Literature Review
Endoscopic Third Ventriculostomy versus Ventriculoperitoneal Shunt in Patients with
Obstructive Hydrocephalus: Meta-Analysis of Randomized Controlled Trials
Liang Lu, Hongwu Chen, Shaotao Weng, Yimin Xu
Key words
- Endoscopic third ventriculostomy
- Obstructive hydrocephalus
- Ventriculoperitoneal shunt
Abbreviations and Acronyms
CI: 95% Confidence interval
CSF: Cerebrospinal fluid
ETV: Endoscopic third ventriculostomy
OR: Odds ratio
RR: Risk ratio
SF: Cerebrospinal fluid
VPS: Ventriculoperitoneal shunt
Department of Neurosurgery, The First Affiliated Hospital of
Shantou University Medical College, Shantou, Guangdong,
PR China
To whom correspondence should be addressed:
Yimin Xu, M.D., Ph.D.
[E-mail: xuym777@163.com]
Citation: World Neurosurg. (2019) 129:334-340.
https://doi.org/10.1016/j.wneu.2019.04.255
Journal homepage: www.journals.elsevier.com/world-
neurosurgery
Available online: www.sciencedirect.com
1878-8750/$ - see front matter ª 2019 Published by Elsevier
Inc.
INTRODUCTION
Hydrocephalus can occur at any age but is
most common in infants and adults aged
60 and older. This health condition is
characterized by excessive accumulation of
cerebrospinal fluid (CSF), leading to ven-
tricular dilatation and increased intracra-
nial pressure, and may be accompanied by
secondary brain parenchymal atrophy.
Ventriculoperitoneal shunt (VPS) remains
the classical treatment for hydrocephalus
due to its relatively high safety features.
Other modes such as ventricle atrial shunt
and lateral ventricleecisterna magna
shunt have rarely been used.1,2
Recently, endoscopic third ven-
triculostomy (ETV) was deemed to be
indispensable for treating obstructive
hydrocephalus, on account of minimal
complications because it avoids foreign-
body associated infections. First reported
by Mixter in 1923 for hydrocephalus,3 the use
of ETV in routine practice before the 1990s
334 www.SCIENCEDIRECT.com
- OBJECTIVE: Endoscopic third ventriculostomy (ETV) and ventriculoperitoneal
shunt (VPS) are 2 surgical methods used for the treatment of obstructive hy-
drocephalus. However, the efficacy and safety of these 2 procedures are un-
known. Therefore we conducted a meta-analysis to compare the safety and
efficiency of ETV and VPS in patients with obstructive hydrocephalus.
- METHODS: In January 2019, a comprehensive search strategy of 3 electronic
databases was initiated. A systematic search from database inception to
December 30, 2018, on clinical outcome, safety, and efficiency of ETV and VPS in
the treatment of obstructive hydrocephalus was conducted. The following
electronic databases were searched: PubMed, Embase, and Cochrane Library for
related randomized controlled trials (RCTs). Main outcomes included post-
operative hematoma, postoperative infection, postoperative cerebrospinal fluid
leakage, blockage rate, and mortality.
- RESULTS: Of the 546 studies identified, 4 RCTs involving 250 patients met our
inclusion criteria. The pooled results show that ETV was associated with lower
incidence of postoperative infection (risk ratio [RR] 0.09, 95% confidence interval
[CI]: 0.02e0.32, P [ 0.0002); postoperative hematoma (RR 0.26, 95% CI: 0.08e0.88,
P [ 0.03); and blockage rate (RR 0.28, 95% CI: 0.13e0.60, P [ 0.001) compared
with VPS. Compared with VPS, ETV had no significant effect on incidence rate of
postoperative cerebrospinal fluid leakage (RR 2.00, 95% CI: 0.30e13.16, P [ 0.47)
and mortality rates (RR 0.19, 95% CI: 0.03e1.09, P [ 0.06), but there have been no
deaths in patients treated with ETV.
- CONCLUSIONS: On the basis of the meta-analysis of RCTs evaluating ETV and
VPS, the incidence of complications and mortality was higher with the VPS
procedure, and therefore greater benefits can be achieved using ETV.
was rather limited due to the lack of meta-analysis of all randomized controlled
effective endoscopic and relevant surgical trials (RCTs) to compare clinical outcomes,
instruments, leading to poor relief from major complications, and mortality of these
hydrocephalus and high mortality rates.4 2 techniques in patients with obstructive
However, rapid advances in neuroimaging, hydrocephalus. It is expected that this study
the endoscopic manufacturing process, will provide guidance for future studies in
operation instruments, and a stereotaxic the treatment of hydrocephalus.
neuronavigation system in recent
years have led to widespread use of
MATERIALS AND METHODS
endoscopy for various procedures including
obstructive hydrocephalus.5,6 Several The methodology adopted in this study is
studies have examined the efficiency and in accordance with Preferred Reporting
safety of VPS and ETV for the treatment of Items for Systematic Reviews and Meta-
obstructive hydrocephalus. Nevertheless, it Analysis guidelines.7
remains unclear as to which of these 2 is
more beneficial to patients because Search Strategy
both methods have advantages and A comprehensive search strategy was
disadvantages. Therefore we conducted a developed wherein 3 electronic databases
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.04.255
 LITERATURE REVIEW
LIANG LU ET AL. ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS

were conducted to identify eligible studies in

the EndNote software program.

Study Selection Criteria

Figure 1. Flow chart of search and study selection. In total, 546 records were
identified from 3 databases. After screening, 4 randomized controlled trials
that fulfilled all the eligibility criteria remained.
were searched. Searches were from database
inception to December 30, 2018. The
following electronic databases were
searched: PubMed, Embase, and the
Cochrane Library. Search terms included
medical subject headings [MeSH]/EMTREE
terms and/or text words): 1) for patients,
“Obstructive Hydrocephalus”[MeSH]; and
2) for intervention, “Ventriculoperitoneal
Shunt” [MeSH] and “Third Ven-
triculostomy”[MeSH]. Manual searches of
reference lists from all the relevant studies
Studies were selected for inclusion in the
meta-analyses of the clinical effectiveness
of the 2 procedures—VPS and ETV—
through predefined and explicit criteria.
The full literature search results were
independently screened by 2 investigators
(L.L. and H.C.) to identify all citations that
possibly met the inclusion criteria, fol-
lowed by independent extraction of data
and quality assessment.
All English-language articles of RCTs
comparing the efficacy and safety of ETV
or VPS for patients with obstructive hy-
drocephalus were selected for the meta-
analysis. Our inclusion criteria were as
follows: 1) studies with randomized
controlled design; 2) all included patients
had obstructive hydrocephalus; and 3)
patients received ETV or VPS. Exclusion
criteria were as follows: 1) preclinical
study; 2) case reports, letters, review arti-
cles, and any other stylistic feature that
cannot provide complete information for
trials; 3) not all patients suffered from
obstructive hydrocephalus; 4) not ran-
domized clinical trial; 5) noneEnglish
language publications; and 6) studies
with the same patient cohorts as in pre-
vious trials. For articles with relevant ab-
stracts, the full-text versions were

Table 1. Characteristics of Included Studies

Intervention Outcomes Measured
Number of Number of
Study Patients Age (years) Males (%) Follow-up
Study Design ETV/VPS ETV/VPS* ETV/VPS Operation Duration Clinical Outcomes Complication

Kamikawa RCT 44/44 4.47
0.73/4.48
23 (52.3)/25 ETV/VPS 24 Months Revision rates, major Bleeding, infection,
et al., 200110 0.86 (56.8) complications blockage, damage to brain
tissue
El-Ghandour, RCT 32/21 6.5
2.1/7.2
2.6 18 (56.3)/12 ETV/VPS 27.4/25 Improvement of symptoms, Bleeding, infection,
201111 (57.1) months postoperative CT, major blockage, CSF leakage
complications, procedure
failure, death
Navaei, et al., RCT 22/27 3.91
3.34/3.60
13 (59.1)/19 ETV/VPS 36 months Major complications, Bleeding, infection, raised
201812 2.95 (months) (70.4) procedure failure, death, intracranial pressure
estimated survival
Rahman, RCT 30/30 18.29
19.74/17.24 NR ETV/VPS 30 days Improvement of symptoms, Bleeding, infection,
et al., 201813
18.56 major complications, blockage, CSF leakage,
procedure failure, death pseudomeningocele

ETV, endoscopic third ventriculostomy; VPS, ventriculoperitoneal shunt; RCT, randomized controlled trial; CT, computed tomography; CSF, cerebrospinal fluid; NR, not reported.
*Age was reported as mean
standard deviation in all of the studies.

WORLD NEUROSURGERY 129: 334-340, SEPTEMBER 2019 www.journals.elsevier.com/world-neurosurgery 335

 LITERATURE REVIEW
LIANG LU ET AL. ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS

ratio (RR) or odds ratio (OR) with a 95%

El-Ghandour, N. M.2011
confidence interval (CI) for dichotomous
Rahman, M. M.2018 variables to reflect event probabilities.

Kamikawa, S.2001
Navaei, A. A.2018
Heterogeneity was assessed by I2 and c2
test metric. A fixed-effect model was
applied when I2 < 50%; otherwise, a
random-effects model was used.9

Random sequence generation (selection bias)

+

RESULTS
Allocation concealment (selection bias)
+

Study Selection
Blinding of participants and personnel (performance bias) A total of 546 publications were identified
+

through our initial database search. After

Blinding of outcome assessment (detection bias)
+

screening the titles, abstracts, and key-

Incomplete outcome data (attrition bias) words and removal of duplicates, 532 ar-
+

ticles were excluded according to our

Selective reporting (reporting bias)
+

prescribed criteria. Fourteen studies were

Other bias considered for the perusal of full text, and
+

4 RCTs10-13 were included for a detailed

Figure 2. Risk of bias summary. assessment. The Preferred Reporting
Items for Systematic Reviews and Meta-
Analysis flow diagram of the publication
obtained and assessed independently on Washington, USA). Surgical success was selection is presented in Figure 1.
the basis of criteria mentioned previously. defined as partial or complete relief of
symptoms or need for no further surgery. Study Characteristics
Table 1 summarizes the main
Data Extraction characteristics of the included studies.10-13
Two reviewers independently extracted the Quality Assessment In total, 250 patients (ETV group: 128; VPS
relevant information according to a pre- The Cochrane Collaboration tool was used group: 122) were included. The mean or
pared form that included authors, publi- to assess the risk of biases8—selection median age of patients ranged from 3.6
cation years, study design, number of bias (random sequence generation and months to 16.3 years. Of the 4 RCTs
patients per group (intervention and con- allocation concealment), performance, included, information on patients’ sex was
trol groups), participants’ age and sex, detection, attrition, reporting bias, and missing in 1,13 while the remaining 3
surgical procedures, follow-up duration, others. Two reviewers (L.L. and H.C.) studies were predominantly male (57.9%).
number of successful cases, mortality, and independently assessed the quality of the Three studies11-13 provided a specific
complications. We focused on the number included RCTs. Any disagreement was number of successful operations and the
of postoperative complications such as resolved by discussion and consultation total number of complications except for
infection, hematoma, CSF leakage, and in order to avoid errors in the process. the study by Kamikawa in 2001.10 The
blockage, referring to stoma malformation deaths of 6 patients, all of whom received
in the ETV group or shunt blockage in the Statistical Analysis VPS, were noted in all studies. Adverse
VPS group after operation. Data extracted All statistical analysis was conducted by effects were recorded: 28 patients with
were distilled into an Excel database Rev-Man software (version 5.3). Results blockage in 3 RCTs10,11,13 (ETV group: 6;
(Microsoft Excel; Microsoft, Redmond, from the RCTs were presented as the risk VPS group: 22); 4 patients with CSF

ETV VPS Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
El-Ghandour, N. M.2011 0 32 2 21 10.8% 0.13 [0.01, 2.65]
Kamikawa, S.2001 0 44 14 44 52.2% 0.03 [0.00, 0.56]
Navaei, A. A.2018 0 22 2 27 8.1% 0.24 [0.01, 4.82]
Rahman, M. M.2018 1 30 8 30 28.8% 0.13 [0.02, 0.94]

Total (95% CI) 128 122 100.0% 0.09 [0.02, 0.32]

Total events 1 26
Heterogeneity: Chi² = 1.07, df = 3 (P = 0.78); I² = 0%
0.01 0.1 1 10 100
Test for overall effect: Z = 3.73 (P = 0.0002)
Favours [ETV] Favours [VPS]

Figure 3. Forest plot of postoperative infection rates for shunt for treating obstructive hydrocephalus.
endoscopic third ventriculostomy and ventriculoperitoneal

336 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.04.255

 LITERATURE REVIEW
LIANG LU ET AL. ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS

ETV VPS Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
El-Ghandour, N. M.2011 2 32 3 21 30.5% 0.44 [0.08, 2.40]
Kamikawa, S.2001 0 44 3 44 29.5% 0.14 [0.01, 2.69]
Navaei, A. A.2018 0 22 2 27 19.0% 0.24 [0.01, 4.82]
Rahman, M. M.2018 0 30 2 30 21.0% 0.20 [0.01, 4.00]

Total (95% CI) 128 122 100.0% 0.26 [0.08, 0.88]

Total events 2 10
Heterogeneity: Chi² = 0.54, df = 3 (P = 0.91); I² = 0%
0.01 0.1 1 10 100
Test for overall effect: Z = 2.17 (P = 0.03)
Favours [ETV] Favours [VPS]

Figure 4. Forest plot of postoperative hematoma rates for shunt for treating obstructive hydrocephalus.
endoscopic third ventriculostomy and ventriculoperitoneal

leakage in 2 RCTs11,13 (ETV group: 3; VPS
group: 1); 12 patients (ETV group: 2; VPS
group: 10) with postoperative hematoma;
and 27 patients (ETV group: 1; VPS group:
26) with postoperative infection.
Quality Assessment
Assessment of bias risk for included
studies was conducted according to the
Cochrane Handbook for Systematic Re-
views of Intervention (version 5.0.2):
Criteria for judging the risk of bias in the
“risk of bias” assessment tool. Randomi-
zation methods were described in all 4
RCTs. Because it was impossible to blind
key study personnel and participants
because patients had to provide written
informed consent before surgery, we
judged that blinding is unlikely to influ-
ence the outcome. One of the RCTs13 was
of high quality with a low risk of biases in
all domains, while it was unclear in the
remaining 310-12 because these RCTs
lacked information on random sequence
generation and allocation concealment.
Reporting bias was of high risk in
Kamikawa10 without results from
successful cases. Quality assessments for
the eligible studies are shown in Figure 2.
Postoperative Infection
Studies10-13 that reported dates of post-
operative infection were included in the
analysis. In the ETV group, only 1 patient
with postoperative infection was re-
ported13 and those in the VPS group
numbered 26 and included ventriculitis
and peritonitis. The pooled data showed
significantly lower postoperative infection
rates in the ETV group than VPS group
(RR 0.09, 95% CI: 0.02e0.32, P ¼
0.0002) without heterogeneity (I2 ¼ 0%,
P ¼ 0.78) (Figure 3).
Postoperative Hematoma
All RCTs reported data on postoperative
hematoma,10-13 and the incidence of
postoperative hematoma was 1.6% in the
ETV group and 8.2% in the VPS group,
without heterogeneity in the studies (I2 ¼
0%, P ¼ 0.91). Postoperative hematoma
incidence rates were significantly lower in
the ETV group than in the VPS group (RR
0.26, 95% CI: 0.08e0.88, P ¼ 0.03)
(Figure 4).
Blockage Rate
Three10,11,13 of the included RCTs provided
data on the blockage. There were 22 cases
of shunt blockage in the VPS group and 6
cases with stoma malformation in the ETV
group, with a significant difference be-
tween the 2 groups (RR 0.28, 95% CI:
0.13e0.60, P ¼ 0.001) and a moderate-
level heterogeneity (I2 ¼ 44%, P ¼ 0.17)
(Figure 5).
Postoperative Cerebrospinal Fluid
Leakage
Two studies11,13 reported data on post-
operative CSF leakage for the 2 groups,
and 4 of 113 cases were identified. Three of
the 4 were treated by ETV and 1 by VPS.
The pooled results showed that CSF
leakage was not significantly different be-
tween groups (RR 2.00, 95% CI: 0.30e
13.16, P ¼ 0.47) without heterogeneity
(I2 ¼ 0%, P ¼ 1.00) (Figure 6).
Mortality
All 4 RCTs reported data on mortality,10-13
but only VPS groups had associated mor-
tality. When mortality data were pooled,
there was no significant difference for
overall mortality between ETV and VPS
treatment in patients with obstructive hy-
drocephalus (RR 0.19, 95% CI: 0.03e1.09,
P ¼ 0.06) without heterogeneity (I2 ¼ 0%;
P ¼ 0.96) (Figure 7).

ETV VPS Risk Ratio Risk Ratio

Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
El-Ghandour, N. M.2011 0 32 6 21 32.1% 0.05 [0.00, 0.87]
Kamikawa, S.2001 0 44 4 44 18.5% 0.11 [0.01, 2.00]
Rahman, M. M.2018 6 30 12 30 49.4% 0.50 [0.22, 1.16]

Total (95% CI) 106 95 100.0% 0.28 [0.13, 0.60]

Total events 6 22
Heterogeneity: Chi² = 3.56, df = 2 (P = 0.17); I² = 44%
0.01 0.1 1 10 100
Test for overall effect: Z = 3.29 (P = 0.001)
Favours [ETV] Favours [VPS]

Figure 5. Forest plot of blockage rates for endoscopic third ventriculostomy and ventriculoperitoneal shunt for treating obstructive
hydrocephalus.

WORLD NEUROSURGERY 129: 334-340, SEPTEMBER 2019 www.journals.elsevier.com/world-neurosurgery 337

 LITERATURE REVIEW
LIANG LU ET AL. ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS

E TV VPS R i s k R a ti o R i s k R a ti o
Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
Rahman, M. M.2018 2 30 1 30 62.5% 2.00 [0.19, 20.90]
El-Ghandour, N. M.2011 1 32 0 21 37.5% 2.00 [0.09, 46.90]

Total (95% CI) 62 51 100.0% 2.00 [0.30, 13.16]

Total events 3 1
Heterogeneity: Chi² = 0.00, df = 1 (P = 1.00); I² = 0%
0.01 0.1 1 10 100
Test for overall effect: Z = 0.72 (P = 0.47)
Favours [ETV] Favours [VPS]

Figure 6. Forest plot of postoperative cerebrospinal fluid ventriculoperitoneal shunt for treating obstructive
leakage rates for endoscopic third ventriculostomy and hydrocephalus.

Postoperative Success
Three studies11-13 provided a specific
number of successful surgeries (118 of 162
patients), but not the study by Kami-
kawa.10 The rate of success was 72.84%,
which was close (75%) to the study by
Kamikawa.10 The pooled data showed
that there was no significant difference
in the rate of success between ETV and
VPS (RR 1.30, 95% CI: 0.72e2.37, P ¼
0.38). However, the chi-squared test
showed that the studies had a consider-
able degree of heterogeneity (I2 ¼ 87%;
P ¼ 0.0006) (Figure 8).
DISCUSSION
Hydrocephalus is due to the accumulation
of an excessive amount of CSF within the
cerebral ventricles and/or subarachnoid
space leading to their dilation. The
disturbance of CSF formation, flow, or
absorption on account of obstacles,
congenital or acquired, leads to an in-
crease in overall CSF volume in the central
nervous system. As well, premature com-
plications such as intraventricular hemor-
rhage and diseases such as tumors,
meningitis, traumatic head injury, or
subarachnoid hemorrhage can also block
CSF circulation.14,15 With recent advances
in endoscopy, ETV has become an indis-
pensable tool for obtaining relief from
obstructive hydrocephalus. Nevertheless,
comparisons between ETV and VPS in the
domains of symptom improvement and
postoperative complication are lacking.
Although a previous comparative analysis
was observational in nature, we report for
the first time a meta-analysis of all ETV-
and VPS-related RCT studies. The overall
aim of the meta-analysis was to enable
physicians with guidance for future
studies in treatment for hydrocephalus
and to investigate the safety and efficiency
of ETV and VPS in patients with obstruc-
tive hydrocephalus. Although current
RCTs do not recommend ETV vs. VPS, our
meta-analysis showed that ETV could
effectively reduce the incidence of com-
plications and mortality compared with
VPS.
Despite its frequent use in the treatment
for hydrocephalus, VPS has a higher
complication rate than ETV. There were 26
cases of infection in the VPS group
compared with 1 in the ETV group in all 4
RCTs, which was significantly different.
Strikingly, 6 patients died after VPS sur-
gery, 5 from shunt infection and severe
sepsis, while there was no mortality after
ETV. Few plausible explanations for this
difference are the larger surgical area of
VPS compared with that of ETV, as well as
the use of implantation materials in VPS,
which presents a higher risk of bacterial
infection. In addition, the patients’ own
immune status and surgical process may
have an impact on postoperative infec-
tion.16 Thus postoperative infection is the
most important fatal complication of
surgery in patients with obstructive
hydrocephalus.
Both shunt blockage and stoma mal-
formation obstruct CSF drainage, and our
pooled data showed significantly lower
blockage rates in the ETV group than in
the VPS group. As the most common
complication, shunt blockage can easily
lead to the failure of VPS, and severe cases
require shunt replacement twice or even
more. The lack of detailed information has
impeded further analysis into the reason
for shunt blockage in our study; however,
literature reports on the causes of
blockage cite 1) the number of cells with
high protein content in CSF enables the
deposit of sediment easily; 2) intraven-
tricular choroid plexus cell injury can
cause the accumulation of cell debris in
the shunt tube, resulting in obstructions;
and 3) the greater omentum of pseudocyst
surrounding the peritoneal end.17,18

E TV VPS R i s k R a ti o R i s k R a ti o
Study or Subgroup Events Total Events Total Weight M-H, Fixed, 95% CI M-H, Fixed, 95% CI
El-Ghandour, N. M.2011 0 32 1 21 23.8% 0.22 [0.01, 5.21]
Kamikawa, S.2001 0 44 0 44 Not estimable
Navaei, A. A.2018 0 22 2 27 29.8% 0.24 [0.01, 4.82]
Rahman, M. M.2018 0 30 3 30 46.3% 0.14 [0.01, 2.65]

Total (95% CI) 128 122 100.0% 0.19 [0.03, 1.09]

Total events 0 6
Heterogeneity: Chi² = 0.07, df = 2 (P = 0.96); I² = 0%
0.01 0.1 1 10 100
Test for overall effect: Z = 1.86 (P = 0.06)
Favours [ETV] Favours [VPS]

Figure 7. Forest plot of mortality rates for endoscopic third ventriculostomy and ventriculoperitoneal shunt for treating obstructive
hydrocephalus.

338 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.04.255


LIANG LU ET AL.
Study or Subgroup
El-Ghandour, N. M.2011
Navaei, A. A.2018
Rahman, M. M.2018
Total (95% CI)
Total events
Heterogeneity: Tau² = 0.24; Chi² = 15.26, df = 2 (P = 0.0005); I² = 87%
Test for overall effect: Z = 0.88 (P = 0.38)
Figure 8. Forest plot of postoperative success rates for endoscopic third ventriculostomy and ventriculoperitoneal shunt for
treating obstructive hydrocephalus.
Postoperative hematoma is a rare and
serious complication, which can lead to
neurologic dysfunction. The incidence of
postoperative hematoma in the current
study is 8.2%, which is higher than the
1.08% incidence reported by Zhou and Liu 5
years ago.19 Postoperative hematoma
includes intraventricular hematoma,
subdural hematoma, and epidural
hematoma, of which the former 2 are
common. Intraventricular hemorrhage is
often caused by multiple catheterization or
venous injury.20 Subdural hematoma and
epidural hematoma are caused by excessive
drainage. In our study, there were 5 cases
of intraventricular hematoma, 3 cases of
subdural hematoma, and 2 cases of
epidural hematoma in the VPS group. In
comparison, a study by El-Ghandour
showed 2 patients with intraventricular he-
matoma.11 Our study confirmed that ETV
could significantly reduce the risk of
postoperative hematoma compared with
the VPS group.
Further, we found 4 cases of post-
operative CSF leakage that stopped spon-
taneously without treatment. Others
including the study by Kamikawa10 showed
many other complications, such as bowel
perforation (2 cases), skull deformity (7
cases), and slitlike ventricle (24 cases) in
the VPS group, but not in the ETV group.
A study by Rahman13 reported on 1 case
with excessive drainage. All these results
suggest that ETV outperformed VPS in its
ability to reduce the incidence of
complications.
Our study is not without a few limita-
tions: 1) the small number of RCTs
included in our meta-analysis can
compromise the generalizability of the re-
sults; 2) information on the specific causes
of hydrocephalus was not available, and
WORLD NEUROSURGERY 129: 334-340, SEPTEMBER 2019
ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS
E TV VPS R i s k R a ti o
Events Total Events Total Weight M-H, Random, 95% CI
29 32 13 21 34.0% 1.46 [1.03, 2.08]
15 22 24 27 34.8% 0.77 [0.56, 1.05]
25 30 12 30 31.2% 2.08 [1.31, 3.32]
84 78 100.0% 1.30 [0.72, 2.37]
69 49
0.01
Favours [ETV] Favours [VPS]
treatment effects between ETV and VPS
according to specific causes were not
calculated; 3) information on the quality of
long-term survival and prognosis was
insufficient; and 4) age may affect the
outcome of these 2 methods; however, data
on the age of patients with hydrocephalus
was lacking. Despite these limitations, the
results of our meta-analysis are rigorous
and can be used to guide future research.
CONCLUSION
On the basis of the meta-analysis of RCTs
evaluating ETV and VPS, the incidence of
complications and mortality was higher
with the VPS procedure and therefore
greater benefits can be achieved using
ETV. These data should be confirmed in
large-scale high-quality RCTs in the
future.
REFERENCES
1. Kousi M, Katsanis N. The genetic basis of hy-
drocephalus. Annu Rev Neurosci. 2016;39:409-435.
2. Kalani MY, Turner JD, Nakaji P. Treatment of
refractory low-pressure hydrocephalus with an
active pumping negative-pressure shunt system.
J Clin Neurosci. 2013;20:462-466.
3. Mixter W. Ventriculoscopy and puncture of the
floor of the third ventricle. Boston Med Surg J. 1923;
188:277-278.
4. Vieneke P, Lutze T. Technologies for micro-
endoscopies of the future: the MINOP project.
Min Invas Ther Allied. 1998;731:233-239.
5. Takasuna H, Goto T, Kakizawa Y, et al. Use of a
micromanipulator system (NeuRobot) in endo-
scopic neurosurgery. J Clin Neurosci. 2012;19:
1553-1557.
6. Cage TA, Auguste KI, Wrensch M, et al. Self-re-
ported functional outcome after surgical inter-
vention in patients with idiopathic normal
pressure hydrocephalus. J Clin Neurosci. 2011;18:
649-654.
www.journals.elsevier.com/world-neurosurgery
LITERATURE REVIEW
R i s k R a ti o
M-H, Random, 95% CI
0.1 1 10 100
7. Liberati A, Altman DG, Tetzlaff J, et al. The
PRISMA statement for reporting systematic re-
views and meta-analyses of studies that evaluate
health care interventions: explanation and elabo-
ration. J Clin Epidemiol. 2009;62:e1-e34.
8. Higgins JPT, Altman DG, Gotzsche PC, et al. The
Cochrane Collaboration’s tool for assessing risk of
bias in randomised trials. BMJ. 2011;343:d5928.
9. Higgins JP, Thompson SG, Deeks JJ, et al.
Measuring inconsistency in meta-analyses. BMJ.
2003;327:557-560.
10. Kamikawa S, Inui A, Kobayashi N, et al. Endo-
scopic treatment of hydrocephalus in children: a
controlled study using newly developed Yamadori-
type ventriculoscopes. Minim Invasive Neurosurg.
2001;44:25-30.
11. El-Ghandour NM. Endoscopic third ven-
triculostomy versus ventriculoperitoneal shunt in
the treatment of obstructive hydrocephalus due to
posterior fossa tumors in children. Childs Nerv Syst.
2011;27:117-126.
12. Navaei AA, Hanaei S, Habibi Z, et al. Controlled
trial to compare therapeutic efficacy of endoscopic
third ventriculostomy plus choroid plexus cauter-
ization with ventriculoperitoneal shunt in infants
with obstructive hydrocephalus. Asian J Neurosurg.
2018;13:1042-1047.
13. Rahman MM, Salam MA, Uddin K, et al. Early
surgical outcome of endoscopic third ven-
triculostomy in the management of obstructive
hydrocephalus: a randomized control trial. Asian J
Neurosurg. 2018;13:1001-1004.
14. Rekate HL. A contemporary definition and clas-
sification of hydrocephalus. Semin Pediatr Neurol.
2009;16:9-15.
15. Jiang L, Gao G, Zhou Y. Endoscopic third ven-
triculostomy and ventriculoperitoneal shunt for
patients with noncommunicating hydrocephalus.
Medicine. 2018;97:e12139.
16. Simon TD, Butler J, Whitlock KB, et al. Risk fac-
tors for first cerebrospinal fluid shunt infection:
findings from a multi-center prospective cohort
study. J Pediatr. 2014;164:1462-1468.
17. Reddy GK, Bollam P, Caldito G. Long-term out-
comes of ventriculoperitoneal shunt surgery in
patients with hydrocephalus. World Neurosurg.
2013;81:404-410.
339
 LITERATURE REVIEW
LIANG LU ET AL. ETV VS. VPS FOR OBSTRUCTIVE HYDROCEPHALUS

18. Reddy GK, Bollam P, Caldito G, et al. Ven-
triculoperitoneal shunt complications in hydro-
cephalus patients with intracranial tumors: an
analysis of relevant risk factors. J Neurooncol. 2010;
103:333-342.
19. Zhou F, Liu Q, Ying G, et al. Delayed intracerebral
hemorrhage secondary to ventriculoperitoneal
shunt: two case reports and a literature review. Int
J Med Sci. 2012;9:65-67.
20. Calayag M, Paul AR, Adamo MA. Intraventricular
hemorrhage after ventriculoperitoneal shunt revi-
sion: a retrospective review. J Neurosurg Pediatr.
2015;16:42-45.
Conflict of interest statement: We certify that there are no
conflicts of interest in the submission of this manuscript.
Liang Lu and Hongwu Chen are cofirst authors.
Received 27 February 2019; accepted 30 April 2019
Citation: World Neurosurg. (2019) 129:334-340.
https://doi.org/10.1016/j.wneu.2019.04.255
Journal homepage: www.journals.elsevier.com/world-
neurosurgery
Available online: www.sciencedirect.com
1878-8750/$ - see front matter ª 2019 Published by Elsevier
Inc.

340 www.SCIENCEDIRECT.com WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2019.04.255