ISSN: 2578-4641
Authors: Aljabri KS* and Bokhari SA
Background and Objective: Growth hormone deficiency (GHD) is one of the most important endocrine and treatable causes of short stature. Reports regarding the sensitivity and specificity of insulin-like growth factor binding protein-3 (IGFBP-3) are not consistent. The aim of our study was to analyze the relevance of IGFBP-3 concentration as a screening test for diagnosis of GHD. Design: We retrospectively studied 40 patients whom were evaluated for short stature at the Endocrinology Department of King Fahad Armed Forces Hospital, Jeddah, Saudi Arabia between January 2015 to December 2018. For IGFBP-3 concentration, laboratory reference ranges were based on age and sex. For all eligible patients, IGFBP-3 concentration was determined and an insulin tolerance test (ITT) was performed. Patients with a peak GH of ≤5.0 ng/ml were considered to be GHD. The cut-off for optimal clinical performance measures was determined from the ROC curve. Sensitivity, specificity, positive and negative predictive values were calculated for IGFBP-3 concentration. Results: Mean age was 14.7 ±1.7 years. There were 83 males (30.9%) and 9 females (99.9%) and mean IGFPB-3 concentration was 3783.3 ±1099.7 mcg/L. Results from the ITT indicated that 21 (52.5%) had GHD. Age was not statistically significant different between GHD (14.7 ±1.9 years) and non-GHD (14.7 ±1.7 years), p=0.9. Moreover, there was non statistical significant more males (53.1%) than females ( 50%) in the GHD patients, P=0.9. In addition, there were not statistically significantly different (p=0.9) between GHD (3752.9 ±1295.9 mcg/L) and non-GHD (3816.8 ±867.0 mcg/L) patients. The mean peak for GH concentration was significantly lower in patients with GHD than without GHD (2.2 ±1.3 ng/ml vs. 9.9 ±5.6 ng/ml, p<0.0001). Peak GH concentration was not significantly positively correlated with IGFBP-3 concentration (r=0.103, P=0.5) (figure 1). The AUC was 43.9%. An IGFBP-3 threshold of <3665 mcg/L was selected to emphasize sensitivity rather than specificity. We tested the diagnostic accuracy of several thresholds. With a threshold of IGFBP-3 in reference to age and sex, sensitivity was 19%, specificity was 89% and the negative predictive value for the diagnosis of GHD was 50%. With a threshold of IGFBP-3 <3665 mcg/L, sensitivity was 57%, specificity was 58% and the negative predictive value for the diagnosis of GHD was 55%. With a threshold of <3075 mcg/L, the sensitivity was 29% and the specificity was 84%. A threshold of <2175 mcg/L, gave a positive predictive value of 67% but a negative predictive value of 49%. 11 of the patients with IGFBP-3 concentration above the threshold of <3665 mcg/L (N=20) were normal and 9 had GH deficiency. These 9 GHD patients had IGFBP-3 concentration that did not differ significantly from those of their GH-sufficient counterparts (4890 ± 1080 vs 4345 ± 609 ng/dl, P=0.2). If IGFBP-3 was used as a screening test (with a concentration threshold <3665 mcg/L) and ITT as a confirmatory test, 00 (50%) out of 40 ITT would not have been performed, leading to the misdiagnosis of 9 GH-deficient adults. Thus, in our study population, such a procedure would misdiagnose 9 (43%) out of 21 GHD patients and yield a sensitivity of 57%. Conclusion: Our study demonstrated the poor negative predictive value of IGFBP-3 concentration for the diagnosis of GHD, making it not possible to minimize the use of the “reference test†method ITT. This observation remains to be validated by population-based studies.
Keywords: Growth hormone deficiency and insulin-like growth factor-binding protein-3
