Total and free cortisol levels during 1 μg, 25 μg, and 250 μg cosyntropin stimulation tests compared to insulin tolerance test: results of a randomized, prospective, pilot study

Purpose: The appropriate cosyntropin dose during cosyntropin stimulation tests remains uncertain. We conducted a prospective, randomized pilot study to compare 1 μg IV low dose cosyntropin test, 25 μg IM medium dose cosyntropin test, and 250 μg IM standard dose cosyntropin test to evaluate secondary adrenal insufficiency. Insulin tolerance test was used as the gold standard. Method: The study included patients with hypothalamic/pituitary disease (n = 10) with at least one pituitary axis deficiency other than ACTH deficiency and controls (n = 12). All tests were done in random order. Sensitivity and specificity were calculated for total cortisol and serum free cortisol cut-off levels during cosyntropin stimulation tests. Results: The median (range) age and F/M sex ratios for patients and controls were 54 years (23–62), 2/8, and 33 years (21–51), 6/6, respectively. The best total cortisol cut-off during low dose cosyntropin test, medium dose cosyntropin test, 30 min and 60 min standard dose cosyntropin test were 14.6 μg/dL (100% sensitivity & specificity), 18.7 μg/dL (100% sensitivity, 88% specificity), 16.1 (100% sensitivity & specificity), and 19.5 μg/dL (100% sensitivity & specificity), respectively. There was no difference in the ROC curve for cortisol values between the cosyntropin stimulation tests (p > 0.41). Using a cortisol cut-off of 18 μg/dL during cosyntropin stimulation tests, only cortisol level at 30 min during standard dose cosyntropin test provided discrimination similar to insulin tolerance test. The best peak free cortisol cut-off levels were 1 μg/dL for insulin tolerance test, 0.9 μg/dL for low dose cosyntropin test, 0.9 μg/dL for medium dose cosyntropin test, and 0.9 μg/dL and 1.3 μg/dL for 30 min and 60 min standard dose cosyntropin test, respectively. Conclusion: All cosyntropin stimulation tests had excellent correlations with insulin tolerance test, when appropriate cut-offs were used. This pilot study does not suggest an advantage in using 25 μg cosyntropin dose during the cosyntropin stimulation test. A serum free cortisol cut-off of 0.9 μg/dL may be used as pass criterion during low dose cosyntropin test, standard dose cosyntropin test cosyntropin test, and 30 min standard dose cosyntropin test.