Cigarette smoking and risk of bladder cancer: a dose–response meta-analysis

Introduction The disease burden of bladder cancer is increasing worldwide; therefore, to deal with this situation, many studies on bladder cancer have been carried out extensively. Among these studies, the risk factors studies may provide a possible way to reduce the incidence of bladder cancer. Meta-analyses and original researches have confirmed that smoking is a risk factor of bladder cancer. However, the specific dose–response relationship between smoking and bladder cancer risk was still unclear. This meta-analysis addresses this research gap by summarizing the accumulated evidences, quantitatively. Methods Relevant studies were obtained by searching PubMed, Embase and Web of Science database since inception until August 10, 2021, without restrictions. To obtain more comprehensive data, reference lists of identified articles were also browsed. Studies that reported risk estimates (relative risks (RR) or odds ratio (OR)) with 95% confidence intervals (CIs) focusing on the association between cigarette smoking and risk of bladder cancer were included in a dose–response meta-analysis. Results A non-linear dose–response relationship was confirmed between cigarette smoking and risk of bladder cancer on the basis of 8 cohorts and 44 case–control studies. The summary relative risk of developing bladder cancer for 1 more cigarettes/day (7 cohorts and 24 case–control studies) was 1.039 (95% CI 1.038–1.040, I 2 = 0%), for 1 more pack-year (3 cohorts and 21 case–control studies) was 1.017(95% CI 1.016–1.019, I 2 = 0%) and for 1 more year of exposure (16 case–control studies) was 1.021 (95% CI 1.020–1.023, I 2 = 0%). Conclusion A positive non-linear dose–response relationship is confirmed between all smoking intensity, pack-years of smoking, smoking duration(years) and the risk of bladder cancer, but the plateau only occurred when smoking intensity reached 20 cigarettes/day. Further studies should report more detailed results, including those for subtypes of gender, age, region and be stratified by other risk factors to rule out residual confounding.