Poor Performance of Current Prognostic Scores for Early Risk of Recurrence After Minor

Stroke
Arvind Chandratheva, Olivia C. Geraghty and Peter M. Rothwell

Stroke. 2011;42:632-637; originally published online January 27, 2011;

doi: 10.1161/STROKEAHA.110.593301
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2011 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/42/3/632

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:

http://stroke.ahajournals.org//subscriptions/

Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012

 Poor Performance of Current Prognostic Scores for Early
Risk of Recurrence After Minor Stroke
Arvind Chandratheva, MRCP; Olivia C. Geraghty, MRCP; Peter M. Rothwell, MD, PhD, FRCP, FMedSci

Background and Purpose—The ABCD2 score predicts the early risk of stroke after transient ischemic attack. The early
risk of recurrence after minor stroke is as high but the only validated prognostic scores for use in minor stroke predict
long-term risk of recurrence: the Essen Stroke Risk Score and the Stroke Prognosis Instrument II.
Methods—We determined the prognostic value of the ABCD2 score, Essen Stroke Risk Score, and Stroke Prognosis
Instrument II in a prospective population-based study in Oxfordshire, UK, of all incident and recurrent stroke (Oxford
Vascular Study). Minor stroke was defined as an National Institutes of Health Stroke Scale score ⱕ5 at the time of first
assessment. The 90-day risks of recurrent stroke were determined in relation to each score. Areas under the receiver
operator curves indicated predictive value.
Results—Of 1247 first events in the study period, 488 were transient ischemic attacks, 520 were minor strokes, and 239
were major strokes. The ABCD2 score was modestly predictive (area under the receiver operator curve, 0.64; 0.53 to
0.74; P⫽0.03) of recurrence at 7 days after minor stroke and at 90 days (0.62; 0.54 to 0.70; P⫽0.004). Neither Essen
Stroke Risk Score (0.50; 0.42 to 0.59; P⫽0.95) nor Stroke Prognosis Instrument II (0.48; 0.39 to 0.60; P⫽0.92) were
predictive of 7-day or 90-day risk of recurrent stroke. Of the traditional vascular risk factors, etiologic classification
(Trial of ORG 10172 in Acute Stroke Treatment) and variables in the ABCD2 score, only blood pressure
⬎140/90 mm Hg (hazard ratio, 2.75; 1.18 to 6.38; P⫽0.02) and large artery disease (hazard ratio, 2.21; 1.00 to 4.88;
P⫽0.05) were predictive of 90-day risk.
Conclusions—The predictive power of the ABCD2 score is modest in patients with minor stroke, and neither the Essen
Stroke Risk Score nor the Stroke Prognosis Instrument II predicts early recurrence. More reliable early risk prediction
after minor stroke is required. (Stroke. 2011;42:632-637.)
Key Words: minor stroke 䡲 prognosis 䡲 risk factors

A t least 48 000 transient ischemic attacks (TIAs) and

43 000 minor strokes are managed as outpatients each
year in England alone and approximately 150 000 suspected
TIAs and minor strokes are referred to secondary care for
assessment and investigation,1 with rates similar in the United
States.2 Recent hospital- and population-based studies have
shown high early risks of recurrent stroke after TIA.2–7
Similar risks have been observed after minor stroke.8 –10
Clinical risk scores have been developed to stratify and select
patients at highest risk of early recurrence after TIA. One
such score, the ABCD score, is based on clinical character-
istics detected at the time of first assessment. This score has
been refined further to include diabetes (ABCD2)11 and more
recently imaging (ABCD2-I) and has been incorporated into
clinical guidelines to assist triaging and risk stratifying
patients.12–14 The ABCD2 score is now recommended for use
in triaging patients with suspected TIA, but its role in minor
stroke is yet to be determined.12–17
Other validated clinical risk prediction tools exist for
long-term risk of recurrent stroke after TIA or stroke. These
include the Essen Stroke Risk Score (ESRS), a 10-point scale
used to predict 1-year risk of recurrent stroke derived from
the data subset of 6433 cerebrovascular patients in the
Clopidogrel versus Aspirin in patients at Risk of Ischemic
events (CAPRIE) trial,18,19 and the Stroke Prognosis Instru-
ment II (SPI-II) a 15-point score consisting of 7 factors used
to predict 2-year risk of recurrent stroke validated in 4 test
cohorts,20 but neither has been validated for 90-day risk of
stroke recurrence. In addition, previous validations were
based on data collected from trials in which patients were
recruited weeks or months after the initial event, which would
not therefore have included many early recurrences. The aim
of our study was to determine whether the ABCD,2 ESRS,
and SPI-II scores were predictive of recurrence in the acute
and subacute phases for patients presenting with minor
stroke.

Received June 11, 2010; final revision received October 1, 2010; accepted October 4, 2010.
From the Department of Clinical Neurology, University of Oxford, on behalf of the Oxford Vascular Study, Stroke Prevention Research Unit, Oxford
University Department of Clinical Neurology, John Radcliffe Hospital, Oxford, UK.
Correspondence to Peter M. Rothwell, MD, PhD, FRCP, FMedSci, Professor of Clinical Neurology, University of Oxford, on behalf of the Oxford
Vascular Study, Stroke Prevention Research Unit, Oxford University Department of Clinical Neurology, Level 6, West Wing, John Radcliffe Hospital,
OX3 9DU, Oxford, UK. E-mail peter.rothwell@clneuro.ox.ac.uk
© 2011 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.110.593301

Downloaded from http://stroke.ahajournals.org/

632 by guest on December 21, 2012
 Chandratheva et al
Methods
The Oxford Vascular Study (OXVASC) is a population-based study
of all stroke and TIAd in 91 105 individuals of all ages registered
with 63 general practitioners in Oxfordshire, UK. This article
includes all events ascertained from April 1, 2002, to March 31,
2007. The study methods have been described elsewhere.21,22
Briefly, multiple overlapping methods of “hot” and “cold” pursuit
were used to achieve near complete ascertainment of all individuals
with TIA or stroke.21–23 These include: (1) a daily weekday, rapid
access “TIA clinic” to which participating general practitioners and
the local accident and emergency department refer all individuals
with suspected TIA or minor stroke whom they would not normally
admit directly to the hospital; (2) daily searches of admissions to the
medical, stroke, neurology, and other relevant wards; and (3) daily
searches of the local accident and emergency department
attendance register.
To not miss patients who presented late or had events out of the
area, patients who were referred to other services, or patients who
were not referred to secondary care, we also performed monthly
computerized searches of family doctor diagnostic coding, hospital
discharge codes, and all cranial and carotid imaging studies per-
formed in local hospitals.
All patients were consented and seen by study physicians as soon
as possible after their initial presentation. Overall, 70% were seen
within 24 hours of the onset of symptoms. Event characteristics and
risk factors were recorded and most patients underwent brain
imaging (91%). All cases were subsequently reviewed by the study
senior neurologist (P.M.R.) and classified using SDs.21,22
All patients were followed up formally face to face by a study
nurse or physician. Recurrent symptoms, medications, and disability
scores were recorded. Recurrent strokes that presented to medical
attention would also be identified acutely by ongoing daily case
ascertainment within OXVASC.
Baseline characteristics were recorded in all patients and assess-
ments were made for severity of event (using National Institutes of
Health Stroke Scale24), territory, and clinical features. All patients
with recurrent events were reassessed by a study physician and then
case-reviewed by P.M.R. with most receiving repeat brain imaging.
Severity of the recurrent event was based on clinical examination at
the time of assessment by the study physician.
Analysis
We restricted analysis to the risk of stroke after the first minor stroke
in the study period. Events were classified as minor stroke if there
was a focal neurological deficit lasting ⬎24 hours and a National
Institutes of Health Stroke Scale score ⱕ5 at the time of assessment
by a study physician. A cutoff of National Institutes of Health Stroke
Scale score ⱕ5 was chosen because this included the majority (97%)
of strokes seen in the outpatient setting. The study denominator and
recurrence rate were defined both from time of the event that led to
presentation and from the time that the patient first sought medical
attention. We included patients regardless of whether they sought
medical attention for their initial symptoms, that is, if patients
described symptoms of acute onset focal neurological deficit con-
sistent with minor stroke preceding the presenting event but had not
sought medical attention. However, we analyzed the risk of stroke
within 7 and 90 days after minor stroke both including and excluding
patients who did not seek medical attention for their initial symptoms
and only presented after the recurrent stroke. In the analysis
excluding patients who only sought medical attention after a recur-
rent stroke, 508 (98%) patients were prospectively assessed before
the recurrent stroke. Of the 61 recurrent strokes, 49 (80%) were
prospectively assessed before their recurrence.
A recurrent stroke was defined as a new and persistent neurolog-
ical symptom in a patient in whom the initial symptoms had already
substantially or fully recovered; most of these patients had repeat
brain imaging.
The risk of recurrent stroke was stratified by the 3 scores: (1) the
ABCD2 score, a validated clinical risk prediction tool for triage
assessment of TIA from 2 to 7 days,7,11 was calculated as follows:
Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012
Poor Performance of Current Prognostic Scores 633
ABCD2 score⫺age ⬎60 years⫽1, blood pressure ⬎140 and/or
ⱖ90 mm Hg⫽1, clinical: unilateral weakness⫽2, speech distur-
bance⫽1, duration of symptoms: ⱖ60 minutes⫽2, 10 to 59⫽1, ⬍10
minutes⫽0 and diabetes presence of which⫽1; (2) the ESRS, a
validated 10-point linear scale to predict 1-year risk of recurrent
stroke: age ⬎75 years (2 points), ⱖ65 to 75 (1 point), arterial
hypertension (1 point), diabetes mellitus (1 point), previous myocar-
dial infarction (1 point), other cardiovascular disease except myo-
cardial infarction and atrial fibrillation (1 point), peripheral arterial
disease (1 point), current or past smoking ⬍5 years (1 point), and
previous TIA or ischemic stroke in addition to qualifying event (1
point). The score was categorized to low risk (0 to 2) and high risk
(ⱖ3); and (3) the SPI-II, a validated 15-point score of 7 factors
validated for 2-year risk of recurrent stroke after TIA and minor
stroke: congestive heart failure (3 points), prior stroke (3 points),
diabetes (3 points), age ⬎70 years (2 points), stroke for the index
event (2 points), hypertension (1 point), and coronary artery disease
(1 point). Risk Groups I, II, III comprised patients with 0 to 3, 4 to
7, and 8 to 15 points, respectively.
Areas under the receiver operator curves and 95% CIs were
calculated as a measure of predictive ability with ideal prediction
producing a value of 1.00, whereas no better than chance prediction
is represented by 0.50. Probability values that we quote are based on
the difference between the observed predictive value (AUROC) and
chance (ie, no predictive value).
Results
Of 1247 first events in the study period, 488 were TIAs, 520
were minor strokes, and 239 were major strokes. The mean
age at first event was 73 years (range, 24 to 98 years).
Forty-seven percent of patients were female. There were 142
recurrent strokes within 90 days (81, 111, and 142 within 7,
30, and 90 days, respectively).
Of the 142 recurrent strokes within 90 days of a first event,
72 (51%) were preceded by a TIA, 61 (43%) by a minor
stroke, and 9 (6%) by major strokes. The 90-day risks of
stroke after TIA, minor stroke, and major stroke were 14.8%
(11.7 to 17.9), 11.7% (9.0 to 14.4), and 3.8% (1.4 to 6.2),
respectively.
Table 1 shows the baseline characteristics of all cases of
minor stroke. Data were missing on blood pressure at first
assessment in 6 patients to complete calculation of the
ABCD2 score leaving 514 of 520 patients with minor stroke
(99%). Table 2 shows the 90-day risk of recurrent stroke after
minor stroke stratified by ABCD2 score. The score was
poorly predictive of stroke recurrence within 90 days in
patients presenting with minor stroke (AUROC, 0.60; 0.52 to
0.67; P⫽0.02), although there was a trend toward higher risks
in patients with the highest scores (ABCD2 ⱖ5: 47 of 349
[13.5%] versus 14 of 165 [8.0%]; P⫽0.11) and the score had
limited predictive value at 7 days (AUROC, 0.57; 0.47 to
0.68; P⫽0.20). The equivalent results measured from time of
first calling for medical attention after the minor stroke were
7-day AUROC⫽0.64 (0.53 to 0.74; P⫽0.03) and 90-day
AUROC⫽0.62 (0.54 to 0.70; P⫽0.006). Neither ESRS (7-
day AUROC, 0.49; 0.35 to 0.62; P⫽0.81; 90-day AUROC,
0.50; 0.42 to 0.59; P⫽0.95) nor SPI-II (7-day AUROC, 0.50;
0.37 to 0.64; P⫽0.99; 90-day AUROC, 0.48; 0.39 to 0.60; P⫽0.92)
predicted risk of recurrent stroke (Table 3).
Table 4 shows the associations between the individual
components of the ABCD2 score and 90-day risk of recurrent
stroke. Only blood pressure ⱖ140/90 mm Hg significantly
predicted 7-day risk (hazard ratio, 8.14; 95% CI, 1.11 to
634 Stroke March 2011

Table 1. Baseline Characteristics of Patients Presenting With Minor Strokes Stratified According to
Clinical Characteristics at Presentation, Vascular Risk Factors, and Premorbid Medication
Minor Stroke Minor Stroke
With Recurrence Without Recurrence
Total (n⫽520) (n⫽61) (n⫽459)
Age, mean years (SD) 73.4 (12.3) 74.3 (12.9) 73.3 (12.3)
Male sex 241 (54%) 23 (42%) 218 (56%)
Characteristics of the index event
Unilateral weakness 303 (58%) 40 (66%) 263 (57%)
Speech disturbance without weakness 72 (14%) 9 (15%) 63 (14%)
No speech disturbance or weakness 145 (28%) 12 (20%) 133 (29%)
Vascular risk factors
Systolic BP, mean mm Hg (SD) 156.6 (28.2) 160.9 (27.4) 156.0 (28.1)
Diastolic BP, mean mm Hg (SD) 83.4 (15.4) 85.3 (17.9) 83.1 (15.0)
Hypertension 303 (58%) 35 (57%) 268 (58%)
Diabetes 58 (11%) 11 (18%) 47 (10%)
Coronary heart disease 111 (21%) 11 (18%) 100 (22%)
Heart failure 46 (9%) 3 (5%) 43 (9%)
Peripheral vascular disease 35 (8%) 4 (7%) 39 (9%)
Previous TIA 54 (10%) 7 (12%) 47 (10%)
Previous stroke 84 (16%) 9 (15%) 75 (16%)
Previously diagnosed atrial fibrillation 88 (17%) 6 (10%) 82 (18%)
Current smoker 79 (15%) 15 (25%) 64 (14%)
Prior antiplatelet 175 (34%) 18 (30%) 157 (34%)
Prior statin 96 (19%) 8 (13%) 88 (19%)
Prior antihypertensive 272 (52%) 32 (53%) 240 (52%)
BP indicates blood pressure.

59.81; P⫽0.04) and 90-day risk (hazard ratio, 2.75; 95% CI,
1.18 to 6.38; P⫽0.02). Ninety-day stroke risk was also
stratified according to underlying etiology (Trial of ORG
10172 in Acute Stroke Treatment classification) and premor-
bid vascular risk factors. Large artery disease was predictive
(2.21; 1.00 to 4.88; P⫽0.05), but cardioembolic (1.23; 0.58 to
2.60; P⫽0.60) and small vessel disease (1.08; 0.51 to 2.28;
P⫽0.68) were not. Previous cerebral, peripheral, or cardiac
disease was not predictive of recurrent stroke.
By comparison in patients presenting with TIA, the
ABCD2 score was predictive of stroke recurrence within 7
days (AUROC, 0.71; 0.63 to 0.79; P⬍0.001) and moderately
predictive at 90 days (AUROC, 0.64; 0.56 to 0.71; P⬍0.001).
Neither ESRS (7-day AUROC, 0.51; 0.44 to 0.59; P⫽0.74;
90-day AUROC, 0.47; 0.36 to 0.59; P⫽0.60) nor SPI-II
(7-day AUROC, 0.50; 0.42 to 0.58; P⫽0.96; 90-day AUROC,
0.49; 0.38 to 0.60; P⫽0.84) was predictive at 7 or 90 days,
respectively.

Table 2. Seven- and 90-Day Risk of Stroke Stratified According to ABCD2 Score at First
Assessment in Patients Presenting With Minor Stroke
Stroke ⱕ7 Days After Stroke ⱕ90 Days After
Seeking Medical Attention Seeking Medical Attention

ABCD2 Patients (%) Strokes (%) Percent Risk (95% CI) Strokes (%) Percent Risk (95% CI)
1 0 0 0 0 0
2 5 (1%) 0 0 0 0
3 65 (13%) 1 (5%) 1.5 (0–4.6) 2 (4%) 3.1 (0–7.4)
4 95 (18%) 2 (9%) 2.1 (0–5.0) 7 (14%) 7.4 (2.1–12.7)
5 135 (26%) 5 (23%) 3.7 (0–6.8) 11 (22%) 8.1 (3.4–12.8)
6 196 (38%) 13 (59%) 6.6 (3.1–10.1) 26 (53%) 13.3 (8.6–18.0)
7 18 (4%) 1 (5%) 5.6 (0–16.2) 3 (6%) 16.7 (0–34.3)
Total 514 (100%) 22 (100%) 4.2 (2.4–6.0) 49 (100%) 9.4 (6.9–11.9)
P value for trend 0.030 0.004
AUROC 0.64 (0.53–0.74) 0.62 (0.54–0.70)
Total no. of minor strokes was 520. Complete data required for completion of ABCD2 score unavailable for 6 cases.

Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012

 Chandratheva et al Poor Performance of Current Prognostic Scores 635

Table 3. Risk of Recurrent Stroke After Minor Stroke at 90 after minor stroke up to 90 days and that neither the ESRS nor
Days Stratified by ESRS and SPI-II Scores* the SPI-II was predictive.
Patients Strokes Percent Risk Previous studies of clinical risk scoring in minor stroke
Clinical Score (%) (%) (95% CI) have varied in methodology with most excluding the acute
ESRS
phase and also those patients who may have had a preceding
event for which they did not seek medical attention before
⬍3 236 (45%) 21 (43%) 8.9 (5.2–12.6)
hospital admission.10,18,20,27 They also vary in population size,
ⱖ3 284 (55%) 28 (57%) 9.9 (6.4–13.4)
follow-up, and completeness of case ascertainment within the
0 25 (5%) 1 (2%) 4.0 (0–11.6) defined population. We performed the first prospective
1 87 (17%) 10 (20%) 11.5 (4.6–18.4) population-based study of risk scoring in patients with minor
2 124 (24%) 10 (20%) 8.1 (3.2–13.0) stroke.
3 138 (27%) 15 (31%) 10.9 (5.6–16.2) Although the ABCD2 score was modestly predictive after
4 80 (15%) 6 (12%) 7.5 (1.6–13.4) minor stroke, 1 argument is that the score may act as a
5 49 (9%) 5 (10%) 10.2 (1.8–18.6)
diagnostic tool to identify patients more likely to have a true
TIA or minor stroke because diagnosis is not completely
6 13 (3%) 2 (4%) 15.4 (0–35.0)
reliable28 and some patients with seizure or migraine are
7 3 (0.6%) 0 0 indistinguishable from those due to brain or retinal ische-
8 1 (0.2%) 0 0 mia.29 However, diagnostic accuracy is likely to be greater in
9 0 0 0 patients with minor stroke compared with those with transient
Total 520 (100%) 49 (100%) symptoms. Moreover, previous work has shown that in
P value for trend 0.84 analysis limited to neurologist-confirmed TIA, the ABCD2
AUROC 0.51 (0.42–0.59) score remains predictive.11 Also, in a recent multicenter
international collaborative analysis of patients with diffusion-
SPI–II
weighted imaging-positive TIA, the ABCD2 score was still
I 314 (61%) 31 (63%) 9.9 (6.6–13.2)
predictive of early recurrent stroke.30
II 172 (33%) 15 (31%) 8.7 (4.4–13.0) In a univariate analysis of the individual components of the
III 28 (5%) 3 (6%) 10.7 (0–22.1) ABCD2 score, vascular risk factors, and Trial of ORG 10172
0 0 0 0 in Acute Stroke Treatment criteria, only hypertension and
1 0 0 0 large artery disease were predictive of recurrent stroke with a
2 96 (19%) 11 (22%) 11.5 (5.0–18.0) nonsignificant trend for a combination of motor and speech
3 30 (6%) 2 (4%) 6.7 (0–15.9) symptoms, diabetes, and anterior circulation of event. In a
recent prospective hospital series by Ois et al, recurrence after
4 149 (29%) 16 (33%) 10.7 (5.6–15.8)
minor stroke was associated with the presence of weakness,
5 82 (16%) 6 (12%) 7.3 (1.6–13.0)
speech disturbance, high alcohol intake, previous TIA, and
6 21 (4%) 0 0 severe symptomatic extra- or intracranial arterial disease.10
7 63 (12%) 8 (16%) 12.7 (4.3–21.1) Cerebral and vascular imaging studies could enhance
8 50 (10%) 2 (4%) 4.0 (0–9.7) prediction of stroke after minor stroke. The presence of new
9 7 (1%) 2 (4%) 28.6 (0–62.1) ischemic lesions on MRI or CT has been associated with an
10 6 (1%) 0 0 increased short-term risk of stroke as has severe symptomatic
11 9 (2%) 1 (2%) 11.1 (0–31.7) arterial disease.10,31–33 Our study did not look at the role of
MRI in improving risk stratification in patients with minor
12 0 0 0
stroke. However, MRI as the first and immediate assessment
13 1 (0.2%) 1 (2%) 0
tool is not yet commonplace in daily clinical practice.34 One
14 0 0 0 benefit of purely clinical scoring systems with simple, easily
15 0 0 0 obtainable information is that they can be used at the point of
Total 514 (100%) 49 (100%) first presentation to facilitate rapid triage by first-line health-
P value for trend 0.89 care professionals. Further research is required to determine
AUROC 0.48 (0.39–0.60) whether newer technologies such as biomarkers of cerebral
*Total no. of minor strokes was 520. Complete data required for completion
ischemia or thrombosis can add prognostically useful infor-
of SPI-II score unavailable for 6 cases. mation in patients presenting with minor stroke.
Our study had some potential limitations. First, we cannot
be completely precise about the risk of stroke in the acute
Discussion phase because an unknown proportion of patients with minor
Minor strokes account for a significant proportion of patients stroke will never seek medical attention. Second, we might
with acute cerebrovascular events1 and the early risk of have underestimated the early risk of stroke slightly because
recurrence is similar to that after TIA.8 Although the ABCD2 some patients with major stroke with a preceding minor
score has been shown to be effective in prediction of recurrent stroke may not have been identified because we excluded
stroke after TIA in the acute phase,11,24 –26 our study has those in whom it was impossible to obtain a definite history
shown that it is a relatively poor predictor of recurrent stroke of TIA or minor stroke because they were aphasic, confused,
Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012
636 Stroke March 2011

Table 4. Ninety-Day Risk of Stroke and HR (95% CI) for Minor Stroke in Relation to Potential
Risk Factors
7-Day Risk 90-Day Risk
HR (95% CI) P HR (95% CI) P
Age 0.77 (0.30 –2.03) 0.60 1.07 (0.51–2.25) 0.86
Blood pressure ⬎140/90 mm Hg 8.14 (1.11–59.81) 0.04 2.75 (1.18–6.38) 0.02
Motor symptoms 1.25 (0.52–3.00) 0.62 1.62 (0.85–3.09) 0.14
Speech symptoms 0.94 (0.33–2.69) 0.90 1.52 (0.64–3.61) 0.34
Combined motor or speech symptoms 1.24 (0.53–2.91) 0.53 1.62 (0.86–3.05) 0.13
Diabetes 2.15 (0.69–4.83) 0.10 1.84 (0.96–3.59) 0.70
Coronary heart disease 1.43 (0.63–3.23) 0.39 0.82 (0.43–1.57) 0.54
Heart failure 0.36 (0.05–2.64) 0.31 0.50 (0.16–1.61) 0.25
Stroke 0.83 (0.29–2.37) 0.72 0.89 (0.44–1.81) 0.76
TIA 1.41 (0.49–4.06) 0.52 1.13 (0.52–2.49) 0.76
Peripheral vascular disease 0.39 (0.05–2.88) 0.36 0.76 (0.28–2.10) 0.60
Hypertension 0.88 (0.42–1.82) 0.72 0.96 (0.58–1.59) 0.87
Atrial fibrillation 0.56 (0.17–1.84) 0.34 0.52 (0.22–1.21) 0.13
Prior antiplatelet 0.74 (0.33–1.66) 0.46 0.80 (0.46–1.39) 0.43
Prior statin 0.50 (0.15–1.66) 0.26 0.65 (0.31–1.37) 0.26
Prior antihypertensive 1.12 (0.54–2.34) 0.76 1.00 (0.61–1.66) 0.99
Vascular territory
Anterior circulation 1.51 (0.64–3.57) 0.35 1.45 (0.80–2.62) 0.22
Posterior circulation 0.79 (0.32–1.95) 0.74 0.83 (0.45–1.55) 0.56
TOAST subtype
Ischemic
Undetermined 1 1
Small vessel disease 1.44 (0.50–4.10) 0.50 1.08 (0.51–2.28) 0.85
Large artery disease 2.49 (0.79–7.86) 0.12 2.21 (1.00–4.88) 0.05
Cardioembolic 1.15 (0.37–3.64) 0.81 1.23 (0.58–2.60) 0.60
Other 0 1.11 (0.15–8.33) 0.79
Unknown 1.67 (0.49–5.69) 0.42 1.32 (0.51–2.28) 0.55
Hemorrhagic 0.42 (0.06–3.05) 0.39 0.59 (0.18–1.87) 0.37
HR indicates hazard ratio; TOAST, Trial of Org 10172 in Acute Stroke Treatment.

or unconscious. Third, our validation of the clinical scores for
recurrences in the acute phase was based on relatively small
numbers of outcomes and so further studies would help to
confirm or refute our findings. Fourth, our study included
some patients recruited into the Effect of urgent treatment of
transient ischemic attack and minor stroke on early recurrent
stroke (EXPRESS) study,9 Phase 2, which involved more
urgent investigation and treatment and which will have
reduced the absolute risk of events and possibly influenced
the association with particular risk factors.
In summary, although the ABCD2 score is highly predic-
tive of recurrent stroke after TIA in the acute phase, it is less
predictive after minor stroke, and neither the ESRS nor the
SPI-II predict 90-day recurrence. More reliable early predic-
tion after minor stroke is required.
Acknowledgments
We thank all primary care practices and physicians who collaborate
with the Oxford Vascular Study, the details of which have been
published previously.19,20
Sources of Funding
The study was funded by the UK Medical Research Council, the
National Institute of Health Research (NIHR), the Stroke Associa-
tion, the Dunhill Medical Trust, and the NIHR Biomedical Research
Centre, Oxford.
Disclosures
None.
References
1. Giles MF, Rothwell PM. Substantial underestimation of the need for
outpatient services for TIA and minor stroke. Age Ageing. 2007;36:
676 – 680.
2. Kleindorfer D, Panagos P, Pancioli A, Khoury J, Kissela B, Woo D,
Schneider A, Alwell K, Jauch E, Miller R, Moomaw C, Shukla R,
Broderick JP. Incidence and short-term prognosis of transient ischemic
attack in a population-based study. Stroke. 2005;36:720 –723.
3. Hill MD, Yiannakoulias N, Jeerakathil T, Tu JV, Svenson LW, Schop-
flocher DP. The high risk of stroke immediately after transient ischemic
attack: a population-based study. Neurology. 2004;62:2015–2020.
4. Johnston SC, Gress DR, Browner WS, Sidney S. Short-term prognosis
after emergency department diagnosis of TIA. JAMA. 2000;284:
2901–2906.

Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012

 Chandratheva et al
5. Lisabeth LD, Ireland JK, Risser JM, Brown DL, Smith MA, Garcia NM,
Morgenstern LB. Stroke risk after transient ischemic attack in a
population-based setting. Stroke. 2004;35:1842–1846.
6. Lovett JK, Dennis MS, Sandercock PA, Bamford J, Warlow CP, Rothwell
PM. Very early risk of stroke after a first transient ischemic attack. Stroke.
2003;34:e138 –140.
7. Rothwell PM, Giles MF, Flossmann E, Lovelock CE, Redgrave JN,
Warlow CP, Mehta Z. A simple score (ABCD) to identify individuals at
high early risk of stroke after transient ischaemic attack. Lancet. 2005;
366:29 –36.
8. Coull AJ, Lovett JK, Rothwell PM. Population based study of early risk
of stroke after transient ischaemic attack or minor stroke: implications for
public education and organisation of services. BMJ. 2004;328:326.
9. Rothwell PM, Giles MF, Chandratheva A, Marquardt L, Geraghty O,
Redgrave JN, Lovelock CE, Binney LE, Bul LM, Cuthbertson FC, Welch
SJ, Bosch S, Carasco-Alexander F, Silver LE, Gutnikov SA, Mehta Z.
Effect of urgent treatment of transient ischaemic attack and minor stroke
on early recurrent stroke (EXPRESS study): a prospective
population-based sequential comparison. Lancet. 2007;370:1432–1442.
10. Ois A, Gomis M, Rodriguez-Campello A, Cuadrado-Godia E,
Jimenez-Conde J, Pont-Sunyer C, Cuccurella G, Roquer J. Factors asso-
ciated with a high risk of recurrence in patients with transient ischemic
attack or minor stroke. Stroke. 2008;39:1717–1721.
11. Johnston SC, Rothwell PM, Nguyen-Huynh MN, Giles MF, Elkins JS,
Bernstein AL, Sidney S. Validation and refinement of scores to predict
very early stroke risk after transient ischaemic attack. Lancet. 2007;369:
283–292.
12. Scottish Intercollegiate Guidelines Network. Management of Patients
With Stroke: Assessment, Investigation, Immediate and Secondary Pre-
vention (SIGN 108). Edinburgh, UK: The Network; 2008. Available at:
www.sign.ac.uk/pdf/sign108.pdf. Accessed October 1, 2010.
13. Department of Health. National Stroke Strategy. 2007. Available at:
www.dh.gov.uk/stroke. Accessed October 1, 2010.
14. National Institute for Health and Clinical Excellence TIA and Stroke
Guidelines. 2008. Available at: www.nice.org.uk. Accessed October 1,
2010.
15. Clinical Guidelines for Acute Stroke Management. 2009. Available at:
www.nhmrc.gov.au/publications. Accessed October 1, 2010.
16. Adams RJ, Albers G, Alberts MJ, Benavente O, Furie K, Goldstein LB,
Gorelick P, Halperin J, Harbaugh R, Johnston SC, Katzan I, Kelly-Hayes
M, Kenton EJ, Marks M, Sacco RL, Schwamm LH. Update to the
AHA/ASA recommendations for the prevention of stroke in patients with
stroke and transient ischemic attack. Stroke. 2008;39:1647–1652.
17. Sacco RL, Adams R, Albers G, Alberts MJ, Benavente O, Furie K,
Goldstein LB, Gorelick P, Halperin J, Harbaugh R, Johnston SC, Katzan
I, Kelly-Hayes M, Kenton EJ, Marks M, Scwamm LH, Tomsick T.
Guidelines for prevention of stroke in patients with ischemic stroke or
transient ischemic attack: a statement for healthcare professionals from
the American Heart Association/American Stroke Association Council on
Stroke: co-sponsored by the Council on Cardiovascular Radiology and
Intervention: the American Academy of Neurology affirms the value of
this guideline. Stroke. 2006;37:577– 617.
18. Diener HC, Ringleb PA, Savi P. Clopidogrel for the secondary prevention
of stroke. Expert Opin Pharmacother. 2005;6:755–764.
19. A randomised, blinded, trial of clopidogrel versus aspirin in patients at
risk of ischaemic events (CAPRIE). CAPRIE Steering Committee.
Lancet. 1996;348:1329 –1339.
20. Kernan WN, Viscoli CM, Brass LM, Makuch RW, Sarrel PM, Roberts
RS, Gent M, Rothwell P, Sacco RL, Liu RC, Boden-Albala B, Horwitz
Downloaded from http://stroke.ahajournals.org/ by guest on December 21, 2012
Poor Performance of Current Prognostic Scores 637
RI. The Stroke Prognosis Instrument II (SPI-II): a clinical prediction
instrument for patients with transient ischemia and nondisabling ischemic
stroke. Stroke. 2000;31:456 – 462.
21. Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM,
Gutnikov SA, Edwards P, Mant D, Sackley CM, Farmer A, Sandercock
PA, Dennis MS, Warlow CP, Bamford JM, Anslow P. Change in stroke
incidence, mortality, case-fatality, severity, and risk factors in
Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet.
2004;363:1925–1933.
22. Rothwell PM, Coull AJ, Silver LE, Fairhead JF, Giles MF, Lovelock CE,
Redgrave JN, Bull LM, Welch SJ, Cuthbertson FC, Binney LE, Gutnikov
SA, Anslow P, Banning AP, Mant D, Mehta Z. Population-based study of
event-rate, incidence, case fatality, and mortality for all acute vascular
events in all arterial territories (Oxford Vascular Study). Lancet. 2005;
366:1773–1783.
23. Coull AJ, Silver LE, Bull LM, Giles MF, Rothwell PM. Direct assessment
of completeness of ascertainment in a stroke incidence study. Stroke.
2004;35:2041–2045.
24. Asimos AW, Johnson AM, Rosamond WD, Price MF, Rose KM,
Catellier D, Murphy CV, Singh S, Tegeler CH, Felix A. A multicenter
evaluation of the ABCD(2) score’s accuracy for predicting early ischemic
stroke in admitted patients with transient ischemic attack. Ann Emerg
Med. 2010;55:201–210.
25. Josephson SA, Sidney S, Pham TN, Bernstein AL, Johnston SC. Higher
ABCD2 score predicts patients most likely to have true transient ischemic
attack. Stroke. 2008;39:3096 –3098.
26. Fothergill A, Christianson TJ, Brown RD Jr, Rabinstein AA. Validation
and refinement of the ABCD2 score: a population-based analysis. Stroke.
2009;40:2669 –2673.
27. Hankey GJ, Slattery JM, Warlow CP. Transient ischaemic attacks: which
patients are at high (and low) risk of serious vascular events? J Neurol
Neurosurg Psychiatry. 1992;55:640 – 652.
28. Koudstaal PJ, Gerritsma JG, van Gijn J. Clinical disagreement on the
diagnosis of transient ischemic attack: is the patient or the doctor to
blame? Stroke. 1989;20:300 –301.
29. Johnston SC, Sidney S, Bernstein AL, Gress DR. A comparison of risk
factors for recurrent TIA and stroke in patients diagnosed with TIA.
Neurology. 2003;60:280 –285.
30. Giles MF, Greg AW, Amarenco P, Arsava MM, Asimos A, Ay H, Calvet
D, Coutts S, Cucchiara BL, Demchuk AM, Johnston SC, Kelly PJ, Kim
AS, Labreuche JL, Lavallee PC, Mas JL, Merwick A, Olivot JM, Purroy
F, Rosamond WD, Sciolla R, Rothwell PM. Addition of brain infarction
to the ABCD2 score (ABCD2I). A collaborative analysis of unpublished
data on 4574 patients. Stroke. 2010;41:1907–1913.
31. Brott T, Adams HP Jr, Olinger CP, Marler JR, Barsan WG, Biller J,
Spilker J, Holleran R, Eberle R, Hertzberg V. Measurements of acute
cerebral infarction: a clinical examination scale. Stroke. 1989;20:
864 – 870.
32. Coutts SB, Hill MD, Campos CR, Choi YB, Subramaniam S, Kosior JC,
Demchuk AM. Recurrent events in transient ischemic attack and minor
stroke: what events are happening and to which patients? Stroke. 2008;
39:2461–2466.
33. Coutts SB, Hill MD, Simon JE, Sohn CH, Scott JN, Demchuk AM. Silent
ischemia in minor stroke and TIA patients identified on MR imaging.
Neurology. 2005;65:513–517.
34. Chalela JA, Kidwell CS, Nentwich LM, Luby M, Butman JA, Demchuk
AM, Hill MD, Patronas N, Latour L, Warach S. Magnetic resonance
imaging and computed tomography in emergency assessment of patients
with suspected acute stroke: a prospective comparison. Lancet. 2007;369:
293–298.