New Cortisol Thresholds for the Diagnosis of Adrenal Insufficiency Using the Low-Dose Synacthen Test in Children on Long-Term Corticosteroids: A North African Pilot Study

Background: The low-dose Synacthen stimulation test (LD-SST) is the reference dynamic test for diagnosing glucocorticoid-induced adrenal insufficiency (AI) in children, but it is resource-intensive and costly. This study aimed to establish morning cortisol thresholds predictive of hypothalamic–pituitary–adrenal (HPA) axis response to LD-SST in pediatric patients receiving chronic corticosteroid therapy. Methods: We conducted a prospective study including 71 children (mean age 6.23 ± 3.49 years; 57.7% male) receiving prolonged oral or inhaled corticosteroids. All patients underwent LD-SST with cortisol measurements at 0, 30, and 60 min. Patients were classified as having AI (peak cortisol < 18 μg/dL at T30 or T60, n = 39) or normal adrenal function (peak cortisol ≥ 18 μg/dL, n = 32). ROC curve analysis determined optimal cortisol thresholds for predicting AI. Results: Among the 71 patients, 44 received inhaled corticosteroids (62%) and 27 oral corticosteroids (38%). Asthma (61.9%) and autoimmune diseases (18.3%) were the main indications. The prevalence of AI was 54.9% (n = 39). Mean morning cortisol was significantly lower in the AI group compared to the normal group (6.69 ± 1.99 vs. 9.21 ± 2.49 μg/dL, p < 0.001). ROC analysis identified a morning cortisol <6 μg/dL as predictive of AI with 96.9% sensitivity and 46.2% specificity (AUC = 0.823), while a threshold >13 μg/dL predicted normal HPA function with 100% specificity. Using these thresholds would have avoided LD-SST in 29.6% of patients. The cortisol increment following stimulation also demonstrated diagnostic value (AUC = 0.822), with an increment <9 μg/dL suggesting AI and ≥9 μg/dL likely excluding the diagnosis. Conclusions: In this North African pilot study, morning cortisol measurement at 7 days after oral corticosteroid withdrawal predicted LD-SST response. A threshold of ≤6 μg/dL identified patients with persistent HPA suppression with high sensitivity (97.4%), while ≥13 μg/dL identified patients with preserved function with high specificity (100%). However, the 7-day washout period might be insufficient for complete HPA recovery; therefore, these thresholds reflect residual suppression rather than permanent adrenal insufficiency.