The Association between Hourly Systolic Blood Pressure Variability and Outcomes in Patients with Intracerebral Hemorrhage is Time-Dependent: Post-hoc Analysis of the ATACH-2 Trial.

Abstract

Background: Systolic blood pressure (SBP) variability has been associated with an increase in rates of death or disability in patients with intracerebral hemorrhage (ICH). We analyzed data from the Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH)-2 trial to determine whether the association between SBP variability and death or disability at 90 days is dependent on the time from randomization.

Methods: The difference between maximum and minimum SBP (hourly SBP range) for the first 24 h after enrollment was used to calculate the hourly SBP variability. The effect of hourly SBP variability was evaluated in logistic regression models on: (1) death or disability (modified Rankin scale score of 4-6 at 90 days), (2) hematoma expansion (increase of > 33% in volume on the computed tomography scan obtained at 24 h) within 24 h, (3) neurological deterioration within 24 h, and (4) acute kidney injury within 72 h after enrollment. We adjusted for age, baseline Glasgow Coma Scale score, intraventricular hemorrhage, hematoma volume, and maximum SBP values for each time window.

Results: A total of 961 patients (mean age ± standard deviation [SD], 62 ± 13 years; 61.9% were men) who were enrolled at a mean ± SD time of 184 ± 56 min from symptom onset were analyzed. The mean ± SD hourly SBP variability was 15.6 ± 16 mm Hg. The hourly SBP variability became significantly lower with increasing time intervals from randomization (ranging from 41.8 ± 23.3 at hour 1 to 12.4 ± 14.0 at hour 24, P < 0.0001). SBP variability at five hours (P = 0.014) and six hours (P = 0.014) after enrollment was significantly associated with death or disability at 90 days, with positive but not statistically significant associations observed at all other points up to eight hours after randomization. Risk of neurological deterioration within 24 h was highly associated with SBP variability, with the largest association observed between one (P < 0.001) and five (P < 0.001) hours following randomization, with significant associations observed up to 22 h following randomization. Risk of hematoma expansion was associated with SBP variability between three (P = 0.015) and eight (P = 0.002) hours after randomization. Statistically significant associations between SPB variability and risk of acute kidney injury were not observed.

Conclusions: Reduced SBP variability within the first eight hours following randomization appears most impactful on both short-term and long-term outcomes in patients with ICH, and the first eight hours may represent a time window for future interventions directed at reducing SBP variability in patients with ICH.