SEARCH

SEARCH BY CITATION

Keywords:

  • Interleukin-6;
  • type 2 diabetes mellitus;
  • polymorphism;
  • meta-analysis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

The association between the interleukin-6 (IL-6) gene −572 C/G (rs1800796) polymorphism and type 2 diabetes mellitus (T2DM) risk remains controversial. Thus, we performed this meta-analysis by searching PubMed, Embase, Web of Science, CBMdisc and CNKI databases until January 30, 2012. In addition, hand searching of the references of identified articles was performed. A total of 10 case–control studies including 11,681 subjects were selected to evaluate the possible association. Our results showed evidence for significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk (for G allele vs. C allele: odds ratio [OR] = 1.29, 95% confidence interval [CI] = 1.09–1.52, P = 0.002, P = 0.008 after Bonferroni testing; for G/G vs. C/C: OR = 1.89, 95% CI = 1.51–2.37, P < 0.00001, P < 0.00004 after Bonferroni testing; for GG vs. G/C + C/C: OR = 1.75, 95% CI = 1.20–2.56, P = 0.004, P = 0.016 after Bonferroni testing; for G/G + G/C vs. C/C: OR = 1.32, 95% CI = 1.11–1.57, P = 0.001, P = 0.004 after Bonferroni testing). In addition, similar results were obtained in the subgroup analysis based on ethnicity. In summary, the present meta-analysis suggests a significant association between the IL-6 gene −572 G allele and increased risk of T2DM.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Diabetes mellitus (DM) is one of the most common noncommunicable diseases that causes significant morbidity and mortality worldwide (Gulliford & Charlton, 2009; International Diabetes Federation, 2012). The International Diabetes Federation estimates that 366 million adults worldwide have DM in 2011, and this number will increase to 552 million by 2030 (International Diabetes Federation, 2012). As the main type of DM, type 2 diabetes mellitus (T2DM) seriously affects the quality of life of the patients, and imposes a large economic burden on the national health care system (Borges et al., 2011; International Diabetes Federation, 2012). T2DM is a complex disease with various involved factors (Mathias et al., 2009). Previous studies have reported that the prevalence rates of T2DM vary substantially throughout the world (International Diabetes Federation, 2012), and relatives of T2DM patients are at greater risk for developing the disease than the general population (Newman et al., 1987; Mathias et al., 2009; Almgren et al., 2011). Newman et al. found that 58% of monozygotic co-twins of diabetic twins were themselves diabetic compared with an expected prevalence of 10% (Newman et al., 1987). Stumvoll et al. found that the heritability of T2DM was estimated to range from 30% to 70% depending on the population investigated (Stumvoll et al., 2005). Furthermore, in 2011, Almgren et al. found the strongest heritability for T2DM in patients with age at onset 35–60 years (h2 = 0.69) (Almgren et al., 2011). Therefore, genetic factors play a key role in the development of T2DM.

It is well known that inflammatory response is an essential part of the pathogenesis of T2DM ( Pickup et al., 1997; Festa et al., 2000; Fernández-Real & Ricart, 2003). Several epidemiological studies have recently demonstrated increased serum levels of inflammatory markers, such as interleukin-6 (IL-6), being associated with increased risk of T2DM (Pradhan et al., 2001; Spranger et al., 2003), suggesting that genetic factors involving cytokines may play an important role in the development of T2DM.

IL-6, as a multifunctional cytokine secreted by both immune cells and adipose tissue (Mohamed-Ali et al., 1997; Fried et al., 1998), plays a key role in the inflammatory response that is associated with insulin-resistant states and T2DM (Pickup et al., 1997). The gene encoding IL-6 is located on chromosome 7p21, and displays a single nucleotide polymorphism (SNP) in the promoter region (−572 C/G (rs1800796), formerly known as -634 C/G) which could affect IL-6 gene transcription and its serum levels (Brull et al., 2001; Kitamura et al., 2002). Recently, a variety of epidemiological studies have been performed to evaluate the association between IL-6 gene −572 C/G polymorphism and T2DM risk in diverse populations. However, results of different studies have been inconsistent. Thus, whether the IL-6 gene −572 C/G polymorphism is associated with increased T2DM risk remains controversial. To better clarify the association between the IL-6 gene −572 C/G polymorphism and T2DM risk, we performed this meta-analysis by collecting and sorting the previously published studies.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Data Sources

This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) criteria (Moher et al., 2009). We identified relevant studies by systematically searching PubMed, Embase, Web of Science, CBMdisc and CNKI databases from the first available year to January 30, 2012. The search terms used were: (“interleukin-6” or “IL-6”) and (“polymorphism” or “mutation” or “variant” or “genotype”) and (“type 2 diabetes mellitus” or “type 2 diabetes” or “diabetes mellitus” or “diabetic patients” or “T2DM”). In addition, hand searching of the references of identified articles was performed.

Inclusion Criteria

To enter this meta-analysis, studies had to satisfy the following criteria: (i) Evaluation of the association between the IL-6 gene −572 C/G polymorphism and T2DM risk; (ii) Study was designed using the methodology of a case–control study; (iii) Published studies with full text articles; (iv) There was sufficient data for the computation of odds ratios (ORs) and corresponding 95% confidence intervals (CIs) and (v) Contain original, not republished, data.

Data Extraction

Two authors (Yin YW and Sun QQ) of this article independently extracted data from included studies, using a predetermined extraction form, for (i) name of the first author; (ii) year of publication; (iii) country of origin; (iv) ethnicity of the studied population; (v) source of controls; (vi) number of cases and controls and (vii) information of genotype and allele frequency. In case of disagreement, a third author (Hu AM) examined such articles, and the assessors’ findings were discussed until consensus was reached. In addition, evidence of Hardy–Weinberg equilibrium (HWE) was collected (P < 0.05 was considered significant deviation from HWE).

Quality Assessment for Individual Studies

The quality of all eligible studies was evaluated independently by two authors (Liu HL and Wang Q) of this article using the Newcastle–Ottawa Scale (NOS), which is applicable for both case–control and cohort studies (Wells et al., 2011). The NOS ranges between zero and nine stars. We defined studies of high quality as those that scored the maximum nine stars on the NOS; studies of medium quality scored seven or eight stars. In case of disagreement, the authors achieved consensus through discussion.

Statistical Methods

Crude ORs with their 95% CI were estimated and used to assess the strength of association between the IL-6 gene −572 C/G polymorphism and T2DM risk. The pooled ORs were obtained from combinations of single studies by the comparisons of allelic model (G allele vs. C allele), additive model (G/G vs. C/C), recessive model (G/G vs. G/C + C/C) and dominant model (G/G + G/C vs. C/C), respectively. Cochran's Q statistic and the I2 statistic were used to assess statistical heterogeneity among studies (P < 0.10 and I2 > 50% indicated evidence of heterogeneity) (Berkey et al., 1995; Higgins et al., 1995). If there was no statistical heterogeneity among studies, the fixed-effects model was used to estimate the summary ORs. Otherwise, the random-effects model was used (Mantel & Haenszel, 1959; DerSimonian & Laird, 1986). The Galbraith plot was used to detect potential sources of heterogeneity.

Considering that potential ethnic difference might be associated with the distribution of genotypes, we also performed subgroup analysis by ethnicity of study population. Sensitivity analysis was performed by limiting the meta-analysis to studies conforming to HWE. Publication bias was examined with Begg's funnel plot and Egger's regression test, with P < 0.05 taken as an indication of publication bias (Egger et al., 1997). Moreover, to adjust for multiple comparisons, we applied the Bonferroni method, which controls for false positive error rate. All the statistical tests were performed using the software Review Manager 5.1.2 (Cochrane Collaboration, The Nordic Cochrane Centre, Copenhagen) and Stata 11.0 (StataCorp LP, College Station, TX).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Characteristics of Included Studies

Figure 1 presents a flow chart of the article search and inclusion process. Based on our search strategy, 954 potentially eligible articles were identified in our initial search. From the titles and, when available, the abstracts of these articles, 68 articles were deemed relevant. Of these, 58 articles were excluded through reading the full text. Subsequently, 10 articles, with a total of 4335 T2MD cases and 7346 controls were included in the final meta-analysis (Kitamura et al., 2002; Deng et al., 2004; Hamid et al., 2005; An et al., 2007; Chen et al., 2009; Koh et al., 2009; Xiao et al., 2009; Tang et al., 2010; Wang et al., 2010; Zhang et al., 2011). From the 10 articles, nine focused on Asians, and one on Europeans. The countries of these studies included China, Denmark, Japan and Korea. The genotype distributions among the controls of all studies followed the HWE except for the studies of Hamid et al. and Wang et al. (P < 0.05) (Hamid et al., 2005; Wang et al., 2010). The NOS results showed that the average score was 7.7, which indicated that the methodological quality was generally good. Table 1 shows the studies included in the meta-analysis and their main characteristics.

Table 1. Characteristics of studies included in this meta-analysis
      Genotypes distribution (case/control)  
First authorYearCountryEthnicitySource of controlsSample size (case/control)C/CG/CG/GCGHWE Y/N (P)Score
  1. PB, population-based; HWE, Hardy–Weinberg equilibrium; Y, yes; N, no.

Kitamura2002JapanAsianPB454/142304/102122/3628/4730/240178/44Y(0.705)8
Deng2004ChinaAsianPB267/126180/9169/3218/3429/214105/38Y(0.925)8
Hamid2005DenmarkEuropeanPB1361/43825/20123/3251233/4037133/3652589/8399N(0.000)7
An2007ChinaAsianPB246/101113/6090/3742/4316/157174/45Y(0.561)8
Xiao2009ChinaAsianPB198/134114/9675/369/2303/22893/40Y(0.503)8
Chen2009ChinaAsianPB183/10537/2894/5256/25168/108206/102Y(0.929)8
Koh2009KoreaAsianPB476/1477275/862164/54737/68714/2271238/683Y(0.109)8
Tang2010ChinaAsianPB280/7090/38138/2652/6318/102242/38Y(0.610)7
Wang2010ChinaAsianPB358/326212/208129/11217/6553/528163/124N(0.037)8
Zhang2011ChinaAsianPB512/483253/244210/19749/42716/685308/281Y(0.803)7
image

Figure 1. Flow diagram of the study selection process.

Download figure to PowerPoint

Quantitative Synthesis

The association between the IL-6 gene −572 C/G polymorphism and T2DM risk was investigated in 10 studies with a total of 4335 T2MD cases and 7346 controls. The results of the present meta-analysis showed that there was significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk (for G allele vs. C allele: OR = 1.29, 95% CI = 1.09–1.52, P = 0.002, P = 0.008 after Bonferroni testing; for G/G vs. C/C: OR = 1.89, 95% CI = 1.51–2.37, P < 0.00001, P < 0.00004 after Bonferroni testing; for GG vs. G/C + C/C: OR = 1.75, 95% CI = 1.20–2.56, P = 0.004, P = 0.016 after Bonferroni testing; for G/G + G/C vs. C/C: OR = 1.32, 95% CI = 1.11–1.57, P = 0.001, P = 0.004 after Bonferroni testing). The main results of the meta-analysis are shown in Table 2 and Figure 2.

Table 2. Meta-analysis of the IL-6 gene −572 C/G polymorphism and the risk of T2DM in each subgroup
  G versus CG/G versus C/CG/G versus G/C + C/CG/G + G/C versus C/C
CategorySample size (case/control)OR (95% CI)P2OR (95% CI)P2OR (95% CI)P2OR (95%CI)P2
  1. 1Significant heterogeneity: the random-effects model was chosen to summarize the result.

  2. 2P values for heterogeneity from Q-test.

  3. BH: based on HWE (Studies without HWE were excluded); BE: based on language (Studies in the English language); BC: based on language (Studies in the Chinese language).

Overall4335/73461.29 [1.09,1.52]10.00031.89 [1.51,2.37]0.11001.75 [1.20,2.56]10.000101.32 [1.11,1.57]10.0500
Asians2974/29641.35 [1.16,1.58]1< 0.00012.13 [1.50,3.01]1< 0.00011.78 [1.42,2.21]1< 0.000011.33 [1.11,1.59]10.0300
Sensitivity analysis         
 BH2616/26381.38 [1.16,1.65]10.00032.09 [1.44,3.05]10.08001.73 [1.38,2.17]1< 0.000011.37 [1.11,1.69]10.0300
 BE3001/66181.12 [0.93,1.35]10.02001.50 [1.14,1.98]0.39001.32 [0.85,2.05]10.00501.12 [0.97,1.29]0.2100
 BC1334/7281.48 [1.27,1.73]0.17002.90 [1.94,4.33]0.40002.25[1.56,3.23]0.22001.46[1.20,1.77]0.1900
image

Figure 2. Forest plots for the IL-6 gene −572 C/G polymorphism and T2DM risk in different genetic models. (A) allelic model: G allele versus C allele; (B) additive model: G/G versus C/C; (C) recessive model: G/G versus G/C + C/C; (D) dominant model: G/G + G/C versus C/C.

Download figure to PowerPoint

In the subgroup analysis by ethnicity, we only analysed the Asian population as only one study involved Europeans. We obtained similar results, suggesting that there was significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk (for G allele vs. C allele: OR = 1.35, 95% CI = 1.16–1.58, P < 0.0001, P < 0.0004 after Bonferroni testing; for G/G vs. C/C: OR = 2.13, 95% CI = 1.50–3.01, P < 0.0001, P < 0.0004 after Bonferroni testing; for GG vs. G/C + C/C: OR = 1.78, 95% CI = 1.42–2.21, P < 0.00001, P < 0.00004 after Bonferroni testing; for G/G + G/C vs. C/C: OR = 1.33, 95% CI = 1.11–1.59, P = 0.002, P = 0.008 after Bonferroni testing). The main results of the subgroup analysis are shown in Table 2.

Sensitivity Analysis

We first performed sensitivity analysis by limiting the meta-analysis to studies conforming to HWE. Two studies without HWE (P < 0.05) were excluded from the sensitivity analysis (Hamid et al., 2005; Wang et al., 2010), and none of the results were materially altered. We next performed sensitivity analyses by limiting the meta-analysis to studies published in the English or Chinese language. Although the results were materially altered in the allelic model, recessive model and dominant model by limiting the studies to those in the English language, a positive result was still obtained in the additive model. Furthermore, the results were not materially altered in the sensitivity analysis based on studies in the Chinese language. Therefore, the most noteworthy finding was that the above results of sensitivity analyses further strengthened the conclusion that the IL-6 gene −572 G allele could increase the risk of T2DM. The results of sensitivity analyses are shown in Table 2.

Heterogeneity Analysis

We detected significant between-study heterogeneity in the allelic model (PQ = 0.0003, I2 = 71%), the recessive model (PQ = 0.0001, I2 = 73%) and the dominant model (PQ = 0.05, I2 = 46%). In contrast, the additive model did not present significant heterogeneity (PQ = 0.11, I2 = 37%). To explore the sources of heterogeneity, we first performed subgroup and sensitivity analyses. However, we failed to remove the heterogeneity (Table 2).We next created a Galbraith plot to graphically assess the sources of heterogeneity. Three studies were identified as the main contributor to heterogeneity in the allelic model (Hamid et al., 2005; An et al., 2007; Tang et al., 2010 ), two studies were identified as the main contributor to heterogeneity in the recessive model (Hamid et al., 2005; An et al., 2007) and one study was identified as the main contributor to heterogeneity in the dominant model (Tang et al., 2010) (Fig. 3). After excluding the outlier studies, the heterogeneity was effectively removed (for the allelic model: PQ = 0.32, I2 = 15%; for the recessive model: PQ = 0.40, I2 = 4%; for the dominant model: PQ = 0.30, I2 = 16%), respectively (Fig. 4).

image

Figure 3. Galbraith plot for the IL-6 gene −572 C/G polymorphism and T2DM risk. (A) The studies of Tang et al. (2010), Xiao et al. (2009) and An et al. (2007) were outliers in the allelic model. (B) The studies of An et al. (2007) and Hamid et al. (2005) were outliers in the recessive model. (C) The study of Tang et al. (2010) was an outlier in the dominant model.

Download figure to PowerPoint

image

Figure 4. Forest plots for the IL-6 gene −572 C/G polymorphism and T2DM risk after excluding the outlier studies. (A) allelic model: G allele versus C allele; (B) recessive model: G/G versus G/C + C/C; (C) dominant model: G/G + G/C versus C/C.

Download figure to PowerPoint

Publication Bias

The shapes of the funnel plots revealed obvious asymmetry in all genetic models (Fig. 5), suggesting that there was obvious publication bias. Meanwhile, the results of Egger's regression test also provided sufficient evidence for publication bias (P = 0.004 for allelic model, P = 0.020 for additive model, P = 0.000 for recessive model and P = 0.017 for dominant model, respectively).

image

Figure 5. Funnel plots for the IL-6 gene −572 C/G polymorphism and T2DM risk. (A) allelic model: G allele versus C allele; (B) additive model: G/G versus C/C; (C) recessive model: G/G versus G/C + C/C; (D) dominant model: G/G + G/C versus C/C.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References

Several epidemiological studies have reported the associations between several genetic variants of IL-6 and risk of T2DM, for instance, -174G/C, −572C/G, -597G/A and -1363G/T (Kitamura et al., 2002; Deng et al., 2004; Hamid et al., 2005; An et al., 2007; Chen et al., 2009; Koh et al., 2009; Xiao et al., 2009; Tang et al., 2010; Wang et al., 2010; Zhang et al., 2011). Of these studies, although the association between the IL-6 gene −572 C/G polymorphism and T2DM risk has been intensively studied, the results have been inconsistent. Most studies demonstrated that the IL-6 gene −572 G allele was associated with an increased risk of T2DM (Deng et al., 2004; Hamid et al., 2005; An et al., 2007; Chen et al., 2009; Koh et al., 2009; Tang et al., 2010; Wang et al., 2010). However, several other studies showed that there was no significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk (Kitamura et al., 2002; Xiao et al., 2009; Zhang et al., 2011). A previous meta-analysis showed that the IL-6 gene -174 G/C polymorphism was not associated with T2DM risk (Qi et al., 2006). Here, we focused on the association between the IL-6 gene −572 C/G polymorphism and T2DM risk due to the scarcity of studies on other SNPs of the IL-6 gene. In 2006, Huth et al. (2006) performed a meta-analysis mainly based on unpublished studies and showed that the IL-6 gene −572 C/G polymorphism was not associated with T2DM risk. By comparison, the present meta-analysis obtained a conclusion inconsistent with that study. In this meta-analysis, we examined the IL-6 gene −572 C/G polymorphism and its relationship with the risk of T2DM in four genetic models. All of the results showed that there was significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk, suggesting that the G allele was associated with an increased risk of T2DM. The result of the allelic model showed that the risk of developing T2DM in G allele carriers was 1.29-fold higher than in those with the C allele. Furthermore, individuals with the G/G genotype had a significantly higher risk for developing T2DM (for OR = 1.89 in the additive model and OR = 1.75 in the recessive model) compared to those with the G/C genotype or the C/C genotype. Moreover, the result of the dominant model suggested that the risk of developing T2DM in G allele carriers was 1.32-fold higher than in those with the C/C genotype. Considering that multiple comparisons may cause a false positive error rate, we also applied the Bonferroni method to correct P values, and there was still significant association between the IL-6 gene −572 G allele and T2DM risk.

In the subgroup analysis by ethnicity, significant association between the IL-6 gene −572 C/G polymorphism and T2DM risk was observed among Asians. As for the Europeans, only one study was included in our meta-analysis; we therefore did not perform subgroup analysis among Europeans. Actually, the minor allele frequency (MAF) of the IL-6 gene −572 C/G is very low among Utah residents with Northern and Western European ancestry (CEU) (MAF = 0.04 in HapMap CEU, International HapMap Project, 2012). However, the IL-6 gene −572 C/G polymorphism is common among Han Chinese in Beijing, China (CHB) and among Japanese in Tokyo, Japan (JPT) (MAF = 0.23 and 0.14, respectively, International HapMap Project, 2012). Therefore, the abundance of studies among Asians and the scarcity of studies among Europeans could contribute to the genetic diversity among ethnicities. Furthermore, considering that the results produced from genetic association case–control studies may be spurious when the genotype distribution of controls deviates from HWE (Zintzaras & Lau, 2008), we performed sensitivity analysis by limiting the meta-analysis to studies conforming to HWE, and similar results to that of the overall study were obtained. The results of subgroup and sensitivity analyses further strengthened the conclusion that the IL-6 gene −572 C/G polymorphism was associated with increased risk of T2DM. Also, HWE should not be considered as a factor influencing the overall results.

Between-study heterogeneity should not be ignored in this meta-analysis. Attempts were made to explore the sources of heterogeneity by subgroup and sensitivity analyses. However, we did not effectively remove the heterogeneity, indicating the presence of other sources of heterogeneity. The heterogeneity had a wide range of potential sources, such as case definition and sampling, sample sizes, methods of genotyping and difference of ethnicity. To further clarify the reasons for heterogeneity, we then created a Galbraith plot to assess the sources of heterogeneity. Four studies were identified as the main contributors to heterogeneity, and the above-mentioned heterogeneity was therefore removed after excluding these outlier studies. Furthermore, the pooled ORs were not materially altered, which suggested that our results were statistically robust.

Publication bias is one of the most important sources of bias in meta-analysis, which might influence the interpretation of our final results supporting the role of the IL-6 gene −572 G allele in T2DM. Asymmetrical “missing” data, which was in the lower part of the funnel plot (Fig. 5), suggested that the negative results of studies between the IL-6 gene −572 G allele and T2DM may not have been reported or published. The preference to publish studies with positive results (significant association) rather than negative results (nonsignificant association) seems to have become the main source of publication bias in the current research. Moreover, only full text articles published in English and Chinese were included in this meta-analysis, thus some eligible studies which were unpublished or were reported in other languages may have been missed. Therefore, some inevitable publication bias may exist in the results. For this reason, projections from the literature of who is at risk for IL-6 gene attributable T2DM and who would benefit from IL-6 gene-targeted therapies should therefore be approached with caution.

Our studies had some limitations that need to be taken into consideration when interpreting the findings. First, this meta-analysis was based predominantly on studies carried out in the Asian population. Only one study involved Europeans, and no studies from other parts of the world were found, which may have led to incomplete results. Secondly, there was considerable heterogeneity among the included studies. Heterogeneity may affect the precision of results, despite the use of appropriate meta-analytic techniques with a random-effects model. Thirdly, there was obvious publication bias in the present meta-analysis, which can cause false positive results. Moreover, we could not obtain sufficient original data and therefore were unable to explore whether the susceptibility to T2DM was affected by individual factors such as age, sex and smoking habits.

In conclusion, this meta-analysis indicated that the IL-6 gene −572 G allele was associated with increased T2DM risk. SNPs in this locus may be considered to act as candidate biomarkers for T2DM screening, diagnosis and therapy in the future. However, the result should be interpreted with caution because of its limitations. We anticipate that the results reported here will stimulate further well-designed studies to advance exploration in this field.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. References
  • Almgren, P., Lehtovirta, M., Isomaa, B., Sarelin, L., Taskinen, M. R., Lyssenko, V., Tuomi, T., & Groop, L.; Botnia Study Group. (2011) Heritability and familiality of type 2 diabetes and related quantitative traits in the Botnia Study. Diabetologia 54, 28112819.
  • An, X. H., Song, D. P., Liu, H., Wang, Y. M. & Duan, Y. (2007) The association of IL-6 gene promoter region -643 C/G polymorphism with diabetic nephropathy. Chin J Diabetes 15, 289291.
  • Berkey, C. S., Hoaglin, D. C., Mosteller, F. & Colditz, G. A. (1995) A random-effects regression model for meta-analysis. Stat Med 14, 395411.
  • Borges, A. P., Guidoni, C. M., de Freitas, O. & Pereira, L. R. (2011) Economic evaluation of outpatients with type 2 diabetes mellitus assisted by a pharmaceutical care service. Arq Bras Endocrinol Metabol 55, 686691.
  • Brull, D. J., Montgomery, H. E., Sanders, J., Dhamrait, S., Luong, L., Rumley, A., Lowe, G. D. & Humphries, S. E. (2001) Interleukin-6 gene –174G > C and –572G > C promoter polymorphisms are strong predictors of plasma interleukin-6 levels after coronary artery bypass surgery. Arterioscler Thromb Vasc Biol 21, 14581463.
  • Chen, J. N., Yin, Y. S. & Li, X. L. (2009) The genotype of IL-6 gene in patients with type 2 diabetic nephropathy. Acta Med Sin 22, 801804.
  • Deng, Y. P., Xu, Y. C. & Zhang, Y. (2004) Association of IL-6 gene polymorphism with diabetic nephropathy in type 2 diabetes. J N Med 14, 2123.
  • DerSimonian, R. & Laird, N. (1986) Meta-analysis in clinical trials. Control Clin Trials 7, 177188.
  • Egger, M., Davey Smith, G., Schneider, M. & Minder, C. (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315, 629634.
  • Fernández-Real, J. M. & Ricart, W. (2003) Insulin resistance and chronic cardiovascular inflammatory syndrome. Endocr Rev 24, 278301.
  • Festa, A., D'Agostino, R. Jr., Howard, G., Mykkänen, L., Tracy, R. P. & Haffner SM. (2000) Chronic subclinical inflammation as part of the insulin resistance syndrome: The insulin resistance atherosclerosis study (IRAS). Circulation 102, 4247.
  • Fried, S. K., Bunkin, D. A. & Greenberg, A. S. (1998) Omental and subcutaneous adipose tissues of obese subjects release interleukin-6: Depot difference and regulation by glucocorticoid. J Clin Endocrinol Metab 83, 847850.
  • Gulliford, M. C. & Charlton, J. (2009) Is relative mortality of type 2 diabetes mellitus decreasing? Am J Epidemiol 169, 455461.
  • Hamid, Y. H., Rose, C. S., Urhammer, S. A., Glümer, C., Nolsøe, R., Kristiansen, O. P., Mandrup-Poulsen, T., Borch-Johnsen, K., Jorgensen, T., Hansen, T. & Pedersen O. (2005) Variations of the interleukin-6 promoter are associated with features of the metabolic syndrome in Caucasian Danes. Diabetologia 48, 251260.
  • Higgins, J. P., Thompson, S. G., Deeks, J. J. & Altman, D. G. (1995) Measuring inconsistency in meta-analyses. BMJ 327, 557560.
  • Huth, C., Heid, I. M., Vollmert, C., Gieger, C., Grallert, H., Wolford, J. K., Langer, B., Thorand, B., Klopp, N., Hamid, Y. H., Pedersen, O., Hansen, T., Lyssenko, V., Groop, L., Meisinger, C., Döring, A., Löwel, H., Lieb, W., Hengstenberg, C., Rathmann, W., Martin, S., Stephens, J. W., Ireland, H., Mather, H., Miller, G. J., Stringham, H. M., Boehnke, M., Tuomilehto, J., Boeing, H., Möhlig, M., Spranger, J., Pfeiffer, A., Wernstedt, I., Niklason, A., López-Bermejo, A., Fernández-Real, J. M., Hanson, R. L., Gallart, L., Vendrell, J., Tsiavou, A., Hatziagelaki, E., Humphries, S. E., Wichmann, H. E., Herder, C., Illig, T. (2006) IL6 gene promoter polymorphisms and type 2 diabetes: Joint analysis of individual participants’ data from 21 studies. Diabetes 55, 29152921.
  • International Diabetes Federation. (2012) Available at: http://www.diabetesatlas.org/content/diabetes. Accessed Feb. 10, 2012.
  • International HapMap Project. (2012) Available at: http://hapmap.ncbi.nlm.nih.gov. Accessed June 25, 2012.
  • Kitamura, A., Hasegawa, G., Obayashi, H., Kamiuchi, K., Ishii, M., Yano, M., Tanaka, T., Yamaguchi, M., Shigeta, H., Ogata, M., Nakamura, N. & Yoshikawa, T. (2002) Interleukin-6 polymorphism (–634C/G) in the promoter region and the progression of diabetic nephropathy in type 2 diabetes. Diabet Med 19, 10001005.
  • Koh, S. J., Jang, Y., Hyun, Y. J., Park, J. Y., Song, Y. D., Shin, K. K., Chae, J. S., Kim, B. K., Ordovas, J. M. & Lee, J. H. (2009) Interleukin-6 (IL-6) -572C–>G promoter polymorphism is associated with type 2 diabetes risk in Koreans. Clin Endocrinol (Oxf) 70, 238244.
  • Mantel, N. & Haenszel, W. (1959) Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22, 719748.
  • Mathias, R. A., Deepa, M., Deepa, R., Wilson, A. F. & Mohan, V. (2009) Heritability of quantitative traits associated with type 2 diabetes mellitus in large multiplex families from South India. Metabolism 58, 14391445.
  • Mohamed-Ali, V., Goodrick, S., Rawesh, A., Katz, D. R., Miles, J. M., Yudkin, J. S., Klein, S. & Coppack, S. W. (1997) Subcutaneous adipose tissue releases interleukin-6, but not tumor necrosis factor-alpha, in vivo. J Clin Endocrinol Metab 82, 41964200.
  • Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G.; PRISMA Group. (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Ann Intern Med 151,264269.
  • Newman, B., Selby, J. V., King, M. C., Slemenda, C., Fabsitz, R., Friedman, G. D. (1987) Concordance for type 2 (non-insulin-dependent) diabetes mellitus in male twins. Diabetologia 30, 763768.
  • Pickup, J. C., Mattock, M. B., Chusney, G. D. & Burt, D. (1997) NIDDM as a disease of the innate immune system: Association of acute-phase reactants and interleukin-6 with metabolic syndrome X. Diabetologia 40, 12861292.
  • Pradhan, A. D., Manson, J. E., Rifai, N., Buring J. E. & Ridker, P. M. (2001) C-reactive protein, interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA 286, 327334.
  • Qi, L., van, Dam, R. M., Meigs, J. B., Manson, J. E., Hunter, D. & Hu, F. B. (2006) Genetic variation in IL6 gene and type 2 diabetes: Tagging-SNP haplotype analysis in large-scale case-control study and meta-analysis. Hum Mol Genet 15, 19141920.
  • Spranger, J., Kroke, A., Mohlig, M., Hoffmann, K., Bergmann, M. M., Ristow, M., Boeing, H. & Pfeiffer, A. F. (2003) Inflammatory cytokines and the risk to develop type 2 diabetes: Results of the prospective population-based European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study. Diabetes 52, 812817.
  • Stumvoll, M., Goldstein, B. J. & van Haeften, T. W. (2005) Type 2 diabetes: Principles of pathogenesis and therapy. Lancet 365, 13331346.
  • Tang, S. F., Wei, H. Y. & Zhang, P. (2010) The polymorphism of the interleukin-6 promoter in patients with type 2 diabetes mellitus complicated with low-extremity vascular disease. Chin J Prev Contr Chron Dis 18, 156160.
  • Wang, Y. X., Kong, L. F. & Wang X. Y. (2010) The association of IL-6 gene 572C/G polymorphism and type 2 diabetes. J China Med Univ 39, 10451054.
  • Wells, G. A., Shea, B. & O'Connell, D. (2011) The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Ottawa Health Research Institute. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed Oct. 20, 2011.
  • Xiao, L. M., Yan, Y. X., Xie, C. J., Fan, W. H., Xuan, D. Y., Wang, C. X., Chen, L., Sun, S. Y., Xie, B. Y. & Zhang, J. C. (2009) Association among interleukin-6 gene polymorphism, diabetes and periodontitis in a Chinese population. Oral Dis 15, 547553.
  • Zhang, X., Ma, L., Peng, F., Wu, Y., Chen, Y., Yu, L., Lei, Z. & Zhang, C. (2011) The endothelial dysfunction in patients with type 2 diabetes mellitus is associated with IL-6 gene promoter polymorphism in Chinese population. Endocrine 40, 124129.
  • Zintzaras, E. & Lau, J. (2008) Synthesis of genetic association studies for pertinent gene-disease associations requires appropriate methodological and statistical approaches. J Clin Epidemiol 61, 634.