Risk of age-related fractures in patients with unprovoked seizures.

The incidence of age-related fractures was determined in a cohort of 467 Rochester, Minnesota residents with unprovoked seizures. The 30 initial hip fractures observed were significantly more than the 13.19 expected [standardized morbidity ratio (SMR) = 2.3; 95% confidence interval (CI) 1.5-3.3]. The overall incidence of distal forearm (Colles) fractures was not significantly increased (SMR = 1.6; 95% CI 0.9-2.5), based on 17 cases. Hip fracture incidence was increased during the first 10 years after seizure diagnosis but was not related to anti-epileptic drug (AED) use. In contrast, the increase in Colles' fracture incidence was associated with duration of therapy, as the SMR for greater than or equal to 10 years of AED use was 2.4 (95% CI 1.0-5.0). The incidence of vertebral fractures was slightly higher among treated patients but was not associated with duration of AED use. Thus, there was no consistent pattern to support the contention that fracture incidence is greatly increased by long-term AED use.