ESPE Abstracts

Objective: To evaluate the accuracy of IGF-1 in diagnosing children with abnormal (low) growth hormone.

Design: a national multicenter cross-sectional study

Method: Two hundred and seventeen patients with short stature (140 boys and 77 girls), age (5-18 years) were evaluated. Anthropometric measurements and pubertal stage evaluations were performed for all children. All patients underwent laboratory tests (CBC, thyroid function, serum IGF-1) and wrist X-ray to determine bone age by using the method of Greulich and Pyle. A provocative GH test was performed with clonidine and glucagon or insulin. Abnormal (low) GH is defined as a peak level of GH < 10 μg/l. the patients were distributed into two groups based on the level of GH. The normal group has normal GH and the abnormal group has low GH response to provocation tests

Results: A significant difference in BMI (P < 0.05), between GHD and ISS groups, BMI was higher in the abnormal GH group than normal GH group however, age (P = 0.821) and BA (P = 0.479) did not. A significant difference in age means (P = 0.001) was found between abnormal and normal IGF-1 groups. The sensitivity and specificity of IGF-I were 79% and 38.6%, respectively using -2SD as the cutoff value for IGF-1. The distribution of IGF-1 z scores looks similar for both groups in the range of ±1SD

Conclusion: Our study suggests that IGF-1 levels had a poor predicting value for abnormal GH. The presence of abnormal IGF-I levels in a short child with normal GH response to provocative tests and the presence of normal IGF-1 in children with abnormal GH response to provocative tests indicate the presence of other strong factors that influence the IG1 levels rather than GH. Increasing the cut-off value of IGF=1 to -1SD as a pilot study did not increase the prediction of children and.