Introduction

Periodontitis is considered as a chronic inflammatory disease by bacterial infection and it is classified into two categories, namely chronic (CP) and aggressive (AP) [1]. Periodontitis is caused by microorganisms that adhere to and grow on tooth surfaces, along with an overly aggressive immune response against these microorganisms [1]. In some patients, periodontitis may develop with malfunctioning of some components of the immunity system (i.e. under the situation of impaired/insufficient immune response).

Tumor necrosis factor-α (TNF-α) is a potent pro-inflammatory cytokine that plays important roles in inflammatory and immune responses, including periodontitis [2]. Several single nucleotide polymorphisms have been identified in its promoter [3]. Of these polymorphisms, G-to-A substitutions at positions −308 and −238 have been intensively studied; these allelic variations could be of functional significance [4]. TNF-α has been found at high levels in the gingival crevicular fluid and inflamed tissues in periodontitis [5]. TNF-α stimulates bone resportion by osteoclast [6] and induces tissue destruction by releasing of the matrix metalloproteinases, which are destructive to the extracellular matrix of ginviva, peritodontal ligaments and alveolar bone [7]. Numerous studies have examined the potential contributions made by TNF-α promoter polymorphisms to periodontitis susceptibility [8–20].

Interleukin-6 (IL-6) is a multifunctional cytokine involved in the inflammatory response and in the modulation of immune responses, including B cell and T cell differentiation [21]. IL-6 is regulated mainly at the transcriptional level by regulatory elements in the 5′ flanking region of the gene [22]. In this region, the −174 G/C and −572 G/C polymorphisms are important regulators of transcription [23]. For example, a luciferase reporter vector assay showed that the −174 G construct was expressed significantly higher than the corresponding −174 C construct [24]. The IL-6 −572 G/C polymorphism lies near a potential glucocorticoid receptor element at position −557 to −552 [25]. Furthermore, the −572 G/C polymorphism has been reported to influence levels of circulating C-reactive protein and bone resorption markers in postmenopausal women, and to be associated with several diseases [26, 27]. Subsequently many studies did examine relationships between IL-6 and TNF-α polymorphisms and periodontitis [16–22, 24–26], but studies have produced diverse results.

A meta-analysis provides a powerful means of overcoming the problem of small sample size and inadequate statistical power, especially in genetic studies of complex traits [28–31]. In the present study, we investigated whether the TNF-α promoter −308 A/G and −238 A/G, IL-6 −174 G/C and −572 G/C polymorphisms contribute to the susceptibility of periodontitis by performing a meta-analysis using published data.

Materials and methods

Identification of eligible studies and data extraction

A literature search was conducted for studies that examined associations between the TNF-α promoter −308 A/G and −238 A/G polymorphisms and IL-6 polymorphisms and periodontitis. We utilized the PubMed and Embase citation indexes to identify articles in which the TNF-α polymorphisms were determined in periodontitis patients and controls (up to July 2012). In addition, all references mentioned in the identified articles were reviewed to identify additional studies not indexed by PubMed and Embase. The following key words and subject terms were searched: ‘Tumor necrosis factor’, ‘TNF-alpha’, ‘polymorphism’, “interleukin-6,” “IL-6,” and ‘periodontitis’. The following information was extracted from each study: author, year of publication, ethnicity of the study population, demographics, numbers of cases and controls, and the A allele frequencies of the TNF-α promoter −308 A/G and −238 A/G polymorphisms. Allele frequencies were calculated from the corresponding genotype distributions.

Evaluation of statistical associations

A Chi squared test was used to determine whether the observed genotype frequencies conformed to Hardy–Weinberg (H–W) expectations (http://ihg2.helmholtz-muenchen.de/cgi-bin/hw/hwa1.pl). Meta-analyses were performed using allelic contrast, homozygote contrast, recessive model, and homozygote model. Point estimates of risk, odds ratios (ORs), and 95 % confidence intervals (CIs) were determined for each study. Cochran’s Q-statistic was used to assess within and between study variations and heterogeneities [32]. When the heterogeneity test assessed the probability of the null hypothesis, all studies were evaluating the same effect. If a significant Q-statistic (p < 0.10) indicated heterogeneity across studies, the random effects model was used for the meta-analysis, and when heterogeneity was not indicated across studies, the fixed effects model was used. The fixed effects model assumes that genetic factors have similar effects on periodontitis susceptibility across all studies, and that observed variations between studies are caused by chance alone [33]. The random effects model assumes that different studies show substantial diversity, and assesses both within-study sampling errors and between-study variances [34]. The random effects model was used in the presence of significant between-study heterogeneity. The effect of heterogeneity was quantified using a recently developed measure, I2 = 100 % × (Q − df)/Q, [35] where I2 ranges between 0 and 100 % and represents the proportion of inter-study variability attributable to heterogeneity rather than chance. I2 values of 25, 50, and 75 % were defined as low, moderate, and high estimates, respectively. Statistical manipulations were undertaken using the Comprehensive Meta-Analysis computer program (Biosta, Englewood, NJ, USA).

Evaluation of publication bias

Funnel plots are used to detect publication bias, but they require a range of studies of varying sizes and subjective judgments. Thus, in the present study, we evaluated publication bias using Egger’s linear regression test [36], which measures funnel plot asymmetry using a natural logarithm scale of ORs.

Results

Studies included in the meta-analysis

Thirteen studies met the inclusion criteria [8–20]. However, one of these studies contained data of two different groups such as aggressive or chronic periodontitis, and they were treated independently [10]. Fourteen and three separate comparisons of the TNF-α −308 A/G and −238 A/G polymorphisms were considered in the meta-analysis. A total of 17 separate comparisons were considered in our meta-analysis consisting of 1,120 patients with periodontitis and 1,592 controls (14 studies with 884 cases and 1,160 controls for the TNF-α −308 A/G polymorphism; three studies with 236 cases and 432 controls for the TNF-α −238 A/G polymorphism), involving six European, five Asian, four Brazilian, and two Turkish populations. Ethnicity-specific meta-analysis was conducted on these populations. Selected characteristics of the relationships found between the TNF-α polymorphisms and periodontitis are summarized in Table 1.

Table 1 Characteristics of the studies included in the meta-analysis
Full size table

Seventeen studies met the inclusion criteria [8, 9, 14, 37–50]. However, one of these studies contained data on three different groups for the IL-6 −174 G/C polymorphism and the IL-6 −572 G/C polymorphism [41], and three studies contained the data of both the IL-6 −174 G/C and −572 G/C polymorphisms [16, 17, 27]; these studies were treated independently. A total of 26 separate comparisons were considered in our meta-analysis consisting of 3,323 patients with periodontitis and 2,488 controls (16 studies with 1,887cases and 1,490 controls for the IL-6 −174 G/C polymorphism; 10 studies with 1,436 cases and 998 controls for the IL-6 −572 G/C polymorphism), involving seven European, three Asian, two Brazilian, two Indian, one African, and one Turkish population for the −174 G/C polymorphism and two European, six Asian, one Indian, and one African population for the −572 G/C polymorphism. Ethnicity-specific meta-analysis was conducted on these populations. Selected characteristics of the relationships found between the IL-6 polymorphisms and periodontitis are summarized in Table 2.

Table 2 Characteristics of the studies included in the meta-analysis
Full size table

Frequencies of the A alleles of the TNF-α promoter −308 A/G polymorphism in the different control groups

The study by Babel et al. showed AA and AG + GG genotype data, but not allele data [14]. Thus, the allele data from the study by Babel et al. was not included in Table 2. The mean frequency of the A allele of the TNF-α −308 A/G polymorphism was 16.0 % among all controls. Asian controls had a lower A allele prevalence (11.6 %) than the controls of the other ethnic groups, in which frequencies varied from 12.9 to 27.1 %. The Brazilian control population had the highest A allele frequency (27.1 %) (Table 3).

Table 3 Prevalences of the A allele of the TNF-α −308 G/A polymorphism
Full size table

Meta-analysis of the association between the TNF-α −308 A/G polymorphism and periodontitis

Meta-analysis was performed on all periodontitis patients and on the patients in each ethnic group. A summary of our meta-analysis findings of the relation between the TNF-α −308 A/G polymorphism and periodontitis is provided in Table 4. Meta-analysis showed no association between the A allele and periodontitis in all subjects (OR = 0.846, 95 % CI = 0.629–1.138, p = 0.269) with between-study heterogeneity (I2 = 49.0 %) (Table 4; Fig. 1). Analysis after stratification by ethnicity indicated that the A allele was associated with periodontitis in Brazilian, Asian and Turkish populations (in respective order: OR = 0.637, 95 % CI = 0.447–0.907, p = 0.013; OR = 0.403, 95 % CI = 0.204–0.707, p = 0.009; OR = 1.818, 95 % CI = 1.036–3.189, p = 0.037), but not in Europeans (OR = 1.034, 95 % CI = 0.772–1.385, p = 0.822) (Table 4; Fig. 2). Stratification by disease type indicated no association between the −308 A allele and aggressive or chronic periodontitis (Table 4).

Table 4 Meta-analysis of the association between the TNF-α −308 G/A polymorphism and periodontitis
Full size table
Fig. 1
figure 1

ORs and 95 % CIs of individual studies and pooled data for the association between the A allele of the TNF-α −308 (a) and TNF-α −238 (b) and periodontitis in all study subjects

Full size image
Fig. 2
figure 2

ORs and 95 % CIs of individual studies and of pooled data for the association between the A allele of the TNF-α –308 and periodontitis in each ethnic group

Full size image

Meta-analysis of the association between the TNF-α −238 A/G polymorphism and periodontitis

Meta-analysis showed no association between the A allele of the TNF-α −238 A/G polymorphism and periodontitis in all patients (OR 0.757, 95 % CI 0.309–1.740, p = 0.512) without between-study heterogeneity (I2 = 8.56 %) (Table 5). Meta-analyses of the TNF-α −238 A/G polymorphism using the dominant, additive model, or homozygote model showed the same pattern as meta-analysis of the TNF-α −238 A allele, indicating no significant association with periodontitis (Table 5).

Table 5 Meta-analysis of the association between the TNF-α −238 G/A polymorphism and periodontitis
Full size table

Meta-analysis of the IL-6 −174 G/C polymorphism and periodontitis susceptibility

Meta-analysis containing studies in HWE revealed an association between periodontitis and the IL-6 −174 G allele (OR = 1.387, 95 % CI = 1.040–1.849, p = 0.026) (Fig. 3; Table 6). Stratification by ethnicity and disease type indicated no association between the IL-6 −174 G allele and periodontitis in each ethnic group and chronic periodontitis (Table 6). Meta-analysis using the recessive and the homozygote models showed an association of the IL-6 −174 G/C polymorphisms with periodontitis in Brazilian populations (OR for GG + GC = 2.394, 95 % CI = 1.081–5.302, p = 0.031; OR for GG vs. CC = 2.498, 95 % CI = 1.108–5.628, p = 0.027, respectively), but not in European, Asian, and Indian populations (Table 6).

Fig. 3
figure 3

ORs and 95 % CIs of the individual studies and pooled data for the association between the G allele of the IL-6 promoter −174 G/C and periodontitis in all study subjects in HWE

Full size image
Table 6 Meta-analysis of the association between the IL-6 promoter −174 G/C polymorphism and periodontitis
Full size table

Meta-analysis of the IL-6 −572 G/C polymorphism and periodontitis susceptibility

Meta-analysis stratified by ethnicity and disease type indicated an association between the IL-6 −572 G allele and chronic periodontitis (OR = 1.585, 95 % CI = 1.030–2.439, p = 0.036), and periodontitis in the European group (OR = 2.118, 95 % CI = 1.254–3.577, p = 0.005) (Fig. 4; Table 7). Analysis using the recessive model showed the same pattern for the IL-6 −174 G allele, showing an association between the IL-6 −572 GG genotype and chronic periodontitis, and periodontitis in the European group (Table 7).

Fig. 4
figure 4

ORs and 95 % CIs of individual studies and pooled data for the association between the GG genotype of the IL-6 promoter −572 G/C and periodontitis in each ethnic group

Full size image
Table 7 Meta-analysis of the association between the IL-6 promoter −572 G/C polymorphism and periodontitis
Full size table

Heterogeneity and publication bias

Between-study heterogeneity was found during analyses of the TNF-α −308 A/G polymorphism, such as −308 A allele and the AA + GA genotype in overall and aggressive periodontitis, the AA +GA genotype in chronic periodontitis (Table 4). However, there was no between-study heterogeneity during analyses of the TNF-α −238 A/G polymorphisms (Table 4). The distribution of the TNF-α −308 A/G polymorphism in normal controls was not consistent with H–W equilibrium in one study [11]. Deviation from H–W equilibrium among controls implies potential bias during control selection or genotyping errors. However, excluding the study did not materially affect our results.

The distribution of genotypes of the IL-6 polymorphisms in control groups was consistent with the HWE, except for four studies on the IL-6 −174 G/C polymorphism [39, 40, 44, 45]. There were two studies that did not show HWE [14, 42]. When we excluded these studies, overall results were substantially affected (OR = 1.387, 95 % CI = 1.040–1.849, p = 0.026) (Table 6). Between-study heterogeneity for the IL-6 −174 G/C polymorphism was found among all study subjects and some subgroup analyses. Accordingly, meta-analysis was performed using a random effects model when heterogeneity was present in a population (Table 7). The funnel plot, which is usually used to detect publication bias, was difficult to interpret because we included a relatively small number of studies. Egger’s regression test showed no evidence of publication bias in the meta-analysis (Egger’s regression test p values > 0.1).

Discussion

In the present study, we addressed associations between the TNF-α promoter −308 and −238 A/G polymorphisms and periodontitis susceptibility. Ethnicity-specific meta-analysis showed an association between the TNF-α −308 A allele and periodontitis in Brazilian, Asian, and Turkish populations but not in Europeans. Stratification by disease type indicated no association between the −308 A allele and aggressive or chronic periodontitis and no association was found between periodontitis and the TNF-α −238 A/G polymorphism. We found an association an association between periodontitis and the IL-6 −174 G allele in all study subjects in HWE (OR = 1.387, 95 % CI = 1.040–1.849, p = 0.026). Ethnicity-specific meta-analysis showed an association of the IL-6 −174 G/C polymorphisms with periodontitis in Brazilian populations. We identified an association of the IL-6 −572 G/C polymorphisms with periodontitis in all study subjects (OR for GG + GC = 1.507, 95 % CI = 1.050–2.162). Furthermore, stratification by ethnicity and disease type indicated an association between the IL-6 −572 G allele and chronic periodontitis, and periodontitis in Europeans.

TNF-α is increased in the gingival crevicular fluid and inflamed tissues in periodontitis. Our meta-analysis showed that the TNF-α −308 A/G polymorphism may influence the susceptibility to periodontitis in various ethnic groups. However, it needs to be explained why the TNF-α –308 A/G polymorphism is not associated with periodontitis in European population and plays a different role in different ethnic populations. It can be related with genetic heterogeneity in different populations. Clinical heterogeneity may also explain the discrepancy. Different linkage disequilibrium (LD) patterns may contribute to the discrepancy. This polymorphism may be in LD with a nearby causal variant in one ethnic group but not in another [51]. The difference might arise purely by chance, such as type I error, or because of multiple testing which inflates the type I error. TNF polymorphisms may occur with distinct patterns in ethnically distinct populations [52]. The difference of the allele frequency of the TNF-α −308 A/G polymorphism exists among ethnic groups. In our study, Asian controls had a lower A allele prevalence (11.6 %) than the controls of the other ethnic groups, in which frequencies varied from 12.9 to 27.1 %. The Brazilian control population had the highest A allele frequency (27.1 %).

As IL-6 −174 G/C and −572 G/C polymorphisms increase IL-6 expression, they may be associated with susceptibility to periodontitis [5–7]. IL-6 transcription is influenced by complex interactions determined by these haplotypes. The IL-6 promoter region influences IL-6 gene transcription not by a simple mechanism but rather through complex interactions determined by the haplotype [21]. Because IL-6 polymorphisms and haplotypes are associated with an increased inflammatory response, specifically in the presence of periodontopathogenic bacteria, IL-6 gene may play a key role in the pathogenesis of periodontitis.

This meta-analysis differs from a previous meta-analysis on the relation between the IL-6 polymorphisms and periodontitis risk [53], because in the present study seven more studies were included, and meta-analysis stratified by ethnicity was further conducted. The result of this meta-analysis regarding an association between the IL-6 −174 G/C and −572 G/C polymorphisms and susceptibility to periodontitis is in agreement with the previous study. In addition, our meta-analysis revealed an association between IL-6 −174 G/C polymorphism and periodontitis in Brazilians, and between IL-6 −572 G/C polymorphism and periodontitis in Europeans.

The present study had some limitations that should be considered. First, publication bias and heterogeneity may have distorted the meta-analysis. Second, our ethnicity-specific meta-analysis included data only from European, Brazilian, Asian, and Turkish patients. Thus, our results are applicable only to these ethnic groups. Additional studies are required in other ethnic groups to determine whether the TNF-α polymorphisms confer a risk of periodontitis. Third, the numbers of subjects and studies included in the ethnicity-specific meta-analysis were small. There were only two studies for Turkish population. This analysis may not have enough power to explore the association between TNF-α –308 A/G polymorphism and periodontitis in each ethnic group. Small sample size may have distorted the meta-analysis of the TNF-α −238 A/G polymorphism. There were only three studies available for the meta-analysis of the TNF-α −238 A/G polymorphism. The number of studies included in the meta-analysis was too small to draw definitive conclusion. Interpretation of the data should be done cautiously. Fourth, data were not stratified by sex, disease severity, or environmental variables because of insufficient data.

In conclusion, this meta-analysis demonstrates that the TNF-α −308 A/G polymorphism confers susceptibility to periodontitis in Brazilian, Asian, and Turkish populations, and the IL-6 −174 G/C polymorphism may confer susceptibility to periodontitis in Brazilians, and that the IL-6 −572 G/C polymorphism is associated with susceptibility to periodontitis in Europeans, and chronic periodontitis. Further larger scale studies are required to explore the roles played by the TNF-α and IL-6 genes in the pathogenesis of periodontitis.