Introduction: Androgen deficiency is a risk factor for osteoporosis in men. However, illness and acute fracture may lead to a reduction in circulating testosterone.

Methods: To explore this possibility, we conducted a case-control study of 240 men presenting to the Emergency Department (ED) with a radiologically confirmed minimal trauma fracture (MTF), and of 75 controls.

Results: Compared to controls, cases had lower total testosterone (TT, 7.1 vs 13.2 nmol/L, p < 0.0001) and calculated free testosterone (cFT, 113 vs 172 pmol/L, p < 0.01). Cases were older (74 vs 68 years, p < 0.001), had lower lumbar spine T-score
(-0.6 vs 0.0, p = 0.04), femoral neck T-score (-1.7 vs -1.1, p < 0.0001), and renal function (eGFR 78 vs 82 ml/min, p= 0.02). There was no difference in BMI (27.4 vs 27.9 kg/m2), and 25OH-vitamin D (58 vs 64 nmol/L), p = n.s.. Lower TT remained associated with a higher fracture risk after adjustment for differences between groups including age and bone mineral density (OR 1.21 p< 0.0001). Of the cases, the 142 admitted to the hospital had lower TT than the 98 discharged from the ED (4.6 vs 10.3 nmol/L, p < 0.0001), and lower cFT (78 vs 151 pmol, p < 0.0001). There was also a difference in TT between cases discharged from ED and controls (10.3 vs 13.2 nmol/L p < 0.0001); but not in cFT (151 vs 172 pmol/L, p = n.s.). In the 34 cases with follow-up testosterone (median of 4 months after the initial testosterone), follow-up TT was 8.5 vs 5.1 nmol/L, and cFT was 127 vs 81 pmol/L, both significantly (p <0.001) higher compared to the initial testosterone.

Conclusions: The diagnosis of hypogonadism and appropriate commencement of androgen replacement in men is challenging. Neither should it be based on measurements following minimal trauma where, at least in part, deficits in serum testosterone may be effects of an acute, fracture-associated, stress response.