Assessment of risk factors for second hip fractures in Japanese elderly

In an attempt to identify a cohort with a high risk of suffering a fracture of the contralateral hip (second hip fracture), we assessed patients who had suffered hip fracture. A total of 714 patients (130 men and 584 women) were prospectively followed to determine those who suffered a second hip fracture. Pathologic hip fractures and fractures that emerged from high-energy trauma were excluded from the analysis. Age, gender, Singh Index (SI), fracture type, cognitive impairment, and comorbid medical conditions were investigated as medical predictors. The 714 patients were observed for 1,579.5 person-years (mean: 2.4 +/- 1.4 years per patient). During the observation period, 45 second hip fractures were identified (bilateral group), giving an overall incidence of 0.029 per person-year. The annual incidence rate declined linearly from the occasion of the initial fracture. Furthermore, the second hip fracture tended to occur increasingly within 8 months after the initial hip fracture. The second hip fracture was of the same type (trochanteric or cervical) in 79% of the trochanteric and 71% of the cervical fractures. There was no significant difference in the incidence of second hip fracture by gender or age. In addition, there was no significant difference in the distribution of SI grades of the unfractured hip at the initial hip fracture between the 669 patients who had not suffered a second hip fracture (unilateral group) and the bilateral group. Cox proportional hazard regression analysis revealed that increased risk of a second hip fracture was associated with senile dementia and Parkinson's disease. We concluded that careful follow-up of hip fracture patients associated with senile dementia and Parkinson's disease might effectively prevent the incidence of a second hip fracture.