Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Inhaled steroids for COPD

Review
Results
Exacerbations
Death
FEV1
Adverse events
Comment

COPD is a big problem, and one that is getting bigger. This is partly because of ageing populations, but is also a reflection of past industry and smoking. The WHO predicts that COPD, by 2020, will be the fifth most prevalent disease and the third most common cause of death.

Inhaled steroids are used to combat airways inflammation. Meta analysis has shown that, in patients with clearly defined moderately severe COPD, relatively high daily doses of inhaled corticosteroids improve FEV1 over two years of treatment [1]. But the issue is not just one of ventilatory outcomes, but of important clinical outcomes, like exacerbations of COPD, or hospital admissions, or death, as well as adverse events. A new systematic review [2] adds a little to this story.

Review

Five electronic databases, including the Cochrane Library, were searched for placebo-controlled trials of inhaled corticosteroids for at least six months in COPD. Experts were also contacted to identify any other studies. The primary outcome sought was the frequency (or risk) of respiratory exacerbations. The definition of exacerbation was that used by the trials, and the frequency of exacerbations was calculated per patient-month of treatment. Other outcomes were the rate of decline in FEV1 and all-cause mortality.

Results

Nine randomised, placebo-controlled studies were found. All those reporting exacerbations were of sufficient quality so as to minimise bias. They differed greatly in size, in duration and in definition of exacerbation. Size varied from 26 to 1277 patients, with 3,926 in total. Duration varied from six to 40 months, with three trials lasting six months and the remainder at least two years.

Exacerbations were defined and reported in six trials. The most common definition used was worsening of symptoms, usually with changing treatment and including use of oral steroids and/or antibiotics. One used cough and phlegm more than usual, and for another hospital admission for a respiratory condition.

The treatment most commonly used was inhaled budesonide at 800-1,600 μg/day (five trials), with fluticasone 1000 μg/day in two, and triamcinalone 1,200 μg/day and beclamethasone 1,500 μg/day each used in one trial. Control in all cases was usual care with placebo.

The average age of patients in the trials was 54-66 years, with variable percentages of current smokers, and with baseline FEV1 of about 1 to 2.5 litres.

Exacerbations

In six trials, exacerbations occurred at an average rate of 0.07 per month (0.8 a year) with placebo, with a reduced rate of 0.05 per month (0.6 a year) with inhaled corticosteroid (Figure 1). The relative risk for exacerbation was 0.68 (0.64 to 0.72). On an annual basis, the number needed to treat was 4.8 (4.0 to 5.9). This means that treating five patients with COPD with inhaled corticosteroids will prevent one exacerbation.

Figure 1: Exacerbations with inhaled steroid and placebo

Death

Mortality in six trials averaged 4.1% with placebo, and 3.4% with inhaled corticosteroids (Figure 2). The difference was not statistically significant, with a relative risk of 0.84 (0.6 to 1.2).

Figure 2: Mortality with inhaled steroid and placebo

FEV1

The rate of decline in FEV1 with placebo was very variable, with average reductions in trials of 12 to 180 mL. Over the nine trials, the mean difference with inhaled corticosteroid was to effect a weighted average reduction of 28 mL in the decline of FEV1.

Adverse events

These were not reported in detail, and absolute rates were not given. We are told that the frequency of oropharyngeal candidiasis was increased with inhaled corticosteroids, with a relative risk of 2.1 (1.5 to 3.1), and that a similar increase in risk was seen for skin bruising, with a relative risk of 2.1 (1.6 to 2.8).

Comment

Interesting stuff, and generally well done. The review highlights some important issues, though. First was that the definition of those things that affect patients healthcare providers, exacerbations in respiratory symptoms, were neither well nor consistently defined. Perhaps we need two definitions, relating to worsening in condition, and another more serious outcome of hospital admission because of worsening respiratory function.

Frustratingly little detail was given about adverse events like candidiasis. Knowing that it occurs twice as often with inhaled corticosteroids is much less useful than knowing how often it occurs. Recent attempts to elicit a number from knowledgeable folk has elicited responses ranging from rare to 15%.

What we do have here is a splendid example of systematic reviews telling us something about a treatment (though not nearly enough in terms of the balance of benefit and harm). More importantly, it helps to define the research agenda. The review almost tells us how to do better and more informative studies in future. Let's hope that someone takes notice.

References:

  1. PM van Grunsven et al. Long term effects of inhaled corticosteroids in chronic obstructive pulmonary disease: a meta-analysis. Thorax 1999 54: 7-14.
  2. A Alsaeedi et al. The effects of inhaled corticosteroids in chronic obstructive pulmonary disease: a systematic review of randomized placebo-controlled trials. American Journal of Medicine 2002 113: 59-65.

previous or next story