Associations of Urinary Total Arsenic and Arsenic Species and Periodontitis

Abstract

Aims

Arsenic exposure is a significant global public health concern and has been implicated in endocrine disruption and increased oxidative stress, both of which are crucial pathogenic mechanisms of periodontitis. This study aimed to investigate the association of urinary total arsenic and arsenic species with periodontitis and to further explore the potential mediating roles of sex hormones and oxidative stress indicators.

Methods

Data used in this study were derived from the 2013–2014 National Health and Nutrition Examination Survey (NHANES) in the US population. In all, 1063 participants with complete data were included in this study. Weighted logistic regression analyses were used to evaluate the relationship between urinary arsenic and periodontitis. Mediation analyses were used to explore the effects of potential mediators on these associations.

Results

High concentrations of urinary dimethylarsinic acid (DMA), monomethylarsonic acid (MMA), 2 types of toxic urinary arsenic (TUA2), and 4 types of toxic urinary arsenic (TUA4) were positively related to periodontitis (P < .05). After adjusting for potential confounders, the positive association remained significant (odds ratio, 1.32; 95% confidence interval, 1.01–1.71). Testosterone may partially mediate the relationship between MMA and periodontitis, with mediating effects of 21.78% and 39.73% of the total effect. No significant mediation effect of oxidative stress indicators was found for this relationship.

Conclusions

This study reports a positive association between urinary MMA and periodontitis, and testosterone may mediate this relationship. Our findings serve as a call for action to avoid the deployment of arsenic-containing therapeutic agents as treatment modalities for oral afflictions.

Introduction

Periodontitis is a common chronic inflammatory disease that affected nearly 1.1 billion people worldwide from 1990 to 2019, and it has become an important dental public health problem.¹ It can lead to the destruction of tooth-supporting tissues, resorption of the alveolar bone, and eventual tooth loss as well as masticatory dysfunction, all of which severely affect the nutrition, quality of life, and self-esteem of individuals.² In addition, periodontitis was reported to be associated with various chronic diseases, including diabetes, cardiovascular disease, cancer, and other common systemic diseases.³ Several risk factors are known to be associated with periodontitis, such as lifestyle (smoking, lack of oral hygiene) and dietary habits.⁴ In recent years, the association between exposure to environmental pollutants and the development of periodontitis has also received much attention.⁵ However, there are still many underlying factors that have not been identified. Therefore, actively exploring risk factors for periodontitis is warranted.

Previous studies have shown that higher concentrations of certain trace elements (such as lead and mercury) in the human body can be potentially harmful to oral health.⁶ Arsenic is a naturally occurring trace element that ubiquitously exists in both organic and inorganic forms in the environment,⁷ including soil, air, food, and water. Exposure to arsenic is a major public health concern globally,⁸ and it was estimated that more than 200 million people worldwide might be chronically exposed to arsenic in drinking water at concentrations above the World Health Organization's safety standard of 10 µg/L.⁹ Accumulating evidence has suggested that arsenic toxicity is governed by chemical speciation, which is mainly contributed by inorganic forms.¹⁰ Exposure to arsenic can increase oxidative stress and inflammatory responses and cause endocrine disruption,11, 12, 13 resulting in a variety of chronic diseases.

Given that inflammation and oxidative stress are important pathogenic mechanisms of periodontitis,14, 15, 16 it is reasonable to hypothesise that arsenic exposure may be a potential risk factor for periodontitis. Although arsenic-containing drugs have been used in traditional Chinese medicine as treatments for periodontal disease and oral ulcers, the use of arsenic-containing drugs to treat periodontitis and oral ulcers should now be considered a health risk.^17,18 Nevertheless, until now, few studies have investigated the relationship between arsenic species and periodontitis. Therefore, using the National Health and Nutrition Examination Survey (NHANES) database, this study aimed to explore the associations amongst urinary total arsenic, 7 arsenic species, and periodontitis in American adults older than 30 years and to further explore whether there are potential factors mediating this association.

Materials and methods

Study population

This study was conducted in accordance with the Declaration of Helsinki of 1975. The data were from NHANES, a national survey approved by the National Center for Health Statistics (NCHS) Research Ethics Review Board to represent the civilian, noninstitutionalised population of the US, and all participants provided written informed consent (Continuation of Protocol #2005-2006, 99 https://www.cdc.gov/nchs/nhanes/irba98.htm). In NHANES 2013–2014, a total of 3624 participants received comprehensive oral examinations. After excluding those who were pregnant or lactating or had a history of cancer, 1063 participants with complete data were included in the final analyses (Figure 1). Data are publicly available in NHANES and can be accessed at: https://www.cdc.gov/nchs/nhanes/index.htm.

Fig. 1.

Flowchart for the selection of eligible participants.

Periodontitis assessment

Participants receiving a full-mouth periodontal examination (FMPE) were selected from adults aged 30 years and older with at least 1 natural permanent tooth present. A HU-Friedy periodontal probe colour-coded and graduated in 2-mm increments was used to assess gingival recession and pocket depth at 6 sites per tooth (distal-facial [DF], mid-facial [BF], mesio-facial [MF], distal-lingual [DL], mid-lingual [BL], and mesio-lingual [ML] sites).¹⁹ Periodontitis was defined using the recommended cases from the Centers for Disease Control and Prevention and the American Academy of Periodontology (CDC/AAP),²⁰ combined with clinical attachment loss (CAL) and periodontal probing depth (PPD). Severe periodontitis was defined as having 2 or more interproximal sites with ≥6 mm CAL (not on the same tooth) and 1 or more interproximal site(s) with ≥5 mm PPD. Moderate periodontitis was defined as 2 or more interproximal sites with ≥4 mm CAL (not on the same tooth) or 2 or more interproximal sites with PPD ≥5 mm (also not on the same tooth), and interproximal sites with ≥4 mm PPD (not on the same tooth) or 1 site with ≥5 mm PPD were defined as mild periodontitis.²⁰

Arsenic exposure assessment

Arsenic-Total-Urine (UTAS_H) and Arsenics-Speciated-Urine (UAS_H) datasets in NHANES 2013–2014 were used for analysis in this study. High-performance liquid chromatography (HPLC) was used to isolate the species, followed by ICP-DRC-MS to detect urinary total arsenic and arsenic species concentrations.^21,22 Seven types of urinary arsenic—total arsenic, arsenic acid, arsenous acid, monomethylarsonic acid (MMA), dimethylarsinic acid (DMA), arsenobetaine, and arsenocholine—were measured with this method. Since DMA, MMA, arsenic acid, and arsenous acid are considered to be the more toxic species of arsenic²³ and DMA and MMA are the most detectable, 4 types of toxic urinary arsenic (TUA4; ie, the sum of arsenic acid, arsenous acid, MMA, and DMA) and 2 types of toxic urinary arsenic (TUA2; ie, the sum of MMA and DMA) were further calculated as a complementary method to assess total arsenic exposure. Concentration values below the detection limit were substituted by values of the lower limit of detection (LLOD)/√2. LLOD and detection rates are shown in Supplementary Table 1.

Covariates

In NHANES 2013–2014, potential covariates included age, sex, race, education, marital status, annual family income, body mass index (BMI), smoking, drinking, diabetes, hypertension, last time visited a dentist, mouthwash use, bone loss around teeth, and creatinine. Sex was divided into male and female. Race was divided into non-Hispanic White and other races. Education was divided into less than high school and high school or above. Marital status was divided into never married, married, and unmarried but have/had partner. Annual family income was divided into less than $25,000 and $25,000 and over. BMI, computed as weight in kilograms divided by height in meters squared, kg/m², was divided into normal (<25 kg/m²) and overweight or obese (≥25 kg/m²). Smoking was divided into never smoking (<100 cigarettes in life) and former or current smoking (≥100 cigarettes in life). Last time visited a dentist was divided into 1 year or less and more than 1 year. Diabetes, hypertension, mouthwash use, and bone loss around teeth were all obtained through interviews with structured questionnaires, divided into yes or no.

Statistical analyses

To achieve representativeness for the US population, all data analyses were conducted using the appropriate sample weights (wtsa2yr, provided by the NHANES 2013–2014 cycle), according to NHANES guidelines (https://wwwn.cdc.gov/nchs/nhanes/tutorials/Weighting.aspx). To describe the differences in baseline characteristics between participants with and without periodontitis, weighted Pearson chi-square test was used to compare the categorical variables that were represented by numerical and frequency distributions, and the weighted Mann–Whitney U test was used to compare the nonnormally distributed continuous variables that were represented by medians and interquartile ranges. Weighted binary logistic regression was used to evaluate the association between urinary total arsenic, 7 arsenic species, and periodontitis. Model 1 did not include any adjustment variables and model 2 adjusted for age, sex, education, hypertension, diabetes and BMI, smoking, last time visited a dentist, mouthwash use, and bone loss around teeth. Stratified analyses were further performed to explore the heterogeneity of associations in different subgroups, including basic characteristics (age, sex, smoking, and BMI) and some chronic diseases (hypertension, diabetes, and chronic bronchitis). Restricted cubic splines (RCS) were used to accommodate the nonlinear relationship amongst MMA, DMA, arsenous acid, and periodontitis. Furthermore, to explore the potential effects of hormone interference (as indicated by sex hormone-binding globulin [SHGB], unconjugated [free and protein-bound] testosterone [TT], and estradiol [E2, the most prevalent of the estrogens with estrogenic activity]) and oxidative stress (as indicated by total bilirubin, albumin, and iron²⁴), mediation analyses were performed by using the R package “mediation” with 5000 bootstrap samples. The following criteria were defined as having mediating effects: (1) the total effect was significant, (2) the indirect effect was significant, and (3) the proportion mediated was in the positive direction.²⁵ E values were calculated as a sensitivity analysis to assess the potential residual unmeasured confounding.²⁶ All statistical analyses were performed in R (version 4.2.1). The level of significance was set at a P value < .05.

Results

The baseline characteristics of 1063 participants in NHANES 2013–2014 with an average age of 49.68 are presented in Table 1, of whom 51.55% were male and 48.45% were female. Education level, BMI, and bone loss around teeth were significantly different between the nonperiodontitis and periodontitis groups (P < .05). Individuals with periodontitis were more likely to be smokers, have a higher proportion of diabetes or hypertension, and visit the dentist much less frequently than individuals without periodontitis. No significant differences in race, marital status, annual family income, mouthwash use, or urine creatinine were seen between these 2 groups (P > .05).

Table 1.

Characteristics of the study population in the National Health and Nutrition Examination Survey (NHANES) 2013–2014.

Characteristics	Total N = 1063	Nonperiodontitis n = 594	Periodontitis n = 469	P
Age, y	48.00 (39.00, 59.00)	46.00 (38.00, 56.00)	52.00 (42.00, 63.00)	.003**
Sex				.002**
Female	515 (48.45)	338 (54.81)	177 (38.63)
Male	548 (51.55)	256 (45.19)	292 (61.37)
Race				.057
Non-Hispanic White	442 (41.58)	267 (68.62)	175 (59.99)
Other	621 (58.42)	327 (31.38)	294 (40.01)
Education				<.001***
Less than high school	226 (21.26)	83 (9.39)	143 (23.24)
High school or above	837 (78.74)	511 (90.61)	326 (76.76)
Marital				.084
Never married	147 (13.83)	78 (10.72)	69 (15.19)
Married	615 (57.86)	359 (64.81)	256 (56.16)
Unmarried but have/had partner	301 (28.32)	157 (24.47)	144 (28.65)
Annual family income				.053
<$25,000	547 (53.733)	302 (53.454)	245 (43.491)
≥$25,000	471 (46.267)	264 (46.546)	207 (56.509)
Body mass index				.026*
Normal	299 (28.23)	179 (29.93)	120 (20.96)
Overweight or obese	760 (71.77)	414 (70.07)	346 (79.04)
Smoking				<.001***
Never	597 (56.16)	385 (65.77)	212 (42.86)
Former or current	466 (43.84)	209 (34.23)	257 (57.14)
Drinking				.836
No	145 (14.573)	80 (11.156)	65 (10.610)
Yes	850 (85.427)	481 (88.844)	369 (89.390)
Diabetes				.003**
No	934 (87.95)	533 (92.45)	401 (85.52)
Yes	128 (12.05)	60 (7.55)	68 (14.48)
Hypertension				<.001***
No	609 (57.29)	388 (67.13)	221 (48.72)
Yes	454 (42.71)	206 (32.87)	248 (51.28)
Last time visit a dentist				<.001***
≤1 y	635 (59.737)	413 (74.586)	222 (46.654)
>1 y	428 (40.263)	181 (25.414)	247 (53.346)
Mouthwash use				.799
No	445 (41.863)	246 (42.554)	199 (43.488)
Yes	618 (58.137)	348 (57.446)	270 (56.512)
Bone loss around teeth				.024*
No	929 (87.807)	530 (91.735)	399 (84.584)
Yes	129 (12.193)	61 (8.625)	68 (15.416)
Creatinine	94.00 (51.00, 150.00)	87.00 (48.00, 149.00)	106.00 (61.00, 153.00)	.093

Continuous variables are represented by median (25% and 75% quantile) using Mann–Whitney U test; categorical variables are represented by n (%) using Pearson's chi-square test.

^⁎P < .05.

^⁎⁎P < .01.

^⁎⁎⁎P < .001.

As shown in Table 2, high concentrations of urinary DMA, MMA, TUA2, and TUA4 were positively associated with periodontitis (P < .05). Amongst them, MMA exhibited the highest magnitude of association with periodontitis (odds ratio (OR), 1.31; 95% confidence interval (CI), 1.05–1.63). Although the positive associations for other arsenic species were attenuated after adjusting for potential confounders, MMA still showed a significant association with periodontitis (OR, 1.32; 95% CI, 1.01–1.71).

Table 2.

Association between urinary total arsenic, arsenic species, and periodontitis.

Variablesa	Model 1b		Model 2c
	OR (95% CI)	P	OR (95% CI)	P
Urinary total arsenic	1.00 (0.99–1.00)	.600	1.00 (0.99–1.01)	.691
Arsenous acid	1.33 (0.99–1.80)	.059	1.44 (0.86–2.40)	.122
Arsenic acid	0.93 (0.61–1.42)	.721	1.14 (0.66–1.98)	.546
DMA	1.02 (1.00–1.04)	.029*	1.01 (0.99–1.04)	.168
MMA	1.31 (1.05–1.63)	.021*	1.32 (1.01–1.71)	.043*
Arsenobetaine	1.00 (0.99–1.01)	.479	1.00 (0.98–1.01)	.595
Arsenocholine	0.96 (0.82–1.12)	.579	0.92 (0.76–1.10)	.261
TUA2	1.02 (1.00–1.04)	.022*	1.02 (0.99–1.04)	.129
TUA4	1.02 (1.00–1.04)	.017*	1.02 (0.99–1.04)	.110

^⁎P < .05.

^aAll variables were treated as continuous variables, and ORs were presented per 1 unit increase in each arsenic indicator.

^bModel 1: Unadjusted.

^cModel 2: Adjusted for age, sex, education, hypertension, diabetes, body mass index, smoking, last time visited a dentist, mouthwash use, and bone loss around teeth.

DMA, dimethylarsinic acid; MMA, monomethylarsonic acid; TUA2, 2 types of toxic urinary arsenic (the combination of DMA and MMA); TUA4, 4 types of toxic urinary arsenic (the combination of DMA, MMA, arsenous acid, and arsenic acid).

When stratified by sex, age, smoking, and BMI, as presented in Figure 2, a stronger significant association was observed between MMA and periodontitis amongst never smokers (OR, 1.75; 95% CI, 1.26–2.43). A similar significant association persisted after adjustment for confounders (OR, 2.10; 95% CI, 1.31–3.36). Of note, there was a significant interaction between MMA and smoking (P for interaction < .001). DMA and arsenous acids were also found to be associated with periodontitis amongst never smokers in both unadjusted and adjusted models (Supplementary Figure 1). Consistent interaction patterns were also observed (all P’s for interaction < .05). Additionally, when stratified by several chronic diseases (Supplementary Table 2), the association between MMA and periodontitis was significant only in the individuals without diabetes (P for interaction < .05).

Fig. 2.

Forest plot for the association between urinary monomethylarsonic acid (MMA) and periodontitis, stratified by sex, age, smoking, and body mass index (BMI). ^aAdjusted for age, sex, education, hypertension, diabetes, BMI, smoking, last time visited a dentist, mouthwash use, and bone loss around teeth.

Although RCS did not reveal a nonlinear relationship amongst MMA, DMA, arsenous acid, and periodontitis, the risk of periodontitis seemed to increase as the exposure levels increased (Figure 3A and Supplementary Figure 2). Of note, when results were further stratified by smoking, RCS displayed a nonlinear relationship between MMA concentrations and periodontitis amongst smokers (P for nonlinear = .043). Similar positive dose-response relationships were observed amongst female participants, male participants, and never smokers, although the nonlinear pattern was less apparent (P for nonlinear > .05; Figure 3B and Figure 3C).

Fig. 3.

Restricted cubic splines of the association between urinary monomethylarsonic acid (MMA) concentrations and periodontitis. A, For all participants. B, Stratified by sex. C, Stratified by smoking status.

To further explore whether hormone interference or oxidative stress has a mediating role in the association between MMA and periodontitis, mediation models for the potential mediators (testosterone, estrogens, SHBG, total bilirubin, albumin, and iron) were assessed. As depicted in Figure 4 and Table 3, testosterone had a significant mediating effect on the association between MMA and periodontitis, whether as continuous or categorical variables (both P for indirect effect < .001), and the proportion of the mediating effect was 21.78% and 39.73%, respectively. However, no mediating role was observed for any oxidative stress indicators in the association between MMA and periodontitis (Table 3).

Fig. 4.

The mediation effect of testosterone on the association between urinary monomethylarsonic acid (MMA) and periodontitis. A, Testosterone as a continuous variable. B, Testosterone as a categorical variable.

Table 3.

Mediation analyses on the association between urinary MMA and periodontitis.

Mediator	Indirect (ab)		Direct (c)		Total (c)		Prop. Mediated (%)
	β (95% CI)	P	β(95% CI)	P	β(95% CI)	P
As continuous variable
Endocrine disruption
Testosterone	0.013 (0.005 to 0.030)	<.001	0.047 (−0.003 to 0.100)	.064	0.062 (0.016 to 0.114)	.010	21.78
Estrogens	0.002 (−0.004 to 0.009)	.364	0.062 (0.014 to 0.116)	.018	0.064 (0.016 to 0.120)	.010	3.87
Sex hormone-binding globulin	0.002 (−0.002 to 0.006)	.327	0.064 (0.014 to 0.117)	.010	0.065 (0.017 to 0.118)	.010	2.45
Oxidative stress
Total bilirubin	−0.004 (−0.014 to 0.002)	.176	0.072 (0.023 to 0.124)	.004	0.067 (0.019 to 0.120)	.007	6.28
Albumin	−0.002 (−0.008 to 0.001)	.169	0.070 (0.023 to 0.123)	.006	0.068 (0.019 to 0.120)	.007	3.48
As categorical variable
Endocrine disruption
Testosterone	0.029 (0.009 to 0.046)	<.001	0.044 (−0.005 to 0.096)	.077	0.073 (0.020 to 0.121)	.010	39.73
Estrogens	0.002 (−0.009 to 0.009)	.945	0.067 (0.017 to 0.119)	.009	0.069 (0.016 to 0.120)	.010	2.90
Sex hormone-binding globulin	0.000 (−0.004 to 0.003)	.947	0.066 (0.016 to 0.119)	.010	0.065 (0.017 to 0.119)	.010	0.00
Oxidative stress
Total bilirubin	−0.005 (−0.014 to 0.000)	.052	0.072 (0.024 to 0.124)	.004	0.067 (0.019 to 0.119)	.008	7.46
Albumin	−0.005 (−0.013 to 0.001)	.141	0.071 (0.023 to 0.123)	.005	0.066 (0.019 to 0.119)	.007	7.58

Discussion

In this large cross-sectional study based on NHANES (2013–2014), we found that a high urinary concentration of organic arsenic MMA was associated with periodontitis, and the relationship was even stronger amongst never smokers. There was a significant interaction between MMA and smoking. Of note, testosterone may mediate the association between MMA and periodontitis.

Most inorganic arsenic entering the body is metabolised into MMA and DMA by methylation and exits the body through urination.²⁷ MMA is the main component of arsenic metabolites and is one of the most toxic forms amongst arsenical species in the human body,^27,28 whilst several types of organic arsenic, such as arsenobetaine and arsenocholine, have not been found to induce serious harm in humans.²⁹ In this study, a significant association between MMA and periodontitis was found. However, the mechanism of this association has not been elucidated. Although we did not observe any significant mediating effect of oxidative stress–related indicators in this study, previous studies have shown that arsenic can increase the production of reactive oxygen species (ROS)³⁰ and subsequently induce widespread damage, such as inflammatory responses, alterations in gene expression, and impaired nitric oxide signaling.³¹ ROS are known for their antimicrobial effects. The effect has 2 sides: ROS can help kill invading pathogenic microorganisms when present in small amounts, but they can become cytotoxic to host cells when overabundant and reduce antioxidant capacity, leading to oxidative stress and resulting in the destruction of periodontal tissues.^15,16 Arsenic exposure can lead to an increase in proinflammatory cytokines such as interleukin (IL)-6, IL-10, and tumour necrosis factor α, which are closely related to the inflammatory response.³² Given that these proinflammatory mediators have been proven to be key players in the destruction of periodontal tissues,³³ they may indirectly explain the association we observed.

Another underlying mechanism of the relationship between MMA and periodontitis might be hormone interference. Arsenic is an endocrine disruptor that can interfere with the synthesis, metabolism, and transport of hormones in the body³⁴ and has adverse effects on humans by hindering the binding of hormones and their receptors.¹³ Our study found that total testosterone has a mediating role in the association between MMA and periodontitis. A recent study also indicated that MMA was positively associated with total testosterone in children and adolescents in the US population.¹³ A similar positive correlation amongst urinary arsenic, including MMA, and total testosterone was reported in a Chinese male cohort.³⁵ It seems to have been widely accepted that some endocrine disruptors, especially testosterone, affect oral tissues.36, 37, 38 A previous study found that anabolic androgen steroid users are more likely to have severe periodontitis.³⁹ Therefore, interfering with androgens may be a plausible explanation for the positive association between MMA and periodontitis, but this postulation still needs to be confirmed by further research. However, we did not observe a mediating role for estradiol. There is still controversy regarding the relationship between arsenic and estrogens. One population-based study revealed a positive correlation amongst total arsenic, arsenic metabolites, and estradiol,⁴⁰ whilst another experimental study demonstrated a significant reduction in estradiol levels following arsenic exposure.⁴¹ Two US population-based studies explored the link between sex hormones and periodontitis, with neither study finding an association between estradiol and periodontitis.^38,42 These previous studies may provide a possible explanation for the absence of evidence regarding the potential mediating role of estrogen in our study. Further experimental studies are required to elucidate the specific mechanisms involved.

Notably, a more pronounced positive association was found between urinary arsenic levels and periodontitis amongst never smokers or individuals without diabetes, but not amongst smokers and those with diabetes. This finding was unexpected, and until now, there was no direct evidence to explain the heterogeneous associations. Given that previous studies have shown that smoking and diabetes are important risk factors for periodontitis,43, 44, 45, 46, 47 we speculate that these strong risk factors may have led to the statistical association between arsenic exposure and periodontitis being obscured. Further experiments, such as animal and in vitro cell experiments, are necessary to clarify the mechanism. On the other hand, few studies have focused on risk factors for periodontitis in nonsmokers and individuals without diabetes. Our findings may also serve as a reminder to pay more attention to arsenic exposure in these groups.

To our knowledge, the present study is the first to investigate the association between total arsenic and arsenic species and periodontitis in a nationally representative survey with a large sample. Nonetheless, there are some limitations to the study. First, because of the cross-sectional study design, potential reverse causality cannot be excluded. Therefore, a prospective cohort study is needed for further validation. Second, our study was unable to assess the levels of arsenic exposure because NHANES only measured the levels of urinary arsenic. Third, smoking was not included in the adjusted model because arsenic exposure in tobacco may lead to an obscured association. Fourth, we only adjusted for some common confounders and cannot exclude the possibility of residual confounding. However, the E value calculated for MMA (E value, 1.55; lower limit of 95% CI, 1.08) indicated that unmeasured confounding factors were unlikely to change the observed association in this study. Finally, in most of chronic disease strata, no significant associations were found between arsenic exposure and periodontitis. Given the intricate biological interplay amongst arsenic, periodontitis, and chronic diseases, a broader biological approach is required to unravel the underlying mechanisms in the future.

Conclusions

This study provides preliminary evidence that urinary MMA was positively associated with periodontitis, and testosterone may mediate the observed association. The specific mechanism still needs to be elucidated by further investigations. In light of the known hazards associated with the utilisation of arsenic-containing therapeutic agents in the management of oral diseases,^17,18 our findings serve as a call to action to avoid the deployment of such compounds as treatment modalities for oral afflictions.

Author contributions

FC and FY conceived the study and provided overall guidance. FC, YW, and JW prepared the first draft and finalised the manuscript based on comments from all other authors. YW and FY had major roles in statistical analysis. All other authors contributed to the analysis and reviewed the manuscript. Han Yang and Jing Wang contributed equally.

Conflict of interest

None disclosed.

ASSOCIATIONS BETWEEN ARSENIC AND PERIODONTIS