Bond Julia C, McDonough Robert, Alshihayb Talal S, Kaye Elizabeth K, Garcia Raul I, Heaton Brenda
Department of Health Policy and Health Services Research, Boston University Henry M. Goldman School of Dental Medicine, Boston, Massachusetts, USA.
Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts, USA.
J Clin Periodontol. 2023 Jan;50(1):71-79. doi: 10.1111/jcpe.13722. Epub 2022 Oct 4.
To evaluate the association between periodontal disease and all-cause mortality in a longitudinal cohort study over 50 years.
Participants (N = 1156) in the Veterans Affairs Dental Longitudinal Study, aged 25-85 years at enrollment in 1968, received comprehensive medical and oral exams approximately every 3 years through 2007. Periodontal status was defined using person-level, mean whole-mouth radiographic alveolar bone loss (ABL) scores using a five-point Schei ruler, each unit representing 20% increments of ABL. Time-varying Cox regression models estimated hazard ratios (HRs) for the association between continuous and categorical ABL and mortality, adjusting for covariates.
Each one-unit increase in mean ABL score was associated with a 14% increase in the hazard of mortality (adjusted HR = 1.14, 95% confidence interval [CI] 1.02, 1.27). When assessed categorically, HRs for average scores of 2 to <3 and 3 to ≤5 showed increasing associations with hazard of mortality, relative to 0 to <1 (adjusted HR = 1.17, 95% CI 0.94, 1.46; and HR = 1.65, 95% CI 0.94, 2.85, respectively). By contrast, we observed null associations for average scores of 1 to <2 relative to 0 to <1 (adjusted HR = 1.00, 95% CI 0.86, 1.17).
Time-varying periodontal status assessed using radiographic ABL was positively associated with all-cause mortality even after confounder adjustment.
在一项为期50年的纵向队列研究中评估牙周疾病与全因死亡率之间的关联。
退伍军人事务部牙科纵向研究的参与者（N = 1156），于1968年入组时年龄在25 - 85岁之间，在2007年之前大约每3年接受一次全面的医学和口腔检查。牙周状况使用个人层面的平均全口放射学牙槽骨吸收（ABL）分数来定义，采用五点Schei尺，每个单位代表ABL增加20%。时变Cox回归模型估计连续和分类ABL与死亡率之间关联的风险比（HRs），并对协变量进行调整。
平均ABL分数每增加一个单位，死亡风险增加14%（调整后的HR = 1.14，95%置信区间[CI] 1.02, 1.27）。当进行分类评估时，相对于0至<1，平均分数为2至<3和3至≤5的HRs显示与死亡风险的关联增加（调整后的HR分别为1.17，95% CI 0.94, 1.46；以及HR = 1.65，95% CI 0.94, 2.85）。相比之下，我们观察到相对于0至<1，平均分数为1至<2时无关联（调整后的HR = 1.00，95% CI 0.86, 1.17）。
即使在对混杂因素进行调整后，使用放射学ABL评估的时变牙周状况与全因死亡率呈正相关。
