ESPE Abstracts

Background: Early onset of androgen excess in pre-pubertal children is usually caused by called premature adrenarche (PA). PA is characterised by the development of pubic and axillary hair, adult-type body odour in girls <8yrs and in boys <9yrs. PA may be a forerunner condition of polycystic ovary syndrome (PCOS), a complex metabolic disorder also characterised by androgen excess. Women with PCOS have higher risk of developing type 2 diabetes, dyslipidaemia, hypertension, and fatty liver disease. Androgens have been identified to accelerate metabolic risk in PCOS. Whilst the link between PCOS and metabolic dysfunction is well established, studies in PA to assess metabolic risk are limited to small numbers and heterogenous methodology.

Objective: We conducted a systematic review and meta-analysis on metabolic risks in children with PA.

Methods: Cross-sectional case/control studies were identified using Medline, Embase and Cochrane Library. Key outcome measures: Mean differences (MD) in anthropometric measures (height, weight and BMI [SDS/ z-scores]), indices of glucose metabolism (fasting insulin and glucose, HOMA-IR, HbA1c) and lipid metabolism (total cholesterol, high-/low-density lipoprotein cholesterol [HDL, LDL], triglycerides) in children with PA and in healthy controls.

Results: From 4,640 records identified via electronic databases, 25 cross-sectional studies published between 1990 and 2024 were selected, reporting on 823 children with PA and 707 healthy controls. Height SDS (MD 0.66; 95% confidence interval [CI₉₅] 0.36-0.95; I²= 79%) and weight SDS (MD: 0.63; CI₉₅ 0.33 -0.93; I²= 53%) were significantly higher in PA group. These findings persist after subgroup analysis by PA definition or BMI matching, and sensitivity analysis for gender and risk of bias assessment. PA children have higher fasting insulin levels than controls (MD: 15.04; CI₉₅ 5.27-24.81; I²= 91%), however, this difference did not remain significant in the subgroup and sensitivity analyses. Differences in fasting glucose, HOMA-IR, mean serum glucose, mean serum insulin and HbA1c were not statistically significant between PA and controls. Total cholesterol, low- and high-density lipoprotein cholesterol were similar in both groups.

Conclusion: In a sample of more than 800 cases and 700 controls, children with PA have higher fasting insulin levels but no other significant metabolic disturbances. However, we observed great statistical and methodological heterogeneity. A standardised approach in the assessment of metabolic risks in children with idiopathic early-onset androgen excess is warranted, as well as long-term follow-up studies to identify those children with PA who are at higher risk of metabolic dysfunction and may progress to develop PCOS.