Journal of Image Guided Surgery

ISSN: 1078-7844 (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/icsu19

Tumor Resection by Stereotactic Craniotomy Using

the Brown-Roberts-Wells System

Robert J. Maciunas

To cite this article: Robert J. Maciunas (1995) Tumor Resection by Stereotactic Craniotomy
Using the Brown-Roberts-Wells System, Journal of Image Guided Surgery, 1:4, 208-216

To link to this article: http://dx.doi.org/10.3109/10929089509106326

1995 Informa UK Ltd All rights reserved:

reproduction in whole or part not permitted

Published online: 06 Jan 2010.

Submit your article to this journal

Article views: 16

View related articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=icsu19

Download by: [36.72.231.171] Date: 20 May 2016, At: 18:05

 Journal of Image Guided Surgery 1:208-2 I 6 (1 995)

Laboratow Investigation
Tumor Resection by Stereotactic Craniotomy Using
the Brown-Roberts-Wells System
Robert J. Maciunas, M.D., F.A.C.S.
Department of Neurological Surgery and VanderbiltBrain Tumor Center, Vanderbilt UniversityMedical
Center;Department of Biomedical Engineering, Vanderbilt University;and Neurosurgical Service,
Nashville, VeterawAdministration Hospital, Nashville, Tennessee
Precise localization of subcortical targets contributes to the technical challenge of craniotomies. To
Downloaded by [36.72.231.171] at 18:05 20 May 2016

address this challenge, the application of readily available stereotactic localization techniques to open
craniotomies was investigated. Over a 2-year period, 6 2 consecutive stereotactic craniotomies were
performed successfully using the CT-compatible Brown-Roberts-Wells (BRW) apparatus. Standard
BRW hardware and software were employed. This series consists of craniotomies in 50 patients for
resection of subcortical mass lesions. Targets were consistently and precisely localized by the stereo-
tactic frame. Pathology revealed 32 metastases, 18 glial tumors, 5 nonglial tumors, and 7 nonneoplas-
tic lesions. Histology differed from presumptive diagnoses by neurodiagnostic imaging studies in
30.6% of cases. The average volume of tumors resected was 55,903 mm3.Gross total resection of all
solid tumor tissue was consistently confirmed by postoperative contrast-enhanced CT. Postopera-
tively, 38 patients with masses were neurologically improved, 22 were unchanged, and 2 were worse.
Median postoperative survival for glioblastoma multiforme after adjuvant therapy was 58.7 weeks and
for metastases was 39.2 weeks. There were no postoperative deaths. Overall surgical morbidity was
3.7%. CT-directed stereotactic craniotomy using the BRW system is a safe, efficacious, and readily
available technique. It successfully confers the precision of stereotactic methodology on open
microneurosurgicalprocedures. J Image &id Surg 1:208-216 (1995). 91996 Wiey-Liss, I~C.

Key words: stereotaxis, stereotactic surgery, computed tomography, brain tumor, glioma,
metastatic tumor

INTRODUCTION
Precise localization of subcortical targets contributes monly, however, stereotacticmethodology is applied
to the technical challengeof craniotomies.8*22.24~26*27,29
to performance of biopsies, interstitialbrachytherapy,
Mass lesions may be deep seated or in eloquent re- or creation of lesions for functional applications
gions of brain; they may be irregularly shaped and rather than to directing the volumetric resection of a
poorly demarcated from surrounding viable paren- mass lesion.1 2 7* 11,U + 5

chyma. Potentially, these characteristics can limit the Several investigators have anecdotally de-
degree of surgical resection during a conventional scribed the incorporation of various stereotactic de-
open c r a n i ~ t o m y . ~The
~ . ~consistent
~ - ~ ~ accuracy and vices into the performance of open cranial proce-
precision of image-directed stereotaxy offers a po- dures..12 Shelden and colleaguesz8 developed a
tential aid to localization of such targets. Most com- modification of the Reichert-Mundiger stereotactic

Received original September 26, 1995; accepted December 1, 1995.

Address correspondencekeprint requests to Robert 1. Maciunas, M.D., Associate Professor, Department of Neurological Sur-
gery, T-4224 Medical Center North, Vanderbilt University, Nashville, TN 37232. E-mail: robert.maciunas@mcmail.vanderbilt.edu.
01996 Wiley-Liss, Inc.

frame that was driven by computerized image pro-
cessing capabilities and coupled to an illuminated
tulip-type endoscopic retractor termed the "resecto-
scope." Kelly and colleague^^^-^' designed an espe-
cially elegant system of computer-assisted volumet-
ric stereotaxy. This degree of advanced technology
is not commonly available, however, precluding the
general availability of these particular systems for
combining stereotactic localization and open micro-
surgical technique.17 The Brown-Roberts-Wells
(BRW) apparatus is the most prevalent stereotactic
system in the United States today.*s7 Heilbrun,
Apuzzo, and others have occasionally used the BRW
system to guide corticotomies and endoscopic ven-
tricular procedure^.^^^*'^ The present clinical series is
reviewed in order to assess the safety and efficacy of
the BRW stereotactic system when it is used for guid-
Downloaded by [36.72.231.171] at 18:05 20 May 2016
ance during open craniotomies.
MATERIALS AND METHODS
Standard, commercially available BRW stereotactic
hardware and software were employed in this series,
with only minor modifications.
Frame Application
After administration of adequate intravenous sedation
with midazolam to the patient in the anesthesiaprepa-
ration area, a CT-compatible BRW head frame was
affixed to the patient's scalp in a routine fashion. The
head was shaved and prepared with povidone-iodine
(Betadine) solution. Four fixation pin sites were in-
jected subcutaneously with a total of 10 cc of a solu-
tion containing equal parts of 1% lidocaine with
1:100,000 epinephrine and 0.75% bupivicaine. Alu-
minum-tipped pins in short sleeves were used to pre-
clude CT artifact.= The frame was a x e d such that
the anterior aspect of the ring was situated below the
nasion for lesions anterior to the central sulcus and
below the inion for targets posterior to the central sul-
cus. This allowed the most stable configuration for
attachment to the Mayfield-type head support at sur-
gery. For posterior fossa approachesto the lateral cer-
ebellar hemisphere and for targets in the temporal lobe,
the base ring was canted steeply to gain access to this
region. Under these circumstances,the ability to angle
the orientation of the support posts was beneficial.
Database Acquisition and
Treatment Planning
In the course of preoperative evaluation, most pa-
tients underwent MR imaging scans of their mass
lesions and preoperative angiography. Angiography
was deemed beneficial in assessing most adequately
the superficial vessels as well as the perilesional vas-
Maciunas: Stereotactic Craniotomy 209
culature. Subsequent to stereotactic frame applica-
tions, all patients underwent CT scanning on a DR-
H Siemens Somatom CT unit for purposes of stereo-
tactic target localization. CT scan slices of 4 mm
thickness were obtained. Digitization of fiducials
proceeded at the CT console in a clockwise direc-
tion, starting with the fiducial with the largest cross
section. In patients undergoing adrenal transplanta-
tion, the target selected on uncontrasted CT scans
was the interface between the frontal horn of the right
lateral ventricle and the dorsomedial ependymal sur-
face of the head of the caudate. For mass lesions, the
center of gravity for deep-seated and symmetric le-
sions on double-dosed contrast-enhanced CT scans
was typically selected as the target, whereas several
targets were chosen about the periphery of irregular
and superficial lesions abutting eloquent regions of
brain. Entry points were selected from CT images,
based on surface landmarks or superficial aspects of
the mass lesions, such that the surgical trajectory,
calculated to access the target point(s) via the entry
point(s), would traverse the shortest path through
nonessential brain parenchyma. This information was
used to define one or more vectors describing local-
ization of the target point or mass in stereotactic
space. The standard stereotactic software package,
as provided for the Epson HX-20 microcomputer
(Radionics, Burlington, MA), was used for target
localization in all cases. All trajectories and target
points were simulated on the phantom base unit.
Surgical Procedure
All craniotomies were performed with the patient
under general anesthesia. After CT scanning,
nasotracheal fiber optic intubation was perf~rmed.*~
The patient was appropriately positioned and the base
ring affixed to a Mayfield-type head support. Special
care was directed to sterile draping about the base ring.
The Epson HX-20 microcomputer and its as-
sociated software package currently available for the
BRW system support definition of a point or line in
three-dimensional space but not of a complete target
~ o l u m e . ~To
* ~localize
*'~ stereotactically the target
volumes during craniotomies, each of the clinically
relevant margins of the CT-defined target volume was
defined such that a vector along the surgical approach
trajectory could delimit that margin at surgery. For
some of the more extensive and irregularly shaped
tumors, this resulted in three or more vectors being
described. For deep-seated and fairly spherical le-
sions, a single vector was adequate to orient the sur-
geon by localizing the center of the target volume.
Each vector was defined as the line passing through
two points, a "target point" chosen from a lower CT
 21 0 Man'unas: Stereotactic Craniotomy

scan slice and an "entry point" from a more superfi-

cial CT scan slice. Ventricular catheters were then
directed stereotacticallyalong these preselected vec-
tors to act as guide markers during the surgical pro-
cedure (Fig. 1). This "vector projection method" has
served as an acceptable interim approximationof true
volumetric stereotactic image processing until more
extensive intraoperative computer capabilities be-
come commonly available.
The appropriate stereotactic arc assembly ad-
justments were made to access the trajectory defined
for a pair of target and entry points. The arc assem-
bly was used to guide the scalp incision and cranial
trephination. When appropriate, a single trajectory
was used to direct placement of a pilot hole through
scalp and skull to center the opening. A 6-cm linear Fig. 2. Demonstration of need for offset of small trephine
scalp incision was made, and a 1.75-inch trephine craniotomy, to prevent inadequate exposure due to asymmetry
Downloaded by [36.72.231.171] at 18:05 20 May 2016

craniotomy was elevated. The precision conferred by
stereotacticguidance permitted smaller skin incisions
and skull openings, minimizingtissue dissection. The
use of smaller cranial openings carries with it a cau-
tion, however, related to proper positioning about the
central approach vector to the lesion. Especially for
superficial lesions approached via oblique trajecto-
ries, the cranial trephine must be offset in relation to
the pilot hole to prevent an obscuring "overhang" of
bone. This is due to the inherently asymmetric pro-
jection of edge vectors from the target volume onto
the slope of the skull (Fig. 2).
of skull relative to projections of target edge vectors.
Standard microsurgical technique was used for
the surgical approach. A corticotomy was developed
along the shaft of the guide catheter and maintained
with self-retaining 5-mm refractors. The ultrasonic
aspirator, carbon dioxide, and KTP-532 lasers were
employed as circumstances indicated.
CLINICAL MATERIAL
Over a 2-year period, 62 stereotactic craniotomies
were performed on 50 patients. These cases were
performed for the resection of subcortical intracra-
nial mass lesions. Thirty-three men and seventeen
women were included in this study. The average age
was 48.4 years, with a range of 8 to I4 years.
Presenting Symptoms
Among this group of 50 patients, none had an en-
tirely normal neurologic examination at presentation.
The clinical presentations of these patients are de-
tailed in Table 1. Some patients demonstrated more
than one finding. The durations of symptoms for spe-
cific histologic diagnoses are detailed in Table 2.
The median preoperative Karnofsky score for
patients undergoing resection of mass lesions was

Table 1. Neurologic Findings

Preoperative Postoperative
Symptoms (No. of cases) (No. of cases)
Hemiparesis 47 25
Seizures 32 12
lncreased ICP 21 6
Visual field defects 19 15
Aphasia 18 11
Fig. 1. Stereotectically directed catheter localizing target Dementia 16 14
volume.
 Maciunas: Stereotactic Craniotomy 21 1

Table 2. Duration of Presenting Complaints Table 4. Calculated Lesion Volumes

Presenting complaints Duration (weeks) Volume
Metastatic tumors 15.3 Histology (mm)
Glioblastoma multifome 9.8 Glial tumors 87.667
Other glial tumors 61.1 Metastatic tumors 45,212
All glial tumors 29.8 Other 40.083
Benign tumors 6.2
Nonneoplastic lesions 10.9
All mass lesions 18.1

Transfemoral cerebral angiography was per-

70, with a range of 20 to 90 and a mean of 7 1.3. The
lowest initial Karnofsky score was held by an 8-year-
old boy, who underwent an emergent stereotactic
craniotomy for evacuation of a posttraumatic left
thalamic intracerebral hematoma. The associated
mass effect resulted in uncontrollable intracranial
hypertension and rapidly declining neurologic func-
Downloaded by [36.72.231.171] at 18:05 20 May 2016
formed in 41 patients with mass lesions, demonstrat-
ing neovascularity in 15 cases and only mass effect
in 26. Preoperatively, neuroradiologic diagnoses for
each mass lesion, based on all imaging studies avail-
able, were recorded for comparison with the actual
histologic diagnoses obtained at surgery.
Histology

tion, despite aggressive medical measures. Postopera-

tively, his neurologic status improved, and he was able In all cases, a frozen section analysis of representa-
to pursue rehabilitative therapy outside the hospital. tive tissue was performed to confirm lesion localiza-
tion. Regardless of tumor histology, once tentatively
Neurodiagnostic Imaging verified by frozen section, an effort was made to re-
Preoperative CT scans with and without contrast were sect all solid tumor tissue as visualized on micro-
obtained for all patients. CT demonstrated a contrast- scopic illumination. All therapeutic decisions regard-
enhancing mass lesion in all patients undergoing re- ing adjuvant therapies awaited permanent histologic
section. This enhancement was homogenous in 22 preparations. The ultimate histopathologic diagnoses
cases, whereas a pattern of ring enhancement was obtained are described in Table 5.
seen in 39 cases. Perilesionallow-attenuation changes
suggestive of edema were present in 55 cases. Calci- Table 5. Histopathologic
fications were present in eight lesions. - Diagnoses
-
of Mass Lesions-
Target locations for the 81 stereotactic cran- Histology
Histoloev NO.
No.
iotomies were as described in Table 3. Calculated lMetastasis
VlCulsULslS 32L
J
mean lesion volumes from contrast-enhanced CT Adenocarcinoma, lung primary I
boundaries for 62 cases are shown in Table 4. Squamous cell carcinoma I
MR imaging scans were performed in 48 pa- Adenocarcinoma, unknown primary 5
Small cell carcinoma 3
tients with mass lesions and tended to corroboratefind- Breast 2
ings from CT scans. T1-weightedpulse sequencesbest Adrenal 2
defined solid tumor tissue components. T2-weighted Renal cell carcinoma 2
images reflected peritumoral edema. T1-weighted Testicular carcinoma 1
images after gadolinium enhancement demonstrated Colon 1
Prostate 1
focal disruption of the blood-brain barrier. Melanoma 1
Glioma 18
Glioblastoma multiforme 11
Anaplastic astrocytoma 2
Table 3. Location of 62 Targets for Two- Malignant oligodendroglioma 3
Dimensional Stereotactic Craniotomy Astrocytoma (pilocytic) 2
No. Location Right Left Lymphoma 2
2 Frontal - 2 Epidermoid 1
15 Frontoparietal 5 10 Meningioma 2
13 Parietal 5 8 Radiation necrosis 2
9 Occipitoparietotemporal 4 8 Arteriovenousmalformation 1
5 Temporal - 5 Abscess 3
6 Occipital 1 5 Mixed flora bacterial 2
8 Basal ganglidinternal capsule 4 4 Nocardial 1
2 Thalamus - 2 Intracerebral hemorrhage 1
2 Cerebellum - 2
 212 Maciunas: Stereotactic Craniotomy
RESULTS
Operative Time
The incorporation of stereotactic methodology US-
ing the BRW system did not significantly alter the
duration of operative procedures. Although frame
application and stereotacticCT localization required
some additional time, the added confidenceconferred
by the localization subsequently expedited access to
the target. Nasotracheal fiber optic intubation greatly
facilitated the induction of general endotracheal an-
esthesia for surgery, while allowing data acquisition
to proceed with the patient awake.
Target Localization
Targets were consistently and precisely localized by
the BRW stereotactic frame. Positive histopathologic
Downloaded by [36.72.231.171] at 18:05 20 May 2016
diagnoses were defined in all cases. In three cases,
the surgical trajectory as computed from CT-derived
data would have transgressed a major cortical vessel
(vein of Trolard in two cases and vein of Labbe in
one). Accordingly, the cortical entry point was rede-
fined to avoid vascular injury.
At surgery, the cortical surface demonstrated
some abnormality (e.g., distortion, discoloration,
neovascularity or reddened veins, or frank tumor) in
20 cases. In 61 cases, there were no visual clues from
surface inspection to assist in target localization.
Motor-evoked potential monitoring was per-
formed to define motor cortex in three cases of
perirolandic metastases.Although electrophysiologic
data derived from direct cortical stimulation proved
consistent in localizing the motor cortex, the infor-
mation so obtained did not significantly change the
course of surgical resection in these three cases.
htraoperative Ultrasonography
Prior to opening the dura in 15 cases of mass lesion
resections, a prospective assessment was carried out
on the ability of intraoperativeultrasonographyto vi-
sualize and localize adequately the stereotactically
defined lesion when directed by the arc assembly along
the preselected trajectory. In two cases of glioblas-
toma multiforme and one case of small cell carcinoma
metastasis, the lesion could not be visualized conclu-
sively. Localization with intraoperative ultrasonogra-
phy was not possible in these three cases. At surgery,
these lesions were uniformly soft and poorly demar-
cated from surrounding edematous white matter by
either visual or tactile cues. Ultrasonic localization
proved too shallow by values in excess of 5 mm rela-
tive to the actual tumor location in five other cases. In
the remaining five cases, intraoperativeultrasonogra-
phy successfully localized the lesion to within 5 mm.
Extent of Resection
A gross total resection of all solid tumor tissue com-
ponents of the mass lesions was consistently
achieved,as judged from intraoperativevisual inspec-
tion under microscopic illumination. Postoperative
contrast-enhanced CT was performed within 96 hr
of resection to confirm adequacy of the surgical re-
section. Subsequently, CT scans with and without
contrast were performed at 6 weeks after surgery and
at 3-month intervals thereafter. In all cases, greater
than 90% of the volume of contrast-enhancing le-
sion was removed. Peripheral contrast enhancement
within the resection cavity was noted in 23 cases as
well as in all 19 sites of adrenal medullary trans-
plantation. This presumably represented a local phe-
nomena of blood-brain barrier breakdown related to
the surgical procedure.
Postoperative CT Scanning
Postoperative CT scanning demonstrated complete
absence (59 cases) or resection of more than 90% (3
cases) of the original contrast-enhancing lesion vol-
ume. Postoperative disruption of the blood-brain
barrier in the area of surgical manipulation, result-
ing in faint enhancement on contrasted CT, was seen
in 23 cases of mass lesions. This enhancement was
transient in tumor resection beds, being no longer
visible after 10days, but remained visible in all trans-
plants throughout the follow-up period.
Correlation of Imaging and Histology
The presumptive preoperative diagnoses based on
neuroradiologic criteria in 62 cases of mass lesions
were compared with postoperative histopathologic
results. In 13 cases (20.2%), the preoperative studies
failed to diagnose the lesion correctly. Six cases
(9.8%) of glial tumors were incorrectly judged re-
garding their grade. Therefore, clinically significant
differences were noted in 3 1% of cases between pre-
operative neuroradiologic impressions and ultimate
histopathologic diagnoses.
Postoperative Status
There were no postoperative deaths among 62 cases.
The postoperative neurologic examination showed
improvement in 38 patients, unchanging status in
22, and worsening in 2. One patient with a meta-
static adenocarcinoma to the internal capsule who
presented with a 4/5 left-hand monoparesis dem-
onstrated a 315 left-hand monoparesis postopera-
tively. One patient with an occipital lobe malignant
oligodendroglioma developed a superior quad-
rantanopia postoperatively. Postoperative CT scans
demonstrated only resection of the target lesion,
 Maciunas: Stereotactic Craniotomy 213

without evidence of intracranial hemorrhage. These
neurologic deteriorations presumably were due to
local trauma to perilesional tissue during resection
of the tumor.
Transiently worsened hemipareses were seen
in three patients, including two cases with basal gan-
glia metastases and one case involving a perirolandic
glioblastoma multiforme. Postoperative CT scans
demonstrated lesion resection and postoperative
edema in perilesional parenchyma. Increased dos-
ages of routine postoperative dexamethasone therapy
(10 mg every 6 hr) were instituted. In all three cases,
the patients neurologic examination returned to the
preoperative baseline within 1 week of surgery.
Tables 6 and 7 summarize the postoperative neuro-
logic results after resection of various histologic types
of lesions in different locations.
Downloaded by [36.72.231.171] at 18:05 20 May 2016
nary embolism at 3 months, and two remain alive at
6 and 24 months postoperatively. For the 17 patients
who died after resection of intracranial metastases,
the cause of death was progression of systemic dis-
ease in 11, unrelated medical conditions in 3, and
the sequelae of carcinomatous meningitis in 3.
Six patients underwent subsequentreoperation
for recurrent metastatic lesions in this series.All were
patients who had received several courses of chemo-
therapy and over 4,500 cGy external beam radiation
before their initial stereotactic resections of poorly
differentiated adenocarcinoma metastases. Time to
recurrence averaged 6 months. The median length
of survival for these patients was 56.6 weeks, with
two patients alive at 6 and 15 months postoperatively.
All patients experienced improved neurologic status
and higher Karnofsky scores postoperatively.

No patient incurred a lowered Karnofsky In this series, the median length of survival for
score postoperatively. Preoperative Karnofsky patients with glioblastoma multiforme who under-
scores had a mean value of 71 (range, 20-90). Post- went stereotactic resection and adjuvant therapy was
operatively, the average Karnofsky score was 76 58.7 weeks (mean, 54 weeks; range, 13-106 weeks).
(range, 60-100); 39 patients improved, and 23 had An average follow-up of 20.3 weeks is recorded for
unchanged scores. patients still living. Six patients with glioblastoma
The average length of hospitalization incurred multiforme received postoperative external beam
by patients for CT-directed BRW stereotactic cran- radiotherapy and chemotherapy. Five patients refused
iotomy was 6 days, with a range of 2 to 24 days. The both radiation therapy and chemotherapy; these pa-
postoperative course was routinely quite benign. The tientsmedian length of survival was only 17.4 weeks
limited soft tissue dissection involved in the short (mean, 21.8 weeks; range, 5-31 weeks). Conse-
linear scalp incisions and small cranial trephination quently, an overall median length of survival for the
appeared to heal well. entire group of patients with glioblastoma multiforme
was 33 weeks (range, 5-106 weeks). Four patients
Infections with recurrent glioblastoma after prior surgery and
There were no postoperative infections among 62 adjuvant therapy were operated on stereotactically
cases of mass lesions. in this series; their median length of survival was 70.9
weeks (mean, 83.6 weeks; range, 13-106 weeks).
Postoperative Survival Both patients with anaplastic astrocytomasand
Among 32 patients with metastatic lesions, 28 re- all three patients with malignant oligodendrogliomas
ceived postoperative adjuvant therapy. Among these
28 patients, three underwent external beam radio-
Table 6. Postoperative Neurologic Status and
therapy alone, 10 received chemotherapy only, and
Histology of Lesion
15 underwent combined radiotherapy and chemo- Postoperativeneurologic status
therapy. For all patients with metastatic tumors, the
overall median survival was 39.2 weeks (mean, 43.8 Histology Better Same Worse
weeks; range, 13.2-118.8 weeks), with an average Metastasis 18 13 1
Glioblastoma multiforme 5 6 -
follow-up period of 62.9 weeks for those patients still Anaplastic astrocytoma 1 1 -
living. At 1 year, 41% of patients with cerebral me- Malignant oligodendroglioma 1 1 1
tastases were still alive after resection. Astrocytoma 2 - -
In 16 cases, stereotactic resection was per- Lymphoma 1 1 -
Epidemoid 1 - -
formed after failure of radiation therapy and chemo- - -
Meningioma 2
therapy to control cerebral lesions. The average Radiation necrosis 2 - -
length of survival of these patients was 55.1 weeks, Arteriovenousmalformation 1 - -
with seven patients still alive after a follow-up of 62.2 Abscess 3 - -
weeks. Of the three patients undergoing resection of Intracerebral hemorrhage 1 - -
Total (n = 621 38 22 2
more than one metastatic lesion, one died of pulmo-
 214 Mariunas: Stereotactic Craniotomy

Table 7. Postoperative Neurologic Status and crosis. After resection, the patient's examination re-
Lesion Location sults improved, and she lived for an additional 52
Postoperative neurologic status weeks before succumbing to diffuse infiltrative dis-
Lesion location Better Same Worse ease. Six patients came to stereotactic resections of
Frontal 1 1 - solitary cerebral metastases of poorly differentiated
Frontoparietal 9 6 - adenocarcinoma only after external beam radiation
Parietal 9 4 - therapy and several cycles of chemotherapy had failed
Occipitoparietotemporal 7 2 - to control their intracranial lesions. After stereotac-
Temporal 5 - -
Occipital - 5 1
tic craniotomy for tumor resection, all six patients
Basal ganglidinternal capsule 6 1 1 improved neurologically and functionally. All six
Thalamus 1 1 - then experienced subsequent neurologic decline af-
Cerebellum - 2 - ter an average of 6-months, due to recurrence of their
Right side 12 6 1 tumors in regions adjacent to the site of stereotactic
Left side 26 16 1
Total 38 22 2 resection. Repeated stereotactic resections of these
recurrent mass lesions succeeded in improving the
neurologic and functional status in all six patients,
without additional morbidity.
Downloaded by [36.72.231.171] at 18:05 20 May 2016

underwent postoperative radiotherapy and chemo- DISCUSSION

therapy. Two patients with diagnoses of anaplastic
astrocytoma remain alive with follow-up periods of
26 weeks. The mean length of survival for three pa-
tients with malignant oligodendrogliomas was 97.9
weeks.
Two patients with radiation necrosis survived
an average of 24.2 weeks. None of the patients un-
dergoing stereotactic craniotomy for resection of an
histologicallybenign lesion has suffered a recurrence
or died as the result of neurologic sequelae. Two pa-
tients died of unrelated medical conditions at 5 and
6 months postoperatively.
Multiple and Recurrent Lesions
Among the craniotomies performed for mass lesions,
50 operative cases involved solitary lesions. Three
cases involved the stereotactic resection of two le-
sions at one setting. Two cases involved a
parietooccipital and cerebellar hemisphere metastatic
tumor, and one involved a parietooccipital and fron-
toparietal metastasis. In all cases, patients with no
known primary disease presented with complex neu-
rologic symptoms referable to the mass effect of tu-
mor and edema from both lesions.
Nine patients underwent two stereotactic cran-
iotomies. One patient developed a recurrent mesial
left temporal lobe abscess after resection of a more
superficial multiloculated mixed-flora bacterial ab-
scess. After the successful resection of a left angular
g y m glioblastoma multiforme with postoperative
external beam radiotherapy and chemotherapy, one
patient presented after 52 weeks with anomic apha-
sia and an enhancing lesion in the left parietal lobe
adjacent to the location of the prior solid tumor tis-
sue mass. This proved to be a focus of radiation ne-
Experience in using the BRW apparatus to direct 8 1
stereotactic craniotomies indicates that this is a safe
and efficacious as well as readily available surgical
instrument. No significantmodificationsof hardware
or software were necessary to adapt the BRW sys-
tem to this use. Sufficient flexibility was present for
procedures involving both supratentorial and deep
cerebellar hemispheric targets. In keeping with the
current computational capabilities of this system, a
vector projection method was developed to allow
definition of surgically relevant margins of volumet-
ric targets. Stereotactically positioned catheters
served as guideposts delimiting the boundaries of
resection. In all cases, intracranial targets selected
on the basis of CT images were consistently and pre-
cisely localized by the stereotactic apparatus. Ster-
eotactic localization proved more reliable than in-
traoperative visual inspection or intraoperative
ultrasonography. Furthermore, stereotactic method-
ology was adequate to guide the precise placement
of craniotomies as well as cortical incisions, thereby
minimizing dissection and operating time.
Positive histologic diagnoses were made in all
cases. Preoperative impressions based on
neurodiagnostic imaging studies were not correct in
predicting the histopathologic diagnoses in 21.O%
of cases and were incorrect in assessing tumor grade
in an additional 9.6% of cases. Thus, clinically sig-
nificant errors of diagnosis could have occurred in
30.6% of cases if the lesions had not been biopsied.
This is consistent with previously reported failure
rates for attempted radiographic prediction of histo-
logic findings at stereotactic biopsy.'
Precise localization aided in maintaining three-
dimensional orientation throughout the surgical pro-

cedure, allowing gross total resection in 59 cases and
over 90% resection in the remaining three cases, as
confirmed by postoperative contrast-enhanced CT
scans. The average volume of tumors resected was
calculated to be 55,903 mm3,somewhat greater than
that reported in previous series of stereotactically
directed resection^.^^-^^^^^*^^.
The surgical mortality for this series was 0%.
Postoperative neurologic status was improved in
38 patients, was unchanged in 22, and worsened
in two: one patient had increased left-hand mono-
paresis after resection of a metastatic adenocarci-
noma to the internal capsule, and one patient had
a superior quadrantanopsia after resection of an
occipital malignant oligodendroglioma. Tran-
siently worsened hemipareses were observed af-
ter resection of two internal capsule metastases and
Downloaded by [36.72.231.171] at 18:05 20 May 2016
one perirolandic glioblastoma. Functional status
as judged from Karnofsky scale scores showed
postoperative improvement in 39 patients with
mass lesions and was unchanged in 23 patients.
There were no other complications referable to
stereotactic methodology in this series. Thus, the
overall surgical morbidity was 3.7%.
Postoperative survival for patients with mass
lesions remains a function of the constraints imposed
by the cytobiology of each lesion.2~'7~20~26~29 Median
length of survival for metastatic tumors was 39.2
weeks, with 4 1% of patients alive at 1 year after re-
section. For the 17 patients who died after resection
of metastatic tumors, only 3 were affected by active
intracranial disease. The correlation in this series
between tumor recurrence and a history of attempted
control of solitary intracranial metastases by exter-
nal beam radiotherapy and chemotherapy might sup-
port a more surgically aggressive approach to patients
at initial presentation. Additionally, interstitial
brachytherapy targeting the surgical resection cav-
ity remains to be explored as a therapeutic adjunct in
these patients. For patients with glioblastoma
multiforme, median length of survival after stereo-
tactic resection and adjuvant therapy was 58.7 weeks.
This compares favorably with conventional surgical
results for series of patients with lesions situated in
more easily accessible and presumably more favor-
able l o ~ a t i o n s .Surgery
~ ~ * ~ ~for recurrent glioblastoma
proved efficacious, with median length of survival
afterrecurrence and reoperation of an additional 70.9
weeks. Undoubtably, multiple selection factors con-
tributed to this favorable result. Patients with benign
tumors and nonneoplastic lesions have done well
uniformly, with folIow-up periods averaging 20
months. Thus, the benefits of stereotactic localiza-
tion are most evident for the resection of histologi-
Maciunas: Stereotactic Craniotomy 215
cally circumscribed deep-seated lesions and in the
performance of periventricular operations.
CONCLUSIONS
CT-directed stereotactic craniotomy with the BRW
system is a safe, efficacious, and readily available
technique. It successfully confers the precision of
stereotactic methodology on open microneuro-
surgical procedures. As such, it can serve as a plat-
form for future developments, incorporating the
digital databases of MR imaging and digital sub-
traction angiography into more powerful computer
software programs developed for volumetric ster-
eotactic neurosurgery.
ACKNOWLEDGMENT
The author thanks A. Echerd, MACT, G.S. Allen,
MD, PhD, R.L. Galloway, PhD, A.L. Failinger, MD,
and the Vanderbilt University neurosurgical house
staff. This work was supported in part through MFEU
SDIORA grant N00014-87-C-0146.
REFERENCES
1. Allen GS, Burns RS, Tulipan NB, Parker RA (1989)
Adrenal medullary transplantation to the caudate nucleus
in Parkinson's disease. Arch Neurol46:487-491.
2. Apuzzo M U , Chandrasoma FT, Cohen D, Zee CS,
Zelman V (1987) Computed imaging stereotaxy: Ex-
perience and perspective related to 500 procedures ap-
plied to brain masses. Neurosurgery 20:93&937.
3. Apuzzo MU, Heifetz MD, Weiss MH, Kurze T (1977)
Neurosurgical endoscopy using the side-viewing tele-
scope: Technical note. J Neurosurg 46:398-400.
4. Apuzzo MLJ, Sabshin JK (1983) Computed tomo-
graphic guidance stereotaxis in the management of in-
tracranial mass lesions. Neurosurgery 12:277-285.
5. Black P (1979) Brain metastasis: Current status and
recommended guidelines for management. Neurosur-
gery 5617-631.
6. Carillo R, Sola RG, Ojellon MG, Uria JG, Bravo G
(1986) Stereotactic localization and open microsurgi-
cal approach in the treatment of some intracranial deep
arteriovenousmalformations. Surg Neurol25535-539.
7. Coffey RJ, Friedman WA: Stereotactic applications
(1988) General overview. In Heilbrun MP (ed): Ster-
eotactic Neurosurgery. Baltimore: Williams and
Wilkins, pp 17-54.
8. Galicich JH, Sundaresen N, Thaler HT (1980) Surgi-
cal treatment of single brain metastases: Evaluation of
the results of treatment by computerized tomography
scanning. J Neurosurg 53:63-67.
9. Gildenberg PL, Kaufman HH, Murthy KSK (1982)
Calculation of stereotactic coordinates from the com-
puted tomographic scan. Neurosurgery 10580.
10. Gutin PH, Phillips T, Wara WM, Leibel SA,
Hosobuchi Y, Levin VA, Weaver KA, Lamb S (1983)
Brachytherapy of recurrent malignant brain tumors
 21 6 Maciunus: Stereotactic Craniotomy
with removable high-activity iodine-125 sources. J
Neurosurg 60:61-68.
11. Hariz MI, Fodstad H (1987) Stereotactic localization
of small subcorticalbrain tumors for open surgery. Surg
Neurol28:345-350.
12. Heilbrun MP (1983) Computed tomography-guided
stereotactic systems. Clin Neurosurg 31 :564-581.
13. Heilbrun MP, Roberts TS, Apuzzo MLJ, Wells TH,
Sabshin JK (1983) Preliminary experience with
Brown-Roberts-Wells (BRW) computerized tomog-
raphy stereotaxic guidance system. J Neurosurg
59:2 17-222.
14. Hitchcock ER (1985) Open stereotacticexcisional sur-
gery. Appl Neurophysiol48:8688.
15. Hyman SA, Rogers WD, Smith DW, MaciunasRJ,Allen
GS, Berman ML (1988) Penoperative management for
transplant of autologous adrenal medulla to the brain
Downloaded by [36.72.231.171] at 18:05 20 May 2016
for parkinsonism.Anesthesiology69:6 18-622.
16. Jacques S, SheidenCH, McCann GD, Freshwater DB,
Rand R (1980) Computerized three-dimensional ster-
eotaxic removal of small central nervous system le-
sions in patients. J Neurosurg 53:816820.
17. Kelly PJ (1988) Volumetric stereotacticsurgical resec-
tion of intra-axial brain mass lesions. Mayo Clin Proc
63: 1186-1 198.
18. Kelly PJ, Alker GJ Jr: A stereotactic approach to deep
seated central nervous system neoplasms using the
carbon dioxide laser. Surg Neurol 15:331-334, 1981.
19. Kelly PJ, Kall BA, Goerss SJ (1986)Computer-assisted
stereotactic resection of posterior fossa lesions. Surg
Neurol25:530-534.
20. Kelly PJ, Kall BA, Goerss SJ (1988) Results of com-
puted tomography-based computer-assistedstereotac-
tic resection of metastatic intracranial tumors. Neuro-
surgery 22:7-17.
21. Kelly PJ, Kall BA, Goerss Sl, Earnest F (1986) Com-
puter-assisted stereotaxic resection of intra-axial brain
neoplasms. J Neurosurg 64:427439.
22. Lunsford LD (1988) Diagnosis and treatment of mass
lesions using the Leksell stereotactic system. In
Lunsford LD (ed): Modem Stereotactic Neurosurgery.
Boston: Martinus Nijhoff, pp 145-168.
23. Maciunas RJ (ed): Interactive Image-Guided Neuro-
surgery. Park Ridge, IL (1993) American Association
of Neurological Surgeons.
24. Maciunas RJ: The diagnosis and therapy of ganglion
cell tumors. In Apuzzo MLJ fed):Benign Cerebral Glio-
mas: Neurosurgical Topics, Book 22. Lebanon, NH
(1995) American Associationof NeurologicalSurgeons.
25. Maciunas RJ. Juneau PJ (1988) Limiting artifact in CT
stereotaxicperiventricularprocedures: Technical note.
J Neurosurg 69:459-460.
26. Salcman M (1980) Survival in glioblastoma: Histori-
cal perspective. Neurosurgery 7:435439.
27. Salcman M, Kaplan RS, Ducker TB, Abdo H, Mont-
gomery E (1982) Effect of age and reoperation on sur-
vival of malignant astrocytoma. Neurosurgery
10:454-463.
28. Shelden CH, McCann G, Jacques S , Lutes HR, Frazier
RE, Katz R,Kuki R (1980) Development of a comput-
erized microstereotaxic method for localization and
removal of minute CNS lesions under direct 3-D vi-
sion. J Neurosurg 52:21-27.
29. Sundaresen N, Galicich JH, Beattie EJ (1983) Treat-
ment of brain metastases from lung cancer. J Neurosurg
58:666-67 1.