The Cut-off Limits of Growth Hormone Response to the Insulin Tolerance Test Related to Body Mass Index for the Diagnosis of Adult Growth Hormone Deficiency

Abstract

Introduction: The diagnosis of growth hormone deficiency (GHD) in adults is based on a reduced GH response to provocative tests, such as the insulin tolerance test (ITT) and the GH-releasing hormone (GHRH) + arginine (ARG) test. However, the cut-off limits of peak GH response in lean subjects are not reliable in obese patients; this is noteworthy since adult GHD is often associated with obesity. To date, there are no ITT cut-offs related to body mass index (BMI). Objective: We aimed to evaluate the diagnostic cut-offs of GH response to the ITT in the function of BMI. Methods: The GH response to the ITT was studied in 106 patients with a history of hypothalamic-pituitary disease, a mean age of 48.2 ± 12.4 years, and a mean BMI of 26.8 ± 6.1 kg/m²). Patients were divided into lean, overweight, and obese groups according to their BMI. The lack of GH response to GHRH + ARG test was considered the gold standard for the diagnosis of GHD. The best GH cut-off in the ITT, defined as the one with the best sensitivity (SE) and specificity (SP), was identified using receiver-operating characteristics curve (ROC) analysis. Results: The best GH cut-off in the ITT was 3.5 μg/L in lean subjects (SE 82.1%; SP 85.7%), 1.3 μg/L in overweight subjects (SE 74.1%; SP 85.7%), and 2.2 μg/L in obese subjects (SE 90.0%; SP 50.0%). The diagnostic accuracy was 97.2, 76.5, and 76.7%, respectively. Conclusions: Our data show that the ITT represents a reliable diagnostic tool for the diagnosis of adult GHD in lean subjects if an appropriate cut-off limit is assumed. Overweight and obesity strongly reduce the GH response to the ITT, GH BMI-related cut-off limits, and the diagnostic reliability of the test.