Comparison of low-dose and high-dose cosyntropin stimulation testing in children

Background: There is no consensus among pediatric endocrinologists in using low-dose (LD) versus high-dose (HD) cosyntropin to test for secondary/tertiary adrenal insufficiency. This paper compares LD and HD cosyntropin stimulation testing in children for evaluation of hypothalamic-pituitary-adrenal axis (HPAA) and suggests a new peak cortisol cut-off value for LD stimulation testing to avoid false positivity. Methods: Data of 36 children receiving LD (1 mu g) and HD (249 mu g) cosyntropin consecutively during growth hormone (GH) stimulation testing were analyzed in two groups. Group A were patients who passed GH stimulation testing and were not on oral, inhaled or intranasal steroids (intact hypothalamic-pituitary axis, n = 19). Group B were patients who failed GH stimulation testing and/or were on oral, inhaled or intranasal steroids (impaired hypothalamic-pituitary axis, n = 17). Results: In group A, the mean peak cortisol response in LD cosyntropin was 18.5 +/- 2.4 mu g/dL and that for the HD cosyntropin was 24.8 +/- 3.1 mu g/dL (r: 0.76, P < 0.05). In group B, the mean peak cortisol response in LD cosyntropin was 15.7 +/- 6.1 mu g/dL and that for HD cosyntropin was 21.7 +/- 7.9 mu g/dL (r: 0.98, P < 0.05). When a standard cut-off of 18 mu g/dL was used, 37% of the patients with intact HPAA failed LD cosyntropin testing, but a cut-off of 14 mu g/dL eliminated false positive results. Conclusions: LD cosyntropin stimulation testing results should be interpreted cautiously when used alone to prevent unnecessary long-term treatment. Using a lower cut-off for LD (>= 14 mu g/dL) seems to avoid false positive results and still detects most cases of impaired HPAA.