Benzodiazepines associated with community-acquired pneumonia

Benzodiazepines are associated with risk for and death from community-acquired pneumonia (CAP), according to a new study.

Researchers in the United Kingdom performed a nested case-control study of data from the Health Improvement Network, a database of primary care patients, to determine whether an association existed between CAP and therapy with benzodiazepines. They hypothesized that benzodiazepines might increase CAP risk because of their potential effects on immune function. The authors also examined the effects of zopiclone, a nonbenzodiazepine that has a similar mechanism of action. The study's primary end points were risk of CAP and mortality from CAP at 30 days and over the long term, a median follow-up of 2.8 years. The study results were published online on Dec. 5 by Thorax.

A total of 4,964 cases and 29,697 controls were included in the study. Cases were patients whose medical records indicated a diagnosis of pneumonia between July 1, 2001 and July 1, 2002. Approximately 54% of both cases and controls were women. The authors used conditional logistic regression to detect an association between drug use and CAP and then used Cox regression to determine the drugs' effects on mortality among the 4,964 cases.

An association was found between benzodiazepine use and increased pneumonia risk (odds ratio, 1.54; 95% CI, 1.45 to 1.67) and between use of diazepam, lorazepam or temazepam and increased CAP incidence. No such individual association was seen with chlordiazepoxide, but an increased risk was also associated with zopiclone. Thirty-day mortality (hazard ratio, 1.22; 95% CI, 1.06 to 1.39) and long-term mortality (hazard ratio, 1.32; 95% CI, 1.19 to 1.47) appeared to be higher in patients with CAP who had taken benzodiazepines. Each of the four benzodiazepines studied also appeared to individually affect long-term mortality, but zopiclone did not. Only diazepam and lorazepam appeared to individually affect 30-day mortality.

The authors noted that unmeasured confounders and selection bias may have affected their results and that their findings do not prove a cause-and-effect relationship between benzodiazepines and CAP, among other limitations. However, they concluded that benzodiazepines and zopiclone appear to be associated with higher risk for and death from CAP. “Given the widespread use of benzodiazepine drugs,” they wrote, “further studies are required to evaluate their safety in the context of infection.”