This Article Abstract Full Text (PDF) All Versions of this Article: 37/3/769    most recent 01.STR.0000204041.13466.4cv1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hand, P. J. Articles by Wardlaw, J. M. Search for Related Content PubMed PubMed Citation Articles by Hand, P. J. Articles by Wardlaw, J. M. Related Collections Acute Stroke Syndromes Emergency treatment of Stroke Related Articles

(Stroke. 2006;37:769.)
© 2006 American Heart Association, Inc.

Original Contributions

Distinguishing Between Stroke and Mimic at the Bedside

The Brain Attack Study

Peter J. Hand, MD, FRACP; Joseph Kwan, MD, MRCP; Richard I. Lindley, MD, FRACP; Martin S. Dennis, MD, FRCP Joanna M. Wardlaw, MD, FRCP, FRCR

From the Division of Clinical Neurosciences (M.S.D., J.M.W.), Western General Hospital, Edinburgh, UK; RMH Stroke Centre (P.J.H.), Department of Neurology, Royal Melbourne Hospital, and Department of Medicine (Neuroscience), Monash University (Alfred Hospital Campus), Victoria, Australia; University Department of Geriatric Medicine (J.K.), Level E (MP807), Southampton General Hospital, Southampton, UK; and Department of Geriatric Medicine (R.I.L.), Western Clinical School, University of Sydney, Westmead Hospital, New South Wales, Australia.

Correspondence to Dr Peter Hand, Department of Neurology, c/-Post Office, Royal Melbourne Hospital, Victoria 3050, Australia. E-mail peter.hand{at}mh.org.au

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Background and Purpose— The bedside clinical assessment of the patient with suspected stroke has not been well studied. Improving clinical skills may accelerate patient progress through the emergency department. We aimed to determine the frequency and nature of stroke mimics and to identify the key clinical features that distinguish between stroke and mimic at the bedside.

Methods— Consecutive presentations to an urban teaching hospital with suspected stroke were recruited. A standard bedside clinical assessment was performed. The final diagnosis was determined by an expert panel, which had access to clinical features, brain imaging, and other tests. Univariate and multivariate analyses determined the bedside features that distinguished stroke from mimic.

Results— There were 350 presentations by 336 patients. The final diagnosis was stroke in 241 of 350 (69%) and mimic in 109 (31%). The mimics included 44 events labeled "possible stroke or TIA." Eight items independently predicted the diagnosis in patients presenting with brain attack: cognitive impairment and abnormal signs in other systems suggested a mimic, an exact time of onset, definite focal symptoms, abnormal vascular findings, presence of neurological signs, being able to lateralize the signs to the left or right side of the brain, and being able to determine a clinical stroke subclassification suggested a stroke.

Conclusions— The bedside clinical assessment can be streamlined substantially. This has important implications for teaching less experienced clinicians how to assess the patient with suspected stroke.

Key Words: diagnosis • stroke, acute • stroke assessment

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Stroke is a clinical diagnosis, supported in some cases, but not all, by an appropriate abnormality on brain imaging. Despite its limitations, the clinical assessment directs immediate management of the patient with suspected stroke. For patients to receive time-critical treatments (such as thrombolysis, medical or surgical treatment of intracerebral hematoma, reversal of anticoagulation), they must be brought to hospital rapidly, assessed quickly and accurately, and promptly sent for the appropriate investigation. Many studies show that stroke patients arrive at hospital early.^1,2 One of the major factors that explains the low proportion of patients who are treated with thrombolysis is the delay in processing acute stroke patients through the emergency department and to the scanner.^3–5 Delays may in part be attributable to the uncertainty of trainee doctors (the first contact point in emergency rooms) who lack confidence dealing with acute neurological patients.⁶

We may be able to improve our management of acute stroke by examining the first interaction between patient and medical staff: the bedside assessment. There have been few comprehensive studies of the clinical assessment.⁷ We prospectively examined consecutive patients who presented to our hospital with suspected stroke. Our aims were to determine the frequency and nature of stroke mimics and to identify the key clinical features that distinguish between stroke and mimic at the bedside.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
This was an observational, prospective study of consecutive patients admitted to hospital with possible stroke. It was based in an urban teaching hospital with a 16-bed acute stroke unit, an emergency department, and access to typical investigations (computed tomography [CT]/magnetic resonance, carotid ultrasound, echocardiography, etc). The local ethics committee approved the study, and all patients (or their relatives) provided informed consent.

Inclusion Criteria
We recruited consecutive patients admitted with possible stroke, which we called a "brain attack." This was defined as apparently focal brain dysfunction of apparently abrupt onset. Focal brain dysfunction could be a symptom or a sign. We wanted to study all patients with suspected stroke, so we did not set time limits for inclusion in the study. Subarachnoid hemorrhage was not considered to be a brain attack or stroke in this study.

The primary source of study patients came from the emergency department staff directly paging the research fellow when a suitable patient arrived. Other overlapping sources (admission registers of the emergency department, stroke unit, and neurology ward) were used to ensure that all patients with brain attack were identified.

Patient Assessment
A research fellow assessed patients as soon as possible after hospital presentation, before investigations were performed (or blind to the results). Four research fellows participated in the study. All were between 5 and 9 years postregistration and were undertaking a cerebrovascular fellowship; 2 were trained neurologists, and 2 were trainees in internal medicine.

The research fellow performed a complete bedside assessment of each patient and recorded details on a standard data form. Information collected included: (1) past medical history, including vascular risk factors, previous stroke or transient ischemic attack (TIA), risk factors for a stroke mimic (cognitive impairment, migraine, epilepsy, malignancy, and psychological disturbance); (2) history of the present event, including nature of the neurological symptoms, timing of symptom onset, change in symptoms over time; (3) general examination, including level of consciousness, vascular status, signs in other systems; (4) neurological examination, including the National Institutes of Health Stroke Scale (NIHSS); and (5) diagnostic formulation, including stroke or stroke mimic, Oxfordshire Community Stroke Project (OCSP) subclassification.⁸

Determination of the Final Diagnosis
We used the consensus opinion of a panel of stroke experts to determine the final diagnosis of the event, which was assigned after reviewing anonymized clinical details, brain imaging, and other relevant investigations. The panel comprised 1 stroke neurologist, 2 stroke physicians, a neuroradiologist, and 2 research fellows.

A definite nonstroke was diagnosed when the clinical details did not suggest a vascular etiology, and another convincing explanation for the symptoms was discovered (often requiring supportive investigations, eg, tumor). A definite stroke was diagnosed when the history and examination were considered to be completely typical of a vascular brain event, and there was supportive or noncontradictory brain imaging. A definite TIA required full resolution of symptoms within 24 hours. A probable stroke had clinical features consistent with a vascular etiology, with no alternative explanation. A possible stroke had clinical features that were less convincing, and an alternative explanation for the clinical syndrome may have been present, but there was no definite proof of a nonstroke. A possible TIA was a possible stroke that resolved within 24 hours.

Statistical Analysis
We dichotomized the final diagnosis into stroke or mimic to permit analysis of the clinical features that distinguished between the 2 conditions. Definite and probable stroke (or TIA) were classed as stroke, whereas definite nonstroke and possible stroke (or TIA) were classed as mimic.

Differences between the 2 groups were assessed using descriptive statistics and standard tests of significance (as indicated). The odds ratio (OR) with 95% CIs was calculated for univariate analyses. Analyses were performed on either the total number of episodes or the total number of patients depending on the nature of the data collected (because the same patient could be recruited into the study multiple times, each hospital admission was defined as an episode).

Forward stepwise multiple logistic regression was performed to determine the clinical factors that independently predicted the diagnosis. To ensure that modeling produced reliable results, we conformed to strict methodological principles.^9,10 Only the first recruiting event was used for this analysis. The outcome was a final diagnosis of stroke or mimic, and predictor variables were dichotomized wherever possible. We eliminated predictor variables with too much missing data, in which the event rate was low, the reliability of the item was poor or moderate (Hand et al, unpublished data, 2005), or that duplicated other items and combined several variables to create composite variables. The performance of the model produced by the multivariate analysis was assessed by: (1) the Hosmer and Lemeshow test, which measures goodness of fit of the model to the data set (when the significance is low [P<0.05], the model does not fit the data set); (2) the accuracy of the predictions of the model; and (3) the area under the receiver operating characteristic (ROC) curve, which measures the discrimination of the model (area under the curve can range from 0.5 (no discrimination) to 1.0 (perfect discrimination).

Analyses were performed using Microsoft Excel (version 97 SR-2), SPSS (version 11.0.0) and Confidence Interval Analysis software (Martin J. Gardner and British Medical Journal, 1989).

Sample Size Calculation
Reliable multivariable statistical analysis requires that there are 10 outcome events for each variable modeled.¹⁰ Assuming 20% to 30% frequency of mimics, a sample size of 350 to 400 would recruit 80 mimics, thus permit logistic regression modeling with 8 variables.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
We studied 350 consecutive presentations with brain attack in 336 patients; 8 patients presented twice, and 3 patients presented 3 times. An exact time of onset could be determined in 247 episodes (71%), an approximate time could be established in all but 5 episodes, and in 31%, the symptoms were first noted on waking. A total of 116 presentations (33%) were within 3 hours of symptom onset, but the study clinicians saw only 32 (9%) within 3 hours. Table 1 describes the baseline demographic data for the patients recruited. At the time of examination by the research fellow, 47 of 350 (13%) presentations had no neurological signs.

View this table: [in this window] [in a new window]   TABLE 1. Characteristics of the Study Population, Subdivided Into Stroke and Mimic

What Conditions Cause Brain Attack?
The expert panel determined the final diagnosis as definite stroke in 186 (53%), definite TIA in 17 (5%), and probable stroke in 34 (10%). A definite nonstroke was diagnosed in 65 (19%), possible stroke in 35 (10%), and possible TIA in 13 (4%). Of the 48 presentations of brain attack labeled possible stroke/TIA, there was an alternate, plausible nonstroke diagnosis in all but 4. The dichotomized final diagnosis was stroke in 241 of 350 (69%) and mimic in 109 (31%).

Of the 106 patients (109 episodes) who presented with a stroke mimic, 44 (42%) had experienced a previous stroke (with symptoms completely resolved in 19 of 44), and 27 (26%) were known to have cognitive impairment. The causes of stroke mimic are detailed in Table 2, which is subdivided by time of patient presentation. A total of 62 of 109 (57%) mimics were neurological conditions, and in an additional 20 mimics (syncope, confusional state, dementia), neurological conditions were among the differential diagnoses. The most frequent site of sepsis was the chest, and the most common toxic/metabolic disturbance was hypoglycemia.

View this table: [in this window] [in a new window]   TABLE 2. Causes of Stroke Mimics (n=109)*, Subdivided by Time to Presentation

Features Distinguishing Stroke From Mimic at the Bedside
Table 1 describes the clinical features of patients recruited into the study, subdivided into stroke and mimic. Univariate analyses are shown in Figure 1 (only the significant relationships have been shown). A mimic was more likely if there was a known history of cognitive impairment, the patient lost consciousness or had a seizure at onset, the patient could still walk, there were no lateralizing symptoms, and the examination revealed confusion, signs in other nonvascular systems (eg, chest crackles) and no neurological signs (P<0.05 for all). A mimic was also more likely if the signs were inconsistent with the symptoms or did not conform to known vascular territory.

View larger version (44K): [in this window] [in a new window]   Figure 1. Items from the clinical assessment that were statistically significant in predicting the final diagnosis from univariate analyses. An OR >1 predicts a stroke; <1 predicts a mimic.

Strong bedside pointers to the diagnosis of stroke included definite focal symptoms, the patient was well in the last week, and an exact time of onset could be determined. Stroke was more likely if the patient had almost any focal neurological symptom or sign, although the frequency of the item was often low (eg, the OR for eye deviation was 11.5 [95% CI, 1.53 to 86.3], but it was observed in only 23 patients with stroke). Symptoms and signs suggesting a brain stem lesion, such as vertigo and lower limb ataxia, were not significant predictors because these features were also observed in peripheral vestibular disorders (a common stroke mimic).

The NIHSS was useful in distinguishing mimic from stroke (Figure 2). A low NIHSS predicted a mimic, but 19% of brain attacks with an NIHSS >10 were attributable to a mimic. These were often patients with a previous stroke who presented with an intercurrent infection or metabolic disturbance. The OCSP classification of a large anterior circulation or lacunar subtype predicted a stroke, and a mimic was likely if the OCSP subtype could not be determined (eg, patient presenting with dysarthria only).

View larger version (15K): [in this window] [in a new window]   Figure 2. Proportion of brain attacks attributable to stroke or mimic subdivided by NIHSS score.

After multivariate analyses, 8 items of the bedside assessment independently predicted the diagnosis (Table 3). The model created by multivariate analysis performed well. The model fitted the data set (Hosmer and Lemeshow test P=0.746), there were 83% correct classifications, and area under the ROC curve was 0.87 (95% CI, 0.83 to 0.91). Excluding the 30 events labeled possible or definite TIA, in which symptoms resolved within 24 hours, did not alter the findings on univariate and multivariate analyses.

View this table: [in this window] [in a new window]   TABLE 3. Logistic Regression Model for Predicting the Diagnosis of Brain Attack

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Although laboratory investigations and brain imaging can refine the diagnosis (and are essential for any decisions regarding treatment), the bedside clinical assessment remains important because it is the first step in the diagnostic pathway and often directs the speed at which more complex procedures are undertaken. Brain imaging, even diffusion-weighted MRI, is not infallible and may give confusing results.¹¹ Despite the need for a rapid, confident clinical diagnosis in the thrombolysis era, the clinical assessment has received little formal study.⁷

Of the 350 consecutive presentations of brain attack in our study, 30% were stroke mimics. This figure is somewhat higher than many hospital-based studies (eg, 1.2%¹² to 5.0%¹³), but this may be explained by their more selective entry criteria. Less selective community-based studies reported higher proportions of mimics (eg, 25%,¹⁴ 29%¹⁵), but they may not be as relevant to the hospital-based stroke physician. Libman et al¹⁶ retrospectively identified all patients presenting with "sudden onset of a focal deficit" to a general hospital. A total of 78 of 411 (19%) had a stroke mimic, the same proportion as were diagnosed definite nonstroke in our study. Few studies report figure for possible stroke. This is curious; although some mimics will be definitively diagnosed (by brain imaging or other laboratory tests), it is a clinical reality that many cannot be diagnosed with certainty.

The stroke-mimicking conditions identified in our study were similar to previous reports. Many mimics are seen infrequently, such as transient global amnesia, demyelination, spinal cord lesions, and so on. A total of 82 of 109 (75%) mimics in our study were neurological disorders, yet many had normal brain imaging. Conversely, almost half (42%) of patients with a mimic had experienced a previous stroke, and many of these patients would have an abnormal brain scan. Because stroke is a clinical diagnosis, these data reinforce the need for neurologists, or stroke physicians with adequate neurological training, to be involved in the assessment of patients with brain attack. This has been argued by others since the 1950s^17–20 and remains relevant now.

We identified 47 clinical factors that significantly distinguished between stroke and mimic on univariate analysis. Libman et al¹⁶ found that female gender, abnormal visual fields, diastolic blood pressure >90 mm Hg and atrial fibrillation increased the odds of stroke; and normal eye movements and an abnormal admission neurological examination increased the odds of a mimic. In our study, diastolic blood pressure >90 mm Hg and abnormal visual fields predicted stroke, but female gender, atrial fibrillation, and normal eye movements were not significant predictors, and an abnormal neurological examination actually suggested stroke (rather than mimic). Ferro et al²¹ found that a mimic was more likely if the patient had no vascular risk factors, but we were unable to confirm this. Older studies^22,23 suggested that the temporal evolution of symptoms distinguished vascular from nonvascular events, but this is of little benefit in hyperacute assessment. In our study, an exact time of onset, the patient being able to recall exactly what he/she was doing at symptom onset, and being well in the last week were all strongly predictive of stroke, and all point to an abrupt onset.

We found that 8 items independently predicted the diagnosis in patients presenting with brain attack. The only other multivariate analysis identified just 2 independent predictors: decreased level of consciousness predicted a mimic, and angina predicted a stroke.¹⁶ Our findings show that the bedside clinical assessment can be streamlined substantially. This has important implications for teaching the bedside assessment of suspected stroke to less experienced clinicians. Patients with acute neurological conditions can be daunting for an inexperienced clinician.⁶ With better knowledge of the key features that reliably distinguish stroke from mimic, as identified in our study, the inexperienced clinician’s assessment can be brief but more focused and assured.

This study had a number of limitations. The entry criteria may have been too restrictive (or overly inclusive) but were similar to many other studies.^16,24 We saw few patients within 3 hours, but our aim was to capture all events. Our cohort was older and the stroke severity was milder than other series,^25,26 which might make the bedside diagnosis more difficult. Our gold standard diagnosis was not independent of the research fellow’s assessment, and most patients did not have MRI. It is difficult to determine a gold standard for the diagnosis of stroke.²⁷ Fewer patients with a mimic were scanned, reflecting clinical reality but also introducing bias. Confounding factors included differences in experience and training of the research fellows, their improvement in clinical skills with time, and the inability to obtain key data in some situations (eg, aphasic patient with no relative). Finally, there are many well-described problems with logistic regression modeling, and internal validation does not imply that the model can be generalized to other cohorts of patients.²⁸ The results of our study need to be validated in further prospective studies.

Despite its limitations, our study provides numerical scientific data to support the "art" of the clinical assessment of patients with suspected stroke. Much of what we have shown would be familiar to the experienced stroke clinician. Knowing the stroke mimics and the key clinical features that help discriminate stroke from mimic (and the relative importance of each feature) means this can be taught to inexperienced doctors to help them gain knowledge and skill. The information from this study should be considered complementary to brain imaging and other laboratory tests. Our study provides a method for accelerating the patient’s passage from the emergency department door to the acute stroke unit and treatment, via the CT or MRI scanner.

	Acknowledgments

 
P.J.H. was funded by Chief Scientist Office, Health Department, Scottish Executive, grant reference CZB/4/14. We thank Sarah Keir, Bart Lamont, Janneke Haisma, and Anne Rowat for assisting with patient recruitment, Steff Lewis for statistical advice, and Vera Soosay for developing the computer database. All authors were involved in devising the study, and reviewing drafts of manuscripts. P.J.H. and J.K. recruited and examined the patients. P.J.H. analyzed the data and wrote the manuscript. J.M.W. and M.S.D. obtained funding and were responsible for the overall study.

Received November 13, 2005; revision received December 18, 2005; accepted January 1, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 
1. Harraf F, Sharma AK, Brown MM, Lees KR, Vass RI, Kalra L. A multicentre observational study of presentation and early assessment of acute stroke. BMJ. 2002; 325: 17–22.[Abstract/Free Full Text]

2. Lacy CR, Suh DC, Bueno M, Kostis JB. Delay in presentation and evaluation for acute stroke: Stroke Time Registry for Outcomes Knowledge and Epidemiology (S.T.R.O.K.E.). Stroke. 2001; 32: 63–69.[Abstract/Free Full Text]

3. Kwan J, Hand P, Sandercock P. A systematic review of barriers to delivery of thrombolysis for acute stroke. Age Ageing. 2004; 33: 116–121.[Abstract/Free Full Text]

4. Barber PA, Zhang J, Demchuk AM, Hill MD, Buchan AM. Why are stroke patients excluded from TPA therapy? An analysis of patient eligibility. Neurology. 2001; 56: 1015–1020.[Abstract/Free Full Text]

5. Evenson KR, Rosamond WD, Morris DL. Prehospital and in-hospital delays in acute stroke care. Neuroepidemiology. 2001; 20: 65–76.[CrossRef][Medline] [Order article via Infotrieve]

6. Johnston F, Wardlaw J, Dennis MS, Lewis S, Nimmo G, Lindley RI, Warlow CP. Delays in stroke referrals. Lancet. 1999; 354: 47–48.[CrossRef][Medline] [Order article via Infotrieve]

7. Goldstein LB, Simel DL. Is this patient having a stroke? J Am Med Assoc. 2005; 293: 2391–2402.[Abstract/Free Full Text]

8. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet. 1991; 337: 1521–1526.[CrossRef][Medline] [Order article via Infotrieve]

9. Concato J, Feinstein AR, Holford TR. The risk of determining risk with multivariable models. Ann Intern Med. 1993; 118: 201–210.[Abstract/Free Full Text]

10. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996; 49: 1373–1379.[CrossRef][Medline] [Order article via Infotrieve]

11. Engelter ST, Wetzel SG, Radue EW, Rausch M, Steck AJ, Lyrer PA. The clinical significance of diffusion-weighted MR imaging in infratentorial strokes. Neurology. 2004; 62: 574–580.[Abstract/Free Full Text]

12. O’Brien PA, Ryder DQ, Twomey C. The role of computed tomography brain scan in the diagnosis of acute stroke in the elderly. Age Ageing. 1987; 16: 319–322.[Abstract/Free Full Text]

13. Besson G, Hommel M, Besson S, Guignier M, Robert C. Scoring systems for the differential diagnosis of ischemic and hemorrhagic stroke: response. Stroke. 1996; 27: 337–338.[Medline] [Order article via Infotrieve]

14. Anderson CS, Jamrozik KD, Burvill PW, Chakera TM, Johnson GA, Stewart-Wynne EG. Ascertaining the true incidence of stroke: experience from the Perth Community Stroke Study, 1989–1990. Med J Aust. 1993; 158: 80–84.[Medline] [Order article via Infotrieve]

15. Ellekjaer H, Holmen J, Indredavik B, Terent A. Epidemiology of stroke in Innherred, Norway, 1994 to 1996. Incidence and 30-day case-fatality rate. Stroke. 1997; 28: 2180–2184.[Abstract/Free Full Text]

16. Libman RB, Wirkowski E, Alvir J, Rao TH. Conditions that mimic stroke in the emergency department. Implications for acute stroke trials. Arch Neurol. 1995; 52: 1119–1122.[Abstract/Free Full Text]

17. Rankin J. Cerebral vascular accidents in patients over the age of 60: III. Diagnosis and treatment. Scot Med J. 1957; 2: 255–268.

18. Bratina P, Greenberg L, Pasteur W, Grotta JC. Current emergency department management of stroke in Houston, Texas. Stroke. 1995; 26: 409–414.[Abstract/Free Full Text]

19. Hachinski V. Thrombolysis in stroke: Between the promise and the peril. J Am Med Assoc. 1996; 276: 995–996.[Abstract/Free Full Text]

20. Horowitz SH. Thrombolytic therapy in acute stroke: neurologists, get off your hands! Arch Neurol. 1998; 55: 155–157.[Free Full Text]

21. Ferro JM, Pinto AN, Falcao I, Rodrigues G, Ferreira J, Falcão F, Azevedo E, Canhão P, Melo TP, Rosas MJ, Oliveira V, Salgado AV. Diagnosis of stroke by the nonneurologist. A validation study. Stroke. 1998; 29: 1106–1109.[Abstract/Free Full Text]

22. Allen CM. Clinical diagnosis of the acute stroke syndrome. Q J Med. 1983; 52: 515–523.[Medline] [Order article via Infotrieve]

23. Weisberg LA, Nice CN. Intracranial tumors simulating the presentation of cerebrovascular syndromes. Early detection with cerebral computed tomography (CCT). Am J Med. 1977; 63: 517–524.[CrossRef][Medline] [Order article via Infotrieve]

24. von Arbin M, Britton M, de Faire U, Helmers C, Miah K, Murray V. Validation of admission criteria to a stroke unit. J Chronic Dis. 1980; 33: 215–220.[CrossRef][Medline] [Order article via Infotrieve]

25. Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1000 consecutive patients with first stroke. Stroke. 1988; 19: 1083–1092.[Abstract/Free Full Text]

26. Foulkes MA, Wolf PA, Price TR, Mohr JP, Hier DB. The Stroke Data Bank: design, methods, and baseline characteristics. Stroke. 1988; 19: 547–554.[Abstract/Free Full Text]

27. D’Olhaberriague L, Litvan I, Mitsias P, Mansbach HH. A reappraisal of reliability and validity studies in stroke. Stroke. 1996; 27: 2331–2336.[Abstract/Free Full Text]

28. Braitman LE, Davidoff F. Predicting clinical states in individual patients. Ann Intern Med. 1996; 125: 406–412.[Abstract/Free Full Text]

Related Articles:

Interobserver Agreement for the Bedside Clinical Assessment of Suspected Stroke

Peter J. Hand, Janneke A. Haisma, Joseph Kwan, Richard I. Lindley, Bart Lamont, Martin S. Dennis, and Joanna M. Wardlaw
Stroke 2006 37: 776-780. [Abstract] [Full Text] [PDF]

Improving the Clinical Diagnosis of Stroke

Larry B. Goldstein
Stroke 2006 37: 754-755. [Extract] [Full Text] [PDF]

This article has been cited by other articles:

				C. Foerch, J. Montaner, K. L. Furie, M. M. Ning, and E. H. Lo Invited Article: Searching for oracles?: Blood biomarkers in acute stroke Neurology, August 4, 2009; 73(5): 393 - 399. [Abstract] [Full Text] [PDF]

				A. I. Scher, L. S. Gudmundsson, S. Sigurdsson, A. Ghambaryan, T. Aspelund, G. Eiriksdottir, M. A. van Buchem, V. Gudnason, and L. J. Launer Migraine Headache in Middle Age and Late-Life Brain Infarcts JAMA, June 24, 2009; 301(24): 2563 - 2570. [Abstract] [Full Text] [PDF]

				D. T. Winkler, F. Fluri, P. Fuhr, S. G. Wetzel, P. A. Lyrer, S. Ruegg, and S. T. Engelter Thrombolysis in Stroke Mimics: Frequency, Clinical Characteristics, and Outcome Stroke, April 1, 2009; 40(4): 1522 - 1525. [Abstract] [Full Text] [PDF]

				R I Lindley Commentary on NICE guidelines for alteplase for the treatment of acute ischaemic stroke Heart, December 1, 2007; 93(12): 1617 - 1618. [Full Text] [PDF]

				L. B. Goldstein Acute Ischemic Stroke Treatment in 2007 Circulation, September 25, 2007; 116(13): 1504 - 1514. [Full Text] [PDF]

				A. E. Baird Blood Genomics in Human Stroke Stroke, February 1, 2007; 38(2): 694 - 698. [Abstract] [Full Text] [PDF]

				L. B. Goldstein Improving the Clinical Diagnosis of Stroke Stroke, March 1, 2006; 37(3): 754 - 755. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 37/3/769    most recent 01.STR.0000204041.13466.4cv1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hand, P. J. Articles by Wardlaw, J. M. Search for Related Content PubMed PubMed Citation Articles by Hand, P. J. Articles by Wardlaw, J. M. Related Collections Acute Stroke Syndromes Emergency treatment of Stroke Related Articles