Abstract
CONTEXT: The 2019 American Association of Clinical Endocrinologists guidelines suggested peak GH-cutoffs to glucagon test (GST) of ≤3 and ≤1 µg/L in the diagnosis of permanent GH deficiency (GHD) during the transition phase. OBJECTIVE: The aim of the study was to evaluate the accuracy of GST compared to insulin tolerance test (ITT) in the definition of GHD at adult height achievement. PATIENTS AND METHODS: Ninety-seven subjects with childhood-onset GHD (median age, 17.39 years) underwent ITT, GST, and IGF-1 testing; 44 subjects were idiopathic (isolated GHD), 35 moderate organic GHD (0-2 hormone deficiencies) and 18 severe organic GHD (≥3 hormone deficiencies). RESULTS: Bland and Altman analysis showed a high consistency of GH peak measures after ITT and GST. Receiver operating characteristic analysis identified 7.3 μg/L as the optimal GH peak cutoff to GST [95% confidence interval (CI) 4.15-8.91; sensitivity 95.7%, specificity 88.2%, positive predictive value (PPV) 88.0%, negative predictive value (NPV) 95.7%] able to correctly classify 91.8% of the entire cohort while 5.8 μg/L was the best GH peak cutoff able to correctly classify 91.4% of moderate organic GHD patients (95% CI 3.16-7.39; sensitivity 96.0%, specificity 80.0%, PPV 92.3%, NPV 88.9%). Patients with ≥3 hormone deficiencies showed a GH peak <5 μg/L at ITT and <5.8 μg/L at GST but 1. The optimal cutoff for IGF-1 was -1.4 SD score (95% CI -1.94 to 0.77; sensitivity 75%, specificity 94%, PPV 91.7%, NPV 81.0%) that correctly classified 85.1% of the study population. CONCLUSION: A GH peak to GST <5.8 μg/L represents an accurate diagnostic cutoff for young adults with childhood-onset GHD and high pretest probability of permanent GHD.