Reliability of Agreement between Insulin, Clonidine, and Glucagon Stimulation Tests for the Diagnosis of Growth Hormone Deficiency in Children: A Retrospective Cohort Study

Abstract

Growth hormone (GH) deficiency (GHD) is a rare disorder. The diagnosis of GHD requires a combination of two provocative GH tests. This study aimed to find agreement between commonly used medications to determine which combined tests have high reliability of agreement. This retrospective cohort included 201 children who underwent GH provocation testing from January 2012 to December 2022. The insulin tolerance test (ITT) with the clonidine stimulation test (CST) or glucagon stimulation test (GST) with the CST were performed. We calculated Cohen’s kappa to determine the agreement between the test medications by considering the post-stimulation peak GH level with a cut-off value of 10 ng/mL as the primary outcome. A total of 151 patients underwent the two provocative tests and were included in the analysis. Of these patients, 119 underwent the ITT and CST and 54 (45.3%) were diagnosed with GHD. However, 32 patients underwent the GST and CST and 18 (56.2%) were diagnosed with GHD. The kappa value for ITT and CST was 0.258 (25.8%), indicating fair agreement between clonidine and insulin (p = 0.005). However, the kappa value for CST and GST was 0.178 (17.8%), representing slight agreement. The correlation coefficient revealed a very strong relationship between ITT and CST. Clonidine has fair agreement and a very strong correlation coefficient with ITT when used to diagnose GHD in children. Among the commonly used pharmacological tests for GH provocation in our unit, the CST was considered the best pharmacological test in terms of safety and reduced parental anxiety.

Keywords

• pituitary dwarfism
• inter-observer reliability
• glucagon
• clonidine
• insulin
• growth hormone