|Ahead of print
Multiloculated hydrocephalus: Evolution of treatments and outcome
Neelam Krishnan Venkataramana
Department of Neurosurgery, BRAINS, Neuro Spine Hospital, Bangalore, Karnataka, India
|Date of Submission||04-May-2016|
|Date of Decision||29-Jul-2016|
|Date of Acceptance||28-Oct-2020|
|Date of Web Publication||07-Jan-2022|
Neelam Krishnan Venkataramana,
Department of Neurosurgery, BRAINS Hospital, No. 560, 9th ‘A’ Main, Near Indiranagar Metro Station, Indiranagar, Bangalore 560038, Karnataka.
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Multiloculated hydrocephalus has an overall dismal functional outcome; imaging, early diagnosis, and technological advancements have made new roads in its management. Thirty infants with multiloculated hydrocephalus were studied. Progressive enlargement of the head was the most common presentation in 77%; of these, 47% were treated for neonatal meningitis and septicemia and 20% had suffered birth trauma. The majority required a single ventriculoperitoneal shunt; nine of them required multiple shunts; and six were managed with endoscopic fenestration and endoscopic third ventriculostomy. On follow-up, shunt malfunction was the most common complication. Only 26% of the survivors could achieve normal, neuropsychological developments. The mortality in this series is 6.7%.
Keywords: Endoscopy, management, multiloculated hydrocephalus, outcome, V P shunt
| Introduction|| |
Multiloculated hydrocephalus, also called compartmentalized hydrocephalus, is a well-known entity with multiple cerebro spinal fluid (CSF) compartments of variable sizes that are separated by septations. These septations vary in their thickness and, thus, prevent communication among the compartments, resulting in loculations forming paraventricular cystic areas that are associated with hydrocephalus. This is usually seen as a complication of neonatal meningitis or ventriculitis,,,, and intraventricular hemorrhage. The other etiological factors that can lead to similar loculations are not very clear. Brain tumor, head injuries, and occasionally congenital origin of septae have been reported., Though all these children present with clinical signs of hydrocephalus, the management of it differs totally from the uncomplicated hydrocephalus. Computed tomography (CT) or MRI will demonstrate irregular CSF cavities that are separated by septae,; however, the treatment of these children remains a significant challenge.
The objective of this article is to highlight the problems in the management and the role of the early institution of treatment to achieve better outcomes.
| Clinical Material|| |
Thirty children, treated for multiloculated hydrocephalus were analyzed retrospectively. All were infants and 24 were younger than the age of six months. The youngest child was two months old. Twenty were males, and nine were females. Two had an associated meningomyelocele. The duration of presenting symptoms varied from 20 days to 5.5months. and in 70% it was less than three months. Progressive enlargement of the head was the most common in 23 (77%); fever in two, vomiting in four, and dullness and lethargy in four were the other symptoms at the time of hospitalization.
A review of medical records has shown proven CNS infection in seven children in the neonatal period; one of them developed meningitis after a rupture of meningomyelocele. Systemic infection and septicemia were documented in seven; six out of thirty suffered definite birth trauma; and another 12 had a history of being treated by pediatricians for neonatal seizures.
Head circumference ranged from 37 cm to 53 cm, with a mean of 48 cm. Four children had an asymmetrical head enlargement. A dermal sinus over the parietal region was present in one.
| Radiology|| |
Computerized axial tomography of the head revealed hydrocephalus with a loculated ventricular system. Very few underwent an MRI scan. The child with dermal sinus had an additional underlying dermoid. The loculations were bilateral supratentroial in 23, unilateral supratentorial in five, and subtentorial in four; thirteen had a very thin cortical mantle; and post-contrast enhancement was noted in four [Figure 1]; [Supplementary Figure 1] and [Supplementary Figure 2].
| Management|| |
The management protocol was identical in all the children. First, bilateral ventricular puncture was performed through the open fontanelle and CSF was analyzed to exclude infection.
Elevated CSF protein ranging from 60 to 650 mg was noted in 14 children. Four had active infection, where pathogenic organisms could be isolated in CSF. The organisms grown were E coli, Pseudomonas, Streptococci, and Staph. After ruling out infection, a ventriculoperitoneal shunt was placed in 25 children. By choice, a unilateral shunt was inserted on the side of maximum dilatation of ventricles. In 15 children, craniotomy and perforation of septae were done with microscopic guidance. Nine children underwent shunt and ultrasound-guided perforations with a ventricular cannula, and openings were enlarged by unipolar cautery. To an extent, this was a relatively blind procedure often requiring a bilateral approach. Eventually, the loculi were converted into possible monolocular ventricles and a single shunt was placed. A long ventricular catheter with multiple perforations was placed into the ventricles under the guidance of an intraoperative real-time ultrasound. A few children were operated with a stealth navigation system with an electromagnetic (EM) facility. We find this to be very useful. Nine children underwent endoscopic fenestration, of whom three also had an additional endoscopic third ventriculostomy (ETV).
| Technical Description|| |
Endoscopic fenestration can be performed easily in young children. Proper imaging and intraoperative guidance by either ultrasound or navigation is helpful. Preoperative planning of trajectory and the site of entry to access maximum number of loculi and also to perform interventricular septostomy is crucial. Navigation helps in identifying crucial structures, including the floor of the third ventricle, to facilitate ETV.
A larger burr hole or mini craniotomy will help to manipulate the scope in different trajectories. In young children, lateral expansion of the open fontanelle is adequate and offers excellent access. Very rarely does one need to plan bilateral openings. As the scope is past into the larger loculi, the septae will be visualized. It is essential to maintain the orientation all through and when in doubt to verify with guidance. The thickness of the membranes can vary, requiring a variety of techniques of blunt probing, sharp cutting, or bipolar coagulation. One must be careful to know the structures underneath. With the recent optic systems, a vascular zone of the membranes can be identified as the ideal site of perforation. After the initial perforation, fenestration can be widened by scissors under vision. If there are important structures underneath, it is safer to enlarge the opening with a Fogarty balloon. Before attempting ETV, one must be clear and confident about anatomy of the third ventricle. Through a wide opening in the interventricular septum taking care of the fornix, access can be gained to the opposite ventricle. Choroid plexus coagulation can also be performed if needed [Figure 2], [Figure 3], [Figure 4], [Figure 5]–[Figure 6] and [Supplementary Figure 3], [Supplementary Figure 4], [Supplementary Figure 5], [Supplementary Figure 6], [Supplementary Figure 7], [Supplementary Figure 8]–[Supplementary Figure 9].
Serial CT scans in the postoperative period revealed (1) Bilateral communication and drainage of ventricles, (2) reduction in the ventricular size with apparent enlargement of adjacent loculi, (3) unilateral drainage, and (4) unilocular drainage.
On the basis of these findings, an additional procedure or additional shunt was required in nine children. These children had persistent symptoms requiring reinvestigations. All the children received prophylactic anticonvulsants. Four had subdural collections as a complication of shunt over drainage.
Nine children were treated with endoscope navigation-assisted (EM stealth Medtronic) fenestration, and they were subsequently managed with one shunt only. Three did not require a shunt at all after fenestration and ETV. The operative morbidity and shunt function rates were better in this group.
| Outcome|| |
All 30 children had a close follow-up, ranging from 1.6 to 6 years. Two children died, accounting for a mortality of 6.7%. Seventeen of them required shunt revisions and repositioning of ventricular catheters. Thirteen had multiple neuropsychological assessment, which revealed near-normal development in only eight (26.6%) and the rest were significantly retarded in their development.
| Discussion|| |
Multiloculated hydrocephalus poses several challenges in the diagnosis, techniques adopted, success of procedures, and final outcome. The first report of multiloculated hydrocephalus after neonatal meningitis was by Salman in 1970. Multiloculated hydrocephalus is defined as multiple separated cystic cavities or spaces located in or in relation to the ventricular system and filled with fresh or altered CSF.
Hydrocephalus and the loculation of ventricles with the presence of septae or membranes is known to complicate ventriculitis in children. The exact incidence of this complication is not clearly known. Handler and Wright reported an incidence of 23% of ventricular loculation after meningitis. All infections, such as pyogenic, tubercular, fungal, and shunt infections, are known to cause this complication. However, with a newer generation of antibiotics and a reduction in the mortality of neonatal meningitis, the survivors need to be followed meticulously for early detection of this complication. In 1975, Eller and Pasternak described intraventricular hemorrhage as an additional cause. David Sandburg also reported an association of prematurity. With the advent of CT/MRI scan, similar clinical condition have been described in brain tumors, head injuries, and after intraventricular hemorrhage.,
In our series, seven had evidence of past infection and four of current active pyogenic infection requiring a full course of antibiotics before treatment. Seven had a history of birth trauma, and resultant intraventricular hemorrhage is known to cause hydrocephalus; however, the exact mechanism of multiloculation after intraventricular hematoma is not clear. It is proposed that the high protein and the reaction to the hemorrhage can form thick membranes and the resultant inflammation can convert them into loculi. Prematurity was seen in 10 in our series. In nine children, we could not find any accountable cause for loculated hydrocephalus possibly might have had any of the above risk factors in the initial stages. The high incidence of neonatal seizures in this group suggests probable insult to the immature nervous system from infection or trauma at an early age. In general, the occurrence of seizures, both focal and generalized, was so high in this group that we routinely put them on anticonvulsants.
In the majority, the illness manifested in a short duration after infection or injury. In our series, enlargement of the head was noted as early as 20 days after the treatment of meningitis and in 70% of them the clinical picture became obvious in three months. Similar evolution was as reported in earlier series.,
CT scan is very useful in the diagnosis of this illness., It denotes the number, size, location, and distribution of loculi apart from hydrocephalus and the degree of the parenchymal damage. We advocate contrast-enhanced CT scan, and the presence of ependymal enhancement or enhancing wall of the loculi should be viewed seriously as the presence of active infection. We have found such enhancement in four children and all of them have grown organisms from the CSF. The CSF from different loculi can vary in their protein content. Hence, the need for repeated CSF sampling from different loculi and differential analysis need not be overemphasized to prevent postoperative infection in these children. The intracranial pressure can vary from loculi to loculi locally. Such differential ICP can cause a selective increase in the size of the cysts.,
In recent years, MRI has almost replaced CT. Though CT cisternography is useful in the identification and communication among the locule, MR ventriculography has an added advantage of direct multiplanar imaging, in addition to the absence of ionizing radiation, absence of bony artifacts, and high spatial and contrast resolution. Moreover, the 3D CISS sequence provides the best morphological images due to excellent fluid and tissue contrast. It helps in differentiating the loculation of the ventricles and multiple cysts extending into the parenchyma. In addition, communication to the ventricles can be deciphered at times [Figure 7] and [Supplementary Figure 10].
The treatment of choice is only surgery. The surgical options available are CSF diversion (multiple shunt systems), stereotactic aspiration, craniotomy, and microsurgical fenestration and endoscopic fenestration.,,,,,,,,, Due to noncommunication of the cysts, invariably one will require multiple shunts, which will add to the problems. However, shunts alone will not be successful in this condition. The presence of chronic and persistent inflammation of the ependyma and the surrounding tissues has been attributed to the high incidence of shunt obstructions. Craniotomy and fenestration was first suggested by Rhoton and Gomez. In 1982, Kleinhaus et al. described the use of an endoscope. In 1986, Powers described the role of a flexible endoscope. The decision to conduct fenestration is often made when a single shunt fails to reduce symptoms adequately. Multiple shunts can function effectively but the more the hardware the more the chances of infection. In multiple shunts, it is difficult to establish effective functionality and also revision is a challenge. Often, these shunts get entangled in the gliotic scar tissue of the loculi and pose difficulty in removal as well [Figure 8].
Although Hoffman advocated the requirement of multiple shunts in these children, the decision seems to be governed by many factors. As such, no definite guidelines are available. Rhoton and Gomez treated two patients with craniotomy and surgical resection of loculated cysts. Rhoton described the technique of conversion of loculi into a simple cavity. However, the loculi widely varied in their size, number, and distribution in our series. We feel that unilateral shunt placement is sufficient in the great majority after opening the loculi. Fenestration of septae and a multi-perforated, long ventricular catheter is quite useful. The choice of the side can be determined by the side of the larger ventricle or locule with a preserved cortical mantle. A low-pressure shunt system is preferable, as we have not found a significant increase in the ICP as measured during surgery, probably due to the associated parenchymal damage and fibrosis. The valve requires removal wherever the protein content is high. Only nine children required bilateral shunts in our series. Craniotomy for fenestration allows better visualization, and it facilitates easy fenestration or excision of membranes with a possibility of creating wider commmunication between compartments apart from achieving hemostasis easily. However, it is a major surgical procedure that is often bilateral and certainly increases the incidence of subdural due to over-drainage of CSF.
In recent years, endoscopy has gained prominence in the treatment of multiloculated hydrocephalus and renewed the interest of surgical treatment. Use of the endoscopy technique has claimed superiority by many authors and decreased the number of shunt revisions. Our experience supports this observation. The children who underwent endoscopic fenestration had better results, and some required only one shunt. Some authors used a steerable flexible scope, but the size of fenestration can be a limiting factor. We prefer a rigid endoscope over the flexible one to make better and wider fenestrations. To overcome the maneuverability, a mini-craniotomy or a larger burr hole is advocated. Careful selection of the site of entry and trajectory is very important to access maximum number of loculi bilaterally. Recent high-definition cameras offer greater light intensity and superior optics. To have intraoperative guidance ultrasonography is very useful. The Medtronic frameless EM Navigation system is advantageous in children. However, conventional navigation systems are of no additional advantage as the brain shifts after drainage of the loculi can create significant variation. Navigation can help in many ways: (a) to achieve complete and successful fenestration of locule, (b) bilateral communication of ventricles by proper septostomy, (c) to identify important structures and/ or to plan ETV, and (d) proper placement of a long catheter traversing across many loculi. Schulz reported significant clearance in 56% of children with established CSF diversions with endoscopy alone, and 44% required an additional procedure. Our experience is similar: A shunt was avoided wherever possible with ETV, and the rest were managed with a single shunt. Thulium laser has been found to be very useful in treating multiloculated hydrocephalus. Thulium was introduced to neuroendoscopy in 2007. Its efficacy in both coagulation and tissue ablation makes it very effective and handy in opening the membranes widely and easily. The wavelength of this laser is close to the absorption band of water (1.92 nm). Therefore, the thermal damage is minimal and the micro fiber is flexible and can be used even with a steerable, flexible endoscope.
Though there are no randomized control studies that define the superiority between the craniotomy and endoscopy, the consensus as of today is in favor of endoscopic fenestration. The advantages are minimal invasiveness, minimal brain retraction, minimal blood loss, minimal operating time, and shorter hospital stay. Nowoslawska reported that endoscopy has reduced the number of shunts as well as shunt revisions. Post endoscopy, subdural collections are significantly less in our series though seen in a few. However they were not clinically significant. A similar experience was documented by Heilmen et al. Apart from fenestration of the cysts, septostomy, foraminoplasty, aqueductoplasty, and ETV can be considered simultaneously. We follow the protocol to offer endoscopic fenestration of the loculi from the right side (depending on the access combined with septostomy to communicate to the opposite ventricle). If the ventricular enlargement is significant, simultaneously ETV can be performed. In case of aqueductal stenosis, aqueductoplasty will be considered. The child will be followed clinically and radiologically. A shunt can be considered only in the presence of persistent or progressive symptoms. In addition, CSF diversion through ventriculoperitoneal shunt is preferable based on the merits of the situation. The whole idea is to limit the number of shunts to one if not two on either side [Supplementary Figure 11]A and [B].
|Supplementary Figure 11: (A)Multi loculated hydrocephalus manage with single shunt after endoscopic fenestration. (B) Endoscopic fenestration and ETV|
Click here to view
Wager and Koch described membrane re-occlusion; however, we did not encounter this complication, as we make wider fenestration. Continuous irrigation of ventricles often helps to maintain the endoscopic visibility, provides easy access into the loculi, and makes the fenestration of the membrane possible. In addition, we can wash out protein content as well. Naser et al. reported requirements of repeat endoscopic surgery in 33% of their children over 30 months of follow-up. It is interesting to note that children previously treated with a shunt had a 7.5-fold risk of having repeat endoscopy than the children who underwent endoscopic cyst fenestration primarily. Even after craniotomy, 16% required repeat procedures. This proves the fact that early diagnosis and fenestration are critical in obtaining best results.
CT/ MRI scan has a significant role again during follow-up, than the clinical symptoms and serial head measurements alone. The most common complication encountered is shunt malformation in our children. The presence of high protein in CSF, improper placement of shunt tube, and kinking of ventricular catheter due to the presence of membranes were the causes of shunt malformation requiring repositioning or revision. Intraoperative ultrasound can guide the placement of the ventricular catheter.
The final outcome and intellectual development is not very encouraging in these children, despite the aggressive treatment. Mortality is high, as 70% has been reported earlier. In our study, the mortality is only 6.7%. Neuropsychological examination of the survivors revealed subnormal developmental levels in 73%, and 26% of them were severely retarded. Only 26% could eventually achieve near-normal developmental levels. A high incidence of seizures, cerebral damage due to neonatal meningitis, delay in the diagnosis and treatment, and repeated surgical procedures might be the contributory factors for intellectual deterioration.
The overall outcome depends on the etiology, age of the child, prematurity, presence of active infection, timing of diagnosis, and early intervention. Children with infections are known to undergo parenchymal damage and later on have high CSF protein, leading to repeated shunt failures. Early eradication of infection and treatment is crucial in these children. Schulz et al. noted improvement in neuromotor and cognitive deficits in children with early treatment, probably due to preserved brain tissue at the time of surgery. On the contrary, with significant parenchymal damage the functional outcomes are rather poor.
During active infection, ependymal damage can cause subependymal seepage of infection and toxins, leading to periventricular leucomalacia. The entrapped cysts can continue to harbor bacteria or their toxins for a longer time despite antibiotic treatment. Apart from infection, there could be many other mechanisms accounting for their poorer functional or intellectual outcomes.
Significant parenchymal damage and the gliotic tissue failing to reconstitute itself even after successful treatment could be the major cause. Infection-associated vasculitis and occlusions of Virchow–Robin spaces can alter the cerebral micro-circulation, causing ischemic damage. The pattern of raised ICP is different in comparison to pure ventricular dilatation. The parenchyma is often sandwiched between the loculi, in the periphery and the larger ventricles in the middle, and the resultant raised ICP need not be symmetrical but could be multifocal. Despite the child not having overt symptoms of raised ICP, one or more loculi can continue to enlarge, leading to localized raised pressure and focal tissue damage. The intervention is often delayed, as the child may be apparently asymptomatic in spite of the growth of few loculi. This fact is reflected in our study, as the outcomes are better in recent years compared with yester years out of early imaging and intervention. Apart from early diagnosis and intervention advances in imaging, antibiotics with good CNS penetration and early endoscopic fenestration have contributed significantly.
The poorer outcomes often put surgeons in a dilemma of decision making, ethics, as well as the cost-effectiveness of treatment. Since early interventions with endoscopy have shown significantly improved results, one must adopt this procedure for the future as early as feasible.
| Conclusion|| |
Multiloculated hydrocephalus continues to pose several diagnostic and therapeutic challenges. The overall outcomes remain to be poor. However, children with adequate treatment of meningitis, early detection of hydrocephalus with fewer loculi, preferably unilateral with preserved parenchyma, can achieve their near-normal psychological and functional development. Early endoscopic fenestration should proceed the shunt. This fact highlights the need for serial follow-up of the high-risk infants, early detection, and treatment with endoscopic fenestration for superior results.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Sandberg DI, McComb JG, Krieger MD. Craniotomy for fenestration of multiloculated hydrocephalus in pediatric patients. Operative Neurosurg2005;57:100-6.
Albanese V, Tomasello F, Sampaolo S, Picozzi P. Neuroradiological findings in multiloculated hydrocephalus. Acta Neurochir (Wien) 1982;60:297-311.
Albanese V, Tomasello F, Sampaolo S. Multiloculated hydrocephalus in infants. Neurosurgery 1981;8:641-6.
Brown LW, Zimmerman RA, Bilaniuk LT. Polycystic brain disease complicating neonatal meningitis: documentation of evolution by computed tomography. J Pediatr 1979;94:757-9.
Handler LC, Wright MG. Postmeningitic hydrocephalus in infancy. Ventriculography with special reference to ventricular septa. Neuroradiology 1978;16:31-5.
Kalsback JE, Desouse AL, Kleiman MB, Goodman JM, Franker EA. Compartmentalization of cerebral ventricles as a sequel of neonatal meningitis. J Neurosurg 1980;52;547-52.
Savolaine ER, Gerber AM. Computerized tomography studies of congenital acquired cerebral intraventricular membrane, report of two cases. J Neurosurg 1981;54;388-91.
Nida TY, Haines SJ. Multiloculated hydrocephalus: craniotomy and fenestration of intraventricular septations. J Neurosurg 1993;78:70-6.
Salmon JH. Isolated unilateral hydrocephalus following ventriculoatrial shunt. J Neurosurg 1970;32:219-26.
Eshra MA. Endoscopic management of septated, multiloculated hydrocephalus-Alex. J Med2013, doi:10.1016/j.ajme.2013.11.002
Lazareff JA, Sadowinski S. The probable role of hydrocephalus in the development of intraventricular septa. An observation of one case. Childs Nerv Syst 1992;8:139-41.
Eller TW, Pasternak JF. Isolated ventricles following intraventricular hemorrhage. J Neurosurg 1985;62:357-62.
Theodore WE, Josephe FP. Isolated ventricles following intraventricular haemorrhage. J Neurosurg 1985;62;357-62.
Schultz P, Leeds NE. Intraventricular septations complicating neonatal meningitis. J Neurosurg 1973;38:620-6.
Rekate HL. Treatment of hydrocephalus. In: Cheek WR, editor. Pediatric Neurosurgery, Surgery of the Developing Nervous System. 3rd ed. Philadelphia, PA: WB Saunders; 1994. pp. 202-20.
Oi S, Abbott R. Loculated ventricles and isolated compartments in hydrocephalus: their pathophysiology and the efficacy of neuroendoscopic surgery. Neurosurg Clin N Am 2004;15:77-87.
Rifkinson-Mann S, Sachdev VP, Huang YP. Congenital fourth ventricular midline outlet obstruction. Report of two cases. J Neurosurg 1987;67:595-9.
Gandhoke GS, Frassanito P, Chandra N, Ojha BK, Singh A. Role of magnetic resonance ventriculography in multiloculated hydrocephalus. J Neurosurg Pediatr 2013;11:697-703.
Fritsch MJ, Mehdorn M. Endoscopic intraventricular surgery for treatment of hydrocephalus and loculated CSF space in children less than one year of age. Pediatr Neurosurg 2002;36: 183-8.
Heilman CB, Cohen AR. Endoscopic ventricular fenestration using a “saline torch”. J Neurosurg 1991;74:224-9.
Kleinhaus S, Germann R, Sheran M, Shapiro K, Boley SJ. A role for endoscopy in the placement of ventriculoperitoneal shunts. Surg Neurol 1982;18:179-80.
Lewis AI, Keiper GL Jr, Crone KR. Endoscopic treatment of loculated hydrocephalus. J Neurosurg 1995;82:780-5.
Nowosławska E, Polis L, Kaniewska D, Mikołajczyk W, Krawczyk J, Szymański W, et al
. Effectiveness of neuroendoscopic procedures in the treatment of complex compartmentalized hydrocephalus in children. Childs Nerv Syst 2003;19:659-65.
Oi S, Hidaka M, Honda Y, Togo K, Shinoda M, Shimoda M, et al
. Neuroendoscopic surgery for specific forms of hydrocephalus. Childs Nerv Syst 1999;15:56-68.
Powers SK. Fenestration of intraventricular cysts using a flexible, steerable endoscope. Acta Neurochir Suppl (Wien) 1992;54:42-6.
El-Ghandour NMF. Endoscopic cyst fenestration in the treatment of multiloculated hydrocephalus in children. J Neurosurg Pediatr 2008;1:217-22.
Schulz M,Bohner G, Knaus H, Haberl H, Thomale U-W . Navigated endoscopic surgery for multiloculated hydrocephalus in children. J Neurosurg Pediatr 2010;5:434-42.
Foda et al
. Endoscopic management of loculated hydrocephalus. Thesis. Faculty of Medicine, University of Alexandria; 2009.
Spennato P, Cinalli G, Carannante G, Ruggiero C, Del ML. Multiloculated hydrocephalus. In: Cinalli G, Maixner WJ, Sainte-Rose C, editors. Pediatric Hydrocephalus. Milan: Springer; 2004. pp. 219-44.
Rhoton AL, Gomez MR. Conversion of multilocular hydrocephalus to unilocular. J Neurosurgery 1972;36:348-50.
Powers SK. Fenestration of intraventricular cysts using a flexible, steerable endoscope and the argon laser. Neurosurgery 1986;18:637-41.
Hoffman JC Jr. Application of computerized axial tomography (CT) in pediatric neurosurgery. In: ‘O’ brain MS, editor. Pediatric Neurological Surgery. New York: Raven Press; 1978.
Lee YH, Kwon YS, Yang KH. Multiloculated hydrocephalus: open craniotomy or endoscopy? J Korean Neurosurg Soc 2017;60:301-5.
Oppido PA, Cattani F. Neurosurgery, National Cancer Institute, Regina Elena, Rome, Italy - The Utility of Thulium Laser in Neuroendoscopy. Available from: http://neurosurgery.imedpub.com/archive.php. [Last accessed on 2020 October 23].
Wagner W, Koch D. Mechanisms of failure after endoscopic third ventriculostomy in young infants. J Neurosurg 2005;103:43-9.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10], [Figure 11], [Figure 12], [Figure 13], [Figure 14], [Figure 15], [Figure 16], [Figure 17], [Figure 18], [Figure 19]