close

Вход

Забыли?

вход по аккаунту

?

Magnetic resonance angiography in reversible cerebral vasoconstriction syndromes.

код для вставкиСкачать
ORIGINAL ARTICLE
Magnetic Resonance Angiography in
Reversible Cerebral Vasoconstriction
Syndromes
Shih-Pin Chen, MD,1,2,3 Jong-Ling Fuh, MD,2,3 Shuu-Jiun Wang, MD,3,2
Feng-Chi Chang, MD,2,4 Jiing-Feng Lirng, MD,2,4 Ying-Chen Fang, MD,5
Ben-Chang Shia, PhD,6 and Jaw-Ching Wu, MD, PhD1
Objective: To investigate the evolution and clinical significance of vasoconstriction on magnetic resonance angiography (MRA) in patients with reversible cerebral vasoconstriction syndromes (RCVS).
Methods: Patients with RCVS were recruited and followed up with MRA examinations until normalization of
vasoconstriction or for 6 months. The vasoconstriction severity of the major cerebral arterial segments (M1, M2,
A1, A2, P1, P2, and basilar artery) was scored on a 5-point scale: 0 (0 –⬍10%), 1 (10 –⬍25%), 2 (25–⬍50%), 3
(50 –⬍75%), and 4 (ⱖ75%). Subjects with at least 1 segment with a vasoconstriction score ⱖ2 were eligible for the
study. Initial mean scores of single or combined arterial segments were used to predict ischemic complications.
Results: Seventy-seven patients with RCVS (8 men/69 women; average age 47.7 ⫾ 11.6 years) finished the study
with a total of 225 MRAs performed. The mean number of arterial segments involved was 5.3 ⫾ 3.0 in the initial
MRA. Vasoconstriction scores reached their maximum 16.3 ⫾ 10.2 days after headache onset, close to the average
timing of headache resolution (16.7 ⫾ 8.6 days). Vasoconstriction evolved in a parallel trend among different
arterial segments. Seven (9.1%) patients developed posterior reversible encephalopathy syndromes (PRES). Six
(7.8%) patients had ischemic stroke. A logistic regression model demonstrated that the M1–P2 combined score
was associated with highest risk of PRES (odds ratio [OR], 11.6, p ⫽ 0.005) and ischemic stroke (OR, 3.4; p ⫽
0.026).
Interpretation: MRA evaluation in patients with RCVS is valid. Vasoconstriction was pervasive and outlasted headache resolution. Vasoconstrictions in M1 and P2 are important determinants for PRES and ischemic stroke.
ANN NEUROL 2010;67:648 – 656
R
eversible cerebral vasoconstriction syndromes (RCVS)
comprise a group of disorders characterized by recurrent severe headaches (mostly thunderclap headaches) and
reversible cerebral vasoconstriction.1,2 RCVS can be spontaneous3,4 or evoked by various factors such as puerperium or exposure to vasoactive substances.5 Patients with
spontaneous RCVS tend to be middle-aged women,1,3
whereas such female preponderance is less distinct in patients with secondary RCVS.5 More than 31 of patients
have blood pressure (BP) surges (systolic BP ⬎
160mmHg) during headache attacks.3,5 RCVS are usually
self-limited to 3 months. Open-labeled trials suggest that
thunderclap headaches in patients with RCVS might be
responsive to calcium channel blockers, but vasoconstrictions might persist despite initiation of treatment.6 Some
patients are complicated by posterior reversible encephalopathy syndromes (PRES), ischemic strokes over watershed zones, cortical subarachnoid hemorrhage (SAH), or
intracranial hemorrhage.3,5 Our recent publication on
transcranial color-coded sonographic studies (TCCS)
demonstrated that the risk of PRES or ischemic stroke
was higher in those with a mean flow velocity of the middle cerebral artery (MCA) ⬎120m/s and a Lindegaard index (LI) ⬎3.4
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.21951
Received Aug 18, 2009, and in revised form Dec 8. Accepted for publication Dec 8, 2009.
Address correspondence to Dr Wang, Neurological Institute, Taipei Veterans General Hospital, Taipei, Taiwan, 112. E-mail: [email protected]
From the 1Institute of Clinical Medicine and 2Faculty of Medicine, National Yang-Ming University School of Medicine; 3Department of Neurology,
Neurological Institute and 4Department of Radiology, Taipei Veterans General Hospital; 5Department of Radiology, Hoping Branch, Taipei City
Hospital; and 6Department of Statistics and Information Science, Fu-Jen Catholic University, Taipei, Taiwan.
Additional Supporting Information may be found in the online version of this article.
648
© 2010 American Neurological Association
Chen et al: MRA Vasoconstrictions in RCVS
An angiographic finding of reversible cerebral vasoconstriction is essential for the diagnosis of RCVS. Conventional angiography has been the gold standard for
evaluating intracranial vessels; however, it is invasive and
not feasible for frequent follow-ups.7 Magnetic resonance
angiography (MRA), surpassing the limitations of conventional angiography, is a noninferior tool widely used to
evaluate vasoconstriction in patients with RCVS.3,5 The
temporal evolution and clinical significance of the angiographic findings of RCVS remain unclear. In this study,
we performed sequential MRA studies to investigate the
evolution of vasoconstriction in RCVS patients and to
clarify the clinical significance of these angiographic findings.
Subjects and Methods
Subjects
We recruited consecutive patients presenting with recurrent
thunderclap headaches and reversible cerebral vasoconstriction
from the headache clinic at Taipei Veterans General Hospital
(TVGH) between 2002 and 2009. The diagnosis of RCVS was
based on the criteria of the syndrome of “benign (or reversible)
angiography of the central nervous system” proposed by the International Classification of Headache Disorders, second edition
(code 6.7.3), except for the duration criterion D (Table 1).8 The
duration criterion was not followed because some patients took
⬎2 months to recover.3 The study protocol was approved by
the TVGH Institutional Review Board. All patients provided
informed consent before entering the study.
Therapeutic Strategies
We treated RCVS as an emergency condition so that the diagnostic evaluations, including the MRA, TCCS, and/or spinal
taps, were done on the first day when patients were seen. Nimodipine therapy was initiated immediately after confirmation
of the diagnosis except for those with spontaneous resolution of
headaches prior to presentation. All patients were given oral ni-
modipine 30 to 60mg every 4 hours, adjusted according to the
severity of vasoconstriction on the initial MRA. When thunderclap headaches or vasoconstrictions worsened despite oral nimodipine, or in the presence of PRES and/or ischemic stroke, we
used intravenous nimodipine (0.5–2mg/h) instead via a central
venous line, with BP monitored every 2 to 4 hours. Hypotension (systolic BP ⬍ 100mmHg), if it occurred, was corrected
immediately by dose tapering and hydration. BP surge, if it occurred during treatment, was treated by dose escalation and
closer BP monitoring.
Magnetic Resonance Studies
All subjects underwent sequential brain magnetic resonance imaging (MRI) with adequate sequences to exclude intracranial lesions or SAH,9,10 a procedure that has been detailed elsewhere.4
MRA was obtained using 3-dimensional time-of-flight magnetic
resonance technique (repetition time, default; echo time, minimum; flip angle, 45°; field of view, 18 –22cm; matrix, 224 ⫻
226; number of acquisitions, 1) with multislab reconstruction.
Postprocessing was performed by using a maximum intensity
projection (MIP) technique, generating in total 22 16.4° incremental rotations in somersault and clockwise rotation.
Vasoconstriction Grading and Scoring
The measurements of MRA were performed electronically on a
commercially available workstation (GE Advantage Workstation
Version 4.3, General Electric, Fairfield, CT). Two neuroradiologists (F.-C.C. and Y.-C.F) blinded to the clinical findings independently graded the degree of stenosis of intracranial vessels
based on source and MIP reconstruction images. If both images
visualized a vessel, the image demonstrating the less severely diseased vessel was measured to avoid overestimation.11 Once an
image type was selected for a given vessel, the single view with
the highest percentage of stenosis was measured. The extent of
vasoconstriction was derived from the following formula:
([Dn ⫺ Ds]/Dn) ⫻ 100, where Ds is the stenosed diameter and
Dn is the normal diameter of the arterial segment proximal to
the stenotic site. If the proximal segment was diseased, contin-
TABLE 1: Current Diagnostic Criteria of Benign (or Reversible) Angiopathy of the CNS in the International
Classification of Headache Disorders, 2nd Edition
6.7.3 Headache attributed to benign (or reversible) angiopathy of the CNS
A. Diffuse, severe headache of abrupt or progressive onset, with or without foca1 neurological deficits and/or
seizures and fulfilling criteria C and D
B. “Strings and beads” appearance on angiography and subarachnoid hemorrhage ruled out by appropriate
investigations
C. One or both of the following:
1. Headache develops simultaneously with neurological deficits and/or seizures
2. Headache leads to angiography and discovery of “strings and beads” appearance
D. Headache (and neurological deficits, if present) resolves spontaneously within 2 months
CNS ⫽ central nervous system.
May, 2010
649
ANNALS
of Neurology
gency sites were chosen to measure Dn: the distal artery (second
choice) and the feeding artery (third choice).12 The first and
second segments of the anterior cerebral artery (A1 and A2),
MCA (M1 and M2), and posterior cerebral artery (P1 and P2),
as well as the basilar artery (BA), were evaluated (Supplementary
Fig 1). The superior and inferior divisions of M2 were considered as a single arterial segment. The severity of vasoconstriction
was graded on a 5-point scale: 0 (0 –⬍10 %), 1 (10 –⬍25%), 2
(25–⬍50%), 3 (50 –⬍75%), and 4 (ⱖ75%). To avoid overestimation, this study defined ⭌1 arterial segment with grade ⭌2
vasoconstriction as a preselection criterion for RCVS. When
multiple vasoconstrictions coexisted in the same vascular segment, the most severe 1 was graded. The severity grading of
vasoconstriction was designated the vasoconstriction score of each
arterial segment for computations. For arterial segments with the
same designations, such as bilateral M1, M2, A1, A2, P1, and
P2, the vasoconstriction scores of both sides were averaged to
derive a mean score. To evaluate the effect of the combination
of different arterial segments, the mean scores of different arterial segments were averaged to derive a combined score. For example, the M1 mean score was derived by averaging bilateral
M1 vasoconstriction scores; the M1–P2 combined score denotes
the average of the mean scores of bilateral M1 and P2.
MRA Follow-up
Sequential MRAs were performed in all subjects to determine
whether the focal narrowing of cerebral arteries was reversible or
attributable to atherosclerosis or dissection. If an arterial seg-
ment had uniformly decreased caliber or was invisible throughout its length, and had no interval changes during sequential
follow-ups, it was considered hypoplastic, and no grading was
assigned.13 The follow-up intervals depended on patient clinical
conditions and magnetic resonance scanner availability. Reversibility was considered present if the vasoconstriction score of ⱖ1
arterial segment returned from higher scores (ie, 2– 4) to score 1
or 0. Follow-up MRA examinations were conducted until normalization of vasoconstrictions or for 6 months. To the increase
homogeneity of the study cohort, subjects who had their initial
MRA done 30 days after headache onset or had a delayed recovery of cerebral arteries (beyond 3 months) were excluded
from the final analysis.
Extracranial and Transcranial Color-Coded
Sonographic Studies
Patients recruited after October 2003 also received extracranial
color-coded sonographic studies (ECCS) and TCCS. A detailed
protocol has been previously described.4 Mean flow velocity of
MCA (VMCA) and LI, which was calculated by dividing the
VMCA by the mean flow velocity of the ipsilateral distal extracranial internal carotid artery (ICA), obtained within 1 day of
MRA performance, were employed to correlate with the grading
of MRA vasoconstriction. Although not the gold standard study,
ECCS were performed in every patient to exclude cervical artery
dissection. Only patients who had significant abnormalities disclosed by ECCS would receive further investigations.
Clinical Follow-up
All eligible patients were followed up until their headaches subsided or until they reached MRA follow-up endpoint.
Statistics
Descriptive statistics were presented as median, mean ⫾ standard deviation or percentages. Means were compared using the
unpaired t tests, and proportions were compared using chisquare tests. Inter-rater agreement of MRA vasoconstriction
grading was evaluated with quadratic weighted kappa. Spearman
correlation was used to evaluate the correlation of the vasoconstriction grading on MRA with VMCA and LI. The distribution
of vasoconstriction grading at different time periods was evaluated with Kruskal-Wallis analysis of variance tests. The time
trend curve of mean scores from each arterial segment was derived by the distance-weighted least squares method. A forward
logistic regression model was employed to evaluate the association of PRES and ischemic stroke with the mean score of a
selected arterial segment or the combined scores from different
arterial segments obtained from the initial MRA. All calculated p
values were 2-tailed, and statistical significance was defined as a
p value ⬍0.05.
Results
FIGURE 1: Screening scheme for eligible patients. RCVS ⴝ
reversible cerebral vasoconstriction syndromes; MRA ⴝ
magnetic resonance angiography.
650
Participants and Their Characteristics
During the study period, we recruited 138 patients with
recurrent thunderclap headaches. The screening process is
Volume 67, No. 5
Chen et al: MRA Vasoconstrictions in RCVS
illustrated in Figure 1. Seventy-seven patients (8 men and
69 women, mean age 47.7 ⫾ 11.6 years [range, 10 –76])
constituted the final sample for analysis. Among the 77 eligible subjects, 19 (24.7%) had a history of hypertension,
and 35 (45.5%) had BP surge during headache attacks.
The mean maximal systolic BP during the headache attacks
was 156.9 ⫾ 30.2 mmHg (range, 101–220mmHg). Three
subjects (3.9%) had type 2 diabetes mellitus, 13 (16.9%)
had migraine, and none had coronary artery disease.
Thirty-three (47.8%) women were postmenopausal, and 13
(16.9%) were under hormone therapy at disease onset. Six
patients had some possible secondary causes or associated
conditions, including the use of selective serotonin reuptake
inhibitor, the ingestion of pseudoephedrine, an unruptured
saccular aneurysm of the intracranial ICA, left distal vertebral artery dissection, a postpartum state, and microangiopathic hemolytic anemia. None of the patients had neck
stiffness. Fourteen (18.2%) patients received cerebrospinal
fluid studies, all of whom had normal findings. On average,
the patients had 7.0 ⫾ 5.4 (median, 5; range, 2–30) thunderclap headache attacks in a mean period of 16.7 ⫾ 8.6
(median, 15; range, 5– 42) days. Triggers were identifiable
in 62 (80.5%) patients.
MRI and MRA Findings
MRA PERFORMANCE AND FOLLOW-UP. In total,
225 MRAs including 2,925 arterial segments were evaluated, with a mean number of 2.9 (median, 3; range, 2–9)
MRA exams per subject. The initial MRA was performed
a mean of 10.4 ⫾ 7.0 (median, 9; range, 1–30) days after
headache onset. The mean follow-up interval was 26.5 ⫾
21.1 (median, 23.0; range, 1–75) days, and the mean
follow-up duration was 71.8 ⫾ 24.7 (range, 40 –90) days.
INTER-RATER AGREEMENT OF SCORING SYSTEM
OF MRA VASOCONSTRICTION. The inter-rater agree-
ment of MRA vasoconstriction scores for each arterial segment was calculated by the quadratic weighted kappa statistics as follows: M1 0.743, M2 0.693, A1 0.791, A2
0.775, P1 0.936, P2 0.710, and BA 0.736. The differences in grading were resolved by consensus.
CORRELATION BETWEEN MRA VASOCONSTRICTION SCORE AND TCCS RESULTS. Higher vasocon-
striction score had higher corresponding VMCA (r ⫽ 0.618,
p ⬍ .001) and LI (r ⫽ 0.504, p ⬍ 0.001) (Supplementary
Fig 2).
DISTRIBUTION AND SEVERITY OF VASOCONSTRICTION. The involved arterial segments and their relevant
mean score in the initial MRA are listed in the Supplementary Table. On average, 39.8% of the assessed arterial
May, 2010
segments had vasoconstrictions scored ⭌2. Most (58.5%)
of these vasoconstrictions had severity scores of 2. The
proportions of arterial segments with vasoconstriction
scores of 3 or 4 were higher in P2 than in the other arterial segments (Table 2). The number of arterial segments involved was 5.3 ⫾ 3.0 in the initial MRA examination.
TEMPORAL EVOLUTION OF VASOCONSTRICTION.
The highest mean numbers of arterial segment involvement occurred on day 16.2 ⫾ 9.4 (range, 2–51) after
headache onset, and the highest mean score was noted on
day 16.3 ⫾ 10.2 (range, 2–53); both were close to the
mean timing of headache resolution. The number of arterial segment involvement and the mean score declined
significantly 1 month after onset (see Table 2). Figure 2
demonstrates the temporal evolution of vasoconstriction
score of each arterial segment. The pattern was similar
among different arterial segments; all arterial curves
peaked within the initial 20 days, followed by an accelerated downsloping curvature. However, the timing and velocity of peaking and downsloping varied slightly among
different arterial segments (see Fig 2). When sequential
MRAs were compared in each individual, up to 62.3% of
the subjects (n ⫽ 48) had unsynchronized evolution of
vasoconstriction, that is, the vasoconstriction(s) in 1 arterial segment improved for at least 1 score, whereas the
vasoconstriction(s) in the other arterial segment(s) worsened for at least 1 score (Fig 3). In 28 patients who had
an early second MRA within the first 3 weeks after headache onset, 24 (85.7%) demonstrated an aggravation of
vasoconstrictions for at least 1 vasoconstriction score in at
least 1 arterial segment; all of them were on nimodipine.
OTHER VASCULAR ANOMALIES. Hypoplasia of the
cerebral arteries was noted in 12 subjects, with the P1
being the most common (see Supplementary Table). No
other arterial or venous anomalies were noted except for
the unruptured saccular aneurysm (n ⫽ 1) and left distal
vertebral artery dissection (n ⫽ 1). The dissection of the
left distal vertebral artery was visualized by brain MRA on
day 3 after headache onset, so that no neck MRA was
performed in this patient. The results of ECCS and
TCCS were also compatible with dissection. The patient’s
headache located at bilateral frontal areas, and no neck
pain was reported.
PRES, STROKES, AND WHITE MATTER HYPERINTENSE LESIONS. PRES was found in MRI at a mean
of 11.0 ⫾ 5.8 (range, 4 –20) days in 7 patients (9.1%).
Six patients developed ischemic strokes, with 4 concomitantly having PRES. The timing and findings of MR im651
ANNALS
of Neurology
FIGURE 2: Temporal trend of vasoconstriction scores of individual arterial segments in the first 90 days of disease course.
Conf. ⴝ confidence; BA ⴝ basilar artery.
652
Volume 67, No. 5
Chen et al: MRA Vasoconstrictions in RCVS
TABLE 2: Mean Number of Involved Arteries and Mean Scores of All Arterial Segments in Relation to the
Time Period
Duration after
Headache Onset, Days
Patient
No.
1-5
28
6-10
32
11-20
50
21-30
30
31-45
29
46-60
22
61-90
27
Number of Arterial Segments Involved,
MeanⴞSD (range)
1st Segment
2nd Segment
1st and 2nd
Segments
2.1 ⫾ 1.6
(0–6)
2.4 ⫾ 1.7
(0–6)
2.7 ⫾ 1.8
(0–6)
2.8 ⫾ 1.6
(0–6)
1.4 ⫾ 1.6
(0–5)a
1.3 ⫾ 1.4
(0–4)
0.8 ⫾ 1.0
(0–3)
2.4 ⫾ 1.5
(0–5)
3.1 ⫾ 1.7
(0–6)
3.1 ⫾ 2.0
(0–6)
2.5 ⫾ 1.9
(0–6)
4.5 ⫾ 2.6
(1–11)
5.6 ⫾ 3.0
(1–12)
5.8 ⫾ 3.5
(0–12)
5.3 ⫾ 3.1
(0–12)
1.3 ⫾ 1.7
(0–6)a
0.9 ⫾ 1.3
(0–5)
0.7 ⫾ 0.9
(0–3)
2.7 ⫾ 2.9
(0–11)a
2.2 ⫾ 1.9
(0–6)
1.5 ⫾ 1.3
(0–4)
Combined Score of 1st and
2nd Segments, MeanⴞSD
(range)
1.05 ⫾ 0.58 (0.81–1.29)
1.26 ⫾ 0.71 (0.99–1.53)
1.29 ⫾ 0.71 (1.09–1.49)
1.09 ⫾ 0.59 (0.87–1.31)
0.66 ⫾ 0.61 (0.38–0.94)a
0.56 ⫾ 0.35 (0.37–0.74)
0.39 ⫾ 0.30 (0.22–0.51)
Significant decrease ( p ⬍ 0.05) compared with the value from the previous time period.
SD ⫽ standard deviation.
a
ages as well as their temporal relationship with thunderclap headaches and nimodipine treatment are summarized
in Table 3. None of the patients had SAH (including cortical SAH) or intracerebral hemorrhage. Except for 1 seizure attack in Patient 4, none of our patients experienced
transient focal signs. Two patients (Patients 6 and 9) were
left with permanent visual field defects at 3-month
follow-up, whereas the others recovered well without clinically significant sequelae. Hyperintense white matter lesions were noted in 36 (46.7%) subjects. These white
matter lesions did not resolve during sequential followups.
Relationship between Vasoconstriction, PRES,
and Stroke
WHICH VESSELS WERE ASSOCIATED WITH RISK OF
PRES OR ISCHEMIC STROKE? Regarding single arte-
FIGURE 3: Vasoconstrictions in 1 patient with unsynchronized resolution. The magnetic resonance angiographies
were performed on days 8 (upper panel), 20 (middle
panel), and 80 (lower panel), respectively. Vasoconstrictions over bilateral M1 (solid arrows) improved on day 20,
but those on bilateral A1 were exacerbated (dashed arrows).
May, 2010
rial segment, the mean score of P2 has the highest explained variance in the association of PRES, and the
mean M1 score has the highest explained variance in the
association of ischemic stroke, as determined by the forward logistic regression model (Table 4). For the combination of 2 different arterial segments, the M1–P2 combined score was found to have the highest explained
variance in the association of PRES and ischemic stroke
653
ANNALS
of Neurology
TABLE 3: Summary of Timing of Thunderclap Headaches, Focal Deficits, Magnetic Resonance Imaging, and
Nimodipine Treatment in Patients with PRES or Ischemic Stroke
Patient
No.
Age,
yr/Sex
Thunderclap Headache
Focal Deficits
Duration
of Illness
(days)
No. of
Attacks
No. of
Days
after
Onset
Symptoms
2
Right limb
weakness, visual
dimming
Initial MRA,
No. of Days
after Onset
MRI Showing Complications
No. of
Days
after
Onset
PRES
Infarction
Nimodipine
Start, No.
of Days
after Onset
3
3
—
PWS
3
1
17/F
8
2
2
46/F
19
3
—
Asymptomatic
4
4
PWS
—
5
3
51/F
11
8
—
Asymptomatic
6
6
PWS
Right cerebellum
6
4
41/M
8
2
Seizure, stupor
8
8
PWS
PWS
8
5
41/F
10
10
Dyschromatopsia
9
9
PWS
—
9
6
24/F
17
17
14
Confusion,
blurred vision,
right hemiparesis
15
15
AWS⫹PWS⫹IWS
AWS⫹PWS⫹IWS
16
7
40/F
21
13
15
Blurred vision
15
15
PWS
—
15
8
54/F
10
5
16
Dizzy, blurred
vision, amnesia
15
16
—
PWS⫹IWS
16
9
51/F
16
24
17
Unsteadiness,
blurred vision
20
20
AWS⫹PWS⫹IWS
AWS⫹PWS⫹IWS
20
5
–
PRES ⫽ posterior reversible encephalopathy syndromes; MRA ⫽ magnetic resonance angiography; MRI ⫽ magnetic resonance
imaging; PWS ⫽ posterior watershed; AWS ⫽ anterior watershed; IWS ⫽ internal watershed.
(see Table 4). Combination of more arterial segments did
not raise the explained variance.
DID THE NUMBER OF INVOLVED ARTERIAL SEGMENTS ALSO REFLECT SEVERITY? The numbers of
involved first segments (total 7) (4.1 ⫾ 1.6 vs 2.2 ⫾ 1.6,
p ⫽ 0.002), second segments (total 6) (4.7 ⫾ 1.1 vs
2.8 ⫾ 1.7, p ⫽ 0.004), and all arterial segments (total 13)
(8.9 ⫾ 2.3 vs 5.0 ⫾ 2.8, p ⫽ 0.001) were significantly
higher in patients with PRES than those without. The
numbers of involved first segments (3.7 ⫾ 2.6 vs 2.3 ⫾
1.6, p ⫽ 0.044) were higher in patients with ischemic
strokes than those without, but the numbers of second
segments (3.5 ⫾ 2.4 vs 2.9 ⫾ 1.7, p ⫽ 0.419) and all
arterial segments (7.2 ⫾ 4.6 vs 5.2 ⫾ 2.8, p ⫽ 0.116) in
patients with ischemic stroke did not significantly differ
from those without.
Relationship between Vasoconstriction and
Blood Pressure
Patients with a history of hypertension had lower mean
scores of P2 (1.25 ⫾ 1.08 vs 1.88 ⫾ 1.14, p ⫽ 0.035)
and M2 (1.05 ⫾ 0.96 vs 1.73 ⫾ 0.96, p ⫽ 0.008) than
those without, whereas patients with BP surges during attacks had higher P2 mean scores than those without
surges (2.16 ⫾ 1.24 vs 1.49 ⫾ 0.95, p ⫽ 0.011). The
mean scores of the other arterial segments did not differ
between patients with or without hypertension or BP
surges during attacks.
TABLE 4: Best Model for PRES and Stroke Using a Forward Regression Model
Odds Ratios (95% CI)
p
Nagelkerke R2
1, P2
2, M1–P2
5.27 (1.69–16.40)
11.57 (2.06–67.85)
0.004
0.005
0.358
0.418
1, M1
2, M1-P2
2.94 (1.13–7.66)
3.38 (1.16–9.87)
0.027
0.026
0.170
0.158
Mean Vasoconstriction Score
PRES
Model
Model
Stroke
Model
Model
PRES ⫽ posterior reversible encephalopathy syndromes; CI ⫽ confidence interval.
654
Volume 67, No. 5
Chen et al: MRA Vasoconstrictions in RCVS
Clinical Follow-up after First Episode of RCVS
During a median follow-up of 25 (range, 10 – 69) months,
6 patients (8%) had recurrence. All of them reported a
clinical recurrence with multiple thunderclap headaches,
and their MRAs showed reversible vasoconstrictions.
Discussion
Our study showed that MRA is valuable for clinical evaluation and risk assessment in patients with RCVS. Both the
severity (mean score of vasoconstrictions) and distribution
(number of arterial segments involved) of vasoconstrictions
revealed by MRA were associated with the complications in
patients with RCVS, such as PRES or ischemic strokes. Vasoconstriction scores of M1–P2 appeared to be the most
important determinant of ischemic strokes and PRES.
These results were reliable because of the large sample size,
the considerable numbers of consecutive MRs performed,
and a long follow-up period in this study.
MRA has not replaced conventional angiography for
evaluating cranial vasculature, especially the distal arterial
branches or arterioles. However, it was reported that up
to 9% of RCVS patients develop transient focal neurological deficits after receiving angiography.5 Although we did
not compare our grading system with conventional angiography directly, the readings by 2 observers using the
MRA vasoconstriction scoring system for RCVS were consistent. The concordance with TCCS measures and the
linkage with complications suggest that this MRA scoring
system is also valid. Over-estimation by MRA in assessing
intracranial stenosis could be of concern.13–16 Nonetheless,
previous studies on vasospasm in SAH using similar grading methods demonstrated that MRA did not overestimate
vasospasm compared with angiography.16 In addition, such
bias could be minimized by employing a standard scoring
protocol, as was done in the present study.
The extent and degree of arterial involvement reached
their maximum during the first month even when the
headaches remitted. This may explain why ischemic attacks
can occur after headache resolution.17,18 In addition, these
seemingly parallel changes of time trend curves suggest that
vasoconstriction is a pervasive process in RCVS. The timing of MRA performance in our patients varied depending
on when the patient sought medical advice; thus, the worst
time point may have been missed. However, our study
showed that the severity of vasoconstriction on initial MRA
provided significant prognostic value. It is a strong reminder to clinicians to implement more aggressive strategies in those who have had severe vasoconstriction upon
initial presentation. However, it should be stressed that increasing nimodipine dosage could possibly worsen the patients’ conditions, and closely monitoring BP and avoidMay, 2010
ance of hypotension should be exercised during dose
escalation. Moreover, our study found that recurrence of
RCVS could occur in a small portion of patients with
RCVS. Further long-term prospective follow-up studies are
warranted to address this issue. It has been demonstrated
that, in patients with PRES, the lesions tended to locate at
watershed zones with posterior preponderance, and regional
cerebral blood flow and perfusion were significantly impaired in these lesions.19,20 In patients with RCVS, the
PRES lesions are of the same anatomical distribution as in
those with PRES due to other causes.17 We hypothesize
that PRES in RCVS patients may have a similar pathogenesis. The finding that more severe vasoconstrictions in M1
and P2 contributed to a higher risk of PRES supports this
hypothesis. A higher proportion of high-grade stenosis over
the P2 segments also augments the relationship between vasoconstrictions and PRES. A history of hypertension may
make P2 more inert to vasoconstriction due to vascular
structural autoregulation or adaptation.21 However, there
was a greater association of posterior circulation (P2) with
BP surge during thunderclap headache attacks. A relative
paucity of sympathetic innervation, which had been considered to be related to the defective autoregulation of posterior circulation,22 may partially explain this selective vulnerability. Because the MCA contributes to the majority of
cerebral blood flow, it is reasonable that hemodynamic
compromise due to severe constriction of M1 was the important determinant for the ischemic infarctions over watershed zones.
Cervical arterial dissection is frequently considered a
secondary cause of thunderclap headache; once identified, it
is treated, and concomitant changes of intracranial vessels
may have been omitted. One of our patients was found to
have left distal vertebral artery dissection. She was considered
as a patient with RCVS because in addition to recurrent
thunderclap headaches, MRA of her cerebral arteries demonstrated multifocal short segmental cerebral vasoconstrictions
that resolved during subsequent follow-ups. A recent report
also identified cervical artery dissection as an associated finding of RCVS.5 It is possible that the reversible underlying
vasculopathy makes arteries more likely to tear, and that may
also lead to superficial SAH in some patients. In other
words, arterial dissection and superficial SAH might be the
consequences rather than causes of RCVS. It is important for
clinicians to recognize these complications. When superficial
local regions of subarachnoid blood or dissections are found,
they should not be directly considered as the causes of the
headache, and unnecessary diagnostic studies and treatment
should be avoided.
Two important differential diagnoses of RCVS
should be addressed, SAH and primary angiitis of the cen655
ANNALS
of Neurology
tral nervous system (PACNS). Both SAH and RCVS could
exhibit thunderclap headaches and vasoconstriction (or vasospasm). However, the vasospasm in SAH tended to be
long-segmental and more prominent around the bleeding
focus,23 whereas the vasoconstrictions in RCVS are generally multiple, short-segmental, and without specific vascular
predilection. Both PACNS and RCVS had string-and-bead
appearances on angiographic findings. Although small vessel involvement is more common in PACNS (92%), large
vessel involvement is not uncommon (71%).24 It is difficult to make a distinction between these 2 syndromes based
solely on angiographic findings at an early stage. Clinical
features such as repetitive thunderclap headaches would be
more valuable for differentiating these 2 disorders.
Our study had some limitations. Because most of
our patients received nimodipine treatment, the clinical
course could have been altered. Because the patients are at
risk of ischemic complications, and a previous open-label
study demonstrated a possible beneficial effect of nimodipine,6 leaving them untreated was unjustified. In addition, this was the first study to use the mean score of
vasoconstrictions to evaluate the effect of combining different arterial segments based on the assumption that each
arterial segment had equal weight. The rationale for this
computation could be oversimplified. However, the predictive value and correlation to blood flow velocity partly
support the validity of this scoring method.
Acknowledgment
This study was supported by the National Science Council of Taiwan (97-2628-B-010-007-MY3 to S-J.W.).
The abstract of this paper was presented at the 14th
International Headache Congress in Philadelphia, PA,
September 10 –13, 2009.
Potential Conflicts of Interest
5.
Ducros A, Boukobza M, Porcher R, et al. The clinical and radiological spectrum of reversible cerebral vasoconstriction syndrome. A
prospective series of 67 patients. Brain 2007;130:3091–3101.
6.
Lu SR, Liao YC, Fuh JL, et al. Nimodipine for treatment of primary thunderclap headache. Neurology 2004;62:1414 –1416.
7.
Dodick DW. Thunderclap headache. J Neurol Neurosurg Psychiatry 2002;72:6 –11.
8.
Headache Classification Subcommittee of the International Headache
Society. The International Classification of Headache Disorders: 2nd
edition. Cephalalgia 2004;24(suppl 1):9–160.
9.
Mitchell P, Wilkinson ID, Hoggard N, et al. Detection of subarachnoid haemorrhage with magnetic resonance imaging.
J Neurol Neurosurg Psychiatry 2001;70:205–211.
10.
da Rocha AJ, da Silva CJ, Gama HP, et al. Comparison of magnetic
resonance imaging sequences with computed tomography to detect low-grade subarachnoid hemorrhage: role of fluid-attenuated
inversion recovery sequence. J Comput Assist Tomogr 2006;30:
295–303.
11.
Stroke Outcomes and Neuroimaging of Intracranial Atherosclerosis
(SONIA) Trail Investigators. Stroke outcome and neuroimaging of
intracranial atherosclerosis (SONIA): design of a prospective, multicenter trial of diagnostic tests. Neuroepidemiology 2004;23:23–32.
12.
Samuels OB, Joseph GJ, Lynn MJ, et al. A standardized method
for measuring intracranial arterial stenosis. AJNR Am J Neuroradiol 2000;21:643– 646.
13.
Bash S, Villablanca JP, Jahan R, et al. Intracranial vascular stenosis
and occlusive disease: evaluation with CT angiography, MR angiography, and digital subtraction angiography. AJNR Am J Neuroradiol
2005;26:1012–1021.
14.
Stock KW, Radue EW, Jacob AL, et al. Intracranial arteries: prospective blinded comparative study of MR angiography and DSA
in 50 patients. Radiology 1995;195:451– 456.
15.
Korogi Y, Takahashi M, Nakagawa T, et al. Intracranial vascular
stenosis and occlusion: MR angiographic findings. AJNR Am J
Neuroradiol 1997;18:135–143.
16.
Grandin CB, Cosnard G, Hammer F, et al. Vasospasm after subarachnoid hemorrhage: diagnosis with MR angiography. AJNR
Am J Neuroradiol 2000;21:1611–1617.
17.
Chen SP, Fuh JL, Lirng JF, Wang SJ. Is vasospasm requisite for
posterior leukoencephalopathy in patients with primary thunderclap headaches? Cephalalgia 2006;26:530 –536.
18.
Sturm JW, Macdonell RA. Recurrent thunderclap headache associated with reversible intracerebral vasospasm causing stroke.
Cephalalgia 2000;20:132–135.
19.
Brubaker LM, Smith JK, Lee YZ, et al. Hemodynamic and permeability changes in posterior reversible encephalopathy syndrome measured by dynamic susceptibility perfusion-weighted
MR imaging. AJNR Am J Neuroradiol 2005;26:825– 830.
20.
Bartynski WS, Boardman JF. Catheter angiography, MR angiography, and MR perfusion in posterior reversible encephalopathy
syndrome. AJNR Am J Neuroradiol 2008;29:447– 455.
Nothing to report.
References
1.
Calabrese LH, Dodick DW, Schwedt TJ, Singhal AB. Narrative
review: reversible cerebral vasoconstriction syndromes. Ann Intern Med 2007;146:34 – 44.
21.
Folkow B. “Structural autoregulation”—the local adaptation of
vascular beds to chronic changes in pressure. Ciba Found Symp
1983;100:56 –79.
2.
Singhal AB, Koroshetz WJ, Caplan LR. Reversible cerebral vasoconstriction syndromes. In: Caplan LR, ed. Uncommon causes of
stroke. 2nd ed. Cambridge, UK: Cambridge University Press,
2008:505–514.
22.
Beausang-Linder M, Bill A. Cerebral circulation in acute arterial
hypertension—protective effects of sympathetic nervous activity.
Acta Physiol Scand 1981;111:193–199.
23.
3.
Chen SP, Fuh JL, Lirng JF, et al. Recurrent primary thunderclap headache and benign CNS angiopathy: spectra of the
same disorder? Neurology 2006;67:2164 –2169.
Aralasmak A, Akyuz M, Ozkaynak C, et al. CT angiography and
perfusion imaging in patients with subarachnoid hemorrhage:
correlation of vasospasm to perfusion abnormality. Neuroradiology 2009;51:85–93.
4.
Chen SP, Fuh JL, Chang FC, et al. Transcranial color Doppler
study for reversible cerebral vasoconstriction syndromes. Ann
Neurol 2008;63:751–757.
24.
Salvarani C, Brown RD Jr, Calamia KT, et al. Primary central
nervous system vasculitis: analysis of 101 patients. Ann Neurol
2007;62:442– 451.
656
Volume 67, No. 5
Документ
Категория
Без категории
Просмотров
5
Размер файла
504 Кб
Теги
angiography, syndrome, reversible, vasoconstriction, magnetic, resonance, cerebral
1/--страниц
Пожаловаться на содержимое документа