Objective: The diagnosis of growth hormone deficiency (GHD) in adults is established through growth hormone (GH) stimulation testing, which is often complex, expensive, time-consuming and may be associated with adverse side effects. The decision to perform GH provocative testing is influenced by clinical findings, medical history and biochemical evidence. We report in this study our experience using the glucagon stimulation test (GST) in assessing GHD in adult patients with traumatic brain injury (TBI) as it relates to baseline serum insulin-like growth factor-1 (IGF-1) concentrations.
Design: A receiver operating characteristic (ROC) curve analysis was performed to determine the optimal IGF-1 cut-off for diagnosis of GHD at different potential diagnostic GST cut-off values (<3, <5, & <10 μg/l).
Patients: One hundred and thirty-eight patients (98 men and 40 women) with a documented history of moderate to severe TBI were assessed for GHD using serum IGF-1 concentrations and the GST.
Measurements: IGF-1 values were compared with peak GH values obtained following the GST.
Results: An IGF-1 cut-off value of 175 μg/l minimized the misclassification of GHD patients and GH-sufficient patients and provided a sensitivity of 83% and specificity of 40%, as well as a negative predictive power of 90% considering a criterion for peak GH response of <3 μg/l.
Conclusions: Our current findings are consistent with previous work assessing peak GH response using the insulin tolerance test (ITT) in a non-TBI sample, suggesting that diagnostic accuracy may be optimized if the GST is used when obtained serum IGF-1 concentrations are below 175 μg/l. While the decision to perform provocative testing to assess GHD in adult patients should be based on the clinician's clinical impression, the findings from this retrospective study can provide useful clinical information and serve as a guide.
© 2011 Blackwell Publishing Ltd.