Incidence rates and case-fatality rates of cerebral vein thrombosis: A population-based study

BACKGROUND AND PURPOSE: Cerebral vein thrombosis (CVT) incidence is estimated to be >10 per 1 000 000 per year. Few population-based studies investigating case-fatality rates (CFRs) and pyogenic/nonpyogenic CVT incidence are available. We assessed trends in CVT incidence between 2002 and 2012, as well as adjusted in-hospital CFRs and incidence of hospital admissions for pyogenic/nonpyogenic CVT in a large Northwestern Italian epidemiological study. METHODS: Primary and secondary discharge diagnoses of pyogenic/nonpyogenic CVT were identified using International Classification of Diseases, Ninth Revision, Clinical Modification codes 325, 671.5, and 437.6. Age, sex, vital status at discharge, length of hospital stay, and up to 5 secondary discharge diagnoses were collected. Concomitant presence of intracerebral hemorrhage (ICH) was registered, and comorbidities were assessed through the Charlson comorbidity index. RESULTS: A total of 1718 patients were hospitalized for CVT (1147 females-66.8%; 810 pyogenic and 908 nonpyogenic CVT, 47.1% and 52.9%, respectively), with 134 patients (7.8%) experiencing a concomitant ICH. The overall incidence rate for CVT was 11.6 per 1 000 000 inhabitants with a sex-specific rate of 15.1 and 7.8 per 1 000 000 in females and males, respectively. CVT incidence significantly increased in women during time of observation (P=0.007), with the highest incidence being at 40 to 44 years (27.0 cases per 1 000 000). In-hospital CFR was 3%, with no difference between pyogenic/nonpyogenic CVT. Patients with concomitant ICH had a higher in-hospital CFR compared with patients without ICH (7.5% versus 2.7%; odds ratio, 2.96 [95% CI, 1.45-6.04]). In-hospital CFR progressively increased with increasing Charlson comorbidity index (P=0.003). Age (odds ratio, 1.03 [95% CI, 1.02-1.05]), Charlson comorbidity index ≥4 (odds ratio, 4.33 [95% CI, 1.29- 14.52]), and ICH (odds ratio, 3.05 [95% CI, 1.40-6.62]) were independent predictors of in-hospital mortality. CONCLUSIONS: In a large epidemiological study, CVT incidence was found to be comparable to the one registered in population-based studies reported after the year 2000. CVT incidence increased among women over time. In-hospital CFR was low, but not negligible, in patients with concomitant ICH. Age, ICH, and a high number of comorbidities were independent predictors of in-hospital mortality. Pyogenic CVT was not a predictor of in-hospital CFR, although its high proportion was not confirmed by internal validation.