Which Parameters of Beat-to-Beat Blood Pressure and Variability Best Predict Early Outcome After Acute Ischemic Stroke?

Abstract

In hypertensive populations, increasing blood pressure (BP) levels and BP variability (BPV) are associated with a greater incidence of target organ damage. After stroke, elevated 24-hour BP levels predict a poor outcome, although it is uncertain whether shorter-length BP recordings assessing mean BP levels and BPV have a similar predictive role. The objectives of this study were to compare the different measures of beat-to-beat BP and BPV on outcome after acute ischemic stroke and assess whether these parameters were affected by stroke subtype. Ninety-two consecutive admissions with a CT-confirmed diagnosis of acute ischemic stroke were recruited, of whom 54 had cortical infarction, 29 subcortical, and 9 posterior circulation infarction. Casual and two 5-minute recordings of beat-to-beat BP (Finapres, Ohmeda) were made under standardized conditions within 72 hours of ictus, with mean BP levels taken as the average of this 10-minute recording and BPV as the standard deviation. Outcome was assessed at 30 days as dead/dependent or independent (Rankin </=2). The effects of BP, BPV, and stroke subtype on outcome were studied with the use of logistic regression. Stroke subjects were subsequently divided by BP quartiles and within each quartile into low- and high-variability groups; the influence of high BPV on outcome was also assessed. The odds ratio for death/dependency was significantly higher in cortical strokes compared with subcortical and posterior circulation strokes even after controlling for differences in BP and BPV (OR 4.19, P=0.002). Beat-to-beat systolic BP (SBP), diastolic BP (DBP), and mean arterial pressure (MAP +/- SD) levels were higher in the dead/dependent group compared with the independent group (MAP 106+/-20.4 mm Hg vs 97+/-19.1 mm Hg, P<0.02), as was MAP variability: 6.1 (interquartile range 4.5 to 7.4 mm Hg) versus 4.9 (3.8 to 6.4 mm Hg, P=0.02). The odds ratio for a poor outcome was 1. 38 (P=0.014) for every 10-mm Hg increase in MAP and 1.32 (P=0.02) for every 1-mm Hg increase in MAP variability. Casual BP measurements had no prognostic significance. For the group as a whole when separated into BP quartiles, those with a high MAP and DBP but not SBP variability within each quartile had a worse prognosis compared with those with a low BPV. A poor outcome at 30 days after ischemic stroke was dependent on stroke subtype, beat-to-beat DBP, and MAP levels and variability. Important prognostic information can be readily obtained from a short period of noninvasive BP monitoring in the acute stroke patient. These findings have important implications, particularly regarding the use of hypotensive agents in the acute stroke period.