Re: Inhaled corticosteroids for chronic obstructive pulmonary disease
In their review of inhaled corticosteroids (ICS) in patients with chronic obstructive pulmonary disease (COPD), Park and colleagues acknowledge that these agents “reduce the risk of exacerbations” when used as an adjunct to long-acting beta-agonists (LABAs) and cause “significant adverse effects.” In addition, Park et al. make reference to clinical practice guidelines as supporting the use of ICS therapy, though these fail to evaluate the benefits of ICS in COPD in the context of our evolving knowledge of their potential harms. To aid clinicians in making informed therapeutic decisions, we wish to clarify these absolute benefits and harms.
The systematic review of ICS therapy in COPD cited by Park and colleagues is from 2008 and is not the latest or most complete look at the evidence. The Cochrane review by Nannini et al. is up-to-date as of November 2011 and includes a more comprehensive overview of randomized-controlled trials (RCT). Park et al. state that based on this 2008 systematic review that six patients with COPD would need to be treated with ICS therapy for a year to prevent one exacerbation. This is a misconception resulting from incorrectly calculating the number needed to treat (NNT) based on the reciprocal of the rate ratio (the ratio of mean exacerbations per year in the ICS group versus the control group).
The appropriate derivation of the NNT involves using the reciprocal of the absolute difference in risk of experiencing the dichotomous outcome of interest (i.e. NNT= 100/ARD); in this case, the standard outcome is the rate of patients experiencing one or more exacerbations. The Cochrane review readily presents the NNT based on rates of patients experiencing one or more exacerbations with ICS with long-acting beta-agonist (LABA) versus LABA alone as 20. However, data for this outcome are not available for over two-thirds of the 11 794 RCT participants – including those from TORCH and TRISTAN, the two largest ICS trials – which seriously compromises the validity of this modest treatment effect.
Similarly, Park et al. incorrectly refer to a 10-20% reduction in the risk of exacerbations with ICS therapy added on to a LABA. The appropriate interpretation of these results, as supported by the Cochrane review, is that patients taking ICS plus LABA therapy have an annual rate of exacerbations that is lower than LABA therapy alone, and that this amounts to an average of 0.16 to 0.32 fewer exacerbations per patient per year. Because this is a difference in continuous data, a NNT cannot be directly derived from these results. Additionally, authors of the Cochrane review caution readers of the low quality of the evidence due to a high risk of bias in this outcome from included trials.
Furthermore, the effects of ICS therapy on day-to-day symptoms and COPD-related quality of life are statistically significant but of questionable clinical importance. Average improvements of 1.6 to 2.7 points in the St George’s Hospital Questionnaire compared to placebo are less than the 4-point change shown to be noticeable to patients.[2,4]
On the other hand, ICS therapy in COPD is associated with potentially important adverse effects that are consistent with their pharmacological profile. In a meta-analysis of RCTs, ICS therapy increased the risk of fractures (odds ratio 1.27, confidence interval 1.01 to 1.58). Applying this risk increase to the control group rate in the TORCH trial, the number needed to harm (NNH) to cause one additional patient to experience a clinical fracture with inhaled corticosteroids over 3 years is 83. Furthermore, inhaled corticosteroids increase the risk of pneumonia, though the true effect of this is currently unclear. Overall, ICS therapy neither increases nor reduces the number of all-cause or respiratory-related hospitalizations in COPD. Finally, observational evidence is surfacing that ICS therapy may increase the risk of diabetes and worsen existing diabetes.
Because of inconsistencies in the reporting of clinical trials, it is not currently possible to directly compare the reductions in exacerbations to the increases in fractures and cases of diabetes over a similar time period. In our opinion, the net effects on quality of life, morbidity and mortality based on the published evidence hardly justify the cost ($150 for one month supply in British Columbia, Canada) of this intervention.
1. Park HY, Man SF, Sin DD. Inhaled corticosteroids for chronic obstructive pulmonary disease. BMJ. 2012;345:e6843.
2. Sobieraj DM, White CM, Coleman CI. Benefits and risks of adjunctive inhaled corticosteroids in chronic obstructive pulmonary disease: a meta-analysis. Clin Ther. 2008;30:1416-25.
3. Nannini LJ, Lasserson TJ, Poole P. Combined corticosteroid and long-acting beta2-agonist in one inhaler versus long-acting beta2-agonists for chronic obstructive pulmonary disease. Cochrane Database of Systematic Reviews 2012, Issue 9. Art. No.: CD006829.DOI: 10.1002/14651858.CD006829.pub2.
4. Jones PW. Health status measurement in chronic obstructive pulmonary disease. Thorax. 2001;56:880-7.
5. Loke YK, Cavallazzi R, Singh S. Risk of fractures with inhaled corticosteroids in COPD: a systematic review and meta-analysis of randomised controlled trials and observational studies. Thorax. 2011;66:699-708.
6. Suissa S, Kezouh A, Ernst P. Inhaled corticosteroids and the risks of diabetes onset and progression. Am J Med. 2010;123:1001-6.
Competing interests: No competing interests