Introduction: The low-dose synacthen test (LDT) is widely used to assess central adrenal insufficiency (CAI); however, the total serum cortisol (C) cut-off value is controversial. A correct diagnosis of CAI is required, but overdiagnosis may lead to unnecessary hormone replacement therapy. Salivary cortisol (SC) reflects the levels of free serum cortisol and is a noninvasive alternative.
Objective: To define a new cut-off value of serum cortisol in pediatric patients evaluated for suspected CAI considering SC of normal responders.
Patients and methods: 145 pediatric patients (88 males) with suspected of CAI were included in the study. Mean age (SD) was 11.3 years (4.88). All patients underwent LDT with intravenous injection of 1 μg/m2 of tetracosactide (Synacthen). Serum C and SC levels were measured at baseline and after 30 and 60 minutes. The highest value of the tested parameters, at either 30 or 60 minutes, was regarded as the maximum response value. Reference cut-off value ≥18 μg/dl of serum cortisol levels was considered as a sufficient response (CAS).
Results: A significant positive correlation between maximum C and SC response was found (r=0.90, P=0.001). Patients were divided according to serum cortisol response into the following groups (Gr): CASGr: n=72, (median (interquartile range) C and SC, 21.3 μg/dl (19.835.2) and 1.14 μg/dl (0.771.58), respectively, and CAIGr: n=73 (median (interquartile range) C and SC, 14.5 μg/dl (12.216.4) and 0.58 μg/dl (0.361.04), respectively. ROC curve analysis established a SC cut off level of <0.61 μg/dl for CAI diagnosis (specificity and sensitivity of 84% and 56.3%, respectively). Considering the lower quartile SC of CASGr (SC ≥0.77 μg/dl), an intermediate (I) Gr (ICAIGr) was established within the CAIGr. ICAIGr: n=28/73 (median (interquartile range) C and CS, 16.35 μg/dl (14.2516.87) and 1.16 μg/dl (0.881.29), respectively. The remaining 45 patients were considered real (R) CAI, median (interquartile range) C and SC 13.2 μg/dl (11.315.3) and 0.41 μg/dl (0.320.55), respectively. Significant differences in maximum serum C level responses were found among CASGr, ICAIGr, and RCAIGr (P<0.001).
Conclusion: A maximum serum C response of <16.35 and SC response of <0.77 μg/dl may be appropriate cut-off values to define RCAI. Recognition of an ICAIGr allows avoiding unnecessary hormone replacement therapy; however, rigorous patient follow-up is required. Finally, the combined evaluation of maximum serum C and SC level responses improves the accuracy of CAI diagnosis in children.
27 - 29 Sep 2018
European Society for Paediatric Endocrinology