A recent observational study of the incidence of
pneumonia in patients with previous
hip fractures found that
bisphosphonate use reduced
pneumonia risk by about one-quarter, in comparisons with those either not receiving
osteoporosis treatment or receiving treatment with non-
bisphosphonate drugs. Mortality from
pneumonia was similarly reduced. It was hypothesized that effects of these drugs on immune or inflammatory function might mediate this effect. We have used the adverse event database from our recent 6-year randomized controlled trial of
zoledronate in 2000 women over the age of 65 years, to determine whether a similar effect is observed using this more rigorous study design. Seventy-five women had at least one episode of
pneumonia (32 [3.2%]
zoledronate, 43 [4.3%] placebo) and 119 women had at least one episode of either
pneumonia or a lower
respiratory tract infection (57 [5.7%]
zoledronate, 62 [6.2%] placebo). There were 93
pneumonia events and 167
pneumonia/lower respiratory
infection events. For
pneumonia, the hazard ratio associated with randomization to
zoledronate was 0.73 (95% confidence interval, 0.46-1.16; P = 0.18) and the rate ratio was 0.69 (0.45, 1.04; P = 0.073). For the composite endpoint of
pneumonia or lower respiratory
infection, the hazard ratio was 0.90 (0.61, 1.30; P = 0.58) and the rate ratio 0.74 (0.54, 0.997; P = 0.048). The proportion of people with events changed approximately linearly over time in both groups, suggesting a progressive divergence in cumulative incidence during the study. In conclusion, these findings lend support to the hypothesis that
bisphosphonate use reduces the number of lower
respiratory tract infections in older women, though the present study is under-powered for this endpoint and the findings are of borderline statistical significance. Further analysis of other trials of
bisphosphonates is necessary to test this possibility further, and exploration of the possible underlying mechanisms is needed.