Neurocritical Care
Copyright © 2007 Humana Press Inc.
All rights of any nature whatsoever are reserved.
ISSN 1541-6933/07/6:40–44 ISSN 1556-0961 (Online)
DOI: 10.1385/Neurocrit. Care 2007;06:40–44

Practical Pearl
Perfusion CT to Evaluate the Effect of Transluminal Angioplasty
on Cerebral Perfusion in the Treatment of Vasospasm After
Subarachnoid Hemorrhage
Charles B. L. M. Majoie,1 Leonard J. van Boven,1 Diederik van de Beek,2,* Henk W. Venema,1,3 and Willem J. van Rooij4
1

Department of Radiology, 2Neurology, and 3Medical Physics, Academic Medical Center, University of Amsterdam,
Amsterdam; the 4Department of Radiology, St. Elisabeth Hospital, Tilburg, the Netherlands

Abstract
Introduction: Delayed ischemic neurologic deficits secondary to vasospasm are a major
cause of morbidity and mortality after subarachnoid hemorrhage (SAH). Treatment
of vasospasm after SAH is associated with complications, and reliable techniques for
evaluating effects of treatment of vasospasm in such patients are warranted. We present
the use of perfusion computed tomography (PTC) to evaluate the effect of transluminal
percutaneous angioplasty in a with SAH and vasospasm-induced ischemia.
Methods: Dynamic PCT with deconvolution produced maps of time-to-peak, mean transit
time, regional cerebral blood flow, and regional cerebral blood volume, with a computerized
automated map of the infarct and penumbra. CT scanners with quadruple detector array
were used before and after angioplasty.
Results: Before angioplasty and intraarterial papaverine, PCT showed normal to decreased
cerebral blood flow and increased cerebral blood volume and mean transit time in the
middle cerebral artery territory of the left hemisphere. After angioplasty and intraarterial
papaverine, PCT showed normalization of perfusion parameters.
Conclusion: PCT can be a useful technique in monitoring angioplasty treatment effects
in patients with vasospasm after SAH.
Key Words: Subarachnoid hemorrhage; perfusion CT; angioplasty; vasospasm.
(Neurocrit. Care 2007;06:40–44)

*Correspondence and Reprint
Requests to:
Diederik van de Beek
Department of Neurology
Academic Medical Center,
University of Amsterdam
P.O. Box 22700, 1100 DE
Amsterdam, The Netherlands
E-mail
d.vandebeek@amc.uva.nl

Introduction
Delayed ischemic neurologic deficits secondary to vasospasm are a major cause of
morbidity and mortality after subarachnoid
hemorrhage (SAH) (1). The development of
vasospasm can be recognized by neurologic
deterioration in combination with radiological techniques, such as conventional digital
subtraction angiography (DSA), transcranial
Doppler (TCD) ultrasound, positron emission tomography (PET), single-photon emission computer tomography (SPECT), and

40

single-slice perfusion computed tomographic
technique (Xe-CT) (1–6). Two recent studies
showed that perfusion CT (PCT) also represents an accurate screening test in patients with suspected vasospasm, although
further refinements are needed before routine
recommendation (6,7). Once symptomatic
vasospasm is evident (with focal neurologic signs), hemodynamic augmentation is
widely employed, although definitive evidence for this therapy is still lacking (1,8,9).
Patients whose condition does not improve

PCT in Vasospasm After SAH

41

with medical therapy undergo cerebral angiography and
transluminal angioplasty (1,10–12). Aggressive treatment of
vasospasm is not without risk (1,8). Hemodynamic augmentation has a complication rate of 10–20% and balloon angioplasty
may be complicated by vessel rupture, branch occlusion, and
hemorrhagic infarction (1,8–11). Therefore, accurate monitoring of treatment effects in patients with vasospasm after SAH
is needed. We report the results of the use of PCT to evaluate
the effect of transluminal percutaneous angioplasty and intraarterial papaverine in a patient with SAH and vasospasminduced ischemia.

Case Report
A 37-year-old woman was admitted with severe sudden
headache, nausea, vomiting, and neck stiffness (Hunt & Hess
grade II). Unenhanced CT showed SAH predominantly in the
left Sylvian fissure and over the left lateral convexity (Fisher
grade 3). DSA showed a left internal carotid artery tip aneurysm,

which was treated with endovascular coil placement (Guglielmi
Detachable Coils, Boston Scientific, Cork, Ireland). After the
procedure she had mild headache and no neurological deficits.
On the fourth day after coil placement, however, she developed
aphasia. TCD revealed severe vasospasm. Despite hemodynamic augmentation and nimodipine, she remained symptomatic and angioplasty was considered. PCT was performed
6 days after coil placement to determine if there was brain tissue at risk for cerebral infarction. Angioplasty was performed
and PCT was repeated after angioplasty.
CT scanners with quadruple detector array were used before and after angioplasty. For PCT, we used a technique described previously (3,4,7). In summary, after unenhanced CT
of the brain, PCT was performed twice at two different levels,
at the roof of the third ventricle and at the roof of the lateral
ventricles. For each examination 40 mL of a nonionic contrast
agent (iodixanol, Visipaque 320 mg of iodine per mL, GE
Health, Cork, Ireland) were injected with a power injector into

Fig. 1. PCT was performed to determine if the patient was at risk for cerebral infarction and whether transluminal angioplasty should be
performed or not (A–C) and was repeated after angioplasty to monitor treatment effects on cerebral perfusion (D–F). (A) Preangioplasty
PCT CBV map demonstrates increased CBV values in the left hemisphere, consistent with ischemia-related activation of cerebral autoregulation (blue to red: CBV = 0 to 20 mL/100g). (B) Preangioplasty PCT MTT map demonstrates increased MTT values in the left hemisphere,
consistent with ischemia-related activation of cerebral autoregulation (red to blue: MTT = 0 to 15 seconds). (C) Preangioplasty PCT CBF
map demonstrates normal to decreased CBF values in the left hemisphere (blue to red: CBF = 0 to 100 mL per 100 g and per minute). (D)
Postangioplasty PCT CBV map demonstrates normalized CBV values in the left hemisphere, symmetric to the normal right side. (E) Postangioplasty perfusion CT MTT map demonstrates normalized MTT values in the left hemisphere, symmetric to the normal right side. (F)
Postangioplasty PCT CBF map demonstrates normalized CBF values in the left hemisphere, symmetric to the normal right side.

Neurocritical Care

♦

Volume 6, 2007

Majoie et al.

42

Fig. 2. DSA images in a 37-year-old woman who developed progressive aphasia 1 day after coiling of a ruptured left internal carotid arterytip aneurysm, despite optimal medical treatment.Transluminal balloon angioplasty was performed to treat the vasospasm. (A) Preangioplasty
DSA demonstrates severe vasospasm in the M1 and M2 segments of left MCA. (B) Postangioplasty DSA shows return to normal caliber of
the M1-segment of the left MCA.

a cubital vein at a rate of 4 mL/second. At 7 seconds after initiation of the injection, a continuous scan was started with the
following parameters: 80 kV, 100 mA (Sensation 4) or 90 kV,
120 mA (Mx8000), 4 × 5-mm collimation, 1 second per rotation
for 46 seconds. The four 5-mm sections were reformatted into
two 10-mm-thick sections. The overall dose required for PCT
was 4.0–4.5 mSv. PCT data were analyzed with commercially
available software developed by Wintermark et al. (14). Imagepostprocessing involved placement of the region of interest
(ROI) in an input artery (anterior cerebral artery) and an input
vein (torcular herophili or superior sagittal sinus) and subsequent generation of contrast-enhancement curves. The software generated color-coded cerebral blood flow (CBF), cerebral
blood volume (CBV) and mean transit time (MTT) maps based
on the CT time attenuation curves for each pixel. After large
vessel removal, four circular ROIs of 1.5 cm in diameter were
placed along the cortical mantle of each hemisphere in the
middle cerebral artery (MCA) territory on the perfusion maps
to generate quantitative data. CBF, CBV, and MTT values in
the normal right hemisphere were compared with those of the
affected left hemisphere.
Unenhanced CT revealed a small left insular infarct. PCT
showed areas of normal or decreased CBF and a large area of
increased CBV and MTT in the MCA territory of the left hemi-

Neurocritical Care

sphere (Figure 1A–C), consistent with ischemia-related activation of autoregulation. The mean value of cortical CBF measured in the MCA territory of the right (normal) hemisphere
was 77 mL per 100 g and per minute (range: 40–174). The mean
value of cortical CBF in the left (affected) hemisphere was 31
mL per 100 g and per min (range: 13–38). The mean value of
cortical CBV measured in the right (normal) hemisphere was
1.4 mL/100 g (range: 1.1–1.8). The mean value of cortical CBV
in the left (affected) hemisphere was 4.7 mL/100 g (range 3.9–
6.0). The mean value of cortical MTT measured in the right
(normal) hemisphere was 1.2 seconds (range 0.6–1.9). The
mean value of cortical MTT in the left (affected) hemisphere
was 9.6 seconds (range 6.7–19.0).
On the same day as the PCT, angioplasty was performed
under general anesthesia with full heparinization via a transfemoral approach using an over the wire (Transend 0.010 in,
Target Therapeutics/Boston Scientific, Boston, MA) silicone
balloon microcatheter (Sentry, Target Therapeutics/Boston
Scientific, Fremont, CA), according to a technique described
previously (11,15). Before angioplasty, DSA showed severe
vasospasm of the M1 segment of the left MCA and of the
proximal left M2 segments (Figure 2A). After angioplasty of
the M1 segment of the left MCA a good angiographic result
was obtained (Figure 2B). In addition, 150 mg of papaverine
♦

Volume 6, 2007

PCT in Vasospasm After SAH

43

(papaverine hydrochloride, Eli Lilly and Co., Indianapolis, IN)
was infused via a microcatheter in the left MCA to treat spasm
of the proximal M2 segments. Within 2 days, her aphasia disappeared. PCT was repeated and showed symmetric, normalized CBV, CBF, and MTT values (Figure 1D–F).

Case Discussion
This case illustrates the value of PCT in demonstrating the
effect of transluminal balloon angioplasty on cerebral perfusion in a patient with vasospasm after SAH. Before angioplasty, decreased to normal CBF and increased CBV and MTT
were identified in the left MCA territory, indicating tissue at
risk for cerebral infarction with intact autoregulation. After angioplasty, perfusion parameters returned to normal values and
were symmetric.
Cerebral vasospasm is a frequent complication in patients
with aneurysmal SAH (1). In 50% of cases it is manifested by
delayed neurologic deficits related to ischemia, with progression to infarction in approximately one-half of the symptomatic
cases (1,8). Angioplasty for vasospasm may significantly reduce
morbidity and mortality if performed early after medical therapy has failed (1,10). It is, albeit rarely, also associated with severe complications, such as vessel rupture, and should therefore only be used in those patients in whom vasospasm causes
cerebral ischemia (1,8). Because delayed neurologic deficits following SAH may have different causes, including hydrocephalus, hemorrhage, edema, metabolic factors, and ischemia (12), it
is important to identify those patients with vasospasm who
most likely will benefit from angioplasty. Although TCD may
aid in the determination of vasospasm, it does not necessarily
assist in the identification of ischemia (12). SPECT (2), Xe-CT
(3), and PET (4,5) are reliable methods for the determination of
cerebral perfusion, but these techniques are cumbersome and
not widely available. PCT can be combined with unenhanced
CT, which is routinely performed for follow-up of patients with
SAH (7). Nabavi et al. used this technique for monitoring
cerebral perfusion after aneurysmal SAH (6). They found significant differences in CBF and CBV between patients with significant and absent or minor vasospasm and between patients
with and without a presumed vasospasm-related infarct. They
could not, however, define a PCT-threshold for cerebral infarct
by using PCT, because of the use of relatively large ROIs, causing volume averaging with normal and ischemic tissue (6,16).
The findings of the current report are consistent with results
described by Grubb et al. (17). The system of autoregulation is
reflected by PCT findings in patients with cerebral ischemia
(18). A mild reduction in cerebral perfusion pressure is associated with a compensatory dilatation of cerebral blood vessels.
Under these circumstances, measurements of CBF remain
within normal range, while those of CBV and MTT increase. As
perfusion pressure falls, normal CBF cannot be maintained,
despite further vasodilatation (oligemia or misery perfusion).
When perfusion pressure falls below a threshold of approximately 20 mL per 100 g and per minute, cerebral metabolism is
severely impaired (ischemia). The development of subsequent
irreversible damage (infarction) is not only dependent on the
absolute perfusion value but also on the time the tissue has
been ischemic. Irreversible damage is associated with loss of
autoregulatory vasodilatation and therefore reduced CBV (18).
PCT may also be valuable in detecting subclinical ischemia,

Neurocritical Care

i.e., areas of vasospasm which are still asymptomatic owing to
intact autoregulation but may become symptomatic over time.
Further research is warranted to determine if angioplasty of
spastic vessels supplying these areas may improve outcome of
patients with SAH.
PCT has several limitations, including restricted anatomic
coverage and lack of full validation of its accuracy, reliability,
and reproducibility of quantitative results (7,16,19,20). With
our multislice CT scanner, we were able to use a collimation of
4 × 5 mm and to obtain 2 × 10 mm reconstructed slices. By performing PCT at two different levels we could obtain 4 cm brain
coverage. With 16-, 32- and 64-slice CT scanners, a larger brain
coverage will be possible. Quantitation is strongly dependent
on the choice of the venous ROI location (19,20). Different
ROI locations within the same vein may result in significantly
different CBV and CBF values (19). The venous ROI with the
greatest peak enhancement value should be chosen, as is done
in the software used in this study to limit variability and increase quality of PCT images (19). With proper ROI selections
a good correlation between quantitative PCT maps and
xenon-CT and positron emission tomography (PET) studies
was found, as described in previous reports (3,5).
This case illustrates the value of PCT for monitoring treatment effects in patients with symptomatic vasospasm. This
minimally invasive technique is widely available and requires
only a modest radiation dose.

Acknowledgments
Diederik van de Beek received funding from the Meningitis
Research Foundation UK, the Jan Meerwaldt Foundation, and
the Netherlands Organization of Scientific Research (Rubicon);
no conflicts of interest.

References
1.
1
2.
2

3.
3

4.
4

5.
5

6.
6

7.
7

8.
8

9.
9

♦

Suarez JI, Tarr RW, Selman WR. Aneurysmal subarachnoid
hemorrhage. N Engl J Med 2006;354:387–396.
Lewis DH, Eskridge JM, Newell DW, et al. Brain SPECT and
effect of cerebral angioplasty in delayed ischemia due to vasospasm. J Nucl Med 1992:33:1789–1796.
Wintermark M, Maeder P, Thiran J-Ph, et al. Simultaneous
measurements of regional cerebral blood flows by perfusion-CT
and stable xenon-CT: a validation study. AJNR Am J Neuroradiol
2001;22:905–914.
Powers WJ, Grubb RL, Baker RP, Mintun MA, Raichle ME.
Regional cerebral blood flow and reversible ischemia due to
vasospasm. Determination by positron emission tomography.
J Neurosurg 1985;62:539–546.
Kudo K, Terae S, Katoh C, et al. Quantitative cerebral blood flow
measurement with dynamic perfusion CT using the vascularpixel elimination method: comparison with H2(15)O positron
emission tomography. AJNR Am J Neuroradiol 2003;24:419–426.
Nabavi DG, LeBlanc LM, Baxter B, et al. Monitoring cerebral
perfusion after subarachnoid hemorrhage using CT. Neuroradiology 2001;43:7–16.
Wintermark M, Ko NU, Smith WS, Liu S, Higashida RT, Dillon
WP. Vasospasm after subarachnoid hemorrhage: utility of perfusion CT and CT angiography on diagnosis and management.
AJNR Am J Neuroradiol 2006;27:26–34.
Sen J, Belli A, Albon H, Morgan L, Petzold A, Kitchen N. Triple-H
therapy in the management of aneurysmal subarchnoid haemorrhage. Lancet Neurology 2003;2:614–620.
Lee Kh, Lukovits T, Friedman JA. “Triple-H” therapy for cerebral
vasospasm following subarachnoid hemorrhage. Neurocrit Care
2006;4:68–76.

Volume 6, 2007

Majoie et al.

44
10.
10

11.
11

12.
12

13.
13

14.
14

Higashida RT, Halbach VV, Dowd CF, Dormandy B, Bell J,
Hieshima GB. Intravascular balloon dilatation therapy for intracranial arterial vasospasm: patient selection, technique and clinical results. Neurosurg Rev 1992;15:89–95.
Eskridge JM, McAuliffe W, Song JK, et al. Balloon angioplasty for
the treatment of vasopasm: results of first 50 cases. Neurosurgery
1998;42:510–517.
Firlik AD, Kaufman AM, Jungreis CA, Yonas H. Effect of transluminal angioplasty on cerebral blood flow in the management of
symptomatic vasospasm following aneurysmal subarachnoid
hemorrhage. J Neurosurg 1997;86:830–839.
Wintermark M, Maeder P, Thiran JP, Schnyder P, Meuli R.
Quantitative assessment of regional cerebral blood flows
by perfusion CT studies at low injection rates: a critical review
of the underlying theoretical models. Eur Radiol 2001;11:
1220–1230.
Wintermark M, Fischbein NJ, Smith WS, Ko NU, Quist M, Dillon
WP. Accuracy of dynamic perfusion CT with deconvolution in

Neurocritical Care

15.
15
16.
16
17.
17

18.
18
19.
19

20.
20

♦

detecting acute hemispheric stroke. AJNR Am J Neuroradiol
2005;26:102–112.
Eskridge JM, Song JK. A practical approach to the treatment of
vasospasm. AJNR Am J Neuroradiol 1997;18:1653–1660.
Hoeffner EG, Case I, Jain R, et al. Cerebral perfusion CT: technique and clinical applications. Radiology 2004;231:632–644.
Grubb RL, Raichle ME, Eichling JO, Gado MH. Effects of
subarachnoid hemorrhage on cerebral blood volume, blood flow
and oxygen utilization in humans. J Neurosurg 1977;46:446–453.
Miles KA, Griffiths MR. Perfusion CT: a worthwile enhancement?
Brit J Radiol;76:220–231.
Kealey SM, Loving VA, Delong DM, Eastwood JD. User-defined
vascular input function curves: influence on mean perfusion
parameter values and signal-to-noise ratio. Radiology 2004;231:
587–593.
Sanelli PC, Lev MH, Eastwood JD, Gonzalez RG, Lee TY. The effect
of varying user-selected input parameters on quantitative values
in CT perfusion maps. Acad Radiol 2004;11:1085–1092.

Volume 6, 2007