Certified testosterone immunoassays for hyperandrogenaemia.

ABSTRACT

BACKGROUND: Testosterone (T) measurement in women is problematic leading to initiatives aiming to improve laboratory standardization of commercial assays. We assessed the impact on the clinical diagnosis of functional hyperandrogenic disorders of a total T immunoassay recently certified by the Centers for Diseases Control and Prevention (CDC). METHODS: We conducted a cross-sectional study including 263 consecutive adult premenopausal women presenting with functional ovarian hyperandrogenism-including polycystic ovary syndrome (PCOS)-and 73 nonhyperandrogenic female volunteers who served to define reference ranges. Total T was measured by a local routine direct radioimmunoassay and by a CDC-certified immunochemiluminescence assay. The main outcome measures were total and calculated free T concentrations and percentage of patients with hyperandrogenaemia. RESULTS: Both assays showed a poor concordance for total and calculated free T measurements. Hence, 147 (56%) and 109 (41%) of women had hyperandrogenaemia with the routine and CDC-certified assays, respectively [κ (95%CI) 0.538 (0.441-0.634)]. Free T levels calculated from total T using both assays showed similar correlations with metabolic variables. Women showing hyperandrogenaemia by both methods had the worst metabolic profiles, yet women presenting with hyperandrogenaemia only when using the CDC-certified assay did not show any significant difference compared to nonhyperandrogenic women in anthropometric or metabolic variables. Those women with hyperandrogenaemia only when using the routine assay were more obese and insulin resistant than normoandrogenaemic hirsute patients. CONCLUSIONS: An isolated androgen measurement, even a very specific one, is unlikely to identify the hyperandrogenic milieu that characterizes patients with functional ovarian hyperandrogenism and PCOS.