ESPE Abstracts

Purpose: We try to analyze the prediction capacity of variable factors to diagnose adrenal insufficiency.

Methods: We analyzed clinical and laboratory data of 22 children (Male =13) who have been checked regular dose ACTH stimulation test for suspected symptoms after HSCT (Lymphoid leukemia=5, Myeloid leukemia=9, Non-malignant=8) at the Catholic HSCT center from Feb 2013 to Feb 2017 at Seoul St. Mary’s Hospital. A normal response of ACTH stimulation test was defined as a stimulated serum cortisol >18.1 mcg/dl and an increment from baseline of at least 7.2 mcg/dl was considered a partial response.

Result: The suspected symptoms of AI that patients had before ACTH test were as follows: Poor feeding=8, Lethargy or sleeping tendency=7, Nausea or vomiting=6, Fever=6, Elevated ACTH or Low cortisol=2, Facial flusing=2, Hyperkalemia=1, Diaphoresis=1, Fatique=1, Hyponatremia=1, Hypoglycemia=1, Edema=1, Epigastric pain=1, Dizziness=1. Sixteen (72.7%) out of 22 patients underwent ACTH test was diagnosed as AI, 3 patients were partial AI and 3 patients were normal. In AI group, serum cortisol levels after 60 minutes (3.82±0.68 vs. 16.14±3.00, P<0.008) was lower and serum K level (4.34±0.75 vs. 3.68±0.20, P<0.018) before ACTH stimulation test was higher than normal or partial response group. There were no significant differences age at HSCT, Age at ACTH test, serum FBS, Na level before ACTH test between AI and normal or partial response group. In a univariate logistic regression analysis, serum K level before ACTH test (OR=317.5; 95% CI, 1.1–191705.1, P < 0.042) and proportion of steroid users at ACTH stimulation (OR=15.0; 95% CI, 1.325–169.87, P < 0.029) were higher was associated with AI.

Conclusion: Serum potassium level and current steroid use could be potential factors for early detection of adrenal insufficiency. These results emphasize the need to pay attention to the small signal including serum normal upper limit K level of children after HSCT for early detecting AI.