Oral, inhaled corticosteroids associated with risk for adrenal insufficiency

Cumulative exposure to glucocorticoids appeared associated with risk for adrenal insufficiency among patients with severe asthma, according to a study published in The Journal of Allergy and Clinical Immunology: In Practice.

Researchers observed the highest risk with both high-dose inhaled corticosteroid (ICS) therapy and intermittent oral corticosteroid (OCS) use.

These results should prompt additional consideration for doctors before prescribing glucocorticoids and educate patients about adrenal insufficiency (AI) risks, Vincent Brennan, MB BCh BAO, MRCPI, doctoral student and research assistant in the department of medicine at Royal College of Surgeons in Dublin, and colleagues wrote.

Previous studies have found associations between intermittent OCS use and regular ICS use and with tertiary AI, the researchers wrote, which is the most common cause of AI.

The risks of AI include the adverse cardiometabolic features of iatrogenic Cushing’s syndrome and increased cardiovascular morbidity and mortality, whereas tertiary AI can cause potentially life-threatening adrenal crises.

The researchers collected data prospectively as part of the 32-week INCA-SUN randomized control trial of 80 patients (58% female; mean age, 48.8 years ± 14 years) with severe asthma. Each patient was prescribed two daily doses of inhaled salmeterol/fluticasone propionate (FP; Seretide 250 or 500 Diskus, GlaxoSmithKline) taken via an electronic inhaler compliance assessment device that recorded adherence, technique and drug delivery.

Also, researchers recorded patient symptoms and quality-of -life parameters at each visit via the asthma control test (ACT), the mini-Asthma Quality of Life Questionnaire and the EQ-5D-3L quality-of-life questionnaire with visual analogue scale (EQ-VAS).

At 32 weeks, the researchers analyzed serum samples because morning serum cortisol concentration is considered an effective screening tool for AI that can reduce the need for short synacthen testing (SST) by 40%.

The 20 patients (25%) with a morning cortisol of less than 130 nmol/L, whom the researchers predicted to fail an SST, were classified at the highest risk for AI (HR-AI). The 21 patients (26%) whose morning cortisol readings were greater than 294 nmol/L were considered likely to pass an SST and classified at the lowest risk for AI (LR-AI). The 39 patients (49%) with morning cortisol concentrations in between these ranges were classified as indeterminate risk for AI (IR-AI).

The HR-AI patients had a greater cumulative OCS exposure, at 470 mg (interquartile range [IQR], 173-1,965) of prednisolone, compared with the LR-AI patients, who had a total exposure of 62.5 mg (IQR, 0-210) of prednisolone (P = .0017).

Also, ten patients receiving maintenance OCS as well as ICS had significantly higher cumulative OCS exposure (2,240 mg prednisolone vs. 187.5 mg; P < .00005) and significantly lower morning cortisol (mean, 92 nmol/l vs. 238 nmol/l; P = .0002) and a greater proportion were HR-AI (60% vs. 19%) compared with patients not on maintenance.

Researchers observed a negative correlation between morning serum cortisol concentration and cumulative OCS exposure in mg of prednisolone (Spearman’s correlation coefficient [r] = –0.4; P = .0002) and weight-based OCS exposure in mg/kg (r = –0.41; P .0002).

Also, the researchers found that the risk for AI increased with each OCS course (OR = 1.41; 95% CI, 0.97-2.05) based on a standard 40 mg/day course of prednisolone for 5 days. However, this risk for AI with a pulse course of OCS therapy lost significance when adjusted for ICS exposure (OR = 1.29; 95% CI, 0.86-1.94).

Compared with the LR-AI group, the patients in the HR-AI group had significantly higher cumulative ICS exposure (212.5 mg FP vs. 163 mg FP; P = .003). High-dose FP prescription increased the risk for AI in a multivariate logistic regression analysis, independent of OCS exposure (OR per 1 mg/kg increase in FP exposure = 2.17; 95% CI, 1.06-4.42).

The HR-AI patients also had the greatest median total cumulative glucocorticoid exposure including both OCS and ICS over the 32-week study period at 1,866 mg of a prednisolone equivalent (IQR, 1,358-3,345). The IR-AI group had a total of 1,354 mg (IQR, 738-1,737) and the LR-AI group had a total of 997 mg (IQR, 863-1,559).

The researchers additionally found an association between each unit increase in ACT score and a 10% decrease in odds for AI (OR = 0.9 per unit change in ACT; 95% CI, 0.8-1). Whereas the median final ACT score was 20 (IQR, 16-22) for the IR-AI group and 20 (IQR, 15-23) for the LR-AI group, the HR-AI group had a lower median final ACT score of 17 (IQR, 12-21).

There was a positive correlation between serum cortisol concentration and mini-AQLQ scores (r = 0.31; P = .0055) but no significant difference between the HR-AI group and the LR-AI and IR-AI groups, individually or combined, in EQ-5D-3L and EQ-VAS scores.

According to the researchers, this study was the first to use an electronic monitoring device to collect individual cumulative ICS exposure data, and the cumulative OCS exposure data allowed for a more nuanced analysis while avoiding selection bias.

The researchers also expressed concern about the high prevalence of undiagnosed AI and the risks it poses these patients. Identifying these at-risk patients would be key to reducing these risks, the researchers wrote.

Further larger studies are necessary, the researchers wrote, into the relationship between patient-reported symptoms, high-dose ICS and OCS prescriptions and resulting adrenal suppression.