Letter to Editor regarding article, ‘Association between dental factors and mortality’

Abstract

Letter to Editor regarding article, 'Association between dental factors and mortality'Dear Editor,We read with great interest the study by Liljestrand et al. (2020) evaluating the association between oral parameters at baseline with cardiovascular mortality and all-cause mortality in a follow-up of approximately 8 years, which demonstrated that number of missing teeth appeared to be the strongest predictor of mortality, and signs of professional intervention such as root canal fillings and removable dentures appeared to be associated with improved survival. After reading this article carefully, we would like to make the following suggestions:First, as described in the methodology, 'The Finnish Parogene cohort consists of 508 Finnish adults (mean age 63.3 years, SD 9.1) with cardiac symptoms, all of whom have undergone coronary angiography for accurate baseline coronary status. A total of n =471 dentate patients were included in the statistical analyses'. Here, we question why the remaining 37 patients were not included in analysis. Does this lead to selection bias?Second, the participants included were patients with coronary artery disease (CAD). Theoretically, these patients are routinely recommended for antiplatelet therapy such as aspirin or clopidogrel, which can reduce mortality but increase the risk of bleeding, including gingival bleeding on probing at the same time (Piepoli et al. 2016). However, the authors did not describe the antiplatelet agents in this study, which might increase the incidence of confounding bias. The results might change when these confounders are adjusted.Third, the mortality of cardiovascular disease patients increases with age (Sidney et al. 2019). According to the World Health Organization (WHO) age classification and general age stratification, we sincerely suggest a subgroup analysis by dividing patients into three groups: (1) <60 years; (2) 60-75 years; (3) >75 years in this article, which will make the results more interesting.Forth, the continuous data were expressed as mean AE SD and compared using Mann-Whitney U-test in 'Table 2'. It is recommended that continuous data should be expressed as mean AE standard deviation if normally distributed and analysed using Student's t-test for unpaired samples (Parfejevs et al. 2018). However, no more description about data was provided. The results might change when the statistical analysis is adjusted.