Models Predict Intracerebral Hemorrhage Growth
Models using 4 or 5 predictors have acceptable to good discrimination for determining additional intracerebral hemorrhage growth in patients with acute intracerebral hemorrhage.
HealthDay News — Models using four or five predictors have acceptable to good discrimination for determining additional intracerebral hemorrhage growth in patients with acute intracerebral hemorrhage, according to a review published in the October issue of The Lancet Neurology.
Rustam Al-Shahi Salman, Ph.D., from the University of Edinburgh in the United Kingdom, and colleagues conducted a systematic review to examine the absolute risk and predictors of intracerebral hemorrhage growth. Data were included from 77 observational cohorts and randomized trials with repeat scanning protocols and at least 10 patients with acute intracerebral hemorrhage. The researchers obtained individual-level data for patients who had brain imaging initially conducted 0.5 to 24 hours after symptom onset and repeated fewer than six days after symptom onset. These patients had a baseline intracerebral hemorrhage volume of <150 mL and had not undergone acute treatment to reduce intracerebral hemorrhage volume.
The researchers found that among 5,435 eligible patients, 5,076 were not taking anticoagulant therapy at symptom onset and 20 percent (1,009) had intracerebral hemorrhage growth. The investigators assessed multivariable models of patients with data on antiplatelet therapy use, anticoagulant therapy use, and assessment of computed tomography (CT) angiography spot sign at symptom onset. Independent predictors of intracerebral hemorrhage growth included time from symptom onset to baseline imaging, intracerebral hemorrhage volume on baseline imaging, antiplatelet use, and anticoagulant use (odds ratios, 0.5, 7.18, 1.68, and 3.48, respectively; C-index, 0.78). The C-index increased by 0.05 with the addition of the CT angiography spot sign (odds ratio, 4.46) to the model.
"These models could inform the location and frequency of observations on patients in clinical practice, explain treatment effects in prior randomized trials, and guide the design of future trials," the authors write.
Several authors disclosed financial ties to the pharmaceutical industry.