I read with interest the recent article by Brus1 on the effects of high-dose inhaled corticosteroids on 24 hour AUC (AUC24) plasma cortisol profiles in healthy adults. A within treatment comparison for placebo at baseline vs steady state showed significant suppression for all inhaled corticosteroids except for flunisolide. This is potentially misleading because flunisolide had the lowest placebo baseline value for AUC24 cortisol, and consequently, it is perhaps not surprising that the percentage suppression would be of a smaller magnitude. Indeed, performing statistical comparisons on placebo values between the 5 treatment groups (using unpaired Student t test) reveals a lower placebo value for flunisolide vs fluticasone proprioate (P<.01), vs budesonide (P<.05), and vs triamcinolone acetonide (.05 <P< .10). Because of the confounding effects of the different placebo baseline values, it is invalid to perform between-treatment comparisons for the percentage suppression of AUC24 cortisol calculated as percentage change from baseline. Indeed, when using a pooled value as the mean placebo value for all 5 treatment groups (ie, pooled value of 1225 ng · mL−1 · h−1), a different picture emerges in terms of the percentage of steady-state suppression: budesonide, 13%; flunisolide, 17%; triamcinolone, 23%; beclomethasone dipropionate, 36%; and fluticasone, 78%.
Lipworth BJ. Relative Effects of Exogenous Inhaled Corticosteroids on Diurnal Cortisol Secretion. Arch Intern Med. 2000;160(16):2546. doi: