Effect of obesity on risk and severity of periodontitis: a cross-sectional study

Abbreviations

AUC: Area under curve

BMI: Body mass index

BOP: Bleeding on probing

CAL: Clinical attachment loss

CI: Confidence of interval

COPD: Chronic obstructive pulmonary disease

IQR: Interquartile range

OHI: Oral hygiene index

OR: Odds ratio

PD: Pocket Depth

PI: Plaque index

ROC: Receiver operating characteristic

Introduction

Chronic periodontitis shares many common risk factors with other chronic diseases such as hypertension, cardiovascular disease, and type 2 diabetes mellitus. Common risk factors include smoking, alcohol consumption, and obesity. The underlying mechanisms related to obesity are thought to contribute to inflammation and thereby to chronic disease.¹

The etiology of chronic periodontitis derives from the series of complex interactions between pathologic microorganisms in the bacterial plaque and the host and modifications by systemic and local factors.² Susceptibility to periodontitis also affected by genetic factors which modulate individual responses to the environment and variations of the immune response.³ There are many proinflammatory cytokines involved in periodontitis subjects, such as IL-1, IL-6, TNF-α which also correlates with proinflammatory cytokines in obese patients⁴ (Makki, 2013). Obesity modulates the host immune response by secreting several proinflammatory factors deriving from adipose tissue (adipocytes) which results in increased bone loss.⁵

Obesity has become a worldwide concern because it’s incidence has nearly tripled since 1975. In 2016 more than 1.9 billion adults 18 years and older were overweight, with 650 millions of those being obese.⁶ The prevalence of obesity among adults over 18 years old in the U.S. was 39.8% according to a National Health and Nutrition Examination Survey conducted in 2015-2016. Asian non-Hispanic adults had the lowest prevalence (12.7%) compare with all other races and Hispanic-origin groups.⁷

In Indonesia according to Indonesian basic health research 2018 (Riskesdas, 2018) the prevalence of obesity in Indonesian adults over 18 years old was 21.8% and the prevalence of periodontitis was 74.1%. Jakarta, the Indonesian capital city, has the second highest obesity prevalence among 34 other provinces at 29,8% whilst the prevalence of periodontitis in urban areas was 71.5%.⁸

Obesity is usually defined by body mass index (BMI) measurement. According to the WHO, ‘overweight’ classification is set at the value of 25.0-29.9 kg/m² and ‘obesity’ defined as a BMI ≥ 30.0 kg/m². However in Asian regions BMIs can be set lower than the existing WHO cut-off point, with the overweight BMI at 22-25 kg/m² and the obesity BMI at 26-31 kg/m².⁹ The WHO proposed classification of a weight by BMI in adult Asians for which obesity 1 is defined at 25-29.9 kg/m² and obesity 2 at ≥ 30 kg/m².¹⁰

Increased BMI may be a potential risk factor for developing periodontitis.^11–13 On the other hand, other studies have shown obesity to be not related to the severity of periodontal disease,¹⁴ but associated with tooth loss, oral hygiene and education level.¹⁵ Obesity is related to adipokines that are secreted from adipose tissue which has an important role in regulating metabolic and vascular biology.¹⁶ Meanwhile, as a result of chronic inflammatory state insulin resistance develops as well as oxidative stress. These factors could be implicated in the possible obesity and periodontitis association.¹⁷ BMI and obesity in non-surgical periodontal therapy also appear to be independent predictors of poor response to the treatment.¹⁸

The aim of this study was to ascertain the odds of having periodontitis in obese individuals among other periodontitis risk factors. We hypothesized that obesity correlated with having higher risk of periodontitis compared to normal-weight individuals.

Methods

Results

A total of 272 subjects were recruited; however, some subjects were not eligible according to the inclusion criteria: such as having fewer than 14 teeth, or age. Missing data were excluded (Figure 1). Therefore, a total of 262 subjects between the ages of 18 and 66 were included in this study. The percentage of female subjects was larger (59.9%) than male subjects (40.1%) as presented in the general profile (Table 1). The age group of 45-54 made up two fifths of the sample size, whereas the fewest subjects were in the 25-34 years age group. Subject’s education status was mostly high school (53.4%) and the primary occupation was housewife (47.3%). The subjects were mostly non-smokers (79.4%) and non-alcohol-consumers (93.5%). 50.8% of subjects had either healthy periodontal scores or mild periodontitis, and the rest had moderate or severe periodontitis.

Figure 1. Flow diagram of the study.

The socio-demographic and BMI analysis showed significant positive correlations for age, sex, hypertension, and periodontal status. Significant negative correlations were shown for occupation and smoking status (coefficient correlation −0.270, p = 0.000; coefficient correlation −0.142, p = 0.021, respectively) (Table 1). Education, alcohol, DM, and plaque index showed no correlations with periodontal status.

The prevalence of obesity in this study was 48.47% and the highest prevalence was in BMI group 25.0-29.9 kg/m² (32.44%), then 18.5-22.9 kg/m² (29.77%). Table 2 presents the correlation between binary socio-demography data, oral hygiene, and BMI as a dependent factor. A significant positive correlation found between periodontitis and age, sex, occupation, and hypertension and while smoking showed a negative correlation. Continuous data of the clinical parameters of periodontitis and BMI is shown in Table 3. PD, CAL, and PBI were found to have significant correlations with BMI.

A significant positive correlation found between BMI and periodontal status. There was and association between mild periodontal status and moderate periodontitis, and between mild periodontal status and severe periodontitis were p = 0.03 and p = 0.04 respectively (Figure 2).

Figure 2. Body mass index and periodontal status; Kruskal-Wallis test *p < 0.05.

Binary periodontal status categorized moderate and severe periodontitis as one group and healthy periodontal and mild periodontitis as another group. Binary periodontal status was then analyzed as a dependent factor towards the various risk factors (Table 4). The multivariate logistic regression analysis was then performed for variables with p < 0.25 (age, sex, smoking, DM, BMI and OHIS). Diabetes mellitus status and smoking had p-values above 0.25, which were p = 0.301 and p = 0.431 respectively, but both were included in the analyses because DM and smoking status are both important risk factors in periodontal classification.

Multivariate logistic regression analysis showed that periodontitis subjects were more likely to have a BMI ≥ 25 kg/m² (p < 0.001; adjusted odds ratio 1.881; 95% CI: 1.050-3.371). The risk of periodontitis was higher in male compare to female (p < 0.001; adjusted OR 6.852; 95%CI: 2.540-18.481), and increased with OHI. Adjusted OR for smoking showed that nonsmoking subjects were more prone to periodontitis. The ROC analyses from multivariate logistic regression had an area under the curve of 0.78 (sensitivity 71.3%, specificity 61.7%), determining that it was acceptable using this approach to discriminate between those individuals with healthy or mild periodontitis and severe periodontitis.

Despite the fixed value of the Asian cut-off point in BMI, we tried to estimate the cut-off point of BMI in our subjects. Figure 3 shows that the BMI special cut point with the periodontal status, was found in the value of 24.785 kg/m² (sensitivity 60.5%, specificity 60.9%, AUC 0.608).

Figure 3. Cut off point body mass index: AUC = 0.629; BMI = 24.785 kg/m²; sensitivity 60.5%, specificity 60.9%.

Discussion

A previous study (in which the age ranged from 25 to 66 years) showed that the prevalence of periodontitis was greater among subjects with obesity and the increased was BMI proposed as potential risk factor for periodontitis.¹² Sources of potential bias were that BMI and not body fat distribution, which can be calculated by waist circumference (WC), was measured. Fat distribution varies between woman and men and between race/ethnic groups.^23,24 Participants were taken from three district in central Jakarta which represent mixed ethnic in Indonesia and can be generalized as Indonesian sub population.

The present study covered a slightly wider range of age from 18 up to 66 years old. The highest group with severe periodontitis was in the 45-54 years age group (46.5%). Indonesian prevalence data for periodontitis also showed the biggest prevalence was in 45-54 years age group (77.85%).⁸

The WHO BMI categories define underweight (<18.5 kg/m²), normal weight (18.5-24.9 kg/m²) and overweight (25.0-29.9 kg/m²), and obese (≥30 kg/m²) while in WHO Asia-Pacific the agreed cut-off for overweight, obese I and obese II category are 23.0 kg/m², 25-29.9 kg/m² and ≥30 kg/m² respectively.¹⁰ In the Asia-Pacific area, the BMI cut-off for obesity is 25 kg/m² rather than the WHO cut-off which is 30 kg/m².²⁵ We have assumed that the BMI classification for the Asia-Pacific region is more suitable for Asian patients.

The specific cut off BMI in this study of 24.785 kg/m² (sensitivity 61.1%, specificity 61%, AUC 0.63) shows resemblance to a study by Suvan et al. which has a cut-off point of 24.32 kg/m² (sensitivity 61.54%, specificity 61.89%).¹² Likelihood ratio (LR) in this study was 1.56 (the disease is 1.56 times more likely in individual with the BMI ≥ 24.78 compare to normal weight < 24.78). This exploration by use of the ROC curve confirmed that using WHO Asian-Pacific standard was more suitable in our study rather than BMI 25-29.99 kg/m² defined for overweight and BMI ≥ 30 kg/m² for obesity.⁶ The BMI increased with increasing age and severity of periodontitis. Both the continuous data and the categorical data showed correlations between BMI and periodontitis. This data was in agreement with previous studies among the population aged 28-55¹¹ and among adults aged 18 to 24 years.²⁶ The clinical parameters of periodontitis (PD, CAL, recession and BOP) presented positive correlations with BMI (p < 0.05). This finding was in line with Zimmermann et al. , a German population cohort study which also found an increased deeper periodontal pocket with higher BMI (OR = 1.6, BMI increase by 5).²⁷

In this study, education level was at the margin of statistical significance (p = 0.053), as in a previous study which showed association with education, tooth loss, and oral hygiene.¹⁵ Tooth loss and oral hygiene were not related to increased BMI in this study, may be because the lowest remaining teeth was 14 teeth as an inclusion criteria while in that study, there was no particular inclusion criteria based on tooth loss. Song et al. in their study found that tooth loss was considered a potential risk indicator for being underweight in Korean adults.²⁸ Regarding poor periodontal health with obesity, subjects had poor compliance towards oral hygiene,¹³ and obese patients had an approximately three fold higher number of bacterial species present compared with normal weight controls with total of 23 species.²⁷ In this study, oral hygiene was not related to BMI although oral hygiene is significantly related with the severity of periodontitis.

Although many studies have shown the relationship between BMI and periodontitis^17,27,29,30 as in this study, there are some studies which have on the contrary showed no difference between no or mild periodontitis, moderate periodontitis and severe periodontitis.^15,31 Also in another study of BMI and periodontitis in postmenopausal women, those with higher BMI had decreased odds (OR) for having periodontitis compared to participants with normal weight (OR: 0.54; 95% CI: 0.27-0.87) although the obese presented significantly higher clinical attachment loss and gingival index compared to normal and overweight subjects (p < 0.01).³² On the contrary, Puspitadewi et al. in the study of postmenopausal women found no significant correlation between age, BMI, bone density and alveolar bone resorption (p > 0.05).³³

Females had a higher BMI compared to male subjects, and according to occupation housewives have the highest BMI compared to employed subjects and students, this corresponds with other studies that have found females to have a higher BMI compared to males.¹⁵ Smoking subjects in this study had a lower BMI than non-smoking subjects. Smoking was not correlated with periodontitis, but after adjusting other covariates, a significant correlation was found between non-smokers and periodontitis. This may be due to the low percentage of smokers (20.6%) compared to non-smokers (79.4%) in our study. BMI has multiple risk factors (Table 2) but after multivariate logistic regression with BMI as dependent factor, only age give a significant result p < 0.001; OR 3.63 (95% CI, 1.86-7.12).

Multivariate regression in role factors in having periodontitis, adjusted OR showed sex, OHI, smoking, and BMI, p < 0.001, OR 6.852 95% CI; p < 0.001, OR 5.189 95% CI 2.752-9.784; p = 0.007, OR 3.803 95% CI 1.449-9.979; and p = 0.034; OR 1.881 95% CI 1.050-3.371 respectively. Together, it means obese male non-smokers with bad oral hygiene are prone to periodontitis.

The odds ratio for BMI was 1.881 (95% CI 1.050-3.371) slightly higher than a previous study where OR was 1.56 (95% CI 1.26-1.92).²⁶ This finding was contrary to a study in Korea by Kim et al. who with multivariate analysis found no association between BMI and periodontitis with BMI ≥ 25, adjusted ratio 0.991 (95% CI 0.806-1.220) but found a significant association between abdominal obesity and periodontitis with an adjusted odds ratio of 1.358 (95% CI 1.003-1.839).¹⁴ Associations between obesity and periodontitis were found more consistently for visceral than general adiposity, suggesting that visceral fat accumulation measurement may be more strongly associated with periodontitis more than BMI.²⁹ So we suggest for further study including waist hip ratio as a comparing method for BMI measurement.

Conclusion

This study showed a significant correlation between BMI and periodontitis in Indonesian adults. Therefore, BMI evaluation can be used as a factor for assessing the risk of periodontitis.

Data availability

Consent

Written informed consent for publication of the patient details was obtained from the patients.

Author contributions

CM: Data Curation, Investigation, Visualization, Writing-Original Draft Preparation, Writing-Review & Editing; EIA: Conceptualization, Recourses, Supervision, Validation, Writing-Review & Editing; SLM: Methodology, Recourses, Supervision; LSK: Formal Analysis, Supervision, Validation; CP: Formal Analysis, Supervision; NS: Data Curation, Formal Analysis.