Am Fam Physician. 2008;78(1):21-22
Author disclosure: Nothing to disclose.
Each month, three presenters review an interesting journal article in a conversational manner. These articles involve “hot topics” that affect family physicians or “bust” commonly held medical myths. The presenters give their opinions about the clinical value of the studies discussed. The opinions reflect the views of the presenters, not those of AFP or the AAFP.
This Month's Article
American Lung Association Asthma Clinical Research Centers, Peters SP, Anthonisen N, et al. Randomized comparison of strategies for reducing treatment in mild persistent asthma [published correction appears in N Engl J Med. 2007;357(7):728]. N Engl J Med. 2007;356(20):2027–2039.
Can patients with mild persistent asthma safely decrease or eliminate inhaled steroids and still maintain adequate asthma control?
Andrea: Inhaled steroids remain the standard of care for patients with mild persistent asthma,1 yet there are concerns about the safety of long-term use, particularly in children. Additionally, many patients with mild asthma only use their controller medications intermittently. Is it possible to safely decrease the use of inhaled steroids, or change the inhaled steroids to a nonsteroidal controller medication, and still maintain adequate asthma control?
What does this article say?
Andrea: This was a 16-week randomized, double-blind trial of fluticasone (Flovent) 100 mcg twice daily (standard care) versus montelukast (Singulair) 5 or 10 mg once daily (based on age) versus fluticasone/salmeterol (Advair) 100 mcg/50 mcg once daily. Patients had to be at least six years of age and have moderate persistent asthma.
There were 1,309 patients assessed for eligibility; 787 were enrolled during a four- to six-week run-in period. During the run-in period, all patients received fluticasone 100 mcg twice daily. Inclusion criteria for randomization after the run-in period was 75 percent compliance with medication, attainment of mild persistent asthma classification, no urgent treatment or escalation of asthma treatment during the run-in period, and no fever within 24 hours of randomization. Only 500 patients met inclusion criteria.
Treatment failure was defined as hospitalization or urgent visit related to asthma, use of systemic corticosteroids, open-label introduction of additional inhaled corticosteroids, use of 10 or more puffs of albuterol over 48 hours, FEV1 of at least 20 percent below baseline, or peak expiratory flow rate of less than 35 percent of baseline.
Treatment failure occurred in 20.2 percent of patients in the fluticasone group, 20.4 percent in the fluticasone/salmeterol group, and 30.3 percent in the montelukast group. The most common reason was a decrease in FEV1 of at least 20 percent below baseline (47.8 percent of events). There was no difference between groups in outcomes that matter to our patients (e.g., hospitalization, introduction of systemic steroids). The authors concluded it was safe and effective to transition patients with well-controlled, mild persistent asthma to once-daily fluticasone/salmeterol.
Should we believe this study?
Andrea: I don't think these results are generalizable to most patients with mild persistent asthma. The authors enrolled patients with moderate persistent asthma, then performed a run-in period before randomization to assess for compliance. Run-in periods for compliance jeopardize the applicability of the results to an average patient population. Physicians are not able to reliably determine which patients will be compliant with a medication regimen, so it's not possible to separate out the subset of your patients that could potentially benefit from the intervention (once-daily fluticasone/salmeterol) as suggested in this trial.2 The other problem with this run-in period is that only patients who were responsive to steroids were included in the final study, which stacks the deck in favor of steroids (and against montelukast).
To make this point in another way: what if the authors had included all of the eligible patients in the treatment arms (rather than just a highly compliant, steroid-responsive subset)? Would there have been any differences among any of the three medication regimens?
Mark: It also wasn't a true step down of inhaled steroid use. The patients were moved from twice-daily fluticasone to once-daily fluticasone plus salmeterol. It would have been more helpful if they had stepped down from twice-daily fluticasone to once-daily fluticasone. This is especially important given the FDA Public Health Advisory on long-acting beta2 agonists. It specifically recommends against the addition of these agents in patients whose asthma is adequately controlled on low to moderate doses of inhaled steroids because of the increased risk of severe asthma episodes and death.3 Low to moderate dosing of fluticasone is 88 to 440 mcg daily in adults.1
Bob: I'm concerned about the way the authors reported their treatment failure rates. The failure rate was 20 percent for twice-daily fluticasone and once-daily fluticasone/salmeterol, and 30 percent for montelukast. However, the authors state that the failure rate was approximately 60 percent higher in the montelukast group compared with the other two groups. This is a classic example of trying to inflate a drug or intervention's benefit by publicizing the relative risk reduction instead of the absolute risk reduction. We want to know absolute risk reduction. In this case, it was 30 to 20 percent, which is a 10 percent absolute risk reduction (number needed to treat [NNT] = 10). Relative risk reduction is statistical shenanigans; it expresses the relative change from a baseline. In this case, the authors use the montelukast failure rate of 30 percent as a baseline, and consider the 20 percent failure rate with once-daily fluticasone/salmeterol a 60 percent improvement. Tricky, huh? Yet, relative risk reduction is commonly used by sales representatives and authors who are trying to sell you something. The physician should ignore all values expressed as relative risk reduction—it's an exaggeration of the facts. Instead, physicians should look for absolute risk data.
What should the family physician do?
Andre: This study reminds me to consider stepping my patients down to the lowest effective dose of controller medication, but it doesn't provide much guidance about the best medication to use when making the switch. It also emphasizes that treatment failure with any medication for persistent asthma is fairly common, which is why adequate follow-up for patients with well-controlled asthma should occur at least every six months.
Mark: Remember to classify and treat patients with asthma according to the 2007 National Asthma Education and Prevention Program guidelines.1 Inhaled steroids are still the mainstay of treatment for persistent asthma. Further, I'd hesitate to add a long-acting beta2 agonist to a regimen just to decrease inhaled steroid use because (1) there is a broad dosing range for low to moderate steroid use; (2) leukotriene inhibitors are available for patients who do not tolerate inhaled steroids; and (3) there is concern about the safety of long-acting beta2 agonists.
Bob: I agree with those points. Always ask whether you are being given the relative risk reduction or absolute risk reduction. What you want is the NNT, the number needed to harm, and the magnitude of the benefit. Anything less does not help you make an educated decision.
Use the 2007 National Asthma Education and Prevention Program guidelines to appropriately classify and treat patients with asthma to improve asthma outcomes.
Avoid adding long-acting beta2 agonists in patients with mild or moderate asthma unless they are failing treatment with low to moderate doses of inhaled steroids.
Inhaled steroids are the preferred anti-inflammatory in asthma. Leukotriene inhibitors are an option for those patients who are unable to comply with inhaled steroid use.
Run-in periods to assess compliance and ensure treatment responsiveness create a bias in favor of the treatment in question, and yield data on a patient population that is not going to behave like your patients. Thus, the results of these studies may not be generalizable to the average family physician's practice.
The number needed to treat, number needed to harm, and the magnitude of the benefit are critical information if you are going to make an educated decision about treatment options.
Absolute risk reduction quantifies the actual difference between two outcomes; this is what the physician should be interested in. Relative risk reduction demonstrates the change in outcome relative to a baseline or control; it will often exaggerate a benefit.