Asthma: fatal exacerbations with long-acting beta2 agonist therapy without concomitant steroid therapy.

(1) In France, two long-acting inhaled beta2 agonists (salmeterol and formoterol) are marketed for long term symptomatic treatment of asthma, in combination with antiinflammatory treatment with an inhaled steroid. (2) A double-blind placebo-controlled trial of salmeterol in about 26 000 patients was stopped prematurely in 2003 due to an excess of serious asthma attacks and asthma-related deaths in the salmeterol group. There were 13 asthma-related deaths after 28 weeks of salmeterol therapy, versus only three in the placebo group, for a relative risk of about 4 and a 95% confidence interval of 1.2 to 15.3. (3) This increased mortality was mainly observed in the subgroup of patients who were not using a steroid (53% of all patients). Nine of these patients died while taking salmeterol versus none on placebo. (4) Three smaller trials showed an increase in severe asthma attacks in patients using formoterol without a steroid. For example, in one trial nine of 271 patients on formoterol had severe asthma attacks, versus two of 277 patients on placebo. A meta-analysis of 19 placebo-controlled trials of long-acting beta2 agonists showed an increase in severe asthma attacks in both adults and children. Exacerbations have also been attributed to salmeterol, including in patients using a steroid. (5) In practice, for long-term treatment, long-acting beta2 agonists must never be used without concomitant inhaled steroid therapy.