RESUMO
Background: The stimulation test with gonadotropin-releasing factor (GnRH) is considered the gold standard to study the activation of the hypothalamic-pituitary-gonadal axis (HPG). A maximum luteinizing hormone (LH) response of more than > 5 mIU/min in girls confirms the activation of the HPG axis. With the advent of more sensitive methods of LH measurement, such as immuno chemiluminescence (ICMA), basal LH values have been proposed to predict the response and replace the GnRH test. Aim: To determine which basal LH value correlates with a GnRH test > 5 uUI/ml. Material and Methods: We retrospectively analyzed LH values, measured by ICMA, in the basal period and after GnRH stimulation, in 120 girls aged 8 +/- 1 years (60 with activated axis and 60 without activation), during 2009. The sensitivity and specificity of the different values were determined by receiver operating characteristic (ROC) curves. Results: In patients with and without activation of the HPG axis, basal LH was 0.5 +/- 0.53 and 0.15 +/- 0.12 mUI/ml, respectively (p < 0.01). According to ROC curves, a basal LH value greater or equal than 0.4 mUI/ml had a specificity of 93.2 percent and a sensitivity of 61 percent to determine activation of the HPG axis. Conclusions: A basal LH > 0.4 mUI/ml in girls with precocious puberty has a specificity of 93.2 percent to determine activation of the HPG axis.