S. Gevaert

Air pollution and ST-elevation myocardial infarction: A case-crossover study of the Belgian STEMI registry 2009–2013

BACKGROUND Previous studies have shown that air pollution particulate matter (PM) is associated with an increased risk for myocardial infarction. The effects of air pollution on the risk of ST-elevation myocardial infarction (STEMI), in particular the role of gaseous air pollutants such as NO2 and O3 and the susceptibility of specific populations, are still under debate. METHODS All patients entered in the Belgian prospective STEMI registry between 2009 and 2013 were included. Based on a validated spatial interpolation model from the Belgian Environment Agency, a national index was used to address the background level of air pollution exposure of Belgian population. A time-stratified and temperature-matched case-crossover analysis of the risk of STEMI was performed. RESULTS A total of 11,428 STEMI patients were included in the study. Each 10μg/m3 increase in PM10, PM2.5 and NO2 was associated with an increased odds ratio (ORs) of STEMI of 1.026 (CI 95%: 1.005-1.048), 1.028 (CI 95%: 1.003-1.054) and 1.051 (CI 95%: 1.018-1.084), respectively. No effect of O3 was found. STEMI was associated with PM10 exposure in patients ≥75y.o. (OR: 1.046, CI 95%: 1.002-1.092) and with NO2 in patients ≤54y.o. (OR: 1.071, CI 95%: 1.010-1.136). No effect of air pollution on cardiac arrest or in-hospital STEMI mortality was found. CONCLUSION PM2.5 and NO2 exposures incrementally increase the risk of STEMI. The risk related to PM appears to be greater in the elderly, while younger patients appear to be more susceptible to NO2 exposure.

Inter-hospital variation in length of hospital stay after ST-elevation myocardial infarction: results from the Belgian STEMI registry.

OBJECTIVE The aim of this paper was to assess the determinants of and variations in length of hospital stay (LOS) in Belgium after ST-elevation myocardial infarction (STEMI). METHODS AND RESULTS Data on LOS were collected from 2079 STEMI patients who were discharged alive from 33 Belgian hospitals (21 with PCI facilities) during 2010-201 1. Early discharge was defined as hospital discharge within 4 days after admission, and the hospitals were clustered according to their LOS for low-risk patients. Determinants of LOS were calculated by means of a negative binomial regression model. LOS was, on average, 6.5 days with a median of 5 days (IQR 4). Baseline risk profiles and reperfusion treatment explained only 13% of the LOS variation. Additional analysis revealed major in-hospital variations independent of the case mix of patients. For comparable baseline risk profiles, the average LOS in a cluster of 11 hospitals with short discharge policies was 5.3 + 5.6 days, with an early discharge rate of 58%, while in the cluster of 11 hospitals with long discharge policies, the average LOS was 7.9 + 8.5 days with an early discharge rate of 22% (P <0.0001). Among the clustered hospitals, there were no differences with regard to logistics (PCI facility, academic affiliation) or volume of STEMI patients. The 1-month mortality rate was less than 0.5% in the different clusters of hospitals (p = NS). CONCLUSIONS Length of hospital stay is not only determined by baseline risk profiles of patients but is also highly dependent on hospital discharge policy, which seems to be unrelated to medical or logistical factors.