Association of ambient particulate matter with heart failure incidence and all-cause readmissions in Tasmania: an observational study

Objectives We sought to investigate the relationship between air quality and heart failure (HF) incidence and rehospitalisation to elucidate whether there is a threshold in this relationship and whether this relationship differs for HF incidence and rehospitalisation. Methods This retrospective observational study was performed in an Australian state-wide setting, where air pollution is mainly associated with wood-burning for winter heating. Data included all 1246 patients with a first-ever HF hospitalisation and their 3011 subsequent allcause readmissions during 2009-2012. Daily particulate matter <2.5 mu m (PM2.5), temperature, relative humidity and influenza infection were recorded. Poisson regression was used, with adjustment for time trend, public and school holiday and day of week. Results Tasmania has excellent air quality (median PM2.5=2.9 mu g/m(3) (IOR: 1.8-6.0)). Greater HF incidences and readmissions occurred in winter than in other seasons (p<0.001). PM2.5 was detrimentally associated with HF incidence (risk ratio (RR)=1.29 (1.15-1.42)) and weakly so with readmission (RR=1.07 (1.02-1.17)), with 1 day time lag. In multivariable analyses, PM2.5 significantly predicted HF incidence (RR=1.12 (1.01-1.24)) but not readmission (RR=0.96 (0.89-1.04)). HF incidence was similarly low when PM <4 mu g/m(3) and only started to rise when PM2.5 >= 4 mu g/m(3) . Stratified analyses showed that PM2.5 was associated with readmissions among patients not taking beta-blockers but not among those taking beta-blockers (P-interaction=0.011). Conclusions PM2.5 predicted HF incidence, independent of other environmental factors. A possible threshold of PM2.5 = 4 mu g/m(3) is far below the daily Australian national standard of 25 mu g/m(3). Our data suggest that beta-blockers might play a role in preventing adverse association between air pollution and patients with HF.