Clinical Neurology and Neurosurgery 124 (2014) 97101

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery

journal homepage: www.elsevier.com/locate/clineuro

Protective effect of external ventricular drainage on cerebral

vasospasm. A retrospective study on aneurysmal SAH treated
endovascularly
G.M. Della Pepa *, A. Scerrati, A. Albanese, E. Marchese, G. Maira, G. Sabatino
Institute of Neurosurgery, Catholic University of Rome, Largo A. Gemelli 8, 00168, Italy

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 1 April 2014
Received in revised form 17 June 2014
Accepted 22 June 2014
Available online 28 June 2014

Objective: Cerebral vasospasm (VS) is one of the factors that can most signicantly worsen the prognosis
after aneurysmal subarachnoid hemorrhage (SAH). A substantial body of evidence supports the idea that
CSF diversion could prevent VS, even if this issue is still much debated. External ventricular drainage
(EVD) is the recommended procedure for post-hemorrhagic hydrocephalus. In this study we analyzed
whether EVD, placed for acute hydrocephalus, is effective in reducing the incidence of clinical and
radiological cerebral vasospasm in patients who underwent endovascular treatment for aneurysmal SAH.
Patients and methods: We retrospectively studied the incidence of radiologically conrmed VS in 141
patients treated endovascularly for aneurysmal SAH: 80 underwent EVD for hydrocephalus, 61 did not
undergo EVD.
Results: VS occurred in 8.75% of cases (7 patients) in the rst groups, while in 22.95% (14 patients) in the
second group. In addition, patients not treated with EVD display a prevalence of VS in lower Fisher grades
compared to the other group.
Conclusion: Our data indicate that CSF drainage reduces the risk of vasospasms in patients with
endovascular treatment for aneurysmal SAH.
2014 Elsevier B.V. All rights reserved.

Keywords:
Vasospasm
SAH
External ventricular drainage

1. Introduction
Cerebral vasospasm is a major cause of morbidity and mortality
associated with subarachnoid hemorrhage (SAH) [1,2]. Advances in
neuroimaging and development of newer intraparenchymal
monitoring devices have improved the prediction and diagnosis
of cerebral vasospasm signicantly [3]. However, VS still contributes to poor outcome in approximately 1040% of patients with
SAH [3]. There is still no effective means to prevent vasospasm, and
clinical practice focus on management of its effects while the
disease runs its course; as a matter of fact it continues to be one of
the factors that can most signicantly worsen the prognosis
afteraneurysmal SAH [16].
The pathogenesis of cerebral vasospasm has yet to be fully
claried, but there is no doubt that the presence of blood in the
subarachnoid space leads to a narrowing of the intracranial arteries
[7]. Because hemolysis is the primary cause of vasospasm, it follows
that clot removal will decrease cerebral vasospasm, and clinical trials
of clot removal during early aneurysm surgery have been performed

* Corresponding author. Tel.: +39 0630154120; fax: +39 063051343.

E-mail address: gdellapepa@hotmail.com (G.M. Della Pepa).
http://dx.doi.org/10.1016/j.clineuro.2014.06.030
0303-8467/ 2014 Elsevier B.V. All rights reserved.

since the early 1980s [7]. The idea that removing the subarachnoid
blood earlyafter the hemorrhage could prevent cerebral vasospasm
was proposed by Johnson and colleagues in 1958 [7,8]. Clot removal,
either surgically or by draining blood out of the basal cisterns,
reduces vasospasm in animals and probably in man [9]. This is based
on the theory that the subarachnoid clot causes vasospasm, which is
supported by a substantial body of evidence [7,9].
Recently a cerebral vasospasm after SAH has undergone a
renewed interest and a series of papers have been published on the
benecial effects of lumbar drainage in removing spasmogens in the
cerebrospinal uid (CSF) thus reducing the incidence of symptomatic
vasospasm [9,10]. Other papers, instead suggest thatCSF drainage is
detrimental [7,10] while some other experiences, instead, underline
that draining CSF (independently from the drainage method) might
even delayclot clearance and worsen vasospasm [10]. Other studies
show that lumbar drain (LD) is superior than external ventricular
drainage (EVD) in the clearance of subarachnoid clots [11].
1.1. Rationale for the study
Accordingly, external ventricular drainage (EVD) is recommended in SAH cases presenting with acute hydrocephalus [12

98

G.M. Della Pepa et al. / Clinical Neurology and Neurosurgery 124 (2014) 97101

14]. Interestingly ndings indicate that lumbar drainage (LD) may

be superior to the EVD in the clearance of subarachnoid clots [11].
However, the indication for CSF drainage following SAH is still a
matter of debate, even detrimental effects have been reported
[7,10], suggesting that EVD or LD might even delay clot clearance
and may worsen vasospasm [10].
Therefore we retrospectively investigated the incidence of
vasospasm in a cohort of patients who underwent endovascular
treatment for aneurysmal SAH. Patients who were treated with
EVD for acute posthemorrhagic hydrocephalus are compared with
those without EVD.

2.4. Ventricular catheter placement and EVD

2. Patients and methods

In all cases a right frontal ventriculostomy was performed; a

Codman Bactiseal catheter is usually placed in the most recent
years, while a CodmandCordis catheter was in the older cases.
Antibiotic prophylaxis is usually used perioperatively (Cepahzoline). The EVD is placed at 1418 cm H2O. EVD is placed generally
right after endovascular procedure, that is, in our institute, carried
out in an emergency setting within few hours after hemorrhage. In
all cases both endovascular treatment and ventriculostomy were
carried out within the rst 24 h from hemorrhage. In all
casesventriculostomy is performed in the operating room by a
neurosurgeon or a neurosurgical resident.

2.1. Patient population

2.5. Statistical analysis

In this study we restrospectivelyanalysed our Institution

database on nontraumatic SAH in the period going between
January 2003 and December 2013. The analysis regarded only
patients with aneurysmal SAH who underwent endovascular
treatment of the aneurysmal malformation. Clinical records from
admission upto the time of discharge were available and were
reviewed.We chose to focus on patients treated by endovascular
methods in order to standardize population sample, to reduce
possible confounding factors such as arterial manipulation during
surgery and cisternal irrigation, all which could introduce bias into
the data. One hundred and forty one patients were included in the
study. Characteristics of the study did not require Institutional
Review Board approval.
EVD was placed when at admission there was CT evidence of
post-hemorrhagic hydrocephalus (right frontal ventriculostomy).
Thus we collected two separated populations wit aneurysmal SAH:
one who underwent EVD for acute hydrocephalus (called VD
GROUP), the other without radiological signs of hydrocephalus not
harbouring and EVD (called NVD GROUP).
Only patients placed in EVD within the rst 24 h from
admission were included in the study. We excluded from analysis
all patients who died before clinical vasospasm could develop (all
patients died before 14 days from onset).

Signicance of data concerning VS occurrence between the two

groups was analysed by Chi-square test; a probability of 0.05 or less
was considered signicant.

2.2. Denition of vasospasm

In the study we decided to focus on those patients in whom
suspected clinical VS was eventually conrmed by the association
of at least the association of two of the following tests: CT scan
(displaying hypodensities not showed before), angiography (focal/
diffuse vessels narrowing) or TCD ultrasonography.
Vasospasm was diagnosed using the criteria dened in most of
the relevant literature [1517]: (1) onset of new neurological
decitssuch as confusion, disorientation, drowsiness, or focalmotor decit during posthemorrhage Days 414; (2) negativendings on CT scans obtained to rule out other causesof neurological
deterioration such as hemorrhage, cerebraledema, or hydrocephalus; (3) no other identiable causeof neurological deterioration
such as hyponatremia, hypoxia, drug toxicity, infection, or seizures.

3. Results
In the 10 year period in our institution 141 patients were treated
endovascularly for aneurysmal SAH satisfying the inclusion
criteria. 80 underwent EVD for hydrocephalus (VD group), 61
did not undergo EVD (NVD group). In the VD group mean age of
patient was 58.71 years, while it was 58.15 in the NVD group; the
VD group included 51 females (63.75%) and 29 males (36.25%),
while the NVD group 43 females (70.49%) and 18 males (29.51%)
Table 1. Aneurysm location is reported in Table 2. Smokers were
22.7% in NVD group while 18.2% in VD group.
Hunt-Hess (HH) grade was at follows for the VD group: HH 127
patients (33.75%), HH 29 patients (11.25%), HH 312 patients
(15.00%), HH 426 patients (32.50%), HH 56 patients (7.50%); for
the NVD group: HH 144 patients (72.13%), HH 212 patients
(19.67%), HH 33 patients (4.92%), HH 41 patient (1.64%), HH 51
patient (1.64%) Table 3.
Fisher grade for the VD group was the following: Fisher 210
patients (12.50%), Fisher 319 patients (23.75%), Fisher 451
patients (63.75%); for the NVD group: Fisher 219 patients
(31.15%), Fisher 328 patients (45.90%), Fisher 414 patients
(22.95%) Table 3.
We observed 21 cases of clinical vasospasm that was
radiological conrmed (association of CT or angiography or
TCD).Seven occurred in the VD group (8.75% of cases), while 14
in the NVD group (22.95% of cases). Table 4This difference showed
to be signicant at chi-square test (p = 0,03).

Table 1
Patients characteristics.
Group

No. of patients

F (%)

M (%)

Age

VD
NVD
Overall

80
61
141

51 (63.75%)
43 (70.49%)
94 (66.67%)

29 (36.25%)
18 (29.51%)
47 (33.33%)

58.71
58.15
58.47

2.3. Medications
The medical approach among the two groups was exactly the
same: after aneurysm coiling volemia and blood pressure were
sustained by means of intravascular volume expansion (isotonic
saline 3 l/day, adjusted on oral intake and electrolytes examinations); Nimodipine was infused with infusion velocity regulated
according systolic pressure (if possible up to 10 ml/h or 60 mg
orally every 4 h for 3 weeks, if systolic pressure was not dropping).

Table 2
Aneurysm location.
Group

AComA

MCA

ICA

BA

VA/PICA/SCA

Other

Multiple

VD
(80 patients)
NVD
(61 patients)

42

12

48

14

15

34 (25.3%)

31

16

14 (14.6%)

G.M. Della Pepa et al. / Clinical Neurology and Neurosurgery 124 (2014) 97101
Table 3
Hunt Hess distribution and Fisher grades at admission.
HH

1
2
3
4
5
Fisher
2
3
4

VD group

99

4. Discussion
4.1. VS and CSF drain: literature background.

NVD group

Patients

Patients

27
9
12
26
6

33.75
11.25
15.00
32.50
7.50

44
12
3
1
1

72.13
19.67
4.92
1.64
1.64

10
19
51

12.50
23.75
63.75

19
28
14

31.15
45.90
22.95

Diagnosis was made by means of CT + TCD in 5/7 cases,

CT + angiography + TCD in the other 2 cases in the VD group. In
the NVD-group in 10/14 cases CT + TCD was used, while in the
remaining 4 cases CT + angiography + TCD was used. All patients
in which VS was suspected underwent CT scan and TCD
routinely. Angiography was performed in those cases in which
symptoms were particularly severe and in those in which an
angiographic control was necessary to assess aneurysm
exclusion.
Of observed cases of vasospasm in the VD group in 5 out of 7
cases a CT evident ischemia (not present before) was observed and
5 patients out of 7 displayed a motor decit (71.42% of VS cases)
and at discharge a VS-related neurological decitwas present in
3out of 7 (42.85% of VS cases). Equally in the NVD group an
ischemia not evident before was demonstrated in 11out of 14 cases,
12 out of 14 cases displayed a motor decit (85.71% of VS cases) and
5 patients displayed at discharge a VS-related neurological decit
(35.7%). Table 1 Differences in terms of motor decitis and of VSrelated neurological decit at discharge is not statistically
signicant between the two groups (p = 1).
Of the 7 cases occurring in the VD group these all occurred in
Fisher 4 SAHs; in the 14 cases occurring in the NVD group 5
occurred in Fisher grade 2 SAHs, 7 in Fisher grade 3, 2 in Fisher
grade 4. Table 5 and Fig. 1 show the overall incidence of
vasospasm in the two groups and its distribution according Fisher
grade.
Mean time in EVD was 20.67 days; for patients without VS is
24.46 while 20.17 for patient with VS (statistically not signicant).
VS does not affect the overall stay in EVD. There is a difference in
the length of EVD stay difference according Fisher grades: Fisher 4
26.90 days, Fisher 320.02 days, Fisher 217.5 days.

The relation between CSF drainage and occurrence of VS is still a

matter of debate. Only few published studies, almost applying LD,
address this topic [7,9,11] claiming some positive effect of CSF
drainage on VS. A recent radiological study comparing LD and EVD
in SAH patients showed that the clearance of blood clots is more
rapid with LD and even the occurrence of new low density area on
CT scans is reduced with LD [11]. However, other reported negative
effects of CSF drainage, independt from the drainage method
(cisternal, ventricular lumbar drainage or combinations of these)
[10]. Nevertheless, since hydrocephalus is not uncommon following SAH, CSF drainage is still considered as a routine procedure
[12,14].
Previous studies on the relation between CSF drainage and
vasospasm are mostly based on patients who were surgically
treated. The present clinical observations are the rst focusing on
cases undergoing endovascular treatment.
However acute hydrocephalus occurring after SAH is notan
uncommon nding, and EVD is still considered to be the most
effective means to treat it [1214]. The present study is the rst
attempt to evaluate whether EVD has some effect on cerebral
vasospasm in patients with SAH undergoing endovascular
treatment.
4.2. Differential diagnosis between VS-related neurological
deterioration and micro-embolism related to endovascular procedure.
An important point in the study is to assess that neurological
deterioration or transient ischemia is not due to other possible
causes related to the endovascular treatment tiself. Our study
focuses on a population of patients affected by SAH treated
endovascularly. Thus it appears fundamental to determine if
symptoms are due to embolism after endovascular procedure.
Such event is commonly found on MR imaging scans after
endovascular procedures and is thought to develop more readily
if the aneurysm is large, incompletely coiled due to a greater
volume of turbulent ow and possible clot formation, or both
[18,19]. Stent assisted coiling, also contributes signicantly to the
increase in embolic risk because of the increase in thrombogenic
surface, as well as vessel wall manipulation [18]. However,
ischaemia from embolic sources related to the endovascular
implants has a different timing from vasospasm and occurs
generally within rst hours from the procedure and can be

Table 4
Incidence of vasospasm.
Vasospasm

VD group
NVD group
Total

No. of cases (%)

Ischemic area at CT-scan

Motor decit

Vasospasm-related neurological decit at discharge

7 (8.75)
14 (22.95)
21 (14.89)

5 (71.42%)
11 (78.57%)

5 (71.42%)
12 (85.71%)

3 (42.85%)
5 (35.71%)

Table 5
Distribution of vasospasm per Fisher grade.
VD group

NVD group

Fisher

Vasospasm

Distribution %

Fisher

Vasospasm

Distribution %

2 (10 patients)
3 (19 patients)
4 (51 patients)
Overall VS

0
0
7
7 (8.75%)

0
0
100

2 (19 patients)
3 (28patients)
4 (14 patients)
Total

5
7
2
14 (22.95%)

35.71
50
14.28

100

G.M. Della Pepa et al. / Clinical Neurology and Neurosurgery 124 (2014) 97101

similar among the two groups, meaning that in nearly a half of

cases of VS related neurological decits are transient and generally
improve during hospitalization.
Conclusions

Fig. 1. Overall incidence of vasospasm in the two groups and its distribution
according Fisher grade.

distinguished from VS-related neurological deterioration that has a

more delayed timing (414th day) [20].
Studies have shown that microembolisation after endovascular
procedure, clot formation in or adjacent to the aneurysm or to the
stent seems to occur mostly within the rst 612 h after the coiling
[19,21]. The risk persists later after the treatment (12 h) and it is
demonstrated that the total amount of micro-embolisations is then
already very low, and new adverse events are generally not
observed at that time [19,21]. Thus timing of neurological
symptoms onset is the key to differentiate the cause of
neurological deterioration in the setting of a patient treated with
an endovascular procedure.
4.3. Protective effect of CSF diversion on VS
In our series we analyzed 141 cases of aneurysmal SAH
undergoing endovascular treatment. Cases harboring acute hydrocephalus were treated with transient EVD, while other did not
undergo any CSF drainage procedure. Twenty one cases of clinical
vasospasm that was radiological conrmed (CT or angiography or
TCD or association of them) were recorded. Overall radiologically
conrmed VS occurred in nearly 15% of cases, and this is
substantially similar to data reported in literature [9,17].
Interestingly the prevalence of VS signicantly differs among
the VD and NVD groups: in patients undergoing EVD the
occurrence of VS is nearly three folds lower (8.75% vs 22.95% of
cases).
Noteworthy is also the different distribution of VS cases in
relationship with Fisher grades. In patients undergoing EVD (VD
group) VS cases denitely distribute towards higher Fisher grades
(all cases Fished grade 4), while in those patients not undergoing
EVD (NVD group) the VS distribution in relationship to Fisher
grades displays a prevalence also in the lower grades (almost a half
of cases in Fisher grade 2).
The difference in terms of VS among the two population is even
more relevant if we take into account that theoretically the two
groups do not have the same VS predisposition ab initio: as also
conrmed by our data, patients that present acute post-hemorrhagic hydrocephalus often also display a more severe SAH both in
HH and Fisher grades with a greater amount of sub-arachnoid
blood and are thus more prone towards VS.In other words the
intrinsic characteristics of patients enrolled in the VD group preset
a greater propensity towards VS. This further magnies the nding
of a nearly three-folds lower occurrence of VS in this group.
On the other hand, while we observed a clear difference in
terms of incidence of VS among the two groups it must be noted
that this does not necessarily correspond to a greater number of
neurological decits (related to VS) at discharge, and that this is

Still no clear data exists on the role of a protective effect of CSF

drain towards VS after aneurysmal SAH in literature. Some
convincing evidence in literature support the idea that LD might
be more effective in removing sub-arachnoid blood. However in
many cases in neurosurgical practice to place an EVD is still
necessary because of acute hydrocephalus. Our data display that
EVD plays a protective role towards VS after SAH and thus we
recommend in those cases in which it is not possible to place a LD
and in those in which acute hydrocephalus is present at onset.
Our results, albeit on a limited number of patients, would
support the concept of CSF diversion in patients with endovascular
treatment for aneurysmal SAH: in the form of a LD (in cases
without hydrocephalus, as suggested by literature) or in the form
of an EVD (in cases with hydrocephalus) CSF diversion reduces the
risk of VS.
Disclosure statement: the authors declare no interest to
disclose, they have any personal or institutional nancial interest
in drugs, materials, or devices described in this submission and
that they did not receive any specic funding. The paper and any of
its contents have been presented previously.
References
[1] Otite F, Mink S, Tan CO, Puri A, Zamani AA, Mehregan A, et al. Impaired cerebral
autoregulation is associated with vasospasm and delayed cerebral ischemia in
subarachnoid hemorrhage. Stroke 2014.
[2] Mir DI, Gupta A, Dunning A, Puchi L, Robinson CL, Epstein HA, et al. CT
perfusion for detection of delayed cerebral ischemia in aneurysmal
subarachnoid hemorrhage: a systematic review and meta-analysis. Am J
Neuroradiol 2013.
[3] Cremers CH, van der Schaaf IC, Wensink E, Greving JP, Rinkel GJ, Velthuis BK,
et al. CT perfusion and delayed cerebral ischemia in aneurysmal subarachnoid
hemorrhage: a systematic review and meta-analysis. J Cereb Blood Flow Metab
2013.
[4] Sabatino G, Della Pepa GM, Albanese A, Marchese E. Angiogram-negative
subarachnoid hemorrhage: is it a correct denition? Acta Neurochir (Wien)
2013;155:11279.
[5] Sabatino G, Rigante L, Minella D, Novelli G, Della Pepa GM, Esposito G, et al.
Transcriptional prole characterization for the identication of peripheral
blood biomarkers in patients with cerebral aneurysms. J Biol Regul Homeost
Agents 2013;27:72938.
[6] Sabatino G, Della Pepa GM, Scerrati A, Maira G, Rollo M, Albanese A, et al.
Anatomical variants of the basal vein of Rosenthal: prevalence in idiopathic
subarachnoid hemorrhage. Acta Neurochir (Wien) 2014;156:4551.
[7] Macdonald RL. Lumbar drainage after subarachnoid hemorrhage: does it
reduce vasospasm and delayed hydrocephalus? Neurocrit Care 2007;7:12.
[8] Johnson R, Potter JM, Reid RG. Arterial spasm in subarachnoid haemorrhage:
mechanical considerations. J Neurol Neurosurg Psychiatr 1958;21.
[9] Klimo Jr. P, Kestle JR, MacDonald JD, Schmidt RH. Marked reduction of cerebral
vasospasm with lumbar drainage of cerebrospinal uid after subarachnoid
hemorrhage. J Neurosurg 2004;100:21524.
[10] Kasuya H, Shimizu T, Kagawa M. The effect of continuous drainage of
cerebrospinal uid in patients with subarachnoid hemorrhage: a retrospective
analysis of 108 patients. Neurosurgery 1991;28:569.
[11] Maeda Y, Shirao S, Yoneda H, Ishihara H, Shinoyama M, Oka F, et al. Comparison
of lumbar drainage and external ventricular drainage for clearance of
subarachnoid clots after Guglielmi detachable coil embolization for aneurysmal subarachnoid hemorrhage. Clin Neurol Neurosurg 2013;115:96570.
[12] Steiner T, Juvela S, Unterberg A, Jung C, Forsting M, Rinkel G. European Stroke
Organization guidelines for the management of intracranial aneurysms and
subarachnoid haemorrhage. Cerebrovasc Dis 2013;35:93112.
[13] Mehta V, Holness RO, Connolly K, Walling S, Hall R. Acute hydrocephalus
following aneurysmal subarachnoid hemorrhage. Can J Neurol Sci
1996;23:405.
[14] Burrows AM, Korumilli R, Lanzino G. How we do it: acute management of
subarachnoid hemorrhage. Neurol Res 2013;35:1116.
[15] Haley Jr. EC, Kassell NF, Alves WM, Weir BK, Hansen CA. Phase II trial of
tirilazad in aneurysmal subarachnoid hemorrhage. A report of the cooperative
aneurysm study. J Neurosurg 1995;82:78690.
[16] Haley Jr. EC, Kassell NF, Apperson-Hansen C, Maile MH, Alves WM. A
randomized, double-blind, vehicle-controlled trial of tirilazad mesylate in

G.M. Della Pepa et al. / Clinical Neurology and Neurosurgery 124 (2014) 97101
patients with aneurysmal subarachnoid hemorrhage: a cooperative study in
North America. J Neurosurg 1997;86:46774.
[17] Lanzino G, Kassell NF, Dorsch NW, Pasqualin A, Brandt L, Schmiedek P, et al.
Double-blind, randomized, vehicle-controlled study of high-dose tirilazad
mesylate in women with aneurysmal subarachnoid hemorrhage. Part I. A
cooperative study in Europe, Australia, New Zealand, and South Africa. J
Neurosurg 1999;90:10117.
[18] Heller RS, Dandamudi V, Lanfranchi M, Malek AM. Effect of antiplatelet
therapy on thromboembolism after ow diversion with the pipeline
embolization device. J Neurosurg 2013;119:160310.

101

[19] Schubert GA, Thome C, Seiz M, Douville C, Eskridge J. Microembolic signal

monitoring after coiling of unruptured cerebral aneurysms: an observational
analysis of 123 cases. Am J Neuroradiol 2011;32:138691.
[20] Lindegaard KF. The role of transcranial Doppler in the management of patients
with subarachnoid haemorrhagea review. Acta Neurochir Suppl 1999;72:59
71.
[21] Albayram S, Selcuk D. Thromboembolic events associated with Guglielmi
detachable coil embolization of asymptomatic cerebral aneurysms: evaluation
of 66 consecutive cases with use of diffusion-weighted MR imaging. Am J
Neuroradiol 2004;25:15960.

Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.