Symptomatic Intracranial Hemorrhage After Stroke Thrombolysis
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Symptomatic Intracranial Hemorrhage After
Stroke Thrombolysis
Comparison of Prediction Scores
Daniel Strbian, MD, PhD, MSc (Stroke Med); Patrik Michel, MD; David J. Seiffge, MD;
Jeffrey L. Saver, MD, FAHA; Heikki Numminen, MD, PhD;
Atte Meretoja, MD, PhD, MSc (Stroke Med); Kei Murao, MD; Bruno Weder, MD;
Nina Forss, MD, PhD; Anna-Kaisa Parkkila, MD, PhD; Ashraf Eskandari, RN;
Charlotte Cordonnier, MD, PhD; Stephen M. Davis, MD, PhD; Stefan T. Engelter, MD, PhD;
Turgut Tatlisumak, MD, PhD
Background and Purpose—Several prognostic scores have been developed to predict the risk of symptomatic intracranial
hemorrhage (sICH) after ischemic stroke thrombolysis. We compared the performance of these scores in a multicenter cohort.
Methods—We merged prospectively collected data of patients with consecutive ischemic stroke who received intravenous
Downloaded from http://stroke.ahajournals.org/ by guest on August 3, 2017
thrombolysis in 7 stroke centers. We identified and evaluated 6 scores that can provide an estimate of the risk of sICH in
hyperacute settings: MSS (Multicenter Stroke Survey); HAT (Hemorrhage After Thrombolysis); SEDAN (blood sugar, early
infarct signs, [hyper]dense cerebral artery sign, age, NIH Stroke Scale); GRASPS (glucose at presentation, race [Asian],
age, sex [male], systolic blood pressure at presentation, and severity of stroke at presentation [NIH Stroke Scale]); SITS
(Safe Implementation of Thrombolysis in Stroke); and SPAN (stroke prognostication using age and NIH Stroke Scale)-100
positive index. We included only patients with available variables for all scores. We calculated the area under the receiver
operating characteristic curve (AUC-ROC) and also performed logistic regression and the Hosmer–Lemeshow test.
Results—The final cohort comprised 3012 eligible patients, of whom 221 (7.3%) had sICH per National Institute of
Neurological Disorders and Stroke, 141 (4.7%) per European Cooperative Acute Stroke Study II, and 86 (2.9%) per Safe
Implementation of Thrombolysis in Stroke criteria. The performance of the scores assessed with AUC-ROC for predicting
European Cooperative Acute Stroke Study II sICH was: MSS, 0.63 (95% confidence interval, 0.58–0.68); HAT, 0.65 (0.60–
0.70); SEDAN, 0.70 (0.66–0.73); GRASPS, 0.67 (0.62–0.72); SITS, 0.64 (0.59–0.69); and SPAN-100 positive index, 0.56
(0.50–0.61). SEDAN had significantly higher AUC-ROC values compared with all other scores, except for GRASPS where
the difference was nonsignificant. SPAN-100 performed significantly worse compared with other scores. The discriminative
ranking of the scores was the same for the National Institute of Neurological Disorders and Stroke, and Safe Implementation
of Thrombolysis in Stroke definitions, with SEDAN performing best, GRASPS second, and SPAN-100 worst.
Conclusions—SPAN-100 had the worst predictive power, and SEDAN constantly the highest predictive power. However,
none of the scores had better than moderate performance. (Stroke. 2014;45:752-758.)
Key Word: intracranial hemorrhages
T he only approved clot-busting medical treatment in isch-
emic stroke, intravenous thrombolysis (IVT), is not without
complications. One of the major reasons for withholding the
worsening of outcome (≥1 grade on modified Rankin Scale)
ranges from 29.7 to 40.1.2 There are several scoring systems
for predicting the risk of sICH.3–8 In an ideal situation, a predic-
therapy remains fear of symptomatic intracranial hemorrhage tion score could identify patients with very high risk of post-
(sICH), which can worsen patients’ outcomes.1 The number thrombolysis sICH. We aimed to compare the performance of
needed for IVT to cause fatal sICH is 36.5, and to cause any existing risk prediction scores in a large multicenter cohort.
Received October 9, 2013; final revision received December 4, 2013; accepted December 17, 2013.
From the Departments of Neurology and Stroke Units, Helsinki University Central Hospital, Helsinki, Finland (D.S., A.M., N.F., T.T.); Centre
Hospitalier Universitaire Vaudois and University of Lausanne, Lausanne, Switzerland (P.M., A.E.); University Hospital Basel, Basel, Switzerland (D.J.S.,
S.T.E.); Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA (J.L.S.); Tampere University Hospital, Tampere, Finland
(H.N., A.-K.P.); The Royal Melbourne Hospital, Parkville, Australia (A.M., S.M.D.); University Lille Nord de France, Lille, France (K.M., C.C.); and
Kantonsspital St Gallen, St Gallen, Switzerland (B.W.).
Guest Editor for this article was Tatjana Rundek, MD, PhD.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.
113.003806/-/DC1.
Correspondence to Daniel Strbian, MD, PhD, Department of Neurology, Helsinki University Central Hospital, Haartmaninkatu 4, 00290 Helsinki,
Finland. E-mail daniel.strbian@hus.fi
© 2014 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.113.003806
752Strbian et al Post-Thrombolytic Hemorrhage Prediction Scores 753
Patients and Methods
Study Setting
Table 1. Description of the Scores The current analysis comprises data from 7 centers. The study was
Score/Derivation Parameters in the Cut-Off Values, Score approved by the relevant authorities in each participating center per
local requirements. This study was approved in the coordinating cen-
Cohort Score Points, pt
ter (Helsinki) as a quality registry and did not require review by the
MSS (NINDS sICH): Age, y >60 y, 1 pt ethical board. All patients were prospectively included in the database.
derivation cohort, NIHSS score >10, 1 pt Data from individual consecutive patients receiving IVT within a 4.5-
n=481; validation, hour time window for acute ischemic stroke were collected using a
Blood glucose, mmol/L >8.325, 1 pt standardized form with predefined variables. For sICH definitions, ra-
n=965
Platelet count/mm3 20, 2 pt
chymal hemorrhage 1 and 2) was collected and used to prospectively
assign sICH for the current study by 1 of the study authors in a blinded
derivation, n=302; Blood glucose or DM >200 mg/dL or DM, 1 pt fashion. Data from all the centers were compiled in the coordinating
validation, n=98
Hypodensity on CT 75, 1 pt terior circulation. None of the patients underwent endovascular proce-
sICH): derivation, NIHSS score ≥10, 1 pt dure. None of the patients in the current analysis was included in the
n=974; validation, derivation cohort of any of the scores/indices.
Blood glucose, mmol/L 8.1–12.0, 1 pt; >12.0,
Downloaded from http://stroke.ahajournals.org/ by guest on August 3, 2017
n=828
2 pt
Early infarct signs Yes, 1 pt Selection Criteria for sICH Risk Scores
Hyperdense artery signs Yes, 1 pt First, we only considered scores and indices but not regression models,
GRASPS (NINDS Age, y ≤60, 8 pt; 61–70, because our aim was to evaluate only tools that are suitable for quick
bedside calculations without depending on potentially time-consum-
sICH): derivation, 11 pt; 71–80, 15 pt;
ing, computer-based systems. Second, we considered only scores and
n=7169; validation, >80, 17 pt
indices based on parameters available shortly after admission, before
n=3073 NIHSS score 0–5, 25 pt; 6–10, 27 pt;
11–15, 34 pt; 16–20,
Table 2. Demographics, Baseline Characteristics, and
40 pt; >20, 42 pt
Frequencies of Symptomatic Intracranial Hemorrhage
Blood glucose, mg/dL754 Stroke March 2014
administration of thrombolysis. Finally, we included scores that had University of Southampton). A 2-tailed P valueStrbian et al Post-Thrombolytic Hemorrhage Prediction Scores 755
Downloaded from http://stroke.ahajournals.org/ by guest on August 3, 2017
Figure 2. Frequencies of symptomatic intracranial hemorrhage according to the criteria of European Cooperative Acute Stroke Study II
for each point level of the predictive scores. Number in parentheses represents patients who reached that specific score points. Because
of the complexity of GRASPS, not all n for particular score points are shown. GRASPS indicates glucose at presentation, race (Asian),
age, sex (male), systolic blood pressure at presentation, and severity of stroke at presentation (NIH Stroke Scale); HAT, Hemorrhage
After Thrombolysis; MSS, Multicenter Stroke Survey; SEDAN, blood sugar, early infarct signs, (hyper)dense cerebral artery sign, age, NIH
Stroke Scale; SITS, Safe Implementation of Thrombolysis in Stroke; and SPAN, stroke prognostication using age and NIH Stroke Scale.
9.3% (ECASS-II criteria), from 4.4% to 9.9% (NINDS), and highest nominal predictive performance in all comparisons,
from 2.2% to 5.1% (SITS). most of which were statistically significant, except for the com-
Frequencies of sICH, according to the 3 criteria,1 per point parison with GRASPS, which showed the second highest AUC-
increase of the scores are outlined in Figures 1–3 and Figure I in ROC values. In 2-way comparisons, the differences between
the online-only Data Supplement. Based on the logistic regres- GRASPS and other scores were frequently nonsignificant.
sion analysis, all scores were associated with sICH according to We observed rather low frequencies of post-thrombolytic
all 3 criteria (Table I in the online-only Data Supplement). The sICH in the current merged cohort with considerable inter-
results of the Hosmer–Lemeshow test showed worst model fit center differences. This contributes to the relatively low risk
for GRASPS in case of ECASS-II sICH. Because SPAN-100 is of sICH even with the worst scores compared with the original
a binary index, the test could not have been calculated. reports (perhaps with the exception of MSS and HAT, rela-
Score comparisons by means of AUC-ROCs are presented tively smaller number of patients scored the highest points).
separately for each sICH definition (Table 3). SEDAN had Nonetheless, what is crucial is that the relative risk of high-risk
the highest absolute values of AUC-ROC in all analyses, and patients compared with low-risk patients remained similar.
except for the comparison with GRASPS, these differences For example, according to the original report,5 a patient with
were statistically significant. SPAN-100 positive index had the SEDAN of 5 had almost 4-fold higher risk of sICH (33.3%)
lowest AUC-ROC values in all comparisons. compared with SEDAN of 2 (8.5%) and >20-fold higher risk
compared with SEDAN of 0 (1.4%). Here, the magnitude of
Discussion these relative risks remained similar, 4 and 18, respectively.
With comprehensive data from several dedicated stroke centers, Taken together, the scores consist of parameters related to
we had a unique opportunity to perform a head-to-head compari- (1) underlying parenchymal injury, microangiopathy (age,
son of the existing sICH prediction scores. In general, SPAN- history of hypertension and diabetes mellitus, blood glucose
100 showed poor predictive power, and all other scores moderate as a marker of diabetes mellitus history), (2) degree of acute
predictive power. Of all scores, SEDAN had constantly the parenchymal injury (CT findings and, to certain level, also the756 Stroke March 2014
Downloaded from http://stroke.ahajournals.org/ by guest on August 3, 2017
Figure 3. Frequencies of symptomatic intracranial hemorrhage according to the criteria of Safe Implementation of Thrombolysis in Stroke
for each point level of the predictive scores. Number in parentheses represents patients who reached that specific score points. Because
of the complexity of GRASPS, not all n for particular score points are shown. GRASPS indicates glucose at presentation, race (Asian),
age, sex (male), systolic blood pressure at presentation, and severity of stroke at presentation (NIH Stroke Scale); HAT, Hemorrhage
After Thrombolysis; MSS, Multicenter Stroke Survey; SEDAN, blood sugar, early infarct signs, (hyper)dense cerebral artery sign, age, NIH
Stroke Scale; SITS, Safe Implementation of Thrombolysis in Stroke; and SPAN, stroke prognostication using age and NIH Stroke Scale.
National Institutes of Health Stroke Scale [NIHSS] and onset- calculation of HAT and SEDAN) seem to improve the perfor-
to-treatment time), (3) coagulation process (platelet count, use mance of outcome prediction scores.12 Although their assess-
of antiplatelet agents, and perhaps patient’s weight determin- ing requires training, we think it is readily achievable with
ing the dose of alteplase), (4) physical factors (systolic blood continuous education in centers delivering IVT. Interestingly,
pressure), and (5) sex and ethnicity. In fact, modest differ- platelet count was included in 1 score only (MSS), but it did
ences in the AUC-ROC values among the scores reflect that not, for example, improve the model of SEDAN (data not
most of the scores include similar components: age, NIHSS, shown). One possible explanation is the fact that a vast major-
and baseline glucose level being the most common (Tables 1 ity of patients in its derivation cohort had similar platelet
and 3). The differences are often in the relative weighting counts, being in the physiological range.
given to individual components. SPAN-100 (consisting of age Another source of differences in the performances of scores
and NIHSS) had rather low AUC-ROC values according to all may reflect the fact that they were derived to predict particular
sICH criteria (0.55–0.56) as compared with 0.73 per NINDS definitions of sICH. For example, SITS had higher AUC-ROC
criteria in the original report.8 Potential explanations may values than for ECASS-II or NINDS criteria (Table 3).
be that SPAN-100 was postulated rather than derived from SEDAN had the highest AUC-ROC value for sICH per
a specific cohort. Also, validation was performed in a rather ECASS-II definition, for which it was developed. Whereas,
small cohort of 312 patients with IVT from the NINDS trial. GRASPS had almost identical AUC-ROC values for each def-
Furthermore, the timing of treatment in the present study was, inition. Another aspect influencing the performance of scores
on average, somewhat later than in the NINDS patients, half of is the number of component items. Scores derived from larger
whom were treatedStrbian et al Post-Thrombolytic Hemorrhage Prediction Scores 757
Table 3. Areas Under the Curves (AUC) and P Values for Their Comparisons
MSS HAT SEDAN GRASPS SITS SPAN-100
sICH NINDS study
AUC (95% CI) 0.62 (0.58–0.66) 0.65 (0.62–0.69) 0.69 (0.66–0.73) 0.67 (0.63–0.70) 0.61 (0.58–0.65) 0.55 (0.51–0.59)
MSS … 0.10758 Stroke March 2014
for ≥1 of the scores, but they did not differ from the included fees from Boehringer-Ingelheim (modest), and advisory board com-
patients in demographics and baseline characteristics. Each pensation from Boehringer-Ingelheim (modest). Dr Davis received
travel grants from EVER Neuropharma (modest) and Sanofi (mod-
center performed its imaging read. However, our study rep- est), as well as speakers fees from Boehringer-Ingelheim (modest).
resents a large sample size with patients from several cen- Dr Tatlisumak received honoraria from Boehringer-Ingelheim (mod-
ters operating under different conditions. Unfortunately, we est) and advisory board compensations from Boehringer-Ingelheim
could not analyze the scores separately in men and women (modest) and H Lundbeck A/S (modest). The other authors have no
or in patients treatedSymptomatic Intracranial Hemorrhage After Stroke Thrombolysis: Comparison of
Prediction Scores
Daniel Strbian, Patrik Michel, David J. Seiffge, Jeffrey L. Saver, Heikki Numminen, Atte
Meretoja, Kei Murao, Bruno Weder, Nina Forss, Anna-Kaisa Parkkila, Ashraf Eskandari,
Charlotte Cordonnier, Stephen M. Davis, Stefan T. Engelter and Turgut Tatlisumak
Downloaded from http://stroke.ahajournals.org/ by guest on August 3, 2017
Stroke. 2014;45:752-758; originally published online January 28, 2014;
doi: 10.1161/STROKEAHA.113.003806
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2014 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/45/3/752
Data Supplement (unedited) at:
http://stroke.ahajournals.org/content/suppl/2014/01/28/STROKEAHA.113.003806.DC1
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/SUPPLEMENTAL MATERIAL
Symptomatic intracranial hemorrhage after stroke thrombolysis: comparison of prediction scores
1
Daniel Strbian, MD, PhD; 2Patrik Michel, MD; 3David J Seiffge, MD; 4Jeffrey L. Saver, MD, FAHA; 5Heikki
Numminen, MD, PhD; 1,6Atte Meretoja, MD, PhD, MSc (Stroke Med); 7Kei Murao; 8Bruno Weder, MD;
1
Nina Forss, MD, PhD; 5Anna-Kaisa Parkkila, MD, PhD; 2Ashraf Eskandari, RN; 7Charlotte Cordonnier, MD,
PhD; 6Stephen M Davis, MD, PhD; 3Stefan T Engelter, MD, PhD; 1Turgut Tatlisumak, MD, PhDSupplementary Table I. Logistic regression coefficients per score point and the Hosmer-Lemeshow
test.
OR (95% CI) Hosmer-Lemeshow test P
sICH (NINDS)
MSS 1.72 (1.47-2.02) 0.85 0.66
HAT 1.75 (1.54-1.98) 0.67 0.72
SEDAN 1.72 (1.54-1.92) 0.94 0.82
GRASPS 1.07 (1.05-1.09) 9.44 0.31
SITS 1.24 (1.15-1.32) 3.78 0.71
SPAN-100 + 2.43 (1.70-3.48)Supplementary Figure I a) b)
c)
Legend: Frequencies of symptomatic intracranial hemorrhage according to the criteria of a) NINDS, b)
ECASS-II, and c) SITS for each point level of the predictive scores. Number in the brackets repre-
sent patients who reached the particular score points. In addition to the full range of possible
GRASPS points presented in the main text of the manuscript, we show here 10 categories of the
GRASPS score (45-49, 50-54, 55-59, 60-64, 65-69, 70-74, 75-79, 80-84, 85-89, 90-94).You can also read
