Hirschsprung's Disease in the Preterm Infant: Implications for Diagnosis and Outcome KATHERINE J. BAXTER, B.S., AMINA M.

BHATIA, M.D. From the Division of Pdiatrie Surgery, Emory University and Children's Healthcare of Atlanta, Atlanta, Georgia Hirschsprung's disease (HD), congenital absence of ganglion cells, is considered uncommon in preterm infants. The aim was to describe the frequency, presentation, and surgic al outcomes of preterm infants with HD. A retrospective cohort study was conducted of all patie nts diagnosed with HD from 2002 to 2012 at a single children's hospital. Clinical presentation and surgical outcomes were obtained for term (37 weeks of gestation or greater) and preterm i nfants. One hundred twenty-nine subjects with HD were identified, 24 (19%) preterm and 105 ( 81%) term. Preterm infants were more likely to be diagnosed after 30 days of life (66.7 vs 37.1%, P < 0.01; median age 2.9 vs 0.3 months, P < 0.05) and to have associated major congenital anomalies (45.8 vs 20.0%, P < 0.01). Fewer preterm infants had primary pull-through operations (45. 8 vs 76.2%, P < 0.005). Preterm infants were more likely to have an episode of Hirschsprung's-as sociated enterocolitis (45.8 us 24.0%, P < 0.05) but were not more likely to die from any cause (8.3 vs 5.8%, P = 0.64). HD may be more common in preterm infants than previously recognized, and increased comorbidities in these patients may lead to delayed diagnosis and increased morb idity. HD should be considered in the preterm infant presenting with a bowel obstruction, especially when accompanied by associated anomalies. H H IRSCHSPRUNG'S DISEASE (HD) is a congenital absence of ganglion cells in the distal bowel, which extiends varying lengths into the more proximal large bowel and less commonly into the small bowel. Hirschsprung's disease commonly presents as neonatal bowel obstruction with delayed passage of meconium (DPM), abdominal distension, and bilious emesis. Because of these obstructive signs and symptoms, the majority of infants with HD are diagnosed within the first month of life. ' HD has been generally considered uncommon in preterm infants. Although recent evidence is sparse, previous estimates for the proportion of children with a history of prematurity among children with HD range from 3 to 9 per cent.'^--^ However, because preterm infants frequently have feeding and stooling difficulties, the classic signs of intestinal obstruction may not be recognized secondary to the effects of prematurity itself Preterm infants often have complicated medical courses including respiratory failure, bradycardia, feeding intolerance, and a higher burden of

congenital anomalies.^ DPM is common in the preterm Presented at the Annual Scientific Meeting and Postgraduate Course Program, Southeastern Surgical Congress, Jacksonville, FL, February 9-12, 2013. Address correspondence and reprint requests to Katherine J. Baxter, B.S., Medical Student, Pdiatrie Surgery, Emory University, 1424 Tuxworth Circle, Decatur, GA 30033. E-mail: katiej.baxter gmail.com. infant with approximately 32 per cent being delayed past 48 hours.^ Frequent use of orogastric tubes and parenteral nutrition in preterm infants may blunt the appearance of distension and emesis. In addition, signs of a bowel obstruction and/or colitis caused by HD are difficult to differentiate from necrotizing enterocolitis (NEC),^ among other causes. Therefore, diagnosis of HD may be more difficult in the preterm infant, leading to delay or misdiagnosis of neonatal obstruction. Delayed diagnosis of HD beyond 1 week after birth significantly increases the risk of serious complications in neonatal patients.^ Patients may be at higher risk of Hirschsprung's-associated enterocolitis (HAEC), which can progress to sepsis and become life-threatening. Delayed diagnosis also affects the surgical treatment of HD.^ These patients are more likely to need an ostomy as opposed to a primary pull-through as a result of massive colonie dilation secondary to chronic constipation or scarring and adhesions from HAEC. We hypothesized that HD is more common in preterm infants than previously recognized and that diagnosis is more often delayed in these patients compared with term infants, resulting in increased morbidity and differential surgical management. Patients and Methods A retrospective chart review was conducted for patients diagnosed with HD from lanuary 2002 to January 734

No. 7 HIRSCHSPRUNG'S DISEASE IN THE PRETERM INFANT Baxter and Bhatia 2012 at Children's Healthcare of Adanta, Egleston Hospital. Internal Review Board approval was obtained. Charts were reviewed for demographics, birth history, presentadon history, surgical treatment, and outcomes. Term gestadon was defined as greater than or equal to 37 weeks of gestadon. An episode of HAEC was defined using the criteria developed by Pastor et al.' HAEC was diagnosed if all four of the following were present: diarrhea or obstipadon, explosive stool on rectal examination, abdominal distension, and radiographie evidenee of bowel obstruction. Emergent operation was defined as an operadon within 24 hours of admission before confirmation of diagnosis by biopsy. Variables were eompared between term and preterm infants. The X^ test of independence was applied to eategorieal variables with greater than five subjects per variable value. Eisher's exaet test was used to compare proportions with less than five subjects per cell. The two-sample independent test was used for normally distributed eontinuous variables, and the nonparametric median test was used for nonnormal condnuous variables. Significance was defined as P < 0.05. Results Chart review idendfied 132 subjeets diagnosed with HD from 2002 to 2012. Three pafients were excluded from the term and preterm comparison as a result of missing gestafional age data but were included in the overall cohort. Among 129 padents with HD with complete gestadonal age informadon, 24 (19%) were preterm and 105 (81%) were term. The overall cohort was 81 per cent male (4.28:1 rado) and the median age at diagnosis was 0.49 months. The most eommon presendng signs of HD were abdominal distension (91.5%) and emesis (56.6%). Delayed passage of meconium was documented in 47.7 per cent of padents. A family history of HD was reported in 5.6 per cent of cases. A barium enema was posidve for a transidon zone in 62.9 per eent of padents. Associated major congenital anomalies were found in 24.4 per cent of padents. The most eommon assoeiated major congenital anomahes were trisomy 21 in eight (6.1%) patients and congenital central hypovendlation syndrome (CCHS) in five (3.8%) padents. Other abdominal anomalies were found in six padents and included malrotadon, duodenal atresia, ileal atresia, pyloric stenosis, and gastroschisis. The most common site of aganglionosis was the sigmoid colon (31.8%). A primary pull-through operadon was performed in 70.5 per eent of patients, and the Soave technique was used in 77.3 per cent of padents. A postsurgical stricture occurred in three (2.3%) padents and anastomode leak oecurred in two (1.5%) padents. Dmographie and presentation data for term and

preterm subjects are shown in Table 1. Preterm infants had lower 1-minute (6.36 vs 7.60, P ^ 0.014) and 4-minute (7.50 v^ 8.65, P < 0.01) Apgar scores and were more likely to be intubated as a neonate (50.0 vs TABLE 1. Characteristics of Term and Preterm Infants with Hirschsprung's Disease * Sex Male Female Rarp White Black Other Family history of Hirschsprung's Gestational age at birth (mean in weeks) Age at diagnosis (median in months) Transition zone Rectosigmoid Long segment colon Total colon Congenital anomalies Presentation Emesis Distension Delayed passage of meconium t Mass Constipation Positive contrast study Term(n = 105) Preterm (n = 24) P value 84 (80.0) 20 (83.3) 0.709 21 (20.0) 4 (16.7) 25 (25.5) 12(57.1) < 0.0001 + 62 (63.3) 3 (14.3) 11 (11.2) 6 (28.6) 5 (4.9) 2 (9.5) 0.091 38.97 33.45 < 0.0001 0.33 2.88 0.017 74 (70.5) 18 (75.0) o.iiot 24 (22.9) 2 (8.3) 7 (6.7) 4 (16.7) 21 (21.0) 11 (45.8) 0.008 63 (61.8) 9 (37.5) 0.031 95 (92.2) 22(91.7) 0.926 52 (57.8) 11 (73.3) 0.255 1 (1.0) 2 (8.3) 0.120 44 (42.3) 15 (62.5) 0.074 70 (66.7) 12 (50.0) 0.126 * Data presented as number (percentage) except where otherwise noted.

t All percentages are calculated from nonmissing data. Delayed passage of meconi um data were more often missing from the record of preterm infants than term infants (37.5 v.5 14.7%). \ P value for the overall x^ test of independence. In pairwise x^ comparison with Bonferroni correction denotes a significantly lar ger proportion in that column.

736 THE AMERICAN SURGEON July 2013 Vol. 79 7.6%, P < 0.0001). Preterm infants were diagnosed significantly later in life than term infants with a median age at diagnosis of 2.9 months compared with 0.3 months in term infants {P < 0.01) and, similarly, were more likely to be diagnosed after 30 days of life (66.7 v.y 37.1%, P < 0.01). Preterm infants were more likely to have associated major congenital anomalies (45.8 vs 20.0%, P < 0.01). With respect to the two most common associated anomalies, preterm infants were more likely to have trisomy 21 (16.7 vs 3.8%, P = 0.039) or CCHS (12.5 vs 1.9%, P = 0.044). All associated congenital anomalies in term and preterm infants are shown in Table 2. There was a nonsignificant trend toward more emergent operations in preterm versus term infants (20.0 vs 9.5%, P = 0.128) and fewer positive barium enema examinadons (50.0 v^' 66.7%, P = 0.104). There was also a trend toward more frequent total colon aganglionosis in preterm infants (16.7 vs 5.7%, P = 0.0891). Fewer preterm infants had primary pull-through operadons (45.8 vs 76.2%, P < 0.005), but the frequency of permanent stomas was not stadsdcally different between preterm and term infants (12.5 vs 4.8%, P = 0.37) (Table 3). A total of nine subjects had permanent stomas: five for total colon aganglionosis, two for chronic enterocolids, one for fatal perforated enterocolids, and one for associated pulmonary insufficiency. Twenty-eight subjects had temporary stomas either separately or as part of their two-stage pullthrough; indicadons included enterocolids (eight), long segment aganglionosis (six), delayed diagnosis (four), overly dilated proximal bowel (three), preoperative perforadon (three), meconium peritonids (one), and dusky appearance of distal anastomosis (one). Preterm infants were more likely to have an episode of HAEC either preor postoperadvely (45.8 vs 24.0%, P < 0.05) but were not more likely to die from any cause (8.3 v^' 5.8%, P = 0.64). Discussion Congenital megacolon or HD is one of the most common causes of neonatal distal bowel obstruction in the term neonate. However, as a result of their intestinal immaturity, preterm infants are vulnerable to disorders that are unique to the preterm infant. In preterm infants, the differential for bowel obstruction includes not only funcdonal bowel dysmodlity, but also mechanical disease processes such as meconium obstruction of prematurity (meconium plug syndrome), milk curd obstrucdon secondary to high-caloric density feeds, NEC, and spontaneous intestinal perforation. Hence, HD is less commonly considered in the differendal diagnosis of preterm intestinal obstruction than in the term neonate. In this study, we found that HD

TABLE 2. Associated Major Congenital Anomalies in Term and Preterm Infants with Hirschsprung's Disease Term Preterm Anomaly (n = 105) (n = 24) CCHS 2(1.9) 3(12.5) Trisomy 21 4 (3.8) 4 (16.7) Other syndromes* 5 (4.8) 1 (4.2) Other genetic anomalies! 1 (10) 0 Intestinal anomalies i: 4 (3.8) 2 (8.3) Cardiopulmonary anomalies 7 (6.7) 3 (12.5) Renal anomalies|| 3 (2.9) 0 * Mowat-Wilson (two), Rubenstein-Taybi, cardiofacial cutaneous syndrome, DiGeorge syndrome, Waardenburg syndrome. t 13q deletion. t Pyloric stenosis, malrotation/nonrotation (three), duodenal stenosis, ileal atresia, gastroschisis. Aortic root dilation, complete arteriovenous canal, ventricular septal defect (four), transposition of great arteries, hypoplastic aortic arch, double outlet right ventricle, hypoplastic left ventricle, pulmonary vein stenosis, agenesis of left lung. II Left renal agenesis, hypoplastic right kidney, renal duplication cyst. CCHS, congenital central hypoventilation syndrome. TABLE 3. Outcomes of Term and Preterm Infants with Hirschsprung 's Disease * Term Preterm (n = 105) (n = 24) P Value Primary pull-through 80 (76.2) 11 (45.8) 0.0003 Permanent stoma 5 (4.8) 3 (12.5) 0.367 Enterocolitis Preoperative 14(13.7) 5 (20.8) 0.381 Postoperative 14(13.5) 8 (33.3) 0.020 Any 25 (24.0) 11 (45.8) 0.032 Operative 2 (1.9) 3 (12.5) 0.045 complications! Mortality 6 (5.8) 2 (8.3) 0.643 * Hirschsprung's disease-associated enterocolitis defined based on the criteria developed by Pastor et al."^ t Stricture (three), anastomotic leak (two). may be more common in preterm infants than previously recognized. We found that 19 per cent of our cohort was preterm infants of less than 37 weeks of gestadon. As expected, in our study, preterm infants had lower Apgar scores and were more likely to be intubated as neonates. Overall complexity of a preterm infant's condidon and the intensity of their care may contribute to the

delay in diagnosis of HD. Other complications of prematurity are numerous but include feeding and stooling difficulty as well as other abdominal disease processes such as NEC, intestinal atresia, and meconium plug syndrome that overlap in presentation with and may be difficult to disdnguish from HD.''- ' ' Associated congenital anomalies also increase the complexity of an infant's presentation, ' ^ but if condidons associated with HD are recognized, they may instead help to expedite a diagnosis. The overall percentage of subjects with associated major anomalies in our study (24.4%) was consistent with previous studies ranging

No. 7 HIRSCHSPRUNG'S DISEASE IN THE PRETERM INFANT Baxter and Bhatia from 11 to 32 per cent with an average of 21 per cent.'^ The increased prevalence of anomalies among preterm infants (45.8%) in our study is similar to the 47 per cent found by Klein et al. in 1993^ but appears to be the first fime this finding is supported by stafisfical tesfing. Idenfiftcation and epidemiology of associated anomalies with HD impacts diagnosis in both term and preterm infants and may lead to clues in the pathogenesis of HD. The association of HD with other neurocristopathies such as CCHS (Haddad syndrome) is well recognized, and the genefic locus at PH0X2B has been idenfified, but this rare syndrome has not been previously associated with preterm birth except in a case report."^' '"* '^ Down syndrome (DS) or trisomy 21 is also known to be associated with HD, and in fact, the incidence of DS in pafients with HD may be increasing. Our overall prevalence of DS was 6.2 per cent, which is shghfiy lower than previous reports of 8 to 13 per cent, ' ^ ' ^ but among preterm infants, it was significantly higher at 16.7 per cent. Although reports differ on whether DS is associated with worse outcomes in HD, DS has been associated with increased HAEC,'^-'^ and this may parfially explain our finding of increased morbidity in the preterm group. The associafion in our cohort of DS and HD with prematurity has not been previously described. Mowat-Wilson syndrome, found in two subjects in this study, is a rare condition characterized by dysmorphic features, intellectual disability, and the presence of HD, most often caused by a de novo mutafion of the gene ZEB2?^ Associafion of RubensteinTaybi syndrome with HD was described in a case report but appears to be quite rare.^' Our results also agree with previous reports of an associafion of HD with other gastrointestinal tract anomalies (chiefiy malrotation and intestinal atresia), which may provide clues to the etiology of HD through common developmental signaling pathways.' '' ^^ Recent trends in the treatment of HD, including more frequent use of primary puU-through and predominance of the Soave endorectal technique,^^ are consistent with our institution's experience and agree with other recent HD case series.'-'^ The reasons for fewer primary operafions in preterm infants with HD are unclear. It seems hkely that this different treatment strategy is related to delayed diagnosis, increased comorbidities, greater prevalence of HAEC, and possibly greater extent of aganghonosis in preterm patients. Comphcations of delayed diagnosis in HD were examined by Lee et al.,^ and their findings of increased HAEC agreed with ours despite differences in the length of delay and definition of HAEC used. Future studies are needed to examine whether early diagnosis would ameliorate this increased morbidity or if it is driven by prematurity itself and associated comphcations.

Despite the prevaiUng belief that HD is uncommon in preterm infants, there is a paucity of evidence on the subject. The prevalence of preterm infants in our HD case series is quite high compared with previous reports, which have ranged from 3 to 9 per cent,^' 2'*' ^^ although most of these case series were limited by sample size (n less than 35) and were published before more recent advances in the care of preterm infants. In the largest case series to report on preterm prevalence to our knowledge, Klein et al.^ found 8 per cent preterm infants in 250 patients with HD. Possible contributing factors to our much higher preterm prevalence are the increased incidence of prematurity over fime^^ and improved diagnostic tesfing for HD. Additionally, the single-institution nature of this study, its terfiary referral status, and large Level III neonatal intensive care unit is an important potential bias. Although we cannot comment on incidence in this retrospective study, it is interesting to note that the most recent estimate of the preterm birth rate in the United States is 12.3 per cent of all births,^^ and when compared with the rate among pafients with HD in this study of 19 per cent, this suggests that HD is in fact associated with prematurity. Further studies are needed to investigate the incidence of HD in preterm infants, but our study suggests that it is higher than previously thought and that there are important implications of delayed diagnosis in these complicated pafients. In conclusion, we present novel evidence that HD is relatively frequently associated with prematurity and that preterm patients with HD have significant morbidities in the form of delayed diagnosis and increased incidence of HAEC. Associated comorbidifies are more common in preterm infants with HD, including CCHS, trisomy 21, and intesfinal anomalies. HD should be considered in the preterm infant presenfing with a neonatal bowel obstruction, especially when accompanied by associated anomalies. REFERENCES 1. Singh SJ, Croaker GD, Manglick P, et al. Hirschsprung's disease: the Australian Paediatric Surveillance Unit's experience. Pediatr Surg Int 2003;19:247-50. 2. Polley TZ, Coran AG. Hirschsprung's disease in the newborn. Pediatr Surg Int 1986; 1:80-3. 3. Klein MD, Philippart AI. Hirschsprung's disease: three decades' experience at a single institution. J Pediatr Surg 1993;28: 1291-3; discussion 1293-^. 4. Bajaj R, Smith J, Trochet D, et al. Congenital central hypoventilation syndrome and Hirschsprung's disease in an extremely preterm infant. Pediatrics 2005;l 15:e737-8. 5. Honein MA, Kirby RS, Meyer RE, et al. The association between major birth defects and preterm birth. Matem Child Health J2009;13:164-75. 6. Weaver LT, Lucas A. Development of bowel habit in preterm infants. Arch Dis Child 1993;68:317-20.

738 THE AMERICAN SURGEON July 2013 Vol. 79 7. Raboei EH. Necrotizing enterocolitis in full-term neonates: is it aganglionosis? Eur J Pediatr Surg 2009;19:101^. 8. Lee CC, Lien R, Chian MC, et al. Clinical impacts of delayed diagnosis of Hirschsprung's disease in newborn infants. Pediatr Neonato! 20l2;53:133-7. 9. Ekenze SO, Ngaikedi C, Obasi AA. Problems and outcome of Hirschsprung's disease presenting after 1 year of age in a developing country. Wodd J Surg 2011;35:22-6. 10. Pastor AC, Osman F, Teitelbaum DH, et al. Development of a standardized definition for Hirschsprung's-associated enterocolitis: a Delphi analysis. J Pediatr Surg 2009;44:251-6. 11. Keckler SJ, St Peter SD, Spilde TL, et al. Current significance of meconium plug syndrome. J Pediatr Surg 2008;43:896-8. 12. Das K, Alladi A, Kini U, et al. Hirschsprung's disease, associated rare congenital anomalies. Indian J Pediatr 2001 ;68: 835-7. 13. Moore SW. The contribution of associated congenital anomalies in understanding Hirschsprung's disease. Pediatr Surg Int 2006; 22:305-15. 14. Haddad GG, Mazza NM, Defendini R, et al. Congenital failure of automatic control of ventilation, gastrointestinal motility and heart rate. Medicine 1978;57:517-26. 15. Lai D, Schroer B. Haddad syndrome: a case of an infant with central congenital hypoventilation syndrome and Hirschsprung disease. J Child Neurol 2008;23:341-3. 16. Hackam DJ, Reblock K, Barksdale EM, et al. The influence of Down's syndrome on the management and outcome of children with Hirschsprung's disease. J Pediatr Surg 2003;38: 946-9. 17. Morabito A, Lall A, Gull S, et al. The impact of Down's syndrome on the immediate and long-term outcomes of children with Hirschsprung's disease. Pediatr Surg Int 2006;22:179-81. 18. Ieiri S, Higashi M, Teshiba R, et al. Clinical features of Hirschsprung's disease associated with Down syndrome: a 30-year retrospective nationwide survey in Japan. J Pediatr Surg 2009;44: 2347-51. 19. Quinn FM, Surana R, Puri P. The influence of trisomy 21 on outcome in children with Hirschsprung's disease. J Pediatr Surg 1994;29:781-3. 20. Adam MP, Bean LJH, Miller VR. Mowat-Wilson syndrome, tn: Pagon RA, Bird TD, Dolan CR, et al., eds. GeneReviews. National Center for Biotechnology Information (NCBI). University of Washington. Seattle, WA; 1993. 21. Isidor B, Podevin G, Camby C, et al. Rubinstein-Taybi syndrome and Hirschsprung disease in a patient harboring an intragenic deletion of the CREBBP gene. Am J Med Genet A 2010; 152A: 1847-8. 22. Berger S, Ziebell P, Offsler M, Hofmann-von Kap-herr S. Congenital malformations and perinatal morbidity associated with intestinal neuronal dysplasia. Pediatr Surg Int 1998;13:474-9. 23. Georgeson KE, Cohen RD, Hebra A, et al. Primary laparoscopicassisted endorectal colon pull-through for Hirschsprung's disease: a new gold standard. Ann Surg 1999;229:678-82; discussion 682-3. 24. Bowring AC, Kern IB. The management of Hirschsprung's disease in the neonate. Aust Paediatr J 1972;8:121-7. 25. Lister J.- Hirschsprung: the man and the disease. J R Coll

Surg Edinb 1977;22:378-84. 26. Beck S, Wojdyla D, Say L, et al. The woridwide incidence of preterm birth: a systematic review of maternal mortality and morbidity. Bull Wodd Health Organ 2010;88:31-8. 27. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Mathews TJ, Osterman MJ. Births: final data for 2008. Nati Vital Stat Rep 2010;59:1,3-71.

Copyright of American Surgeon is the property of Southeastern Surgical Congress and its content may not be copied or emailed to multiple sites or posted to a listserv w ithout the copyright holder's express written permission. However, users may print, downloa d, or email articles for individual use.