They investigators state that they have found that women with osteopenia (low bone density) are at an increased risk of stroke. Each standard deviation decrease in bone density at the calcaneus (0.10 g/cm2) was associated with a 1.31-fold increase in incident stroke (95% confidence interval: 1.03 - 1.65), and a 1.7-fold increase (95% CI: 1.1 - 2.7) in fatal stroke, adjusted for age and follow-up time, a result that has now also been seen in two separate cohort studies. There have also been several new advances in cellular biology establishing links between calcium-regulating hormones, the vascular system, and thrombosis. They propose a nested case-cohort study within the Study of Osteoporotic Fractures (SOF) to look for biologic explanations for this association; for example, perhaps elevated levels of parathyroid hormone, are associated with stroke. Participants in SOF (n=9704 women ages 65 years and older) have been followed since 1986-88. Bone density was measured using photon absorptiometric techniques. Serum samples were obtained from these women at baseline, and frozen at -190 degrees Celsius since then. Women with an incident ischemic non-cardiogenic stroke during follow-up (n = 300) will be compared with controls (n = 300). Cases of stroke will be identified by review of annual questionnaires, Medicare and health maintenance organization data tapes, and death certificates, and will be validated by blinded review of medical records using preset criteria. """"""""Controls"""""""" will be randomly selected from the cohort. They will measure several factors in serum (including parathyroid hormone, endogenous estrogens, calcitriol, insulin-like growth factor (IGF)-1, homocysteine, osteocalcin, calcium, phosphate, creatinine and albumin) that may be related to stroke. With 300 cases and 300 controls, the study will have 90% power (at two-sided alpha = 0.05) to detect whether one of these factors is associated with stroke at a relative hazard of 1.26 per standard deviation or more. Should they identify new predictor(s) of stroke, they will compare levels in black women participating in SOF (n = 130) with those in the controls, and will attempt to determine whether there are racial differences in these predictor(s). The investigators point out that many of these potential new risk factors for stroke are amenable to relatively simple treatments. For example, high levels of parathyroid hormone in the elderly can be treated with calcium supplementation. Thus, they state that determining a biologic explanation for the osteopenia-stroke association may suggest new ways to prevent tow of the most common disease of elderly women.
