Abstract 101: Is Intracerebral Hemorrhage-Associated Ischemia a Consequence of Blood Pressure Lowering?

Abstract

Listen

Translate article

Introduction: The cause of (presumed) ischemic lesions associated with intracerebral hemorrhage (ICH) is poorly understood. We investigated the relationship between BP lowering and the incidence of ipsilateral diffusion weighted imaging (DWI) lesions in a prospective ICH cohort. Methods: We prospectively enrolled consecutive ICH patients in the NIH-funded DiAgnostic Utility of MRI in Spontaneous Intracerebral Hemorrhage (DASH) study. Two neuroradiologists reviewed the MRIs for evidence of ischemia, defined as: reduced diffusivity ipsilateral to the ICH without evidence of blood products on FLAIR or GRE. Only DWI lesions attributed to tissue compression; vessel compression; or hypoperfusion were included. Patients with post-operative MRIs or insufficient BP data were excluded. Mean arterial blood pressures (MAP) were recorded on admission, and at 6, 12, 18, and 24 hours. Chi-square and t-tests were used as appropriate. Receiver operator characteristic (ROC) curves were created to assess accuracy of predicting DWI lesions. Results: Of 160 patients, 136 met inclusion criteria (median age: 63 (IQR 50-77); median ICH volume: 10 (IQR 4-33cc); median NIHSS: 6 (IQR 2-16); median GCS: 15 (IQR 10-15); median onset to MRI 40 hrs (IQR 25-75). DWI lesions were observed in 78 (57%) patients. Patients with DWI lesions had higher ICH volumes (32 vs 12cc, p < 0.001); higher admission MAP (125 vs 113mmHg, p=0.006); higher maximal MAP reduction (46 vs 33mmHg, p=0.008); and higher mean %MAP reduction (25 vs 17% p=0.006). DWI lesions were not associated with lowest MAP (80 vs 79mmHg, p=0.97) or mean MAP (90 vs 91, p=0.62). ICH volume and maximum MAP reduction predicted DWI lesions with an area under curve (AUC) of 0.70 (95% CI: 0.61-0.78) and 0.63 (95% CI: 0.53-0.72) respectively. Controlling for ICH volume using logistic regression: for every 10% reduction in MAP the risk of DWI lesions increased substantially (OR 1.28, 95% CI: 1.01-1.62). Similarly, each 10% reduction in mean MAP over the first 24 hours had an increased risk of detecting DWI lesions (OR 1.3, 95% CI: 1.01-1.69). The likelihood of having a DWI lesion was highest in patients with > 30mmHg drop in MAP (OR 2.3, 95% CI: 1.09-4.6). In ICH < 10cc (N=70), DWI lesions were not associated with ICH volume (4.1 vs 4.8cc, p=0.40) but with higher admission MAP (125 vs 112mmHg, p=0.045); maximum MAP reduction (45 vs 31 mmHg, p=0.03); and maximum % MAP reduction (34 vs 25%, p=0.03). Conclusions: ICH volume and large BP reductions are both associated with the presence of DWI lesions. The likelihood of having a DWI lesion went up by 30% for each 10% drop in MAP from admission, and was 230% higher in patients with > 30 mmHg reduction in MAP. These data suggest that aggressive BP reduction may contribute to ICH associated ischemia, and that percentage-based BP goals may be more appropriate than “one-size fits all” for clinical trial design. Future studies are needed to clarify causation.