OBJECTIVE: Spontaneous intraparenchymal hemorrhage (IPH) is a common neurosurgical emergency, with hemorrhage size and expansion associated with poor clinical outcomes. In this study, radiologic risk factors and specific IPH volume thresholds were calculated to identify heightened risk of neurologic deterioration and mortality. METHODS: A consecutive review of all patients with nontraumatic IPH transferred to a tertiary academic neurosurgery service was performed over 2 years. IPH volume, hemorrhage location, presence of intraventricular hemorrhage, hydrocephalus, anticoagulant use, and neurologic status were reviewed. A maximum Youden index was calculated to determine thresholds of initial IPH volume and expansion most predictive of deterioration and mortality. RESULTS: A total of 202 transfers were studied. Both initial IPH volume at the outside hospital and IPH expansion were correlated with neurologic deterioration and death. The most predictive threshold for mortality was 32 mL of initial IPH volume (area under the curve 0.758, P < 0.001, confidence interval 1.012-1.035) and 1 mL of expansion. The threshold for neurologic deterioration was 18 mL of initial volume (area under the curve 0.690, P = 0.004, confidence interval 1.004-1.025) and 1 mL of expansion. Both intraventricular hemorrhage and hydrocephalus were independently associated with elevated risk for deterioration and mortality, while anticoagulant use was associated with neurologic deterioration. CONCLUSIONS: Volume and growth of IPH are significant predictors of neurologic deterioration and death. An initial volume over 32 mL is associated with increased mortality risk, whereas risk of neurologic deterioration appears to peak at a smaller volume of 18 mL. Any measurable IPH expansion suggests elevated risk of deterioration and mortality.
OBJECTIVE: Spontaneous intraparenchymal hemorrhage (IPH) is a common neurosurgical emergency, with hemorrhage size and expansion associated with poor clinical outcomes. In this study, radiologic risk factors and specific IPH volume thresholds were calculated to identify heightened risk of neurologic deterioration and mortality. METHODS: A consecutive review of all patients with nontraumatic IPH transferred to a tertiary academic neurosurgery service was performed over 2 years. IPH volume, hemorrhage location, presence of intraventricular hemorrhage, hydrocephalus, anticoagulant use, and neurologic status were reviewed. A maximum Youden index was calculated to determine thresholds of initial IPH volume and expansion most predictive of deterioration and mortality. RESULTS: A total of 202 transfers were studied. Both initial IPH volume at the outside hospital and IPH expansion were correlated with neurologic deterioration and death. The most predictive threshold for mortality was 32 mL of initial IPH volume (area under the curve 0.758, P < 0.001, confidence interval 1.012-1.035) and 1 mL of expansion. The threshold for neurologic deterioration was 18 mL of initial volume (area under the curve 0.690, P = 0.004, confidence interval 1.004-1.025) and 1 mL of expansion. Both intraventricular hemorrhage and hydrocephalus were independently associated with elevated risk for deterioration and mortality, while anticoagulant use was associated with neurologic deterioration. CONCLUSIONS: Volume and growth of IPH are significant predictors of neurologic deterioration and death. An initial volume over 32 mL is associated with increased mortality risk, whereas risk of neurologic deterioration appears to peak at a smaller volume of 18 mL. Any measurable IPH expansion suggests elevated risk of deterioration and mortality.