A long-term follow-up analysis of associations between tooth loss and multiple cancers in the Linxian General Population cohort

Abstract

Poor oral health, indicated by tooth loss and periodontal disease, may be an important risk factor for various cancers. Prior studies have found inconsistent associations between tooth loss and several cancer types. Here, we examined the relationship between tooth loss and incident cases of multiple cancers in the Linxian General Population Nutrition Intervention Trial cohort. In this large prospective cohort of over 29,000 participants, there were 3101, 1701, 626, 327, 348, and 179 incident esophageal, gastric cardia, gastric noncardia, liver, lung, and colorectal cancer cases, respectively, over 30 years of follow-up. Adjusted Cox proportional hazards regression models with time-varying covariates were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between tooth loss and cancer outcomes during three time intervals: ≤ 5 years (early), > 5 and ≤ 10 years (mid), > 10 years (late). Tooth loss was assessed as quartiles of the number of lost teeth in excess of the loess-smoothed, age-specific median number of teeth lost. For esophageal cancer, the increase in risk associated with the highest quartile of tooth loss was 25% (95% CI: 1.02, 1.52) in the mid time interval, but the association weakened thereafter. For gastric cardia cancer, the increase in risk associated with the highest quartile of tooth loss was 1.34 in both the early (95% CI: 1.06, 1.71) and mid time intervals (95% CI: 1.02, 1.76), with no significant associations in the late interval. Gastric noncardia cancer was only associated with the second quartile of tooth loss in the late time interval (HR = 1.54; 95% CI: 1.16, 2.04). All associations between tooth loss and liver, lung, and colorectal cancers were null. Tooth loss was associated with risk of esophageal and gastric cancers in this updated analysis from the cohort.