Utility and duration of leuprolide stimulation testing in children.

Objective Basal (unstimulated) LH levels and leuprolide stimulation tests are used to define pubertal status of children presenting with signs of early puberty. The primary aims of this study were to (i) confirm utility of detectable basal LH levels in precluding the need for leuprolide stimulation testing, and, (ii) determine whether duration of testing could be abbreviated from usual 3 h test without compromising sensitivity. Methods We reviewed morning basal and leuprolide-stimulated LH levels in 105 children, aged 1-9 years (mean 6.9 years, SD 1.8) who were seen for concerns of precocious puberty and received a leuprolide stimulation test between June 2006 and March 2017. Results A pubertal basal LH level had high specificity and poor sensitivity for the following outcome measures: (1) peak stimulated LH≥5 mIU/mL (2) treatment with GnRHa; and (3) a composite outcome of (1) and/or (2). Following leuprolide stimulation, LH response was highest at 180 min in most children (n=78, 74.3%). Using a single cutoff of LH≥5 mIU/mL at any timepoint, 25% of children would have been misdiagnosed with an abbreviated 60 min test. A single sample at 180 min would have correctly identified 97% of patients. Conclusions A pubertal basal LH level is sufficient to distinguish children with precocious puberty without stimulation testing. However, prepubertal basal LH had relatively poor negative predictive value to refute CPP, necessitating clinical follow-up and/or a leuprolide stimulation test. For a cutoff of LH≥5 mIU/mL at any timepoint, test duration of 180 min maximizes sensitivity.