Background
Periodontitis is a chronic infectious disease with a high incidence that is amajor cause of tooth loss. Periodontitis is classified into two major types: chronic periodontitis (CP) and aggressive periodontitis (AgP) [
1,
2]. Periodontitis is also recognized as one of the most common diseases worldwide, with a prevalence of 15–20% [
3]. Research has indicated that the occurrence of periodontitis is the result of multiple factors. Geneticvariation may be a factor that contributes to individual susceptibility to periodontitis [
4]. Recent research has indicated that polymorphisms in multiple genes are associated with susceptibility to periodontitis [
5,
6]. Interleukin-13 (IL-13) is a pleiotropic cytokine produced by type 2 helper T (TH2) lymphocytes. Single nucleotide polymorphisms (SNPs) have been identified in the interleukin (IL)-13 gene, including two SNPs (−1512A/C and -1112C-T) in the 5′-terminal regulatory region and one SNP (+1923C–T) in the third intron [
7]. These SNPs have extremely complex biological effects and are involved in various physiological processes; additionally, they can exhibit different effects under different conditions and can even play dual roles in the same pathophysiological process. The association between the IL-13 -1112 gene polymorphism and the susceptibility to periodontitis has been studied recently. However, inconsistencies exist in the conclusions due to the limited sample sizes and possible differences between studies. These studies are based on small samples in case-control studies; thus, the credibility of the conclusions has been questioned. Therefore, the IL-13 -1112 gene polymorphism and its possible correlation with periodontitis must be analyzed from an evidence-based medicine perspective. A meta-analysis is a combination of multiple effects of the same purpose to overcome the shortcomings of small sample studies by performing a comprehensive analysis and objectively evaluating the statistical methods. This approach can obtain more credible conclusions than a single study. Recently, meta-analyses have been widely used in the field of genetic polymorphism research. The aim of this study was to use a meta-analysis to evaluate the relationship between the IL-13-1112C/T polymorphism and susceptibility to periodontal disease.
Methods
Search strategy
Data were collected from the CNKI (China National Knowledge Infrastructure), Wanfang, PubMed (PubMed is an interface for the Medline database), Embase and Cochrane Library databases. Published articles in Chinese and English that addressed the IL-13-1112C/T SNP and susceptibility to periodontitis were retrieved. The retrieval process involved the use of Chinese and English keywords, including “periodontitis”, “periodontal disease”, “IL 13”, “IL-13”, and “interleukin-13”. Relevant publications were also identified via searches supplemented by a literature review. The meta-analysis was restricted to articles published on or before November 30,2016.The search strategy is asfollows (Table
1):
Search (((“Periodontitis”[Mesh]) OR (((Periodontitides[Title/Abstract]) OR Pericementitis[Title/Abstract]) OR Pericementitides[Title/Abstract]))) AND ((“Interleukin-13”[Mesh]) OR ((((IL-13[Title/Abstract]) OR Interleukin 13[Title/Abstract]) OR IL 13[Title/Abstract]) OR IL13[Title/Abstract])) | ‘periodontitis’/exp. OR ‘periodontitides’:ti,ab OR ‘pericementitis’:ti,ab OR ‘pericementitides’:ti,ab AND (‘interleukin 13’/exp. OR ‘il-13’:ti,ab OR ‘interleukin-13’:ti,ab OR ‘il 13’:ti,ab OR ‘il13’:ti,ab) | #1:MeSH descriptor: [Periodontitis] explode all trees #2:Periodontitides:ti,ab,kw or Pericementitis:ti,ab,kw or Pericementitides:ti,ab,kw (Word variations have been searched) #3: #1 or #2 #4:MeSH descriptor: [Interleukin-13] explode all trees #5: IL-13:ti,ab,kw or Interleukin 13:ti,ab,kw or IL 13:ti,ab,kw or IL13:ti,ab,kw (Word variations have been searched) #6: #4 or #5 | Subject = Periodontitis or Topic = Periodontal Disease and Subject = Interleukin 13 or Subject = IL-13 or Subject = IL 13 or Subject = Interleukin-13 (Exact Match) | Subject: (periodontitis + periodontal disease) * Subject: (interleukin 13+ IL-13 + IL 13+ Interleukin-13) * |
Conditions for study inclusion
All included studieswere required to satisfy the following three conditions.
Inclusion: 1. The literature was limited to domestic and foreign published cases of periodontal disease associated with the IL-13 -1112C/T gene polymorphism in Chinese and English; 2. The case group of patients had a clinical diagnosis of periodontitis, including CP patients and AgP patients, whereas the control group was the periodontal healthy population; additionally, the case and control groups were not associated with other systemic diseases; 3. Indicators were provided in the study, including the IL-13-1112C/T odds ratio (OR) and the 95% confidence interval (95% CI), for the genotype frequency and the allele frequency, or the original data provided were sufficient to perform the calculations; and 4. When multiple document data were identical or overlapping, the most recently published literature was selected.
Exclusion: 1. Non-genetic polymorphism research was excluded; 2. Polymorphic loci genotype distribution frequencies and specific data descriptions were unclear; 3. The case group suffered not only from periodontitis but also from other diseases; 4. The study did not set up a control group; and 5. The study was conference literature.
Data were independently extracted from the evaluated information and cross-checked by two reviewers (WB Zhang and QH Mao). Inconsistencies were resolved by discussion or by consulting a third researcher (YN Cheng). The assessed information included the name of the first author, the date of publication, the distribution area of the study subjects, the genotype distributions, the Hardy-Weinberg balance index and the disease descriptions [
8]. The quality of the literature included in the study was evaluated using the NOS scale (Newcastle-Ottawa quality assessment scale) [
9]. The table was defined and selected from the case and control groups to assess the comparability between the two groups. The identification of exposure factors was based on three aspects of the score, with a literature quality score ≥ 6 representing high quality.
Data analysis
Odds ratios (ORs) and the corresponding 95% confidence intervals (CIs) for the pooled data were used to determine the relationship between the assessed IL-13 -1112C/T SNP and susceptibility to periodontal disease.
We used five models, including an allele model (C vs T), co-dominant models (CT vs TT and CC vs TT), a dominant model (CC vs CT + TT) and a recessive model (TT vs CT + CC), and subgroup analyses were performed based on the case type. The Chi-square test and Q-test were utilized to assess the heterogeneity of the data. If P > 0.10 or I2<50%, the merged data exhibited good homogeneity, and a fixed-effects model was used for the analysis; otherwise, a random-effects model was employed. A subgroup analysis was performed based on the disease type. The software used for the aforementioned analysis was STATA12.0 (Stata Corp. LP, College Station, TX, USA). Finally, each study was removed individually for the sensitivity analysis prior to the conclusion of the study.
Discussion
Periodontitis is an inflammatory infectious disease that is the main cause of the loss of human teeth; its main features are gingival inflammation, periodontal pocket formation, alveolar bone resorption and tooth loosening and displacement [
14]. An early sign of gingivitis is dental plaque accumulation, which with further development of inflammation may progress into periodontitis. However, plaque reactions are not consistent between individuals; some plaque accumulation occurs when gingivitis develops, followed by further development of periodontitis, whereas other individuals exhibit plaque accumulation but do not progress to periodontitis. Therefore, genetic factors play an important role in the development and progression of periodontitis. Recent evidence has suggested that individual differences in susceptibility to periodontitis are associated with genes [
15,
16].
The IL-13 gene is located on chromosome 5 (5q23.31). The gene is 4.6 kb in length, contains 4 exons, 3 introns and 2 methionine initiation codons and encodes a 132-amino acid protein [
17]. The non-glycosylated protein has a relative molecular weight of 12,000 Da. IL-13 is a multi-potent cytokine with various biological effects [
18]. For instance, IL-13 is a key inducer of many pathological processes that are determined by class II cytokines [
19]. It can regulate inflammation, mucus production, tissue remodeling and fibrosis and as a result has become a therapeutic target for many diseases, including asthma, idiopathic pulmonary fibrosis, ulcerative colitis and other IL-13 overexpression diseases [
20‐
22].
During the process of periodontitis, type II helper T lymphocytes (TH2) secrete IL-4, IL-13 and other cytokines that are conducive to B cell humoral immunity and improve the symptoms of inflammation. These cytokines can inhibit the secretion of tumor necrosis factor-α (TNF-α) by mononuclear cells to achieve some anti-inflammatory effects and can activate transforming growth factor-β1 (TGF-β1) and thus produce fibrosis. IL-13 can affect bone resorption by affecting the bone protection factor system and can directly promote IgE and thus cytokine synthesis in humans. Studies have found that periodontitis in the gingival tissue is associated with a significant increase in IgE and that IL-13 can stimulate cells in periodontal lesions to produce and fight against periodontal pathogens that express lipopolysaccharide. Therefore, understanding the biological role of IL-13 and its gene polymorphisms is necessary for an in-depth study of periodontitis. Of the single nucleotide polymorphic loci of IL-13, the − 1112 locus was the first discovered and the most studied. In 1999, the Dutch scholar Van der Pour Kranfirst [
23] confirmed that the site could be enhanced. The binding of the nucleoprotein to the promoter region increased IL-13 synthesis. Nie et al. [
24] showed that the IL-13 -1112 gene polymorphism was associated with the occurrence of asthma in Caucasians; however, this site was not associated with susceptibility to asthma among Asians and African Americans. A number of studies have shown that the IL-13 -1112 gene polymorphism is associated with a variety of diseases, including type 2 diabetes, chronic obstructive pulmonary disease, allergic rhinitis, inflammatory bowel disease and colorectal cancer [
25,
26]. The IL-13 -1112 gene polymorphism is associated with susceptibility to periodontitis, which has been studied in recent years.
Currently, genetic polymorphisms and periodontitis are studied using case-control studies. However, case-control studies can have insufficient sample sizes, neglect confounding factors and include multiple statistical measurements. Additionally, case-control studies themselves can demonstrate weakness and other defects [
27], resulting in reduced reliability of the results. A meta-analysis [
28] is a systematic quantitative analysis method for the consolidation of multiple similar studies. The advantage of this approach is that it can increase the sample content from the statistical perspective and improve the test efficiency, especially when multiple studies are included. When the results are not significant or the conclusion is inconsistent, the meta-analysis method can be closer to the real situation of the system analysis results. Meta-analysis results have been used as best evidence in health decision-making and clinical practice and have played an increasingly important role. The present study on the association between the IL-13 -1112C/T gene polymorphism and periodontitis is different. Therefore, this meta-analysis of the IL-13-1112C/T gene polymorphism and periodontitis susceptibility used case-control studies to conduct a systematic quantitative analysis of the IL-13-1112C/T gene polymorphism and susceptibility to periodontitis and to provide evidence-based medical evidence.
This study systematically collected relevant case-control studies, merged them to discuss the relevance of the site and the incidence of periodontitis and provided meta-analysis results. Five studies were included, with 2 studies conducted on AgP. Gonzales [
10] suggested that the IL-13-1112 genotype and the C and T allele frequencies were not associated with susceptibility to AgP, whereas Wu et al. [
11] suggested that the CC genotype of IL-13 -1112 was a risk factor for AgP compared to the CT/TT genotypes. Three studies investigated CP. Zhang et al. [
12] showed that the IL-13 -1112 allele frequency and genotype frequency were independent of the CP susceptibility, which was the same as the finding of Wu et al. [
11]. Chen et al. [
13]. showed that the IL-13-1112 gene polymorphism might be associated with susceptibility to CP. All of the included studies had quality scores greater than 6 points, indicating that the quality of the included literature was high. To maximize the correlation between the genetic model and periodontitis, the CC vs TT, CC vs CT and TT vs CT models were used to analyze the correlation between the models. The correlation for the TT model + CC model was significant. In the subgroup analysis for the disease-based groups, the differences between the CT vs TT, CC vs TT and TT vs CT + CC models were significant in the CP group. The heterogeneity was low, and the conclusion was reliable, suggesting that the IL-13 -1112 gene polymorphism was associated with CP susceptibility. The five models in the AgP group showed no significant differences. According to the results of this study, the association of the IL-13 -1112 gene polymorphism with periodontitis may be due to differences in the disease (CP and AgP), which are presumably due to factors such as race, region, predilection, suspicious disease-causing bacteria and etiology. The IL-13-1112 gene polymorphism may play different roles in the pathogenesis of CP and AgP; thus, the same gene polymorphic sites may occur in patients with CP and AgP, underscoring the relevance of the differences that lead to different susceptibilities. Heterogeneity is one factor that affects the reliability of our results. The statistical analysis of the CC vs CT + TT model (
P = 0.050, I
2 = 57.8%) with the C vs T model (
P = 0.012, I
2 = 68.9%) suggests that
P<0.1 or I
2 > 50% indicates that the merged data are heterogeneous, although the sensitivity analysis failed to find the source of heterogeneity. Possible sources of heterogeneity may include the regional population, genetic testing, quantity and the research process.
This study has the following limitations. 1. The number of original studies included in this study is small. Additionally, one of the studies used MALDI-TOF-MS, whereas the others used PCR-RFLP. The sensitivity of the two methods results in a certain degree of inconsistency. The control group in another study did not meet the HWE balance; thus, the control group subjects may exhibit population bias. The above issues limit the credibility of the results of this study but also show the need for the current study. More high-quality studies should be conducted to verify the conclusions of this study. 2. Of the studies included in this meta-analysis, one case-control study investigated a German population, and four studies investigated Chinese populations. Because the study’s language limitations were English and Chinese, relevant literature on different racial groups published in other languages may exist and requires further analysis. 3. In this meta-analysis study, only two cases included patients with AgP. The number is small; therefore, studies with a larger sample size are needed to confirm the relationship between the IL-13 -1112 gene polymorphism and AgP. 4. The commonly used detection methods for publication bias are Egger’s regression method [
29], the funnel regression method [
30], Begg’s rank method [
31] and the trim and fill method [
32]. Some scholars have noted that at least five independent studies (the point of the funnel map > 5) [
33] are required to obtain a persuasive result with the funnel regression method. Additionally, some scholars believe that Begg’s rank method and Egger’s regression method are less sensitive for the identification of publication bias when they are included in the study [
34]. Certain risks are inherent in the identification of publication bias; for instance, publication bias may be overestimated when the study is too small. Therefore, these methods should be used with caution [
35]. Since the number of studies included in this study is small, some bias is likely to exist in this study because several commonly used bias detection methods are limited in this case.