Background: Maintenance immunosuppression of pediatric kidney transplant (PKT) recipients often includes corticosteroid medications. Prolonged corticosteroid treatment has been associated with adrenal insufficiency (AI); however, little is known about AI in the context of PKT. We report on the AI prevalence, risk factors, and potential adverse effects in a PKT recipient cohort.
Methods: We conducted a retrospective, single-center cohort review of 50 PKT recipients. Patients without a prior AI diagnosis were included if they had a minimum of one valid morning cortisol screening (MCS) or ACTH stimulation test performed. MCS was routinely used clinically to screen for AI. For MCS <240 nmol/L, diagnosis was confirmed by ACTH stimulation test. Baseline and contemporaneous characteristics were tested for univariate association with MCS levels and AI diagnosis, stratified as AI risk factors or potential adverse effects.
Results: The cohort included 54 transplants, with 33 males (61.1%) and median transplant age of 7.0 (IQR 3.8, 12.7) years. Thirteen (24.1%) were diagnosed with AI; however, only 28/54 had definitive testing.  Comparing only tested patients (Table 1), 13 (47%) received an AI+ diagnosis at mean age of 12.3 ± 5.3 years, and median 2.4 (IQR 1.4, 5.6) years post-transplant. In the AI+ group, steroid exposure-related risk factors included current prednisone dosage (p = 0.001), prior 6-month prednisone exposure (p = 0.02), daily (vs. not daily) prednisone administration (p = 0.002), and number of rejection episodes since transplant (p = 0.001).  Eighty-three MCS samples (mean 1.7 ± 0.9 samples per patient) were evaluated (Table 2) at a median age of 11.8 (IQR 7.3, 15.7) years old, and median-time post-transplant of 2.5 (IQR 1.1, 5.1) years. Mean cortisol was 246 ± 121 nmol/L and was significantly associated with steroid exposure-related risk factors, including current prednisone dosage (ρ = -0.40, p < 0.001), prior 6–month exposure (ρ = -0.34, p = 0.002), daily (vs. not daily) prednisone administration (p = 0.006), and number of rejection episodes since transplant (ρ = -0.33, p = 0.002). MCS was associated with potential adverse effects related to measures of medical acuity/complexity: number of medications (ρ = -0.37, p < 0.001), prior 6-month hospitalizations (ρ = -0.27, p = 0.01) and hospitalization days (ρ = -0.33, p = 0.002). MCS was also associated with measures of acute kidney injury risk (AKI): prior 6-month CV for serum creatinine (ρ = -0.30, p = 0.005) and AKI episodes with >50% increase from baseline creatinine (p = 0.004).  
Conclusion: Several measures of elevated steroid exposure were associated with both AI diagnosis and the level of morning cortisol. Potential AI-related adverse effects included greater hospitalization risk increased medical complexity and greater risk of AKI. These findings identify need for more systematic surveillance of AI-risk in pediatric transplant recipients.