Risk-benefit profile of inhaled corticosteroids in COPD is still unclearBMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e8360 (Published 20 December 2012) Cite this as: BMJ 2012;345:e8360
- 1Lower Mainland Pharmacy Services, Vancouver, BC, Canada V5Z 1M9
- 2Therapeutics Initiative (UBC), Vancouver, BC, Canada
Park and colleagues’ review acknowledges the harms associated with using inhaled corticosteroids for chronic obstructive pulmonary disease (COPD).1 To help clinicians make informed decisions, we wish to clarify the absolute benefits and harms.
Estimates of benefit were less optimistic in an up-to-date and comprehensive Cochrane review,2 partly because of Park and colleagues’ unconventional derivation of the number needed to treat (NNT) using rate ratios. From the Cochrane review, the NNT for inhaled corticosteroids based on rates of patients experiencing one or more exacerbations when added to a long acting β agonist is 20. Cochrane authors caution that this estimate is at high risk of bias because no data were available for more than two thirds of randomised controlled trial participants, which severely compromises the validity of this estimate of modest benefit.
Furthermore, the effects of inhaled corticosteroids on symptoms and quality of life (QOL) are statistically, but not clinically, significant. Average improvements of about 2 points on the QOL questionnaire compared with placebo are less than the 4 point change that is noticeable to patients.2 3
Moreover, inhaled corticosteroids are associated with serious adverse effects. In a meta-analysis,4 these drugs increased the risk of fractures. Applying the odds ratio of 1.27 from this study to TORCH results in a number needed to harm of 83 over three years. Inhaled corticosteroids also increased the risk of pneumonia,2 leading to no net effect of treatment on respiratory related hospital admissions for COPD.2 Finally, observational evidence shows a possible increased risk of diabetes.5
Because of inconsistencies in trial reporting, the impact of harms and benefits resulting from inhaled corticosteroids cannot be directly compared. In our opinion, the net effects hardly justify the cost of this widespread intervention.
Cite this as: BMJ 2012;345:e8360
Competing interests: None declared.