Salmeterol: A study by prescription-event monitoring in a UK cohort of 15,407 patients

Objective. To examine the safety of the long acting beta(2)-agonist, salmeterol, in general medical practice in the U.K. Design. In Prescription-Event Monitoring (PEM) the exposure data are derived from prescriptions confidentially provided by the Prescription Pricing Authority. Outcome data are obtained by questionnaires (green forms) on which prescribing medical practitioners recorded event (safety) data. Additional information is obtained from the patients' lifetime clinical records, from follow-up of all pregnancies, and from death certificates. Setting. PEM provides an observational cohort study in which there is no interference with the medical decision on which drug is prescribed for individual patients. Subjects. A total of 15,407 patients were given salmeterol and observed for a minimum of 1 year. Results. Headache, tremor, and palpitations were associated with the use of salmeterol. No unexpected major adverse events were noted, although in this study there was a total of 1022 deaths. Of these, 73 were due to asthma, but only 39 of these patients were taking salmeterol in the last month of life. All of these deaths appear on careful examination of the clinical records to have been due to natural causes, although in a minority (four, 10%) of the group of 39 subjects, a temporal relationship between death and the use of salmeterol makes it difficult to exclude a possible association. The study suggests that advanced age and severity of disease were the most likely factors contributing to asthma mortality in the population studied. Conclusion. Headache, tremor, and palpitations occur as adverse effects of salmeterol and lead to withdrawal of the drug in some subjects. No unexpected adverse reaction has been noted and paradoxical bronchospasm was reported in only three subjects. The study has produced no evidence that salmeterol contributed to death in any of the patients examined.