Periodontitis in patients with psoriasis: A systematic review and meta‐analysis

Abstract Objective The present study aimed to summarize and update the evidence regarding the association between periodontitis and psoriasis. Methods The present systematic review was conducted under the guidelines of Transparent Reporting of Systematic Reviews and Meta‐Analyses (PRISMA) and was recorded in the PROSPERO database, under registration number CRD42017063799. Three databases (MEDLINE, Embase, and Cochrane Library) were searched up to March 2020. Case–control or cohort studies assessing the association between periodontitis and psoriasis were identified. Quantitative synthesis was conducted with meta‐analysis. Results A total of 13 studies (11 case–control and two cohort studies) assessing the association between periodontitis and psoriasis were included. Of these 13 articles, 9 showed the prevalence of periodontitis or psoriasis. Therefore, meta‐analyses were conducted with data retrieved from the nine studies included. Pooled effect estimate for nine studies showed that patients with periodontitis associated with a higher risk of psoriasis with a pooled OR of 2.87 (95% CI, 1.75–4.69). Conclusions This systematic review demonstrated a positive association between periodontitis and psoriasis; however, a causal relationship cannot be established. Due to the weak evidence, caution should be taken when interpreting the results regarding periodontal parameters. Well‐designed prospective studies are necessary to evaluate interactions between both diseases.

Psoriasis is a systemic inflammatory disorder manifesting in the skin or joints or both, with an estimated global prevalence ranging from 0.5% to 4.6% (Lebwohl, 2003). Although the pathogenesis of psoriasis is still not fully understood, the immune system is now recognized as having a prominent role in the development of this disease (Nickoloff, Qin, & Nestle, 2007).
Since psoriasis and periodontitis have similar pathogenic mechanisms and associated conditions in common, there has been a renewed interest in the research of possible links between psoriasis and periodontitis. The current hypotheses of common etiopathogenesis pathways between psoriasis and periodontitis involve several possible mechanisms such as an amplified inflammatory reaction and T-cell activation (Pischon et al., 2008) and lower concentration of salivary IgA and Iysozyme (Ganzetti et al., 2014); this may result in an increased inflammatory response. Other hypotheses indicated specific characteristics of the salivary microbiota and reduced levels of neutrophil gelatinase-associated lipocalin and transferrin in saliva of patients with psoriasis, which might result in an abnormal host response to the oral microbiota and impact oral homeostasis (Belstrom et al., 2020). Several clinical studies reported increased alveolar bone loss, increased severity of periodontal disease, and a higher number of missing teeth in patients with psoriasis compared with match controls (Egeberg, Mallbris, Gislason, Hansen, & Mrowietz, 2017;Preus, Khanifam, Kolltveit, Mork, & Gjermo, 2010;Su, Huang, Hu, Yu, & Chang, 2017), while a few studies showed no significant relationship (Fadel et al., 2013;Sezer et al., 2016). A systematic review and meta-analysis published in 2017 evaluating five observational studies (2 cohort and 3 case-control studies) reported that patients with periodontitis presented a significantly elevated risk for psoriasis (pooled RR = 1.55, 95% CI: 1.35-1.77; p < .001; Ungprasert, Wijarnpreecha, & Wetter, 2017). This review was built on limited evidence, and the influence of insecure periodontitis diagnostic criteria was not taken into consideration when performing data synthesis. Therefore, the present review aimed to update the meta-analysis with additional studies published through to March 2020 to provide further evidence about the association between periodontitis and the risk of psoriasis.

| Protocol and registration
The present systematic review was recorded in the PROSPERO database, under registration number CRD42017063799, and was conducted under the PRISMA guidelines (Liberati et al., 2009).

| Study design and eligibility criteria
This review aimed to answer the following question: Is there an association between periodontitis and psoriasis demonstrated in observational studies? The PECO strategy for the question was as follows: in patients diagnosed with psoriasis and/or periodontitis (P), the exposure was the presence of periodontitis and/or psoriasis (E), compared with patients with no periodontitis and/or no psoriasis (C).
The outcome was any measure of the prevalence of psoriasis and periodontitis or other forms of periodontal evaluation (O).
The included studies had to meet the following criteria: casecontrol, cross-sectional, or cohort (retrospective or prospective) design; carried out in adult subjects; report the prevalence of psoriasis or periodontitis. We excluded studies lacking raw data after contacting the authors for complete data, as well as review articles, case reports, and articles without the results of interest.

| Search strategy
Articles published before March 2020 were systematically searched through MEDLINE, Embase, and the Cochrane Library. References of included articles and related review articles were manually searched. Eligible articles were independently identified by two reviewers (XZZ and YYS) for possible inclusion in the systematic review and meta-analysis. Full texts were obtained and evaluated independently for articles appearing to meet the inclusion criteria and for those lacking sufficient information in their title and abstract to allow clear decisions. Disagreements were resolved by discussion.

| Data extraction and data synthesis
Data were extracted independently by two reviewers (XZZ and YYS) and included study identifier, study design, number of participants per group, outcome measures, and diagnosis of periodontitis and psoriasis. Disagreements were settled with discussion, or by a third author (SX). For binary outcomes, we first extracted the multivariable-adjusted effect measurements of odds ratio (OR), relative risk (RR), or hazard ratio (HR) and their 95% confidence intervals, if these were not available, we would extract crude effective measurements or use estimated measure based on data reported.

| Quality assessment
The quality assessment for observational studies included in present review was based on the Newcastle-Ottawa Scale (NOS) tool evaluating the following criteria: adequacy of case definition (periodontitis/ psoriasis); representativeness of the cases; selection and definition of controls; comparability of cases and controls with respect to confounding factors; assessment of outcome (periodontitis/psoriasis).

| Statistical analysis
Cochrane test was used to assess heterogeneity across studies. Also, we calculated the I 2 statistic to measure the consistency between trials with values of 25%, 50%, and 75% representing low, moderate, and high degrees of heterogeneity, respectively. Pooled odds ratio (OR) was estimated using fixed effects models when there was low heterogeneity or moderate heterogeneity (I 2 < 50%), and randomized effects models when there was more than moderate heterogeneity (I 2 ≥ 50%). Forest plots were generated for graphical presentations of outcomes. All p-values were two-tailed with the statistical significance set at <.05. All analyses were conducted using R software (R version 3.6.1).

| Sensitivity analyses
To test for the robustness of the results, several sensitivity analyses were performed. Subgroup analysis results were reported according to the study type (case-control or cohort). We also considered subgroup analysis by different levels of diagnostic criteria. Descriptive analyses were conducted about parameters involving in the assessment of periodontitis, such as remaining teeth, missing teeth, probing depth (PD), and clinical attachment level (CAL).

| Meta-analysis
In total, 314,121 participants were included in the main out-

| Analyses of periodontal clinical parameters
In addition to the association between periodontitis and psoriasis, parameters involving in the assessment of periodontitis were also analyzed in our systematic review. Since the number of studies was too small to perform a meta-analysis, only descriptive analyses were conducted. For the tooth loss analysis, two studies (Fadel et   Using CDC/AAP definitions. Severe periodontitis was defined as having two or more interproximal sites with ≥CAL 6 mm (not on the same tooth) and one or more interproximal site(s) with PD ≥ 5 mm. Moderate periodontitis, defined as two or more interproximal sites with CAL ≥ 4 mm (not on the same tooth) or two or more interproximal sites with PD ≥ 5 mm, also not on the same tooth; and mild periodontitis, defined as ≥2 interproximal sites with CAL ≥3 mm and ≥2 interproximal sites with PD ≥4 mm (not on the same tooth) or 1 site with PD ≥5 mm  with substantial heterogeneity, which may mainly be attributed to variability across the included studies, in terms of study design and periodontitis diagnostic criteria. The quality assessment has shown that the quality of most of the case-control studies was lower than that of the cohort studies. Low methodological quality may contribute to the significant heterogeneity among case-control studies. When case-control and cohort studies were separately input in meta-analysis, both the heterogeneity and the pooled ORs changed, indicating that heterogeneity of the study designs could modify the magnitude of the association and restrict its interpretation. In addition, since a large number of risk factors are shared between periodontitis and psoriasis, it is essential to evaluate other independent factors that may confound the real association between the two diseases. The majority of the original studies analyzed in the present meta-analysis have included age and gender (Keller & Lin, 2012;Lazaridou et al., 2013;Mendes et al., 2019;Nakib et al., 2013;Painsi et al., 2017;Skudutyte-Rysstad et al., 2014). The other adjusted confounding factors were smoking (Antal et al., 2014;Lazaridou et al., 2013;Mendes et al., 2019;Nakib et al., 2013;Skudutyte-Rysstad et al., 2014), alcohol intake (Mendes et al., 2019;Nakib et al., 2013), body mass index (Nakib et al., 2013;Skudutyte-Rysstad et al., 2014), medicaments use (Mendes et al., 2019;Skudutyte-Rysstad et al., 2014), and socioeconomic factors (i.e. education, income, or geographical region) (Keller & Lin, 2012;Skudutyte-Rysstad et al., 2014). Two studies have not included any confounding factors (Ganzetti et al., 2015;Preus et al., 2010). This may also result in the heterogeneity of the studies. Federation of Periodontology (Papapanou et al., 2018). Besides, CPI is widely used for periodontal assessment in epidemiological studies with large samples, although it is criticized for the validity of the estimates (Oppermann, Haas, Rosing, & Susin, 2015). In this review, included studies were classified as secure or insecure according to the periodontal diagnosis criteria used. Seven studies which defined periodontitis by clinical/radiographic methods were considered as secure (Antal et al., 2014;Ganzetti et al., 2015;Keller & Lin, 2012;Lazaridou et al., 2013;Mendes et al., 2019;Nakib et al., 2013;Skudutyte-Rysstad et al., 2014); two using self-reported diagnosis method or no description were classified as insecure (Painsi et al., 2017;Preus et al., 2010). When meta-analysis was conducted regarding the diagnosis classification (secure/insecure), a decrease in heterogeneity was observed for secure studies, while heterogeneity was still significant for insecure studies, indicating that studies using secure periodontal diagnosis presented more reliable estimates.

| D ISCUSS I ON
Previous systematic reviews have studied the relationship between periodontitis and psoriasis (Qiao et al., 2019;Ungprasert et al., 2017

| CON CLUS IONS
In conclusion, this systematic review with meta-analysis reveals an association between periodontitis and psoriasis. However, this review was conducted with a relatively limited number of studies, which resulted in limitations for the pooled estimation of periodontal clinical parameters. Due to the relatively low quality of included studies, our review also highlights the necessity of more studies, particularly well-designed prospective studies in different populations with follow-up periods to confirm the evidence.

ACK N OWLED G EM ENTS
The authors declare no potential conflicts of interest. This work was supported by grants from the National Natural Science Foundation of China (81600829 to Y.S).

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/odi.13617.