Suppression of the hypothalamus-pituitary-adrenal axis in asthmatics on inhaled steroids treated with itraconazole

Itraconazole increases glucocorticoid levels through inhibition of CYP3A4 in the liver. Recent case-reports have shown that the co-administration of itraconazole with inhaled glucocorticoids in asthmatics may result in suppression of their HPA axis and a Cushing’s phenotype. We have evaluated the impact of this interaction on the cortisol axis in a double-blind randomised placebo-controlled study. Twenty-seven patients with severe asthma (17 women; mean±S.D. age, 54.2±15.6 years) taking inhaled steroids were recruited and randomized to receive oral itraconazole 200 mg bd or placebo. Plasma cortisol and ACTH were measured at baseline, 4 and 28 weeks on treatment, and 16 weeks after stopping treatment. The study was approved by the Local Ethics Committee. There was no difference between groups in baseline cortisol (median 183 nmol/l, range 92–393 vs 251 nmol/l, 75–770; P=0.42). After 4 weeks, median cortisol was 69.5 nmol/l (0–642) and 240 (155–1026) respectively (P=0.015), with 6/13 itraconazole-treated patients having undetectable plasma cortisol (<50). ACTH was 8.4±9 ng/l at baseline and 4.5±5.8 after 4 weeks (P<0.05), but was unchanged in the placebo group. Cortisol remained significantly lower in the active group after 28 weeks (135.7±125.3 vs 255.1±61.76, P=0.025). Sixteen weeks after stopping treatment plasma cortisol was not different between the active (383 nmol/l, 109–907) and the placebo group (285.5 nmol/l, 107–539; P=0.43) with recovery in all suppressed patients remaining in the study. Two patients in the active group withdrew from the study with symptoms suggestive of increased glucocorticoid exposure. No patients in the placebo group had an undetectable cortisol at any study time-point. We have shown that treatment with itraconazole and inhaled steroids commonly causes suppression of cortisol secretion within 4 weeks of treatment, with clinical evidence of Cushing’s syndrome in some. While full recovery of the HPA axis was documented 4 months following treatment withdrawal, this may not be the case with more prolonged treatment.