The Reply

We thank Grau and George for their letter. We agree with their comments that the evidence for the beneficial pleiotropic effects of statins in community-acquired pneumonia is now difficult to deny, with a large number of studies of different designs demonstrating these benefits.1, 2, 3 It also seems increasingly unlikely that healthy user bias can account for all of the effects reported in these studies.4

We agree that the influence of pneumococcal and influenza vaccination may affect the results. The literature is, however, conflicting in this area. Recent studies suggest, conversely, that including vaccination status may have increased, rather than decreased, confounding. They showed no reduction in 30-day mortality or intensive care unit admission associated with influenza vaccination use as the result of a large degree of healthy user confounding.5, 6 No previous study has identified a relationship between vaccination status and complicated parapneumonic effusion or empyema.

We believe that only randomized controlled trials will be able to account for all of the potential confounders.

Further studies are required to elucidate the mechanism of the protective effects of statins and to demonstrate if these effects are truly independent of all potential confounders.