Role of fasting plasma glucose, glycated haemoglobin and homeostatic model assessment in the detection of glucose intolerance in adult hypopituitary patients during growth hormone replacement therapy.

AIMS: This study was designed to determine the sensitivity and specificity of conventional criteria for diagnosis of impaired glucose tolerance (IGT) in a high-risk population of GH-treated GH deficient (GHD) adults.

METHODS: 33 hypopituitary GHD patients with HbA(1c) >5.1% and 13 gender- and age-matched control GHD patients were selected. Oral glucose tolerance test (OGTT), fasting plasma glucose (FPG), HbA(1c), and homeostatic model assessment (HOMA) parameters were determined in all patients. Receiver operator characteristic curves were used to determined sensitivity and specificity for the detection of glucose intolerance as defined by plasma glucose >7.8 mmol/l at 120 min during OGTT.

RESULTS: Sensitivity and specificity for this purpose for HbA(1c) (>5.1%) were 89 and 17%; for FPG (>5.5 mmol/l): 78 and 67%; for FPG (>6.1 mmol/l): 56 and 89%; for HOMA-derived beta-cell function (betaCF) (<40%): 78 and 58%; for HOMA-derived insulin sensitivity (IS) (<70%): 11 and 89%, and for betaCF-IS hyperbolic product (betaCF-IS) (<54%): 89 and 75%, respectively.

CONCLUSIONS: This study shows that FPG (>5.5 mmol/l) and betaCF-IS have high sensitivity and relatively high specificity for the detection of IGT and confirms that measurement of FPG or calculation of betaCF-IS provides appropriate safety surveillance in hypopituitary patients on GH replacement.