Avoiding False-Positive Adrenal Insufficiency Diagnoses in Children: Insights from Cortisol Kinetics During Pediatric Low-Dose ACTH Stimulation Test

Background: The low-dose ACTH stimulation test (LDST) is widely used to evaluate hypothalamic–pituitary–adrenal (HPA) axis function in children; however, optimal cortisol sampling times and interpretation strategies remain controversial. Reliance on early or single time-point measurements may lead to false-positive diagnoses of adrenal insufficiency (AI). Objective: To characterize the timing of peak cortisol responses during LDST in children without AI and to assess the incremental diagnostic contribution of extended sampling and time-specific cortisol thresholds. Methods: We retrospectively analyzed 177 pediatric patients who underwent LDST for suspected central adrenal insufficiency at a single tertiary center. Serum cortisol was measured at baseline and at 15, 30, 45, and 60 minutes following intravenous administration of 1 µg ACTH. Adrenal sufficiency was defined as a peak cortisol ≥ 18 µg/dL. Peak timing distribution, basal predictors, incremental diagnostic contribution of additional time points, number needed to test (NNT), and false-positive rates using fixed versus time-specific cut-offs were evaluated. Results: Peak cortisol occurred most frequently at 15 minutes (48.6%), followed by 30 minutes (28.8%), baseline (10.7%), 45 minutes (9.0%), and 60 minutes (2.8%). Termination of testing at 30 minutes would have misclassified 11.9% of patients as insufficient despite normal later responses. Extension to 45 minutes provided meaningful diagnostic improvement, whereas routine extension to 60 minutes yielded only marginal additional benefit (NNT = 30). Higher basal cortisol levels were independently associated with earlier peak responses (p = 0.021), while demographic and auxological factors showed no association. Application of time-specific, percentile-based cortisol thresholds reduced false-positive classifications nearly five-fold at 30 minutes compared with a uniform 18 µg/dL cut-off. Conclusions: LDST cortisol responses in children show substantial interindividual variability in peak timing. Extension of sampling to 45 minutes and use of time-specific interpretation thresholds significantly improve diagnostic accuracy and reduce false-positive AI diagnoses in pediatric practice.