Background: Nocturnal profile of GH secretion is proposed to be a screening diagnostic procedure for GH deficiency (GHD) in short children. Every screening procedure must be characterised by very high sensitivity and good specificity. The data concerning the cut-off level between normal and subnormal nocturnal GH secretion are scarce and non-consistent.
The aim of the study was to assess the clinical utility of 2-h nocturnal profile of GH secretion in diagnosing GHD in children.
Material and methods: Nocturnal GH peak during 2-h assessment (5 samples every 30) was compared in the 2 groups of children, classified as: 1/ GHD decreased both GH peak in 2 stimulating tests (<10 ng/ml) and IGF-I secretion (IGF-I SDS <−1.5 for age and sex (n=24), 2/ idiopathic short stature normal both GH peak in stimulating tests and IGF-I secretion (n=27). ROC analysis was performed in order to assess the sensitivity and specificity of nocturnal GH peak for different cut-off levels.
Results: The best accuracy of 2-h nocturnal GH profile was found for the cut-off level of GH peak=16 ng/ml, with the sensitivity of 87.5% and specificity of 59.3%. The sensitivity on the level of 95% (acceptable for screening procedures) required increasing the cut-off value up to 20 ng/ml. Using the same cut-off level=10 ng/ml decreases the sensitivity of the examination to only 56.2%, with a slight increase of specificity up to 74.1%.
Conclusions: Nocturnal 2-h profile of GH secretion may be a screening tool in diagnosing GHD, however with the cut-off level higher than for GH stimulating tests. There is no evidence for using the same cut-off values for GH peak in nocturnal profile as that established for stimulating tests.
03 - 07 May 2008
European Society of Endocrinology