HIP FRACTURE IN ELDERLY MEN - THE IMPORTANCE OF SUBCLINICAL VITAMIN-DDEFICIENCY AND HYPOGONADISM

Objective: To determine the major risk factors for hip fracture in eld erly men. Design: Prospective recruitment, followed by analysis of cli nical and biochemical variables. Patients and setting: Men aged 60 yea rs and older who presented to St George Hospital (a 650-bed tertiary-c are centre) in 1995, comprising all 41 men with hip fractures, as well as 41 hospital inpatient and 41 outpatient control subjects without h ip fractures. Main outcome measures: Osteoporotic risk factors (includ ing age, body weight, comorbid illnesses, alcohol intake, cigarettes s moked, and corticosteroid use) and serum concentrations of creatinine, urea, calcium, albumin, alkaline phosphatase, parathyroid hormone, 25 -hydroxyvitamin D and free testosterone. Results There were no signifi cant differences between the hip fracture and two control groups on an y oi the osteoporotic risk factors. Men with hip fracture had signific antly lower mean serum 25-hydroxyvitamin D concentration (45.6 nmol/L; 95% confidence interval [CI], 36.9-52.3 nmol/L) than both inpatient ( 61.1 nmol/L; 95% Ci, 50.0-72.2 nmol/L) and outpatient: (65.9 nmol/L; 9 5% CI, 59.0-72.8 nmol/L) controls (P = 0.007). Subclinical vitamin D d eficiency (defined as , 50 nmol/L serum 25-hydroxyvitamin D) was 63% i n the fracture group, compared with 25% in the control groups combined (odds ratio, 3.9; 95% CI, 1.74-8.78; P = 0.0007). Inpatients with and without hip fractures had significantly lower mean serum albumin, cal cium and free testosterone concentrations than outpatients (P < 0.05). In a multiple regression analysis, subclinical vitamin D deficiency w as the strongest predictor of hip fracture (beta [regression coefficie nt], 0.34 +/- 0.19; P = 0.013). Conclusions: Subclinical vitamin D def iciency in Australian men may contribute significantly to the developm ent of hip fracture through the effects of secondary hyperparathyroidi sm, resulting in increased bone loss.