Peer-Review Reports

ETV for Idiopathic Aqueductal StenosisPietro Spennato1, Sanna Tazi2, Olivier Bekaert 2, Giuseppe Cinalli1, Philippe Decq2

INTRODUCTIONThe sylvian aqueduct is the most commonsite of intraventricular blockage of the cere-brospinal fluid (CSF). Stenosis of the aque-duct is responsible of 6% to 66% of cases ofhydrocephalus in children and 5% to 49%in adults (30, 37). There are two peaks ofdistribution for age, one in the first year oflife and the other in adolescence (30).

The sylvian aqueduct may become ste-notic as a consequence of compressionfrom mass lesions or because of intrinsicpathology. Intrinsic aqueductal stenosismay be congenital (Figure 1) or acquired.Russell (57) in her histopathological classi-fication in 1949 subdivided nontumoral aq-ueduct anomalies into 4 types: stenosis,forking, septum formation, and gliosis(Figures 2 through 4). In about three quar-ters of patients, the etiology of the disorder isnot known (idiopathic aqueductal stenosis[IAS]) (37). In the remaining cases it can beattributed to different causes: genetic factors(X-linked hydrocephalus [4]), bacterial and

viral infections (both intrauterine and infan-tile) (13, 37), hemorrhage (intraventricularhemorrhage of the prematurity, subarach-noid hemorrhage), central nervous systemmalformations (Arnold-Chiari, spina bifida,encephaloceles, Dandy-Walker) (13, 48, 37)(Figures 2 and 5).

Aqueductal stenosis should not be con-sidered a stable condition. Often it is welltolerated for years. Many hypotheses havebeen advanced to explain this phenome-non. Lapras et al. (47) suggested that occlu-sion can be worsened by head traumas, sub-arachnoid hemorrhages, or viral infectionsduring life. They also suggested that partialstenosis could be completed by functionalmechanisms (46): accumulation of fluid inthe supratentorial ventricular system maycause distortion of the brain stem and kink-

ing of the aqueduct, worsening the stenosisin a vicious cycle.

Recently, aqueductal stenosis has beendescribed in a condition known as LOVA(longstanding overt ventriculomegaly inadults) (53). This is a condition with rootsin childhood, but that manifests in middle-age adults: the patients have severe triven-triculomegaly and macrocephalus, andbecome symptomatic during adulthood,usually mimicking a normal-pressure hy-drocephalus syndrome (dementia, gait dis-turbances, and urinary incontinence).

The movement of CSF through the aque-duct has a pulsatile nature, with a systolicand diastolic to-and-fro displacement. Thenet outward movement is equivalent to CSFproduction, which occurs predominantly inthe lateral and third ventricles. During sys-

! BACKGROUND: Idiopathic aqueductal stenosis is a cause of noncommuni-cating hydrocephalus, which actual treatment with endoscopic third ventriculo-stomy (ETV) could assess without any interference with the etiology. The resultsof ETV in this indication therefore could be interpreted as the result of thesurgical procedure alone, without any additional factors related to the etiologyof the cerebrospinal fluid pathway obstruction, such as hemorrhage, infection,brain malformations, or brain tumors or cysts.

! METHODS: After a brief description of pathogenesis of hydrocephalus inaqueductal stenosis, the authors review the literature for studies on ETV,extrapolating patients with idiopathic aqueductal stenosis in infancy, childhood,and adulthood. Differences in outcome between patients treated with ETV andpatients treated with ventriculoperitoneal shunt (VPS) are also reviewed.

! RESULTS: The overall success rates of ETV range between 23% to 94%, witha mean of 68%; when only patients affected by obstructive triventricularhydrocephalus secondary to aqueductal stenosis are considered, the successrate is actually quite homogeneous and stable, being above 60% at any age, evenif a trend in lower success rate in very young infants (younger than 6 months ofage) is noticeable. The few reports on intellectual outcome failed to demonstratedifferences between ETV and VPS.

! CONCLUSIONS: Several issues, such as the cause of failures in well-selectedpatients, long-term outcome in infants treated with ETV, effects of persistentventriculomegaly on neuropsychological developmental, remain unanswered.Larger and more detailed studies are needed.

Key words! Hydrocephalus! Intellectual outcome! Neuroendoscopy! Sylvian aqueduct

Abbreviations and AcronymsCSF: Cerebrospinal fluidETV: Endoscopic third ventriculostomyIAS: Idiopathic aqueductal stenosisMRI: Magnetic resonance imagingVPS: Ventriculoperitoneal shunt

From the 1Department of PediatricNeurosurgery, Santobono-Paulipon

Pediatric Hospital, Naples, Italy; and 2UPEC, APHP, GroupeHospitalier Henri Mondor, Service de Neurochirurgie,Créteil, France

To whom correspondence should be addressed:Pietro Spennato, M.D. [E-mail: spennato2@libero.it]

Citation: World Neurosurg. (2012).DOI: 10.1016/j.wneu.2012.02.007

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter © 2012 Elsevier Inc.All rights reserved.

WORLD NEUROSURGERY xx [x]: xxx, MONTH 2012 www.WORLDNEUROSURGERY.org 1

12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

AQ: au

F1

F2-F4

123456789

10111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758

AQ: 2

F5

AQ 1

tole an increased amount of blood volumereaches the cranial cavity; thereafter, ac-cording to the Monro-Kellie doctrine, thesame volume of venous blood and CSFleaves the cranial cavity. Thereafter CSF isforced through the aqueduct and foramenmagnum. This is the consequence of thedistensibility of the lumbar sac, which actsas a buffer to the changes in intracranialvolume. During diastole, the decrease inbrain blood volume and recoil of CSF dis-placed in the lumbar sac reverses the CSFdisplacement (14, 56). In normal circum-stances, CSF flow in the aqueduct is lami-

nar, even when the direction is reversedduring systolic and diastolic to-and-fro dis-placement. The shape of the aqueduct, driv-ing a core of fluid centrally away from thewall effect, is important in maintaining alaminar flow (35).

Changes in aqueductal size and shapewill alter the volume flow rate and the flowpattern. As the aqueduct narrows, the flowdecreases for a given pressure differenceand the velocity increases (34). To maintainthe flow through the aqueduct, a pressuregradient between the third and fourth ven-tricles may develop. Thereafter, with the in-

crease of the pressure inside the aqueductand of the wall shear stresses, gliosis withfurther narrowing may develop (34).

This pressure gradient may lead to typicalanatomical deformations of the third ven-tricle [enlargement of suprapineal, laminaterminalis and infundibular recesses, andformation of atrial diverticula (13)], and tosymptoms specific for hydrocephalus sec-ondary to aqueductal stenosis [endocrino-logic and visual disturbances due to masseffect on the chiasmatic hypothalamic re-gion, and ocular disturbances, extrapyra-midal signs, and global rostral midbraindysfunction due to a mass effect on the qua-drigeminal-pineal region (13)].

The importance of CSF pulsation in thedevelopment of ventricular dilatation hasbeen widely discussed (28, 49) and leads tothe proposed pathophysiological mecha-nism: when the aqueduct is stenotic, thesystolic displacement of CSF is impaired.The CSF pulse during systole is transmittedto the ependymal walls, leading to periven-tricular edema, ischemia, tissue damage,and ventricular dilatation (49). Greitz (28)hypothesizes different mechanisms in ven-tricular dilatation in acute and in chronicphases. In case of acute intraventricularblock, the CSF is prevented from reachingits main absorption site. The ventricles in-crease in volume because the periventricu-lar capillaries can only absorb part of theCSF produced. The pressure in the ventri-cles and brain become elevated. The ven-

Figure 1. Intrauterine MRI (A) showing hydrocephalus secondary to congenital aqueductal stenosis.Postnatal midsagittal MRI (B) after endoscopic third ventriculostomy performed in the second day oflife. MRI, magnetic resonance imaging.

Figure 2. Midsagittal T1 MRI (A) and axial T2 MRI (B) showing stenosis of the whole aqueductassociated with supependymal glial nodules in a patient with drug-resistant epilepsy. MRI, magneticresonance imaging.

Figure 3. Midsagittal T2 MRI showing septalstenosis in the aqueduct. MRI, magneticresonance imaging.

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

2 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2012.02.007

5960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116

5960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116

tricular dilatation displaces the brain to-ward the skull and compresses the corticalveins. This leads to venous congestion withincreased blood volume and further in-creased intracranial pressure. The venouscongestion and the brain swelling counter-act the ventricular dilatation. At some point,a new equilibrium is reached at higher pres-sure. The arteriolar and capillary regulationof fluid absorption in the periventricularbrain capillaries finally balances the pro-duction of CSF inside the isolated ventri-

cles. The CSF pressure decreases, and a newequilibrium is established at near-normalpressure in the chronic phase of obstructivehydrocephalus (28). In this phase, the ve-nous outflow compression and venous con-gestion disappears; thereafter, there is noforce to counteract ventricular dilatation.Because the ventricles are isolated fromthe cranial and spinal subarachnoid spaces,the CSF pulse during systole increases thetransmantle pulsatile stress, which contin-ues to dilate the ventricles even if the meanCSF pressure is normal (28).

Following that previous hypothesis, en-doscopic third ventriculostomy (ETV) inobstructive hydrocephalus has 2 main pur-poses: to provide an alternative route to CSFin order to bypass the obstructed or stenoticaqueduct, and to increase ventricular com-pliance, restoring communication betweenthe ventricles and the cranial and spinalsubarachnoid spaces, reducing the trans-mantle pulsatile stress.

Definition of IAS HydrocephalusBy its restrictive definition, IAS hydroceph-alus is a noncommunicating hydrocephalusin which CSF pathway obstruction is lo-cated at the site of the aqueduct of Sylvius(on its proximal or distal part, or intersect-ing the all aqueduct [Figures 2 through 4])without any additional extrinsic compres-sion (such as a tumor or a hematoma) andwithout any etiology on the medical history(such as meningitis or a genetic disease)

that is known to generate an aqueductal ste-nosis. Unfortunately, the term IAS is rarelyfound in the literature. We mostly foundaqueductal stenosis or primary aqueductalstenosis, but these terms are rarely defined.Aqueductal stenosis alone could be used inall cases in which the CSF obstruction islocated in the aqueduct, whatever the cause.The term primary aqueductal stenosis maysuppose that all of the extrinsic compres-sions have been excluded (such as a tumorcompression), but often nothing is knownabout the previous medical history. This isthe reason why it is difficult to assess theresults of third ventriculostomies per-formed in series in which all of the cases aremixed, not only the age but also the etiolo-gies.

Indications for ETVThe indications to perform an ETV ratherthan a shunt operation have been increasedin recent years. Initially, the ideal candi-dates had to meet these criteria (32, 40):acquired aqueductal stenosis, adequate sizeof third ventricle (at least 1 cm bicoronaldiameter), extension of the floor of the thirdventricle behind and below the dorsum sel-lae, and potential patency of subarachnoidspaces. Thus, patients previously shuntedand patients who experienced meningitis,subarachnoid hemorrhages, or associatedspinal dysraphism were considered poorcandidates (31, 54, 69). Recently, there hasbeen a tendency to include more patients in

Figure 4. Segmental aqueductal stenosis. (A) Midsagittal T1 MRI showingstenosis of the proximal third of the aqueduct, associated with expansionof the suprapineal recess. (B) Midsagittal T2 MRI showing stenosis of the

distal third of the aqueduct. (C) Midsagittal T2 MRI showing stenosis ofthe middle third of the aqueduct. Note the flow artifact following thirdventriculostomy (arrow). MRI, magnetic resonance imaging.

Figure 5. Midsagittal T2 MRI showingaqueductal stenosis associated withdysplasia of the tectal plate. MRI, magneticresonance imaging.

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

WORLD NEUROSURGERY xx [x]: xxx, MONTH 2012 www.WORLDNEUROSURGERY.org 3

117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174

117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174

AQ: 3

all cases when absence of communicationbetween the ventricles and the subarach-noid spaces is suspected, including caseswith myelomeningocele, Chiari malforma-tion, congenital aqueductal stenosis, steno-sis of the foramina of Magendie andLuschka, previous meningitis, age youngerthan 2 years, and prior ventriculoperitonealshunt (VPS) (11, 39, 42, 51, 60, 64). In pa-tients who suffered meningitis, shunt in-fection, subarachnoid and/or intraven-tricular hemorrhage, and hydrocephalusassociated with spinal dysraphism, ETV iseffective in approximately two thirds ofpatients (10, 33). Previous shunting couldeven increase the success rate of ETV bydecreasing the transmantle pressure, al-lowing arachnoid granulations to open andmature (33, 51). Controversial are still pres-ent in patients younger than 1 year old (10,24 33) because of underdeveloped granula-tions. Warf (72) suggested associating en-doscopic coagulation of the choroid plexuswith ETV in order to increase the successrate in children younger than 1 year old. Asno clinical tests, even an invasive infusiontest, may predict outcome, because theopening of the subarachnoid spaces afterthe operation may take days or even weeks(31), ETV should be suggested for all casesof allowable anatomy, giving patients thechance to remain or to become shunt-free.However, adequate information regardingmildly higher surgical risks and lower suc-cess rates (particularly in the first year of lifeand in cases with a history of meningitis orhemorrhage) than a shunting operationshould be addressed in the preoperative in-formed consent.

Preferred Surgical TechniqueThe procedure is usually performed undergeneral anesthesia. Through a burr holeplaced 0.5 cm in front of the coronal sutureand 2.5 cm away from the midline (usuallyon the right side), the right (or left) lateralventricle is penetrated with the sheath of theendoscope. A rigid rod lens endoscope(Decq endoscope, Karl Storz GmbH & Co.,Tuttlingen, Germany) is inserted in thesheath after stylet removing. The anatomi-cal landmarks of the lateral ventricle (cho-roid plexus, thalamostriate and septalveins, foramen of Monro) are carefullyidentified (17). Through the foramen ofMonro, the third ventricle is entered withthe endoscope, and the anatomical land-

marks of the floor, in particular the mam-millary bodies and the infundibular recess,are carefully visualized. The floor, usuallyvisible as a translucent membrane, is perfo-rated between the infundibular recess infront and the mammillary bodies behind.Several techniques have been described toperforate the floor and enlarge the stoma(13, 18). We personally prefer to perforatethe floor and enlarge the stoma with a singleinstrument, such as the Decq forceps (16).If further dilatation is needed, the stoma isdilated by inflating a Fogarty or double bal-loon catheter. Thereafter, the subarachnoidspace is entered with the endoscope. If anarachnoid membrane (Liliequist mem-brane) prevents the free flow of CSF, it isfenestrated with a delicate surgical maneu-ver, avoiding sharp instruments and mono-polar coagulation. Balloon dilatation is thepreferred method. The clear vision of thestructures of the interpeduncular cistern(basilar artery and/or its branches, brains-tem, third cranial nerve, and dura of theclivus) is the goal of surgery, indicating thatthe third ventricle freely communicateswith the basal cisterns. After removing theendoscope out of the stoma, the visualiza-tion of the pulsation of the third ventricularfloor, and especially the pulsatile move-ments of the edges of the stoma, is an indi-rect marker of the CSF flowing through theperformed stoma.

RESULTSThe results are difficult to assess because inalmost all of the series, the patients (adultsand children) as well as the etiologies (idio-pathic stenosis, extrinsic aqueductal com-pression, etc.) are mixed together. ETV, byperforming a communication between thethird ventricle and the subarachnoid space,appears to be able to restore an almost nor-mal physiological CSF circulation in casesof noncommunicating hydrocephalus.Nevertheless, several reports, despite theaccuracy of patient selection, show a signif-icant number of patients in which ETVfailed to control the hydrocephalus, and ashunt was finally inserted (10, 30). This canbe more frequently observed in the immedi-ate postoperative period (early failures) orcan sporadically be observed later, usuallywithin 5 to 6 years from the operation (latefailures). Feng (21) calculated a proportionof 75% to 80% functioning ETV at 1 year,

Cinalli et al. (10) a functioning rate of 72%at 15 years, excluding technical failures.These data are comfortable if comparedwith those of shunt surgery, in which a 50%failure rate within 2 years is expected (19).

The overall success rates of ETV rangebetween 23% (8) and 94% (29), with a meanof 68% (18); when only patients affected bynoncommunicating triventricular hydro-cephalus secondary to aqueductal stenosisare considered, the success rate is actuallyquite homogeneous and stable, being above60% at any age (1, 10, 25, 26, 36). Only Tuliet al. (67) reported a 44% failure rate. Mostauthors include neuroradiological criteriato confirm the clinical success (3, 6, 10) (Ta-ble 1). According to these authors, the ven-tricular system must decrease in size or atleast remain stable, and radiographic signsof active hydrocephalus must disappear. Incase of progressive dilatation of the ventri-cles (despite the absence of symptomatol-ogy), another procedure is warranted (10).The cause of ETV failures in idiopathic aq-ueductal stenosis remains unknown, possi-bly reflecting a multifactorial etiology of hy-drocephalus in these cases.

The long-term effect of persistent ven-triculomegaly [present in about 30% of pa-tients (27)] on neuropsychological develop-mental is not fully known. Teo (65)highlights the importance of maintaining aclose vigilance by means of neuropsycho-logical testing on patients whose ventriclesremain large. He recommended shuntingthose individuals in whom a neuropsycho-logical deterioration is documented duringthe follow up, despite the absence of symp-toms and signs of acute intracranial hyper-tension.

ETV in Infants with Aqueductal StenosisThe role of age in predicting success of ETVis very controversial in the literature. Manyauthors claim that ETV results in infants areas good as those in adults, ranging from70% to 100%, and do not consider age to bea limiting factor in proposing ETV, even innewborns (3, 10, 26, 36, 73). Other authorsdescribe intermediate results, ranging from47% to 52% (2, 70-72), and still proposeETV as the first-line treatment in cases inwhich clinical and radiological featuresclearly show hydrocephalus of an obstruc-tive nature. Other authors have experiencedvery poor results (10% to 30% success rate)and consider an age younger than 12

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

4 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2012.02.007

175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232

AQ: 4

175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232

AQ: 5

T1

months as a contraindication to ETV (9, 41,44). These data should be considered criti-cally; in fact, when analyzing only patientswith primary aqueductal stenosis, the suc-cess rate generally corresponds to an aver-age of 50% to 70% with few exceptions (Ta-ble 2).

However, a trend toward a lower successrate in very young infants (under 6 monthsof age) is noticeable (Table 2). Wagner et al.(70) observed a higher tendency to form

new arachnoid membranes in younger pa-tients, thus producing obstruction of thestoma. An immature CSF absorption capac-ity by the arachnoid villi is also postulated asa possible hypothesis of ETV failure by oth-ers (36, 52).

In infants, a higher pressure gradient be-tween the subarachnoid space and the su-perior sagittal sinus, required to drive theCSF through the arachnoid villi during theadaptive period after ETV, is not possible. In

fact, increases in intracranial pressure maylead to an increase in head circumference, ifcranial sutures are open, lowering intracra-nial pressure. Lumbar taps in the postoper-ative period may gain time until reopeningof the arachnoid villi (12). Considering thatshunt surgery is burdened by a higher rateof failures and complications in very youngchildren (68), ETV should be preferred inwell-selected cases of noncommunicatinghydrocephalus in this age group.

Reobstruction of the StomaPrimary ETV may fail if the stoma becomesblocked by a clot, by debris, or by the devel-opment of new membranes (23). A repeatETV can be considered despite the failure ofthe first procedure if the patients had expe-rienced an excellent clinical response to thefirst ETV and have a favorable radiologicalpresentation (ETV not patent on specificflow magnetic resonance imaging [MRI]sequences and presence of typical anatomicdeformation of the third ventricle) (55). Re-opening or enlargement of the stoma car-ries the same success rate as the primarytreatment (!65%) (23, 43, 55, 62) andshould be preferred to shunt implantationin the first instance. The presence at surgeryof membranes and arachnoid adhesions inthe subarachnoid cisterns and an ageyounger than 2 years at first ETV are impor-tant predictors of failure of the second ETV(55).

Table 2. Results of ETV in Younger Children, Including Only Patients with IdiopathicAqueductal Stenosis

AuthorNumber of

Patients Age Successful ETV

Jones, 1994 15 "6 months 7 (47%)

Buxton et al., 1998 (9) 4 "1 year 2 (50%)

Cinalli et al., 1999 (10) 21 "6 months Same outcome (measured usingKaplan-Meier survival analysisscoring) as did children olderthan 6 months

Beems and Grotenhuis, 2002 (3) 16 "2 years 14 (87%)

Garayeb, 2004 11 "1 year 6 (55%)

Fritsch et al., 2005 (22) 3 "1 year 3 (100%)

Koch-Wiewrodt and Wagner, 2006 (44) 13 "1 year 7 (54%)

Baldauf et al., 2007 (2) 8 "2 years 4 (50%)

Faggin et al., 2009 (20) 5 "6 months 4 (80%)

Sufianov et al., 2010 (61) 2 "2 years 2 (100%)

Ogiwara, 2010 11 "6 months 4 (36%)

Table 1. Reported Series of ETV in Aqueductal Stenosis with Long-Term Follow-up (!3 years)

Author Number of Patients Age Mean Follow-Up (years) Overall Rate of Success

Cinalli et al., 1999 (10) 119 Children 6 72.00%

Tuli et al., 1999 (67) 32 Children 3.5 56.00%

Fukuhara et al., 2000 (24) 37 Adults and children 4.7 68.00%

Tisell et al., 2000 (66) 16 Adults 3 50.00%

Buxton et al., 2001 (8) 18 Adults 3 74.00%

Amini and Schmidt, 2005 (1) 32 Adults 3.7 72.00%

Bognar et al., 2005 (5) 76 Adults and children 3 71.00%

O’ Brien et al., 2005 (51) 54 Adults and children 3 69.00%

Gangemi et al., 2007 (25) 88 Adults and children 8.4 87.00%

Drake, 2007 107 Children 65% and 52% success rates at 1 and 5 years

Jenkinson et al., 2009 (38) 33 Adults 3 88.00%

Sacko et al., 2010 (58) 64 Adults and children 4 61.00%

Adapted with permission from Gangemi et al. (25).

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

WORLD NEUROSURGERY xx [x]: xxx, MONTH 2012 www.WORLDNEUROSURGERY.org 5

233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290

T2

AQ: 13

AQ: 10

AQ: 11

AQ: 12

233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290

AQ: 6

AQ: 7

Intellectual OutcomeOnly a few studies focusing on neuropsy-chological outcomes are available. Burt-scher et al. (7) in a prospective study evalu-ated 6 patients affected by late-onsetaqueductal stenosis using standardizedpsychometric testing preoperatively andpostoperatively. They found improvementof preoperative deficits (usually combineddeficits of memory and frontal-executivefunction) in all patients (full recovery in 2,good recovery in 3, and moderate recoveryin 1). Ventricular size diminished in all ofthese cases but never reached normal size.Hirsch et al. (30) reviewed 114 children af-fected by aqueductal stenosis, 70 treated byshunting, and 44 by percutaneous thirdventriculostomy: the postoperative IQ wasnot significantly different in the 2 groups.Sainte-Rose (59) compared 2 identical se-ries of patients with aqueductal stenosis: 38were treated by insertion of VPS and 30 bythird ventriculostomy. From neurological,endocrinological, social, and behavioralpoints of view, he found no statistical differ-ence between the 2 groups. Similar resultshave been achieved by Tuli et al. (67).

Takahashi (63) reviewed long-term out-comes and neurological development in chil-dren affected by obstructive hydrocephalus,treated with ETV when they were youngerthan 9 months. Takahashi (63) also relatedoutcomes with radiological appearance of hy-drocephalus at presentation. The study foundthat ETV was sufficiently effective andachieved normal development in patientswithout secondary brain damage and cerebralmorphology close to normal at MRI. Infantswho had poor cerebral development or sec-ondary brain damage underwent shuntingwithin 1 year after ETV because there was noappreciable improvement of development at 6months after the initial procedure.

ETV vs. ShuntThe main advantages of shunting are that itcan be used for all types of hydrocephalus, itis technically straightforward, and the mor-tality rate is very low. The benefits of ETVare that the mechanical problems of shunts,such as disconnection, occlusion, overd-rainage, and valve dysfunction, can beavoided. Also the risk of infection is signif-icantly lower. However, few series compareETV and VPS in obstructive hydrocephalus.

Recently de Ribaupierre et al. (15) ana-lyzed outcomes and failure rates in 2 groups

of pediatric patients with noncommunicat-ing hydrocephalus, 1 treated with ETV andthe other with VPS. They found in the groupof ETV patients fewer revisions and a largerrevision-free time. The majority of patientsin the ETV group underwent only 1 proce-dure, although some had a second proce-dure after closure of the first ventriculos-tomy. In the VPS group, even if a majorityunderwent only 1 procedure, some patientshad to undergo more than 10 operations. At5 years of follow-up, the failure rate of ETVwas 26%, as compared with 42% for the VPSgroup. This trend was also found in the pe-diatric series that the authors reviewed (27peer-reviewed articles, with more than 1500patients in each group). However, morbid-ity associated with the procedure washigher in the ETV group than in VPS. Theyconcluded that in carefully selected patientswith noncommunicating hydrocephalus,ETV can be considered the treatment ofchoice, but the family and the patientshould be alerted regarding the risk of lateobstruction.

Because traditionally neurosurgeons havereserved ETV for older patients with favor-able etiologies, Kulkarni et al. (45) per-formed a comparative analysis of ETV andshunts using advanced statistical methods(propensity score) to adjust for treatmentselection bias (age and origin of hydroceph-alus) in a large cohort of patients from dif-ferent international centers over a long pe-riod of time. They demonstrated that therelative risk of ETV failure is initially higherthan that for shunt, but after about 3months after surgery, the relative risk be-comes progressively lower for ETV. At 2years, the risk of ETV failure is roughly halfthe risk of shunt failure. Therefore, if pa-tients survive the early high-risk period ofETV failure, they could experience a long-term treatment survival advantage com-pared with patients receiving shunt. Theearly high risk of failure of ETV likely re-flects the selection of patients for whomETV was physiologically unsuitable.

In another study, Kulkarni et al. (46)compared quality of life in 2 groups of pa-tients treated with ETV or VPS. The patientswere between ages 5 and 18 years, and wereaffected by noncommunicating hydroceph-alus with no history of hemorrhage, menin-gitis, brain malformations, brain surgeryfor tumor resection, chemotherapy, and ra-diotherapy. The ETV group had fewer cases

requiring repeat surgery, had larger ventri-cle size at last follow-up, spent fewer days inthe hospital for CSF obstruction, and spentfewer days per year in the hospital for anyhydrocephalus-related complications. Theoutcome measures of quality of life [Hydro-cephalus Outcome Questionnaire (46)]were slightly lower in the ETV group. AlsoIQ scores, measured with the Wechsler In-telligence Scale for Children–Fourth Edi-tion (WISC-IV), were slightly lower in theETV group. However, none of these correla-tions was statistically significant. The au-thors concluded that there was no obviousdifference in quality of life between shuntand ETV patients, but also that because ofthe limited sample of the patients, largerand more detailed studies are needed.

UNCITED REFERENCEThis section consists of references that areincluded in the reference list but are notcited in the article text. Please either citeeach of these references in the text or, alter-natively, delete it from the reference list. Ifyou do not provide further instruction forthis reference, we will retain it in its currentform and publish it as an “un-cited refer-ence” with your article (4, 50).

REFERENCES

1. Amini A, Schmidt RH: Endoscopic third ventriculo-stomy in a series of 36 adult patients. NeurosurgFocus 19:E9, 2005.

2. Baldauf J, Oertel J, Gaab MR, Schroeder HW: Endo-scopic third ventriculostomy in children less thantwo years of age. Childs Nerv Syst 23:623-626, 2007.

3. Beems T, Grotenhuis A: Is the success rate of thirdventriculostomy age dependent? An analysis of theresults of endoscopic third ventriculostomy inyoung children. Childs Nerv Syst 18:605-608, 2002.

4. Bickers DS, Adams RD: Hereditary stenosis of aque-duct of Sylvius as a cause of congenital hydrocepha-lus. Brain 72:246-262, 1949.

5. Bognar L, Markia B, Novak L: Retrospective analysisof 400 neuroendoscopic interventions: the Hungar-ian experience. Neurosurg Focus 19:E10, 2005.

6. Brockmeyer D, Abtin K, Carey L, Walker M: Endo-scopic third ventriculostomy: an outcome analysis.Pediatr Neurosurg 28:236-240, 1998.

7. Burtscher J, Bartha L, Twerdy K, Eisner W, Benke T:Effect of endoscopic third ventriculostomy on neu-ropsychological outcome in late onset idiopathic

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

6 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2012.02.007

291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348

291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348

aqueductal stenosis: a prospective study. J NeurolNeurosurg Psychiatry 74:222-225, 2003.

8. Buxton N, Ho KJ, Macarthur D, Vloeberghs M, PuntJ, Robertson I: Neuroendoscopic third ventriculos-tomy for hydrocephalus in adults: report of a singleunit’s experience with 63 cases. Surg Neurol 55:74-78, 2001.

9. Buxton N, Macarthur D, Mallucci C, Punt J, Vloe-berghs M: Neuroendoscopic third ventriculostomyin patients less than 1 year old. Pediatr Neurosurg29:73-76, 1998.

10. Cinalli G, Sainte-Rose C, Chumas P, Zerah M,Brunelle F, Lot G, Pierre-Kahn A, Renier D: Failureof third ventriculostomy in the treatment of aque-ductal stenosis in children. J Neurosurg 90:448-454,1999.

11. Cinalli G, Salazar C, Mallucci C, Yada JZ, Zerah M,Sainte-Rose C: The role of third ventriculostomy inthe management of shunt malfunction. Neurosur-gery 43:1323-1329, 1998.

12. Cinalli G, Spennato P, Ruggiero C, Aliberti F, ZerahM, Trischitta V, Cianciulli E, Maggi G: Intracranialpressure monitoring and lumbar puncture after en-doscopic third ventriculostomy in children. Neuro-surgery 58:126-136, 2006.

13. Cinalli G, Spennato P, Cianciulli E, D’Armiento M:Hydrocephalus and aqueductal stenosis. In: CinalliG, Maixner WJ, Saint-Rose C, eds. Pediatric Hydro-cephalus. Milan: Springer-Verlag; 2004:279-293.

14. Czosnyka M, Czosnyka ZH, Whitfield PC, et al: Ce-rebrospinal fluid dynamics. In: Cinalli C, MaixnerWJ, Sainte-Rose C, eds. Pediatric Hydrocephalus.Milan: Springer-Verlag; 2004:47-63.

15. de Ribaupierre S, Rilliet B, Vernet O, Regli L, Vill-emure J-G: Third ventriculostomy vs ventriculoperi-toneal shunt in pediatric obstructive hydrocephalus:results from a Swiss series and literature review.Childs Nerv Syst 23:527-533, 2007.

16. Decq P, Leguerinel C, Palfi S, Djindjian M, Keravel Y,Nguyen JP: A new device for endoscopic third ven-triculostomy. J Neurosurg 93:509-512, 2000.

17. Decq P: Endoscopic anatomy of the ventricles. In:Cinalli G, Maixner WJ, Saint-Rose C, eds. PediatricHydrocephalus. Milan: Springer-Verlag; 2004:351-359.

18. Di Rocco C, Cinalli G, Massimi L, Spennato P, Cian-ciulli E, Tamburrini G: Endoscopic third ventriculo-stomy in the treatment of hydrocephalus in pediatricpatients. Adv Tech Stand Neurosurg 31:119-219,2006.

19. Drake JM: Ventriculostomy for treatment of hydro-cephalus. Neurosurg Clin North Am 4:657-666,1993.

20. Faggin R, Bernardo A, Stieg P, Perilongo G, d’AvellaD: Hydrocephalus in infants less than six months ofage: effectiveness of endoscopic third ventriculos-tomy. Eur J Pediatr Surg 19:216-219, 2009.

21. Feng H, Huang G, Liao X, Fu K, Tab H, Pu H, ChengY, Liu W, Zhao D: Endoscopic third ventriculostomyin the management of obstructive hydrocephalus:

an outcome analysis. J Neurosurg 100:626-633,2004.

22. Fritsch MJ, Kienke S, Ankermann T, Padoin M, Me-hdorn HM: Endoscopic third ventriculostomy in in-fants. J Neurosurg 103(suppl):50-53, 2005.

23. Fukuhara T, Luciano M, Kowalski RJ: Clinical fea-tures of third ventriculostomy failures classified byfenestration patency. Surg Neurol 58:102-110, 2002.

24. Fukuhara T, Vorster SJ, Luciano MG: Risk factorsfor failure of endoscopic third ventriculostomy forobstructive hydrocephalus. Neurosurgery 46:1100-1111, 2000.

25. Gangemi M, Mascari C, Maiuri F, Godano U, DonatiP, Longatti PL: Long-term outcome of endoscopicthird ventriculostomy in obstructive hydrocephalus.Minim Invasive Neurosurg 50:265-269, 2007.

26. Gorayeb RP, Cavalheiro S, Zymberg SY: Endoscopicthird ventriculostomy in children younger than 1year of age. J Neurosurg 100:427-429, 2004.

27. Goumnerova LC, Frim D: Treatment of hydroceph-alus with third ventriculostomy: outcome and CSFflow patterns. Pediatr Neurosurg 27:149-152, 1997.

28. Greitz D: Radiological assessment of hydrocepha-lus: new theories and implications for therapy. Neu-rosurg Rev 27:145-165, 2004.

29. Hellwig D, Heinemann A, Riegel T: Endoscopicthird ventriculostomy in treatment of obstructivehydrocephalus caused by primary aqueductal steno-sis. In: Hellwig D, Bauer B, eds. Minimally InvasiveTechniques for Neurosurgery. Berlin: Springer;1998:65-72.

30. Hirsch JF, Hirsch E, Sainte-Rose C, Renier D, Pierre-Kahn A: Stenosis of the aqueduct of Sylvius (etiologyand treatment). J Neurosurg Sci 30:29-39, 1986.

31. Hirsch JF: Percutaneous ventriculocisternostomiesin non-communicating hydrocephalus. MonogrNeurol Sci 8:170-178, 1982.

32. Høffman HJ, Harwood-Nash D, Gilday DL: Percuta-neous third ventriculostomy in the management ofnon-communicating hydrocephalus. Neurosurgery7:313-321, 1980.

33. Hopf NJ, Grunert P, Fries G, Resch KD, PerneczkyA: Endoscopic third ventriculostomy: outcomeanalysis of 100 consecutive procedures. Neurosur-gery 44:795-804, 1999.

34. Jacobson EE, Fletcher DF, Morgan MK, JohnstonIH: Computer modelling of the CSF flow dynamicsof aqueductal stenosis. Med Biol Eng Comput 37:59-63, 1999.

35. Jacobson EE, Fletcher DF, Morgan MK, JohnstonIH: Fluid dynamics of the cerebral aqueduct. PediatrNeurosurg 24:229-236, 1996.

36. Javadpour M, Mallucci C, Brodbelt A, Golash A, MayP: The impact of endoscopic third ventriculostomyon the management of newly diagnosed hydroceph-alus in infants. Pediatr Neurosurg 35:131-135, 2001.

37. Jellinger G: Anatomopathology of nontumoral aq-ueductal stenosis. J Neurosurg Sci 30:1-16, 1986.

38. Jenkinson MD, Hayhurst C, Al-Jumaily M, Kan-dasamy J, Clark S, Mallucci CL: The role of endo-scopic third ventriculostomy in adult patients withhydrocephalus. J Neurosurg 110:861-866, 2009.

39. Jones RF, Kwok BC, Stening WA, Vonau M: Thirdventriculostomy in patients with spinal dysraphism.Indications and contraindications. Eur J PediatrSurg 6(suppl):5-6, 1996.

40. Jones RFC, Stening WA, Brydon M: Endoscopicthird ventriculostomy. Neurosurgery 26:86-92,1990.

41. Kadrian D, van Gelder J, Florida D, Jones R, VonauM, Teo C, Stening W, Kwok B: Long-term reliabilityof endoscopic third ventriculostomy. Neurosurgery56:1272–1278, 2005.

42. Karachi C, Le Guérinel C, Brugières P, Melon E,Decq P: Hydrocephalus due to idiopathic stenosis ofthe foramina of Magendie and Luschka. Report ofthree cases. J Neurosurg 98:897-902, 2003.

43. Koch D, Grunert P, Filippi R, Hopf N: Re-ventricu-lostomy for treatment of obstructive hydrocephalusin cases of stoma dysfunction. Minim Invasive Neu-rosurg 45:158-163, 2002.

44. Koch-Wiewrodt D, Wagner W: Success and failureof endoscopic third ventriculostomy in young in-fants: are there different age distributions? ChildsNerv Syst 22:1537-1541, 2006.

45. Kulkarni AV, Drake JM, Kestle JR, Mallucci CL,Sgouros S, Constantini S, Canadian Pediatric Neu-rosurgery Study Group: Endoscopic third ventricu-lostomy vs cerebrospinal fluid shunt in the treat-ment of hydrocephalus in children: a propensityscore-adjusted analysis. Neurosurgery 67:588-593,2010.

46. Kulkarni AV, Hui S, Shams I, Donnelly R: Quality oflife in obstructive hydrocephalus: endoscopic thirdventriculostomy compared to cerebrospinal fluidshunt. Childs Nerv Syst 26:75-79, 2010.

47. Lapras C, Bret P, Patet JD, Huppert J, Honorato D:Hydrocephalus and aqueductal stenosis. Direct sur-gical treatment by interventriculostomy (aqueductcannulation). J Neurosurg Sci 30:47-53, 1986.

48. McFarlane A, Maloney AFJ: The appearance of aque-duct and its relationship with hydrocephalus in theArnold-Chiari malformation. Brain 80:479-491,1957.

49. Mise B, Klarica M, Seiwerth S, Bulat M: Experimen-tal hydrocephalus and hydromyelia: a new insight inmechanism of their development. Acta Neurochir138:862-869, 1996.

50. Murshid WR: Endoscopic third ventriculostomy: to-wards more indications for the treatment of non-communicating hydrocephalus. Minim InvasiveNeurosurg 43:75-82, 2000.

51. O’Brien DF, Javadpour M, Collins DR, Spennato P,Mallucci CL: Endoscopic third ventriculostomy: anoutcome analysis of primary cases and proceduresperformed after ventriculoperitoneal shunt mal-function. J Neurosurg 103(5 suppl):393-400, 2005.

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

WORLD NEUROSURGERY xx [x]: xxx, MONTH 2012 www.WORLDNEUROSURGERY.org 7

349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406

AQ: 8

349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406

52. Oi S, Di Rocco C: Proposal of “evolution theory incerebrospinal fluid dynamics” and minor pathwayhydrocephalus in developing immature brain.Childs Nerv Syst 22:662-669, 2006.

53. Oi S, Shimoda M, Shibata M, Honda Y, Togo K,Shinoda M, Tsugane R, Sato O: Pathophysiology oflong-standing overt ventriculomegaly in adults. JNeurosurg 92:933-940, 2000.

54. Oka K, Yamamoto M, Ikeda K, Tomonaga M: Flexi-ble endoneurosurgical therapy for aqueductal ste-nosis. Neurosurgery 33:236-243, 1993.

55. Peretta P, Cinalli G, Spennato P, Ragazzi P, Rug-giero C, Aliberti F, Carlino C, Cianciulli E: Long-term results of a second endoscopic third ventricu-lostomy in children: retrospective analysis of 40cases. Neurosurgery 65:539-547, 2009.

56. Quencer RM, Post MJ, Hinks RS: Cine MR in theevaluation of normal and abnormal CSF flow: intra-cranial and intraspinal studies. Neuroradiology 32:371-391, 1990.

57. Russell DS: Observations on the pathology of hy-drocephalus. Medical Res Counsil, spec rep seriesNo. 265. His Majesty’s Stationery Office, London,1949.

58. Sacko O, Boetto S, Lauwers-Cances V, Dupuy M,Roux FE: Endoscopic third ventriculostomy: out-come analysis in 368 procedures. J Neurosurg Pedi-atr 5:68-74, 2010.

59. Sainte-Rose C: Third ventriculostomy. In: Manwar-ing KH, Crone KR, eds: Neuroendoscopy. NewYork: Mary Ann Liebert; 1992:47-62.

60. Siomin V, Cinalli G, Grotenhuis A, Golash A, Oi S,Kothbauer K, Weiner H, Roth J, Beni-Adani L,Pierre-Kahn A, Takahashi Y, Mallucci C, Abbott R,

Wisoff J, Constantini S: Endoscopic thirdventriculostomy in patients with cerebrospinal fluidinfection and/or hemorrhage. J Neurosurg 97:519-524, 2002.

61. Sufianov AA, Sufianova GZ, Iakimov IA: Endoscopicthird ventriculostomy in patients younger than 2years: outcome analysis of 41 hydrocephalus cases. JNeurosurg Pediatrics 5:392-401, 2010.

62. Surash S, Chumas P, Bhargava D, Crimmins D,Straiton J, Tyagi A: A retrospective analysis of revi-sion endoscopic third ventriculostomy. Childs NervSyst 26:1693-1698, 2010.

63. Takahashi Y: Long-term outcome and neurologicdevelopment after endoscopic third ventriculos-tomy versus shunting during infancy. Childs NervSyst 22:1591-1602, 2006.

64. Teo C, Jones R: Management of hydrocephalus byendoscopic third ventriculostomy in patients withmyelomeningocele. Pediatr Neurosurg 25:57-63,1996.

65. Teo C: Third ventriculostomy in the treatment ofhydrocephalus: experience with more than 120cases. In: Hellwig D, Bauer B, eds: Minimally inva-sive techniques for neurosurgery. Berlin: Springer;1998:73-76.

66. Tisell M, Almstrom O, Stephensen H, Tullberg M,Wikkelso C: How effective is endoscopic third ven-triculostomy in treating adult hydrocephalus causedby primary aqueductal stenosis? Neurosurgery 46:104-110, 2000.

67. Tuli S, Alshail E, Drake JM: Third ventriculostomyversus cerebrospinal fluid shunt as a first procedurein pediatric hydrocephalus. Pediatr Neurosurg 30:11-15, 1999.

68. Tuli S, Drake J, Lawless J, Wigg M, Lamberti-Pas-culli M: Risk factors for repeated cerebrospinalshunt failures in pediatric patients with hydroceph-alus. J Neurosurg 92:31-38, 2000.

69. Vries JK, Friedmann WA: A quantitative assessmentof CSF absorption in infants with meningomyelo-cele. Surg Neurol 13:38-40, 1980.

70. Wagner W, Koch D: Mechanisms of failure afterendoscopic third ventriculostomy in young infants.J Neurosurg 103:43-49, 2005.

71. Warf BC: Hydrocephalus in Uganda: the predomi-nance of infectious origin and primary managementwith endoscopic third ventriculostomy. J Neurosurg102:1-15, 2005.

72. Warf BC: Comparison of endoscopic third ventricu-lostomy alone and combined with choroid plexuscauterization in infants younger than 1 year of age: aprospective study in 550 African children. J Neuro-surg 103(6 suppl):475-481, 2005.

73. Yadav YR, Jaiswal S, Adam N, Basoor A, Jain G:Endoscopic third ventriculostomy in infants. Neu-rol India 54:161-163, 2006.

Conflict of interest statement: The authors declare that thearticle content was composed in the absence of anycommercial or financial relationships that could beconstrued as a potential conflict of interest.

received 16 August 2011; accepted 02 February 2012

Citation: World Neurosurg. (2012).DOI: 10.1016/j.wneu.2012.02.007

Journal homepage: www.WORLDNEUROSURGERY.org

Available online: www.sciencedirect.com

1878-8750/$ - see front matter © 2012 Elsevier Inc.All rights reserved.

PEER-REVIEW REPORTS

PIETRO SPENNATO ET AL. ETV FOR IDIOPATHIC AQUEDUCTAL STENOSIS

8 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, DOI:10.1016/j.wneu.2012.02.007

407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464

407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464

AQ: 9