Hypoglycemia insulin test in the assessment of the hypothalamic-pituitary-adrenal function

BACKGROUND: Recent advances in sensitivity and specificity of hormone immunoanalysis and their automatization have brought about changes in clinical laboratories that led us to review the need for some endocrine dynamic tests. The specific aim of this study was to determine the basal cortisol values predicting a normal or impaired response to the insulin hypoglycemia test (ITT). SUBJECTS AND METHOD: We retrospectively analysed cortisol responses to ITT in 320 subjects. Moreover, we studied the impact the use of a new strategy has on economic cost, by means of a relative value unit (RVU) calculation. RESULTS: No patient with a basal cortisol < 6 mug/dl (13%) responded to the test whereas all those with a basal cortisol > 18 mug/dl responded in full. The rest of patients exhibited a basal cortisol level between 6-18 mug/dl, 39% of them responding and 16% showing an inadequate response. Baseline and peak cortisol concentrations were strongly correlated (r = 0.74; p < 0.0001). The cost of ITT was 131.6 RVU as compared to 17.8 RVU for cortisol. CONCLUSIONS: Basal cortisol levels below 6 mug/dl or above 18 mug/dl make the test unnecessary. Cortisol measurement by automated methods, along with subsequent reduced assay times, allows us to apply new diagnosis strategies. Considering that the cost of ITT is 15 fold higher than that of single cortisol measurement, the potentially generated saving is significant.