Effect of periodontal therapy on COPD outcomes: a systematic review

This review was registered in the center for Reviews and Dissemination, University of York, (protocol number: UK #CRD42020158481). The inclusion/exclusion criteria were specified a priori and followed the PICO criteria and Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocol (PRISMA-P) guidelines [18].

The following databases were searched by one author (IA) using a combination of keywords for COPD, periodontitis and periodontal therapy: PubMed, Scopus, Virtual Health Library, ScienceDirect, Wiley Online Library, Web of Science, ProQuest Dissertation and Theses Global and Google Scholar (Table 1). Search was limited to papers involving human subjects and published in the English language between 01/01/2000 to 31/03/2020. The original search was created in PubMed and adjusted thereafter to the other databases.

Randomized controlled clinical studies, cross-sectional, cohort and case–control studies reporting data on COPD patients with periodontal disease and its treatment were considered eligible for inclusion. Both prospective and retrospective studies were included and assessed separately. Case reports, case series, and systematic reviews were excluded.

COPD diagnosis was based on Global Initiative for Chronic Obstructive Lung Disease report[1], that is a post bronchodilator ratio of forced expiratory volume in 1st second (FEV₁) to forced vital capacity (FVC) (FEV₁/FVC < 70%)[1] or on previous track records and physicians’ reports. For the diagnosis of periodontitis, we used the criteria employed by the authors of the individual papers as long as they were validated.

Interventions included non-surgical or surgical periodontal therapy, which was established by hand instruments or ultrasonic devices. The effectiveness of periodontal treatment was assessed according to the results of before-after periodontal treatment studies and from reports which compared COPD patients who received periodontal treatment against those who received no treatment.

The primary outcome measures included the effect of non-surgical, surgical and supportive periodontal therapy on the frequency of COPD exacerbations, lung function (FEV₁, FEV₁/FVC) and quality of life. The secondary outcome measures included: oral hygiene and periodontal indices, radiographic alveolar bone loss, levels of inflammatory biomarkers, periodontal bacterial burden, comorbidities, hospitalizations and mortality.

Literature search results were uploaded to Distiller Systematic Review (DSR) software, an Internet based software program that facilitates collaboration among reviewers during the study selection process. After duplicates’ removal, the team developed screening questions and forms for level 1 and 2 assessments based on the eligibility criteria. Citation abstracts and screening questions were uploaded to DSR.

Studies were screened by two independent reviewers (IA, AV) who assessed the titles and abstracts of each article. Disagreements were resolved by discussion, but if decision was inconclusive, the assessment of the contentious study continued to full text screening level. At that stage, the same authors carried out the independent appraisal of each article, reasons for exclusion were recorded and in case of disagreement a third author adjudicated unresolved quarrel.

Data were then extracted from each article using a custom data extraction sheet and included the following: principal author, publication year, country, study design, study population, periodontal variables, periodontal therapy techniques, COPD assessment methods and main results.

The risk of bias of randomized clinical trials was assessed with Cochrane risk of bias tool (RoB2), which assesses the following five domains: bias arising from randomization process, bias due to deviation from intervention, bias due to missing outcome data, bias in measurement of the outcome and bias in selection of reported results. Possible risk of bias judgments were “low risk of bias”, “some concerns” and “high risk of bias” [19].

Risk of bias assessment was performed independently by two review authors (IA and AV) and any disagreement was resolved by discussion. The risk of bias of non-randomized studies was assessed with Cochrane risk of bias tool for non-randomized studies of intervention (ROBINS-I tool) which assesses the following 7 domains of bias: (1) confounding (2) selection of participants into the study (3) classification of interventions (4) deviations from intended interventions (5) missing data (6) measurement of outcomes and (7) selection of the reported result. Possible risk of bias judgments were: “low risk of bias”, “moderate risk of bias”, “serious risk of bias”, “critical risk of bias” and “no information” [20].

If a sufficient number of trials were identified (n ≥ 10), analyses were planned to identify reporting biases (“small-study effects” and/or “publication bias”), through the inspection of a contour-enhanced funnel plot and through respective statistical tests for funnel plot asymmetry and identification of missing studies [21].

The overall quality of evidence (confidence in effect estimates) was judged using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) working group methodology [22], according to a recent guidance on combining randomized with nonrandomized studies [23]. A table with a summary of findings was constructed using the format by Carrasco‐Labra et al. [24] GRADEpro software [25] was used to assess the quality of evidence across the domains of risk of bias, consistency, directness, precision and publication bias. Quality was adjudicated as “high” (further research is very unlikely to change our confidence in the effect estimate), “moderate” (further research is likely to have an important impact on our confidence in the effect estimate and may change the estimate), “low” (further research is very likely to have an important impact on our confidence in the effect estimate and is likely to change the estimate) and “very low” (very uncertain about the estimate effect).

A flow diagram (Fig. 1) depicts the process that we employed in order to identify relevant articles. Out of 1442 initially retrieved studies, 7 studies were considered eligible for inclusion in this systematic review. Table 2 demonstrates the major characteristics of the 7 studies. Three of the included studies were RCTs [27‐29], 2 were case–control studies [30, 31], 1 was cross-sectional study [32] and 1 was a before-after treatment cohort study [33]. These studies were conducted in India [27, 33], Taiwan [30], China [29, 32], Turkey [31] and USA [28] and took place at university and hospital settings. Due to both small sample sizes from most included studies, but most importantly due to differences in the study design, types of interventions and comparisons as well as their methods of reporting results and because of the low quality of the eligible articles, a meta-analysis could not be performed. Hence, the respective results of each study are presented without being statistically processed.

The diagnosis of COPD was based on GOLD criteria [27, 29, 31‐33], medical records [28] and ICD-9 classification [30]. The diagnosis of COPD exacerbations was based on the GOLD criteria [29, 32], ICD-9 classification [30] and symptom worsening requiring treatment adjustment and/or hospitalization [31]. COPD exacerbations were reported by the patients [32], by the investigators [29] or were reported by the patient and confirmed by the investigator [31]; while 2 studies have also provided data on COPD exacerbations prior to periodontal interventions based on patients’ recall [29, 31].

The diagnosis of periodontitis was based on a mean attachment loss (MAL) ≥ 1.5 mm[28], a community periodontal index (CPI) > 3 and a loss of attachment (LOA) > 1[33], the presence of at least one site with a pocket probing depth (PPD) > 3 mm and periodontal clinical attachment loss (CAL) > 3 mm[29], the ICD-9 classification[30], the American Academy of Periodontology Workshop 1999 guidelines[31, 34] and on oral/periodontal health indices [27, 32]. Quality of life was assessed with the Saint George Respiratory Questionnaire (SGRQ) in 2 studies [27, 28]. The effect of periodontal treatment on lung function variables was investigated in 2 other studies [29, 33].

Periodontal therapy was established through scaling and root planning using hand instruments [27], supragingival scaling, hygiene instructions and preprocedural antibiotics [33], subgingival curettage, root planning and periodontal flap surgery (if needed) [30], oral hygiene instructions, supra- and subgingival scaling using hand instruments and ultrasonic devices [28, 29, 31]. In another study, information about the frequency of previous supragingival scaling procedures was obtained by interview [32].

Risk of bias for RCTs was judged as “some concerns” in one trial[28] and “high” in two trials [27, 29]. With regard to the rest of the studies, serious methodological limitations were found for two studies [32, 33] and moderate for another two [30, 31]. Risk of bias summary plots are depicted in Figs. 2 and 3.

Two studies [27, 28] evaluated the effect of periodontal therapy on quality of life. Das et al. [28] randomized 35 COPD patients (diagnosed according to the GOLD spirometric criteria) to receive either periodontal treatment (treatment group, n = 18) or no treatment (control group, n = 17). Interventions included full-mouth scaling and root planning with hand instruments as well as oral hygiene instructions, while the effect of periodontal therapy on quality of life was assessed with the SGRQ score. All patients were reassessed at 1-year post-treatment for significant changes in periodontal and oral parameters and in SGRQ scores. With respect to the intervention group in this study significant improvements were observed in the Oral Hygiene Index-Simplified (OHI-S), Gingival Index (GI) and CAL, but not in PPD, while no differences in any of the appraised parameters were noted in the control group. The SGRQ “activity” subscore significantly improved in the treatment group post treatment (53.68 ± 16.37 vs. 38.20 ± 13.18; p = 0.005), but no differences in any of the SGRQ subscores were observed in the control group. At the end of the 1-year follow-up period, no differences were noted in the SGRQ subscores between the two groups with the exception of the “activity” subscore which was significantly better in the intervention group (38.20 ± 13.18 vs. 56.48 ± 18.31; p < 0.01).

Agado et al. [28] randomized 30 COPD patients (recruited from medical databases) with chronic periodontitis (MAL ≥ 1.5 mm) into periodontal therapy with ultrasonic instrumentation (n = 10), periodontal therapy with hand instruments (n = 10) or no treatment (n = 10) for a follow-up period of 4–6 weeks; periodontal treatment was completed when all clinically detectable deposits had been removed. The effect of periodontal therapy on quality of life was evaluated with the American English modified SGRQ score (SGRQ-A), while patients graded their perception of health status on a 5-point Likert scale from “very poor” to “very good”. In addition, all participants provided a yes/no response to a customary made 7-item “illness questionnaire” which interrogated respiratory or other illness within the previous 4 weeks, physician visits, antibiotic use, use of respiratory medications, respiratory problems after dental care and avoidance of dental care due to respiratory ailments. All questionnaires were issued by an investigator blinded to the type of intervention. At the end of the follow-up period no significant difference was noted between the 3 groups in SGRQ-A total scores, self-assessment of health scores and “illness questionnaire” and no improvement between pre- and post-test. However, each group showed a significant improvement in the SGRQ-A “activity” subscore but with no interactions between groups.

Two studies [29, 33] evaluated the effect of periodontal therapy on lung function. Madalli et al.²⁹ conducted a before-after treatment prospective cohort study of 30 COPD patients with periodontal disease to investigate the effect of periodontal treatment on FEV₁/FVC variables. COPD diagnosis was based on the GOLD spirometric criteria and the diagnostic criteria for periodontal disease included a CPI score ≥ 3 and a LOA score ≥ 1. Interventions included supragingival scaling and oral hygiene instructions and all patients were re-assessed after 3–5 months. This study observed no significant changes in CPI and LOA scores, while no significant improvement in FEV₁/FVC was also noted (48.08 ± 12.01 vs. 51.25 ± 12.29; p = NS). Zhou et al. [29] randomized 60 moderate-to-severe COPD patients (diagnosed according to the GOLD criteria) into scaling, root planning and maintenance care (SRP group, n = 20), supragingival scaling alone and maintenance supragingival scaling (scaling group, n = 20) or no treatment (control group, n = 20). Periodontal disease was considered based on the presence of at least one site with a PPD > 3 mm and a CAL > 3 mm. All patients were re-assessed for changes in periodontal variables, namely PPD, CAL, bleeding index (BI) and plaque index (PI) at 6 months, 1 year and 2 years, while lung function variables (FEV₁, FEV₁/FVC) were re-evaluated at 1 year and 2 years. This trial observed significant improvements in both treatment groups at 6 months, 1 year and 2 years for all the appraised periodontal parameters, whereas no improvements were noted in the control group; all periodontal variables were worse in the control group as compared to each of the treatment groups at any of the investigated time points. With respect to the lung function variables, when compared to baseline, the control group exhibited a significant decrease in FEV₁ (% predicted) at 2 years (56.3 ± 16.4 vs. 51.6 ± 18.4, p < 0.05) and in FEV₁/FVC at 1 and 2 years (0.55 ± 0.11 vs. 0.54 ± 0.11 vs. 0.53 ± 0.11, respectively; p < 0.05). When compared to the control group, the SRP group showed significantly improved levels of FEV₁ (% predicted) and FEV₁/FVC at 1 year (55.9 ± 16 vs. 53.6 ± 18.7 and 0.59 ± 0.09 vs. 0.54 ± 0.11, respectively; p < 0.05) and at 2 years (57.1 ± 19 vs. 51.6 ± 18.4 and 0.57 ± 0.10 vs. 0.53 ± 0.11, respectively; p < 0.05). Likewise, when the scaling group was compared to controls, it exhibited a significantly higher FEV₁ (% predicted) at 1 year (59.6 ± 17.1 vs. 53.6 ± 18.7, p < 0.005) and FEV₁/FVC at 2 years (0.56 ± 0.11 vs. 0.53 ± 0.11, p < 0.05). No differences in lung function were observed between the two treatment groups at any of the abovementioned time points.

Four studies [29‐32] evaluated the effect of periodontal therapy on the frequency of COPD exacerbations. In the previously mentioned RCT by Zhou et al., COPD exacerbations post-randomization were defined according to the GOLD criteria and their frequencies were assessed at 6 months, 1 year and 2 years, while their frequencies during the last year prior to randomization were self-reported by the patients; patients with ≥ 2 events /year were defined as “frequent” exacerbators. At 2 years, the SPR and the scaling groups showed a lower proportion of “frequent” exacerbations as compared to controls (30% vs. 15.8% vs. 66.7%, respectively; p < 0.004). In a multivariate logistic regression model, the calculated OR for “frequent” exacerbations was 0.29 (95% CI 0.10–0.84; p = 0.02) for the SRP group and 0.004 (95% CI 0.003–0.64; p = 0.02) for the scaling group after adjusting for age, gender, body mass index (BMI) and baseline frequency of exacerbations. No differences were noted in the frequency of COPD exacerbations between the two treatment groups.

Shen et al. [30] conducted a population-based, retrospective case–control study based on the Taiwan National Health Insurance claims data. COPD and periodontal disease diagnosis, periodontal interventions and adverse respiratory effects were identified based on the database’s ICD-9 classification. The study population consisted of 5562 COPD individuals that received periodontal therapy and 5562 propensity-matched controls. All subjects were followed up to a maximum of 5 years or until death, occurrence of an adverse respiratory event or withdrawal from the insurance system. Periodontal interventions included supragingival scaling and root planning as well as periodontal flap surgery, depending on the severity of the disease. The primary outcomes included emergency room (ER) visit and hospitalization rates following acute COPD exacerbation, pneumonia or acute respiratory failure. Overall, all types of adverse respiratory events were less frequent in the treatment group as opposed to the control group. After adjusting for confounders, including age, gender, residential area, occupation, income and comorbidities, the ER visit rate per 100 person-years for COPD exacerbation was 2.54 for the treatment group versus 2.88 for the control group, for an adjusted incidence rate ratio (IRR) of 0.86 (95% CI 0.78–0.94; p < 0.001).

Kucukcoskun et al. [31] conducted a prospective case–control study consisting of 40 patients with COPD, diagnosed according to the GOLD criteria with a history of ≥ 1 exacerbation during the previous year and with moderate-to-severe chronic periodontitis based on the 1999 American Academy of Periodontology Workshop guidelines. COPD exacerbation was defined as a worsening in the baseline respiratory symptoms lasting at least 2 days and necessitating oral steroids and antibiotics (moderate exacerbation) or hospitalization (severe exacerbation). Patients were allotted to periodontal treatment (oral hygiene, full mouth scaling and root planning with hand instruments and ultrasonic devices completed over 3 visits, n = 20) or no treatment (n = 20) according to their access to periodontal services. Exacerbations were self-reported by the patients and confirmed by the investigator, while the number of exacerbations and hospitalizations during the last year before enrollment was also registered. Periodontal procedures were performed by a dentist blind to the study design and COPD severity and all patients were re-evaluated at 6 and 12 months. With respect to the periodontal parameters, when compared to baseline, the treatment group exhibited significant improvements in BP, GI, PPD and CAL at 6 months; at 12 months, this improvement remained significant only for CAL. On the other hand, in the control group most of the periodontal parameters showed significant deterioration at 6 (GI, PPD, PI and CAL) and 12 months (GI, PPD, and CAL). The frequency of exacerbations per patient-year was 1.95 events in the treatment group versus 3.25 events in the control group (p = 0.01). After 1 year of follow-up, exacerbation frequency, was significantly reduced in the treatment group (3 ± 1.83 vs. 1.95 ± 1.46; p = 0.01) but remained unchanged in the control group (3.5 ± 4.62 vs. 3.25 ± 3.35; p = NS).

Liu et al. [32] conducted a cross-sectional study investigating the role of periodontal health on the frequency of COPD exacerbations. The diagnosis and definition of COPD exacerbation were based on GOLD criteria. Overall, 392 COPD patient were assessed for the presence and severity of periodontal disease and were interviewed about oral health habits, history of dental visits and frequency of supragingival scaling procedures (≥ 1 time/year vs. < 1 time/year) during the last 12 months. Periodontal examination was conducted by an investigator blinded to COPD severity. Patients also provided information about the frequency of exacerbations during the previous 12 months with patients with < 2 events classified as “infrequent” exacerbators (n = 209), whereas those with ≥ 2 events considered “frequent” exacerbators (n = 183). In a multivariate logistic regression model adjusted for age, gender, BMI and smoking, ≤ 25 remaining teeth (adjusted OR 1.63, 95% CI 1.01–2.61; p = 0.045), a PI > 2 (adjusted OR 1.93, 95% CI 1.09–3.42; p = 0.02) and < 1/day brushing frequency (adjusted OR 4.5, 95% CI 1.6–12.7; p = 0.004) were the independent predictors of “frequent” exacerbations. When COPD severity and dyspnea scores were additionally considered in the model, this association was still significant only for fewer remaining teeth (adjusted OR 1.69, 95% CI 1.03–2.77; p = 0.04) and lower daily brushing frequency (adjusted OR 4.19, 95% CI 1.44–12.1; p = 0.008). The frequency of supragingival treatment was not included in any of the models, however the number of patients who underwent frequent treatment (≥ 1 time/year) was very low (4%).

The quality of evidence for all outcomes ranged from moderate to very low (Table 3). The main reason for downgrading the quality of evidence pertained to the inclusion of non-randomized studies with serious methodological issues that most probably introduce bias and to the imprecision of estimates due to narrative synthesis. This means that further research with well-designed studies is very likely to have an important impact, and is likely to change our current estimates of effect.

The present systematic review addresses the effects of non-surgical or surgical periodontal treatment on various clinical outcomes of patients suffering from COPD and periodontal disease. To the best of our knowledge this is the first study to systematically assess the direct effect of periodontal therapy on the outcomes of patients with COPD. In summary, the accumulated evidence suggests that effective periodontal treatment in patients with COPD and periodontitis is associated with a reduced frequency of COPD exacerbations and slower lung function decline, although it is unclear whether it can also improve the quality of life. In addition, based on the results of a large population-based retrospective case–control study, periodontal treatment reduces the rate of hospitalizations for respiratory events (including COPD exacerbations) and is associated with lower all-cause mortality.

COPD is associated with systemic inflammation [35] and oxidative stress [36] and is now considered a systemic disease with frequent manifestations “outside the lungs” [37]. Thus, several comorbidities including cardiometabolic ones, lung cancer, osteoporosis, anxiety and depression, obstructive sleep apnea and gastroesophageal reflux occur more frequently in COPD patients compared to the general population [38]. Considering these findings, there is an ongoing need to further identify relevant comorbidities which share common pathogenetic factors with COPD and thus could exert a negative effect on its clinical outcomes. A growing body of studies over the last years has established an association between periodontitis and COPD [11, 15‐17]. Both diseases are characterized by neutrophilic inflammation with subsequent proteolytic destruction of connective tissue and share common risk factors such as age, smoking and social deprivation [9]. Accumulating evidence has identified several putative pathophysiologic mechanisms linking the two disorders including aspiration of cytokines, bacteria and neutrophils [9, 10], hematogenous dissemination of periodontal pathogens via gingival micro-ulcerations [39] and hyper-reactivity of peripheral blood neutrophils to bacterial stimuli and reactive oxygen species [40]. Among these, the most compelling evidence supports the hypothesis of neutrophilic inflammation and neutrophil-mediated tissue damage [9, 10]. Although a direct causal link cannot be unequivocally suggested based on current evidence, some studies have shown a higher loss of lung function in patients with more severe periodontal disease³⁴. Interestingly, COPD patients receiving corticosteroids (CS) are exposed at higher risk for periodontal disease as compared with those who are not; an additional issue to consider regarding the puzzling associations between COPD and periodontitis and specifically the additive effects of COPD treatment on onset and progression of periodontal disease [41, 42].

Several mechanisms have been proposed regarding the beneficial effects of periodontal treatment on COPD outcomes. First, periodontal therapy has been shown to reduce the concentration of various inflammatory biomarkers, potentially leading to less severe inflammatory response in patients with COPD [43, 44]. Second, removal of dental plaque may lead to a reduction of oral mucosa colonization with pathogens which are commonly isolated in respiratory and dental specimens of patients with COPD, especially during exacerbations [45, 46]. On the other hand, some authors have reported a deterioration of respiratory symptoms due to bacterial contamination caused by aerosols produced during dental treatment with ultrasonic devices [33, 47]. Antimicrobial preprocedural mouth rinse, high volume evacuation suction and air polishing avoidance have been proposed as precautionary measures [47].

The present systematic review underscores the need for collaboration between pulmonologists and dental practitioners in the field of COPD management. Poor oral health is one of the primary risk factors for COPD exacerbations and could stand on its own right as a potentially modifiable condition [48, 49]. On the other hand, the presence of periodontal disease should alert dental practitioners to the possibility of coexistent COPD and prompt them to refer the patient for further evaluation [29]. Future clinical trials should aim to investigate whether COPD patients will benefit from routine oral/dental assessment, and to provide important insights on the pathophysiological link between periodontal disease and COPD.

This systematic review was conducted according to an a priori registered protocol and has several strengths, including a comprehensive literature search, inclusion of randomized and non-randomized studies and the application of the GRADE approach to assess the overall quality of the evidence.

However, there are also several limitations that should be acknowledged. First, our search strategy was restricted to the last 20 years. However, it was our aim to evaluate the most recent data pertaining to the impact of periodontal treatment on COPD outcomes given that the most widely accepted diagnostic criteria for periodontitis and COPD were proposed in 1999 [34] and in 2001 [50], respectively and continue to be in use. In addition, the overall quality of evidence was very low for the effect of periodontal therapy on lung function and quality of life and moderate regarding its effect on COPD exacerbation frequency. Other limitations include the differences in study design and their methods of reporting results, the lack of uniformly applied definitions for COPD and periodontitis and the moderate-to-high risk of bias in individual studies. It should also be noted that in some studies exacerbations were self-reported by patients and not objectively verified by the investigator. Importantly, neither of the studies made adjustments for COPD severity according to FEV₁, nor explored the effect of periodontal treatment on respiratory symptoms, both considered powerful predictors of future exacerbations and hospitalizations [51]. Of note, exacerbations and their related risk of hospitalization are strongly associated with greater lung function decline among COPD patients [52]. Interestingly, the study by Shen et al. [30] demonstrated a beneficial role of periodontal treatment both on exacerbation rate and on FEV₁ decline_, a finding that should be examined more concisely in future studies. Additionally, only one study investigated the role of supportive follow-up periodontal therapy on COPD outcomes [29, 53]. Finally, the impact of periodontal treatment on prevalence and severity of COPD comorbidities has not been investigated.