Association between periodontal disease and coronary heart disease risk in Chinese population: evidence from a meta-analysis

Aims More and more research is focused on studying the relationship of periodontal disease with susceptibility to coronary heart disease (CHD), among which discrepancies have risen. The current study aimed to evaluate the possible relationship about periodontal disease and CHD in certain Chinese population. Methods The PubMed, Embase, Web of Science, the Cochrane Library, the Chinese Biomedical Database, and the Chinese National Knowledge Infrastructure were searched up to December 2018. The overall and subgroup analyses were performed using �xed or random-effect model depending on heterogeneity. Results A total of 2241 CHD cases and 2160 in 15 studies were used in the current study. Overall, a signi�cantly elevated relationship about periodontal disease on CHD risk was found in all included Chinese subjects (OR=3.04, 95% CI, 2.37-3.92). Subgroup analyses by geographic location and sources of control were also performed; all the results in subgroup analyses were consistent with the overall result. Meanwhile, the signi�cant association was found among the measures of periodontal disease with attachment loss, periodontal index, two or more clinical indications, and gum recession as well as CHD assessment method with WHO criteria and coronary arteriongraphy. Conclusions Collectively, the meta-analysis indicated that periodontal disease was capable of causing CHD susceptibility among Chinese.


Introduction
At present, cardiovascular disease accounts for 30% of all deaths worldwide.Moreover, the great majority of this disease burden is currently occurred in developing countries [1].The death of coronary heart disease (CHD) dropped signi cantly in United States after peaking around 1968 [2].But, CHD had become a public health problem during the past few decades in China.The mortality rate of CHD (every 100000 people per year) in China has risen from 95.3 in 1999 to 103.4 in 2008 [3].In 2008, the crude incidence rate of ischemic heart disease in Chinese urban inhabitants was 12.7% [4].Although some factors which may affect the risk of CHD had been explored, a large proportion of CHD has not been explained by the traditional risk factor.In the recent publications, several evidences had involved the cause of chronic in ammation in CHD [5].Interestingly, periodontal disease had been studied to be related with exaltation of several markers of chronic in ammation, therefore, we assumed an etiologic association about periodontal disease on CHD risk.The association about periodontal disease on CHD risk had been studied since a century ago, and many systematic reviews and metaanalyses had assessed the potential association of periodontal disease with the increased risk of CHD [6][7][8][9][10][11][12][13][14].Although a large amount of evidence was showed, a scienti c statement declaration by the American Heart Association in 2012 expounded that it has not enough evidence to determine the causal-related association about periodontal disease on the risk of heart disease [15].It was still unclear whether this relationship differs among people from different race backgrounds.Therefore, the current study employed a meta-analysis to quantitatively clarify the association about periodontal disease on CHD risk based on included Chinese population.

Identi cation of eligible studies
Meta-analysis was conducted according to the guidelines of the PRISMA group.Search scopes included articles from PubMed, Embase, Web of Science, the Cochrane Library, the Chinese Biomedical Database [CBM], and the Chinese National Knowledge Infrastructure [CNKI] databases published until December 2018.The search keywords were ("coronary heart disease"OR "coronary disease"OR "heart disease"OR "ischemic heart disease"OR "myocardial infarction") AND ("periodontal disease"OR "periodontitis") AND ("China" OR "Chinese") .No language limitation was applied.To available identify relevant studies; we also manually search for references cited in the eligible articles.A study should be included if it met all the following criteria: (1) it was a case-control study to investigate the association about periodontal disease on CHD risk; (2) it provided odds ratios (ORs) and 95% con dence intervals (CIs); (3) it evaluated the population in Chinese; (4) the study which had the most complete data was used if two or more studies had partial overlaps.The study was excluded if it was review, letter, case report, and editorial article.

Data extraction
We ltered the titles and abstracts of all possible related papers to identify their relevancy.The full paper would be reviewed if it cannot be sure depending on the title and abstract.Information from all eligible publications was independently extracted with two investigators.The following data were extracted: rst author's name; publication years; geographic location; sources of control; number of participants; and measures of periodontal disease.Sources of control were divided into hospital-based or population-based studies.The following measures for periodontal disease in dental examinations were: periodontal index, gum bleeding, attachment loss, gum recession, teeth loosening, in ammation/gingivitis, and tooth loss.

Statistical analysis
The pooled results about periodontal disease on CHD risk were calculated by ORs with 95% CIs.Z test was adopted to explore whether the pooled ORs were signi cant.Regarding potential heterogeneity among studies, which assessed by Cochran's Q-statistic, we de ned signi cant heterogeneity at the levels P < 0.10.A random-effect model was used on the analyses when signi cant heterogeneity was observed, otherwise xed-effect model was used.Sensitivity analysis was carried out by compared the results of xed-effect model with random-effect model.Egger's test was used to detect the publication bias.Our meta-analysis was calculated by Stata 12.0 (College Station, TX, USA); p < 0.05 was known as signi cant.In addition, subgroup analyses by geographic location and source of controls to further explore the association about periodontal disease on CHD risk.Furthermore, we pooled the ORs about dental and CHD respectively if they assessed the periodontal disease differently to explore in uence of the measures.

Research characteristics
One hundred and eighty-eight articles which evaluated the association about periodontal disease on CHD risk were searched.At last, fteen articles [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30] were included in our nal analysis.All the eligible articles were published between 2001 and 2017.Figure 1 shows the owchart of included or excluded reason.Totally, there were 2241 CHD cases and 2160 controls involved in our analysis, which were included to assess the potential association about periodontal disease on CHD risk in Chinese.The characteristics of each individual article are presented in Table 1.

Meta-analysis results
In the results of the overall analysis, periodontal disease could signi cantly increase the risk of CHD (ORs= 3.04, 95% CI= 2.37-3.92; Figure 2).Meanwhile, the heterogeneity was detected signi cantly (I 2 = 69.8%,p<0.001).Detailed results summarized in table 2 found that all the results in subgroup analyses were consistent with the overall analysis, suggesting no single factor had essential impact on the whole result.Furthermore, subgroup analysis was also performed to explore the sources of heterogeneity.The further subgroup study by geographic location and source of controls suggested that signi cant association was found in northern and southern of China, as well as population-based studies and hospitalbased studies.Some articles had explained the categories for dental examinations and CHD assessment method, as follows: attachment loss; periodontal index; two or more clinical indications; gum recession; alveolar bone loss; WHO criteria and coronary arteriongraphy.The signi cant association was found among the measures of periodontal disease with attachment loss, periodontal index, two or more clinical indications, and gum recession as well as CHD assessment method with WHO criteria and coronary arteriongraphy (Table 2).However, there was non-signi cant association was found in the dental assessment method with alveolar bone loss.

Sensitivity analysis and publication bias
Sensitivity analysis which described in the methods was performed.All the pooled ORs were signi cantly both in the total and subgroup analyses.Hence, results of the sensitivity analysis suggested that the results in the current study were relatively stable (Table 2).Publication bias was detected by Begg's funnel plot and Egger's test.As shown in Figure 3, funnel plot indicated that the shape was symmetrical.At the same time, the Egger's test (t = 3.21, P = 0.007; Figure 4) did not detect any publication bias.

Discussion
The associations about periodontal disease on both CHD and cardiovascular disease risk have been explored for many years.These diseases appear to have a number of characteristics in common such as high incidence in older persons, smokers and hypertensive patients etc [31].Since periodontal disease and CHD are common, quantifying their association is of signi cant public health importance [9].Our study assessing the association about periodontal disease, including attachment loss, periodontal index, gum recession, and alveolar bone loss, suggested that periodontal disease could be signi cantly increased the risk of CHD in Chinese.These analyses also provided the currently assessment of the relationship about periodontal disease on CHD risk.The present meta-analysis was carried out to clarify the effects for Chinese only and therefore to reduce the in uence of different geographic location background or lifestyle.
Our meta-analysis used 15 studies involving 2241 CHD cases and 2160 controls.Pooled results from the study suggested that periodontal disease cases had an overall 3.04-fold increased risk of CHD when compared to those with no periodontal disease or the general people in China.Subgroup analyses were also conducted to further explore the effective of geographic location and sources of control.Signi cant association was also found about periodontal disease on CHD risk in northern and southern of China, as well as studies with hospital-based controls and population-based controls.In addition, the signi cant association was found among the measures of periodontal disease with attachment loss, periodontal index, two or more clinical indications, and gum recession as well as CHD assessment method with WHO criteria and coronary arteriongraphy.
There have been several published meta-analyses which re ected that periodontal disease may contribute to the risk of CHD [6][7][8][9][10][11][12][13][14].It is obvious lack of standardized protocols for determining periodontal disease and CHD.Some research used self-report, which may underestimate the level of disease, while other studies including gingivitis and periodontitis may overestimate disease prevalence.Thus, this point should be addressed with a meta-analysis in order to reach a consensus.On the other hand, we only included Chinese population with relatively consistent evaluation criteria for periodontal disease and CHD, and conducting subgroup analyses by geographic location and source of controls.Two more databases (CBM and CNKI) in Chinese language were searched and more Chinese participants included in our study than the previous metaanalyses.Furthermore, the pooled OR about the association of periodontal disease on CHD risk in our study was higher than those published meat-analyses with populations except Chinese.Sensitivity analysis con rmed the reliability and stability of this study.Thus, ndings from the present paper strongly supported that periodontal disease could signi cantly increase CHD risk in Chinese.Some limitations existed in our study should be attention.Firstly, a publication bias has been indicated in our study, as all included studies were positive results and they may be easier to publish.Secondly, the subgroup analyses by aging, cigarette smoking, alcohol consumption, and dietary structure were not performed because there was no enough information in each individual study.Thirdly, CHD is a complex disease with multiple factors, containing environmental and genetic factors.However, some studies in our meta-analysis didn't consider the effects of environmental factors on CHD.
In conclusion, the meta-analysis indicated that periodontal disease was capable of causing CHD susceptibility among Chinese.While some limitations exited in the current paper, more studies with large participants of different background are warranted to validate this association reported in our analysis.It is recommendation to prevent the heart disease with the maintenance of periodontal health due to the high mortality rate of cardiovascular diseases and the relatively minor morbidity of periodontal therapy.

Figures
Figures

Figure 1 Flow
Figure 1