Periodontitis is a chronic inflammatory process that starts as gingivitis and slowly progresses over time. It is initiated by a dysbiotic biofilm, but it can be modified by multiple factors such as diabetes and cigarette smoking. Due to its chronic nature, inflammation develops unnoticed leading to the loss of periodontal support and eventually, tooth loss. In addition, the effects of periodontitis not only affect the tooth support but accumulate with time. This causes systemic inflammatory changes that have associated periodontitis with systemic conditions such as pregnancy complications, cardiovascular disease, and rheumatoid arthritis. 1

Rheumatoid arthritis (RA) is also a chronic inflammatory disease of the joints with an autoimmune etiology. Pain and swelling of the joints ultimately lead to significant disability and reduced quality of life. RA is known to have a genetic basis, but it is also affected by environmental factors such as hormonal changes, diet, cigarette smoking and other medical conditions that include diabetes.2,3

Periodontitis and RA have similar immunologic features that result in the destruction of periodontal tissues and joints. Some authors have proposed a “two-hit” model where periodontitis represents the first hit and other unknown factors, the second hit that in the long run could influence the development of RA in a susceptible individual. 4 These features include high levels of inflammatory cytokines (IL-1, IL-6, IL-17, TNF) and systemic inflammatory biomarkers such as C-reactive protein (CRP).5 Although they have different etiologic origins, some studies suggest that periodontitis is associated with RA through the inflammatory axis and Porphyromonas gingivalis.6 The specific production of a peptidylarginine deiminase enzyme by P. gingivalis (PPAD), led to the proposal that the citrullination of proteins by PPAD at inflamed periodontal tissues could result in the production of anti-citrullinated protein antibodies (ACPAs) which is a key feature of RA. These pathological mechanisms are thought to be the connecting link between periodontitis and AR.

The systemic changes caused by periodontitis could have important detrimental effects on the health of individuals. Results from the First National Health and Nutrition Examination Survey (NHANES I) and its epidemiological follow-up study (NHEFS) showed that participants with periodontal disease had higher odds of prevalent/incident RA.7 Nevertheless, the odds ratios (ORs) were not statistically significant. In contrast, other studies have suggested that periodontitis is a significant risk factor for RA resulting in controversial results.8,9 Therefore, since the association between periodontitis and RA has not been fully clarified, this study aimed to determine the association between clinical parameters of periodontitis and rheumatoid arthritis.

Table 1 lists the demographic and clinical characteristics of the participants of the study. There were more female participants, and this was specially observed in the RA groups (P<0.05). The mean age of participants was 52.8 years old (95% CI: 50.3–55.3) but periodontitis patients without RA were younger than RA patients (P<0.05). The duration of RA was between 5 and 14 years although patients with periodontitis had higher duration (11.3; 95% CI: 7.7–14.6) than non-periodontitis patients (P<0.05). No differences regarding the frequency of diabetes, hypertension and smoking were observed. In contrast, osteoporosis was more frequent in RA patients (P<0.05). The use of hypertensive medication was similar in all groups. Medications for RA which include DMARDs, NSAIDs and corticosteroids are reported.

In general, participants were comparable regarding the number of teeth present (Table 2). Nonetheless, the periodontitis group presented mainly stage IV periodontitis with worse periodontal condition characterized by increased loss of CAL, presence of periodontal pockets, BoP, and periodontal inflammatory burden than patients with RA and this difference was statistically significant (P<0.05). Participants in the periodontitis with RA group presented less periodontal destruction while the non-periodontitis with RA group presented minimal inflammation on a reduced periodontium. The highest levels of ACPAs and RF were detected in non-periodontitis patients with RA followed by periodontitis patients with RA (P<0.05). The levels of ACPAs and RF were minimal in participants without RA (P<0.05). A similar trend was observed regarding CRP levels, but the difference was not statistically significant. The levels of ACPAs were higher in P. gingivalis negative patients as compared to P. gingivalis positive patients (P<0.05; Table 3).

The association between covariates and RA is presented in Table 4. Covariates such as age, P. gingivalis, diabetes, smoking, osteoporosis, and use of medication were not associated with RA. The model explained 90.3% in variance of the dependent variable (RA).

The correlation between relevant clinical variables and biochemical markers of RA is presented in Table 5. In general, all periodontal variables expressed a negative correlation with ACPAs, RF and CRP (P<0.05). Age presented a slight positive correlation with biochemical markers of AR but was not statistically significant. The duration of RA was positively correlated with ACPAs and specially with RF (P<0.05) but not with CRP. Furthermore, in the multivariate model (Table 6), periodontal clinical variables such as CAL, PD and PIBI presented correlation coefficients were less than 0.50, suggesting that these indicators could only partially explain the correlation between periodontitis and ACPAs and CRP. The detection of P. gingivalis was a negative predictor variable for ACPAs in the model. Conversely, P. gingivalis was a positive predictor variable for RF. The highest statistical weight observed was between periodontal variables and RF. The independent variables described explained between 2 and 6% of the scores obtained for RA biomarkers. Nonetheless, these correlations did not reach statistical significance.

The results of our study produced contrasting results on the association between periodontitis and RA as opposed to previous studies. However, several aspects related to the study of periodontitis and systemic relationships warrant discussion. It is our contention not to refute previous evidence, but to show that perhaps, the association between periodontitis and RA is not clear.

The periodontal condition of periodontitis patients without RA was significantly worse than patients with RA, even with periodontitis. This means higher loss of CAL, deep periodontal pockets, and BoP. In this study, we defined a case of periodontitis as the presence of at least 2 interproximal non-adjacent sites with concurrent loss of CAL, PD≥4mm and BoP. This definition was based on the latest classification of periodontal diseases to differentiate between periodontitis and a reduced periodontium. Previous studies that have reported an association between periodontitis and RA, defined periodontitis on the sole basis of loss of CAL. But loss of CAL represents previous multifactorial exposures and is not exclusive of periodontitis. In fact, the frequency of sites with loss of CAL increases with age and it is now accepted that a healthy/stable periodontium can exist in the presence of loss of CAL and a probing of ≤4mm without BoP.11 For argumentation purposes, we even analyzed the data considering all patients with RA as having periodontitis, but the periodontal parameters were still lower than patients without RA. Although loss of CAL is a sequelae of periodontitis, it is the periodontal pocket (≥4mm with BoP) the sign of current periodontitis. This marks a clear difference between our study and others and the resulting groups for analysis.

We could not find evidence that patients with RA have a more severe periodontitis or loss of CAL than patients without RA as previous reports have suggested.15,16 However, it is worth noting that such studies reported a <0.5mm difference in CAL and PD in subjects with RA as compared to subjects without RA, which is not considered clinically significant. As it occurs in most of the studies, patients with RA are recruited from specialized rheumatology clinics. The same occurs for periodontitis patients from dental clinics. Inclusion parameters were defined to select patients with and without RA and afterwards the periodontal condition was determined. Even though all groups were comparable regarding age and number of teeth, periodontal parameters were more severe in subjects without RA. This may be due to the fact that our patients without RA resulted from the dental clinics of a public university where people with low income are seeking treatment for their dental condition and have limited access to health services. In comparison, RA patients were monitored periodically at their institution and had better access to medical and dental treatments. Hence, it is possible that appropriate access to health services influenced the periodontal condition.

The chronic use of special RA medication could have an impact on periodontal parameters. The results from our study, showed that patients with RA presented less CAL and this difference may be due to the chronic use (>5 years) of DMARDs which was exclusive of RA patients. Studies in animals have shown that the administration of chloroquine reduces periodontal inflammation and bone loss in experimental periodontitis.17 Furthermore, studies in humans revealed that the use of DMARDs and corticosteroids are associated with improvement in CAL in patients with RA and ankylosing spondylitis18,19 DMARDs have potent anti-inflammatory effects and have been recently considered as prospects co-adjuvants in the treatment of periodontitis.20 The study by Heredia-P et al. (2019) showed that DMARD medication in early RA patients was associated with less progression in the loss of CAL after 1-year follow up period suggesting a possible effect on periodontal parameters.21 Similar results were obtained by Rovas et al. (2021) and Punceviciene et al. (2021).16,22 However, the mechanisms of this phenomenon are unknown and could not be analyzed in our study.

We could not demonstrate that periodontitis was associated with RA as previously suggested. A study in 75 patients with RA, 60% presented periodontitis but there was no association between probing depths and disease activity of RA.23 The study by Chen et al. (2013) in a large population-based study reported that the association was still weak after adjusting for confounding variables (OR 1.16; 95% CI 1.12–1.20).24 Another large population-based study revealed no evidence of an increased prevalence of periodontitis in patients with established RA compared to healthy controls.25 But demonstrating the bidirectional relationship between periodontitis and RA is still controversial as the results of the combined evidence does not have a strong correlation between periodontitis and biochemical markers of RA such as ACPAs.23 Wen et al. (2019) found that not all studies report a significant association between periodontal parameters and RA and therefore the results are inconclusive.26 A recent systematic review and meta-analysis showed that there is no substantial effect (<0.5mm) of RA on periodontal probing depth and CAL after combining the available studies.27 In addition, another systematic review and meta-analysis that evaluated whether periodontitis represents a risk for RA, reported higher prevalence of RA among patients with periodontitis as compared to controls (OR 1.97; 95% CI 1.68–2.31).28 Nonetheless, the association is moderate and stems from few high heterogeneity studies (I2=96%). Strong associations are generally accepted with ORs greater than 3 and no study has reliably proven the association between periodontitis and RA. In addition, the detection of P. gingivalis does not support its role in RA in this study. P. gingivalis is an important periodontal pathogen that produces a peptidylarginine deiminase enzyme (PPAD) that induce ACPAs formation and therefore could aggravate RA. Another periodontal pathogen such as Aggregatibacter actinomycetemcomitans has also been recently implicated with the development of RA through the hyperactivation of host peptidylarginine deiminase enzymes and therefore suggesting a more complex interaction between periodontal pathogens and the host.29,30 However, the role of these periodontal pathogens still warrants further studies. In general, the collective evidence gathered to this day is controversial due to consistency, strength, and specificity, leaving the association between periodontitis and RA inconclusive for the moment.

Periodontitis was not associated with RA. Furthermore, there was no correlation between periodontal clinical parameters and biochemical markers of RA.

APL, EBG and JEB contributed to the concept and design of the study. EBG, RAP and JEB contributed to the methodology of the study. AP and JEB contributed to the analysis of data. All authors contributed equally to the drafting of the manuscript.

The study was reviewed and approved by the institutional review board (05-2016).

This study was funded by a grant [CODI 2015-6263] from the Universidad de Antioquia.

Dr. Botero received a grant from Universidad de Antioquia. The rest of the authors have nothing to declare.