The effect of time of day testing and utility of 30 and 60 minute cortisol values in the 250 mcg ACTH stimulation test

Background: Despite widespread use of the 250-mcg Cosyntropin test (ACTH test) for the diagnosis of adrenal insufficiency (AI), the effect of time of day and the utility of performing both 30- and 60-min serum cortisol values remains unclear. Methods: We conducted a retrospective cohort study of all ACTH testing at the Halifax Neuropituitary Program, Nova Scotia, Canada between January 2006 and April 2016. Data were collected on age, gender, medication history, serum cortisol levels at 0, 30 and 60 min after ACTH administration, as well as time of and indication for testing. Results: There were 336 tests performed, divided by time of day (0800–1000 h, 1001–1200 h, and after 1200 h). There were no differences in mean cortisol levels at 30 (574.5, 559, 534 nmol/L, respectively; p = 0.25) and 60 min (642, 623, 619 nmol/L, respectively; p = 0.63) between groups. When comparing 30- vs. 60-min values using a cut-off of ≥500 nmol/L, 45 patients (13.4%) failed to reach the cut-off at 30 min but met the cut-off at 60 min. Conversely, only 2 patients (0.6%) who met the cut-off at 30 min failed to reach it at 60 min. Conclusion: We found that outcomes from ACTH testing are not affected by time of day. Furthermore, using a 30-min cortisol level in isolation results in more than one in seven patients having a false positive diagnosis of AI; a 60-min value of ≥500 nmol/L alone may be sufficient to diagnose AI in >99% of cases.