoral ring, where it is stitched circumferentially. Munshi and Wantz [11] have recently pointed out that this repair would not be suitable for perivascular defects since the prosthesis cannot be stitched to the external iliac vessels. They report the successful use of the unilateral GPRVS procedure, developed primarily for inguinal herniation, in 69 such difficult primary (15) or recurrent (54) femoral hernias. Of particular interest to our study is that in six of these patients, who were elderly and too debilitated for general anesthesia, an emergency operation was successfully completed from below using local anesthesia. It could be argued that the presence of a femoral hernia so stretches the fascial confines of the canal, especially its ring and orifice, that it can be more easily entered and worked in. Further, aging and loss of body fat with laxity of the abdominal musculature might allow surgical manipulations not possible in those with a normal femoral canal, possibly including patients presenting with inguinal herniation. Nevertheless, in the most successful fascial repair for this condition being performed around the world today, the Shouldice procedure, the femoral canal is routinely fingered from below to rule out an associated femoral hernia. This is done through the same 1-cm transverse incision in the fascia lata and anterior femoral sheath, immediately below the inguinal ligament, we have used (Fig. 2) except that the subcutaneous fat in the groin incision is retracted inferiorly to avoid having to make a separate femoral wound. Our experience is similar in that, in our patients with inguinal herniation (or cadavers without) at this level, but not below at the foramen ovale, the femoral canal was open (Fig. 2) despite the absence of femoral herniation. The intent has not been that of Wantzto perform prosthetic repair through the femoral canalbut just to initiate the operation, which is then carried out laparoscopically from above, using the posterior approach to the groin in the usual manner. Classical anatomy, using data often obtained before the use of antibiotics or even antisepsis, teaches that the femoral canal contains lymph nodes which, presumably, if enlarged, with associated preperitoneal fat, could block passage; however, sepsis is much less common today. Further, failure of the transversalis fascia, the most common cause of inguinal herniation in the adult, enlarges the femoral ring by damaging its elastic transversalis fascial components, thereby producing a significant incidence of associated femoral herniation. Thus, the space of Bogros has two exits (inguinal and femoral canals) through the myopectineal orifice or hole of Fruchaud, which enlarges in patients with inguinal herniation. The use of the umbilicus as the portal for insufflation derives from the traditional approach to transperitoneal laparoscopy. A high periumbilical incision in the abdominal wall (which usually requires general anesthesia), dissection of the rectus sheath, and distension of the space of Retzius (the latter only mandated, as unilateral GPRVS has shown, in the more unusual, bilateral herniation) add little to unilateral repair itself. Dissection of the lateral suprainguinal preperitoneal space of Bogros via the femoral

691

canal transfers insufflation for total extraperitoneal laparoscopy from the umbilicus to the groin, thus allowing the routine use of local epidural or spinal anesthesia. Our preliminary experience indicates that this goal can be reached. The purpose of fingering the transversalis fascia lined femoral canal is to break through the flimsy femoral septum which covers the femoral ring, thereby entering the suprainguinal parietoperitoneal plane (Fig. 2). It is important to realize that, as Hureau has stressed, this space of Bogros is a cleavable space devoid of any real structure but for a scant amount of adipose-cellular tissue. . . . Vessels, nerves and other elements run along the wall [5]. This parietal wall is the translucent but tough posterior lamina of transversalis fascia originally described in 1807 by Cooper but later lumped in with the extraperitoneal fatty layer until rediscovered by Read in 1992 [13]. Unfortunately, Bendavid has recently described the deep inguinal venous plexus (tributaries of the inferior epigastric vein) as arching within the space of Bogros, thereby raising unnecessary fears of bleeding if this plane is entered, as it has to be for adequate repair. This area is markedly vascular and prone to bleeding and hematomas [3]. Accumulated experience with the unilateral or bilateral Stoppa posterior preperitoneal procedure performed either open or laparoscopically within the space of Bogros has shown unequivocally that such veins are not encountered. Our experience with balloon distension of this extraperitoneal space from below is consonant in that bleeding does not occur. Thus the undisturbed parietoperitoneal spaces of Bogros and Retzius are essentially avascular and can therefore serve as fatty shock absorbers protecting the bladder and peritoneal content from stress and strain. The epigastric vasculature runs external to these parietoperitoneal planes in between the two laminae of transversalis fascia, the deepest layer of the abdominal wall, and is not encountered except with the classical anterior approach from the groin or the unilateral posterior preperitoneal incision. Even though the approach of McEvedy has been considered equivalent to the classical Cheatle-Henry midline incision, anatomically it is not. In the latter the operation is conducted through the extraperitoneal space of Retzius internal to the posterior lamina of the transversalis fascia on which the inferior epigastric vasculature, in the parietes lies surrounded by fat. These vessels are retracted anterolaterally with the rectus muscles. By contrast, McEvedy originally described his incision as being within the rectus sheath [2]. Thus, initially after transecting the substantial outer lamina of transversalis fascia the surgeon comes down onto the epigastric vessels resting upon the posterior lamina of transversalis fascia. Nyhus, who popularized the McEvedy incision, initially ligated the inferior epigastric artery and vein before penetrating the underlying posterior lamina transversalis fascia in order to enter the space of Bogros and repair the defect. Other surgeons using the posterior preperitoneal approach, either open or laparoscopically, have debated whether the prosthesis should be inserted superficial to or deep to the epigastric vasculature (most settling for the latter). Gilbert reports [personal communication] that prosthetic plugs introduced through the internal inguinal ring, into the preperitoneal plane, from the classic anterior approach can cause bleeding which may extravasate into bizarre locations; presumably Bendavids suprainguinal ve-

nous plexus is being torn in the transversalis fascial roof to the space of Bogros. Thus, laparoscopic exposure of the space of Bogros, from above or below, has an important advantage over the unilateral open preperitoneal approach: Regardless of patient comfort, the epigastric vasculature is not encountered on the way in; this asset needs to be more widely appreciated. It is important to realize that instrumentation developed to mount balloons inserted through the standard transumbilical approach cannot be automatically applied to this new route. Here the peritoneum overlies the femoral ring (Fig. 1) and can be easily fingered from the infrainguinal incision (Fig. 3). This compares with a distance of three finger lengths or more between the umbilicus and the pubes. Our experience in the cadaver indicated that mandrels should not be used from below, and we have therefore made it a rule to insert the balloons digitally. Just recently, after this initial series of patients had been operated upon and followed up for presentation, we contravened this principle in a 75-yearold patient who presented with a history of multiply recurrent bilateral inguinal herniation and a fresh protrusion in the left groin. Since he had undergone a suprapubic prostatectomy we attempted to obtain a higher posterolateral inflation of the space of Bogros, hoping thereby to avoid the scarred space of Retzius. After fingering the cavity through the femoral canal, a curved rigid mandrel-mounted balloon was inserted and inflated. Excellent distension, albeit unilateral, was obtained. However, when the balloon was withdrawn, bleeding ensued from the incision. We induced general anesthesia and converted to open repair by incising his midline scar. The external iliac artery was fibrosed with surrounding lymph nodes. It had been deviated anterolaterally, presumably by the blunt tip of the rod. Removal of a hematoma showed a leak distal to a calcified plaque at the takeoff from the common iliac artery. This was sutured and a GPRVS (unilateral) repair was carried out. Fortunately, he has done well. Thus, by its breach, the principle of using only free-standing, digitally placed everting balloons was confirmed. Further, a history of suprapubic prostatectomy remains a contraindication to laparoscopic repair since reaction to it is not restricted to the space of Retzius but extends unpredictably throughout the posterolateral extraperitoneal space. However, as indicated above, we have been able to use this novel approach in other patients with infraumbilical scarring not extending down to the pubis. An unexpected advantage of transfemoral inflation is that the distending force is applied more from below than in front of the peritoneal cavity. Coupled with the Trendelenburg position, this has resulted in more of Henrys [4] peritoneal horns of the snail being lifted out of hernial defects. This enhanced reduction of hernial sacs has reduced the amount of preperitoneal fascial funnel dissection required to separate the cord from the peritoneum and to parietalize it before placement of the prosthesis. Hopefully, the incidence of peritoneal tears and damage to the vas deferens will be reduced. It is interesting that Bogros, whose original observations are central to this new approach, was a pioneer vascular surgeon. Exponents [18] of his specialty in 1972 used the normal femoral canal as a new route for the femorofemoral arterial bypass, demonstrating that, even in the absence of herniation, inguinal or femoral, there is easy communication with the space of Bogros, again not

692

from as far down as the foramen ovale. A prosthetic conduit was introduced via the femoral canal from the good leg; it was then passed digitally across the space of Retzius retropubically to the space of Bogros on the ischemic side. By then bringing the graft down the contralateral femoral canal and anastomosing its two ends to the femoral vessels a new source of blood flow was provided. In the future, laparoscopic dissection of the space of Bogros via the femoral canal may facilitate retroperitoneal exposure of the terminal aorta and iliac vessels, thereby preventing iatrogenic herniation through the groin secondary to splitting of the inguinal ligament and abdominal musculature.
Acknowledgment. We appreciate the assistance provided to this study by the GSI Corporation, Palo Alto, California, and in particular Dr. Thomas J. Fogarty, Dennis Benassi, George Hermann, Ken Mollenauer, and Shelly Monfort. The illustrations were the work of Patricia ONeal from the University of Arkansas for Medical Sciences Media Department.

6. 7.

8. 9. 10.

11.

12. 13. 14.

References
1. Abernethy J (1811) Surgical observations on the constitutional origin and treatment of local diseases and on aneurysms. 2nd ed. Fry and Kammerer, Philadelphia 2. Bendavid R (1987) A femoral umbrella for femoral hernia repair. Surg Gynecol Obstet 65: 153156 3. Bendavid R (1992) The space of Bogros and the deep inguinal venous circulation. Surg Gynecol Obstet 174: 355358 4. Henry AK (1936) Operation for femoral hernia by a mid-line extraperitoneal approach: with a preliminary note on the use of this route for reducible inguinal hernia. Lancet 1: 531533 5. Hureau J (1994) The space of Bogros and the interparieto-peritoneal

15. 16.

17. 18. 19.

spaces. In: Bendavid R (ed) Prostheses and abdominal wall hernias. R G Landes, Austin, TX, pp 7481 Koontz AR (1963) Historical analysis of femoral hernia. Surgery 53: 551555 Mackay JY (1889) The relations of the aponeuroses of the transversalis and internal oblique muscles to the deep epigastric artery and to the inguinal canal. In: Cleland J (ed) Memoirs and memoranda in anatomy, vol 1. Williams and Norgate, London, pp 142145 McEvedy PG (1950) Femoral hernia. Ann R Coll Surg Engl 7: 484 496 McVay CB, Anson BJ (1940) Composition of the rectus sheath. Anat Rec 77: 213225 Monroe A (1964) Femoral hernia: the lower approach. In: Nyhus LM, Harkins HN (eds) Hernia. J B Lippincott, Philadelphia, PA, pp 199 207 Munshi IA, Wantz GE (1996) Management of recurrent and perivascular femoral hernias by giant prosthetic reinforcement of the visceral sac (GPRVS). J Am Coll Surg 182(5): 417422 Read RC (1995) Surgical comments on the Bogros thesis. Postgrad Gen Surg 6: 1517 Read RC (1992) Coopers posterior lamina of transversalis fascia. Surg Gynecol Obstet 174: 426434 Retzius AA (1858) Some remarks on the proper design of the semilunar lines of Douglas. Edinburgh Med J 3: 685687 Rouviere H (1912) Anatomie humaine descriptive, topographique et fonctionelle. Masson, Paris Stoppa RE (1995) The preperitoneal approach and prosthetic repair of groin. Hernias. In: Nyhus LM, Condon RE (eds) Hernia. 4th ed. JB Lippincott, Philadelphia, PA, pp 188210 Testut L, Latarjet A (1948) Traite danatomie humaineToure premier. G Dovin, Paris Tyson RR, Reichle FA (1972) Retropubic femorofemoral bypass a new route through the space of Retzius. Surgery 72: 401403 Wantz GE (1993) The technique of giant prosthetic reinforcement of the visceral sac performed through an anterior groin incision. Surg Gynecol Obstet 176: 497500

Surg Endosc (1997) 11: 643644

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Laparoscopic bowel mobilization combined with intraoperative colonoscopic polypectomy in patients with an inaccessible polyp of the colon
K. Smedh,1 S. Skullman,2 A. Kald,2 B. Anderberg,2 P.-O. Nystro m2
1 2

Department of Surgery, Central Hospital, Central Hospital, S-721 89 Va sters, Sweden Department of Surgery, Colorectal Unit, University Hospital, Linko ping, Sweden

Received: 5 May 1996/Accepted: 19 September 1996

Abstract Background: The purpose of this report was to describe a simple technique suitable for polyps where circumstances of the bowel anatomy prevent complete access and control of the colonoscopic procedure. Methods: By combining laparoscopic mobilization of the bowel with colonoscopic polypectomy, previously inaccessible polyps could be snared in two patients. Results: Both patients had 3-cm large sessile adenomas in the sigmoid colon safely removed, and they returned home within a day. Conclusions: The described procedure increases the safety of the otherwise difficult polypectomy and also avoids laparotomy with enterotomy or bowel resection as the alternative. Key words: Laparoscopy Colonoscopy Colonic polyp

surgical procedure is described whereby laparoscopic bowel mobilization alone has been combined with colonoscopic polypectomy in order to reduce the operative procedure.

Technique
The patient is placed in the lithotomy position. After insufflation of gas the camera port is placed just above or beneath the umbilicus and additional ports are introduced just outside the rectus muscle in each fossa. The colonic segment is mobilized from its lateral peritoneal attachments so that it can be easily moved by atraumatic Babcock forceps. The colonoscope is introduced through the anus and the polyp is identified. By rotating and stretching the bowel with Babcocks it is now possible to get a good view to assess the resectability of the polyp and to procede with snare polypectomy if justified.

Results We have performed this combined laparoscopic and endoscopic technique in two patients, both with 3-cm large sigmoid polyps not possible to remove by simple colonoscopy due to their localization. Both our patients had uneventful recovery and histology showed tubulovillous adenomas with moderate dysplasia.

In patients with colonic adenomas that are located behind a mucosal fold or a sharp bowel bend, it sometimes can be impossible to get a good endoscopic view to snare the polyp safely. This is most common in the sigmoid colon. It can be difficult to assess the full extension of the polyp and its potential malignancy or to get representative biopsies. In such cases the polyp is usually resected at laparotomy, or, in recent years, by laparoscopically assisted techniques [2, 3]. A laparoscopic-assisted full-thickness endoscopic polypectomy has recently been described for colonic lesions too large to be safely removed colonoscopically [1]. In patients with impaired access to colonic polyps a
Correspondence to: K. Smedh

Discussion This simple technique of combining laparoscopic mobilization of the bowel with colonoscopic polypectomy is suitable for polyps which are resectable with colonoscopy in principle but where circumstances of the bowel anatomy prevent complete access and control of the procedure. After the bowel has been mobilized there is always the possibility to proceed to resection if the polyp appears malignant or larger and more sessile than expected. There is also excellent con-

644

trol of potential transmural thermal injury as the external bowel surface of the polypectomy site is under direct inspection with the laparoscope. By avoiding a bowel resection the postoperative course is much easier and more straightforward and the patient can return home within a day. The described procedure increases the safety of the otherwise difficult polypectomy while it also avoids bowel resection as the alternative. We submit that it will be costeffective, too, because it avoids uncertainty concerning the

completeness of the polypectomy and hence repetetive endoscopies are unnecessary. References
1. Beck DE, Karulf RE (1993) Laparoscopic-assisted full-thickness endoscopic polypectomy. Dis Colon Rectum 36: 693695 2. Guillou PJ, Darzi A, Monzon JR (1993) Experience with laparoscopic colorectal surgery for malignant disease. Surg Oncol 2(Suppl 1): 4349 3. Monson JRT, Hill ADK, Darzi A (1995) Laparoscopic colonic surgery. Br J Surg 82: 150157

Letters to the editor

Surg Endosc (1997) 11: 696

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The totally extraperitoneal laparoscopic hernia repair

We read with interest the paper by Vanclooster and colleagues [11] and commend their contribution to this procedure. However, we would offer three comments: First, totally extraperitoneal laparoscopic hernia repair was developed by Dulucq [13] in 1989/90 and by McKernan [7] not a great deal later. While it seems legitimate for others to publish their own technical variations, developments, and outcomes, we strongly support recent reminders [6] that journal editors and their peer referees owe the reader a duty of diligence: they should insist that authors exercise proper scholarship by giving credit where it is due. Otherwise the uninformed reader may assume originality and the informed may infer plagiarism, where the author intended neither. Second, the mesh configuration suggested by Vanclooster et al. was presented by one of us several years ago [4, 10]. However, the concept of amputating the inferior and lateral corner so the mesh fits better on the iliac vessels and the psoas muscle is flawed, since it has subsequently been reported [12] that recurrences may occur dorsal/ inferior to this inferolateral corner. The most extensive possible coverage of the psoas muscle belly is therefore appropriate. Third, the need for mesh fixation remains debatable. However, to fix the cranial border to prevent early migration or slipping is illogical: In our joint experience of over 1,000 cases and, to our knowledge in all reports in the world literature, recurrences pass uniformly caudal to the inferior border of the prosthesis. Fixation of the inferior medial part of the mesh to Astley Coopers ligament alone [5] may not offend against the original tension-free notion of Stoppa [8, 9] nor interfere with the mechanics of prosthesis retention. To fix the superior border to points that move relative to one another within a musculofascial structure contravenes both principles. Finally, on a minor point, if the structure annotated as D in Fig. 1 is the testicular vascular bundle, where is the vas deferens? Despite these comments we congratulate the authors on their low complication rate. References
1. Dulucq J-L (1991) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritone ale en retrope ritone oscopie. Cah Chir 79: 1516 2. Dulucq J-L (1992) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritoneal en pre -pe ritoneoscopie. Chirurgie 118(12): 8385 3. Dulucq J-L (1992) The treatment of inguinal hernias by implantation of mesh through retroperitoneoscopy. Postgrad Gen Surg 4: 173174 4. Fiennes AGTW, Taylor RS (1994) Learning laparoscopic hernia repair. In: Inguinal hernia, advances or controversies? Arregui M, Nagan R (eds) Radcliffe, Oxford, pp 475482 5. Fiennes AGTW, Himpens J, Dulucq J-L (1994) Preperitoneoscopic groin hernioplasty, current synthesis. Surg Endosc 8(8): 989 6. Horton R, Smith R (1996) Time to redefine authorship (editorial). Br Med J 312: 723 7. McKernan JB (1993) Laparoscopic extraperitoneal repair of inguinofemoral herniation. Endosc Surg Allied Tech 1(4): 198203 8. Stoppa R, Petit J, Abourachid H (1973) Proce de original de plastie des hernies de laine. Linterposition sans fixation dune prothe ` se en tulle de Dacron par voie me diane pre pe ritone ale. Chirurgie 99: 119 9. Stoppa RE, Rives JL, Warlaumont CR (1984) The use of Dacron in the repair of hernias of the groin. Surg Clin North Am 64: 269285 10. Taylor RS, Fiennes AGTW (1992) A tension free modification of the Dulucq preperitoneal laparoscopic hernioplasty. Min Inv Ther 1(Suppl 1): 101 11. Vanclooster P, Meersmann AL, de Gheldere CA, Van de Ven CK (1996) The totally extraperitoneal laparoscopic hernia repair. Surg Endosc 10: 332335 12. Vivekanadan S, Fiennes AGTW (1995) Totally extraperitoneal groin hernioplasty: mechanism of delayed indirect recurrence. Min Inv Ther 4(Suppl 1): 55

A. Fiennes
Department of Surgery St Georges Hospital Medical School Cranmer Terrace London, SW17 ORE, United Kingdom

J. Himpens
Department of Digestive Surgery University Hospital Ste Pierre Rue Haute 201 B-1000 Brussels, Belgium

Surg Endosc (1997) 11: 697

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank you for the opportunity to answer the letter of Mr. Fiennes. First of all, we certainly do acknowledge that Mr. Dulucq and Mr. Mc Kernan were the pioneers of the extraperitoneal laparoscopic hernia repair. Honor to whom honor is due. We did not intend to pretend to be the pioneers of this technique. We wished only to describe the technique the way we perform it, to describe our own findings, and to give a fair report of our preliminary results. The reason for cutting the inferolateral corner is not just that we think it fits better on the iliopsoas but also because we are afraid to cause damage to the nerves running on it by dissecting unnecessarily high on the muscle. Since the mesh measures 15 15 cm, we do not think we compromise the strength of the repair by merely removing a small piece of its inferolateral corner. We think that dissecting very high on the muscle just to position the whole inferolateral corner of the mesh flat on the muscle is unnecessary and dangerous. We do agree totally that fixation of the mesh is unnecessary provided the mesh is large enough, which is obviously the case when using a 15 15 cm mesh. In fact, we have not fixed the mesh since January 1996. We also agree that the vas deferens is not clearly seen on Fig. 1. We chose this shot because of the clearly visible large direct defect.

C. de Gheldere
Heilig Hart Ziekenhuis Kolveniersvest 20B-2500 LIER Belgium

P. Vanclooster
Bouwelsesteenweg 6 2560 Nijlen Belgium

Surg Endosc (1997) 11: 698

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The influence of pneumoperitoneum on the peritoneal implantation of free intraperitoneal cancer cells
Recently Hubens et al. published an interesting article entitled The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells [2]. They reported on the possible implantation of cancer cells at trocar wounds at the moment of deflation as cells are forced through these wounds by the pressure gradient created by the pneumoperitoneum during laparoscopic surgery for malignant disease. We would like to point out that this chimney effect, as originally described by us, can occur during the entire laparoscopic procedure and not only at the moment of deflation, as leakage of CO2 alongside trocars during surgery is impossible to prevent with the existing trocars [3]. Consequently, deflation of the pneumoperitoneum by letting CO2 escape through one of the trocars before pulling these trocars out of the abdomen will not prevent the occurrence of entrapment of cancer cells in the trocar wounds. We fully agree with the authors on the possible advantage of gasless laparoscopy as this could prevent the chimney effect. In our experimental work we found significantly less tumor growth at the port sites following gasless laparoscopic surgery for colon cancer in the rat as compared to laparoscopic surgery using a CO2 pneumoperitoneum. This technique seems promising to treat malignant disease laparoscopically.
Correspondence to: G. Kazemier

References
1. Bouvy ND, Marquet RL, Jeekel J, Bonjer HJ (1996) Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Surg Endosc 10: 551 2. Hubens G, Pauwels M, Hubens A, Vermeulen P, Van Marck E, Eyskens E (1996) The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells. Surg Endosc 10: 809812 3. Kazemier G, Bonjer HJ, Berends FJ, Lange JF (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 11411142

G. Kazemier1 F. J. Berends1 N. D. Bouvy1 J. F. Lange2 H. J. Bonjer1

1

Department of Surgery University Hospital Rotterdam-Dijkzigt Dr Molewaterplein 40 3015 GD, Rotterdam The Netherlands 2 Department of Surgery St. Clara Hospital Rotterdam The Netherlands

Surg Endosc (1997) 11: 699

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank Dr. Kazemier et al. for their kind remarks and fully agree with them that gas leakage can occur during the entire procedure with subsequent implantation of tumor cells at the trocar sites. At the moment we are conducting further experimental studies on the possible effects of gas leakage on tumor cell implantation and the chimney effect, as they have called it. Results will be ready for publication soon. G. Hubens
Department of Surgery University Hospital University of Antwerp Wilrijkstraat 10 2650 Edegem Belgium

News and notices

Surg Endosc (1997) 11: 700702

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

New Address for the European Association for Endoscopic Surgery (E.A.E.S.)
Effective January 1, 1997, the new correspondence, telephone, and fax numbers of the E.A.E.S. office are: E.A.E.S. Office, c/o Mrs. Ria Palmen Luchthavenweg 81 Unit 1.42 5657 EA Eindhoven The Netherlands or: P.O. Box 335 5500 AH Veldhoven The Netherlands Tel: +31 40 2525288 Fax: +31 40 2523102

Essentials of Laparoscopic Surgery Surgical Skills Unit University of Dundee Scotland, UK

Under the direction of Professor A. Cuschieri the Surgical Skills Unit is offering a three-day practical course designed for surgeons who wish to undertake the procedures such as laparoscopic cholecystectomy. This intensely practical program develops the necessary operating skills, emphasizes safe practice, and highlights the common pitfalls and difficulties encountered when starting out. Each workshop has a maximum of 18 participants who will learn both camera and instrument-manipulation skills in a purpose-built skills laboratory. During the course there is a live demonstration of a laparoscopic cholecystectomy. The unit has a large library of operative videos edited by Professor Cuschieri, and the latest books on endoscopic surgery are on display in our Resource area. Course fee including lunch and course materials is $860. For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042

Volunteer Surgeons Needed Northwestern Nicaragua Laparoscopic Surgery Teaching Program, Leon, Nicaragua
Volunteer surgeons are needed to tutor laparoscopic cholecystectomy for this non-profit collaboration between the Nicaraguan Ministry of Health, the National Autonomous University of Nicaragua, and Medical Training Worldwide. The program consists of tutoring general surgeons who have already undergone a basic laparoscopic cholecystectomy course. Medical Training Worldwide will provide donated equipment and supplies when needed. For further information, please contact: Medical Training Worldwide Ramon Berguer, MD, Chairman Tel: 707-423-5192 Fax: 707-423-7578 e-mail: berguer.r@martinez.va.gov

Advanced Endoscopic Skills Surgical Skills Unit University of Dundee Scotland, UK

Each month Professor Cuschieri Surgical Skills Unit offers a 412 day course in Advanced Endoscopic Skills. The course is intensely practical with hands on experience on a range of simulated models. The program is designed for experienced endoscopic surgeons and covers advanced dissection techniques, extracorporeal knotting techniques, needle control, suturing, internal tying technique, stapling, and anastomotic technique. Individual workstations and a maximum course number of 10 participants allows for personal tuition. The unit offers an extensive collection of surgical videos and the latest books and publications on endoscopic surgery. In addition, participating surgeons will have the opportunity to see live advanced laparoscopic and/or thoracoscopic procedures conducted by Professor Cuschieri and his team. The course is endorsed by SAGES. Course fee including lunch and course materials is $1850. For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042

Fellowship in Minimally Invasive Surgery George Washington Medical Center Washington, DC USA
A one-year fellowship is being offered at the George Washington University Medical Center. Interested candidates will be exposed to a broad range of endosurgical Education and Research Center. Active participation in clinical and basic science research projects is also encouraged. For further information, please contact: Carole Smith 202-994-8425 or, send curriculum vitae to: Dr. Jonathan M. Sackier Director of Endosurgical Education and Research George Washington University Medical Center Department of Surgery 2150 Pennsylvania Avenue, N.W. 6B-417 Washington, DC 20037, USA

The Practical Aspects of Laparoscopic Fundoplication Surgical Skills Unit University of Dundee Scotland, UK
A three-day course, led by Professor Cuschieri, designed for experienced laparoscopists wishing to include fundoplication in their practice. The

701 course covers the technical details of total and partial fundoplication using small group format and personal tuition on detailed simulated models. There will be an opportunity to observe one of these procedures live during the course. Maximum course number is six. Course fee including lunch is $1850. For further details and a brochure please contact: Julie Struthers, Unit Co-ordinator Surgical Skills Unit Ninewells Hospital and Medical School Dundee DD1 9SY Tel: +44 382 645857 Fax: +44 382 646042 to enable the surgeon to improve his or her laparoscopic dexterity, efficiency, and creativity. Exact and meticulous technique is emphasized so that the surgeon can apply these skills with confidence. Personal instruction is provided by Zoltan Szabo, Ph.D., F.I.C.S., Director of the MOET Institute, and surgeons are allowed to progress their own pace. Each participant has sole use of a laparoscopic training station equipped with high-quality clinical laparoscopic equipment and instrumentation. Inanimate, animal tissue, and optional live animal models are utilized. Features of these program include: fluently choreographed instrument movements; economy of movement and flawless technique; needle and suture handling skills (2-0 to 7-0); precision suturing, knotting, ligature, and anastomosis techniques; atraumatic, hemostatic tissue handling and dissection; optimal angles of approach (coaxial alignment of setup and geometry of port positioning); laparoscopic surgical strategy, technical nuances, and troubleshooting; visual perception problems and solutions; magnified eye-hand coordination; and two-handed (ambidextrous) technique. Courses are offered year-round by individual arrangement. The MOET Institute is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians and designates these CME activities for 2040 credit hours in Category 1 of the Physicians Recognition Award of the American Medical Association. These programs are also endorsed by the Society of Gastrointestinal Endoscopic Surgeons (SAGES). For further information, please contact: Wanda Toy, Program Administrator Microsurgery & Operative Endoscopy Training (MOET) Institute 153 States Street San Francisco, CA 94114, USA Tel: (415) 626-3400 Fax: (415) 626-3444

Courses at George Washington University Endosurgical Educational and Research Center

George Washington University Endosurgical Educational and Research Center is proud to offer a wide range of surgical endoscopy courses. These courses include advanced laparoscopic skills such as Nissen fundoplication, colon resection, common bile duct exploration, suturing, as well as subspecialty courses. Individual surgeons needs can be met with private tuition. The Washington D.C. area is a marvelous destination to visit for recreational pursuits which can be arranged by the facility to suit your personal agenda. For further details please contact: Carole Smith: Department of Surgery 2150 Pennsylvania Avenue NW 6B Washington, DC 20037, USA Tel: (202) 994-8425

Courses at the Royal Adelaide Centre for Endoscopic Surgery

Basic and Advanced Laparoscopic Skills Courses are conducted by the Royal Adelaide Centre for Endoscopic Surgery on a regular basis. The courses are limited to six places to maximize skill development and tuition. Basic courses are conducted over two days for trainees and surgeons seeking an introduction to laparoscopic cholecystectomy. Animal viscera in simulators is used to develop practical skills. Advanced courses are conducted over four days for surgeons already experienced in laparoscopic cholecystectomy who wish to undertake more advanced procedures. A wide range of procedures are included, although practical sessions can be tailored to one or two procedures at the participants request. Practical skills are developed using training simulators and anaesthetised pigs. Course fees: $A300 ($US225) for the basic course and $A1,600 ($US1,200) for the advanced course. For further details and brochure, please contact: Dr. D. I. Watson or Professor G. G. Jamieson The Royal Adelaide Centre for Endoscopic Surgery Department of Surgery Royal Adelaide Hospital Adelaide SA 5000 Australia Tel: +61 8 224 5516 Fax: +61 8 232 3471

Courses at WISE Washington Institute for Surgical Endoscopy Washington, DC, USA
The Washington Institute of Surgical Endoscopy is pleased to offer the following courses: Laparoscopic antireflux and hiatal hernia surgery (July 1415, 1997); Laparoscopic management of the common bile duct and difficult cholecystectomy (May 1516, August 1112, November 1011, 1997); Laparoscopic colon and rectal surgery (June 2021, September 1516, December 45, 1997). Also, courses for operating room nurses and technicians will be run on a monthly basis and personal instruction and preceptorship is available. For further information, please call: Carole Smith Washington Institute of Surgical Endoscopy 2150 Pennsylvania Avenue, N.W. Washington, DC 20037 Tel: 202-994-9425

4th International Meeting on Laparoscopic Surgery May 17, 1997 Berne, Switzerland
Main topic: Acute appendicitis: Standard treatment and the role of laparoscopic surgery For further information, please contact: Mrs. Caroline Zrcher Klinik fr Viszerale und Transplantationschirurgie Universitt Bern Inselspital CH-3010 Bern, Switzerland Tel: +41 31 632 97 22 Fax: +41 31 632 97 23

Advanced Laparoscopic Suturing and Surgical Skills Courses MOET Institute San Francisco, CA, USA
This intensive hands-on training program is intended to help the surgeon develop proficiency in the essential laparoscopic surgical techniques. A sequence of progressively challenging exercises has been designed

702

European Course on Laparoscopic Surgery (French language) May 1316, 1997 (English language) November 1821, 1997 Brussels, Belgium
Course director: G.B. Cadiere For further information, please contact: Administrative Secretariat Conference Services s.a. Avenue de lObservatoire, 3 bte 17 B-1180 Bruxelles Tel: (32 2) 375 16 48 Fax: (32 2) 375 32 99

Fellowships in Laparoscopic Surgery Staten Island University Hospital Staten Island, NY USA
A one year fellowship, to start July 1, 1997, in advanced laparoscopic surgery is being offered at Staten Island University Hospital. The selected fellow will be exposed to many advanced general laparoscopic surgeries including: hiatal hernia repair, splenectomy, adrenalectomy, bowel resection, and others. Participation in research projects will be encouraged. For further information, please contact: Barbara Coleman Coordinator, Surgical residency program Tel: 718-226-9508

Joint Euro Asian Congress of Endoscopic Surgery 5th Annual Congress of the European Association for Endoscopic surgery (EAES) 3rd Asian-Pacific Congress of Endoscopic Surgery June 1721, 1997 Istanbul, Turkey
The Congress will include a joint postgraduate course EAES/SAGES/ ELSA on June 17th. For information and registration: SETUR Congress Department Cumhuriyet Cad. No. 107 80230 Elmadag Istanbul, Turkey Tel: (90.212) 23003 36 Fax: (90.212) 240 82 37

9th International Meeting Society for Minimally Invasive Therapy July 1416, 1997 Kyoto, Japan
Scientific program to include: Plenary, Parallel, Poster, and Video sessions. Host Chairman: Professor Osamu Yoshida, Department of Urology, Kyoto University, 54 Shogoin Kawahara-sho, Sakyo, Kyoto 606, Japan. Phone: +81 75 751-3328, Fax: +81 75 751-3740. This meeting coincides with the Gion Festival in Kyoto, one of the greatest festivals in Japan. For further information, please contact: Secretariat of SMIT 9th Annual International Meeting c/o Academic Conference Planning 383 Murakami-cho Fushimika, Kyoto 612 Japan Tel: +81 75 611-2008 Fax: +81 75 603-3816

Fellowships in Minimally Invasive Surgery The University of Pittsburgh Medical Center Pittsburgh, PA, USA
One year fellowships in advanced minimally invasive surgery in both general and thoracic surgery are being offered at the University of Pittsburgh Medical Center beginning on July 1, 1997. Requirements include completion of residency training programs in the desired area. The fellowships include a competitive salary and travel allowance. Interested candidates should send a letter of inquiry with curriculum vitae to: Philip R. Schauer, MD (General Surgery) or James Luketich, MD (Thoracic Surgery) The University of Pittsburgh Medical Center 3471 Fifth Avenue Suite 300 Pittsburgh, PA 15213-3221

6th World Congress of Endoscopy Surgery Roma 98 6th International Congress of European Association for Endoscopic Surgery June 36, 1998 Rome, Italy
The program will include: the latest, original high quality research; symposia; plenary lectures; abstract presentations (video, oral, and posters); EAES and SAGES postgraduate courses, OMED postgraduate course on therapeutic endoscopy; working team reports; educational center and learning corner; meeting of the International Society of Nurses and Associates; original and non original scientific reports; and a world expo of new technology in surgery. For further information, please contact: Congress Secretariat: Studio EGA Viale Tiziano, 19 00196 Rome, Italy Tel: +39 6 322-1806 Fax: +39 6 324-0143

Surg Endosc (1997) 11: 658662

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

First results of laparoscopic gastrostomy

K. Peitgen, M. K. Walz, U. Krause, F. W. Eigler
Department of General Surgery, University of Essen, Hufelandstrae 55, D-45122 Essen, Germany Received: 5 March 1996/Accepted: 31 July 1996

Abstract Background: Laparoscopic gastrostomy as an alternative to open gastrostomy was introduced with various technical variants 5 years ago. However, long-term results of these new methods are still lacking. Methods: From 4/1993 to 2/1996, laparoscopic gastrostomies were performed on 42 patients (50.9 15.6 [2471] years) with esophageal stenosis in locally advanced hypopharyngeal (17 patients) or oropharyngeal (nine patients) carcinoma, incurable esophageal carcinoma (13 patients) and cerebral dyspagia (three patients). Operating time was 38 11 min [1565 min]. Procedure-related mortality was 0%. Major complications occurred in 2/42 (4.7%) patients; minor complications were found in 4/42 (9.4%) patients. During a total usage time of 427 months, 14 stoma infections occurred (0.11 infections/100 days). Conclusion: Laparoscopic gastrostomy allows a safe, fast, and cheap reestablishment of enteral nutrition. The procedure is minimally invasive and can also be performed under local anesthesia. It has become our method of choice in patients with malignant, nonresectable subtotal stenosis of the hypopharynx or esophagus. Key words: Laparoscopy Gastrostomy Enteral nutrition

24]. We introduced laparoscopic gastrostomy as a routine procedure for patients with total or subtotal esophageal stenosis under palliative therapy 30 months ago [23] and hereby present our first long-term results. Methods
From April 1993 to February 1996, laparoscopic gastrostomies have been performed in 42 patients (24 male, 18 female; age 50.9 15.6 [2471] years). Twenty-eight patients presented with locally advanced hypopharyngeal (17 patients) or oropharyngeal (nine patients) carcinoma and 13 patients suffered from incurable esophageal carcinoma with subtotal stenosis or complete obstruction of the esophagus. Oral nutrition and gastroscopic pertubation were not possible in all these cases. One patient suffered from a neurologically caused dysphagia after severe head injury; two other patients had severe cerebral dysphagia after meningitis (one patient) or removal of a cerebral tumor (one patient). Three patients suffered from liver cirrhosis due to alcohol abuse. One patient with a hypopharyngeal carcinoma had a previous PEG that was removed 3 months prior to laparoscopic gastrostomy. At laparoscopy, only a thin occluded scar approximately 5 cm long at the former PEG insertion site was found, and it was left in place. Here, laparoscopic gastrostomy could be performed without problems 2 cm distal to the former PEG insertion site. Thirty-four patients had adjuvant or palliative therapy concerning their primary disease (10 patients chemotherapy, 15 patients radiotherapy, nine patients combined chemo-/radiotherapy) prior to laparoscopic gastrostomy. Six patients had previous upper abdominal surgery: Three had gastric resections (Billroth I) 8, 11, and 15 years prior to gastrostomy; two other patients had an open cholecystectomy 5 and 10 years before gastrostomy. Another patient had an open cholecystectomy with common bile duct exploration 15 years ago. In these six cases, we performed an open laparoscopy, choosing a 1.5-cm incision in the umbilical region. After opening the peritoneum under visual control, we then inserted a trocar-sheath and insufflated the abdomen under laparoscopic control. The procedures were performed by eight surgeons. Forty-one patients had general anesthesia during the procedure; one patient had local anesthesia with 60 ml of 0.25% bupivacaine and intravenous sedation with 200 mg propofole. Our operative technique for the laparoscopically assisted gastrostomy has been described in detail recently [23]. As experience progressed, this technique has been slightly modified: After creating a pneumoperitoneum of 12 mmHg using a Veress needle under general anesthesia, two 10-mm trocars and one 5-mm trocar are placed into the abdominal cavity. The first 10-mm trocar is inserted below the umbilicus, the second 10-mm trocar is placed into the left lower quadrant under laparoscopic control. The optic equipment is then adapted to this second trocar. This modification allows a good exposure of the stomach. A 5-mm trocar is placed into the right upper quadrant for an atraumatic grasper.

Percutaneous endoscopic gastrostomy (PEG) has become a safe and effective alternative to surgical gastrostomy in patients who require enteral access or gastric decompression [11]. Technical considerations, however, limit the application of PEG. In patients with subtotal stenosis or a complete obstruction of the hypopharynx or esophagus, e.g., as seen in locally advanced hypopharyngeal or esophageal carcinoma, surgical gastrostomy is indicated [3]. Recently, laparoscopic gastrostomy tube implantation has been introduced in miscellaneous technical variants as an alternative to surgical open gastrostomy and to PEG [6, 8, 16, 20, 23,
Correspondence to: K. Peitgen

659

Fig. 1. The three stay sutures are placed, creating a triangle at the stomach wall for the future three-point gastropexy. The stomach is punctured and the guidewire inserted. Fig. 2. Radiography of a laparoscopic gastrostomy (patient No. 3, total esophageal stenosis due to hypoharyngeal carcinoma). Fig. 3. Radiography of a laparoscopic gastrostomy after Billroth I resection (patient No. 26, total esophageal stenosis due to hypoharyngeal carcinoma). The anterior wall of the stomach is identified at the region of the corpus. Using a needle-holder and a grasper, three full-thickness stitches (2-0; e.g., Dexon) are applied, creating a triangle at the level of the stomach wall. In order to fix the corresponding ones at the end of the procedure subcutaneously, both ends of the sutures are pulled through the abdominal wall by a special awl beneath the left costal margin (Storz, Germany; No. 26173 AK). A 1-cm stab incision of the skin and fascia is then performed in the center of the skin-level triangle built by the three sutures. Under laparoscopic control, the anterior wall of the stomach is punctured percutaneously, exactly in the center of the two corresponding triangles at the skin level and at the stomach level in a strict rectangle to the stomach wall. Through the 18-gauge needle, a guidewire is inserted into the stomach (Fig. 1). A 26-French dilator with a peel-away sheath (e.g., C-Pli-26-38; Cook, Mo nchengladbach, Germany) is pushed over the guidewire into the stomach percutaneously. After removing the inner dilator, a 24-French urinary catheter (e.g., Foley-catheter, Ru sch, Kernen, Germany) is placed through the maintaining peel-away sheath into the stomach. After inflating the balloon of the catheter, the stomach is then pulled to the abdominal wall and the sutures are fixed under traction and progressive reduction of the pneumoperitoneum. At the end of the procedure the correct placement of the tube is confirmed radiographically (Figs. 2, 3). Feeding through the tube commenced 12 to 24 h postoperatively. After 2 days of feeding with water, tea, and other liquids, the patients were instructed to apply normal, mashed foods into the tube and to maintain their nutritive habits as close as possible to the time prior to their disease. The Foley catheter was routinely changed once after 10 to 14 days. Further exchanges depended on actual needs. The last 32 of our 42 patients received a skin-level gastrostomy button device 2 weeks after laparoscopic gastrostomy during outpatient treatment. Data of all patients have been collected prospectively. All patients were routinely seen once a week in the first month. Further follow-ups were kept up by contact with the patients family doctors. If possible, data are expressed as mean mean standard error (M SEM).

Results The laparoscopic gastrostomies could be performed successfully in all cases. The mean duration of the procedure was 38 11 min (1565 min). In the three patients after gastric resection, remarkable laparoscopic adhesiolysis was necessary in order to identify the corpus region and the gastroduodenal anastomosis. In the three patients after cholecystectomy, only moderate adhesiolysis was necessary. Operating time (53 6 [4575] min; first cut to last suture) in the six patients with prior surgery was significantly longer (p < 0.05, Students t-test) compared to the 36 patients without prior surgery to the abdomen (34 4 [2060] min). None of the patients died related to the procedure. One patient with a hypopharyngeal carcinoma died of a myocardial infarct 3 days after surgery; another patient with an esophageal carcinoma died 4 days after surgery due to a severe pneumonia based on an esophagopulmonary fistula which developed 2 days after surgery.

660

In one patient with a complete stenosis of the lower esophagus due to a cardia carcinoma, a gastric perforation was noticed radiologically 5 h after gastrostomy. This patient suffered from acute abdominal pain. At laparotomy, a small lesion of the gastric wall was found at a 5-cm distance medially to the stoma. After suturing of this lesion, the further course of this patient was uneventful. The genesis of this injury remains unclear, but most likely was caused by unskilled use of the grasping forceps. In another patient, a leakage of contrast fluid was suspected intraoperatively by a falsely interpreted radiography. The suspected perforation of the posterior wall could be excluded by a small midline incision. No further major postoperative complications related to the procedure were observed. In four procedures, the transgastric fixation sutures were cut by the stab incision (one case) or torn apart during catheter placement (three cases). In three cases, the sutures could be repeated laparoscopically without problems. In one case, the suture was refixed after moderate enlargement through the stoma incision. In one patient with a hypertrophic left liver lobe due to severe cirrhosis, an additional 5-mm trocar had to be inserted in order to hold the lobe aside with a palpation probe. The mean usage time is now 10.4 1.3 (125) months. To date (February 28th, 1996), 12 patients have died of their primary disease after 7.8 6.2 (115) months with their gastrostomies; 30 patients are still alive with their gastrostomies after 9.2 8.1 (131) months. During a total usage time of 12,819 days (427 months), 14 stoma infections occurred (0.11 infections in 100 days). One diabetic patient developed a dumping syndrome with attacks of hypoglycemia after stoma feeding. These were treated by insulinedose adaptation and antisecretoric medication. No bleeding events, episodes of diarrhea, tube displacements, stoma leakages or other complications occurred during long-term follow-up. One gastrostomy was removed without problems in the patient with severe head injury after 4 weeks when dysphagia resolved. Thirty-six of 40 patients (excluding two patients who died within 3 days after surgery) were able to nourish themselves mainly on normal sieved foods. Four of the surviving patients were unable to use mashed foods, mostly due to social reasons, and therefore applied pharmaceutically prepared liquid food. Bolus feeding and continuous feeding were both possible in all patients.

Table 1. Indications for laparoscopic gastrostomy in 42 patients (March 1993February 1996) Indication Oropharyngeal/ laryngeal carcinoma Hypopharyngeal carcinoma Esophageal carcinoma Dysphagia Patients 9 17 Abdominal pathology or prior surgery 1 patient after gastric resection (Billroth I) 2 patients after gastric resections (Billroth I) 1 patient after open cholecystectomy 3 patients with liver cirrhosis 2 patients after open cholecystectomy 1 polytrauma, 1 patient after cerebral tumor removal, 1 patient after meningitis

13

Discussion Placement of feeding tubes is a common procedure for general surgeons [3, 12, 22]. The introduction of PEG has changed and improved surgical practice [5, 19]. However, this technique is contraindicated or not feasible under certain conditions. Ascites, previous gastric or other surgery to the upper abdomen, severe esophageal stenosis due to tumors of the oro-/epi-/hypopharynx or the esophagus present contraindications to PEG. In these cases, a surgical gastrostomy is still indicated (Table 1). We have developed a laparoscopic variant of the classic Kader-/Stamm gastrostomy [23] on patients with subtotal or total esophageal and used it routinely in the last 34 months. The essential features of our procedure are the transabdom-

inal fixation sutures, the wire-guided dilator (which ensures safe and tight catheter placement), and the large catheter sizes (which allow easy feeding with normal mashed foods). Prior surgery to the upper abdomen does not limit the feasibility of laparoscopic gastrostomy. After only minor technical modifications (open laparoscopy, laparoscopic adhesiolysis) we were able to perform laparoscopic gastrostomies in six patients after previous upper abdominal operations. No further trocar accesses were needed. The operating time was lengthened significantly, but postoperative complications, food intake, and hospital stay were equal compared to the patients without previous surgery. To our knowledge, there is no other description of laparoscopic gastrostomy after upper abdominal surgery. The laparoscopically applied and percutaneously placed sutures are most important for safety reasons. They allow easy and precise placement of the needle, guidewire, and catheter. Furthermore, they prevent peritonitis in case of an early accidental tube loss. Additionally, tube changes are facilitated. In comparison to other variants of laparoscopic gastrostomies [8, 9], nonabsorbable fixation devices are not needed. Thus the risk of secondary perforation due to gastric wall necrosis should be reduced. In case of inadvertent intraoperative rupture of these sutures, laparoscopic reapplication is possible. No special catheters are needed as proposed by other authors. In our experience, intraoperative radiography control using a water-soluble contrast liquid is mandatory in order to exclude leakage either due to inadvertent gastric perforation or insufficient fixation. In the patient (number 36 in our series) with a gastric wall rupture, this lesion was not detected by laparoscopic control. Routine intraoperative radiography would have detected the lesion immediately. Many technical modifications of laparoscopic gastrostomies have been published since the first description in 1991 [24] (Table 2). Most authors used modifications of the conventional Stamm-type gastrostomy first described by Russell [26] using stay sutures [1, 21, 30], T-fasteners or anchors [8, 9], a combination of stay and pursestring sutures [14, 17], fibrin glue [29], and various catheters (Malecot, Foley, suprapubic catheters, Moss tube) in various sizes from 12 to 18 French. The majority of authors [1, 6, 20, 21, 30] used general anesthesia for their procedures; only two [10, 17] also report

661 Table 2. Published clinical results of laparoscopic gastrostomy 19921996 Reference 7 Year 1992 n 3 Catheter type/size 18 Fr
a

Trocars 3

Technical features Anisoperistaltic Janeway gastrostomy, two 30-mm endoscopic staplers 2 stay sutures No data 4 T-fasteners 2 stay sutures 24 anchors, dilator-set, peel-away sheath 2 stay sutures 2 pursestring sutures + 2 stay sutures Isoperistaltic Janeway gastrostomy, two 30-mm endoscopic staplers 3 stay sutures, guidewire, dilator set, peel-away sheath

Anesthesia General

Time (minutes) No data

Complications None

21 6 8 1 10

1992 1993 1993 1993 1994

9 13 3 3 22

Various/18 Fr Foley Foley/18 Fr Moss tube Foley or Malecot/ 1214 Fr Suprapubic catheter/ 15 Fr MIC silicon gastrostomy tube 20 Fra

3 3 2 3 23

General General General General 17 general, 5 local General 6 general, 3 local General

3560 No data No data 30 20.5 (1025)

30 17 18

1994 1994 1995

10 9 5

4 2 2

No data 22 (1533) 30 (2535)

1 gastric bleeding ( laparotomy) None None No data 1 dislocation ( laparotomy) 1 death due to peritonitis None 1 dislodgment ( laparotomy) None

This study

1996

42

Foley/24 Fr

41 general, 1 local

38 (1565)

1 gastric rupture ( laparotomy)

20-Fr Foley catheter for the first days and at regular intervals for enteral feeding

on laparoscopic gastrostomies under local anesthesia in individual indications. Gasless laparoscopy as an alternative approach to gastrostomy has also been reported [29]. The laparoscopically modified Janeway gastrostomy [13, 18, 20] forms a gastric tube by using a special linear stapling device and thus forms a potentially continent and irrigable stoma. Since a laparoscopic stapler and two magazines are required for this technique, this procedure is more difficult to perform and more expensive than the Stammtype procedures. Long-term results have not yet been reported. Until 1995, only 11 cases in two different modifications had been documented in the literature [13, 18, 20]. Furthermore, the potential continence of this variant remains debatable [27]. Usually, laparoscopic gastrostomy is a permanent procedure. However, it can be easily undone by simple removal of the catheter, if indicated. Complete restitution is achieved after three to five days without complications. In our series, one gastrostomy in a patient with cerebral dysphagia after polytrauma with severe brain damage was removed after 4 weeks without problems. In contrast to our method, removal of Janeway-type gastrostomies requires additional surgery [7, 13, 16, 18]. In comparison to the traditional gastrostomy, postoperative pain is reduced, enteral feeding is possible within 12 h, and hospitalization is short [10]. These facts are of major importance, especially in patients with palliative treatment. Additionally, palliative therapy (e.g., chemotherapy, radiation therapy)if indicatedcan be applied earlier compared to open gastrostomy. In contrast to the PEG, normal mashed food can principally be used for nutrition due to the large catheter lumen. This is of psychological and practical importance for the patients and also reduces costs. Also, endoscopic techniques do not uniformly secure the tube to the abdominal wall. The

latter may induce possible complications, including leakage, accidental early tube removal, bowel obstruction, intraperitoneal catheter migration, intestinal fistulas, and necrosis of the stomach or abdominal wall because of excessive traction. There is some information that suggests that the outcome of PEG is experience- and discipline-related [2, 28]. Many references indicate a lower morbidity of PEG in short-term studies compared to conventional gastrostomies [12, 25, 28] or nasogastric feeding [22]. However, only few data exist concerning results of long-term use of PEG and its complications [15, 22]. Our long-term complication rate of 0.11 on 100 usage days reflects the usually overestimated morbidity of surgical gastrostomies [12, 28] quite well. On the contrary, other series reporting on conventional Stammtype gastrostomies demonstrated a significantly lower intraoperative morbidity but no significantly different long-term morbidity compared to PEG [2, 22]. Complications related to aspiration [4] should not occur in our patients, as all but three patients were unable to swallow due to complete or subtotal esophageal stenosis. Our results after 42 laparoscopic gastrostomies show that this technique allows a safe, fast, and cheap reestablishment of enteral nutrition in patients with hypopharyngeal or esophageal stenosis. The procedure has the wellknown advantages of minimally invasive surgical techniques and can also be performed under local anesthesia. It has become our method of choice in patients with malignant, nonresectable subtotal stenosis of the hypopharynx or esophagus. Regarding the practical advantages of laparoscopic gastrostomies and the possibility of using normal mashed foods, further indications for laparoscopic gastrostomies under local anesthesia should be considered in cases of dysphagic patients requiring long-term gastrostomy feeding.

662

References
1. Albrink MH, Hagan K, Rosemurgy AS (1993) Laparoscopic insertion of the Moss feeding tube. J Laparoendosc Surg 3: 531534 2. Apelgren KN, Zambos J (1989) Is percutaneous better than open gastrostomy? A clinical study in one surgical department. Am Surg 55: 596600 3. Au FC (1993) The Stamm gastrostomy: a sound procedure. Am Surg 59: 674675 4. Burtch GD, Shatney CH (1985) Feeding gastrostomy: assistant or assassin? Am Surg 51: 204207 5. Campos AC, Butters M, Meguid MM (1990) Home enteral nutrition via gastrostomy in advanced head and neck cancer patients. Head Neck 12: 137142 6. Chambon JP, Wurtz A, Saudemont A, Bosse JL, Chevalier D, Quandalle P (1993) Celioscopic gastrostomies and jejunostomies. Ann Chir 47: 311315 7. Cossa JP, Marmuse JP, Lecomte P, Le Goff JY, Johanet H, Benhamou G (1992) Tubular gastrostomy using celioscopy. Presse Med 21: 15191521 8. Duh QY, Way LW (1993) Laparoscopic gastrostomy using t-fasteners as retractors and anchors. Surg Endosc 7: 6063 9. Edelman DS, Unger SW, Russin DR (1991) Laparoscopic gastrostomy. Surg Laparosc Endosc 1: 251253 10. Edelman DS, Arroyo PJ, Unger SW (1994) Laparoscopic gastrostomy versus percutaneous endoscopic gastrostomy: a comparison. Surg Endosc 8: 4749 11. Gauderer MW (1992) Gastrostomy techniques and devices. Surg Clin North Am 72: 12851298 12. Grant JP (1988) Comparison of percutaneous endoscopic gastrostomy with Stamm gastrostomy. Ann Surg 207: 598603 13. Haggie JA (1992) Laparoscopic tube gastrostomy. Ann R Coll Surg Engl 74: 258259 14. Horbach T, Reck T, Ko ckerling F (1994) Perkutane laparoskopische Gastrostomie (PLG) unter Verwendung einer laparoskopischen Tabaksbeutelklemme. Min Inv Chir 3: 179183 15. Hull MA, Rawlings J, Murray FE, Field J (1993) Audit of outcome of 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

26. 27. 28.

29. 30.

long-term enteral nutrition by percutaneous endoscopic gastrostomy. Lancet 341: 869872 Lathrop JC, Felix EJ, Lauber D (1991) Laparoscopic Janeway gastrostomy utilizing an endoscopic stapling device. J Laparoendosc Surg 1: 355359 Lee WJ, Chao SH, Chang S, Chen KM (1994) Laparoscopic assisted gastrostomy tube placement. J Laparoendosc Surg 4: 201204 Meyer G, Rau H, Strauss T (1995) Die laparoskopische Gastrostomie nach Janeway. Chirurg 66: 719723 Miller RE, Castlemain B, Lacqua FJ, Kotler DP (1989) Percutaneous endoscopic gastrostomy results in 316 patients and review of literature. Surg Endosc 3: 186190 Modesto VL, Harkins B, Calton WC Jr, Martindale RG (1994) Laparoscopic gastrostomy using four-point fixation. Am J Surg 167: 273 276 Neufang T, Ko hler H, Lepsien G, Lu dtke FE (1992) Die perkutane laparoskopische Gastrostomie (PLG). Min Inv Chir 1: 34 Pargger H, Kern C (1995) Technische Aspekte der Rehydrierung. Chir Prax 50: 1322 Peitgen K, Walz MK, Krause U, Kruschke A, Eigler FW (1994) Laparoscopic Kader fistula. Chirurg 65: 213216 Reiner DS, Leitman IM, Ward RJ (1991) Laparoscopic Stamm gastrostomy with gastropexy. Surg Laparosc Endosc 1: 189192 Ruge J, Vazquez RM (1986) An analysis of the advantages of Stamm and percutaneous endoscopic gastrostomy. Surg Gynecol Obstet 162: 1316 Russell TR, Brotman M, Norris F (1984) Percutaneous gastrostomy: a new simplified and cost-effective technique. Am J Surg 184: 132137 Shellito PC, Malt RA (1985) Tube gastrostomy: techniques and complications. Ann Surg 201: 180185 Stern JS (1986) Comparison of percutaneous endoscopic gastrostomy with surgical gastrostomy at a community hospital. Am J Gastroenterol 81: 11711173 Viani PM, Poggi RV, Pinto A (1995) Gasless laparoscopic gastrostomy. J Laparoendosc Surg 5: 245249 Wolfram T, Albrecht R, Voigtsberger P (1994) Perkutane laparoskopische Gastrostomie. Chir Prax 48: 6368

Surg Endosc (1997) 11: 653654

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Port site electrosurgical (diathermy) burns during surgical laparoscopy

P. D. Willson,1 J. D. van der Walt,2 D. Moxon,1 J. Rogers1
1 2

Academic Surgical Unit, Royal London Hospital, London, E1 1BB, United Kingdom Department of Histopathology, Royal London Hospital, London, E1 1BB, United Kingdom

Received: 12 August 1996/Accepted 26 November 1996

Abstract Background: Direct and capacitive coupling of diathermy current have been reported as causes of occult injury during surgical laparoscopy. Methods: In order to determine the incidence of electrosurgical injury adjacent to metal and plastic cannulas, skin biopsies at 19 port sites used for monopolar electrosurgery were analyzed for coagulative necrosis. Prior to surgery the cannulas were randomized to either metal or plastic. Results: Coagulative necrosis was observed at nine electrosurgery port sites compared to only one control (2 4.872; df 1; 0.05 > p > 0.02). Plastic cannulas afforded no greater protection from skin burns than metal cannulas. Conclusions: Burns may be the result of direct or capacitive coupling to metal cannulas or capacitive coupling to the skin edge across plastic cannulas. The potential exists for burns to other tissues also in close proximity to a cannula used for electrosurgery. Key words: Electrosurgery Surgical diathermy Capacitive coupling Direct coupling Patient burns

Following surgery, skin biopsies were taken from the lower edge of the cannula wound. This was fixed in formalin, prepared for histological sectioning and stained with hematoxylin and eosin. Control biopsies were taken in the same patient from the lower edge of a cannula wound not used for electrosurgery. Sections were viewed after staining by the same consultant pathologist (J.D.W.) who was blind to both the cannula type and biopsy site. Coagulative necrosis within the epidermis was graded 0III (Table 1).

Results Eleven plastic and eight metal cannulae were used for electrosurgical instruments. Nine of the 19 electrosurgery port wounds had evidence of coagulative necrosis, nine showed no coagulative necrosis, and one biopsy was equivocal. Thirteen of the 15 control biopsies showed no coagulative necrosis, one revealed coagulative necrosis, and one was equivocal (Table 2). Equivocal biopsies were affected by crush artefact. The higher incidence of coagulative necrosis in the diathermy cannula group is statistically significant both when all diathermy wounds are included together (2 4.872 [Yates corrected]; df 1; 0.05 > p > 0.02) and when separated out by the type of cannula used (2 7.342; df 2; 0.05 > p > 0.02). Equivocal biopsies were incorporated with the group showing no necrosis.

Inadvertent monopolar electrosurgical burns occurring outside laparoscopic view have been reported following surgical laparoscopy [5, 6]. Although these have been attributed to direct coupling (live contact) or capacitive coupling (electromagnetic induction) the true incidence with which these phenomena cause patient injury is unknown. This study presents definitive evidence of burns caused by electrosurgery during surgical laparoscopy. Methods
Nineteen port sites in 15 consecutive patients undergoing laparoscopic cholecystectomy or fundoplication were randomized to receive either a metal or plastic cannula for electrosurgical instruments.

Discussion Coagulative necrosis occurs after sudden tissue ischemia, chemical injury, and thermal injury. The only likely cause for the coagulative necrosis seen in these patients is thermal injury from monopolar diathermy. Monopolar electrosurgery involves current passing between an active electrode and a return plate across the patient. The cannula represents one of several return paths to the patient plate. Burns may occur along any part of the return path if the power density in the tissue exceeds 7.5 W/cm2 [3]. The burns demonstrated in this study show that inadvertent coupling to the cannula occurs during electrosurgical laparoscopy and is more commonly associated with

Correspondence to: P. D. Willson, Department of Surgery, St Marys Hospital, Praed Street, London W2 1NY, United Kingdom

654 Table 1. Grading of epidermal burns Grading 0 I II III Definition No coagulative changes Coagulative changes in the epidermis, all layers intact Coagulative changes with loss of up to 50% of the epidermis Coagulative changes with loss of over 50% of the epidermis

Table 2. Epidermal burns by grade and cannula type Electrosurgical cannulas Burn grade 0 I II III Crush artefact: unable to assess Plastic 5 2 2 2 0 Metal 4 1 0 2 1 Nonelectrosurgical cannulas (control) 13 0 0 1 1

to bowel, however, may lead to more serious injury [5]. In addition, the data demonstrate that it is not the characteristics of the cannula or insulation materials which are the important determinant of skin burns but rather the close proximity of the wound to the electrosurgical instrument. Plastic cannulas afford no greater protection than metal cannulas against skin burns although the mechanism with plastic cannulas is capacitive coupling whereas with metal cannulas direct coupling may also be a cause. Theoretically the only electrosurgical instruments which can avoid this potential from both direct and capacitive coupling are bipolar instruments where the return electrode is guaranteed to be at the site of application of the surgical current. Coaxially shielded instruments are an alternative means of reducing the risk from capacitive coupling and insulation breaks but not from direct coupling. In general, reduction in the probability of electrosurgical burns may be achieved by using low diathermy generator settings (30 W or less) and low-voltage wave forms (blend or cut) [1]. References

those ports where diathermy is used. This may be due to direct contact or arcing to a metal cannula from an active electrode (direct coupling) or due to electromagnetic induction to a metal cannula or directly to skin across a plastic cannula (capacitive coupling) [24]. The single burn seen in the control group is most likely due to direct coupling to the control cannula. Although skin burns of the type identified in this study are of little clinical consequence to the patient, the study indicates that monopolar electrosurgical energy can be transmitted across both metal and plastic cannulas and injure tissue in close proximity to the cannula. Such proximity

1. Department of Health (1994) Diathermy injury during laparoscopic surgery. Safety Action Bull SAB(94)38. 2. McAnena OJ, Willson PD (1993) Diathermy in laparoscopic surgery. Br J Surg 80: 10941096 3. Tucker RD, Voyles CR, Silvis SE (1992) Capacitive coupled stray currents during laparoscopic and endoscopic electrosurgical procedures. Biomed Instrum Technol 26: 303311 4. Voyles CR, Tucker RD (1992) Education and engineering solutions for potential problems with laparoscopic monopolar electrosurgery. American J Surg 164: 5762 5. Willson PD, McAnena OJ, Peters EE (1994) A fatal complication of diathermy in laparoscopic surgery. Minim Invasive Ther 3: 1920 6. Willson PD, van der Walt JD, Rogers J (1995) Electrosurgical coupling to a metal cannula causing skin burns during laparoscopic surgery. Minim Invasive Ther 4: 163164

Surg Endosc (1997) 11: 632635

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Prospective evaluation of a minimally invasive approach for treatment of bile-duct calculi in the high-risk patient
M. Trias,1 E. M. Targarona,1 E. Ros,2 J. M. Bordas,3 R. M. Perez Ayuso,2 C. Balague ,1 I. Pros,1 J. Teres2
1 2

Service of General and Digestive Surgery, Hospital Clinic, University of Barcelona, Villarroel 170, 08036 Barcelona, Spain Gastroenterology Service, Hospital Clinic, University of Barcelona, Villarroel 170, 08036 Barcelona, Spain 3 Endoscopy Unit, Hospital Clinic, University of Barcelona, Villarroel 170, 08036 Barcelona, Spain Received: 17 May 1996/Accepted: 12 September 1996

Abstract Background: The best approach to bile duct stones in highrisk patients is controversial. We showed in a randomized trial that open surgery had a morbi-mortality similar to that of endoscopic sphincterotomy alone (ES) and less late biliary complications. The aim of this study was to evaluate a minimally invasive approach to duct stones in high-risk patients compared with open surgery or ES alone. Methods: Sixty high-risk patients (mean age 80 years) suspected of duct stones were treated by ES + laparoscopic cholecystectomy (LC). High-risk factors were: age > 70 years, Goldman cardiac index > 13, chronic pulmonary disease, liver cirrhosis, neurologic deficit, and severe obesity. Results: ERCP success was 87%. Duct stones were found in 75%. LC succeeded in 92%. Post-LC stay was 4 days. Overall morbidity was 19% and mortality was 3%. Recurrent symptoms (mean follow-up: 9 months) was 3.6%. When compared with open surgery or ES alone, ES + LC had a similar morbi-mortality, but shorter postop stay (p < 0.001). Late symptoms appeared in 20% after ES alone vs 4% after open surgery or ES plus LC (p < 0.04). Conclusions: Combined ES + LC is an effective alternative to open surgery or ES alone for treatment of duct stones in high-risk patients. Key words: Bile duct stones Laparoscopic cholecystectomy Endoscopic sphincterotomy High-risk patient Elderly

the gallbladder in situ was proposed as an alternative to open surgery in high-risk patients with bile duct calculi [19]. We recently showed in a prospective and randomized trial that open surgery and ES leaving the gallbladder in situ had a similar immediate morbi-mortality in this subset of patients and that surgery prevented long-term biliary complications derived from the presence of the gallbladder [16]. Logically, the next therapeutic approach to evaluate in these patients would be the applicability of minimally invasive therapy. It has been shown that laparoscopic surgery can be applied safely in elderly or high-risk patients [4, 9]. The best accepted policy for treatment of bile-duct stones in the laparoscopic era is the combined endoscopic-laparoscopic approach [2, 12], but to our knowledge it has never been compared with open surgery or endoscopic treatment. However, a randomized trial becomes difficult once a new technique is well accepted. Therefore, we present in this paper a series of 60 consecutive patients, suspected of having bile duct stones and defined as high risk, treated with a minimally invasive approach. The results of this series have been compared with those of a previous study by the same surgical team [16], with the analogous criteria of high-risk definition, in order to discover the potential advantages of the laparoscopic approach in this selected subset of patients.

Materials and methods

Between September 1991 and September 1994, we performed a prospective and randomized trial in which we compared the early and late results of open surgery vs endoscopic sphincterotomy leaving the gallbladder in situ for treatment of bile-duct stones in high-risk patients; the results have been published elsewhere [16] and are summarized in Tables 13. From October 94 to December 95, and following the same criteria as in the previously cited study (outlined below), we tested the efficacy of a laparoscopic approach, associated with preoperative endoscopic sphincterotomy (ES) if required, in a consecutive series of 60 symptomatic patients with cholecystolithiasis and a clinical diagnosis of bile-duct stones. All high-risk gallstone patients presenting with biliary pain, jaundice,

It has been classically considered that morbidity and mortality following bile duct surgery increase with age and associated disease [15, 18]. Before the rapid spread of laparoscopic surgery, endoscopic sphincterotomy (ES) leaving

Correspondence to: M. Trias

633 pancreatitis, cholangitis, or any combination thereof, and suspected of harboring bile-duct stones, were candidates for inclusion into the study. Patients who had prior surgical or endoscopic intervention of the biliary tree or who presented with severe cholangitis or biliary pancreatitis requiring urgent ES were not considered. Diagnostic criteria for bile-duct stones were biliary pancreatitis or the presence of cholestasis (increased plasma levels of bilirubin, alkaline phosphatases, and gamma glutamyl transpeptidase) associated with evidence of gallbladder stones and a dilated common bile duct (>8 mm) and/or duct stones as assessed by abdominal ultrasonography or endoscopic retrograde cholangiopancreatography (ERCP). Surgical high risk was defined by the presence of at least one of the following [16]. (1) Age over 70 years. (2) Goldman cardiac-risk index > 13. (3) Chronic pulmonary disease with a preevaluated postoperative maximum sustained ventilation (PPO-MSV) <10 l/min. (4) Liver cirrhosis, type B or C, of the Child Pugh classification. (5) Neurologic deficit or articular impairment associated with severely impaired mobilization. (6) Severe obesity, defined by a body mass index > 30. ERCP was performed in all cases, with the exception of patients with biliary pancreatitis in the absence of a dilated bile duct, because in a previous study [17] we showed that the frequency of bile-duct stones was low, and if present, could be dealt with during laparoscopic surgery. ES was performed only if a duct stone was observed. Laparoscopic cholecystectomy was scheduled in the first operating list available. To evaluate early morbidity and mortality, all patients were assessed daily after intervention and during hospitalization. The frequency of late biliary complications was evaluated during follow-up. Patients were scheduled for outpatient visit 30 days after discharge, every 3 months for the 1 year and every 6 months thereafter. Any patient in whom the assigned intervention could not be carried out, or who developed recurrent biliary symptoms within 30 days of intervention, was considered a primary failure. Secondary failure was defined as the relapse of biliary symptoms after the 1 month. The trial was analyzed on an intention to treat basis. Table 1. Comparison of clinical characteristics of patients randomized to surgery or endoscopic sphincterotomy with a consecutive series treated with a minimally invasive approacha Prospective series (199495) Laparoscopic cholecystectomy preoperative endoscopic sphincterotomy 60 79 (6) 18/42 14 (23%) 16 (16%) 23 (50%) 2.8 (2.9) 443 (276) 297 (275) 64 (6) 58 25 20 5 1 6 3

Randomized study [16] (199194) Endoscopic sphincterotomy leaving the gallbladder 50 79 (9) 13/55 18 (36%) 8 (16%) 24 (48%) 3.2 (3.1) 627 (518) 378 (347) 61 (7) 46 33 12 2 1 2 3

Group Number Age (years) Sex (M/F) Clinical manifestations Jaundice Cholangitis Acute pancreatitis Liver function tests Bilirubin mg/l Alkaline phosphatase IU/l Gamma glutamyl transpeptidase Total proteins g/l Surgical Risk Factors Age > 70 years Age > 70 years as sole factor Cardiac Goldman index > 13 Chronic pulmonary disease Liver cirrhosis Limited mobility Severe obesity
a

Open surgery 48 80 (7) 15/33 11 (23%) 11 (23%) 26 (54%) 4 (5) 720 (537) 377 (303) 61 (8) 46 27 7 1 3 3 2

Results Between October 1994 and December 1995, 61 patients fulfilling the inclusion criteria were considered for the study; 60 of them were included. The reason for exclusion of one patient was severe cholangitis requiring emergency surgery. Clinical features, liver function tests, and surgical risk factors are shown in Table 1. Forty-five patients underwent preoperative ERCP, whereas 15 patients with biliary pancreatitis in whom no ductal stones where visualized and who showed no bile duct dilatation were operated without prior ERCP.

Data are mean (SD).

ERCP and endoscopic sphincterotomy The technical efficacy of ERCP was 87%. The procedure failed in six patients for the following reasons: One patient had a cardiac arrest during premedication for ERCP and recovered after resuscitation maneuvers; in one patient the bile duct was not visualized due to the impossibility of cannulating the papilla deeply enough; in two patients the size of the common bile-duct stone precluded its retrieval; and in two patients residual bile-duct stones were observed during intraoperative cholangiography and were retrieved through a choledocholithotomy. All these six patients were operated by laparoscopy with intraoperative cholangiography. Stones were observed in four, and two patients were treated laparoscopically. Two cases were converted due to massive choledocholithiasis in one and an important distortion of biliary anatomy and cholecystoduodenal fistula in the other. Common bile-duct stones were found at ERCP in 32 out of 45 (75%) patients.

Morbidity related to ERCP occurred in five patients (11%). As described, one patient presented a cardiac arrest. Two patients developed acute pancreatitis, which resolved uneventfully, and one patient had massive pulmonary embolism that precluded further surgery. In one patient, a perforation of the duodenal wall occurred during ES; she developed severe cholangitis and underwent urgent surgical drainage, but died in the postoperative period. Laparoscopic cholecystectomy A laparoscopic cholecystectomy could be attempted in 58 patients out of 60 (97%). The interval between ES and LC was 10 6 days. In most cases the delay was due to difficulties with operating room scheduling. In six patients the delay was because ERCP had to be repeated (two), because of endoscopic suspicion of ampuloma (two) and because endoscopic diagnosis of gastric cancer (one) and atrial flutter (one) precluded immediate surgery. In three patients ES and LC were performed on the same day. Laparoscopic cholecystectomy was performed in 53 out of 58 patients. Conversions in five patients (8.6%) were due to adhesions in two, to multiple bile-duct stones in one, to a cholecystoduodenal fistula in one and to omental bleeding secondary to trocar injury in one. Intraoperative cholangiography was

634 Table 2. Technical efficacy and immediate morbidity and mortality; comparison of the results of a randomized trial between open surgery and ES with a consecutive series treated with a minimally invasive approach Prospective series (199495) Laparoscopic cholecystectomy preoperative endoscopic sphincterotomy 7 (11%) 11 (18%) 5 1 (1a) 1 1 (1a) 2 6 1 2 1 1 1 2 (3%) 11 (6) 4.4 (4)* 19 (8)** Table 3. Events during follow-up; comparison of the results of a randomized trial between open surgery and ES with a prospective series treated with a minimally invasive approach Prospective series (199495) Laparoscopic cholecystectomy preoperative endoscopic sphincterotomy 53 2 (3.6%)* 1 1 1 (2%)** 1 (2%)** 1 (2%)

Randomized study [16] (199194) Endoscopic sphincterotomy leaving the gallbladder 6 (12%) 8 (16%) 5 4 (2a) 1 (1a) 3 1 2 3 (6%) 11 (8) 5 (4) 16 (10)

Randomized study [16] (199194) Endoscopic sphincterotomy leaving the gallbladder 46 10 (21%) 2 3 2 3 11 (23%) 7 (15%) 1 (2%)

Group Primary failure Immediate morbidity Severe complications Intraabdominal sepsis Cholangitis Pneumonia Stroke Cardiac arrest Bile-duct injury Bleedingb Minor complications Wound infection Pancreatitis Atelectasis Urinary infection Catheter sepsis Hemoperitoneum Ileus Immediate mortality Pre treatment stay, days Post treatment stay, days Overall stay, days
a

Open surgery 3 (6%) 11 (23%) 4 3 (1a) 1 (1a) 7 4 2 1 2 (4%) 11 (5) 11 (8) 21 (9)

Group Patients followed Biliary complications during follow-up Biliary colic Acute cholecystitis Cholangitisa Jaundiceb Umbilical stone discharge Readmissions for biliary complications Reoperation Endoscopy
a b

Open surgery 43 3 (6%)* 1 2 2 (4%)** ** 2 (4%)

All four patients had common bile-duct calculi. Common bile-duct calculi in two patients. * p < 0.04 vs ES alone. ** < 0.01 vs ES alone.

Discussion The therapeutic approach to bile duct stones has changed dramatically during recent years, especially in the elderly and in patients with associated medical conditions, a subset of patients traditionally considered poor surgical candidates. Surgery is a safe treatment for biliary stones, with mortality rates of less than 0.5% for cholecystectomy and of 1.5% if the bile duct is explored [18]. Several series in the early 1980s suggested that mortality increases sharply (up to 30%) in the elderly or high-risk patient [15, 18], although there was a lack of uniformity in the definition of high risk. Surgical mortality rates have declined, however, in the last decade. A survey of 12 series of patients older than 70 years operated on for bile-duct stones published from 1988 to 1992 showed a mortality range between 0 and 9%, and in eight series it was below 4% [3, 6]. With the aim of reducing this morbidity and mortality, many authors have proposed that the endoscopic extraction of calculi, leaving the gallbladder, could be a useful alternative, with a lower immediate morbidity and mortality [19]. During recent months three prospective, randomized trials have shown the safety of open surgery compared with ES as the only treatment of bile-duct stones in patients with cholecystolithiasis [5, 8, 16]. Two of these studies [5, 16] showed that the incidence of biliary symptoms during follow-up was higher in patients treated with the endoscopic approach in which the gallbladder was preserved than in those operated on. The results of one of these trials have been used as a reference [16] to evaluate the applicability of mimimally invasive techniques in a series of elderly or high-risk patients operated by the same team, and using the same criteria to define high risk. In order to take advantage of a less aggressive surgical approach, it seems logical to apply laparoscopic techniques

Died. b Both patients needed to be reoperated on. * p < 0.001 vs open surgery. ** p < 0.001 vs ES alone.

performed in 49 patients and it showed bile duct stones in six. Two patients were converted, and in the remaining four, the stones were extracted by choledocholithotomy (three cases) or through the cystic duct (one case). Postoperative morbidity occurred in six patients (10%). Three patients had liver bed bleeding and two of them required reoperation. One patient presented a prolonged ileus. One of the patients converted to open surgery developed wound sepsis, and in another, who was converted due to a distorted anatomy and severely scarred gallbladder, a bile-duct transection occurred necessitating a hepaticojejunostomy. The patient died of intraabdominal sepsis 10 days later. Overall morbidity (endoscopic plus laparoscopic) was 19% (11 patients) and overall mortality 3% (two patients). Follow-up A follow-up was carried out in 55 out of 58 (95%) patients who were discharged. Mean follow-up was 9 4 months. Two patients out of 55 (3.6%) developed recurrent symptoms of biliary origin. In one patient with spilled stones during gallbladder retrieval, a stone discharge through the umbilicus was observed and required surgical exploration of the umbilical wound. A second patient developed obstructive jaundice due to a pancreatic cancer not identified during ERCP or surgery, and a percutaneous stent was required.

635

in elderly and high-risk patients. At this moment the preferred approach for bile duct stones is a combined one, with ES preceding laparoscopic cholecystectomy (LC) [2, 10, 11], although several series have shown the feasibility of a purely laparoscopic approach [12, 13]. When applied in the elderly or high-risk patient, there are specific aspects to both ERCP and LC that need to be considered. Regarding ERCP, it is well known that the frequency of ductal calculi and dilated ducts is higher in older than in younger patients; hence, endoscopic treatment is more challenging in the former. In this series ERCP was successfully accomplished in 87% of the patients in whom it was attempted, with a morbidity of 10% and a mortality of 2%. Recently, Deenitchin et al. [1] related the outcome of ES to age in 1,103 patients (921 younger than 80 years and 182 older than 80 years) and found a similar success in stone retrieval although older patients needed more accessory maneuvers than younger ones (lithotripsy or stenting, 19% vs 9%) and had higher morbidity (13% vs 7%) and mortality (1% vs 0%). Another consideration is that laparoscopic surgery could be more difficult in the elderly because these patients have longer and more complicated histories of biliary disease and are more likely to have peritoneal adhesions and distorted anatomy. Also, there is a high proportion of patients with previous upper abdominal surgery, additionally increasing technical difficultly. The conversion rate in this series was 8.6%, which compares favorably with the 10% rates[9] in other series of LC in the elderly. Finally, the increased frequency of associated medical diseases makes the perioperative control of the patient more difficult due to the cardiopulmonary changes induced by pneumoperitoneum. In this series, in which 30% of patients had associated cardiac disease, none had to be converted due to impossibility of maintaining the pneumoperitoneum. In our patients, most bile-duct stones were retrieved by endoscopic means. In a minority of patients, ductal stones were identified and treated during LC but two conversions were necessary because of massive choledocholithiasis in one patient and a cholecystoduodenal fistula in another. The dilemma that arises in trying to avoid a negative preoperative ERCP and pursuing laparoscopic exploration of the bile duct should be solved in future years. During the late 1980s, and before the widespread use of laparoscopic surgery, the general opinion among surgeons, supported by two prospective randomized trials [7, 14], was that the combination of ES plus open cholecystectomy was not superior to open surgery for patients with bile-duct calculi. Although comparative trials are lacking, most surgeons now prefer laparoscopic surgery preceded by ES. The use of randomized trials to scientifically prove the benefits of new alternative therapies is difficult when new treatment modalities become widely used, as occurred with LC. Keeping the objection in mind that this was not a randomized trial, the comparison of the results of this prospective series with a previous randomized trial from the same institution and the same surgical team with analogous inclusion criteria permits some conclusions to be drawn. The endoscopioplus-laparoscopic approach had a similar success rate to open cholecystectomy with bile-duct exploration, with a similar overall morbidity and mortality. Follow-up data also showed minimal morbidity of biliary origin in these patients. The postoperative hospital stay was shorter in the pa-

tients treated with ES followed by LC than in those previously submitted to open surgery (Table 2). However, the overall duration of hospitalization was similar in both cholecystectomy groups. An important shortcoming of the endoscopic-plus-laparoscopic approach in a busy general hospital is that cholecystectomy cannot be scheduled soon after ES. Thus, the mean delay was 10 days in our study. To evaluate the advantages that minimally invasive surgery can offer to the elderly or high-risk patient with an intact gallbladder presenting with choledocolithasis, future trials should compare preoperative endoscopic therapy followed by LC with simultaneous LC and bile-duct exploration [1012]. References
1. Deenitchin GP, Konomi H, Kimura H, Ogawa Y, Naritomi G, Chijiwa K, Trias M, Ikeda S (1995) Reappraisal of safety of endoscopic sphincterotomy for common bile duct stones in the elderly. Am J Surg 170: 5154 2. Erickson RA, Carlson B (1995) The role of ERCP in patients with laparoscopic cholecystectomy. Gastroenterology 109: 252263 3. Escarce JJ, Shea JA, Chen W, Quian Z, Schwartz JS (1995) Outcomes of open cholecystectomy in the elderly: a longitudinal analysis of 21,000 cases in the prelaparoscopic era. Surgery 117: 156164 4. Fried GM, Clas D, Meakins JL (1994) Minimally invasive surgery in the elderly patient. Surg Clin North Am 74: 375387 5. Hammarstro m LE, Holmin, Stridbeck H, Ihse I (1995) Long term follow up of a prospective randomized study of endoscopic versus surgical treatment of bile duct calculi in patients with gallbladder in situ. Br J Surg 82: 15161521 6. Moreaux J (1995) Traditional surgical management of common bile duct stones: a prospective study during a 20 year experience. Am J Surg 169: 220226 7. Neoptolemos JP, CarrLocke DL, Fossard DP (1987) Prospective randomised study of preoperative endoscopic sphincterotomy versus surgery alone for common bile duct stones. BMJ 294: 470474 8. Panis Y, Suc B, Escat J (1995) Surgery versus endoscopic sphincterotomy for choledocholithiasis: results of a prospective randomized study. Gastroenterology 109: A431 9. Passone N, Navez B, Cambier E, Baert G, Richir Ch, Guiot P (1995) Cholecystectomie coelioscopique chez le sujet a ge . Ann Chir 49: 291 295 10. Perissat J (1994) Operative treatment of common bile duct stones: laparoscopic cholecystectomy and endoscopic treatment. Eur J Gastroenterol Hepatol 6: 857863 11. Perissat J, Huibregtse K, Keane FGBV, Russell RCG, Neoptolemos JP (1994) Treatment of bile duct stones in the laparoscopic era. Br J Surg 81: 799810 12. Phillips EH (1994) Controversies in the management of common duct calculi. Surg Clin North Am 74: 931947 13. Phillips EH, Libermann M, Carroll BJ, Fallas MJ, Rosenthal RJ, Hiatt JR (1995) Bile duct stones in the laparoscopic era. Is preoperative sphincterotomy necessary? Arch Surg 130: 880886 14. Stiegman GV, Goff JS, Mansour A, Pearlman N, Reveille RM, Norton L (1992) Precholecystectomy endoscopic cholangiography and stone retrieval is not superior to cholecystectomy, cholangiography and common duct exploration. Am J Surg 163: 227230 15. Targarona EM, Pros I, Trias M (1992) Tratamiento de la coledocolitiasis en el paciente de riesgo elevado. Med Clin (Barc) 98: 6973 16. Targarona EM, Perez Ayuso RM, Bordas JM, Ros E, Pros I, Martinez J, Teres S, Trias M (1996) Randomised trial of endoscopic sphincterotomy with gallbladder left in situ versus open surgery for common bileduct calculi in high-risk. Lancet 347: 926929 17. Targarona EM, Balague C, Espert JJ, Perez Ayuso RM, Ros E, Navarro S, Bordas JM, Teres S, Trias M (1995) Laparoscopic treatment of acute biliary pancreatitis. Int Surg 80: 365368 18. Tierney S, Lillemoe KD, Pitt HA (1995) The current management of common duct stones. Adv Surg 28: 271299 19. Winslet MC, Neoptolemos JP (1991) The place of endoscopy in the management of gallstones. Baillieres Clin Gastroenterol 5: 99129

Original articles
Surg Endosc (1997) 11: 615617

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

How safe is ERCP to the endoscopist?

R. V. Cohen, M. A. Aldred, W. S. Paes, A. M. F. Fausto, J. R. Nucci, E. M. Yoshimura, E. Okuno, M. E. Garcia, L. M. Maruta, E. M. C. Tolosa
Division of Surgical Endoscopy, Health Physics, Occupational Medicine and Dosimetry Laboratory, University of Sa o Paulo, Sa o Paulo, Brazil Received: 15 July 1996/Accepted: 7 November 1996

Abstract Background: Interventional techniques in endoscopy such as endoscopic retrograde cholangiopancreatography (ERCP) have greatly increased since laparoscopic cholecystectomy has become widespread; mainly these techniques deal with common bile duct stones. Fluoroscopy is usually employed, and chronic exposure to X-ray, in spite of the relative low dose, can lead to potentially unhealthy conditions such as malignancies like bone marrow and other solid cancers. A median of 18 years of life is lost per fatal cancer, including the time of latency since exposure. Nor should one forget benign condition such as cataracts that can lead to partial or complete blindness and which surely impair lifes quality. Methods: Simulated examinations were carried at the University Hospital (Sa o Paulo, Brazil) using an anthropomorphic phantom in place of the physician. Four sets of dosimeters were placed in the forehead, neck, torso, and lower abdomen (with and without a lead apron) and standard ERCP fluoroscopic techniques were employed. Results: The dose equivalents were calculated and compared to the recommended exposure doses of national and international boards of radiation protection. Conclusions: Based on the results found and compared to standards, working safely means: (1) A lead (0.5 mm thickness) apron is fundamental. Without it less than one ERCP\month should be performed. (2) With an apron, 23 examinations/month are allowed. (3) No thyroid protection grants only 19 exams/month. (4) Performing ERCP without lead glasses is hazardous to the eye, allowing only seven ERCPs monthly. Key words: ERCP Radiological protection Radiation exposure protection Interventional laparoscopy Common bile duct stones management

The combination of endoscopic retrograde cholangiopancreatography (ERCP) with endoscopic sphincterotomy and stone extraction has acquired an expanded role in the management of choledocholithiasis in the minimal access surgery era [1, 2, 3]. During those procedures, fluoroscopic and radiographic images are taken with paramedical and medical staff near the patients. Even if the level of exposure to X-rays during ERCP is low, it is important to stress that it is a chronic exposure that can lead to potentially unhealthy conditions such as malignancies and benign conditions such as eye disorders (cataract) [7]. The goal of this paper is to evaluate the dose equivalent that the medical staff is exposed to during exams (ERCPs) performed at the Division of Surgical Endoscopy, University Hospital, University of Sa o Paulo, Brazil. The equivalent dose resulting from staff exposure was compared to recommended limits issued by international boards of radiation protection [8] and to Brazilian national standards [4]; then an ideal maximum number of examinations allowed per month and year was established in order to assure to the medical team the best protection against deleterious effects of ionizing radiation.

Methods
The endoscopic procedures were simulated by employing a recipient containing water as the patient and an anthropomorphic phantom (Alderson phantom) as the endoscopist who has control of the fluoroscopic equipment (wearing a 0.5-mm-thickness lead apron). The position of the phantom was chosen to achieve the maximum exposure during ERCP, using data calculated by a radiation monitor (Radcal Corporation, model 9015; Washington, DC) on radiation exposure. The phantom was set up in an upright position at a distance of 55 cm from the examination table. Each dose equivalent was calculated from the measurements of four thermoluminescent dosimeters (TLD 100) located in different sites of the phantom: the forehead, neck, torso and lower abdomen (under and over the apron). The dosimeters were calibrated at the Dosimetry Laboratory, Physics Institute, University of Sa o Paulo [9, 12]. Measurements were performed for fluoroscopic and radiographic exposures, using four sets of TLDs. During fluoroscopy three sets were submitted to three different irradiation times (20, 40, and 60 min). Fluoroscopy was simulated with consecutive displays of 85 kVp and 2.2 mA, breaking each 5 min. The last set of TLDs was submitted to an irradiation

Correspondence to: R. V. Cohen, Rua Wanderley, 1482 ap. 53, 05011-001 Sa o Paulo SP, Brazil

616 Table 1. Equivalent doses for medical and paramedical staff; fluoroscopy and radiographs Equivalent doses HT (mSv) for 1 h fluoroscopy 4.35 5.50 0.15 4.79 Nonmeasurable 0.20 Equivalent doses HT (mSv) 6 films 0.33 0.31 0.03 0.32 Nonmeasurable 0.07 Table 4. ERCPs maximum number (monthly/yearly) where CNEN and ICRP-60 limits are respected Maximum allowed exams (CNEN, Brazil) Organs Crystalline Thyroid Whole body (with apron) Whole body (without apron) Year 89 243 714 27 Month 7 19 57 2 Maximum allowed exams (ICRP-60) Year 89 285 10 Month 7 23 0.8 Table 3. Occupational limits of doses; yearly basis Occupational limits CNEN (yearly) mSv 50 150 500 Occupational limits ICRP60 (yearly) mSv 20 (5-year basis) 150 500

Organ/site Whole body Crystalline Skin

Dosimeters site Forehead Neck Torso (inside apron) Torso (outside apron) Lower abdomen (inside apron) Lower abdomen (outside apron)a
a

Located below the table. Works as protection from ionizing radiation.

Table 2. Equivalent doses per procedure; calculated from Table 1 Equivalent doses (fluoroscopy, mSv) 1.45 1.84 0.050 Equivalent doses (radiographs, mSv) 0.22 0.21 0.02 Equivalent doses (total) 1.67 2.05 0.07

Site/organ Crystalline Thyroid Whole bodyapron Whole bodyno apron Gonadsapron Gonadsno aprona
a

1.60 0.22 1.82 Nonmeasurable Nonmeasurable Nonmeasurable 0.067 0.05 0.12

Discussion The combination of endoscopic retrograde cholangiopancreatography (ERCP) with endoscopic sphincterotomy (ES) and stone extraction has acquired an expanded role in the management of choledocolithiasis. With the rapid growth of laparoscopic cholecystectomy (LC), the evaluation and treatment of suspected common bile duct stones have been limited, and the main approach combined to LC to common duct stones has been until now in many centers the combination of ERCP and ES [1, 6, 11]. To perform those procedures, exposure of the medical and paramedical staff to ionizing radiation is required. Even though the exposition is low, it is continuous and may lead to hazardous consequences. Information on the risk of cancer following radiation exposure comes from a large number of epidemiological studies and has recently been reviewed by the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) and by the US Committee on the Biological Effects of Ionizing Radiation (BEIR V)[5, 14]. The populations that have been studied include: 1. Over 90,000 survivors of the atomic bombings of Hiroshima and Nagasaki 2. 14,000 mostly male patients in the UK treated for ankylosing spondylitis with X-rays 3. 83,000 women in eight countries treated for cervical cancer with X or gamma radiation Very few studies have followed the entire population until the end of life. The temporal pattern of risk was assessed in studies as the Life Span Study [10] of A bomb survivors. Those studies showed that for the main cancer secondary to radiation exposureleukemiathe peak is about 7 years of exposure followed by a tailing off in risk. The relative risks depend on the age at the time of exposure, dose level and rate, and if the individual who was dealing with ionizing

Located below the table. Works as protection from ionizing radiation.

equivalent to six plain radiographic films. The plain films were obtained at 85 kVp and 100 mA. The quantity dose-equivalent (H) is limited to radiation protection applications and is calculated as the product of the absorbed dose (D) by a quality factor (Q) that takes into account the type of radiation (, , , or n). Dose equivalent is measured in units of Sievert (Sv), e.g., 1 mSvone milliSievertis equivalent to an energy of 103 Joules per 1 kilogram of tissue mass.

Results The average dose equivalent, measured during fluoroscopy and performing the six static films, is shown on Table 1. The estimated time of exposure to ionizing radiation during the performance of ERCPs at the Division of Surgical Endoscopy, University of Sa o Paulo, is 20 min and generally four films are taken. Bearing in mind that the points of measurement can be analogous to some anatomic parts, respectively the crystalline, thyroid, gonads, and whole body (torso + abdomen), the total equivalent-dose was calculated (Table 2). Limits established by Brazilian standards (National Commission of Nuclear Energy, CNEN, Sa o Paulo, Brazil) and international boards (International Commission of Radiation ProtectionICRP, London, England) are shown on Table 3. Considering the limits of dose equivalents as a standard of working safely, according to CNEN-NE 3.01 and ICRP 60, and the total doses obtained per procedure at the Division of Surgical Endoscopy (Table 2), the maximum number of procedures that the staff can perform safely was determined (Table 4).

617

radiation got a specific oncogen in his genoma. No one should forget that there are different temporal patterns of risk for different cancer types. For leukemia, after adjustment for age at exposure, the relative risk appeared to be constant over time, mainly if the subject is constantly exposed as the staff that performs endoscopic procedures under fluoroscopy. For solid cancers, such as lung, breast, thyroid, and GI (colon and stomach) cancers, the relative risk decreases about 10 to 20 years following exposure [13]. Following leukemia, solid GI cancers, breast, lung, and thyroid malignancies are the commonest neoplasias following radiation exposure. It is important to stress that those data are only relative to deaths secondary to ionizing-radiation-induced cancers. Quality of life tends to decrease to important levels, if its considered the time taken for the diagnosis and treatment of secondary cancers. Besides neoplastic affections, in spite of the lack of significant statistical data, crystalline injury was described [14], and this benign condition can lead to a feared situation that may follow chronic exposition to radiationcomplete blindness. Keeping in mind all the potential hazards related to chronic exposure to ionizing radiation, and comparing the dose equivalent taken by the phantom to international standards (Table 3), it is concluded that working safely means: 1. Wearing a lead apron (0.5-mm thickness) is fundamental: without it, less than one ERCP should be performed per month; only four examinations should be performed in a period of 5 months. 2. With an apron, 23 exams/month are allowed. 3. No thyroid protection grants only 19 exams per month. 4. Performing ERCP without lead glasses is hazardous to the eye, allowing only seven ERCPs per month. Besides lead aprons, glasses and thyroid shields are important protective devices and should allow an increased number of safe monthly examinations. It must always be emphasized that radiation carcinogenic induction is a process with no threshold dose, e.g., even for small doses the probability of ocurrence is not zero. Specialists set limits as a way to keep the incidence of malignancies at acceptably low levels. However, focusing

on benign effects such as cataracts, the limiting doses might be adopted as threshold doses. References
1. Arregui M, Davis CJ, Akrush AM, Nagan R (1992) Laparoscopic cholecystectomy combined with endoscopic sphincterotomy and stone extraction or laparoscopic choledocoscopy and electrohydraulic lithotripsy for management of cholelithiasis with choledocolithiasis. Surg Endosc 6: 1015 2. Cohen RV, Schiavon CA, Moreia L, Tolosa EMC (1994) What is minimally invasive surgery? In: Cohen RV (ed) Metabolic and systemic responses following interventional laparoscopy. RG Landes Biomedical, Austin, TX 3. Cohen RV, Shiavon CA, Schaffa TD (1995) Laparoscopic cholecystectomy can be safely performed without intraoperative cholangiography. Surg Laparosc Endosc 5(2): 165166 4. Comissa o Nacional de Energia Nuclear (1988-NE 3.01.) Diretrizes Ba sicas de Radioprotec a o, Bras lia, Brasil 5. Committee on the Biological Effects of Ionising Radiation (BEIR V) (1990) Health effects of exposure to low levels of ionising radiation. National Academy of Sciences, National Academy Press, Washington, DC 6. Fink AS (1993) To ERCP or not to ERCP. That is the question. Surg Endosc 7(5): 375376 7. Gray J, Ragozzino M, Gazzi R (1981) Normalized organ doses for various diagnostic radiologic procedures. Am J Radiol 137: 463470 8. International Commission on Radiation Protection ICRP Publication 60 (1990) Recommendations of the International Commission on Radiological Protection. Pergamon Press, Oxford 9. Nucci JR, Aldred WS, Paes AM, Okuna E (1995) Sistema de Monitorac a o Individual de Rotina do Laborato rio de Dosimetria do Instituto de F sica da Universidade de Sa o Paulo. Presented at the Segundo Congresso de Seguridad Radiologica e Nuclear, Cusco, Peru, October 2327 10. Shimizu Y, Kato H, Nagai Y (1988) Life span study report 11. Part II: cancer mortality in the years 19501985 based on the recently revised doses (DS86). Hiroshima, RERF TR5-88 11. Surick B, Washington M, Ghazi A (1993) ERCP in conjunction with laparoscopic cholecystectomy. Surg Endosc 7: 388392 12. Tresniak P, Yoshimura EM, Cruz MT, Okuno E (1990) Brazilian fluorite-base dosimetric pellets: history and post-use review. Radiat Prot Dosim 24: 167170 13. Tucker MA, Meadows AT, Boice JD (1984) Cancer risk following treatment for childhood cancer. In: Boice JD, Fraumeni (eds) Radiation carcinogenesis: epidemiology and biological significance. Raven Press, New York, pp 211224 14. UNNSCEAR (1988) Sources, Effects and Risks of Ionising Radiation. Report to the General Assembly, United Nations, New York

Surg Endosc (1997) 11: 671672

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Long-term results after laparoscopic cholecystotomy in a child with symptomatic gallstone disease
B. M. Ure, E. P. Eypasch, H. Troidl
II. Department of Surgery of the University of Cologne, Ostmerheimerstrae 200, 51109 Cologne, Germany Received: 10 May 1996/Accepted: 29 May 1996

Abstract. Cholecystotomy has been suggested for symptomatic gallstone disease in selected children. This suggestion is supported by a potential reduction in the frequency of the so-called postcholecystotomy syndrome. To our knowledge, laparoscopic cholecystotomy has not been reported yet. However, gallstone recurrence has been reported up to 4 years after conventional cholecystotomy and therefore we waited to publish our results for that period of time. A 12-year-old girl with idiopathic symptomatic gallstone disease and a normal kinetic of the gallbladder underwent laparoscopic cholecystotomy. The laparoscopic technique was similar to laparoscopic cholecystectomy but the gallbladder was left in place and multiple gallstones were removed. Intraoperative cholecystoscopy revealed three additional small stones. They were removed by subsequent lavage of the gallbladder. Choledocholithiasis was excluded by intraoperative cholangiography and the gallbladder was closed using an Endo GIA. There were no intraoperative or postoperative events. The patient is free of complaints without recurrent gallstones on ultrasound examination today, 4 years after the operation. Laparoscopic cholecystotomy represents a feasible alternative to laparoscopic cholecystectomy. Key words: Cholelithiasis Cholecystotomy Laparoscopy Children

aroscopic cholecystectomy [4, 12, 15] numerous advantages of this technique have been proven in adult series [15, 16] and it has been reported to be safe and feasible in children [6, 11, 18]. To our knowledge, laparoscopic cholecystotomy in children and the long-term results of the technique have not been published yet.

Patient and methods

The patient was a 12-year-old girl who had suffered occasional attacks of severe upper abdominal pain since 6 months of age. Pain and nausea were aggravated by food and in particular by fatty meals. There was no history of jaundice and no family history of hemolytic disease. Ultrasound examination revealed multiple stones in the gallbladder. The diameter of the common bile duct was 2.6 mm without evidence of choledocholithiasis. The volume of the gallbladder reduced from 32 ml to 13 ml after a standardized meal. On plain x-ray and laboratory examinations there were no pathological findings. The operation was performed under general anesthesia with the patient in prone position. The surgeon stood between the legs of the patient; one assistant stood on the right and a second assistant on the left side [15]. A 1-cm incision was made in the infraumbilical fold and the pneumoperitoneum was introduced with a Veress needle. The pressure was kept at 12 mmHg maximum. A 10-mm 30 laparoscope was inserted and four additional trocar sites were set in place (Fig. 1). The gallbladder was opened using unipolar electrocoagulation and the gallstones were removed by a grasping forceps. Subsequent cholecystoscopy revealed three additional stones which were removed by lavage and a suction device. Additional stones of the common bile duct were excluded by cholangiography which was performed via a catheter through the cystic duct. The gallbladder was closed using a 11-mm Endo GIA (Auto Suture). The operation time was 140 min; there was no blood loss.

Symptomatic gallstone disease is uncommon in childhood, but in recent years the frequency has been reported to be increasing [1, 5, 7]. Cholecystectomy is recommended as standard treatment for symptomatic gallstones in children. However, some reports on cholecystotomy in selected cases have been published [7, 13]. Since the introduction of lap-

Results The patient had oral fluids the evening after the operation and regular meals on day 1. No postoperative complications were recorded. On discharge at day 3 the patient was free of complaints. Today, 4 years after the operation there are no diet restrictions, no attacks of abdominal pain, and no other abdominal symptoms. On ultrasound examination there is no evidence for recurrence of gallstones.

Correspondence to: B. M. Ure

672

Fig. 1. Placement of five trocar incision sites in laparoscopic cholecystotomy.

candidates for cholecystotomy. The function of the gallbladder in our patient was normal before and after the operation on ultrasound examination. Numerous advantages of laparoscopic compared to conventional cholecystectomy have been reported for adult patients [15, 16]. In small series of children the technique was performed safely and effectively with a short hospital stay and recovery period [6, 11, 18]. We suggest that children with idiopathic symptomatic gallstone disease and a normal function of the gallbladder may be candidates for laparoscopic cholecystotomy. They may profit from the advantages of the laparoscopic as compared to the conventional technique. References
1. Bailey PV, Connors RH, Tracy TF, Cirilo S-A (1989) Changing spectrum of cholelithiasis and cholecystitis in infants and children. Am J Surg 158: 585588 2. Bates T, Ebbs SR, Harrison M, AHern RP (1991) Influence of cholecystectomy on symptoms. Br J Surg 78: 964976 3. Carswell WR, Willis JD (1969) Cholecystitis with gall-stones in infancy and childhood. Br J Surg 56: 547548 4. Dubois F, Ikard PF, Berthelot G, Levard H (1990) Coelioscopic cholecystectomy. Preliminary report of 36 cases. Ann Surg 211: 6062 5. Holcomb GW, Holcomb III GW (1990) Cholelithiasis in infants, children, and adolescents. Pediatr Rev 11: 268274 6. Holcomb III GW, Olsen DO, Sharp KW (1991) Laparoscopic cholecystectomy in the pediatric patient. J Pediatr Surg 10: 11861190 7. Molander M-L, Berhdahl S (1992) Gallbladder disease, primary cholelithiasis, or gallbladder hydrops: review of 32 children. Pediatr Surg Int 7: 328331 8. Moorehead RJ, Kernohan RM, Patterson CC, McKelvey STD, Parks TG (1986) Does cholecystectomy predispose for colorectal cancer? Dis Colon Rectum 29: 3638 9. Moorehead RJ, McKelvey STD (1989) Cholecystectomy and colorectal cancer. Br J Surg 76: 250253 10. Moosa AR (1973) Cholelithiasis in childhood. J R Coll Surg Edinb 18: 4246 11. Newman KD, Marmon LM, Attori R, Evans S (1991) Laparoscopic cholecystectomy in pediatric patients. J Pediatr Surg 10: 11841185 12. Perissat J, Collet D, Belliard R (1990) Gallstones: laparoscopic treatmentcholecystectomy, cholecystotomy and lithotripsy. Surg Endosc 4: 15 13. Robertson JFR, Carachi R, Sweet EM, Raine PAM (1988) Cholelithiasis in childhood: a follow-up study. J Pediatr Surg 23: 246249 14. So derlund S, Zetterstro m B (1962) Cholecystitis and cholelithiasis in children. Arch Dis Child 37: 174180 15. Troidl H, Spangenberger W, Langen R, Al-Jaziri A, Eypasch E, Neugebauer E, Dietrich A (1992) Laparoscopic cholecystectomy: technical performance, safety, and patient benefits. Endoscopy 24: 252261 16. Ure BM, Troidl H, Spangenberger W, Lefering R, Dietrich A, Neugebauer E (1994) Pain after laparoscopic cholecystectomy. Surg Endosc 8: 9096 17. Ure BM, Troidl H, Spangenberger W, Lefering R, Dietrich A, Eypasch E, Neugebauer E (1995) Symptoms more than one year after laparoscopic cholecystectomy. Br J Surg 82: 267270 18. Vinograd I, Halevy A, Klin B, Negri M, Bujanover Y (1993) Laparoscopic cholecystectomy: treatment of choice for cholelithiasis in children. World J Surg 17: 263266

Discussion Childhood cholelithiasis has been associated with several disorders, in particular with hemolytic disease [1, 5]. However, in children with idiopathic gallstone disease gallstone production is believed to be caused by a temporary vulnerability [7]. Thus, removing the gallstones by cholecystotomy and leaving the gallbladder in place has been suggested [7, 13]. This suggestion is supported by a potential reduction in the frequency of the so-called postcholecystectomy syndrome, which has been reported in up to 42% of patients after cholecystectomy [2, 17]. In addition, there is evidence for an association between cholecystectomy and colorectal cancer [9]. An increased relative risk of up to 3.5 times for right-sided colonic cancer in women has been reported [8]. Robertson et al. [13] reported a series of nine patients on whom they successfully performed cholecystotomy, five of whom were children with hereditary spherocytosis who underwent splenectomy in the same session. All nine patients were asymptomatic 0.37.5 years after surgery. However, in one child with spherocytosis an asymptomatic gallstone was detected on ultrasound examination after 2 years. Molander and Bergdahl [7] performed cholecystotomy in one child with congenital spherocytosis and in six children with idiopathic symptomatic gallstones. One of the six children underwent cholecystectomy later on because of dyskinesia of the gallbladder, but there is no follow-up information on the remaining six children. Out of three other case reports [3, 10, 14] one child was free of complaints and free of gallstones after 11 years [14]. Another patient was cholecystectomized after 6 years for persisting symptoms due to a poorly functioning gallbladder [10]. Another patient underwent cholecystectomy 4 years later for recurrent gallstones [3]. Therefore, it was decided to not publish the present case report until 4 years after cholecystotomy. None of these authors had investigated the kinetic of the gallbladder by ultrasound or x-ray examination prior to cholecystotomy. In our opinion, this is essential for selecting

Surg Endosc (1997) 11: 673675

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Endoscopic hydrostatic balloon dilation of ulcer-induced pyloric stenosis in rheumatoid arthritis and secondary amyloidosis
K. Hizawa,1 Y. Ohta,1 H. Satou,2 K. Aoyagi,1 K. Eguchi,3 M. Fujishima1
1 2

Second Department of Internal Medicine, Faculty of Medicine, Kyushu University, Maidashi 3-1-1, Higashi-Ku, Fukuoka 812-82, Japan First Department of Surgery, Faculty of Medicine, Kyushu University, Maidashi 3-1-1, Higashi-Ku, Fukuoka 812-82, Japan 3 Second Department of Pathology, Faculty of Medicine, Kyushu University, Maidashi 3-1-1, Higashi-Ku, Fukuoka 812-82, Japan Received: 11 July 1996/Accepted: 23 July 1996

Abstract. We describe a 50-year-old Japanese woman with rheumatoid arthritis who presented with near-complete gastric outlet obstruction. The patient also suffered from secondary gastrointestinal and cardiac amyloidosis. Gastroscopy revealed multiple huge gastric antral ulcers in which amyloid deposits were identified on histologic examination. The ulcers became scars after treatment with omeprazole, which cause in severe pyloric stenosis. Endoscopic hydrostatic balloon dilation under fluoroscopic guidance was performed twice for 10 min. The pyloric outlet remained sufficiently patent 22 months later. Key words: Rheumatoid arthritis gastric outlet obstruction-pyloric stenosis

Patients with rheumatoid arthritis (RA), who require longterm nonsteroidal antiinflammatory drugs (NSAIDs) and corticosteroids, have a tendency to develop relapsing gastric antral ulcers which infrequently result in pyloric channel stenosis [3, 8, 11, 12, 14]. In addition, secondary gastrointestinal amyloidosis (SGIA) is a well-known serious complication in RA patients [13]. In these patients, surgical intervention should be carefully approached because of an associated hemorrhagic diathesis, impaired wound healing, and a high rate of cardiac and renal involvement [10, 15]. We describe a patient with RA and SGIA in whom a severe gastric outlet obstruction was successfully managed by endoscopic hydrostatic balloon dilation. Case report
A 50-year-old Japanese woman was admitted to our hospital in January 1994 with a complaint of diarrhea of 2 months duration. The patient had

been suffering from RA for 20 years. Her recent medications consisted of oral prednisolone 5 mg and loxoprofen 180 mg per day. According to the American Rheumatism Association, her arthritis was class II and stage IV [1]. There was occult blood in her stool, but no causative agent was detected by culture. Laboratory data revealed a thrombocytosis (55 104/l), but no leucocytosis or anemia. C-reactive protein was 0.1 mg/d1 and serum rheumatoid factor was 1169 IU/ml. Serum chemistry indicated a hypoproteinemia (total protein, 5 g/dl), but no abnormality of liver or renal function. An upper gastrointestinal endoscopy on admission showed extensive granularity in the gastric and duodenal mucosa. Biopsy of these areas revealed amyloid deposits (Fig. 1), and the material was confirmed as amyloid A protein by histochemical investigation [13]. We therefore considered her symptoms to be due to secondary gastrointestinal amyloidosis resulting from long-standing RA. The patient also manifested congestive heart failure, probably due to cardiac amyloidosis as suggested by echocardiography. Total parenteral nutrition therapy (TPN) was administered and the patients diarrhea subsequently improved. However, the patients RA worsened, requiring intermittent administration of diclofenac suppositories and intraarticular injection of triamcinolone acetate. The patient, furthermore, began to experience epigastric pain even with the prophylactic administration of famotidine 40 mg daily. The endoscopy performed 60 days after

Correspondence to: K. Hizawa

Fig. 1. Biopsy specimen shows marked amyloid deposits in the lamina propria mucosa (H.E, 66).

674

Fig. 2. Gastroscopy in May 1994, reveals severe pyloric scarring (arrow, A). The gastric outlet channel remains patent 22 months after the dilation (B).

Fig. 3. Rigiflex balloon catheter threaded over the guide wire is positioned precisely under fluoroscopic guidance (A). The balloon is inflated with a radiopaque solution and remains dilated for 10 min with 3 atm (B).

admission showed multiple and large ulcers in the gastric antrum. Although, 50 days later, the gastric ulcers had healed after the additional treatment with omeprazole 20 mg, pyloric scarring with nearly complete obstruction of the gastric outlet channel had developed (Fig. 2A). Although TPN was continued, her obstructive symptoms of nausea and vomiting remained. Endoscopic balloon dilation was therefore attempted alternative to surgical intervention. Because the pyloric outlet could not be clearly identified on endoscopy, we inserted a radifocus wire (Terumo) through the narrow canal of outlet obstruction into the duodenum under fluoroscopic guidance with torque control. We passed through a small-caliber catheter over the wire and then exchanged the wire for a standard guidewire (Microvasive). Using a 15-mm Rigiflex balloon dilator (Microvasive), hydrostatic dilation was performed, maintaining balloon inflation for 10 min with 3 atm (Fig. 3). One week later, this procedure was repeated, and the diameter of the gastric outlet widened to more than 10 mm. The patient was subsequently able to eat a regular diet. Her clinical course after dilation was uneventful, and she was discharged in August 1994. Since then, the patient has been maintained on acid-suppressive therapy with per oral famotidine and has been free of any symptoms. Follow-up endoscopy performed in March 1996 revealed sufficient patency of the gastric outlet and no recurrence of ulcer (Fig. 2B).

Discussion NSAIDs are well known to be causative agents of gastric mucosal injury. A prospective study by Caruso and Bianchi

Porro [3] has indicated that about 10% of subjects developed a gastric ulcer 3 to 12 months after the administration of NSAIDs. Larkai et al. [8] have reported that a gastric ulcer was found on endoscopy in more than 10% of arthritic patients receiving chronic NSAIDs. They also suggested that the ulcers were characterized by a lack of pain and were located in the gastric prepyloric region [8]. These ulcers might thus lead to pyloric scarring and gastric outlet obstruction as they are recurrent and can be overlooked in asymptomatic patients. In a recent study by Weaver et al. [14], prolonged NSAIDs use was associated with gastric outlet obstruction. Ulcers in RA patients receiving chronic NSAIDs are more frequent when concomitant corticosteroids are used [12]. Ohtsuka [11] has speculated that decreased gastric mucosal blood flow is associated with ulcer formation in RA patients. In addition, RA is occasionally complicated by SGIA [13], which manifests as ischemic lesions in the gut due to perivascular amyloid deposits [10]. Our patient had all these risk factors for developing severe pyloric stenosis. In patients with SGIA, particularly involving the heart and kidney, surgery should be carefully approached because of unfavorable postoperative results [10, 15]. Therefore, we

675

chose endoscopic balloon dilation alternative surgery in this patient, which was managed successfully and safely. Even though the pyloric stenosis was so severe that the outlet canal could not be clearly identified, we were able to pass through the narrow canal using a radifocus guidewire (Terumo) under fluoroscopic guidance with a torque control. Since a successful report by Benjamin et al. in 1984 [2], endoscopic balloon dilation has been applied for benign gastric outlet obstruction, and the efficacy over the short term has been established [4, 6]. However, the safety and long-term results are now being debated. Kuwada and Alexander [7] have reported that the long-term success rate was only 16% during a median observation of 45 months. In a recent study by Lau et al. [9], the ulcer complication-free rate at 3 years was 55%, and one-half of the patients required subsequent surgery because of recurrent obstruction, perforation, or bleeding. Kozarek [5] has stated that the results seem to depend on patient selection, standardization of the procedures (air or liquid insufflation, or under fluoroscopy) and patient compliance with acid-suppressing drugs. Although further follow-up is necessary, our patient maintained her clinical improvement for 22 months after the dilation. References
1. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS (1988) The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31: 315324 2. Benjamin SB, Glass RL, Cattau EL, Miller WB (1984) Preliminary experience with balloon dilation of the pylorus. Gastrointest Endosc 30: 9395

3. Caruso I, Bianchi Porro G (1980) Gastroscopic evaluation of antiinflammatory agents. Br Med J 280: 7578 4. DiSario JA, Fennerty MB, Tietze CC, Hutson WR, Burt RW (1994) Endoscopic balloon dilation for ulcer-induced gastric outlet obstruction. Am J Gastroenterol 89: 868871 5. Kozarek RA (1996) Endotherapy for gastric outlet obstruction. Gastrointest Endosc 43: 173174 6. Kozarek RA, Botoman VA, Patterson DJ (1990) Long-term follow-up in patients who have undergone balloon dilation for gastric outlet obstruction. Gastrointest Endosc 36: 558561 7. Kuwada SK, Alexander GL (1995) Long-term outcome of endoscopic dilation of nonmalignant pyloric stenosis. Gastrointest Endosc 41: 15 17 8. Larkai EN, Smith JL, Lidsky MD, Graham DY (1987) Gastroduodenal mucosa and dyspeptic symptoms in arthritic patients during chronic nonsteroidal anti-inflammatory drug use. Am J Gastroenterol 82: 11531158 9. Lau JYW, Chung SCS, Sung JJY, Chan ACW, Ng EKW, Suen RCY, Li AKC (1996) Through-the-scope balloon dilation for pyloric stenosis: long-term results. Gastrointest Endosc 43: 98101 10. ODoherty DP, Neoptolemos JP, Wood KF (1987) Place of surgery in the management of amyloid disease. Br J Surg 74: 8388 11. Ohtsuka E (1992) Upper gastrointestinal endoscopic findings and gastric mucosal blood flow in patients with rheumatoid arthritis. Fukuoka Acta Med 83: 6271 12. Piper JM, Ray WA, Daugherty JR, Griffin MR (1991) Corticosteroid use and peptic ulcer disease: role of nonsteroidal anti-inflammatory drugs. Ann Intern Med 114: 735740 13. Tada S, Iida M, Yao T, Kawakubo K, Yao T, Okada M, Fujishima M (1994) Endoscopic features in amyloidosis of the small intestine: clinical and morphologic differences between chemical types of amyloid protein. Gastrointest Endosc 40: 4550 14. Weaver GA, Harper RL, Storey JA, Jenkins PL, Marrell NB (1995) Nonsteroidal antiinflammatory drugs are associated with gastric outlet obstruction. J Clin Gastroenterol 20: 196198 15. Yood RA, Skinner M, Rubinow A, Talarico L, Cohen AJ (1983) Bleeding manifestations in 100 patients with amyloidosis. JAMA 249: 13221324

Letters to the editor

Surg Endosc (1997) 11: 696

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The totally extraperitoneal laparoscopic hernia repair

We read with interest the paper by Vanclooster and colleagues [11] and commend their contribution to this procedure. However, we would offer three comments: First, totally extraperitoneal laparoscopic hernia repair was developed by Dulucq [13] in 1989/90 and by McKernan [7] not a great deal later. While it seems legitimate for others to publish their own technical variations, developments, and outcomes, we strongly support recent reminders [6] that journal editors and their peer referees owe the reader a duty of diligence: they should insist that authors exercise proper scholarship by giving credit where it is due. Otherwise the uninformed reader may assume originality and the informed may infer plagiarism, where the author intended neither. Second, the mesh configuration suggested by Vanclooster et al. was presented by one of us several years ago [4, 10]. However, the concept of amputating the inferior and lateral corner so the mesh fits better on the iliac vessels and the psoas muscle is flawed, since it has subsequently been reported [12] that recurrences may occur dorsal/ inferior to this inferolateral corner. The most extensive possible coverage of the psoas muscle belly is therefore appropriate. Third, the need for mesh fixation remains debatable. However, to fix the cranial border to prevent early migration or slipping is illogical: In our joint experience of over 1,000 cases and, to our knowledge in all reports in the world literature, recurrences pass uniformly caudal to the inferior border of the prosthesis. Fixation of the inferior medial part of the mesh to Astley Coopers ligament alone [5] may not offend against the original tension-free notion of Stoppa [8, 9] nor interfere with the mechanics of prosthesis retention. To fix the superior border to points that move relative to one another within a musculofascial structure contravenes both principles. Finally, on a minor point, if the structure annotated as D in Fig. 1 is the testicular vascular bundle, where is the vas deferens? Despite these comments we congratulate the authors on their low complication rate. References
1. Dulucq J-L (1991) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritone ale en retrope ritone oscopie. Cah Chir 79: 1516 2. Dulucq J-L (1992) Traitement des hernies de laine par mise en place dun patch prothe tique sous-pe ritoneal en pre -pe ritoneoscopie. Chirurgie 118(12): 8385 3. Dulucq J-L (1992) The treatment of inguinal hernias by implantation of mesh through retroperitoneoscopy. Postgrad Gen Surg 4: 173174 4. Fiennes AGTW, Taylor RS (1994) Learning laparoscopic hernia repair. In: Inguinal hernia, advances or controversies? Arregui M, Nagan R (eds) Radcliffe, Oxford, pp 475482 5. Fiennes AGTW, Himpens J, Dulucq J-L (1994) Preperitoneoscopic groin hernioplasty, current synthesis. Surg Endosc 8(8): 989 6. Horton R, Smith R (1996) Time to redefine authorship (editorial). Br Med J 312: 723 7. McKernan JB (1993) Laparoscopic extraperitoneal repair of inguinofemoral herniation. Endosc Surg Allied Tech 1(4): 198203 8. Stoppa R, Petit J, Abourachid H (1973) Proce de original de plastie des hernies de laine. Linterposition sans fixation dune prothe ` se en tulle de Dacron par voie me diane pre pe ritone ale. Chirurgie 99: 119 9. Stoppa RE, Rives JL, Warlaumont CR (1984) The use of Dacron in the repair of hernias of the groin. Surg Clin North Am 64: 269285 10. Taylor RS, Fiennes AGTW (1992) A tension free modification of the Dulucq preperitoneal laparoscopic hernioplasty. Min Inv Ther 1(Suppl 1): 101 11. Vanclooster P, Meersmann AL, de Gheldere CA, Van de Ven CK (1996) The totally extraperitoneal laparoscopic hernia repair. Surg Endosc 10: 332335 12. Vivekanadan S, Fiennes AGTW (1995) Totally extraperitoneal groin hernioplasty: mechanism of delayed indirect recurrence. Min Inv Ther 4(Suppl 1): 55

A. Fiennes
Department of Surgery St Georges Hospital Medical School Cranmer Terrace London, SW17 ORE, United Kingdom

J. Himpens
Department of Digestive Surgery University Hospital Ste Pierre Rue Haute 201 B-1000 Brussels, Belgium

Surg Endosc (1997) 11: 697

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank you for the opportunity to answer the letter of Mr. Fiennes. First of all, we certainly do acknowledge that Mr. Dulucq and Mr. Mc Kernan were the pioneers of the extraperitoneal laparoscopic hernia repair. Honor to whom honor is due. We did not intend to pretend to be the pioneers of this technique. We wished only to describe the technique the way we perform it, to describe our own findings, and to give a fair report of our preliminary results. The reason for cutting the inferolateral corner is not just that we think it fits better on the iliopsoas but also because we are afraid to cause damage to the nerves running on it by dissecting unnecessarily high on the muscle. Since the mesh measures 15 15 cm, we do not think we compromise the strength of the repair by merely removing a small piece of its inferolateral corner. We think that dissecting very high on the muscle just to position the whole inferolateral corner of the mesh flat on the muscle is unnecessary and dangerous. We do agree totally that fixation of the mesh is unnecessary provided the mesh is large enough, which is obviously the case when using a 15 15 cm mesh. In fact, we have not fixed the mesh since January 1996. We also agree that the vas deferens is not clearly seen on Fig. 1. We chose this shot because of the clearly visible large direct defect.

C. de Gheldere
Heilig Hart Ziekenhuis Kolveniersvest 20B-2500 LIER Belgium

P. Vanclooster
Bouwelsesteenweg 6 2560 Nijlen Belgium

Surg Endosc (1997) 11: 698

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The influence of pneumoperitoneum on the peritoneal implantation of free intraperitoneal cancer cells
Recently Hubens et al. published an interesting article entitled The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells [2]. They reported on the possible implantation of cancer cells at trocar wounds at the moment of deflation as cells are forced through these wounds by the pressure gradient created by the pneumoperitoneum during laparoscopic surgery for malignant disease. We would like to point out that this chimney effect, as originally described by us, can occur during the entire laparoscopic procedure and not only at the moment of deflation, as leakage of CO2 alongside trocars during surgery is impossible to prevent with the existing trocars [3]. Consequently, deflation of the pneumoperitoneum by letting CO2 escape through one of the trocars before pulling these trocars out of the abdomen will not prevent the occurrence of entrapment of cancer cells in the trocar wounds. We fully agree with the authors on the possible advantage of gasless laparoscopy as this could prevent the chimney effect. In our experimental work we found significantly less tumor growth at the port sites following gasless laparoscopic surgery for colon cancer in the rat as compared to laparoscopic surgery using a CO2 pneumoperitoneum. This technique seems promising to treat malignant disease laparoscopically.
Correspondence to: G. Kazemier

References
1. Bouvy ND, Marquet RL, Jeekel J, Bonjer HJ (1996) Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Surg Endosc 10: 551 2. Hubens G, Pauwels M, Hubens A, Vermeulen P, Van Marck E, Eyskens E (1996) The influence of a pneumoperitoneum on the peritoneal implantation of free intraperitoneal colon cancer cells. Surg Endosc 10: 809812 3. Kazemier G, Bonjer HJ, Berends FJ, Lange JF (1995) Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 82: 11411142

G. Kazemier1 F. J. Berends1 N. D. Bouvy1 J. F. Lange2 H. J. Bonjer1

1

Department of Surgery University Hospital Rotterdam-Dijkzigt Dr Molewaterplein 40 3015 GD, Rotterdam The Netherlands 2 Department of Surgery St. Clara Hospital Rotterdam The Netherlands

Surg Endosc (1997) 11: 699

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

The author replies

We thank Dr. Kazemier et al. for their kind remarks and fully agree with them that gas leakage can occur during the entire procedure with subsequent implantation of tumor cells at the trocar sites. At the moment we are conducting further experimental studies on the possible effects of gas leakage on tumor cell implantation and the chimney effect, as they have called it. Results will be ready for publication soon. G. Hubens
Department of Surgery University Hospital University of Antwerp Wilrijkstraat 10 2650 Edegem Belgium

Surg Endosc (1997) 11: 650652

Surgical Endoscopy
Springer-Verlag New York Inc. 1997

Early international results of laparoscopic gastrectomies

P. M. Y. Goh, A. Alponat, K. Mak, C. K. Kum
Department of Surgery, National University Hospital, 5 Lower Kent Ridge Road, 119074 Singapore Received: 7 August 1996/Accepted: 28 October 1996

Abstract Background: The first totally laparoscopic Billroth II gastrectomy was performed in 1992. To date, laparoscopic gastrectomy has been performed by a small number of surgeons around the world and the laparoscopic approach has been extended to Billroth I and total gastrectomy. The aim of this study is to review the state of laparoscopically performed gastrectomies in the international scene. Methods: Questionnaires were prepared and sent to every surgeon in the world known by the authors or their contacts to have performed a laparoscopic gastrectomy. A questionnaire survey was started in July 1994 and completed by November 1994. Data collected included age, sex, type of gastric resection, technique of reconstruction after resection, average duration of surgery, time to liquid and solid intake, postoperative hospital stay, complications, and opinions of the surgeons. Results: Sixteen surgeons contributed to this study. A total number of 118 cases of laparoscopic gastrectomies, comprising Billroth I (11), Billroth II (87), vagotomy and antrectomy (10), and total gastrectomy (10) had been performed. The indications were gastric and/or duodenal ulcers and benign and malignant gastric tumors. Conclusions: Laparoscopic gastrectomy was found to be superior to the open technique by 10 of 16 surgeons because of faster recovery, less pain, and better cosmesis. The procedure was an expensive and long operation according to four. Two surgeons were uncertain of any benefit because of limited experience. Key words: Laparoscopic gastrectomy International results Questionnaire survey

sity Hospital, laparoscopic gastrectomy began being performed by a number of surgeons around the world. Some modifications to the original procedure have been described [4, 6]. The laparoscopic approach has also been extended to other related operations such as wedge resection, Billroth I gastrectomy, and total gastrectomy. The indications have also been expanded to early gastric cancer and palliative resections for advanced gastric cancer. The aim of this study is to evaluate the early international gastrectomy results. As the data is diverse, uncontrolled, and incomplete, we feel that this study has more value as a historical record of the development of a new procedure than as a scientific testimony to its efficacy at this moment. It would, for instance, be extremely nice to have the data of the initial experience of open gastrectomy around the world during the year 1885, 4 years after its initial development by Theodore Billroth.

Materials and methods

In order to evaluate the state of laparoscopic gastrectomy today, we prepared a questionnaire and sent it to every surgeon in the world known by the authors or their contacts to have performed a laparoscopic gastrectomy. As the procedure had just been described, we assumed that the number would be small and that not many would be left out. The study was started in July 1994 and all questionnaires were returned by 30 November 1994. Several surgeons who stated that they only performed endoscopic mucosal resections or wedge resections were excluded. Our own series of 18 cases with Billroth II gastrectomy were included. Data collected included age, sex, indication for operation, type of gastric resection, technique of reconstruction after resection, numbers of endostaplers (Endo-GIA) used, average duration of surgery, average time to mobilize the patient, time to liquid and solid intake, mean hospital stay, time to return to work, postoperative complications, and surgeons opinion on laparoscopic gastrectomy.

After the first successful totally laparoscopic Billroth II gastrectomy was performed in the Singapore National UniverCorrespondence to: P. M. Y. Goh

Results The questionnaire was answered by all of those surgeons who received it. According to the data collected from these surgeons and our own experience (16 surgeons from 16

651 Table 1. The indications for operation Indication Gastric cancer Gastric ulcer Duodenal ulcer Gastric + duodenal ulcer Benign gastric tumor Non-Hodgkins lymphoma Tubulovillous adenoma Angiodysplasia Gastric volvulus Pancreatic rest Patients, n 118 (%) 46 (38) 35 (30) 23 (19) 5 (4) 3 (3) 2 (2) 1 (1) 1 (1) 1 (1) 1 (1) Mobilization of patient (hours) Liquid intake (days) Solid intake (days) Hospital stay (days) Return to daily activities (days) Table 2. Postoperative recovery Average time (range) 28 (6192) 3 (16) 5 (212) 8 (330) 17 (990)

different centers in 12 countries), the numbers of laparoscopic gastrectomies were 118, comprising 11 Billroth I gastrectomy, 87 Billroth II gastrectomy, 10 vagotomy, and antrectomy and 10 total gastrectomy. The range in number of cases reported by contributors was one to 22. Eighty-four (71%) of the cases were men and 34 (29%) were women; mean age was 49 years ranging from 24 to 91 years. The data provided by contributors represents a consecutive experience. The most common indication for operation was gastric cancer (in 4638% of the cases). The stage of these tumors and lymph node status were not stated in the questionnaire forms. Other common indications were gastric ulcers in 35 and duodenal ulcers in 23 (Table 1). The operations were totally laparoscopic in 94 (79%) and laparoscopic assisted in 24 (21%) of the patients. In laparoscopic-assisted cases, the hand-sewn anastomoses were fashioned between duodenum and proximal transected stomach with the usual technique as in open surgery through a 5-cm transverse muscle-splitting incision over the duodenal cap in patients with Billroth I gastrectomy. In patients with Billroth II gastrectomy, a 5-cm left subcostal musclesplitting incision was made to perform the gastrojejunostomy. The anastomoses were completed totally laparoscopically using staples (Endo GIA) in 60 (51%) patients. The combination of Endo GIA and intracorporeal hand-sewn technique was used in 55 (46%) cases in whom the anastomoses were performed by Endo GIA and side openings were closed with laparoscopic suturing material. The anastomoses were performed totally hand-sewn in three (3%) cases. The mean number of endoscopic staples (Endo GIA) used was five, ranging from four to 16. Conversion to open surgery was mandatory in six (5%) cases because of bleeding in three, stapler failure in one, transection line too close to the esophagus in one, and inability to localize the ulcer site in one. The average duration of the surgery was 215 min, ranging from 90 to 360 min. In the postoperative period, the average time to mobilize patient was 28 h, commencement of liquid intake 3 days, solid food 5 days, and the mean hospital stay 8 days. Return to normal daily activities took an average of 17 days (Table 2). Postoperative complications were anastomotic obstruction in four (4%), bleeding in two (2%), anastomotic leak in three (3%), and sepsis in two (2%). Other complications were duodenal stump leak in one (1%)

and gastric atony in one (1%). There were two operationrelated mortalities; duodenal stump leakage with sepsis in one patient and subhepatic abscess and death at reoperation in the other. One patient died of AIDS-related pneumonia. Laparoscopic gastrectomy was considered superior to open technique by 10 of 16 surgeons contributing to this study because of faster recovery, less pain, and better cosmesis. According to four of those who responded, it was an expensive and long operation. Two surgeons were uncertain due to the small number in their personal series.

Discussion Laparoscopic gastrectomy has been performed by a number of surgeons around the world and many modifications to the original technique have been described. Besides the totally intraperitoneal laparoscopic gastrectomy, laparoscopicassisted gastrectomy either by gasless technique (elevating the abdominal wall with U-shaped retractor) or insufflating the abdomen by CO2 has been practiced by some authors [12]. The laparoscopic approach can now be extended to other related operations such as Billroth I gastrectomy, wedge resection, and total gastrectomy [1, 7, 10, 12 14]. Indications for laparoscopic Billroth II gastrectomy include: (1) chronic duodenal or pyloric channel ulcer which has failed to heal after adequate medication and Helicobacter pylori eradication; biopsy must be taken to exclude malignancy; (2) failure of benign gastric peptic ulcer to heal after 3 months of well-supervised and compliant medical treatment; (3) bleeding gastric ulcer after failure of endoscopic hemostasis; (4) a perforated benign gastric ulcer with minimal soilage; (5) early gastric cancer, and (6) palliative resection in advanced gastric carcinoma [3, 5, 8]. Relative contraindications are previous upper abdominal surgery and severe cardiopulmonary disease. Indications for laparoscopic Billroth I gastrectomies are the same as for Billroth II gastrectomies. Technically, in most cases of Billroth I procedures, dissection and resection phases are carried out under the conditions of pneumoperitoneum, but the gastroduodenostomy is done by open technique through a small incision in the right upper quadrant [1, 12]. Wedge resections can easily be performed with endostaplers and are mainly indicated in benign tumors of the stomach [9] and selected early gastric cancers [8]. The indications for laparoscopic gastrectomy has been expanded to early gastric cancer and palliative resections of

652

advanced gastric cancer. However, the gastric malignancies should be carefully evaluated and selected for laparoscopic approach. Thus, in early gastric cancer, partial resection, intragastric mucosal resection, or distal partial gastrectomy can be performed. Laparoscopic palliative resection for advanced gastric carcinoma is also feasible. The place of the laparoscopic approach in stage II and III gastric carcinomas is still uncertain. In these cases a meticulous and extensive nodal dissection can improve survival. Melotti and Azagra, however, reported encouraging preliminary results in these intermediate stages [2, 11]. The general surgical community remains unconvinced and longer follow-up and bigger trials are required. There are certain advantages to laparoscopic gastrectomy. It holds promise of less pain, less immobility, quicker alimentation, less wound and respiratory complications, and early return to normal daily activities. Therefore, this technique is patient-friendly. On the other hand, with the technology we have today, it is still quite a technical challenge to the surgeon. In addition, the high cost of endoscopic stapling devices is an important drawback of this technique. However, this can be overcome when surgeons improve their hand suturing skills. Many suturing assist devices are now available to make laparoscopic anastomosis less daunting. Most surgeons performing this operation were impressed by the prompt recovery, earlier mobilization, and prompt recovery of gut function in the successful cases. However, the real benefit of this procedure will not be scientifically proven for some time due to difficulties in collecting large numbers of cases and documenting them. A randomized controlled trial comparing open and laparoscopic gastrectomy will also be difficult to carry out as the indications in benign disease are shrinking as a consequence of the proven effectiveness of Helicobacter pylori eradication therapy. This is very preliminary data. Therefore, we can not draw very definitive conclusions from this study. It just documents the preliminary efforts of surgeons around the world experimenting with this new technique. We are sure that in 510 years, the situation will become clearer because more data, and more scientific presentation of data, will be available. This data, however, has historical value because it is the

first worldwide survey of this very new procedure. Fifty to 100 years from now, when people look back, they will at least have some information as to how this procedure started and spread around the world.
Acknowledgment. The authors wish to thank the following surgeons who contributed to this international study: JS Azagra (Belgium), DL Fowler (USA), JR Mauras (France), G Melotti (Italy), Y Nagai (Japan), N Katkhouda (USA), V Lange (Germany), D Litwin (Canada), D Rosin (England), M Suppiah (Malaysia), G Szinicz (Germany), JF Uddo, Jr (USA), DI Watson (Australia), T Wuttichai (Thailand), CZ Zheng (China).

References
1. Anvari M, Park A (1994) Laparoscopic-assisted vagotomy and distal gastrectomy. Surg Endosc 8: 13121315 2. Azagra JS, Goergen M (1994) Laparoscopic total gastrectomy. In: Meinero M, Melotti G, Mouret PH (eds) Laparoscopic surgery. Masson, Milano, pp 289296 3. Goh P (1994) Laparoscopic Billroth II gastrectomy. Semin Laparosc Surg 1(3): 171181 4. Goh P, Kum CK (1993) Laparoscopic Billroth II gastrectomy: a review. Surg Oncol 2(Suppl 1): 1318 5. Goh P, Kum CK (1995) Laparoscopic Billroth II gastrectomy. In: Phillips HP, Rosenthal RJ (eds) Operative strategies in laparoscopic surgery. Springer, New York, pp 155158 6. Goh P, Tekant Y, Kum CK, Isaac J, Ngoi SS (1992) Totally intraabdominal laparoscopic Billroth II gastrectomy. Surg Endosc 6: 160 7. Kitano S, Iso Y, Moriyama M, Sugimachi K (1994) Laparoscopicassisted Billroth I gastrectomy. Surg Laparosc Endosc 4(2): 146148 8. Kitano S, Shimoda K, Miyahara M (1995) Laparoscopic approaches in the management of patients with early gastric carcinomas. Surg Laparosc. Endosc 5(5): 359362 9. Llorente J (1994) Laparoscopic gastric resection for gastric leiomyoma. Surg Endosc 8: 887889 10. Lointer P, Leroux S, Ferrier C, Dapoigny M (1993) A technique of laparoscopic gastrectomy and Billroth II gastrectomy. J Laparoendosc Surg 3(4): 353364 11. Melotti G, Meniero M, Tammborrino E (1994) Gastric resection for cancer. In: Meinero M, Melotti G, Mouret PH (eds) Laparoscopic surgery. Masson, Milano, pp 273282 12. Uyama I, Pgiwara H, Takahara T, Kato Y, Kikuchi K, Iida S (1994) Laparoscopic and minilaparotomy Billroth I gastectomy for gastric ulcer using an abdominal wall-lifting method. J Laproendosc Surg 4(6): 441445 13. Uyama I, Ogiwara H, Takahara T, Kato Y, Kikuchi K, Iida S (1995) Laparoscopic Billroth I gastrectomy for gastric ulcer: technique and case report. Surg Laparosc Endosc 5(3): 209213 14. VanHouden CE (1994) Laparoscopic bilateral truncal vagotomy, antrectomy, and Billroth I anastomosis for prepyloric ulcer. Surg Laparosc Endosc 4(6): 457460