Should we avoid β-agonists for moderate and severe chronic obstructive pulmonary disease?

β-Agonists for obstructive lung diseases

Controversy has surrounded β-agonists since their introduction more than 50 years ago.1 Short-acting β-agonists became widely used for management of asthma and COPD in the 1960s without good scientific data on long-term efficacy and safety. In the 1990s, the long-acting β-agonist salmeterol was introduced, despite evidence that it might be associated with greater risk of respiratory death than short-acting agents were.2 After the US Food and Drug Administration received postmarketing reports of several asthma-related deaths associated with salmeterol, the Salmeterol Multicenter Asthma Research Trial was conducted; the trial found a twofold increase in life-threatening asthma exacerbations and a fourfold increase in asthma deaths compared with placebo.3 Recently, some have asked whether long-acting β-agonists should be taken off the market.4

β-Agonists worsen control of obstructive lung diseases through a negative feedback mechanism that is an adaptive response of the β-adrenergic system.5 Stimulation results in uncoupling and internalization of receptors, known as desensitization, followed by a decrease in receptor density and receptor gene expression, known as down-regulation.5 This tolerance to βagonists could explain the counterintuitive results of trials that demonstrate adverse respiratory effects.

Cardiovascular outcomes

β-Blockers reduce morbidity and mortality among patients with cardiovascular disease or other risk factors. β-Agonists exert physiologic effects opposite to those of β-blockers and might be expected to have deleterious cardiovascular effects. Case-control studies demonstrate an association between β-agonist use and increased risk of myocardial infarction, congestive heart failure, cardiac arrest, and acute cardiac death, with odds ratios ranging from 1.3 to 3.4.6

A meta-analysis pooled results from 33 randomized placebo-controlled trials of patients with obstructive lung disease and found that a single dose of β-agonist increased heart rate by 9 beats/min and reduced potassium concentration by 0.4 mmol/L compared with placebo.7 For trials that lasted from 3 days to 1 year, β-agonist treatment increased the risk of cardiovascular events more than twofold compared with placebo. Adverse events included sinus and ventricular tachycardia, syncope, atrial fibrillation, congestive heart failure, myocardial infarction, cardiac arrest, and sudden death.

Respiratory outcomes

β-Agonists and anticholinergics are generally considered to be equivalent choices for treatment of COPD. Many trials of these agents have concentrated on short-term end points, such as airflow or symptoms. Anticholinergics have been shown to have equal or superior efficacy to β-agonists in improving lung-function parameters.8 Surveys show, however, that prescriptions for β-agonists are 2 to 10 times more common than prescriptions for anticholinergics.9 Anticholinergics have been shown to reduce severe COPD exacerbations by 40% (P < .001), hospitalizations by 30% (P = .001), and respiratory deaths by 70% (P = .02) compared with placebo, without tolerance to their effects over time10; significant tolerance develops to the effects of β-agonists in COPD.11

Two meta-analyses9,10 pooled the results of randomized placebo-controlled trials of β-agonists or anticholinergics for COPD published through December 2005 and showed that β-agonist use increased respiratory deaths more than twofold compared with placebo, without significantly affecting hospitalization or total mortality. Approximately 50% of participants were also taking inhaled corticosteroids. When directly compared with other treatments, β-agonists caused a twofold increase in COPD hospitalizations and a fivefold increase in total mortality compared with anticholinergics, and a twofold increase in total mortality compared with inhaled corticosteroids.9,10 Adding of β-agonists to anticholinergics or inhaled corticosteroids had no beneficial effect on any long-term clinical outcomes.9

Conflicts of interest

Why has it not been clearer to clinicians that β-agonists have such adverse effects on people with obstructive lung diseases? Conflicts of interest can arise from pharmaceutical company sponsorship of trials, as has been found with other medications. A systematic review of β-agonist trials found that 75% of industry-sponsored trials but only 10% of trials without industry support reported that β-agonists were beneficial.12 The P value for interaction was <.00001, indicating a very strong association between pharmaceutical company funding and reporting beneficial results. Most of the trials longer than 3 months were industry sponsored. It is essential to evaluate hard clinical outcomes, such as hospitalizations or deaths.