Surgical Treatment for Patients With
Periodontal Disease Reduces Risk of
End-Stage Renal Disease: A Nationwide
Population-Based Retrospective Cohort
Study
Chun-Feng Lee,* Cheng-Li Lin,† Ming-Chia Lin,‡ Shih-Yi Lin,§ Fung-Chang Sung,†i
and Chia-Hung Kao¶#
Various forms of periodontal disease impair human health, in addition to being evidenced by bone loss and tooth mobility.1,2 These dental diseases are highly prevalent, affecting up to
90% of the worldwide population.
1 In Taiwan, approximately 91% of adults
are affected.3 Periodontal
disease is the destructive inflammatory
reaction of surrounding tooth tissues such as
periodontal ligament, cementum, alveolar bone,
and gingival tissue. As disease progresses, a
deep gingival pocket develops, and eventually,
tooth loosening and loss occurs. The main risk
factors of periodontal disease are poor oral hygiene,
presence of oral
plaque microorganisms, genetics, tobacco and
alcohol consumption, improper nutrition, stress, and impaired host immune response.4 Some systemic diseases, such as
diabetes mellitus, cardiovascular disease, hematologic disease, and psychosomatic disease, may contribute to periodontal disease.3
End-stage renal disease (ESRD) is the most
severe type of chronic renal failure, requiring replacement therapies such as hemodialysis, peritoneal
dialysis, or renal transplantation. ESRD has
been reportedly associated with periodontal disease.
4-7 The prevalence of chronic renal disease is increasing in Western and Asian counties. Chronic kidney disease and chronic periodontitis (CP) might share significant mutual etiological factors.6,8 CP is
suggested to contribute to the overall systemic inflammatory burden and affect the management of
patients with ESRD undergoing hemodialysis therapy.
2,4,7 However, the relationship between periodontal disease management and hemodialysis
progress is not clear. No large population-based study is available on the difference in ESRD risks
between treated and untreated patients with periodontal disease. The present study determines
whether the treatment of periodontal disease with subgingival curettage and/or periodontal
flap surgery reduces ESRD risk. Insurance claims data were obtained from the National Health Insurance system of Taiwan to establish periodontal disease cohorts
with and without treatment.9 The difference in ESRD risk between the two cohorts was then compared.
MATERIALS AND METHODS Data Sources
This study uses a subset of insurance claims data from the Taiwan
National Health Insurance
Database provided by the National Health Research Institute. Implemented in 1996, the universal
health insurance program now covers approximately 99% of the 23.74 million residents of Taiwan.9
The National Health Insurance Database provides comprehensive information on demographic status, diagnostic codes, medical institutions,
outpatient and inpatient orders, and
prescriptions, as well as expenditures for healthcare.
The subset data included claims data from
1996 to 2009, randomly sampled for 1 million insured people from the entire membership. International Classification of Diseases, Ninth
Revision, Clinical Modification (ICD-9-CM) codes are used to define diseases for the study (the Taiwanese insurance code book). The datasets were linked with scrambled identification to protect patient privacy.
Study Participants
Two cohorts of adult patients with periodontal disease
were established based on the claims data from 1997 to 2009. One cohort underwent surgical treatment, and the other cohort did not. Patients were identified who received care for periodontal diseases (ICD-9-CM codes 523.4 and 523.5) at least three times, to exclude patients with minor dental complaints. Procedure codes were based on the Taiwanese insurance code book to link patients who underwent surgery for subgingival curettage (root planing) (91006C [full mouth], 91007C [1/2 arch], and 91008C [localized]) and periodontal flap surgery (91009C [localized] and 91010C [1/3
arch]) as the treatment cohort. The date the periodontal surgery was performed served as the index
date for the patient. Patients with a history of chronic kidney disease by the index date were
excluded. A no-treatment cohort comprised randomly selected patients with periodontal disease
who did not have surgical treatment at any time during follow-up. For each patient in the treatment cohort, four comparisons were randomly selected to the no-treatment cohort: frequency matched with age, sex, and the index date.
Outcome Measures
Follow-up person-years data were measured for each patient from the index date to the date of
ESRD diagnosis (ICD-9-CM code 585), patient withdrawal from the insurance system, or the end of
2009. Participants with proteinuria (code 791.0), diabetes (code 250), hypertension (codes 401 to 405), coronary artery disease (codes 410 to 413, 414.01 to 414.05, 414.8 to 414.9), congestive heart failure (code 428), hyperlipidemia (code 272), and cancer (codes 140 to 208) identified by the index date were considered to have comorbidities.
The study design is thorough, and adequate
control of available confounding factors was ensured.
Furthermore, this study may demonstrate the
natural history of ESRD risk associated with periodontal disease.
Statistical Analyses
Data analysis was performed by comparing the distribution of sex, age, and comorbidities of the patients between the two cohorts. The differences were tested using the x2 test for categorical variables and t test for continuous variables. Incidence
rates of ESRD were estimated by demographic status as well as comorbidity, and the rates in both cohorts were compared. Poisson regression was used to estimate the incidence rate ratio (IRR) and 95% confidence interval (CI) of the treatment cohort to the no-treatment cohort. Cox proportional hazards regression was used to estimate the related
hazard ratio (HR) and 95% CI of ESRD. Given that diabetes, hypertension, and hyperlipidemia were prevalent among the study participants, further data analysis using software** was performed to evaluate the interaction pattern among the three comorbidities.
A two-tailed P value <0.05 was considered statistically significant.
RESULTS
Among patients with periodontal disease who were free of chronic kidney disease at baseline, 35,496 patients were included in the treatment cohort (18,061 males and 17,395 females; mean age – SD:
47.7 – 12.7 years) and 141,824 in the no-treatment cohort (72,244 males and 69,580 females; age:
47.6 – 12.8 years). The mean ages and sex distributions in the two cohorts were similar. Compared
with the no-treatment cohort, diabetes
(10.8% versus 10.3%, P = 0.004) and hyperlipidemia (17.2% versus 16.4%, P <0.0001) were slightly increased in the treatment cohort; hypertension (20.9% versus 21.4%, P = 0.04) and congestive heart failure (0.88% versus 1.18%, P <0.0001) were decreased (Table 1). The prevalence rates of proteinuria,
coronary artery disease, and malignancy were similar in both cohorts.
The overall incidence rate of ESRD was 37% lower in the treatment cohort than in the no-treatment cohort (4.66 versus 7.38 per 10,000 person-years, IRR = 0.63, 95% CI = 0.60 to 0.66), with an adjusted HR of 0.59 (95% CI = 0.46 to 0.75) (Table 2). Sexand age-specific analysis of IRR showed higher
beneficial effect from treatment for men than for women, and the relative benefit increased with age.
However, the adjusted HR was higher for men compared with women and increased with age.
Young adults did not benefit from surgery.
Table 3 presents the incidence of ESRD measured by the comorbidity status, which was consistently lower in the treatment cohort. Patients with
proteinuria and congestive heart failure in the notreatment cohort showed much higher ESRD incidence.
The IRR for ESRD of the treatment cohort
to the no-treatment cohort was the lowest among patients with congestive heart failure (0.25; 95%
CI = 0.13 to 0.46). The corresponding IRRs of patients with proteinuria and diabetes were 0.55 (95% CI = 0.30 to 0.99) and 0.63 (95% CI = 0.56 to 0.71), respectively. Individuals without comorbidities also had lower ESRD incidence in the treatment cohort than in the no-treatment cohort.
Table 4 shows that the incidence of ESRD increased with the number of comorbidities (diabetes,
hypertension, and hyperlipidemia). Patients in the no-treatment cohort with these comorbidities had the highest incidence of 66.8 per 1,000 personyears.
Surgery reduced the incidence by nearly half to 36.9 per 10,000 person-years, with an IRR of 0.55 (95% CI = 0.45 to 0.67). The multivariable Cox proportional hazards regression analysis showed that, compared with individuals without comorbidities, those with these comorbidities simultaneously
had an adjusted HR of 26.2 (95% CI = 19.9 to 34.5).
Figure 1 compares the overall trend of ESRD incidence between treatment and no-treatment cohorts
during the follow-up period. The cumulative incidence of ESRD was significantly lower in the
treatment cohort than in the no-treatment cohort by the end of 12 years of follow-up.
DISCUSSION
Periodontal disease is a common chronic inflammatory condition, initially bacteria-driven, leading to
deep gingival pocket formation. Eventually, the deep collagenous structures of the periodontium and supportive alveolar bone are broken down, leading to premature tooth loss.1,10 The pathogenesis of periodontitis is an immune-inflammatory response that is induced by the presence of plaque bacteria
developing in periodontal tissue. Sustained inflammatory mediators and host cellular components
may release proteolytic enzymes and further damage soft and hard tissues.11 A dental
plaque or biofilm composed of bacteria and bacterial products such as endotoxins/
lipopolysaccharides, as well as an extracellular matrix of polysaccharide,
proteins, and inorganic compounds, colonizes the periodontal pockets. These
components are the possible causes of systemic inflammation from periodontal diseases.12
Some studies revealed that periodontitis is associated with increased
systemic inflammatory burden, which possibly occurs through the acute-phase response mediator C-reactive protein and several cytokines related to inflammatory response.8,13
The present results reveal that periodontal treatment including subgingival
curettage and/or periodontal flap surgery reduces the incidence rate of ESRD to 37%, with an adjusted HR of 0.59 (P <0.01). A possible explanation is the effective elimination of irritants, inflammatory mediators, and bacterial
products from deep gingival pockets by advanced periodontal treatment, because it stops systemic inflammation. Recent studies show that effective periodontal therapy can reduce systemic inflammation, and that the maintenance of oral
hygiene as well as periodontal health decreases the risk of ESRD.14,15 The results also showed that with sex and age
stratification, the IRR of the treatment cohort to the no-treatment cohort was higher in women than in men and decreased with age. The age-specific incidence
rates showed that patients aged 20 to 39 years in the treatment cohort had the highest risk, but the adjusted HR
was not significant. The elderly in the notreatment cohort had the highest risk of
ESRD. These findings indicate that elderly patients benefit the most from surgery.
Analysis of the association between ESRD and comorbidity with periodontal treatment reveals that ESRD risks measured by comorbidity status decreased in
the treatment cohort (Table 3). Patients with hypertension, coronary artery disease, congestive heart failure, and hyperlipidemia who received periodontal
therapy had lower ESRD IRRs (0.59, 0.40, 0.25, and 0.48, respectively; P <0.01).
The results are partially consistent with a previous study.16 The results also indicate that
the maintenance of periodontal health helps prevent cardiovascular disease (CVD) and related
disorders.17,18 Kshirsagar et al.19 have suggested that periodontal disease can be linked to CVD disorders among patients with ESRD. That study also proposed that periodontal disease treatment reduces CVD morbidity and mortality among patients with ESRD. A clear relationship exists
between diabetes and periodontitis via the mechanism of impaired immune function. Recent studies
confirm that diabetes is an important risk factor for periodontitis.20-22 The comorbidity findings stratified by diabetes also suggest that the treatment of periodontal disease for patients with diabetes lowers the ESRD IRRs compared with the no-treatment cohort.
This study has some limitations. First, the comparative analyses were not adjusted for potential
confounders of smoking, drinking, and other unhealthy lifestyle behaviors, because these data are
not available in the Taiwan National Health Insurance Database. However, with such a large study
population, there may be some neutralizing of their possible effect. Second, the data analysis did not control for obesity. In the insurance claims data, obesity is coded only in patients with severe obesity seeking weight loss. The prevalence of obesity is low in the relevant population. Because of the large study sample size and the strong measured benefit from the surgery treatment, the effect of obesity is likely to be neutralized as well.
CONCLUSIONS
This population-based retrospective cohort study finds a remarkably reduced ESRD risk in patients with periodontal disease who received surgery
treatment of subgingival curettage and/or periodontal flap. The periodontal treatment also accords
reduced risk of ESRD to patients with
comorbidity. The underlying mechanism remains unclear and may be related to the elimination of inflammatory mediators and bacterial products, which lead to systemic inflammation. Further large-scale studies are needed to confirm these findings.
