BACKGROUND

A number of studies suggested a nexus between long-term exposure to nitrogen dioxide (NO) and the incidence of cardiovascular disease (CVD), while population-based cohort evidence in low- and middle-income countries was extensively sparse.

METHODS

We carried out an 8-year longitudinal study (2010-2018) in a nationwide dynamic cohort of 36,948 Chinese adult participants, who were free of CVD at baseline. Annual average estimates of NO exposure were predicted using a well-validated spatiotemporal model and assigned to study participants based on their residential counties. Considering death as a competing risk event, Fine-Gray competing risk models with time-varying exposures at an annual scale were used to quantify incident risks of overall CVD, hypertension, and stroke associated with a 10-μg/m rise in NO exposure. Using the meta-analysis approach, we performed a pooled analysis of hazard ratio (HR) drawn from this and prior multinational cohort studies for the assessment of attributable burden. NO-attributable overall CVD incidents in China were evaluated by city and province for years 2010 and 2018, referring to a counterfactual exposure level of 10 μg/m (2021 World Health Organization [WHO] air quality guidelines). A decomposition method was used to decompose net change in NO-attributable CVD incidents during 2010 and 2018 into 3 primary contributions of driving factors (i.e., changes in NO exposure, population size, and incidence rate).

RESULTS

A total of 4428 overall CVD events (hypertension 2448, stroke 1044) occurred during a median follow-up period of 6.1 years. Annual mean NO concentration from 2010 to 2018 was 20.0 μg/m (range: 6.9-57.4 μg/m). An increase of 10-µg/m in NO was associated with an HR of 1.558 (95% confidence interval [CI]: 1.477, 1.642) for overall CVD, 1.521 (95% CI: 1.419, 1.631) for hypertension, and 1.664 (95% CI: 1.485, 1.865) for stroke. Longitudinal associations of NO exposure with incident CVD were nearly linear over the exposure range, suggesting no discernible thresholds. Subgroup analyses indicated significantly higher NO-associated risks of incident CVD among urban residents and overweight/obese individuals. According to pooled HR of NO-CVD association (1.108, 95% CI: [1.007, 1.219]) from 10 multinational cohort studies, we estimated totally 1.44 million incident CVD cases attributable to NO exposure in 2018, representing a substantial decrease of 0.41 million compared to the estimate in 2010 (1.85 million) in mainland of China. Nationally, from 2010 to 2018, the attributable incident cases greatly dropped by 22.4%, which was dominantly driven by declined NO concentration (-47.1%) that had offset far from the rise of CVD incidence rate (+19.6%) and population growth (+5.1%).

CONCLUSIONS

This study provided nationwide cohort evidence for elevated risks of CVD incidence associated with long-term ambient NO exposure among Chinese adults, particularly in urban areas and among overweight/obese individuals. Our findings highlighted that reducing NO exposure below 2021 WHO guideline could help prevent a substantial portion of incident CVD cases in China.