Introduction: Current recommendations for diagnosing central precocious puberty (CPP) in girls suggest using basal LH levels >0.3 IU/l to predict progression into CPP and using stimulated LH values >5 IU/l in the LHRH test to diagnose CPP. Our objectives were to test the efficacy of using basal LH values as well as to establish diagnostic cut-offs for LHRH tests.
Method: Retrospective data collection of LHRH test results of 173 girls between 2 and 10 years old, from a regional paediatric centre between 2005 and 2013. Measure of progression into CPP was based on clinicians judgment following LHRH testing: 56 girls were in progression group and 117 girls in non-progression group.
Results: There was a statistically significant difference in basal LH, basal FSH, LHRH-stimulated LH levels and stimulated LH/FSH ratios between the two groups. A basal LH level ≥3.4 IU/l yielded a negative predictive value (NPV) of 89.2%. When basal LH ≥3.4 IU/l or basal FSH ≥3.4 IU/l were used; there was 80.4% sensitivity and 79.3% specificity. If both conditions were present, sensitivity rose to 92.2%. There were different diagnostic cut-offs between 30 and 60 min, which have not been previously reported. The optimal diagnostic cut-off for stimulated LH levels at 30 and 60 min was ≥5.4 and >4.1 IU/l respectively. Optimal cut-offs for stimulated LH/FSH ratio at 30 and at 60 min were >0.63 and >0.88 respectively. A stimulated LH/FSH ratio at 30 min >0.63 produced 89.3% sensitivity, 85.5% specificity, PPV 74.6% and NPV 96.2%.
Conclusion: Using basal LH and FSH levels together is a useful screening test to rule out CPP in the majority of girls. If an LHRH test is required, we have reported novel cut-offs for LH levels at 30 and 60 min, and have shown how the LH/FSH ratio has overall greatest diagnostic value.
12 - 14 Nov 2014
British Society for Paediatric Endocrinology and Diabetes