Meta-analysis of the association between chronic periodontitis and chronic kidney disease

Abstract

BACKGROUND

Many scholars have performed several clinical studies have investigated the association between chronic periodontitis (CP) and chronic kidney disease (CKD). However, there are still differences between these research results, and there is no unified conclusion. Therefore, a systematic review is required to understand this issue fully.

AIM

To explore the correlation between CP and CKD.

METHODS

Literature on the correlation between CP and CKD, as well as the clinical attachment level (CAL) and pocket probing depth (PPD) of CKD and non-CKD, were retrieved from PubMed, Embase, the Cochrane Library, and Web of Science repositories until January 2024. After the effective data were extracted, data processing and statistics were performed using Stata 12.0.

RESULTS

Of the 22 studies, 13 were related to CP and CKD, and 9 reported CAL and PPD in patients with CKD and healthy controls. Meta-analysis of the correlation between CP and CKD revealed that CKD probability in people with CP was 1. 54 times that of healthy individuals [relative risk = 1.54, 95% confidence interval (CI): 1.40–1.70], and CP incidence in patients with CKD was 1. 98 times that of healthy individuals [overall risk (OR) = 1.98, 95%CI: 1.53–2.57]. Meta-analysis of CAL and PPD evaluations between CKD patients and healthy individuals showed that CAL and PPD levels were higher in CKD patients [standard mean difference (SMD) of CAL = 0.65, 95%CI: 0.29–1.01; SMD of PPD = 0.33, 95%CI: 0.02–0.63].

CONCLUSION

A bidirectional association exists between CP and CKD. CKD risk is increased in CP patients and vice versa. Periodontal tissue or tooth loss risks increase over time in CKD patients.

Key Words: Chronic periodontitis; Chronic kidney disease; Correlation; Meta-analysis

Core Tip: The occurrence and development of chronic periodontitis (CP) is often associated with other systemic diseases. However, the relationship between CP and chronic kidney disease (CKD) remains controversial, and there are still differences among relevant research results from many scholars with no unified conclusion. Based on this, we conducted a meta-analysis of the existing studies on the correlation between CP and CKD, and the results of 22 studies proved that CP is closely related to the incidence of CKD. Over time, patients with CKD are at increased risk of periodontal tissue destruction and tooth loss.

INTRODUCTION

Any congenital or acquired impairment of the periodontal supporting tissue can be defined as periodontal disease, a common oral disease that can eventually lead to tooth loss. World Health Organization statistics from 35 countries show that periodontitis is widely prevalent worldwide, with an incidence of 40%–75%[1]. The deepening body of research shows that chronic periodontitis (CP) is often not independent, and its occurrence and development are often linked to other systemic diseases, such as diabetes and cardiovascular and cerebrovascular diseases. Chronic kidney disease (CKD) is defined as a glomerular filtration rate (GRF) < 60 mL/min/1.73 m² due to kidney injury or unknown causes lasting for more than 3 months[2]. Owing to changes in lifestyles, social environments, and abuse of various drugs, the incidence of CKD has increased annually in recent years and has become a common disease affecting human health. Individuals with CKD have high rates of cardiovascular events and death[3]. However, the pathogenesis of CKD remains unclear. In 2005, a cross-sectional study found for the first time that severe CP was significantly associated with decreased renal function (GRF < 60 mL/min/1.73 m²)[4]. Subsequently, many scholars have performed clinical studies on the correlation between the two; however, there are still discrepancies between the results of these studies, and there is no unified conclusion. Based on this, we investigated the association between CP and CKD using a meta-analysis of the existing literature, aiming to provide a science-based foundation for their clinical treatment and early intervention.

MATERIALS AND METHODS

Literature retrieval

This research was performed according to the Meta-analysis of Observational Studies in Epidemiology statement[5]. Computer retrieval was conducted using PubMed, the Cochrane Library, Embase, and Web of Science from the times of their establishment until January 2024. The keywords used were “chronic kidney disease”, “kidney disease”, “chronic kidney insufficiency”, “pericementitis”, “chronic periodontitis”, “clinical attachment loss”, and “alveolar bone loss”.

Inclusion and exclusion criteria

The inclusion criteria were as follows: (1) Studies exploring the correlation between chronic periodontitis and CKD; (2) studies with clear diagnostic criteria for chronic periodontitis; (3) studies clarifying and reporting the diagnostic criteria related to CKD; and (4) containing or suitable for calculating outcome indicators: Risk estimates, including odds ratio (OR), relative risk (RR), or hazard ratio (HR); or secondary outcome measures: Clinical attachment level (CAL), or pocket probing depth (PPD).

The exclusion criteria were as follows: (1) Outcome indicators could not be counted; (2) the full text of the literature could not be obtained; (3) systematic reviews, clinical cases, reviews, dissertations, conference papers, animal experiments, and other documents; and (4) repeated publications.

Literature screening and data extraction

First, all studies obtained from related databases were imported into EndNote X9.1 software to eliminate duplicates. Then, the appropriate studies were selected based on the title and abstract. Their entire texts were read and papers that met the requirements were selected. Finally, the basic characteristics of the studies meeting the requirements were extracted, including: (1) Research information: First author’s name, publication time, nationality, and research design; (2) Baseline data: Sample size, sex, age, and diagnostic criteria for CP; and (3) Outcome data: OR/RR/HR and the corresponding 95% confidence interval (CI), and the average CAL and PPD values of the study and control groups.

Literature quality evaluation

Literature quality was evaluated using the evaluation criteria recommended by the Newcastle–Ottawa Scale (NOS)[6] and the Agency for Healthcare Research and Quality (AHRQ)[7]. NOS evaluation cohort studies include three aspects: Selection of exposed and unexposed cohorts, comparability of the two cohorts, and results. NOS uses a nine-point scale, with scores of 0–4 categorized as poor quality documentation, 5–6 categorized as medium quality, and > 7 was categorized as high quality. The AHRQ cross-sectional study included 11 items with a total score of 11 points, with scores of 0–3 categorized as poor quality, 4–7 categorized as moderate quality, and 8–11 categorized as high quality.

Statistical analysis

The extracted relevant data were analyzed and processed using Stata 12.0. OR/RR/HR values and their 95%CIs were calculated as a measure of the association between CP and CKD. To summarize the continuous data (CAL and PPD), the average of the differences between the groups was reported [standard mean difference (SMD)]. A heterogeneity test was performed on the data from the final included studies and analyzed for heterogeneity using the I² statistic. The heterogeneity of the studies was distinguished based on the I² value. The greater the I², the greater the heterogeneity. If I² was < 50%, this indicated less heterogeneity, and a fixed-effects model was applied. In contrast, I² ≥ 50% indicates greater heterogeneity, so a random-effects model was applied, and further sensitivity analysis was performed. Sensitivity analyses were performed by individually excluding each study to assess their potential impact on the overall risk estimate. Potential publication bias was identified by analyzing the symmetry of the standard funnel plot, which was evaluated using the Begg correlation test and the Egger’s linear regression method. The observed differences were deemed statistically significant if P < 0.05.

RESULTS

Results of literature screening

A total number of literature retrieved from the relevant databases was 1201. After screening and elimination, 22 eligible studies were finally included[4,8-28]. The literature screening steps are shown in Figure 1.

Figure 1 Literature screening process.

Included basic research information

The essential information included in the studies on the correlation between CKD and chronic periodontitis is presented in Table 1. In terms of study type, five cohort and eight cross-sectional studies were included. Overall 162914 cases were included in the studies, including 7331 patients with CKD and 28598 patients with chronic periodontitis. In each study, a range of measures and classification criteria were used to assess chronic periodontitis, including the Community Periodontal Index or the Community Periodontal Index of Treatment Needs, the guidelines including those from Centers for Disease Control and the American Academy of Periodontology, American Academy of Periodontology and European Federation of Periodontology, the European Federation of Periodontology, or no formal classification of disease. Basic information on studies comparing CAL (mm) and PPD (mm) in the non-CKD healthy control group and CKD cases is presented in Table 2 and Table 3, respectively.

Table 1 Basic characteristics of the literature on the correlation between chronic kidney disease and chronic periodontitis.

Ref.	Research location	Research type	Sample size (CKD/non-CKD)	Outcome indicators (number of cases of chronic periodontitis)	Effect size (95%CI)	CP diagnostic criteria
Yoshioka et al[8]	Japan	Cross-sectional study	66 (10/16)	45	OR = 3.169 (1.031-9.742)	According to CPI
Tsai et al[9]	Taiwan of China	Cross-sectional study	1280 (318/926)	668	OR = 3.80 (1.56-9.27)	According to AAP-EFP 2017 Symposium
Ioannidou and Swede[10]	United States of America	Cross-sectional study	3681 (81/3600)	354	OR = 1.59 (1.14-2.13)	According to CDC-AAP
Han et al[11]	Korea	Cross-sectional study	15729 (252/15477)	5143	OR = 1.39 (1.03-1.89)	According to CPI
Fisher et al[12]	United States of America	Cross-sectional study	12947 (617/12330)	784	OR = 1.60 (1.16-2.21)	According to CPI
Lamba et al[13]	United States of America	Cross-sectional study	210 (105/105)	121	OR = 3.954 (2.09-7.45)	According to the 2017 World Symposium on the Classification of Periodontal and Peri-Implant Diseases and Conditions
Kshirsagar et al[4]	United States of America	Cross-sectional study	5537 (110/5427)	3223	OR = 2.14 (1.19-3.85)	As defined by the Working Group on Periodontal Infection Surveillance
Ioannidou et al[14]	United States of America	Cross-sectional study	3686 (59/3627)	105	OR = 1.75 (1.13-2.71)	According to CDC-AAP
Sharma et al[15]	United States of America	Cohort study	13784 (861/12923)	1809	HR = 1.41 (1.36-1.47)	According to CDC-AAP (Page & Eke 2007)
Grubbs et al[16]	United States of America	Cohort study	761 (56/705)	270	RR = 2.01 (1.21–3.44)	According to European seminar standards
Grubbs et al[17]	United States of America	Cohort study	699 (21/678)	114	RR = 4.18 (1.68-10.39)	According to CDC-AAP 2003
Chen et al[18]	Taiwan of China	Cohort study	100263 (550/99713)	13749	OR = 1.59 (1.37-1.86)	According to CPI (CPITN)
Li et al[19]	United States of America	Cohort study	4271	2213	HR = 1.578 (1.54, 1.61)	According to CDC-AAP

CP: Chronic periodontitis; CKD: Chronic kidney disease; CPI: Community Periodontal Index; CPITN: Community Periodontal Index of Treatment Needs; CDC-AAP: Centers for Disease Control and the American Academy of Periodontology; AAP-EFP: American Academy of Periodontology and European Federation of Periodontology.

Table 2 Basic characteristics of clinical attachment level (mm) level difference between patients with chronic kidney disease and non-chronic kidney disease population.

Ref.	Research location	Research type	Sample size (CKD/non-CKD)	Average CAL (mm, CKD/non-CKD)
Garcez et al[20]	United States of America	Cross-sectional study	80/80	0.54 ± 0.60/0.44 ± 0.56
Gavaldá et al[21]	Spain	Cross-sectional study	105/53	4.9 ± 2.1/4.2 ± 2.5
Frankenthal et al[22]	Israel	Cross-sectional study	35/35	4.43 ± 0.29/4.03 ± 0.25
Cengiz et al[23]	Türkiye	Cross-sectional study	68/41	3.1 ± 1.5/2.4 ± 1.4
Limeres et al[24]	Portugal	Cross-sectional study	44/44	1.04 ± 0.59/0.71 ± 0.55
Camacho-Alonso et al[25]	Spain	Cross-sectional study	91/118	1.77 ± 0.81/1.15 ± 0.41

CKD: Chronic kidney disease; CAL: Clinical attachment level.

Table 3 Basic characteristics of pocket probing depth (mm) level differences between chronic kidney disease patients and non-chronic kidney disease population.

Ref.	Research locaiton	Research type	Sample size (CKD/non-CKD)	Average PPD (mm, CKD/non-CKD)
Garcez et al[20]	United States of America	Cross-sectional study	80/80	0.65 ± 0.75/0.56 ± 0.79
Tollefsen et al[26]	Norway	Cross-sectional study	12/12	2.61 ± 0.85/2.60 ± 0.42
Frankenthal et al[22]	Israel	Cross-sectional study	35/35	2.92 ± 0.14/2.90 ± 0.12
Marakoglu et al[27]	Türkiye	Cross-sectional study	36/36	1.80 ± 0.6/1.8 ± 0.8
Bayraktar et al[28]	Türkiye	Cross-sectional study	116/61	1.89 ± 0.48/1.94 ± 0.62
Cengiz et al[23]	Türkiye	Cross-sectional study	68/41	2.3 ± 0.6/1.6 ± 0.6
Limeres et al[24]	Portugal	Cross-sectional study	44/44	1.55 ± 0.98/0.81 ± 0.94
Camacho-Alonso et al[25]	Spain	Cross-sectional study	91/118	2.33 ± 1.07/1.76 ± 0.71

CKD: Chronic kidney disease; PPD: pocket probing depth

Literature quality evaluation

The NOS and AHRQ scale evaluation results of all the studies showed that the literature selected in this study was of medium and high quality, as shown in Table 4 and Table 5, respectively.

Table 4 Newcastle–Ottawa Scale scores of the cohort study literature.

Ref.	Study population selection				Comparability between groups	Outcome measurement			Total points
	Representativeness of the exposed group	Selection of non-exposed groups	Identification of exposure factors	No outcome indicators were available before the study began	Comparability of the resulting cohort based on design and analysis	Methods of evaluating outcome events	Whether the follow-up time was sufficient	Integrity of follow-up
Sharma et al[15]	1	1	1	1	1	1	1	1	8
Grubbs et al[16]	0	1	1	1	1	1	1	1	7
Grubbs et al[17]	1	1	1	1	1	1	1	1	8
Chen et al[18]	1	1	1	1	1	1	1	1	8
Li et al[19]	1	1	1	1	1	1	1	1	8

Table 5 Agency for Healthcare Research and Quality scores of cross-sectional studies.

Ref.	1	2	3	4	5	6	7	8	9	10	11	Total points
Yoshioka et al[8]	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Uncertain	8
Tsai et al[9]	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	No	Uncertain	Uncertain	8
Ioannidou and Swede[10]	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Yes	Uncertain	8
Han et al[11]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Uncertain	Uncertain	8
Fisher et al[12]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Uncertain	Uncertain	8
Lamba et al[13]	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	No	Uncertain	Uncertain	7
Kshirsagar et al[4]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	Uncertain	Uncertain	8
Ioannidou et al[14]	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	No	Uncertain	Uncertain	7
Garcez et al[20]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Uncertain	No	Uncertain	Uncertain	7
Gavaldá et al[21]	Yes	Yes	Yes	Yes	Yes	Uncertain	No	Uncertain	No	Uncertain	Uncertain	5
Frankenthal et al[22]	Yes	No	Yes	Yes	Yes	Yes	No	No	No	Uncertain	Uncertain	5
Cengiz et al[23]	Yes	Yes	Yes	Yes	Yes	Uncertain	Yes	No	No	Uncertain	Uncertain	6
Limeres et al[24]	Yes	Yes	Yes	Yes	Yes	Yes	No	No	No	Uncertain	Uncertain	6
Camacho-Alonso et al[25]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Uncertain	8
Tollefsen et al[26]	Yes	Yes	Yes	Yes	Yes	Yes	No	No	No	Uncertain	Uncertain	6
Marakoglu et al[27]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	No	No	Uncertain	Uncertain	7
Bayraktar et al[28]	Yes	Yes	No	Yes	Yes	Yes	Yes	No	No	Uncertain	Uncertain	6

1: Are the sources clear? 2: Are the inclusion and exclusion criteria for the two groups clear? 3: Is the time to identify patients clear? 4: Are the subjects continuous? 5: Is the evaluator isolated from other objective measures of the patient? 6: Has quality assurance been assessed? 7: Are excluded objects analyzed? 8: Are confounding factors assessed? 9: Is there any processing of the lost data? 10: Is the data complete? 11: Is there follow-up and analysis?

Results of meta-analysis and forest map

Cohort studes of the association between CP and CKD: Five cohort studies on the correlation between CP and CKD were included[15-19]. The heterogeneity test results showed P < 0.001 and I² = 86. 74%, suggesting heterogeneity. After applying a random effects model analysis, the results indicated that people with CP face a greater risk of developing CKD compared to those with good periodontal health (RR = 1.54, 95%CI: 1.40–1.70; Figure 2A). Sensitivity analysis (Figure 2B) revealed that heterogeneity mainly came from the study by Sharma et al[15] 2016 study, which excluded patients receiving kidney replacement therapy (such as dialysis or kidney transplantation), which was significantly inconsistent with the inclusion and exclusion criteria of other cohort studies. After the study by Sharma et al[15] was removed from the forest map, the scores were I² = 42. 5%, P = 0.157, indicating that the degree of heterogeneity was within the acceptable range. Therefore, a fixed-effect model analysis was selected, and the findings suggest that CP patients had a greater risk of CKD compared to those with good periodontal health (RR = 1.58, 95%CI: 1.55–1.61; Figure 3).

Figure 2 Meta-analysis (A) and sensitivity analysis (B) results of cohort studies on correlation between chronic periodontitis and chronic kidney disease. RR: Relative risk.

Figure 3 Meta-analysis of the correlation between chronic periodontitis and chronic kidney disease cohort study obtained after the study of Sharma et al[15] was removed. RR: Relative risk.

Cross-sectional studies of the correlation between CP and CKD: Eight cross-sectional research studies on the correlation between CP and CKD were included[4,8-14]. The findings of the heterogeneity test showed that P = 0.047 and I² = 50.9%, suggesting the presence of heterogeneity. Using a random effects model analysis, the results indicated that the risk of CP in CKD patients was 1.98 times that of healthy people (OR = 1.98, 95%CI: 1.53–2.57; Figure 4A). The results of the sensitivity analysis revealed (Figure 4B) that heterogeneity was mainly derived from the Lamba et al[13] 2023 study, which relied on patients’ medical records and self-reported data, which introduce possible information biases. After removing the study of Lamba et al[13] in the forest map, I² = 20.9%, P = 0.270, and the degree of heterogeneity was within the acceptable range. Therefore, a fixed-effect model was selected, and the findings suggest that patients with CKD were 1.66 times more likely to develop CP than healthy people (OR = 1.66, 95%CI: 1.41–1.96; Figure 5).

Figure 4 Meta-analysis (A) and sensitivity analysis (B) results of cross-sectional studies on the correlation between chronic periodontitis and chronic kidney disease. OR: Overall risk.

Figure 5 Meta-analysis of cross-sectional studies on the correlation between chronic periodontitis and chronic kidney disease after removing the Lamba et al[13] 2023 study. OR: Overall risk.

The difference of CAL depth between patients with chronic kidney disease and the healthy control group: Six cross-sectional studies (774 samples)[20-25] reported the average CAL of patients with CKD and healthy controls. The heterogeneity test results showed P < 0.001 and I² = 82. 5%, suggesting the presence of heterogeneity. Using random effects model analysis, and results showed that, compared with healthy controls, patients with CKD had a higher CAL depth (SMD = 0.65, 95%CI: 0.29–1.01; Figure 6A). Further sensitivity analysis revealed that (Figure 7A) no directional change occurred in the results after removing any reference, indicating the high stability and reliability of the results.

Figure 6 Differences in clinical attachment level depth (A) and Pocket probing depth (B) between chronic kidney disease patients and non-chronic kidney disease groups. SMD: Standardized mean difference.

Figure 7 Sensitivity analysis results of chronic kidney disease and clinical attachment level depth (A), and pocket probing depth (B).

Differences in PPD between patients with chronic kidney disease and healthy controls: Eight cross-sectional studies (889 samples)[20,22-28] reported the average PPD in patients with CKD and in healthy controls. The heterogeneity test results showed P < 0.001 and I² = 78.7%, indicating the existence of large heterogeneity. Using random effects model analysis, and the results showed that, compared with healthy controls, CKD patients had higher PPD (SMD = 0.33, 95%CI: 0.02–0.63; Figure 6B). Further sensitivity analysis revealed (Figure 7B) that no directional change occurred in the results after removing any references, indicating the high stability and reliability of the results.

Publication bias analysis

Publication bias in studies on the association between CP and CKD was analyzed using Begg’s rank correlation test and Egger’s linear regression method (Figures 8 and 9). A meta-analysis of the correlation cohort study of CP and CKD suggested that there was no publication bias (Egger’s regression test P = 0.777 and Begg’s correlation test P = 0.806); however, the funnel plot showed slight asymmetry on both sides (Figure 8A), which may be associated with the inclusion of fewer studies. The meta-analysis of cross-sectional studies on the correlation between CP and CKD suggests that there is a publication bias (Egger’s regression test P = 0.007 and Begg’s correlation test P = 0.035), and the funnel plot showed bilateral asymmetry (Figure 8B). The reasons were as follows: The small sample size of the individually included studies, the inconsistent follow-up time of each study, and the different instruments used for detection, which all may cause publication bias.

Figure 8 Funnel plot of correlation between chronic periodontitis and chronic kidney disease. A: Cohort study; B: Cross-sectional study.

Figure 9 Egger diagram of the correlation between chronic periodontitis and chronic kidney disease. A: Cohort study; B: Cross-sectional study.

DISCUSSION

According to previous studies, potential risk factors for CKD include age, proteinuria, high blood pressure, diabetes, dyslipidemia, obesity, and smoking status. However, these traditional factors do not fully explain the occurrence of kidney disease. In 2005, Kshirsagar et al[4] first noted that severe chronic periodontitis could lead to decreased renal function (GRF < 60 mL/min/1.73 m²). Subsequently, Fisher et al[12] screened 12947 adults in the United States from the third Health and Nutrition Survey to conduct a cross-sectional study and discovered that the prevalence of CKD was dramatically higher in people with periodontitis than in those without. CP causes inflammation in the human body, and studies have shown that chronic inflammatory responses can lead to hypertension and diabetes[29], both of which are important causative factors of cardiovascular disease and CKD. In mouse model studies, periodontitis can upregulate the expression of TNF-α, IL-1b, IL-17A, and MMP9 in renal tissue with renal interstitial fibrosis, promote the infiltration of neutrophils and aggravate the inflammatory response[30]. Graziani et al[31] believe that basic periodontal treatment not only has positive effects on systemic inflammatory markers, but also has certain positive effects on GFR. Clinical and basic studies have demonstrated a correlation between CP and CKD.

CP may promote CKD through the following means: (1) CP can cause a systemic inflammatory response, which then in turn contributes to the release of inflammatory cytokines (including IL-6, TNF-α, etc.). This can affect the kidney through blood circulation and promote kidney inflammation and fibrosis, thus hastening the progression of CKD[32,33]. Furthermore, inflammation may lead to vascular endothelial dysfunction, affecting the elasticity and permeability of blood vessels, and increasing the risk of vascular sclerosis, which is closely related to the development of CKD as it can lead to reduced renal blood flow and affect kidney filtration[34]; (2) Under chronic inflammation, the level of oxidative stress increases and large amounts of reactive oxygen species are produced, which can damage kidney cells and promote kidney injury and fibrosis[35,36]; and (3) Periodontitis may cause oral bacteria to enter blood circulation, causing bacteremia. These bacteria and their products may damage the kidney tissue directly or indirectly affect kidney health by activating the immune system[37,38]. In patients with CKD, the immune system is disorganized, and blood immune cells tend to be activated and express more pro-inflammatory factors, thus promoting the occurrence of opportunistic infections[39], which can eventually lead to an imbalance between normal periodontal flora and host immunity. Next, due to decreased salivary secretion in patients with CKD, the periodontal flora may be disturbed by changes in salivary urea concentration or PH value[40]. Therefore, the prevalence of CP in patients with CKD may be higher than that of healthy individuals, and the association between CP and CKD may be bidirectional. At present, the literature has shown scattered correlations between CP and CKD, but no research has shown a definitive relationship between them. Thus, we evaluated the correlation between CP and CKD using a meta-analysis.

According to the meta-analysis of cohort studies, the CP population is 1.54 times more likely to develop CKD than the periodontally healthy population (RR = 1.54, 95%CI: 1.40–1.70). Results from cross-sectional studies indicate that the incidence of CP in CKD patients was 1.98 times higher than that in healthy people (OR = 1.98, 95%CI: 1.53–2.57). Whether based on cohort study or cross-sectional study, sensitivity analysis suggests that the two random effects were stable and dependable, which proved that CP was closely related to the incidence of CKD. A systematic evaluation of four observational studies also supported the findings of this paper, which indicates that people with periodontitis have a 65% (CI: 1.35–2.01) increased risk of developing CKD compared to healthy individuals or those with mild periodontitis[41]. Most studies have confirmed a correlation between CP and CKD; however, some have reported inconsistent results. For example, the study results of Castillo et al[42] revealed that there was little difference in the concentration of periodontal bacteria, such as Trypanosoma forsythia, Porphyromonas gingivalis, Prevotella intermedia, and Prevotella nigra, between healthy individuals and patients with CKD, indirectly indicating that there was no substantial relationship between CP and CKD. The association between CKD and CP may also be influenced by the degree of impaired kidney function; however, in this meta-analysis, the practical approach of combining all stages of CKD into one “disease” group may have exaggerated the estimated effect in more than half of the studies. Some of the included studies involved end-stage CKD patients with GFR levels equal to or below 15 mL/min/1.73 m² who had recurrent uremic syndromes without dialysis, which impaired the normal immune function of monocytes and polymorphonuclear lymphocytes. Thus, overgrowth of periodontal bacteria is possible in the presence of end-stage CKD.

Unlike the criteria used to assess the presence of CKD, the definition and classification of CP are controversial, leading to differences in the diagnostic criteria worldwide. Therefore, to include as many studies as possible in the meta-analysis, this study also included the measurement of periodontal status. CAL measures the lack of periodontal tissue caused by periodontitis. Over time, an increase in PPD is associated with the loss of teeth caused by periodontitis[43]. The results of this meta-analysis comparing CAL and PPD assessment between patients with CKD and healthy people showed that CKD was significantly associated with higher CAL and PPD (SMD of CAL = 0.65, 95%CI: 0.29–1.01; SMD of PPD = 0.33, 95%CI: 0.02–0.63); thus, it can be deduced that the risk of periodontal tissue destruction and tooth loss in patients with CKD tends to increase over time.

Limitations of this study included: (1) The definition and severity of CP were not uniform among different studies, and a “dose-effect analysis” could not be performed; (2) The studies included in this meta-analysis lacked intervention studies, which can indicate a bidirectional relationship between CP and CKD, but whether immunosuppression induced by CKD increases susceptibility to CP and whether systemic inflammatory response induced by CP leads to chronic pathological changes of renal function, no causal relationship can be concluded at present. A reliable assessment of causality can only be based on interventional studies, especially randomized controlled trials; and (3) There were differences in the adjustments for potential confounding factors among the included studies, resulting in a certain degree of confounding bias.

CONCLUSION

In summary, we conducted a meta-analysis and found a correlation between CP and CKD, finding that the risk of periodontal tissue destruction and tooth loss in patients with CKD increased over time. Based on the above results, it is clinically advisable to consider renal function when treating periodontal disease in patients with chronic periodontitis to improve their oral hygiene. Likewise attention should be paid to the oral hygiene of patients with CKD.