Peripartum stroke and intracranial venous thrombosis in the National Hospital discharge survey

Abstract
Objectives: To determine nationally representative estimates of the incidence of peripartum stroke and intracranial venous thrombosis and to identify potential risk factors for these conditions. Methods: We analyzed National Hospital Discharge Survey data for the period 1979–1991. Nationally representative estimates of risk were calculated by age, race, method of delivery, presence of pregnancy-related hypertension, census region, hospital ownership, and number of hospital beds. Multivariate models were developed using logistic and Poisson regression. Results: There were an estimated 5484 cases of peripartum stroke and 4454 cases of intracranial venous thrombosis in the United States among 50,110,949 deliveries (risk, 10.3 cases of peripartum stroke and 8.9 cases of intracranial venous thrombosis per 100,000 deliveries). In both univariate and multivariate models, peripartum stroke was associated strongly with cesarean delivery, pregnancy-related hypertension, proprietary hospital ownership, and larger hospital size. Intracranial venous thrombosis was strongly associated with cesarean delivery and less strongly with smaller hospital size. Conclusions: Peripartum stroke and intracranial venous thrombosis are common complications of pregnancy. Although the associations identified in this study are plausible, further studies are required to confirm these associations and elucidate the underlying mechanisms.