Abstract
Objective we assessed the diagnostic accuracy of insulin-like growth factor (IGF) 1 measurements with 1 growth hormone stimulation test (GHST) vs performing 2 GHSTs as the standard test to confirm the diagnosis of growth hormone deficiency (GHD) in children. Methods We retrospectively analyzed the baseline characteristics, anthropometric measurements, and laboratory data of 703 children with short stature, aged 4-14 years (mean age, 8.46 +/- 2.7 years), who had undergone 2 GHSTs. We compared the diagnostic values of IGF-1 levels by using a cut-off value of <= 0 SD score, along with results of a single clonidine stimulation test (CST). We evaluated the false-positive rate, specificity, likelihood ratio, and area under the curve (AUC) of the 2 diagnostic methods. GHD was diagnosed if the peak growth hormone level was Results Of the 724 children, 577 (79.7%) had a low IGF-1 level (mean 104.9 +/- 61.4 ng/mL), and 147 (20.3%) had a normal IGF-1 level (mean 145.9 +/- 86.9 ng/mL). GHD was diagnosed in 187 patients (25.8%), of whom 146 (25.3%) had a low IGF-1 level. An IGF-1 level reflecting <= 0 SDs in combination with results of a single CST had a specificity of 92.6%, a false-positive rate of 5.5%, and an AUC of 0.6088. Using an IFG-1 cut-off level of <=-2 SDs did not alter the diagnostic accuracy. Conclusion Low IGF-1 values of <= 0 SDs or <=-2 SDs in combination with results of a single CST had poor diagnostic accuracy for GHD.