Endocrine Abstracts

Background: Adrenal insufficiency (AI) is a life-threatening condition if left unmanaged. Despite treatment patients can expect a life expectancy that is shortened by 12 years secondary to probable inherent over-replacement associated with oral glucocorticoid regimens. Thrice-daily hydrocortisone is the most common regimen used. Very low-dose prednisolone (2-4 mg) is an alternative with lower uptake due to the absence of evidence for its use. This study fills this literature gap.

Methods: This is a cross-sectional, observational study that recruited 20 healthy volunteers (HV), 20 AI patients on prednisolone, 20 AI patients on hydrocortisone and 9 patients on anti-inflammatory doses of steroids (mainly IV methyl-prednisolone) for other medical conditions. During stereotyped study visits, subjects provided anthropometric data, blood samples, urine samples and SF-36 data. This was used to assess bone health, cardiovascular risk, diabetic risk, immune cell profiles and subjective health between groups.

Results: There was no significant difference between groups in osteocalcin, P1NP or urinary NTX. Parathyroid hormone was significantly elevated in the hydrocortisone group compared to HV at 8.2(±3.2) vs 6.1(±2.3) pmol/L (P=0.04). Calcium and phosphate levels were in the reference range. The waist:hip ratio was significantly lower in HV at 0.83(±0.07) compared to the cohort on hydrocortisone and prednisolone, at 0.90(±0.09) and 0.90(±0.07) respectively (P=0.009). High sensitivity-CRP was 1.6(2.1) mg/L compared to 0.8(1.1) mg/L (P=0.008) in the hydrocortisone and HV groups respectively. Potassium levels were significantly decreased in the hydrocortisone and high dose groups; 4.0(0.4) mmol/L in both compared to 4.2(0.2)mmol/L in HV (P=0.005). Glycaemic markers in both AI groups demonstrated significantly lower fructosamine levels compared to healthy volunteers (P=0.0008). The hydrocortisone group had significantly elevated insulin, c-peptide and HOMA-%β compared to HV (P=0.039, P=0.025, P=0.025 respectively). There was no difference in HbA1C levels between groups. Although infection rates were comparable between groups, all steroid groups had significantly elevated neutrophils compared to HV (P<0.0001). Monocyte count was significantly elevated in the hydrocortisone cohort at 0.6(0.2) x10⁹/L vs 0.4(0.2) x10⁹/L in HV. Flow cytometry showed significant HLA-DR suppression in the monocytes in the high dose group. SF36 data showed impaired energy/fatigue scores in the prednisolone group compared to the healthy volunteers (77.3±(10.2) compared to 58.2(±21.7); P=0.021)

Conclusion: The data suggests that hydrocortisone causes greater steroid exposure as assessed by the markers in this study, when compared to healthy volunteers. There are no overt differences between hydrocortisone or prednisolone cohorts. Both medications should be used interchangeably in the treatment of AI.