Abstract 2802: Age and Stroke Severity Influence the Association between Blood Pressure Variability and Stroke Recovery

Abstract

Blood pressure elevations after ischemic stroke are regarded as the body's response to maintain cerebral perfusion. Extremes of blood pressure have correlated with poor stroke recovery. Recently, blood pressure variability was shown to predict outcome better than mean blood pressures. Variables such as age, National Institute of Health stroke scale (NIHSS) and diabetes are established predictors of poor outcome and may potentially affect the degree of blood pressure variability. Our hypothesis was that age, NIHSS and diabetes would influence the association between blood pressure variability and stroke recovery. Methods: a retrospective chart review of consecutive patients discharged with an acute ischemic stroke based on ICD-9 codes was conducted. Successive blood pressure recordings during the 1st 5 days of hospitalization were noted. Blood pressure variability was calculated for systolic, diastolic and mean arterial pressures separately using published methods. Demographics, comorbidities, and modified Rankin score (mRS) at discharge were abstracted. Results: The study included 295 stroke patients, (median age 60 years; median NIHSS of 4.0). Age >60, NIHSS 4-9, NIHSS ≥ 10 and diabetes were associated with poor recovery (mRS 3-6). The mean pressures in the study group ranged from 105-195 mmHg for systolic blood pressure (SBP), 51-159 mm Hg for diastolic blood pressure and 70-163 mm Hg for mean arterial pressure. The median variability for SBP was 21.0mmHg; diastolic blood pressure was 12.7mmHg; mean arterial pressure was 13.6mmHg. Subjects with higher SBP variability (greater than the median of 21.0) had significantly higher rates of poor recovery. Stratifying by age, the association of SBP variability and poor outcome was significant for age > 60(OR: 2.16; 95%CI 1.06-4.38) but not for the younger subgroup with age≤ 60. Stratifying by stroke severity, the association of SBP variability and poor recovery was significant for NIHSS 4-9 (OR 3.41; 95%CI 1.55-8.23). All patients with high SBP variability in the NIHSS ≥ 10 group had poor recovery. The association was also significant in the subgroup without diabetes (OR 1.97; 95%CI 1.10-3.51), not in the subgroup of diabetic patients. Overall, the odds of a poor recovery due to high SBP variability were highest among subjects > 60 years with moderate to high NIHSS. Conclusions: SBP variability in acute ischemic stroke is especially harmful among the elderly, nondiabetic and the more severe strokes. Extreme caution is advised when prescribing blood pressure lowering medications in these subgroups.