ventricular volume following third ventriculostomy

7/28/2019 Ventricular Volume Following Third Ventriculostomy

http://slidepdf.com/reader/full/ventricular-volume-following-third-ventriculostomy 1/6

ECENT advances in endoscopic technology have ledto a rise in the performance of third ventriculos-tomy for noncommunicating hydrocephalus. Al-

though this technique avoids the risks of infection, mal-function, and overfunction associated with the placementof a mechanical shunt, its reliance on the reabsorptivecapacity of the arachnoid granulations often translates in-to a minimal or unappreciable change in ventricular size.This may be true even in patients who experience symp-tomatic relief following successful surgery. Various in-vasive or costly techniques have been used to evaluatethe patency of the ventriculostomy including radionu-cleotide ventriculography,3,17,22,23,28,32 radioisotope lumbarscanning,15 computerized tomography (CT) ventriculogra-phy,20 sagittal magnetic resonance (MR) imaging,20,26 andtwo-dimensional cine phase–contrast MR imaging.17,22,25,35

However, assessment of the patency of the fenestrationalone may not be relevant to determining the adequacy of treatment. A more relevant, but no less invasive, assess-ment is performed using direct intracranial pressure (ICP)monitoring.20,21,29 This technique has demonstrated thatICP may remain elevated immediately after third ventric-

ulostomy and begins to decrease between 4 to 8 days post-operatively. In an earlier retrospective study,30 we showedthat careful measuring of ventricular diameter on standardCT scans will show a consistent decrease in both thirdand lateral ventricular size if performed later than 1 monthafter successful surgery (that is, in patients with clinicalimprovement). We hypothesized that, because volumechanges in proportion to the cube of the radius, measure-ments of ventricular volume might significantly decreaseas early as the first 3 weeks after surgery; this wouldbe more useful in evaluating adequacy of treatment andwould provide an early indication of successful outcome.

Clinical Material and Methods

Sixteen consecutive patients undergoing stereotactical-ly guided endoscopic third ventriculostomy were prospec-tively entered into this study. The patient’s ages at op-eration ranged from 1 month to 71 years. Indications forsurgery, symptoms, and presence of a previously placedventriculoperitoneal shunt are presented in Table 1. One

patient (Case 8) underwent reoperation for closure of a

J Neurosurg 91:20–25, 1999

20

Ventricular volume following third ventriculostomy

THEODORE H. SCHWARTZ, M.D., BRIAN HO, M.D., CHARLES J. PRESTIGIACOMO, M.D.,JEFFREY N. BRUCE, M.D., NEIL A. FELDSTEIN, M.D., AND ROBERT R. GOODMAN, M.D., PH.D.

Department of Neurological Surgery, The Neurological Institute of New York, Columbia–Presbyterian Medical Center, New York, New York

Object. Ventricular size often shows no obvious change following third ventriculostomy, particularly in the early post-operative period, making postoperative evaluation difficult without expensive and often invasive testing in patients withequivocal clinical responses. The authors hypothesized that performing careful volumetric measurements would show de-creases in size within the first 3 weeks after surgery.

Methods. Volumetric measurements were calculated from standard 3 ϫ 3–mm axial computerized tomography (CT)

scans obtained immediately before and 3 and 21 days after surgery. Two independent investigators measured third ventric-ular volume in a series of 16 patients and lateral ventricular volume in 10 of the patients undergoing stereotactically guid-ed endoscopic third ventriculostomy for noncommunicating hydrocephalus.

Fifteen patients were symptomatically improved at the time the follow-up scan was obtained. Third ventricular volumedecreased in all patients by a mean of 35% (range 7.8–95.1%) and lateral ventricular volume decreased in all patients bya mean of 33% (range 4.5–80.3%). The degree of change correlated with the length of preoperative symptoms (pϽ 0.005).The one patient who experienced no improvement showed no decrease in third ventricular volume. In seven of 10 patients,the decrease in third ventricular volume exceeded the decrease in lateral ventricular volume. Repeated measurements indi-cated that the 95% confidence interval for the authors’ calculations varied around the mean by 2.5% for third ventricularvolume and 1.2% for lateral ventricular volume. Long-term outcome was excellent, with only one case of delayed failure.The mean follow-up duration was 12 months.

Conclusions. Volumetric measurements calculated from standard CT scans will show a demonstrable decrease in ven-tricular volume soon after successful third ventriculostomy and can be helpful in assessing patients postoperatively. Al-though the third ventricle may exhibit a greater decrease, the lateral ventricular measurements are more accurate. Patientswith more indolent symptoms show the smallest change.

KEY WORDS • endoscopy • hydrocephalus • stereotaxis • ventriculostomy

R

J. Neurosurg. / Volume 91 / July, 1999



third ventriculostomy performed 2 years earlier. All pa-tients with mass lesions underwent resective surgery per-formed either before entering the study or following thepostoperative scan.

Operative Procedure

The procedure has been outlined in detail elsewhere.8

All patients underwent CT scanning immediately beforesurgery. In 15 patients this scanning was performed forstereotactic localization. In a 1-month-old infant (Case 9)

third ventriculostomy was performed “free-hand” and theCT scan was obtained for ventricular size assessment.With their heads fixed in a CT-compatible stereotactichead holder patients were administered intravenous iohex-ol so that we could target the foramen of Monro and theinterpeduncular cistern and see the basilar artery. In somepatients, previously obtained MR images were coregis-tered with the CT scans to facilitate targeting.4,8 After gen-eral or local anesthesia had been induced in the patient, theCosman-Robert-Wells arc was used to guide a 2.8-mm-diameter rigid endoscope through a No. 12 French peel-away catheter placed into the lateral ventricle through aburr hole over the coronal suture in the midpupillary line.The scope was maneuvered stereotactically through theforamen of Monro toward the floor of the third ventri-cle by using direct visual guidance. After visualizing andconfirming the relevant anatomical structures, the floor of the third ventricle was punctured with the tip of an 0.9-mm flexible scope. The interpeduncular cistern was exam-ined to fenestrate the membrane of Liliequist. The flexiblescope was then withdrawn and replaced with a No. 3French Fogarty catheter that was inflated to enlarge thefenestration, which was accomplished under direct visual-ization with the rigid scope. In some cases the ventricu-lostomy was performed using the flexible endoscope anda No. 3 French embolectomy catheter via a 2.5-mm-outer-diameter rigid cannula. If patients previously had been de-pendent on a shunt, a temporary external ventricular drain

was placed; if ICP remained low, this remained clampeduntil postoperative Day 2, when it was removed. No pa-tient in this study required shunt placement in the imme-diate postoperative period.

Volumetric Measurements

The goal of the study was to obtain a follow-up scanwithin the first 3 weeks, but at least 3 days after surgeryand at least 3 days after the ventricular catheter was re-moved, if one was placed. In two patients scans were ob-

tained on postoperative Day 3, whereas in the other pa-tients scans were obtained between 1 and 3 weeks aftersurgery. The patients in whom scans were obtained onpostoperative Day 3 did not have a temporary ventricularcatheter. In all 16 patients 3-mm contiguous slices throughthe entire third ventricle were obtained both pre- and post-operatively. After the first six patients had been examined,the decision was made to examine the lateral ventriclesas well. In 10 patients 3-mm contiguous slices throughthe entire lateral ventricles were obtained both pre- andpostoperatively. Computerized tomography scans wereobtained just prior to the procedure by using one of two scanners (9800 high-speed scanner; General ElectricMedical Systems, Milwaukee, WI, or a Somatom Plus 4scanner; Siemens Medical Imaging, Inc., Needham, MA).A similar protocol was used to obtain postoperative scans.

All ventricular measurements were made by two in-dependent observers (T.H.S. and C.J.P.). The images werescanned into a computer (Power Macintosh 6500/250[Apple Computer, Inc., Cupertino, CA] with a ScanJetIICX/T scanner [Hewlett–Packard, Palo Alto, CA]) at600 dpi resolution. Using public domain image-proces-sing software (Image version 1.61; National Institutes of Health, Bethesda, MD) at 2 ϫ magnification, the area (a)of the lateral and third ventricles was calculated for eachaxial slice either by determining a threshold for the pixelvalue of cerebrospinal fluid (CSF) or by manual outliningof a region of interest. The contrast and window settings


Ventricular volume after third ventriculostomy

21

TABLE 1

History and outcome of 16 patients undergoing third ventriculostomy*

Age Duration Early Im- FollowCase (yrs), of Prior provement Up LateNo. Sex Indications for Surgery Symptom(s) Symptoms Shunt (Յ3 wks) (mos) Improvement

1 30, F Chiari I malformation, MM HA, ataxia 30 yrs yes yes 9 yes

2 54, M pineal tumor HA, diplopia 1 yrs no yes 22 yes3 8, F tectal mass HA, N/V, incontinence 6 mos no yes 7 yes

4 23, M cerebellar tumor HA, N/V, lethargy 1 wk no yes 9 yes5 7, F tectal mass developmental delay 6 yrs no yes 16 yes

6 5, F cerebellar tumor HA, N/V 1.5 mos no yes 20 yes7 43, M aqueductal stenosis HA, cognitive decline 15 yrs yes yes 17 yes

8 30, F pineal cyst HA, N/V, diplopia 12 yrs yes yes 8 no, VP shunt9 0.1, F Dandy–Walker syndrome macrocephaly NA no yes 17 yes, sz disorder,

microcephaly

10 36, F Chiari I malformation HA, shunt infection 9 mos yes yes 25 yes11 36, F aqueductal stenosis HA, ataxia, incontinence 1.5 yrs no yes 10 yes

12 16, F pineal tumor HA, N/V, dizziness 9 mos yes yes 8 yes13 71, M tectal met ataxia, confusion 3 wks no yes 8 yes

14 41, M posterior fossa ependymoma apnea, syncope 4 mos no no NA no15 62, F tectal mass tremor, ataxia 3 yrs no yes 7 yes16 50, F pineal tumor incontinence, ataxia 15 mos no yes 7 yes

* HA = headache; met = metastasis; MM = meningomyelocele; NA = not applicable; N/V = nausea and vomiting; sz = seizure.



were held constant. The scale bar of the scan was used asa reference for standardization and to convert from pixelsto cubic centimeters. The area of each ventricle (in squarecentimeters) on each slice on which it appeared (i) wasthen summed (⌺ai) and multiplied by the thickness of the

slice (t) to arrive at a measurement of ventricular volume(V): V = t ϫ ⌺ai.

Because 3-mm slices were obtained for this study theequation for volume (in cubic centimeters) was: V = 0.3cmϫ ⌺ai.

To examine the possibility that early decreases in ven-tricular volume might be attributable to subgaleal absorp-

tion of CSF leaking through the cortical incision, we com-pared those patients in whom follow-up scans had beenobtained within the 1st week with those in whom laterscans were obtained. To rule out the possibility that epen-dymal contrast enhancement on the preoperative scansmight alter volumetric measurements we repeatedly mea-sured, with and without addition of contrast agent, scansobtained on the same day in two control patients. Also, re-peated measurements of single scans provided us with anindication of the reliability of our measurements.

Statistical Analysis

The Wilcoxon signed-rank test, a nonparametric paired

statistic, was used to compare ventricular size before andafter third ventriculostomy. The Spearman correlation testwas used to correlate chronicity of disease with the mag-nitude of the change in ventricular volume and to evaluatethe similarity between the ventricular measurements per-formed by two separate individuals.

Results

Fifteen of the patients showed symptomatic improve-ment at the time the early follow-up scan was obtained.Among these was a 1-month-old infant (Case 9) withDandy–Walker syndrome who was asymptomatic be-fore surgery, but had marked macrocephaly and hydro-

cephalus. Her head circumference stabilized. One patient(Case 14) did not improve. This patient had long-standinghydrocephalus following the resection of a posterior fossaependymoma that had left him severely ataxic and depen-dent on a respirator for apneic episodes (Table 1).

Long-term results were also excellent. At a mean fol-low-up period of 12 months, 14 patients remained symp-tomatically improved. The patient with Dandy–Walkersyndrome (Case 9) developed a seizure disorder but herhead circumference decreased from the 95th to the fifthpercentile. Another patient (Case 8), who had undergonefenestration 2 years earlier and thus underwent reopera-tion, was initially improved but treatment failed after 4months. The patient subsequently underwent shunt place-ment despite a patent fenestration confirmed by direct vi-sualization as well as a CSF flow study. One patient (Case14) never improved after third ventriculostomy. The over-all long-term success rate was 88%.

On visual inspection, early postoperative CT scans of-ten failed to reveal a definite change in ventricular size(Fig. 1). The mean Ϯ standard deviation (SD) preopera-tive third and lateral ventricular volumes were 8 Ϯ 3.9cm3 and 184.3 Ϯ 102.4 cm3, respectively. At follow-upscanning sessions, all 15 patients with improvementshowed a decrease in third ventricular volume accordingto both observers: by 32.9 Ϯ 20.1% (range 7.8–81%) forObserver 1 and by 37.4 Ϯ 21.4% (range 9.1–95.1%) forObserver 2 (p Ͻ 0.0002; Fig. 2A). The interobserver

T. H. Schwartz, et al.

22 J. Neurosurg. / Volume 91 / July, 1999

FIG. 1. A–D: Case 15. Preoperative (A and C) and postopera-tive Day 12 (B and D) CT scans revealing no obvious change inventricular size despite clinical improvement. Volumetric mea-

surements showed an 8% decrease in third ventricular volume (Acompared with B) and a 10% decrease in lateral ventricular volume(C compared with D). E and F: Case 2. Preoperative (E) andpostoperative Day 9 (F) CT scans revealing no obvious change inventricular size despite clinical improvement. Volumetric mea-surements showed a 31% decrease in third ventricular volume (Ecompared with F).



Spearman’s correlation was r = 0.9207 (p Ͻ 0.0005). Themean postoperative third ventricular volume was 5.3 Ϯ3.1 cm3. Lateral ventricular volume also decreased in the10 patients in whom it was measured according to bothobservers: by 32.3Ϯ 23.6% (range 4.5–75.3%) for Obser-ver 1 and by 35.7 Ϯ 27.1% (range 5.4–80.3%) for Obser-ver 2 (p Ͻ 0.008; Fig. 2B). The interobserver Spearman’scorrelation was r = 0.9398 (p Ͻ 0.0005). The mean post-operative lateral ventricular volume was 134.5 Ϯ 110.2cm3. Combining the two ventricles provided the most reli-able measurements: a decrease of 33.4 Ϯ 22.5% (range4.8–74.7%) according to Observer 1 and 35.9 Ϯ 27%(range 5–81.3%) according to Observer 2. The inter-observer Spearman’s correlation was r = 0.9353 (p Ͻ

0.0005). The one patient (Case 14) in whom there was noimprovement showed no change in third ventricular in-volvement according to Observer 1 and a 5.8% increaseaccording to Observer 2. In seven of 10 cases the decreasein third ventricular volume exceeded the decrease in later-al ventricular volume. In three of 10 cases (Cases 11, 13,and 15), the decrease in lateral ventricular volume exceed-ed the decrease in third ventricular volume. The interob-server agreement was 100%.

Patients in whom the smallest decreases in ventricularsize (≤ 10%) were observed had milder symptoms or in-dolent pathological entities. Two patients (Cases 5 and 15)had mass lesions that had caused symptoms for longerthan 3 years, but had not required a shunt. Two other pa-tients (Cases 1 and 7) had chronic pathological conditions(Chiari I malformation, aqueductal stenosis) that had re-quired them to rely on shunts since childhood. For thewhole group, the chronicity of the disease correlated in-versely with the magnitude of the decrease in ventricularsize (Fig. 3; Spearman’s test p Ͻ 0.002).

Ventricular volume was measured within the 1st week after surgery in four patients. The mean change (Ϯ SD) inthird ventricular volume in these patients was 31.2 Ϯ22.1% according to Observer 1 and 40.1 Ϯ 23.6% accord-ing to Observer 2; this was not significantly different fromthe rest of the group, which later was measured at 34.6 Ϯ15% by Observer 1 and 35.5 Ϯ 13.9% by Observer 2. Themean change (Ϯ SD) in lateral ventricular volume for the

early group was 29.7 Ϯ 23.3% according to Observer 1and 32.6 Ϯ 24.5% according to Observer 2, comparedwith 31.3 Ϯ 24.8% (Observer 1) and 37.2 Ϯ 30.4% (Ob-server 2), which was also not significant. In two controlpatients the third and lateral ventricular volume measure-ments, obtained both with and without contrast enhance-ment, showed no significant or consistent change in vol-ume as measured by both observers (p = 0.91).

Repeated measurements (15 iterations) of third and lat-eral ventricular volumes in the same patient indicated thatthe 95% confidence interval (CI) for the third ventricularmeasurements varied by 2.5% around the mean. The 95%CI for the lateral ventricular measurements varied by1.2% around the mean.

Discussion

Radiographic follow-up examination after third ventric-



23

FIG. 2. Bar graphs showing the mean pre- and postoperative third (A) and lateral (B) ventricular volumes for 15 symp-tomatically improved patients as measured by two independent observers. Bars indicate the standard error of the mean.

FIG. 3. Graph showing the magnitude of decrease in ventricularvolume following third ventriculostomy depending on the chronic-

ity of the disease and the length of symptoms.



ular fenestration is a challenge because the ventricles of-ten remain enlarged despite clinical improvement.6,13–15,18,24,

25,32,33 This phenomenon has been attributed to the slow,progressive opening of the subarachnoid space, whichhad been collapsed by long-standing increased ICP,9,14 aswell as to an element of communicating hydrocephaluspresent in many cases of chronic aqueductal obstruction.32

In support of these explanations, we found that patientswho had more chronic symptoms or pathological condi-tions since childhood had minimum decreases in theirventricular volume following “successful” (that is, clin-ically improved) fenestration. Nevertheless, our carefulmeasurements revealed a demonstrable change that waslarger than the variance of the measuring technique in allsuccessful cases.

This study was designed to examine whether clinicallyrelevant ventricular volume changes could be detected inthe early postoperative period. Although there may be atransient leakage of CSF through the cortical track into thesubgaleal space, we do not think this accounts for our

findings for two reasons. First, direct ICP measurementsoften show that ICP often remains elevated initially fol-lowing third ventriculostomy.19–21,29 Second, patients inwhom the ventricles were measured within the 1st week after surgery did not have a more significant decrease involume than those in whom the ventricles were measuredin the 2nd and 3rd week. Finally, we purposefully waitedat least 3 days to perform our measurements to reduce thepossibility of false-positive measurements. Another possi-ble variable in ventricular size was the fact that contrastagent was administered only during preoperative scan-ning. However, measurements on scans with and withoutcontrast enhancement that were obtained in control indi-viduals showed that this was not an influential factor.

Although the intraobserver reliability for this techniquewas excellent and the smallest decrease in ventricular vol-ume exceeded the 95% CI for both third and lateral ven-tricles, the 95% CI for the third ventricular measurementswas twice as large as that for the lateral ventricular mea-surements. Because the overall volume of the third ven-tricle is small with respect to the fixed distance betweenaxial sections, those measurements may not be optimal toassess very small changes in ventricular volume. Inter-observer agreement was also better for lateral ventricularmeasurements than for third ventricular measurements.Hence, we recommend using either the entire ventricularsystem or only the lateral ventricles to assess changes inventricular size to avoid error. In our earlier study,30 we

reported that measured changes in ventricular size weremore notable for the third ventricle than for the lateralventricles. In that study, however, we used a linear meth-od to determine ventricular size that is less sensitive tochange than volume measurements and the reliability of our measurements was never assessed.

Earlier reports of ventricular size after third ventric-ulostomy have not found a consistent decrease. Sainte-Rose29 compared 30 patients after third ventriculostomywith 38 control patients who had shunts and found a 60%rate of persistent ventricular enlargement in the third ven-triculostomy group. In a series composed of 84 patients,Dalrymple and Kelly5 reported that “a decrease in ventric-ular size on CT or MR imaging has not shown itself to be

an accurate indicator of fistula patency—19% of patientsin this series have shown no decrease in ventricular sizeon follow-up scans despite resolution of symptoms anddocumented patency on ancillary studies.” Musolino, etal.,24 presented a series of 23 patients in whom postfenes-tration CT scans had been obtained; in seven (30%) therewas no decrease in ventricular size. Jaksche and Loew18

commented that the ventricles were “unchanged or slight-ly decreased” in a series of 79 patients who underwentventriculostomy and, in a prospective study of ventricularsize using MR imaging, Wilcock and colleagues34 report-ed no decrease in ventricular size in five (38%) of 13 pa-tients despite clinical improvement. Some other studieshave reported more consistent decreases in ventricularsize, but do not specify either how or when these mea-surements were made.14,22,31

We chose to assess changes in ventricular size by usingstandard CT-based volumetric measurements. Subtle butclinically significant changes in ventricular size are diffi-cult to estimate visually because of the irregular shape of

the ventricles and the fact that comparisons are often madebetween scans obtained using different scanners with dif-ferent scales.2 Linear-based methods have been used inthe past, such as Evans’ ratio (right and left anterior hornwidthϬmaximum internal skull width),7,10–12 which false-ly assumes that all ventricles have the same shape and aremuch less sensitive to changes in size because volume-based measurements change in proportion to the cube of the radius. Although there are a variety of segmentationschemes available to measure ventricular volume withMR imaging,16 we relied on CT scans as a less expen-sive, rapid radiographic method of following patients. Ourmethod could be described as a semiautomated segmenta-tion scheme involving the manual outlining of regions of

interest. The computer then automatically counts the pix-els within the region of interest and multiplies them by aknown constant to provide an area that is integrated overthe width of the scan. This technique is similar, althoughmore sensitive, than the stereological method based onCavalieri’s principle, which states that the volume of anyobject can be estimated from a set of two-dimensionalslices through the object, provided that they are parallel,separated by a known distance, and begin randomly with-in the object.1,2,27 Any preoperative scan can be used forbaseline measurements. We used scans obtained for CT-guided stereotactic localization to optimize the trajectoryto avoid trauma to structures surrounding the foramen of Monro and to facilitate perforation of the third ventricular

floor away from the basilar artery. Thus, ventriculostomycould be accomplished even when the thickened floor wasopaque.

Conclusions

In this prospective study we show that both third andlateral ventricular volumes decrease by anywhere from 5to 80% (average 30%) within the first 3 weeks after suc-cessful third ventriculostomy. However, lateral ventricularvolume measurements are more accurate, probably due tothe small size of the third ventricle. These changes can bemeasured as early as 3 days after surgery by using a stan-dard CT scanner and the magnitude of the change corre-

T. H. Schwartz, et al.

24 J. Neurosurg. / Volume 91 / July, 1999



lates inversely with the chronicity of the disease. Early de-creases in ventricular size do not, however, predict long-term patency and, although we had an 88% success rate,the prospect of delayed closure remains a possibility in asmall percentage of patients.

References

1. Cavalieri B: Geometria Degli Indivisibli. Turin: Unione Tipo-graphico-Edetrice, 1966

2. Clatterbuck RE, Sipos EP: The efficient calculation of neuro-surgically relevant volumes from computed tomographic scansusing Cavalieri’s direct estimator. Neurosurgery 40:339–343,1997

3. Cohen AR: Endoscopic ventricular surgery. Pediatr Neuro-surg 19:127–134, 1993

4. Cohen DS, Lustgarten JH, Miller E, et al: Effects of coregistra-tion of MR to CT images on MR stereotactic accuracy. J Neu-rosurg 82:772–779, 1995

5. Dalrymple SJ, Kelly PJ: Computer-assisted stereotactic thirdventriculostomy in the management of noncommunicating hy-drocephalus. Stereotact Funct Neurosurg 59:105–110, 1992

6. Drake JM: Ventriculostomy for treatment of hydrocephalus.Neurosurg Clin North Am 4:657–666, 1993

7. Evans WA Jr: An encephalographic ratio for estimating ven-tricular enlargement and cerebral atrophy. Arch Neurol Psy-chiatry 47:931–937, 1942

8. Goodman RR: Magnetic resonance imaging-directed stereo-tactic endoscopic third ventriculostomy. Neurosurgery 32:1043–1047, 1993

9. Guiot G: Ventriculo-cisternostomy for stenosis of the aqueductof sylvius. Puncture of the floor of the third ventricle with a leu-cotome under television control. Acta Neurochir 28:275–289,1973

10. Gyldensted C: Measurements of the normal ventricular systemand hemispheric sulci of 100 adults with computed tomogra-phy. Neuroradiology 14:183–192, 1977

11. Hahn FJY, Rim K: Frontal ventricular dimensions on normalcomputed tomography. AJR 126:593–596, 197612. Haug G: Age and sex dependence of the sex of normal ventri-

cles on computed tomography. Neuroradiology 14:201–204,1977

13. Hirsch JF: Percutaneous ventriculocisternostomies in non-com-municating hydrocephalus. Monogr Neurol Sci 8:170–178,1982

14. Hirsch JF, Hirsch E, Sainte Rose C, et al: Stenosis of the aque-duct of sylvius. Etiology and treatment. J Neurosurg Sci 30:29–39, 1986

15. Hoffman HJ, Harwood-Nash D, Gilday DL: Percutaneous thirdventriculostomy in the management of noncommunicating hy-drocephalus. Neurosurgery 7:313–321, 1980

16. Jack CR Jr: Brain and cerebrospinal fluid volume: measurementwith MR imaging. Radiology 178:22–24, 1991

17. Jack CR Jr, Kelly PJ: Stereotactic third ventriculostomy: assess-ment of patency with MR imaging. AJNR 10:515–522, 1989

18. Jaksche H, Loew F: Burr hole third ventriculo-cisternostomy.An unpopular but effective procedure for treatment of certainforms of occlusive hydrocephalus. Acta Neurochir 79:48–51,1986

19. Jones RF, Kwok BC, Stening WA, et al: The current status of

endoscopic third ventriculostomy in the management of non-communicating hydrocephalus. Minim Invasive Neurosurg37:28–36, 1994

20. Jones RFC, Kwok BCT, Stening WA, et al: Neuroendoscopicthird ventriculostomy. A practical alternative to extracranialshunts in non-communicating hydrocephalus. Acta NeurochirSuppl 61:79–83, 1994

21. Jones RFC, Stening WA, Brydon M: Endoscopic third ventric-ulostomy. Neurosurgery 26:86–92, 1990

22. Kelly PJ: Stereotactic third ventriculostomy in patients withnontumoral adolescent/adult onset aqueductal stenosis andsymptomatic hydrocephalus. J Neurosurg 75:865–873, 1991

23. Kelly PJ, Goerss S, Kall BA, et al: Computed tomography-based stereotactic third ventriculostomy: technical note. Neuro-surgery 18:791–794, 1986

24. Musolino A, Soria V, Munari C, et al: La ventriculocisternos-tomie stéréotaxique dans le traitement de l’hydrocéphalie ob-structive triventriculaire. A propos de 23 cas. Neurochirurgie34:361–373, 1988

25. Oka K, Go Y, Kin Y, et al: An observation of the third ventri-cle under flexible fiberoptic ventriculoscope: normal structure.Surg Neurol 40:273–277, 1993

26. Oka K, Yamamoto M, Ikeda K, et al: Flexible endoneuro-

surgical therapy for aqueductal stenosis. Neurosurgery 33:236–243, 1993

27. Pakkenberg B, Boesen J, Albeck M, et al: Unbiased and effi-cient estimation of total ventricular volume of the brain ob-tained from CT-scans by a stereological method. Neuroradi-ology 31:413–417, 1989

28. Pierre-Kahn A, Renier D, Bombois B, et al: Place de la ven-triculo-cisternostomie dans la traitement des hydrocéphaliesnoncommunicantes. Neurochirurgie 21:557–569, 1975

29. Sainte-Rose C: Third ventriculostomy, in Manwaring KH,Crone KR (ed): Neuroendoscopy. New York: Mary Anne Lie-bert, 1992, pp 47–62

30. Schwartz TH, Yoon SS, Cutruzzola FW, et al: Third ventricu-lostomy: post-operative ventricular size and outcome. MinimInvasive Neurosurg 39:122–129, 1996

31. Teo C, Jones R: Management of hydrocephalus by endoscopicthird ventriculostomy in patients with myelomeningocele. Pe-diatr Neurosurg 25:57–63, 1996

32. Vries JK, Friedman WA: Postoperative evaluation of thirdventriculostomy patients using 111In-DTPA. Childs Brain 6:200–205, 1980

33. Wilcock DJ, Jaspan T, Punt J: CSF flow through third ventricu-lostomy demonstrated with colour doppler ultrasonography.Clin Radiol 51:127–129, 1996

34. Wilcock DJ, Jaspan T, Worthington BS, et al: Neuro-endoscop-ic third ventriculostomy: evaluation with magnetic resonanceimaging. Clin Radiol 52:50–54, 1997

35. Yamamoto M, Oka K, Ikeda K, et al: Percutaneous flexible neu-roendoscopic ventriculostomy in patients with shunt malfunc-tion as an alternative procedure to shunt revision. Surg Neurol42:218–223, 1994

Manuscript received June 25, 1998.Accepted in final form February 11, 1999.

Address reprint requests to: Robert R. Goodman, M.D., Ph.D.,Department of Neurological Surgery, The Neurological InstituteNI-99, 710 West 168th Street, New York, New York 10032.



25

ventricular volume following third ventriculostomy

Documents