Serum testosterone levels are not elevated in patients with ankylosing spondylitis

Objective. Studies in patients with ankylosing spondylitis (AS) describe sl ightly elevated serum testosterone levels, but these studies were not prope rly controlled for possible confounders. Methods. In a case-control study serum levels of sex steroids, luteinizing hormone, and sex hormone binding globulin (SHGB) were measured in patients with AS and in age and sex matched controls. The body mass index, smoking s tatus, use of alcohol, and fat intake were recorded. Results. Testosterone levels measured in serum extracts did not differ in 5 0 male patients with AS compared to controls (mean +/- SD 16 +/- 4 vs 15 +/ - 5 nmol/l, respectively; p = 0.54). In unextracted serum, however, male pa tients showed elevated testosterone (p < 0.001) and dehydroepiandrosterone sulfate levels (p = 0.003), even after controlling for confounders (p < 0.0 01). One of 10 female patients had an elevated testosterone level in unextr acted serum, The 17 male users and one of the 2 female users of phenylbutaz one had the highest testosterone levels in unextracted serum, and all showe d a significant decline after extraction. Serum levels of other sex steroid s, luteinizing hormone, and SHGB did not differ significantly between patie nts and controls. Conclusion. Serum testosterone levels are not elevated in male patients wit h AS. Spuriously elevated testosterone levels in unextracted serum might be related to the use of phenylbutazone in our patient sample.