BACKGROUND High blood pressure, although the main modifiable risk factor for stroke, may have a beneficial effect in maintaining brain perfusion after acute ischemia. The authors assessed the effects of blood pressure variation in the first 24 hours of stroke onset. METHODS The authors prospectively studied consecutive patients admitted in the first 24 hours after stroke onset. Patients were classified according to the NIH Stroke Scale (NIHSS) and Oxfordshire Stroke Classification Scale (OSCS). Stroke etiology was defined according to Trial of ORG 10172 in Acute Stroke Treatment classification. After 3 months, outcome was assessed using Rankin and Barthel scales, with poor outcome defined as Rankin score > 2 or Barthel score < 70. RESULTS A total of 115 patients were admitted between January 2001 and October 2002. Median NIHSS was 4.5; main stroke etiology was cardioembolism (30%). After 3 months, 44 (39%) patients had a poor outcome. Predictors of poor outcome in univariable analyses (p < 0.05) were as follows: total anterior circulation classification on OSCS, nonlacunar stroke etiology, older age, higher NIHSS score, lack of antiplatelet use, higher body temperature, lower diastolic blood pressure on admission, and a larger degree of systolic blood pressure reduction. In the multivariable analysis, remaining predictors of poor outcome included the following: NIHSS score (OR = 1.55 per 1 point increase; 95% CI = 1.28 to 1.87; p < 0.001) and degree of systolic blood pressure reduction in the first 24 hours (OR = 1.89 per 10% decrease; 95% CI = 1.02 to 3.52; p = 0.047). CONCLUSION Blood pressure reduction in the first 24 hours of stroke onset is independently associated with poor outcome after 3 months.