The term minor stroke is often used for stroke patients

What Is a Minor Stroke? Urs Fischer, MD, MSc; Adrian Baumgartner, MS; Marcel Arnold, MD; Krassen Nedeltchev, MD; Jan Gralla, MD, MSc; Gian Marco De Marchis, MD; Liliane Kappeler, MD; Marie-Luise Mono, MD; Caspar Brekenfeld, MD; Gerhard Schroth, MD; Heinrich P. Mattle, MD Background and Purpose The term minor stroke is often used; however a consensus definition is lacking. We explored the relationship of 6 minor stroke definitions and outcome and tested their validity in subgroups of patients. Methods A total of 760 consecutive patients with acute ischemic strokes were classified according to the following definitions: A, score 1 on every National Institutes of Health Stroke Scale (NIHSS) item and normal consciousness; B, lacunar-like syndrome; C, motor deficits with or without sensory deficits; D, NIHSS 9 excluding those with aphasia, neglect, or decreased consciousness; E, NIHSS 9; and F, NIHSS 3. Short-term outcome was considered favorable when patients were discharged home, and favorable medium-term outcome was defined as a modified Rankin Scale score of 2 at 3 months. The following subgroup analyses were performed by definition: sex, age, anterior versus posterior and right versus left hemispheric stroke, and early (0 to 6 hours) versus late admission (6 to 24 hours) to the hospital. Results Short-term and medium-term outcomes were most favorable in patients with definition A (74% and %, respectively) and F (71% and %, respectively). Patients with definition C and anterior circulation strokes were more likely to be discharged home than patients with posterior circulation strokes (P 0.021). The medium-term outcome of older patients with definition E was less favorable compared with the outcome of younger ones (P 0.001), whereas patients with definition A, D, and F did not show different outcomes in any subgroup. Conclusions Patients fulfilling definition A and F had best short-term and medium-term outcomes. They would be best suited to the definition of minor stroke. (Stroke. 2010;41:661-666.) Key Words: minor stroke definitions outcome NIHSS The term minor stroke is often used for stroke patients with mild and nondisabling symptoms. However, a consensus definition is lacking. We performed a structured literature search of MEDLINE from 1950 to May 2009 with the key word minor stroke(s) and found 670 articles with the term in the abstract (568 articles) or the title (102 articles). Most authors did not define their meaning of minor stroke. In the journal, Stroke, the most relevant specialist journal on cerebrovascular diseases, only 25 of 75 articles provided a definition. All authors giving a definition tried to capture stroke syndromes with mild and nondisabling symptoms, but their definitions varied considerably (see supplemental Appendix, available online at http://stroke.ahajournals.org). Patients are selected for trials and epidemiological studies based on the syndrome of a minor stroke. 1,2 Therefore, a broadly acceptable concept and definition of minor stroke is required. Ideally, a definition of a minor stroke should reflect the following aspects: (1) it should capture patients with mild and nondisabling symptoms in acute stage and favorable shortterm and medium-term outcomes; (2) it should be valid for different subgroups of stroke patients; (3) it should imply both qualitative and quantitative dimensions; (4) it should be simple and useful in daily clinical practice; and (5) it should not overlap with the definition of a transient ischemic attack (TIA). The National Institute of Neurological Disorders and Stroke (NINDS) rt-pa Stroke Study Investigators developed 5 working definitions of minor stroke based on information limited to the CT scan and clinical presentation, specifically, the baseline NIHSS score, baseline NIHSS individual items, and baseline stroke subtype. 3 However, these 5 definitions have not been tested in clinical practice, and none have been adopted in stroke research. We tested these definitions of minor stroke and the definition most often used according to our literature search (NIHSS 3) in the Bernese Stroke cohort. We explored the relationship of the 6 definitions with short-term and medium-term outcomes and tested the validity in different subgroups of stroke patients. Patients and Methods The University Hospital of Bern (Inselspital) is a large teaching hospital providing tertiary care for 1 million people and primary care for the urban area of Bern, with 300 000 inhabitants. This indicates that we receive and admit both severe strokes from distant Received November 12, 2009; final revision received December 15, 2009; accepted December 30, 2009. From the Departments of Neurology (U.R., A.B., M.A., K.N., G.M.D.M., L.K., M.-L.M., H.P.M.) and Neuroradiology (J.G., C.B., G.S.), Inselspital, University Hospital Bern and University of Bern, Switzerland. Correspondence to Heinrich P. Mattle, MD, Department of Neurology, University of Bern, Inselspital, Freiburgstrasse 10, 3010 Bern, Switzerland. E-mail heinrich.mattle@insel.ch 2010 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.109.572883 661

662 Stroke April 2010 areas and mild and severe strokes and TIA patients from the city area. Most patients with a TIA or stroke are usually admitted for treatment and further investigation. From January 2000 to December 2005, 815 patients with acute ischemic strokes (time from symptom onset to presentation 24 hours) were admitted to the ward of the Department of Neurology. Patients with retinal artery occlusions and TIAs, defined as a focal cerebral ischemic event with symptoms lasting 24 hours, were excluded. Some aspects of these patients have been reported previously. 4,5 All patients were examined immediately after admission by a neurologist, and the deficit was scored using the NIHSS. 6 Demographic data and time of symptom onset were recorded. After clinical evaluation, patients underwent a standard investigation protocol in the emergency department, including blood tests, ECG, cranial CT, or MRI. Status of extracranial and intracranial vessels was assessed by neurovascular ultrasound, CT angiography, magnetic resonance angiography, or digital subtraction angiography. Based on the ischemic lesions on the CT or MRI scan, infarctions were classified into anterior and posterior circulation and right and left hemispheric strokes. The following stroke risk factors were assessed: sex, hypertension, diabetes, current cigarette smoking, hypercholesterolemia, coronary heart disease, previous TIA or stroke, and a family history of TIA and stroke. Stroke etiology was classified using the Trial of Org 10172 in Acute Stroke Treatment criteria after a complete diagnostic work-up. 7 Seventeen patients (2%) underwent intravenous therapy ( 3 hours of onset), 146 (32%) intra-arterial thrombolysis ( 6 hours of onset), and 21 (3%) mechanical thromboaspiration, according to international guidelines and our institutional protocol. 8,9 Antithrombotic and secondary preventive therapy was given according to European guidelines. 10,11 All patients were admitted to the neurological ward. Patients with disabling stroke symptoms including neuropsychological deficits such as aphasia, visuospatial deficits, or neglect were then transferred to the rehabilitation unit; those with nondisabling strokes were discharged home. Definitions of Minor Stroke Based on clinical presentation, baseline NIHSS score, and baseline NIHSS items, we classified all patients for this study into yes or no for each definition A through F. The definitions are summarized below. Thirty-eight patients (4.7%) could not be classified because clinical records were inadequate to classify them into all of the 6 definitions. The 6 definitions of minor stroke used in this study: A. all patients with a score 0 or 1 on every baseline NIHSS score item, except level of consciousness items (items 1a to 1c), which must be 0; B. all patients with a lacunar-like syndrome (presumed smallvessel occlusive disease) such as pure sensory syndrome, pure motor hemiparesis, sensorimotor syndrome, ataxic hemiparesis, and dysarthria-clumsy hand syndrome; C. all patients with only motor deficits (can include dysarthria or ataxia) with or without sensory deficits. These patients can have only a combination of motor, coordination, and sensory deficits without any deficits in the spheres of language, level of consciousness, extinction or neglect, horizontal eye movements, or visual fields, deficits generally ascribed to larger territories of focal ischemia; D. all patients with baseline NIHSS in the lowest (least severe) quartile of severity (NIHSS 9), excluding all patients with aphasia, extinction, or neglect, or any points on the level-ofconsciousness questions; E. all patients with baseline NIHSS in the lowest (least severe) quartile of severity (NIHSS 9); and F. all patients with baseline NIHSS 3. Table 1. Baseline Characteristics of 760 Patients With an Acute Ischemic Stroke Characteristics Age, years (SD) 62 (13) Female sex, n (%) 275 (36) Vascular risk factors, n (%) Hypertension 474 (62) Diabetes mellitus 122 (16) Current smoking 197 (26) Hypercholesterolemia 386 (51) Coronary artery disease 313 (41) Previous TIA 108 (14) Previous stroke 76 (10) Family history of stroke 159 (21) Family history of MI 120 (16) Baseline NIHSS score, median (range) 8 (1 37) Territory, n (%) Anterior circulation 580 (76) Posterior circulation 179 (24) Both 1 Hemisphere, n (%) Right 262 (35) Left 333 (44) Both 8 (1) Brain stem 149 (20) Stroke etiology, n (%) Large artery disease 129 (17) Cardioembolic 268 (35) Small artery disease 75 (10) Other determined etiology 82 (11) Unknown etiology 200 (26) More than one cause 5 (1) MI indicates Myocardial infarction. Follow-Up Assessment Short-term outcome was considered favorable when patients were discharged home from our hospital and unfavorable when they died or were referred to a rehabilitation unit or another hospital. Mediumterm outcome was assessed 3 months after stroke using the modified Rankin scale. 12 A total of 41.4% of patients were examined clinically, and 58.6% were interviewed by phone. A good outcome was defined as a modified Rankin Scale score of 0 to 2. Telephone interviews were performed by physicians and study nurses experienced in the use of the modified Rankin Scale. Recurrent vascular events after hospital discharge such as TIA, recurrent stroke, or myocardial infarction were recorded. Seventeen (2.1%) of 815 patients were lost for follow-up. Statistical Analysis Quantitative data are expressed as mean values 1SD. The NIHSS score on admission is given as median value. Data are reported in frequency tables. Effect of different variables on clinical outcome among patients with each of the 6 definitions was assessed using Fisher exact test for comparison of proportions. The following variables were assessed: sex, age ( 65 years versus 65 years), anterior versus posterior circulation strokes, right versus left hemispheric stroke (those with bilateral, cerebellar, or brain stem ischemia were excluded for this subanalysis), and early (0 to 6 hours) versus late presentation (6 to 24 hours). P 0.05 was considered significant.

Fischer et al What Is a Minor Stroke? 663 Table 2. Characteristics of Patients With Minor Stroke Classified According to Different Definitions Def A n 144 Def B n 183 Def C n 191 Def D n 251 Def E n 417 Def F n 160 Baseline variables Age, mean (SD) 62 (13) 62 (13) 62 (13) 62 (13) 62 (13) 62 (14) Sex, women 38 (26) 68 (37) 67 (35) 86 (34) 137 (33) 46 (29) NIHSS, median (range) 2 (1 6) 3 (1 11) 4 (1 13) 3 (1 9) 5 (1 9) 2 (1 3) Stroke etiology, n (%) Large artery disease 32 (22) 28 (15) 29 (15) 42 (17) 78 (19) 34 (21) Cardioembolic 40 (28) 48 (26) 56 (29) 75 (30) 131 (31) 45 (28) Small vessel disease 38 (27) 52 (29) 51 (27) 59 (23) 73 (18) 40 (25) Other cause 9 (6) 11 (6) 9 (5) 19 (7) 35 (8) 12 (8) Unknown etiology 22 (15) 40 (22) 42 (22) 52 (21) 96 (23) 28 (17) More than one cause 3 (2) 4 (2) 4 (2) 4 (2) 4 (1) 1 (1) Thrombolysis, n (%) 3 (2) 11 (6) 15 (8) 15 (6) 64 (15) 4 (3) Territory, n (%) Anterior 98 (68) 133 (73) 140 (73) 151 (60) 282 (68) 99 (62) Posterior 46 (32) 50 (27) 51 (27) 100 (40) 134 (32) 61 (38) Hemisphere, n (%) Right 44 (44) 69 (52) 77 (54) 92 (56) 134 (45) 41 (39) Left 55 (56) 64 (48) 65 (46) 72 (44) 164 (55) 64 (61) Def indicates definition. Results Baseline characteristics, clinical information and stroke etiology of all 760 patients included in this analysis are presented in Table 1. Table 2 summarizes the number of patients fulfilling minor stroke definitions A through F and their characteristics. Median NIHSS on admission was lowest in patients with definitions A and F and highest in those with definition E. Short-term and medium-term outcome was best in patients with definitions A and F and worst in those with definitions C and E (Figure 1). The number of recurrent vascular events at 3 months among patients with different definitions was comparable and low (range, 0.6 to 2.5%). Outcome in nonthrombolyzed patients compared with all patients in the corresponding definition group was similar (Table 3). Supplemental Table I (available online at http:// stroke.ahajournals.org) shows subgroup analyses of shortterm and medium-term outcomes. Patients with definitions B and E presenting after 6 hours were more likely to be discharged home than patients presenting within the first 6 hours (P 0.048 and P 0.019, respectively). Patients with definition C experiencing an anterior circulation stroke were more likely to be discharged home (P 0.021) than those with posterior circulation strokes, and older patients with definition E were more likely to be handicapped at 3 months than younger patients (P 0.001). Patients with definition A, D, and F did not show different outcomes in any subgroup. However, older patients with definition A tended to be handicapped after 3 months more frequently than younger patients (P 0.054). Figure 2 shows short-term and mediumterm outcomes in relation to stroke severity, measured with the NIHSS score at admission. Discussion In this study, we assessed the outcome of 760 stroke patients classified as minor stroke according to 6 different definitions. Definition A (score 0 or 1 on every baseline NIHSS score item Percentage 100 95 85 80 75 70 65 60 55 50 45 40 74 Def A n=144 mrs 0-2 85 57 Def B n=183 83 54 Def C n=191 Discharge Home 88 58 Def D n=251 85 48 Def E n=417 71 Def F n=160 Definition Figure 1. Short-term (Œ) and medium-term (f) outcomes in patients with different definitions (Def) of minor stroke (95% CI indicated by whiskers). mrs indicates modified Rankin Scale.

664 Stroke April 2010 Table 3. Short-Term and Medium-Term Outcomes and Recurrence Events in Patients With Different Definitions of Minor Stroke Short-Term Follow-Up Follow-Up at 3 Months Minor Stroke, n (%) n Discharge Home Death mrs 1 mrs 2 Death Definition A No thrombolysis n 141 105 (74.5) 0 98 (69.5) 127 (.1) 2 (1.4) All patients n 144 106 (73.6) 0 99 (68.8) 129 (89.6) 2 (1.4) Definition B No thrombolysis n 172 100 (58.1) 1 (0.6) 108 (62.8) 145 (84.3) 1 (0.6) All patients n 183 104 (56.8) 1 (0.5) 116 (63.4) 155 (84.7) 1 (0.5) Definition C No thrombolysis n 176 99 (56.3) 2 (0.6) 108 (61.4) 146 (82.9) 2 (0.6) All patients n 191 103 (53.9) 2 (1.0) 117 (61.3) 159 (83.2) 3 (1.6) Definition D No thrombolysis n 236 142 (60.2) 0 154 (65.3) 206 (87.3) 2 (0.8) All patients n 251 146 (58.2) 0 164 (65.3) 221 (88.0) 2 (0.8) Definition E No thrombolysis n 353 188 (53.3) 2 (0.6) 217 (61.5) 300 (84.9) 7 (1.9) All patients n 417 201 (48.2) 5 (1.2) 252 (60.4) 356 (85.4) 11 (2.6) Definition F No thrombolysis n 156 113 (72.4) 0 108 (69.2) 141 (.4) 3 (1.9) All patients n 160 114 (71.3) 0 110 (68.8) 144 () 3 (0.9) mrs indicates modified Rankin Scale. and normal level of consciousness) and definition F (NIHSS 3) would be best suitable to the concept of minor stroke : patients with definition A or F were most likely to be discharged home and to be independent at 3 months. Definition F was valid for all subgroups of stroke patients. In addition, definition F is easily applicable in clinical practice and relies on NIHSS total score. Several eminent groups in stroke research have adopted this definition. 13,14 However, using a specific cut point of the NIHSS for defining a minor stroke might provoke criticism and raise the question why a stroke with an NIHSS score of 3 should be minor but with an NIHSS score of 4 not? Figure 2 relates NIHSS scores and short-term and medium-term outcome: two thirds of patients with an NIHSS score of 3 were discharged home. The difference for patients with an NIHSS score of 4 is not impressive, whereas a significant number of patients with NIHSS scores of 5 have a less favorable short-term outcome compared with patients with scores of 4. This illustrates how arbitrary a cut point is. There is no real scientific reason to choose a cut point of 3 or 4. Therefore, it would be desirable to define a cut point for minor stroke by consensus of several stroke researchers. A minor stroke patient, according to definition F, could have a severe deficit in one NIHSS item or a mild deficit in more than one item. Some stroke physicians would consider the first situation (eg, hemianopia 2 points) more severe than a mild facial weakness combined with dysarthria (ie, 1 plus 1 2 points). Definition A circumvents this problem. According to this definition, minor stroke includes only patients who are conscious and score 1 on each NIHSS item. This means that definition A includes only patients who show mild but no severe functional deficits in any NIHSS item. The limitation of this definition according to our results is that it might not be entirely robust for all subtypes of stroke patients. Patients classified with definition D had worse short-term outcomes than patients with definitions A and F. Nevertheless, definition D merits some discussion; it includes all patients with an NIHSS score of 9 but excludes patients with decreased consciousness or neuropsychological deficits such as aphasia, extinction phenomena, or neglect. Mediumterm outcome of patients with minor stroke definition D was favorable in quasi as many patients as with definition A or F, although the cutoff of the NIHSS score was considerably high. Furthermore, this definition turned out to be robust in many subgroups of stroke patients. Less than 60% of patients with definitions B, C, and E were independent enough to be discharged home. Furthermore, these definitions were not robust for all subgroups of stroke patients. In addition, more patients with definitions B, C, and E compared with definitions A, D, and F had significant disabilities at 3 months. Therefore, these definitions are less suitable to define a minor stroke. Thrombolysis might have influenced short-term and mediumterm outcomes in patients with different definitions of a minor stroke. Therefore, patients without thrombolysis were analyzed separately (Table 2). Outcome in nonthrombolyzed patients compared with all patients in the corresponding definition group was similar. However, only a minority of patients with different definitions of minor stroke underwent thrombolysis, ranging from 2% to 15%. Whether thrombolysis is beneficial in patients with different definitions of a minor stroke cannot be derived from this study. However, in a previous analysis, we assessed outcome in patients with mild and rapidly improving symptoms;

Fischer et al What Is a Minor Stroke? 665 Percentage 100 80 70 60 50 40 mrs 0-2 at 3 months Discharge home score, and baseline NIHSS items by the first author. On the other hand, both the NIHSS, which relies on 4 of the 6 minor stroke definitions, and the modified Rankin Scale, on which we assessed outcome, have been used extensively in many trials. 15 18 Their reproducibility and validity have been analyzed in many studies and are considered as being fair. In conclusion, to date, there is no consensus on the definition of a minor stroke. Our study indicates that patients fulfilling definition A (conscious patients scoring 1 on every NIHSS item) and definition F (patients with NIHSS 3) had the best short-term and medium-term outcomes. They would be suitable best to define minor stroke for clinical and research purposes. Future consensus panels from international stroke organizations should consider a uniform definition of minor stroke to enhance clinical research. 30 20 Acknowledgments We thank Neal Thurley for the support with the systematic literature search and Pietro Ballinari for statistical advice. 10 0 1 2 3 4 5 6 7 8 9 10 11 12 n=48 n=60 n=52 n=47 n=53 n=47 n=51 n=26 n=33 n=24 n=22 n=29 NIHSS score Figure 2. Short-term (Œ) and medium-term (f) outcomes in relation to stroke severity measured with NIHSS scores. mrs indicates modified Rankin Scale. overall, 75% of these patients had a favorable outcome (modified Rankin Scale score 0 to 1). 4 Baseline NIHSS score 10 points increased the odds of unfavorable outcome or death 17-fold, and proximal vessel occlusion increased the odds 7-fold. Therefore, selected patients, especially those with proximal vessel occlusions and baseline NIHSS scores 10 points, are likely to derive a benefit from thrombolysis. In a post hoc subgroup analysis of the NINDS study, no difference in the beneficial effects of tissue plasminogen activator in patients with minor stroke syndromes compared with the overall treatment effects in the entire cohort could be detected. 3 Our current policy is to work up all TIA and minor stroke patients rapidly and base our treatment decisions (IV thrombolysis, endovascular treatment, or no treatment) on the clinical evolution and the results of the ancillary investigations. The present definitions of a minor stroke are based on clinical deficits and exclude information obtained from imaging. CT scans of patients with mild deficits are often normal in the acute stage of a stroke. However, nearly all patients with neurological signs lasting 24 hours show abnormalities on diffusion-weighted imaging. A combination of clinical signs and information from diffusion-weighted imaging might define a minor stroke more accurately than a definition based on clinical information only. However, such a study has yet to be performed. An important issue is the validity and interrater reproducibility of our minor stroke definitions. However, we cannot address this issue with the data of this study. All patients were categorized based on clinical presentation, baseline NIHSS Sources of Funding Urs Fischer was supported by the KK Foundation for Cardiology and Circulation and the Gottfried und Julia Bangerter Foundation. None. Disclosures References 1. Rothwell PM, Giles MF, Chandratheva A, Marquardt L, Geraghty O, Redgrave JN, Lovelock CE, Binney LE, Bull LM, Cuthbertson FC, Welch SJ, Bosch S, Alexander FC, Silver LE, Gutnikov SA, Mehta Z; Early Use of Existing Preventive Strategies for Stroke (EXPRESS) Study. 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. 2. Kennedy J, Hill MD, Ryckborst KJ, Eliasziw M, Demchuk AM, Buchan AM; FASTER Investigators. Fast assessment of stroke and transient ischaemic attack to prevent early recurrence (FASTER): a randomised controlled pilot trial. Lancet Neurol. 2007;6:961 969. 3. National Institute of Neurological Disorders and Stroke rt-pa Stroke Study Group. Recombinant tissue plasminogen activator for minor strokes: the National Institute of Neurological Disorders and Stroke rt-pa Stroke Study Experience. Ann Emerg Med. 2005;46:243 252. 4. Nedeltchev K, Schwegler B, Haefeli T, Brekenfeld C, Gralla J, Fischer U, Arnold M, Remonda L, Schroth G, Mattle HP. Outcome of stroke with mild or rapidly improving symptoms. Stroke. 2007;38:2531 2535. 5. Arnold M, Halpern M, Meier B, Fischer U, Haefeli T, Kappeler L, Brekenfeld C, MD, Mattle HP, Nedeltchev K. Age-dependent differences in demographics, risk factors, comorbidity, aetiology, management, and clinical outcome of acute ischemic stroke. J Neurol. 2008;255:1503 1507. 6. Brott T, Adams HP Jr, Olinger CP, Marler JR, Barsan WG, Biller J, Spilker J, Holleran R, Eberle R, Hertzberg V, Roorick M, Moomaw CJ, Walker M. Measurements of acute cerebral infarction: a clinical examination scale. Stroke. 1989;20:846 870. 7. Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial TOAST Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993;24:35 41. 8. Adams H, Adams R, Del Zoppo G, Goldstein LB; Stroke Council of the American Heart Association; American Stroke Association. Guidelines for the early management of patients with ischemic stroke: 2005 guidelines update a scientific statement from the Stroke Council of the American Heart Association/American Stroke Association. Stroke. 2005;36:916 923. 9. Arnold M, Schroth G, Nedeltchev K, Loher T, Remonda L, Stepper F, Sturzenegger M, Mattle HP. Intra-arterial thrombolysis in 100 patients with acute stroke due to middle cerebral artery occlusion. Stroke. 2002; 33:1828 1833. 10. The European Stroke Initiative Recommendations for Stroke Management. European Stroke Council, European Neurological Society and

666 Stroke April 2010 European Federation of Neurological Societies. Cerebrovasc Dis. 2000; 10:335 351. 11. The European Stroke Initiative Recommendations for Stroke Management Update 2003. Cerebrovasc Dis. 2003;16:311 337. 12. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19:604 607. 13. Luengo-Fernandez R, Gray AM, Rothwell PM. Effect of urgent treatment for transient ischaemic attack and minor stroke on disability and hospital costs (EXPRESS study): a prospective population-based sequential comparison. Lancet Neurol. 2009;8:218 219. 14. Coutts SB, Hill MD, Campos CR, Choi YB, Subramaniam S, Kosior JC, Demchuk AM. Recurrent events in transient ischemic attack and minor stroke. Stroke. 2008;39:2461 2466. 15. Dewey HM, Donnan GA, Freeman EJ, Sharples CM, Macdonell RA, McNeill JJ, Thrift AG. Interrater reliability of the National Institutes of Health Stroke Scale: rating by neurologists and nurses in a community-based stroke incidence study. Cerebrovasc Dis. 1999;9:323 327. 16. Kasner SE, Chalela JA, Lucaino JM, Cucchiara BL, Raps EC, McGarvey ML, Conroy MB, Localio AR. Reliability and validity of estimating the NIH stroke scale from medical records. Stroke. 1999;30:1534 1537. 17. Williams LS, Yilmaz EY, Lopez-Yunez AM. Retrospective assessment of initial stroke severity with the NIH Stroke Scale. Stroke. 2000;31: 858 862. 18. Bushnell CD, Johnston DC, Goldstein LB. Retrospective assessment of initial stroke severity: comparison of the NIH Stroke Scale and the Canadian Neurological Scale. Stroke. 2001;32:656 660.