You are on page 1of 9

Childs Nerv Syst DOI 10.

1007/s00381-011-1475-0

ORIGINAL PAPER

Initial experience with combined endoscopic third ventriculostomy and choroid plexus cauterization for post-hemorrhagic hydrocephalus of prematurity: the importance of prepontine cistern status and the predictive value of FIESTA MRI imaging
Benjamin C. Warf & Jeffrey W. Campbell & Eric Riddle

Received: 12 June 2010 / Accepted: 28 April 2011 # Springer-Verlag 2011

Abstract Purpose Post-hemorrhagic hydrocephalus of prematurity (PHHP) is among the most common causes of infant hydrocephalus in developed nations. This population has a high incidence of shunt failure, infection, and slit ventricle syndrome. Although effective for other etiologies of infant hydrocephalus, the efficacy of combined endoscopic third ventriculostomy and choroid plexus cauterization (ETV/ CPC) in PHHP has not been investigated. This pilot study reports the initial experience. Methods Ten patients (four grade III and six grade IV intraventricular hemorrhage) requiring definitive treatment for PHHP underwent ETV/CPC within 6 months of birth. Seven had a prior ventriculo-subgaleal shunt. Mean age at birth was 12.8 weeks, or 25.2 weeks gestation (24 28 weeks), and at surgery was 1.6 weeks (11 to +11 weeks). Mean weight at surgery was 3.3 (1.0 5.5 kg). Each patient had preoperative magnetic resonance imaging (MRI) with fast imaging employing steady-state acquisition (FIESTA). Results Four of ten (40%) required no further operations related to hydrocephalus (mean follow-up, 29.7 months). Six required another procedure (five ultimately shunted). Prepontine cistern status correlated with outcome (p=0.033).
B. C. Warf (*) Department of Neurosurgery, Childrens Hospital Boston, 300 Longwood Avenue, Boston, MA 02115, USA e-mail: benjamin.warf@childrens.harvard.edu J. W. Campbell : E. Riddle Department of Neurosurgery, Alfred I. duPont Hospital for Children, Wilmington, DE, USA

Procedures in all infants with unobstructed cisterns were successful but failed in six of seven with cisternal obstruction, with the one success having an alternative lamina terminalis endoscopic third ventriculostomy. Preoperative MRI FIESTA images correlated well with intraoperative assessment of the cistern. Conclusions Results from this small homogenous cohort suggest cistern status is an important determinant of outcome. FIESTA imaging correlated with endoscopic observation. Preliminary analysis suggests ETV/CPC as an effective treatment for PHHP, but only when the cistern is unscarred. This information should guide patient selection for future study protocols. Keywords Endoscopic third ventriculostomy . Choroid plexus cauterization . Hydrocephalus . Post-hemorrhagic hydrocephalus of prematurity . FIESTA imaging . Outcome

Introduction Post-hemorrhagic hydrocephalus of prematurity (PHHP) is one of the most common causes of infant hydrocephalus in developed nations, but, it is rarely seen in developing countries because prematurely born infants with very low birth weight (VLBW) generally do not survive. A National Institutes of Health consensus conference held in 2005 concluded that the incidence and prevalence of hydrocephalus is not well known [1]. However, the incidence of PHHP can be estimated. Premature birth with VLBW (<1,500 g) occurs in about 1.4% of live births in the USA [2]. There is a 5 10% incidence of grade III or IV intraventricular hemorrhage (IVH) in VLBW infants, with hydrocephalus reported to

Childs Nerv Syst

develop in around 30% [3, 4]. In a recent multi-center North American study, 40% of infants with grade III or IV IVH of prematurity underwent surgical interventions for hydrocephalus, with 86% undergoing an initial temporizing measure (external ventricular drain, tapping reservoir, or ventriculosubgaleal shunt) and the remaining 14% having permanent ventriculo-peritoneal shunt (VPS) placement as the initial treatment [5]. About 75% of those with temporizing measures were converted to permanent shunts, thus 90% of those undergoing intervention were permanently shunted. For the US population of 307,000,000 (birth rate, 14/1000 population/year) these data (1.4% incidence of VLBW, 5 10% incidence of severe IVH of prematurity and 3040% of those with severe IVH requiring surgery for PHHP) would suggest between 1,000 and 2,000 infants requiring definitive operative treatment for PHHP in the US each year. PHHP arises when blood in the cerebrospinal fluid (CSF) spaces and the subsequent inflammatory response to it inhibit normal CSF circulation and absorption. The most common mechanism of PHHP is thought to be obstructive arachnoiditis in the posterior fossa, but aqueductal obstruction from clot and subsequent ependymal inflammatory reaction is also known to occur. In addition to these, a more diffuse inflammatory reaction affecting the basal cisterns and cortical subarachnoid spaces may also co-exist or at times be the primary etiology [6]. Placement of a VPS is the current standard of treatment for PHHP. In a large multi-center North American study, nearly half of newly placed shunts in children failed within 2 years [7]. However, the complications of shunt dependence may be more common in the PHHP population, with an average of four to five shunt revisions per patient reported in two studies and reported infection rates ranging from 13% to 50% [8, 9]. In one study of developmental outcome in this population, requirement of a ventriculo-peritoneal (VP) shunt was a significant negative predictor of outcome, and a history of shunt infection or multiple shunt revisions was particularly bad in this regard [10]. Another study has demonstrated the number of shunt revisions to be a significant predictor of mortality [11]. Thus, an effective alternative to shunt dependence would potentially be very beneficial for these children. In very small series of patients, it has been variously reported that endoscopic third ventriculostomy in this population fails in 5070% [8, 12] or 90100% [13, 14] of cases. However, in one of these studies, although endoscopic third ventriculostomy (ETV) had failed as a primary procedure, it was later successful in all patients with PHHP presenting with shunt failure [14]. We have previously shown, in a series of 550 patients, that adding choroid plexus cauterization (CPC) to ETV as the primary treatment for hydrocephalus in infants less than 1 year of age significantly increased the rate of success

overall from 47% to 66% [15]. This African study did not include infants with PHHP, but, it did include a large number of infants with post-infectious hydrocephalus secondary to neonatal ventriculitis. There are striking similarities between PHHP and post-infectious hydrocephalus (PIH), and experience with 113 infants less than 1 year of age with PIH showed that combined ETV/CPC was successful in 65% of those with aqueductal obstruction and 57% of those with open aqueducts [15]. These data were not further stratified according to whether the prepontine cistern was scarred, but we subsequently demonstrated a scarred cistern to be a strong independent predictor of failure [16]. There was, then, reason to believe that treatment with combined ETV/CPC might avoid shunt dependence for many infants with PHHP as well, especially among those in whom the prepontine cistern was unaffected. For etiologies other than PHHP, we have shown that age at treatment, extent of choroid plexus cauterization, and etiology of hydrocephalus are independent predictors of ETV outcome [17]. We wanted to assess the efficacy of ETV/CPC among infants requiring definitive treatment for PHHP and to explore how patient selection could be optimized. Here, we provide the first report of results for ETV/CPC in a small but very homogenous group of infants with PHHP and suggest guidelines for patient selection in future treatment protocols.

Methods Data collection and clinical evaluation This study cohort is restricted to prematurely born infants who developed PHHP that required definitive treatment after temporizing measures had failed and in whom ETV/ CPC was performed at duPont Hospital for Children within 6 months of birth. Infants with IVH of prematurity were followed for the development of progressive ventriculomegaly by monitoring serial cranial ultrasounds and head circumference measurements. In the majority that developed hydrocephalus, serial lumbar punctures had been initially performed, and this was often followed by placement of a ventriculo-subgaleal shunt for temporary management in those cases where lumbar puncture ceased to be effective. Definitive treatment for hydrocephalus was considered only for cases of persistent, progressive symptomatic hydrocephalus despite these measures. The criteria for treatment included: increasing ventriculomegaly on serial imaging studies, abnormally accelerated head growth, and full fontanel or other evidence of symptomatically elevated intracranial pressure such as apnea, bradycardia, vomiting, irritability, lethargy, or sunsetting of the eyes. Clinically asymptomatic infants with stable ventriculomegaly received no further treatment. With appropriate institutional review board ap-

Childs Nerv Syst

proval, the ETV/CPC procedure was offered as a possible alternative to primary shunt placement, with the understanding that another operation for shunt placement might ultimately be required. Among those opting for endoscopic management, operative reports, clinical follow-up records, and imaging studies were retrospectively reviewed. Full-term infants with IVH and infants operated upon after 6 months of age were excluded from this analysis. All infants had preoperative magnetic resonance imaging (MRI) of the brain that included sagittal fast imaging employing steady-state acquisition (FIESTA) views. Operative technique The operative technique for ETV/CPC was identical to that previously reported [15]. A single approach was used via a peri-coronal scalp flap centered on the mid-pupillary line and the right lateral corner of the anterior fontanel. After sharply incising the dura, the cortical surface was cauterized and penetrated, followed by introduction of a Karl Storz flexible steerable neuro-endoscope (11282 BN) into the right lateral ventricle. If a ventriculo-subgaleal shunt was present, this was removed and the endoscope passed down the same tract into the ventricle. Typical findings included ependymal hemosiderin staining or old intraventricular blood clot, aqueductal obstruction, and scarring of the third ventricular floor. The Bugby wire was used to penetrate the floor without the use of electrocautery, and further dissection below the floor to fenestrate the Liliequist membrane or webs of arachnoid adhesions was often required. In five cases, the lamina terminalis was also fenestrated, as originally described for infants with PIH [18], because of the presence of a scarred prepontine cistern. The third ventriculostomy was widened by using the flexible endoscope to stretch its margins, or by use of a balloon compression catheter. The choroid plexus was then cauterized bilaterally from the foramen of Monro into the temporal horn using the Bugby wire. In cases where the septum pellucidum was intact, a septostomy was performed in order to enter the contralateral lateral ventricle. Any areas of loculation from intraventricular septations were opened as well to allow free communication of CSF throughout the lateral and third ventricles. Upon withdrawal of the endoscope, the durotomy was closed primarily with a running 40 vicryl suture to prevent CSF leak. Clinical follow-up The infants were followed postoperatively with serial head imaging and head circumference measurements as well as monitoring for fontanel tension or other clinical signs or symptoms of untreated hydrocephalus. Treatment failure was defined as the need for another operation for hydrocephalus, as determined by the same criteria used

for initial treatment: persistent abnormally accelerated rate of head growth, progressively increasing ventriculomegaly, or symptoms of elevated intracranial pressure (e.g., full fontanel, sunsetting of the eyes, vomiting, irritability, or lethargy). Statistical methods Two-tailed probability values, as calculated using the Fisher exact test, were used to assess the significance of outcome differences between groups. A probability value less than 0.05 was considered significant.

Results Ten consecutive patients with PHHP were treated by ETV/ CPC within the first 6 months after birth. Seven of ten had a prior ventriculo-subgaleal shunt, two of which had been placed at other facilities prior to transfer. In five patients, we placed the ventriculo-subgaleal (VSG) shunt at a mean of 3.5 weeks after birth (median, 3 weeks) and performed the ETV/CPC at a mean of 2.7 months (median, 2.6 months) after VSG shunt placement. Hydrocephalus in the other three infants had been initially managed by lumbar punctures alone. IVH grade, age and weight at time of surgery, extent of CPC, site of ETV, and outcome are shown in Table 1. All patients were born between 24 and 28 weeks gestation (mean, 25.2 weeks). Four patients had a grade III hemorrhage while six had grade IV hemorrhages. ETV/CPC was performed between the corrected ages of 11 and + 11 weeks (with 0 being full gestational age). Mean corrected age at treatment was 1.6 weeks. The mean and median ages (time period after birth) at time of ETV/ CPC were 11.2 and 11.5 weeks, respectively. The mean weight at time of treatment was 3.3 (range, 1.05.5 kg). Overall, four of ten patients did not require an additional procedure for hydrocephalus, and five of ten remained shunt-free. The mean follow-up for the successfully treated patients was 29.7 (1637.25 months). The mean time to reoperation for failure was 2.5 (0.75 to 5 months). Figure 1 illustrates pre- and postoperative MRI images. Table 2 demonstrates that, of the six failures, five had a scarred cistern reported at the time of operation. The one failed patient with an unscarred cistern (patient no. 5) had obstruction of the cistern from blood that could not be completely cleared (see below). Four of the six failures underwent fenestration of the lamina terminalis in addition to that on the third ventricular floor because of the presence of scarring in the prepontine cistern. Of the four successes, only patient no. 1 had a scarred prepontine cistern. This patient underwent fenestration of the lamina terminalis into a clean, unscarred subarachnoid

Childs Nerv Syst Table 1 Patient and operative data and clinical outcome Patient Birth gestational age (weeks) 26 25 24 25 24 26 24 28 25 25 ETV/CPC postpartum age (weeks) 6 24 12 15 3 11 8 12 9 12 ETV/CPC corrected age (weeks) 32 49 36 40 27 37 32 40 34 37 (6) (+11) (2) (+2) (11) (1) (6) (+2) (4) (1) ETV/CPC weight (kg) 1.46 5.5 2.68 4.88 1.0 2.71 2.31 4.74 2.8 4.5 IVH grade IV III III IV IV IV IV IV III III CPC extent Bilateral Bilateral Bilateral Bilateral unilateral Bilateral Bilateral Bilateral Bilateral Bilateral ETV site Outcome

1 2 3 4 5 6 7 8 9 10

Floor+LT Floor Floor Floor Floor Floor+LT Floor Floor+LT Floor+LT Floor+LT

Success Success Success Success Failure Failure Failure Failure Failure Failure

Fig. 1 Preoperative and 16 months postoperative MRI for patient no. 4, who had suffered a grade IV IVH with large right frontal intracerebral hemorrhage. a Preop axial MRI; b Postop axial MRI; c Preop sagittal MRI; and, d Postop sagittal MRI (note flow artifact through patent ETV stoma and decrease in fourth ventricle)

Childs Nerv Syst Table 2 Cistern and aqueduct status and clinical outcome Patient Cistern MRI Scarred Open Open Open Open Scarred Obscure Narrow Scarred Scarred Cistern Ventriculoscopy Scarred Open Open Open Open Scarred Scarred Scarred Scarred Scarred Aqueduct MRI Open Open Closed Open Closed Open Obscure Closed Open Open Aqueduct Ventriculoscopy Not noted Not noted Closed Not noted Closed Open Closed Closed Open Not noted Outcome

1 2 3 4 5 6 7 8 9 10

Success Success Success Success Failure Failure Failure Failure Failure Failure

space. All other successful patients were recorded as having open prepontine cisterns that were uninvolved by arachnoid adhesions. Patient no. 2 had a scarred interpeduncular cistern that was dissected through into an open, unscarred prepontine cistern. The presence of an obstructed cistern correlated with ETV/CPC failure whether patient no. 5 was excluded from analysis (p=0.048) or included among patients with an obstructed cistern (p=0.033). The aqueduct was open in six patients and closed in four as suggested by preoperative MRI, but, in four patients, the operative note did not record aqueduct status to provide endoscopic confirmation. There was no apparent correlation between the status of the aqueduct and outcome in this small group of patients. All patients, with the exception of no. 5, had a completed bilateral CPC. All patients, except no. 2, were operated upon at less than or equal to 2 weeks corrected age (e.g., 2 weeks past date of full-term pregnancy). We observed that preoperative MRI imaging with FIESTA sequences strongly suggested the status of the cistern as observed at the time of surgery. Figure 2 shows the preoperative mid-sagittal FIESTA MRI image of every patient in whom the prepontine cistern did not appear to be obstructed by thickened arachnoid membranes at the time of direct endoscopic inspection following completion of the third ventriculostomy. These images demonstrate a distinct prepontine CSF space. This should be compared with Fig. 3, which shows the same preoperative image for every patient in whom the prepontine cistern appeared involved with arachnoid scarring at the time of surgery. In all patients found to have an open cistern at the time of the operation, the preoperative MRI sagittal FIESTA images were consistent with an open prepontine cistern (Fig. 2). In patient no. 2 (Fig. 2a), the preoperative MRI reflected the ventriculoscopic finding of a scarred interpeduncular cistern that was successfully dissected through into an open prepontine cistern. Among the six found to have a scarred prepontine cistern intraoperatively, only one (patient no. 8)

had what appeared to be a distinct prepontine CSF space on the imaging study (Fig. 3d). All patients with an obscured cistern on the preoperative study were found to have cisternal scarring at the time of the operation. In patient no. 5, the smallest and youngest of the series, bleeding during the plexus cauterization in the right lateral ventricle obscured the field so as to prevent completion of the CPC on the contralateral side. Although this was controlled with irrigation and there were no clinical sequelae, thrombus was noted obstructing the prepontine cistern, having migrated through the ETV stoma. Irrigation failed to satisfactorily clear the cistern. Because of this and the incomplete CPC, a ventricular catheter and subgaleal Rickham reservoir were placed, given the likelihood of failure. This patient did fail, but subsequently underwent a successful repeat endoscopy for reopening of ETV and completion of CPC, with an ultimately successful outcome, even though he is counted as a failure for the purposes of this study. No other patients experienced complications, and there were no deaths.

Discussion ETV provides the attractive option of treating hydrocephalus in a permanent way without creating shunt dependence. ETV is a minimally invasive procedure in which an endoscope is used to create a direct communication between the third ventricle and the extra-cerebral subarachnoid spaces by fenestrating the third ventricle floor or the lamina terminalis. Although half of VP shunts can be expected to fail in the first 2 years [7] and around 80% in the first 12 years [19], the long-term failure rate for patients whose treatment by ETV is successful beyond 6 months appears to be very low [1518, 20, 21], with the majority of failures becoming evident prior to that time. This is an important consideration in determining whether ETV or shunt placement should be the primary procedure for

Childs Nerv Syst Fig. 2 Mid-sagittal FIESTA MRI of patients with open cisterns noted at ventriculostomy. a Patient no. 2. b Patient no. 3. c Patient no. 4. d Patient no. 5. Note evidence of aqueduct obstruction in a, b, and d, and of fourth ventricle outlet obstruction in b and c

infants with hydrocephalus. The initial weeks and months following treatment are typically the time of closest clinical surveillance. Also, for infants, this is the safest time to fail, since recurrence of rapid head growth and a full anterior fontanel virtually always precede any lifethreatening effects of elevated intracranial pressure. The flattening of the ETV survival curve around 6 months contrasts with the pattern for shunt failure, which is one of continued failure over time [2224]. The pattern of failure in the series of infants with PHHP reported here was consistent with this. The mean time to failure was 2.5 months, and the longest time between ETV/CPC and reoperation for failure was 5 months. Thus, for a population of infants with known parameters predicting a chance of ETV failure that equalled, or even exceeded, that for shunt failure, ETV might still be supported as the initial treatment since virtually all failures will occur within this safe window and virtually all successes will be durable. In addition, endoscopic treatment of hydrocephalus has a lower risk of infection than does shunt implantation. With the same patient population in the same institution, the incidence of shunt infection was 9% whereas the infection

incidence for endoscopic treatment was <1% [15, 25]. A recent multi-institution study reported a 513% infection rate for shunt implantation in PHHP [5]. Shunt infection has been reported to harm neurocognitive development in the PHHP population [10]. Whereas PIH is a disease of poverty [18], PHHP is a disease of prosperity. We have previously shown, in a series of 550 Ugandan patients, that adding CPC to ETV as the primary treatment for hydrocephalus in infants less than 1 year of age significantly increased the rate of success overall from 47% to 66% [15]. Combined ETV/CPC was demonstrated to be safe, with an operative morbidity and mortality of <1% [15]. That study did not include infants with PHHP, but it did include a large number of infants with post-infectious hydrocephalus secondary to neonatal ventriculitis. There are striking similarities between PHHP and PIH, including both intra- and extra-ventricular obstructive inflammatory etiologies. Both etiologies can manifest intraventricular hemorrhage, obstruction of the cerebral aqueduct, intraventricular scarring and loculation, and scarring of the subarachnoid spaces of the posterior fossa leading to fourth ventricle outlet obstruction [6, 18]. Our observations reported here also demonstrate that, in

Childs Nerv Syst

Fig. 3 Mid-sagittal FIESTA MRI of patients with obstructed cisterns noted at ventriculostomy. a Patient no. 1. b Patient no. 6. c Patient no. 7. d Patient no. 8. e Patient no. 9. f Patient no. 10. Note evidence of

aqueduct obstruction in c and d and, of fourth ventricle outlet obstruction in a, b, and f

PHHP, as in PIH, the prepontine cistern may be obstructed from arachnoid adhesions. It has been proposed that CPC augments the success of ETV by reducing the rate of CSF production and thus addressing a so-called communicating component of the hydrocephalus secondary to an impediment in the rate of CSF absorption. In reality, all PHHP is obstructive, with the potential for multiple sites of partial or complete obstruction from the aqueduct to the fourth ventricle outlet foramina to the basal and convexity subarachnoid spaces. ETV may successfully bypass an obstruction at the aqueduct or in the posterior fossa and yet fail because of obstruction to circulation and absorption elsewhere. Choroid plexus cauterization may function to decrease CSF output sufficiently to offset inefficiencies of extra-axial CSF circulation and absorption. This study does not address whether including CPC with ETV in this particular patient population is necessary. It is possible that ETV alone, in the face of an unscarred prepontine cistern, might be equally effective in PHHP. A randomized controlled trial of ETV versus ETV/CPC in a

large number of infants with PHHP would be necessary to make this determination. ETV/CPC was recommended because of its demonstrated superiority to ETV alone in young infants with other etiologies of hydrocephalus. We have previously reported parameters that independently predict ETV/CPC outcome [16, 17]. The small size and homogeneity of the present study cohort precluded analysis of the impact on outcome of CPC extent or of very small differences in age. However, our experience with patient no. 5 (corrected age 11 weeks and 1,000 g weight at time of surgery) was instructive in that the ependymal surfaces and choroid plexus appeared to be more friable than in other cases, making the procedure technically more challenging. Although this patient initially failed, apparently, in part, secondary to residual thrombus in the prepontine cistern and an incomplete CPC, repeat ETV and completion of the CPC at the time of failure ultimately resulted in success. It is now our practice to delay ETV/CPC until the infant is close to full gestational age whenever possible. Although the small sample size is clearly a limitation of this study, the status of the prepontine cistern appeared to

Childs Nerv Syst

correlate with outcome (p=0.033). This is not surprising, given that we have previously shown the presence of cisternal scar to double the risk of ETV failure in treating hydrocephalus of other etiologies [16]. The procedure was successful in all (three of three) who had an open prepontine cistern at the conclusion of the procedure and in only one of seven that did not. Although retrospective and unblinded, our review of the preoperative MRI studies strongly suggested that sagittal FIESTA images can effectively evaluate the status of the prepontine cistern. This may prove very useful in regard to patient selection and will be incorporated into future investigative protocols. Although the overall rate of success for ETV/CPC as primary treatment for PHHP was only 40% in this pilot study (50% with the inclusion of patient no. 5), had those with evidence of cisternal scar been excluded from the procedure preoperatively on the basis of the MRI alone, the success of ETV/CPC for those selected on the basis of an unobstructed cistern would have been much better (three of four, or four of four if patient no. 5 is included as a success). It is important to note here that the majority of infants with PHHP may not be good candidates for treatment by ETV/CPC, since cisternal obstruction from the original hemorrhage was common in this series (60%). Thus, careful patient selection is important. Furthermore, the addition of lamina terminalis ETV as an alternative in those patients with scarred prepontine cisterns failed in four of five patients and was therefore not effective in its intended purpose of bypassing the basal cistern scarring. Prepontine cistern scar may often be an indicator of more diffuse arachnoid scarring that cannot be effectively bypassed. There was no apparent correlation between aqueduct status and outcome. This might be expected since obstructive hydrocephalus secondary to posterior fossa arachnoid scarring rather than aqueduct obstruction is reportedly a common mechanism of PHHP [6]. The neurocognitive outcome for VLBW infants with PHHP appears to best correlate with the severity of the initial hemorrhage [11], and the majority of these children have significant deficits. Furthermore, as already mentioned, shunt infection and multiple shunt revisions are reported to have a further deleterious effect on developmental outcome in these infants [10, 11]. Treatment by ETV may not lead to the small ventricles that have come to be the expectation with shunting [26], but ventricle size may not correlate with developmental outcome [2731]. In a recent study, we have shown that infants with myelomeningocele whose hydrocephalus was treated by ETV/CPC performed as well as or better than those that were shunted in every testing category of the Bayley Scales of Infant Development, and there was no correlation between ventricle size and outcome [31]. Furthermore, ETV avoids the difficult problems associated

with slit ventricles, which are a common complication of shunting in children with PHHP [32]. However, randomized prospective studies of shunting versus endoscopic treatment would likely be required to provide definitive answers to the question of whether one treatment modality is superior to the other in regard to neurocognitive development in this particular population.

Conclusion This initial study in a small homogenous patient cohort demonstrates the safety and efficacy of ETV/CPC in infants with PHHP. Long-term shunt dependence was avoided in 50%. As we have demonstrated for other etiologies of hydrocephalus, the status of the prepontine cistern appeared to correlate with outcome, although the small number of patients prevents firm conclusions from being drawn. In addition, there may be some advantage in waiting until the infant is closer to full gestational age when possible. These preliminary data suggest that, in cases where the cistern is open, ETV/CPC can be expected to be successful in the majority whereas failure should be anticipated when the cistern is scarred. Preoperative mid-sagittal MRI FIESTA images appear to be a reliable predictor of cisternal status and should be incorporated into protocols as an important guide to patient selection in the future.

References
1. National Institute of Neurologic Disorders and Stroke: Hydrocephalus: Myths, New Facts, Clear Directions. Bethesda, MD: NINDS (2005) http://www.ninds.nih.gov/news_and_events/ proceedings/Hydrocephalus_2005.htm (last accessed June 7, 2010) 2. Child Health USA 2003, United States Department of Health and Human Services (2003) http://mchb.hrsa.gov/chusa03/pages/sta tus.htm (last accessed June 7, 2010) 3. Linder N, Haskin O, Levit O, Klinger G, Prince T, Naor N, Turner P, Sirota L, Karmazyn B (2003) Risk factors for intraventricular hemorrhage in very low birth weight premature infants: a retrospective case-control study (2003). Pediatrics 111(5):590595 4. Martins L, Borges MJ, Gomes SM, Macedo I, Nona J, Pinto F, Faria C, Nascimento O, Valido AM. (2009) Intraventricular hemorrhage risk factors in very low birth weight newborns: a case-control study. Einstein 7 (3 Pt 1): 27983 http://apps.einstein. br/revista/arquivos/PDF/1345-Einstein%20v7n3p279-83.pdf (last accessed June 7, 2010) 5. Wellons JC, Shannon CN, Kuklkarni AV, Simon TD, RivaCambrin J, Whitehead WE, Oakes WJ, Drake JM, Luerssen TG, Walker ML, Kestle JR (2009) A multicenter retrospective comparison of conversion from temporary to permanent cerebrospinal fluid diversion in very low birth weight infants with posthemorrhagic hydrocephalus. J Neurosurg Pediatr 4:5055

Childs Nerv Syst 6. Boop FA (2004) Posthemorrhagic hydrocephalus of prematurity. In: Cinalli G, Maixner WJ, Sainte-Rose C (eds) Pediatric Hydrocephalus. Springer, Milano, pp 121131 7. Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr et al (1998) Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurgery 43:294305 8. Beems T, Grotenhuis JA (2002) Is the success rate of endoscopic third ventriculostomy age-dependent? An analysis of the results of endoscopic third ventriculostomy in young children. Childs Nerv sys 18:605608 9. Boynton B, Boynton C, Merritt A, Vaucher YE, James HE, Bejar RF et al (1986) Ventriculoperitoneal shunts in low birth weight infants with intracranial hemorrhage: neurodevelopmental outcome. Neurosurgery 18:141145 10. Resch B, Gedermann A, Maurer U, Ritschl E, Mller W (1996) Neurodevelopmental outcome of hydrocephalus following intra-/ periventricular hemorrhage in preterm infants: short- and longterm results. Childs Nerv Syst 12:2733 11. Levy M, Masri L, McComb J (1997) Outcome for preterm infants with germinal matrix hemorrhage and progressive hydrocephalus. Neurosurgery 41:11111118 12. Javadpour M, Malluci C, Brodbelt A, Golash A, Mary P (2001) The impact of endoscopic third ventriculostomy on the management of newly diagnosed hydrocephalus in infants. Peditatr Neurosurg 35:131135 13. Buxton N, Macarthur D, Mallucci C, Punt J, Vloeberghs M (1998) Neuroendoscopic third ventriculostomy in patients less than 1 year old. Pediatr Neurosurg 29:7376 14. Siomin V, Cinalli G, Grotenhuis A, Golash A, Oi S, Kothbauer K et al (2002) Endoscopic third ventriculostomy in patients with cerebrospinal fluid infection and/or hemorrhage. J Neurosurg 97:519524 15. Warf BC (2005) Comparison of endoscopic third ventriculostomy alone and combined with choroid plexus cauterization in infants younger than 1 year of age: a prospective study in 550 African children. J Neurosurg 103(6):475481 16. Warf BC, Kulkarni AV (2010) Intraoperative assessment of cerebral aqueduct patency and cisternal scarring: impact on success of endoscopic third ventriculostomy in 403 African children. J Neurosurg Pediatr 5:204209 17. Warf BC, Mugamba J, Kulkarni A (2010) Endoscopic third ventriculostomy in the treatment of childhood hydrocephalus in Uganda: report of a scoring system that predicts success. J Neurosurg Pediatr 5:143148 18. Warf BC (2005) Hydrocephalus in Uganda: the predominance of infectious origin and primary management with endoscopic third ventriculostomy. J Neurosurg 102(1):115 19. Sainte-Rose C, Hoffman HJ, Hirsch JF (1989) Shunt failure. Concepts pediatric neurosurgery 9:720 20. Drake J, Chumas P, Kestle J, Pierre-Kahn A, Vinchon M, Brown J, Pollack IF, Arai H (2006) Late rapid deterioration after endoscopic third ventriculostomy: additional cases and review of the literature. J Neurosurg 105(2 Suppl):118126 Warf BC, Campbell JW (2008) Combined endoscopic third ventriculostomy and choroid plexus cauterization as primary treatment of hydrocephalus for infants with myelomeningocele: long-term results of a prospective intent-to-treat study in 115 East African infants. J Neurosurg Pediatr 2:310316 Casey AT, Kimmings EJ, Kleinlugtebeld AD, Taylor WA, Harkness WF, Hayward RD (1997) The long-term outlook for hydrocephalus in childhood. A ten-year cohort study of 155 patients. Pediatr Neurousrg 27:6370 Piatt JH Jr, Cosgriff M (2007) Monte Carlo simulation of cerebrospinal fluid shunt failure and definition of instability among shunt-treated patients with hydrocephalus. J Neurosurg 107(6 Suppl):474478 Stein SC, Guo W (2007) A mathematical model of survival in a newly inserted ventricular shunt. J Neurosurg 107(6 Suppl):448 454 Warf BC (2005) Comparison of one-year outcomes for the Chhabra and Codman-Hakim Micro Precision shunt systems in Uganda: a prospective study in 195 children. J Neurosurg 102 (4):358362 St George E, Natarajan K, Sgouros S (2004) Changes in ventricular volume in hydrocephalic children following successful endoscopic third ventriculostomy. Childs Nerv Syst 20:834838 Hanlo PW (1995) Noninvasive intracranial pressure monitoring in infantile hydrocephalus and the relationship with transcranial Doppler, myelination and outcome. Thesis, Utrecht, The Netherlands Hanlo PW, Gooskens HJM, Vandertop PW (2004) Hydrocephalus: intracranial pressure, myelination, and neurodevelopment. In: Cinalli G, Maixner WJ, Sainte-Rose C (eds) Pediatric Hydrocephalus. Springer-Verlag Italia, Milan, pp 113119 Hanlo PW, Gooskens RJ, van Schooneveld M, Tulleken CA, van der Knaap MS, Faber JA, Willemse J (1997) The effect of intracranial pressure on myelination and the relationship with neurodevelopment in infantile hydrocephalus. Dev Med Child Neurol 39(5):286291 van der Knaap MS, Valk J, Bakker CJ, Schooneveld M, Faber JA, Willemse J, Gooskens RH (1991) Myelination as an expression of the functional maturity of the brain. Dev Med Child Neurol 33:849857 Warf BC, Ondoma S, Kulkarni AV, Donnelly R, Ampeire M, Akona J, Kabachelor CR, Mulondo R, Nsubuga BK (2009) Neurocognitive outcome and ventricular volume in myelomeningocele children treated for hydrocephalus in Uganda. J Neurosurg Pediatr 4:564570 Oi S, Matsumoto S (1987) Infantile hydrocephalus and the slit ventricle syndrome in early infancy. Childs Nerv Syst 3(3):145 150

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.