Aromatase inhibitors, GnRH analogues and growth hormone for short stature with advanced bone age in adolescent boys: a Bayesian network meta-analysis.

INTRODUCTION: Adolescent boys presenting late with short stature and advanced bone age (BA/CA >1) have guarded height outcomes. Whether aromatase inhibitors (AI) or GnRH analogues (GnRHa) combined with growth hormone (GH) improve height is unknown. We compared AI+GH, GnRHa+GH, AI alone, and GH alone in boys with reduced predicted adult height and advanced bone age. METHODS: We conducted a PROSPERO-registered systematic review and Bayesian network meta-analysis, reported per PRISMA-NMA, with a pre-specified letrozole-versus-anastrozole sensitivity analysis; certainty was rated using CINeMA. Eight observational studies (651 boys) were included. Outcomes were change in predicted adult height (ΔPAH; primary), final adult height, height standard deviation score (SDS) for chronological and bone age, bone-age advancement ratio (ΔBA/ΔCA), and hyperuricemia, ranked by SUCRA with mean differences (95% credible intervals). RESULTS: AI+GH ranked first for ΔPAH (SUCRA 0.99) and was the only regimen significantly superior to GH, improving ΔPAH by +3.58 cm (1.83-5.14) and final adult height by +2.70 cm (0.12-5.16). AI+GH also ranked highest for height SDS for chronological age and ΔBA/ΔCA and tied with GnRHa+GH for height SDS for bone age. GnRHa+GH ranked second but did not differ from GH. Hyperuricemia was more frequent with AI+GH (47%) than with GH or GnRHa+GH (0%). CINeMA certainty was very low across comparisons. DISCUSSION: In adolescent boys with advanced bone age, AI+GH was the highest-ranked strategy for improving predicted and final adult height, ahead of GnRHa+GH and monotherapy, although very low-certainty evidence warrants confirmatory trials.