NYMC Faculty Publications

The Magnitude of Blood Pressure Reduction Predicts Poor In-Hospital Outcome in Acute Intracerebral Hemorrhage

Journal Title

Neurocritical Care

First Page


Last Page


Document Type


Publication Date





BACKGROUND: Early systolic blood pressure (SBP) reduction is believed to improve outcome after spontaneous intracerebral hemorrhage (ICH), but there has been a limited assessment of SBP trajectories in individual patients. We aimed to determine the prognostic significance of SBP trajectories in ICH.

METHODS: We collected routine data on spontaneous ICH patients from two healthcare systems over 10 years. Unsupervised functional principal components analysis (FPCA) was used to characterize SBP trajectories over first 24 h and their relationship to the primary outcome of unfavorable shift on modified Rankin scale (mRS) at hospital discharge, categorized as an ordinal trichotomous variable (mRS 0-2, 3-4, and 5-6 defined as good, poor, and severe, respectively). Ordinal logistic regression models adjusted for baseline SBP and ICH volume were used to determine the prognostic significance of SBP trajectories.

RESULTS: The 757 patients included in the study were 65 ± 23 years old, 56% were men, with a median (IQR) Glasgow come scale of 14 (8). FPCA revealed that mean SBP over 24 h and SBP reduction within the first 6 h accounted for 76.8% of the variation in SBP trajectories. An increase in SBP reduction (per 10 mmHg) was significantly associated with unfavorable outcomes defined as mRS > 2 (adjusted-OR = 1.134; 95% CI 1.044-1.233, P = 0.003). Compared with SBP reduction < 20 mmHg, worse outcomes were observed for SBP reduction = 40-60 mmHg (adjusted-OR = 1.940, 95% CI 1.129-3.353, P = 0.017) and > 60 mmHg, (adjusted-OR = 1.965, 95% CI 1.011, 3.846, P = 0.047). Furthermore, the association of SBP reduction and outcome varied according to initial hematoma volume. Smaller SBP reduction was associated with good outcome (mRS 0-2) in small (< 7.42 mL) and medium-size (≥ 7.42 and < 30.47 mL) hematomas. Furthermore, while the likelihood of good outcome was low in those with large hematomas (≥ 30.47 mL), smaller SBP reduction was associated with decreasing probability of severe outcome (mRS 5-6).

CONCLUSION: Our analyses suggest that in the first 6 h SBP reduction is significantly associated with the in-hospital outcome that varies with initial hematoma volume, and early SBP reduction > 40 mmHg may be harmful in ICH patients. For early SBP reduction to have an effective therapeutic effect, both target levels and optimum SBP reduction goals vis-à-vis hematoma volume should be considered.