Characteristics and outcomes of intracerebral hemorrhage in a population-based stroke registry

Background. Intracerebral hemorrhage (ICH) is the most severe stroke type. Understanding the epidemiology and pathogenesis of ICH is key to design adequate prevention and treatment strategies to improve the dismal prognosis of ICH. Methods. We performed a prospective population-based study in the district of L’Aquila covering the years 2011-2019. ICH incidence, 30-day and 1-year and case-fatality rates (CFRs) were computed in patients residing in the district and suffering a first-ever ICH over the 2011-2017 period (incidence dataset). All the other assessments were performed in residents of the district reporting either a first-ever ICH or an ICH after a stroke during the 2011-2019 period (full dataset). Cases were actively monitored from multiple sources. We classified ICH according to the SMASH-U system. ICH volumes were calculated using the ABC/2 method. The presence of radiological Edinburgh criteria, including associated subarachnoid hemorrhage (aSAH) and finger-like projections (FLPs) was assessed on the first available brain CT of patients with lobar ICH. Crude incidence rates were calculated assuming a Poisson distribution. Incidence rates were also standardized to the European population using the direct method. Univariate comparisons were performed using the chi-square test, t test, ANOVA, Mann-Whitney or Kruskal-Wallis tests, as appropriate. Logistic regression or Cox regression models were used to perform multivariate analyses. Results. We identified 645 patients with ICH (full dataset; 58.6% men; mean age 75.3±13.4 years). The incidence dataset included 514 patients. The crude ICH incidence rate was 24.6 per 100,000 person-years (95% confidence interval [CI] 22.5-26.8); the corresponding rate was 19.4 per 100,000 person-years (95% CI 17.9-20.9) after standardization to the European population. Case-fatality rates were 36.0% at 30 days and 44.6% at 1-year. Compared with the 567 patients with a first-ever ICH, the 78 patients with ICH after a stroke had a higher pre-stroke disability (median modified Rankin Scale score 2, interquartile range [IQR] 1-3, vs 1, IQR 0-2; P<0.001) and higher ICH volume at onset (median 20 cm3, IQR 3-53, vs 8, IQR 2-25; P=0.004), but not higher case-fatality rate. The 34 patients with ICH after hemorrhagic stroke had a higher proportion of lobar location compared with the 44 patients with ICH after ischemic stroke (79.4% vs 40.9%; P=0.009). According to the SMASH-U classification, 39 patients (6.0%) had an ICH attributable to structural lesions, 74 (11.5%) to medication, 41 (6.4%) to systemic or other disease, 217 (33.6%) to amyloid angiopathy, 235 (36.4%) to hypertensive angiopathy, and 39 (6.0%) to undetermined cause. ICH attributable to medication was the only category which independently predicted 30-day (hazard ratio 1.78, 95% CI 1.18-2.67; P=0.006) and 1-year case-fatality (hazard ratio 1.50, 95% CI 1.02-2.19; P=0.038). We included 259 patients with lobar ICH; 87 (33.6%) had both the Edinburgh CT criteria for the classification of probable amyloid angiopathy, i.e. aSAH+FLPs, while 77 (29.7%) had only one and 95 (25.6%) none of the criteria. Patients with aSAH+FLPs also had more severe ICH at onset, higher 30-day (log rank test P=0.009) and 1-year case-fatality (log rank test P=0.003), and higher mRS scores at discharge (P<0.001) as compared to those fulfilling one or none of the Edinburgh criteria. However, the Edinburgh criteria were not independent predictors of case-fatality. Conclusions. In our population-based study, the incidence and outcomes of first-ever ICH were comparable to those reported in similar studies. Applying a classification tool to patients with ICH identified patient categories with different prognosis; however, only ICH attributable to anticoagulant medication was an independent predictor of ICH case-fatality. Besides, the available classification tools are limited by the coexistence of several etiologic factors in the same patient.