Published online before print August 23, 2007, doi: 10.1161/STROKEAHA.107.487744
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2007;38:2706.)
© 2007 American Heart Association, Inc.
Original Contributions |
Clinical Diagnosis of Lacunar Stroke in the First 6 Hours After Symptom Onset
Analysis of Data From the Glycine Antagonist In Neuroprotection (GAIN) Americas Trial
From the Queen Elizabeth II Health Sciences Centre (S.J.P., G.J.G.) and the Departments of Medicine (S.J.P., A.M., G.J.G.) and Community Health and Epidemiology (D.D.), Dalhousie University, Halifax, Nova Scotia, Canada.; the University of Virginia (K.C.J.), Charlottesville, Va; the National Institutes of Neurological Disorders and Stroke (W.J.K.), Bethesda, Md; Massachusetts General Hospital (K.L.F.), Boston, Mass; the University of Toronto and Sunnybrook Health Sciences Centre (S.B.), Toronto, Ontario, Canada.; and Akron General Medical Centre (D.E.H.), Akron, Ohio.
Correspondence to Stephen J. Phillips, MBBS, Division of Neurology, Halifax Infirmary Room 3831, 1796 Summer Street, Halifax, Nova Scotia, Canada B3H 3A7. E-mail stephen.phillips{at}dal.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background and Purpose— Although the pathophysiological heterogeneity of stroke may be highly relevant to the development of acute-phase therapies, discriminating between ischemic stroke subtypes soon after onset remains a challenge. We conducted a study of the accuracy of a clinical diagnosis of lacunar stroke in the first 6 hours after symptom onset.
Methods— We analyzed data from 1367 patients in the Glycine Antagonist In Neuroprotection (GAIN) Americas trial. The Trial of ORG10172 in Acute Stroke Treatment (TOAST) category "small vessel (lacunar)" disease at day 7 or at hospital discharge was used as the reference standard to determine the accuracy of a diagnosis of a lacunar stroke made within 6 hours of symptom onset using the Oxfordshire Community Stroke Project (OCSP) classification "LACS." Outcome was analyzed by comparing the proportions of patients classified as "LACS" at baseline or "small vessel (lacunar)" at 7 days who were dead or dependent at 3 months.
Results— The positive predictive value of an OCSP diagnosis of a lacunar stroke was 76% (95% CI: 69% to 81%; sensitivity 64% [95% CI: 58% to 70%]; specificity 96% [95% CI: 95% to 97%]; negative predictive value 93% [95% CI: 92% to 94%]; accuracy 91% [95% CI: 89% to 92%]). The 3-month outcomes of patients classified as either OCSP "LACS" within 6 hours of onset or TOAST "small vessel (lacunar)" at 7 days were not significantly different.
Conclusions— An OCSP LACS diagnosis made within 6 hours of stroke onset is reasonably predictive of a final diagnosis of "small vessel (lacunar)" disease made using TOAST criteria and has a similar relationship to outcome at 3 months.
Key Words: lacunar infarction • stroke assessment • stroke classification • stroke outcome
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
An acute stroke may be described in several different ways. Although there is much contemporary interest in the use of imaging techniques such as perfusion-weighted and diffusion-weighted MRI,1 clinical descriptors will continue to be important2 because: (1) MRI is not feasible in a substantial minority of patients3–5; (2) the narrow time window for intervention restricts the application of MRI6 and other time-consuming special investigations; (3) the relatively long scanning time for MRI is associated with hypoxia in medically unstable patients with severe strokes4; (4) small vessel infarcts, particularly those located in the brainstem, may not be detected by diffusion-weighted MRI in the early hours after onset7,8; and (5) MRI is not widely available.
Stroke treatments with most impact will be delivered soon after symptom onset and will be available to the majority of patients. To facilitate uptake of new treatments in clinical practice, potential interventions must be studied in the settings in which they will be used. Because most therapies under development are likely to be only modestly beneficial, clinical trials have to be large to ensure that random error (the play of chance) does not mask a true treatment effect.9 The requirement for treatment trials to be large, streamlined, and pragmatic10,11 highlights the need for a clinical classification of ischemic stroke that can be applied easily within the first few hours after symptom onset and which reliably distinguishes between stroke subtypes with substantially different outcomes. Such a classification system may be used in stratification and minimization algorithms to ensure that important prognostic factors are balanced between the treatment groups of a clinical trial.
The pathophysiological heterogeneity of ischemic stroke may be relevant to the development of acute-phase therapies because it is possible that what works for one subtype of stroke may work differently for another. Although no clinical stroke syndrome is absolutely pure with respect to pathophysiology, lacunar syndromes are the most homogeneous.12,13 Lacunar syndromes are usually due to a small subcortical infarct in the territory of a penetrating artery13 caused by in situ microatheroma or lipohyalinosis.13–15 Neurochemical studies suggest that subcortical ischemia may respond differently to hyperacute intervention than cortical ischemia.16 Subgroup analyses in a trial of a putative neuroprotective agent suggested the possibility of (an unexpected) benefit in patients with lacunar strokes.17
The Oxfordshire Community Stroke Project (OCSP) classification of ischemic stroke subtypes18 (including lacunar stroke [LACS]) has reasonable interrater reliability,19 provides information about the topography and vascular pathology of the stroke20–24; and is reasonably predictive of death, dependency, and stroke recurrence during the first year poststroke.18 Although most of our knowledge of the prognostic power of the OCSP classification is derived from the original population-based study in which patients were first assessed days after stroke onset,25 the OCSP classification also lends itself for use within the first few hours after stroke onset because it depends only on the clinical findings and a CT scan (to exclude hemorrhage and stroke mimics).
The Trial of ORG10172 in Acute Stroke Treatment (TOAST) investigators defined criteria for distinguishing between the pathophysiological mechanisms of ischemic stroke, including "small vessel (lacunar)" disease,26 that have since become widely used in clinical stroke studies.27 The TOAST criteria are based on a combination of clinical and imaging findings and the results of ancillary investigations and so cannot be easily applied within the first few hours after stroke onset.
GAIN Americas used the OCSP and TOAST classifications, as well as other evaluation tools, to describe the trial subjects. We analyzed the GAIN Americas data to describe the relationships between these tools, focusing on the accuracy of the OCSP classification in the early diagnosis of a lacunar stroke syndrome.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study Population
Analyses were performed on the primary efficacy (nonhemorrhagic) population (n=1367) of the GAIN Americas trial, which was a randomized, double-blind, placebo-controlled trial of gavestinel, an antagonist of the glycine site of the N-methyl-D-aspartate receptor and putative neuroprotective agent for acute ischemic stroke when administered within 6 hours after stroke onset.28 This substudy used data from both arms of the trial because there was no statistically significant difference in outcome between gavestinel- and placebo-treated patients. Patients were eligible for enrollment in GAIN Americas if they were aged 18 years or older, previously functionally independent, and within 6 hours after the onset of a stroke that caused a predefined level of limb weakness but no major reduction in level of consciousness. Baseline evaluation included assessment of stroke severity using the National Institutes of Health Stroke Scale (NIHSS)29 and ischemic stroke subtype according to the OCSP classification.18 Participating investigators were certified in the use of the NIHSS and oriented to the use of the OCSP at the start of the trial.
Brain imaging was performed using either CT within 18 hours of stroke onset or MRI within 6 hours. A committee of 3 independent neuroradiologists blinded to all clinical data read the scans, but the imaging findings were not linked with the neurological findings on clinical examination at baseline. Repeat scanning was performed in an MRI substudy of cortical strokes,6 but follow-up brain imaging was not mandated by the GAIN Americas main protocol.
At day 7 or at hospital discharge, whichever came first, the most likely cause of the stroke was classified by the local investigator according to TOAST criteria26 using information from all available diagnostic studies. The TOAST subtype was assigned if the patient met either the "probable" or "possible" criteria for that subtype. Participating investigators did not receive any special training in the use of the TOAST classification. The trial protocol did not mandate a specific battery of investigations for study subjects.
The trial primary outcome was functional capability in activities of daily living at 3 months measured by the Barthel Index (BI).30 The modified Rankin Scale (MRS)31 was a secondary outcome measure. Guidelines for administering the BI and MRS were reviewed at prestudy training meetings for investigators.
Statistical Analysis
The TOAST subtype "small vessel (lacunar)" was used as the reference standard in the analysis of the positive and negative predictive values, sensitivity, specificity, and accuracy of a diagnosis of the OCSP "lacunar stroke syndrome (LACS)" in the first 6 hours after stroke onset. Outcome was analyzed by comparing the proportions of patients classified as "LACS" at baseline or "small vessel (lacunar)" at 7 days who were dead or dependent (defined using different cut points on the MRS and BI) at 3 months. The relationship among the NIHSS, BI, and MRS at 3 months was analyzed using the Spearman rank order correlation statistic. Group characteristics were compared using parametric and nonparametric statistics, as appropriate. Confidence intervals were estimated by using the exact binomial procedure with continuity corrections. All statistical testing was performed at the 2-tailed alpha level of 0.05. The data were analyzed using SAS software (SAS Institute Inc, Version 9.1).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Figure 1 summarizes the study in the format of a Standards for Reporting of Diagnostic Accuracy (STARD)32,33 flow diagram. Table 1 describes the primary efficacy (nonhemorrhagic) population (n=1367) of the GAIN Americas trial according to OCSP classification at baseline. Fewer than 2% of patients were unclassified. The timing of the neurological examination was not recorded, but the median time from stroke onset to treatment was 5.2 hours. Stroke severity was greatest for total anterior circulation strokes and lowest for LACS (P<0.0001).
|
|
Table 2 describes the study population according to the TOAST classification at 7 days or at the time of discharge from the hospital. Small vessel (lacunar) strokes were the least severe (P<0.0001). The median NIHSS scores for the other TOAST subtypes differed by only 5 points.
|
Table 3 compares the OCSP and TOAST classifications. The positive predictive value of an OCSP diagnosis of a lacunar stroke syndrome was 76% (95% CI: 69% to 81%; sensitivity 64% [95% CI: 58% to 70%]; specificity 96% [95% CI: 95% to 97%]; negative predictive value 93% [95% CI: 92% to 94%]; accuracy 91% [95% CI: 89% to 92%]). At 3 months poststroke, there was no significant difference in the likelihood of death, dependency, or the proportion of patients living at home for patients classified as either OCSP "LACS" within 6 hours of onset or TOAST "small vessel (lacunar)" at 7 days (Table 4).
|
|
Patients with an acute LACS diagnosis not classified as small vessel (lacunar) on day 7 were equally distributed among large vessel (14 patients), cardioembolic (16 patients), and infarct of unknown cause (14 patients). Of the 230 patients with a small vessel (lacunar) TOAST diagnosis at 7 days, 83 were not diagnosed as LACS initially. The majority of these (50 patients) were called partial anterior stroke. Of the 47 patients acutely classified as posterior circulation stroke, 30% were assigned a final diagnosis of small vessel (lacunar) stroke.
Figure 2 summarizes the NIHSS, BI, and MRS scores of the patients who survived 3 months as well as the proportions living at home displayed according to OCSP subtype at baseline and TOAST subtype at 7 days (or at time of discharge from hospital). "Small vessel (lacunar)" strokes had better outcomes than the other TOAST subtypes, but the TOAST classification system did not clearly distinguish differences in outcomes between the other subtypes.
|
There was a high degree of correlation between the NIHSS and MRS (r=0.82; P<0.001) and between the NIHSS and BI (r=–0.75; P<0.001) at 3 months. For each one-point increase in the MRS, the NIHSS increased by approximately 3 points. For each decile increase in the BI, the NIHSS decreased by approximately 2 points.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The main findings of this study are that an OCSP LACS diagnosis made within 6 hours of stroke onset is reasonably predictive of a final diagnosis of "small vessel (lacunar)" disease made using TOAST criteria and that both diagnoses have a similar relationship to better outcome at 3 months when compared with the other ischemic stroke subtypes. We also demonstrated the overlap in 3-month MRS and BI scores for all OCSP and TOAST subtypes and quantified the correlation between the NIHSS and MRS and BI.
The 76% positive predictive value of a clinical diagnosis of lacunar stroke in GAIN Americas is similar to that found in a prospective population-based cohort of patients with ischemic stroke in the Northern Manhattan Stroke Study (NOMASS).34 Among the 195 patients who presented with a lacunar syndrome (diagnosed within 1 week after stroke onset) and had a lacunar infarct on CT or MRI, 147 (positive predictive value=75%) were caused by small vessel disease as determined by a modified Stroke Data Bank scheme35 (a forerunner of TOAST).
There are a number of limitations to the work presented here. The GAIN Americas cohort was selected and so the findings may not apply to all strokes in general. Lacunar strokes not causing limb weakness were excluded; this may have biased our findings in the conservative direction as pure sensory stroke had a positive predictive value of 100% for detecting a lacunar infarct on brain imaging in NOMASS.34 Our analyses did not include any information derived from brain imaging. Although this is of some concern because studies using MRI36,37 have shown that not all lacunar stroke syndromes are caused by solitary, small, deep, infarcts, our findings demonstrate the pragmatic use of distinguishing lacunar strokes from other stroke syndromes. In NOMASS in which 74% of patients had 2 or more CT or MRI scans, a lacunar syndrome had a positive predictive value of 87% for detecting a lacunar infarct on brain imaging.34
We were unable to assess the extent to which brain imaging at baseline influenced the diagnosis of OCSP subtype, but very few patients had an MRI before randomization6 and acute infarcts are not well seen on CT performed within 6 hours after stroke onset.38 The TOAST criteria for "small vessel (lacunar)" stroke include the clinical findings of a lacunar stroke syndrome, and we were unable to assess how repeat brain imaging and other investigations influenced the investigators’ TOAST categorization.
Despite these limitations, our study suggests that lacunar strokes, which are relatively pathophysiologically homogeneous, can be identified with reasonable accuracy in the first 6 hours after stroke onset. The use of the OCSP classification in acute treatment trials would allow an important minority of patients to be categorized at the time of randomization on the basis of most likely mechanism as well as stroke severity.
|
Acknowledgments |
|---|
Sources of Funding
GAIN Americas was sponsored by GlaxoWellcome (now GlaxoSmithKline) Inc, Research Triangle Park, NC. This study was funded by the Capital Health Research Fund, Halifax, Nova Scotia, Canada.
Disclosures
None.
|
Footnotes |
|---|
Current affiliation for D.D.: Health Core Inc, Wilmington, Del. Current affiliation for D.E.H.: Cuyahoga Falls General Hospital-Summa Health Systems, Akron, Ohio.
Received March 11, 2007; revision received April 8, 2007; accepted April 17, 2007.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Lee LJ, Kidwell CS, Alger J, Starkman S, Saver JL. Impact on stroke subtype diagnosis of early diffusion-weighted magnetic resonance imaging and magnetic resonance angiography. Stroke. 2000; 31: 1081–1089.
2. Bamford JM. The role of the clinical examination in the subclassification of stroke. Cerebrovasc Dis. 2000; 10 (suppl 4): 2–4.[Medline] [Order article via Infotrieve]
3. Singer OC, Sitzer M, du Mesnil de Rochemont R, Neumann-Haefelin T. Practical limitations of acute stroke MRI due to patient-related problems. Neurology. 2004; 62: 1848–1849.
4. Hand PJ, Wardlaw JM, Rowat AM, Haisma JA, Lindley RI, Dennis MS. Magnetic resonance brain imaging in patients with acute stroke: feasibility and patient related difficulties. J Neurol Neurosurg Psychiatry. 2005; 76: 1525–1527.
5. Kang D-W, Chalela JA, Dunn W, Warach S, NIH-Suburban Stroke Center Investigators. MRI screening before standard tissue plasminogen activator therapy is feasible and safe. Stroke. 2005; 36: 1939–1943.
6. Warach S, Kaufman D, Chiu D, Devlin T, Luby M, Rashid A, Clayton L, Kaste M, Lees KR, Sacco R, Fisher M; for the GAIN Americas and GAIN International MRI Investigators. Effect of the glycine antagonist gavestinel on cerebral infarcts in acute stroke patients, a randomized–placebo controlled trial: the GAIN MRI substudy. Cerebrovasc Dis. 2006; 21: 106–111.[Medline] [Order article via Infotrieve]
7. Ay H, Buonanno FS, Rordorf G, Schaefer PW, Schwamm LH, Wu O, Gonzalez RG, Yamada K, Sorensen GA, Koroshetz WJ. Normal diffusion-weighted MR imaging during stroke-like deficits. Neurology. 1999; 52: 1784–1792.
8. 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.[CrossRef][Medline] [Order article via Infotrieve]
9. Baigent C. The need for large-scale randomized evidence. Br J Clin Pharmacol. 1997; 43: 349–353.[CrossRef][Medline] [Order article via Infotrieve]
10. Peto R, Baigent C. Trials: the next 50 years. BMJ. 1998; 317: 1170–1171.
11. Tunis SR, Stryer DB, Clancy CM. Practical clinical trials. Increasing the value of clinical research for decision making in clinical and health policy. JAMA. 2003; 290: 1624–1632.
12. Davis SM, Donnan GA. Why lacunar syndromes are different and important. Stroke. 2004; 35: 1780–1781.
13. Norrving B. Lacunar infarction. Embolism is the key: against. Stroke. 2004; 35: 1779–1780.
14. Fisher CM. Lacunes: small, deep cerebral infarcts. Neurology. 1965; 15: 774–784.
15. Fisher CM. The arterial lesion underlying lacunes. Acta Neuropathol (Berl). 1969; 12: 1–15.[CrossRef]
16. Waxman SG. Molecular mechanisms of subcortical versus cortical infarction. In: Donnan GA, Norrving B, Bamford JM, Bogousslavsky J, eds. Subcortical Stroke, 2nd ed. Oxford, UK: Oxford Medical Publications; 2002: 67–83.
17. Aslanyan S, Weir CJ, Muir KW, Lees KR; for the IMAGES Study Investigators. Magnesium for treatment of acute lacunar stroke syndromes: further analysis of the IMAGES trial. Stroke. 2007; 38: 1269–1273.
18. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet. 1992; 337: 1521–1526.
19. Lindley RI, Warlow CP, Wardlaw JM, Dennis MS, Slattery J, Sandercock PAG. Interobserver reliability of a clinical classification of acute cerebral infarction. Stroke. 1993; 24: 1801–1804.
20. Wardlaw JM, Dennis MS, Lindley RI, Sellar RJ, Warlow CP. The validity of a simple clinical classification of acute ischaemic stroke. J Neurol. 1996; 243: 274–279.[CrossRef][Medline] [Order article via Infotrieve]
21. Al-Buhairi AR, Phillips SJ, Llewellyn G, Jan MMS. Prediction of infarct topography using the Oxfordshire Community Stroke Project classification of stroke subtypes. J Stroke Cerebrovasc Dis. 1998; 7: 339–343.[CrossRef][Medline] [Order article via Infotrieve]
22. Tei H, Uchiyama S, Koshimizu K, Kobayashi M, Ohara K. Correlation between symptomatic, radiological and etiological diagnosis in acute ischemic stroke. Acta Neurol Scand. 1999; 99: 192–195.[Medline] [Order article via Infotrieve]
23. Mead GE, Lewis SC, Wardlaw JM, Dennis MS, Warlow CP. How well does the Oxfordshire Community Stroke Project classification predict the site and size of the infarct on brain imaging? J Neurol Neurosurg Psychiatry. 2000; 68: 558–562.
24. Smith CJ, Emsley HCA, Libetta CM, Hughes DG, Drennan RF, Vail A, Tyrrell PJ. The Oxfordshire Community Stroke Project classification in the early hours of ischemic stroke and relation to infarct site and size on cranial computed tomography. J Stroke Cerebrovasc Dis. 2001; 10: 205–209.[CrossRef][Medline] [Order article via Infotrieve]
25. Bamford J, Sandercock P, Dennis M, Warlow C, Jones L, McPherson K, Vessey M, Fowler G, Molyneux A, Hughes T, et al. A prospective study of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Project 1981–86, I: methodology, demography and incident cases of first-ever stroke. J Neurol Neurosurg Psychiatry. 1988; 51: 1373–1380.
26. Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE, the TOAST Investigators. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. Stroke. 1993; 24: 35–41.
27. Ay H, Furie KL, Singhal A, Smith WS, Sorensen AG, Koroshetz WJ. An evidence-based causative classification system for acute ischemic stroke. Ann Neurol. 2005; 58: 688–697.[CrossRef][Medline] [Order article via Infotrieve]
28. Sacco RL, DeRosa JT, Haley EC Jr, Levin B, Ordronneau P, Phillips SJ, Rundek T, Snipes RG, Thompson JLP; for the GAIN Americas Investigators. Glycine antagonist in neuroprotection for patients with acute stroke. GAIN Americas: a randomized controlled trial. JAMA. 2001; 285: 1719–1728.
29. Brott TG, Adams HP, Olinger CP, Marler JR, Barsan WG, Biller J, Spilker J, Holleran R, Eberle R, Hertzberg V, Rorick M, Moomaw CJ, Walker M. Measurements of acute cerebral infarction: a clinical examination scale. Stroke. 1989; 20: 864–870.
30. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J. 1965; 14: 61–65.[Medline] [Order article via Infotrieve]
31. van Sweiten JC, Koudstaal PJ, Visser MC, Shouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988; 19: 604–607.
32. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, Lijmer JG, Moher D, Rennie D, de Vet HCW; for the STARD Group. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Clin Chem. 2003; 49: 1–6.
33. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, Moher D, Rennie D, de Vet HCW, Lijmer JG. The STARD Statement for reporting studies of diagnostic accuracy: explanation and elaboration. Clin Chem. 2003; 49: 7–18.
34. Gan R, Sacco RL, Kargman DE, Roberts JK, Boden-Albala B, Gu Q. Testing the validity of the lacunar hypothesis: the Northern Manhattan Stroke Study experience. Neurology. 1997; 48: 1204–1211.[Abstract]
35. Foulkes MA, Wolf PA, Price TR, Mohr JP, Hier DB. The Stroke Data Bank: design, methods, and baseline characteristics. Stroke. 1988; 19: 547–554.
36. Gerraty RP, Parsons MW, Barber AB, Darby DG, Desmond PM, Tress BM, Davis SM. Examining the lacunar hypothesis with diffusion and perfusion magnetic resonance imaging. Stroke. 2002; 33: 2019–2024.
37. Seifert T, Enzinger C, Storch MK, Pichler G, Niederkorn K, Fazekas F. Acute small subcortical infarctions on diffusion weighted MRI: clinical presentation and aetiology. J Neurol Neurosurg Psychiatry. 2005; 76: 1520–1524.
38. Wardlaw JM, Lewis SC, Dennis MS, Counsell C, McDowall M. Is visible infarction on computed tomography associated with an adverse prognosis in acute ischemic stroke? Stroke. 1998; 29: 1315–1319.
This article has been cited by other articles:
 |
|
 |
|
  D. Toni, M. L. Sacchetti, and M. Prencipe Early Clinical Diagnosis of Lacunar Strokes Stroke, October 1, 2008; 39(10): e152 - e152. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||