Short stature (SS) is an important referral cause for evaluation in pediatric endocrinology. Growth hormone deficiency (GHD) needs to be considered when other causes of BE are excluded, but there are limitations in establishing its definitive diagnosis, being the subject of several debates and controversies. Although highly questioned, GH stimulation tests are still considered the standard for the diagnostic confirmation of GHD. The present study aimed to evaluate the sensitivity, specificity and accuracy of the different GH peak cut-off points for diagnosis of GHD, in response to stimulus by insulin tolerance test (ITT) and clonidine test, in addition to identifying the best GH peak level to confirm diagnosis using a Receiver Operating Characteristics (ROC) curve analysis. For this purpose, a retrospective and observational study was carried out. Clinical and laboratory data from 62 patients at the endocrinology department of the Hospital Universitário João de Barros Barreto (HUJBB) were collected. The gold standard considered for performance analysis of cut-off points in both GH stimulation tests was the therapeutic response. Thus, 26 patients who achieved a height increase of at least 0,3 standard-deviation at the end of one year of treatment with recombinant human GH (rhGH) were classified as GHD. The remaining patients who did not obtain this gain formed the group called non-GHD. Both groups had similar mean height (p = 0,8155) and gained height at the end of follow-up, but this gain was higher in the GHD group compared to the non-GHG group (20,5 ± 14,8 cm vs. 9,2 ± 6,7 cm, respectively, p = 0,0064). GHD group had a significantly lower meddle GH peak than the non-GHG group in both tests (p <0,0001). Sensitivity, specificity and accuracy of cut-off points 3, 5, 7 and 10 ng/mL were defined in the TTI and in the clonidine test, and there was no superiority of one test over the other. In addition, the cut-off points found were 7,92 ng/mL and 6,78 ng/mL in the TTI and clonidine test, respectively, based on the construction of the ROC curve, representing the most sensitive and specific GH peak levels for the diagnosis of GHD. We conclude that the cut-off points found in this study may represent an emerging tool in the selection of patients who would probably benefit from treatment with rhGH, both in cases of GHD for a known cause and in cases of IGHD.
|
