Identification of lacunar infarcts before thrombolysis in the ECASS I study

BACKGROUND: The identification of lacunar infarcts before thrombolysis would make it possible either to exclude them from treatment or to show that they also may benefit from it. OBJECTIVE: To determine whether clinical presentation or early CT findings of patients enrolled in the first European Cooperative Acute Stroke Study (ECASS I) trial would identify lacunar infarcts before treatment. METHODS: Predictive values, sensitivity, specificity, and accuracy of clinical presentation as pure motor hemiparesis (PMH) or sensorimotor stroke (SMS) syndromes and of baseline CT findings in predicting lacunar infarcts were calculated in the ECASS I patients. RESULTS: Of 514 patients, 44 placebo (17%) and 44 recombinant tissue plasminogen activator (rt-PA) (18%) patients had PMH/SMS involving at least two of three areas. Thirty-one placebo (12%) and 32 rt-PA (13%) patients had PMH/SMS involving three areas. The 7-day CT was compatible with a lacunar infarct in 32 placebo (12%) and 44 rt-PA (18%) patients. PMH/SMS involving at least two areas had a positive predictive value of 30% both in placebo and rt-PA patients, whereas positive predictive values of the involvement of three areas were 23% and 31%. Those of absence of early CT signs were 21% and 30%, and those of leukoaraiosis or previous lacunar infarcts were 21% and 23%. Positive predictive values of PMH/SMS involving at least two areas combined with absence of early CT signs were 36% in placebo and 33% in t-PA patients, and those of PMH/SMS plus leukoaraiosis or previous lacunes were 28% and 7%, respectively. CONCLUSIONS: In the ECASS I trial, lacunar infarcts were not recognizable on clinical grounds, and early CT findings, alone or in combination with the clinical picture, added poorly to the differential diagnosis.