(CSEMP006) CURRENT PRACTICES IN DIAGNOSIS OF ADRENAL INSUFFICIENCY AND ACTH STIMULATION TESTS IN CHILDREN: A SURVEY AMONG NORTH AMERICAN PEDIATRIC ENDOCRINOLOGISTS

Background: Accurately diagnosing adrenal insufficiency (AI) is essential, considering the serious risks of unrecognized AI, and the potential negative consequences of overdiagnosis. Nevertheless, there are still many uncertainties and controversies around this issue, especially in the pediatric population. Therefore, we set out to evaluate current practices of pediatric endocrinologists in North America regarding the diagnosis of AI.

METHODS AND RESULTS: Members of the Pediatric Endocrine Society received an anonymous survey via email in February of 2022. This consisted of 12 main items, and employed branching logic, tailoring questions to respondents' answers. Most questions were multiple choice with the option of free-text responses. RedCap was used to administer the survey and store data.

A total of 221 survey responses were included. Respondents were predominantly pediatric endocrinologists (90%), and the remainder were nurses or trainees. 20% had less than 5 years of experience in pediatric endocrinology, and 35% more than 15 years.

All participants ordered ACTH stimulation tests for the diagnosis of AI; additionally, 82% used morning cortisol levels and 77% considered “critical” cortisol levels (during hypoglycemia, hemodynamic instability). For suspected primary AI, 85% used high-dose ACTH stimulation tests (HDST), but there was more variability for secondary AI: 62% ordered low-dose ACTH stimulation tests (LDST), 22% HDST, and 14% used both tests. The most frequent time points for sampling cortisol levels were 0 and 60 minutes for the HDST (47% of respondents), and 0, 30 and 60 minutes for the LDST (44%).

Regarding interpretation of ACTH stimulation test results, 95% considered peak cortisol levels, 70% the patient’s clinical picture, and 49% the relative increase in cortisol levels from baseline. The median (IQR) cortisol level deemed sufficient after ACTH stimulation testing was 496 (428-496) nmol/L. 17% used different cutoffs for LDST, and 18% had different cutoffs for newborns. For respondents who used non-stimulated cortisol levels (morning, random or during stress) to diagnose AI, median (IQR) level to rule out AI was 276 (221–276) nmol/L. Finally, 47% of participants were unaware of the method used in their institution for cortisol measurements. Further, among the 25% who knew that their laboratory used immunoassays, 31% were not able to specify which immunoassay platform was employed.

Conclusion: We found wide variability regarding ACTH stimulation test protocols and interpretation of results by pediatric endocrinology providers in North America. These exploratory results highlight the need for pediatric-specific recommendations, based on the best available evidence.