Ambient carbon monoxide and cardiovascular-related hospital admissions: A time-series analysis

Background and aims: Although several studies have investigated the association between air pollutants and cardiovascular diseases (CVDs) in recent years, a lack of evidence exists regarding carbon monoxide (CO) exposure, especially in the Eastern Mediterranean's polluted regions. In this study, we aimed to evaluate the short-term effect of CO exposure on daily CVD hospital admissions in Isfahan, a major city in Iran. Methods: Data were extracted from the CAPACITY study on daily CVD hospital admissions in Isfahan from March 2010 to March 2012. The 24-h mean CO concentrations were obtained from four local monitoring stations. In a time-series framework, the association between CO and daily hospitalizations for total and cause-specific CVDs in adults (ischemic heart disease (IHD), heart failure (HF), and cerebrovascular disease) was conducted using Poisson's (or negative binomial) regression, after adjusting for holidays, temperature, dew point, and wind speed, considering different lags and mean lags of CO. The robustness of the results was examined via two- and multiple-pollutant models. Stratified analysis was also conducted for age groups (18-64 and ≥65 years), sex, and seasons (cold and warm). Results: The current study incorporated a total of 24,335 hospitalized patients, (51.6%) male with a mean age of 61.9 ± 16.4 years. The mean CO concentration was 4.5 ± 2.3 mg/m³. For a 1 mg/m³ increase in CO, we found a significant association with the number of CVD hospitalizations. The largest adjusted percent change in HF cases was seen in lag0, 4.61% (2.23, 7.05), while that for total CVDs, IHD, and cerebrovascular diseases occurred in mean lag2-5, 2.31% (1.42, 3.22), 2.23% (1.04, 3.43), and 5.70% (3.59, 7.85), respectively. Results were found to be robust in two- and multiple-pollutant models. Although the associations changed for sex, age groups, and seasons, they remained significant for IHD and total CVD, except for the warm season, and for HF, except for the younger age group and cold seasons. Additionally, the exposure-response relationship curve of the CO concentrations with total and cause-specific CVD admissions showed non-linear relationships for IHD and total CVDs. Conclusions: Our results showed that exposure to CO contributed to an increase in the number of CVD hospitalizations. The associations were not independent of age groups, season, and sex.