Associations between three common single nucleotide polymorphisms (rs266729, rs2241766, and rs1501299) of ADIPOQ and cardiovascular disease: a meta-analysis

Associations between three common single nucleotide polymorphisms (rs266729, rs2241766, and... Background: Inconsistencies have existed in research findings on the association between cardiovascular disease (CVD) and single nucleotide polymorphisms (SNPs) of ADIPOQ, triggering this up-to-date meta-analysis. Methods: We searched for relevant studies in PubMed, EMBASE, Cochrane Library, CNKI, CBM, VIP, and WanFang databases up to 1st July 2017. We included 19,106 cases and 31,629 controls from 65 published articles in this meta-analysis. STATA 12.0 software was used for all statistical analyses. Results: Our results showed that rs266729 polymorphism was associated with the increased risk of CVD in dominant model or in heterozygote model; rs2241766 polymorphism was associated with the increased risk of CVD in the genetic models (allelic, dominant, recessive, heterozygote, and homozygote). In subgroup analysis, significant associations were found in different subgroups with the three SNPs. Meta-regression and subgroup analysis showed that heterogeneity might be explained by other confounding factors. Sensitivity analysis revealed that the results of our meta-analysis were stable and robust. In addition, the results of trial sequential analysis showed that evidences of our results are sufficient to reach concrete conclusions. Conclusions: In conclusion, our meta-analysis found significant increased CVD risk is associated with rs266729 and rs2241766, but not associated with rs1501299. Keywords: ADIPOQ, Single nucleotide polymorphisms, Cardiovascular disease, Association, Meta-analysis Background Adiponectin is involved in CVD: low levels of adipo- Cardiovascular disease (CVD) is the primary cause of nectin (hypoadipoectinemia) positively correlate with the death worldwide, leading to 32% of all deaths worldwide risk of CVD, and higher levels of adiponectin protect in 2013 [1]. Epidemiological and biological evidences against this disease [6–11]. Adiponectin is synthesized demonstrate that multiple environmental and genetic and secreted by adipose tissue [12], osteoblasts [13], factors are implicated in CVD, although the etiology of skeletal muscle [14], and cardiomyocytes [15]. This CVD has not been fully elucidated [2–5]. Identifying protein, as one of the most abundant adipocytokines in CVD-relative risk factors is critical in control of the blood, has anti-atherogenic, cardioprotective, anti- development and progress of CVD. inflammatory, and antithrombotic properties [16–20]. Adiponectin is encoded by ADIPOQ which is located in chromosome 3q27 [21], and adiponectin levels are influenced by single-nucleotide polymorphisms (SNPs) * Correspondence: ywliu@jlu.edu.cn † in ADIPOQ [22]. SNPs in ADIPOQ have been found to Joseph Sam Kanu and Shuang Qiu contributed equally to this work. Department of Epidemiology and Biostatistics, School of Public Health of be associated with CVD [23, 24], diabetes [25, 26], stroke Jilin University, 1163 Xinmin Street, Changchun 130021, China [27, 28], myocardial infarction [29, 30], cancer [31, 32], Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 2 of 21 kidney disease [33, 34], and even gynecological condi- ‘ischemic heart disease’ or ‘angina’ or ‘myocardial infarc- tions [35, 36]. Previous studies have shown the associ- tion (MI)’ or ‘stroke’ or ‘atherosclerosis’ or ‘arteriosclerosis’ ation between SNPs in ADIPOQ (rs3774261, rs1063537, or ‘coronary stenosis’ combined with ‘ADIPOQ’ or ‘APM1’ rs2082940, rs2241766, rs266729, and rs1501299) and or ‘ACDC’ or ‘adiponectin gene’ and ‘polymorphisms’ or CVD/subclinical CVD [30, 37, 38]. The three common ‘variants’ or ‘variations’. Joseph Sam Kanu and Shuang SNPs of ADIPOQ (rs266729, rs2241766, and rs1501299) Qiu independently performed the literature search for were most widely studied. However, findings from previ- potential articles included in this meta-analysis. All articles ous studies on the three SNPs in relation to CVD risk retrieved were first organized in reference manager are inconsistent and inconclusive. software (Endnote 6). For rs266729 (− 11,377 C/G) in ADIPOQ, Du et al. [39] and Zhang et al. [40] found that the SNP is associ- Inclusion and exclusion criteria ated with CVD risk; Stenvinkel et al. [41] revealed that A study included in this meta-analysis was based on the rs266729 is associated with the decreased risk of CVD; following criteria: 1) the study has sufficient data to Zhang et al. [40], Cheong et al. [27], and Chiodini et al. allow association between CVD risk and ADIPOQ SNP [29] found that there is no significant association be- to be assessed; 2) the study included original data (inde- tween rs266729 and CVD. For rs2241766 (+ 45 T/G), pendence among studies); 3) evaluation of the ADIPOQ Pischon et al. [42] and Jung et al. [43] identified no asso- polymorphisms (rs266729, rs2241766, and rs1501299) ciation between rs2241766 and the risk of coronary and CVD risk; 4) the language of the study was English artery disease (CAD) in patients with type 2 diabetic or Chinese; and 5) observed genotype frequencies in mellitus (T2DM); Du et al. [39], Oliveira et al. [44], and controls must be consistent with Hardy–Weinberg Mofarrah et al. [45] found that there is a significant equilibrium (HWE). We excluded a study based on: 1) association between rs2241766 polymorphism and CAD the study contained overlapping data; 2) the study with risk; Chang et al. [46] revealed that rs2241766 is associated missing information (particularly genotype distributions), with the decreased risk of CVD. Moreover, for rs1501299 after corresponding author, who was contacted by us with (+ 276 G / T), Bacci et al. [47] and Esteghamati et al. [48] email, failed to provide the required information; and 3) revealed that rs1501299 is associated with the decreased genome scans investigating linkages with no detailed risk of CAD; Mohammadzadeh et al. [38], however, genotype distributions between cases and controls. Where reported that there is an association between rs1501299 there was a disagreement on the selection of a study, the and CAD risk; Foucan et al. [49] found that there is no issue was resolved by discussion or consensus with significant association between rs1501299 and CAD in pa- the third investigator (Ri Li). For articles with missing tients with T2DM. Thus, those results are inconsistent. data, we emailed the corresponding authors for the Meta-analysis performed by Zhang et al. in 2012 required data. revealed that associations between the SNPs (rs2241766, rs1501299, and rs266729) in ADIPOQ and CVD were Assessment of study quality significant but weak [50]. Since that data, several more We used the NATURE-published guidelines proposed studies have emerged to investigate the association by the NCI-NHGRI Working Group on Replication in between SNPs in ADIPOQ and susceptibility to CVD Association Studies for assessing the quality of each [37, 38, 45]. In this study, we further collected references study included in this meta-analysis [51]. These guide- and updated meta-analysis of association between SNPs lines have a checklist of 53 conditions for authors, jour- (rs2241766, rs1501299, and rs266729) in ADIPOQ and nal editors, and referees to interpret data and results of CVD in order to get a more precise and reliable assess- genome-wide or other genotype–phenotype association ment of the association. studies clearly and unambiguously. We used the first set of 34 conditions in assessing the quality of each study. Methods We allocated a score of 1 point for each condition a Search strategy study met, and no point (0 score) if the condition or We performed an extensive literature search in PubMed, requirement is lacking. Each study was given a total EMBASE, Cochrane Library, CNKI, CBM, VIP, and Quality Score – the sum of all points each study WanFang databases for published articles on the associ- obtained. Study quality assessment was independently ation between ADIPOQ polymorphisms and CVD risk up carried out by Joseph Sam Kanu and Shuang Qiu. to July 1st, 2017. The literature search was done without any language or population restrictions imposed. During Data extraction the literature search, we used various combinations of Joseph Sam Kanu and Shuang Qiu extracted data from keywords, such as ‘coronary heart disease (CHD)’ or each study independently. We summarized the information ‘coronary artery disease’ or ‘cardiovascular disease’ or extracted from each article in Table 1. The characteristics Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 3 of 21 of articles included first author, year of publication, country threshold for statistical significance [57]. In present in which the study was done, study population (ethnicity), meta-analysis, we used trial sequential analysis software numbers of cases and controls, genotyping method, SNPs (TSA, version 0.9; Copenhagen Trial Unit, Copenhagen, investigated, genotype frequency of cases and controls, and Denmark, 2011) by setting an overall type I error of 5%, outcome (Table 1; Additional file 1:TablesS1,S2,and S3). a statistical test power of 80%, and a relative risk reduc- tion of 20% [58, 59]. Statistical analysis If the Z-curve crosses trial sequential monitoring HWE was evaluated for each study using Goodness of fit boundary or RIS has been reached, a sufficient level of Chi-square test in control groups, and P < 0.05 was evidence has been reached and further studies are un- considered as a significant deviation from HWE. The needed; otherwise, additional studies are needed to reach strength of association between the three ADIPOQ poly- a sufficient conclusion. morphisms and CVD susceptibility was assessed using odds ratios (OR) and 95% confidence intervals (95% CI). Results The associations were measured based on five different Overall results genetic models: allelic model (rs266729: G versus C; This meta-analysis included 68 studies from 65 arti- rs2241766: G versus T; rs1501299: T versus G), domin- cles after literature search and critical screening, as ant model (rs266729: GG + GC versus CC; rs2241766: described in methods (Fig. 1). Meta-analysis of the GG + GT versus TT; rs1501299: TT + TG versus GG), rs266729 (− 11,377 C > G), rs2241766 (+ 45 T > G), recessive model (rs266729: GG versus GC + CC; and rs1501299 (+ 276 G > T) variants included 29, 40, rs2241766: GG versus GT + TT; rs1501299: TT versus and 44 studies, respectively. We summarize the char- TG + GG), heterozygote model (rs266729: GC versus acteristics of each primary study in Table 1. Detailed CC; rs2241766: GT versus TT; rs1501299: TG versus characteristics of those studies are further presented in GG), and homozygote model (rs266729: GG versus CC; Additional file 1: Tables S1, S2, and S3, respectively. Over- rs2241766: GG versus TT; rs1501299: TT versus GG). all, this meta-analysis included a total of 50,735 subjects Heterogeneity were evaluated by the Chi-square test (19,106 cases and 31,629 controls). based Q-statistic, and quantified by I -statistic [52]. If there was no substantial statistical heterogeneity (P >0.10, Meta-analysis results I ≤ 50%), data were pooled by fixed-effect model (Mantel Association between rs266729 (− 11,377 C > G) and Haenszel methods); otherwise, the heterogeneity was polymorphism and CVD evaluated by random-effect model (DerSimonian and The meta-analysis of the association between rs266729 Laird methods). Meta-regression analysis was performed (− 11,377 C > G) polymorphism and CVD included 29 to detect main sources of heterogeneity. In addition, sub- studies with 29,021 subjects (10,506 cases and 18,515 group analyses were stratified by population (European, controls). Significant heterogeneity among studies was East Asian, West Asian, and African), genotyping method observed (P < 0.10 or I ≥ 50%). Thus, we selected (PCR-RFLP, TaqMan, and Others), sample size (< 1000 random-effect model, and found that rs266729 poly- and ≥ 1000), and quality score (< 10 and ≥ 10). Sensitivity morphism was associated with the increased risk of analysis was performed to examine stability of our results CVD in dominant model (GG + GC VS CC: OR = 1.129, by omitting each study in each turn. Publication bias was 95% CI = 1.028–1.239, P = 0.011) and in heterozygote measured by funnel plots [53], and quantified by the model (GC VS CC: OR = 1.141, 95% CI = 1.041–1.250, Begg’s and Egger’s tests [54](P <0.05 considered statisti- P = 0.005) (Table 2, Fig. 2). cally significant publication bias). STATA 12.0 software Based on population, genotyping method, sample size, (StataCorp. 2011. Stata Statistical Software: Release 12. and quality score, we performed subgroup analyses. On College Station, TX: StataCorp LP) was used for all statis- the basis of population, rs266729 polymorphism was as- tical analyses. P-value < 0.05 was considered statistically sociated with the increased risk of CVD under dominant significant, except where other-wise specified. A separate model (GG + GC VS CC: OR = 1.198, 95% CI = 1.006–1. analysis was performed for each SNPs included in the 427, P = 0.043) and under heterozygote model (GC VS meta-analysis. CC: OR = 1.184, 95% CI = 1.002–1.398, P = 0.048) in East Asian. On the basis of genotyping methods, a significant Trial sequential analysis (TSA) risk association between rs266729 polymorphism and Traditional meta-analysis may result in type I and type CVD was found when genotyping was performed using II errors owing to dispersed data and repeated signifi- PCR-RFLP method under dominant model (GG + GC cance testing [55, 56]. To reduce the risk of type I error, VS CC: OR = 1.276, 95% CI = 1.014–1.607, P = 0.038) TSA was used to estimate required information size and under heterozygote model (GC VS CC: OR = 1.282, (RIS) and confirm statistical reliability with an adjusted 95% CI = 1.032–1.592, P = 0.025). On the basis of sample Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 4 of 21 Table 1 Characteristics of included studies Study ID Year Country Population Outcome Sample size Genotyping Quality Method Score Cases Controls Lacquemant Swiss 1 2004 Switzerland European CAD 107 181 Other 9 Lacquemant French 2 2004 France European CAD 55 134 Other 9 Bacci 3 2004 Italy European CAD 142 234 Other 8 Ohashi 4 2004 Japan East Asian CAD 383 368 TaqMan 7 Stenvinkel 5 2004 America European CVD 63 141 Other 6 Filippi 6 2005 Italy European CAD 580 466 Other 9 Ru Y 7 2005 China East Asian CHD 131 136 TaqMan 6 Qi1 8 2005 America European CVD 239 640 TaqMan 10 Qi2 9 2006 America European CVD 285 704 TaqMan 10 Wang JN 10 2006 China East Asian CHD 120 131 PCR-RFLP 7 Hegener 1 11 2006 America European MI 341 341 TaqMan 11 Hegener 2 12 2006 America European Stroke 259 259 TaqMan 11 Jung 13 2006 Korea East Asian CAD 88 68 TaqMan 8 Gable 1 14 2007 UK European CVD 266 2,727 PCR-RFLP 11 Gable 2 15 2007 UK European MI 530 564 PCR-RFLP 12 Pischon 16 2007 America European CHD 1,036 2,071 TaqMan 11 Lu F 17 2007 China East Asian CHD 135 131 PCR-RFLP 7 Hoefle 18 2007 Austria European CHD 277 125 TaqMan 7 Yamada 19 2008 Japan East Asian ACI 313 971 Other 9 Oguri 20 2009 Japan East Asian MI 773 1,114 Other 10 Chang 21 2009 China East Asian CAD 600 718 PCR-RFLP 9 Zhang XL 22 2009 China East Asian CHD 205 135 PCR-RFLP 8 Zhong C 23 2010 China East Asian CAD 198 237 TaqMan 10 Foucan 1 24 2010 France African CAD 57 159 TaqMan 7 Xu L 25 2010 China East Asian CHD 153 73 PCR-RFLP 8 Chiodini 26 2010 Italy European MI 503 503 TaqMan 10 Persson 27 2010 Sweden European MI 244 244 TaqMan 9 Chen XL 28 2010 China East Asian Stroke 357 345 TaqMan 8 Luo SX 29 2010 China East Asian CHD 221 100 PCR-RFLP 8 Caterina 30 2011 Italy European MI 1,864 1,864 Other 13 Al-Daghri 31 2011 Saudi A. West Asian CAD 123 295 PCR-RFLP 8 Prior 32 2011 UK European CHD 85 298 PCR-RFLP 7 Leu 33 2011 China East Asian Stroke 80 3,330 Other 10 Liu F 34 2011 China East Asian Stroke 302 338 PCR-RFLP 9 Rodriguez 35 2011 Spain European CVD 119 555 TaqMan 9 Chen F 36 2011 China East Asian CHD 93 102 PCR-RFLP 8 Maimaitiyiming 37 2011 China East Asian CHD 196 124 PCR-RFLP 8 Hu HH 38 2011 China East Asian CHD 150 152 Other 8 Zhang YM 39 2011 China East Asian CHD 149 167 PCR-RFLP 8 Zhou NN 40 2011 China East Asian CAD 358 65 PCR-RFLP 8 Sabouri 41 2011 UK European CAD 329 106 PCR-RFLP 8 Boumaiza 42 2011 Tunisia African CAD 212 104 PCR-RFLP 10 Chengang 43 2012 China East Asian CAD 267 250 PCR-RFLP 8 Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 5 of 21 Table 1 Characteristics of included studies (Continued) Study ID Year Country Population Outcome Sample size Genotyping Quality Method Score Cases Controls Esteghamati 44 2012 Iran West Asia CAD 114 127 PCR-RFLP 10 Gui 45 2012 China East Asian CAD 438 443 TaqMan 10 Katakami 46 2012 Japan East Asian CVD 213 2,424 Other 12 Oliveira 47 2012 Brazil European CAD 450 153 Other 10 Shi KL 48 2012 China East Asian CAD 396 292 Other 8 Zhang HF 49 2012 China East Asian ATHERO 394 118 PCR-RFLP 8 Nannan 50 2012 China East Asian CAD 213 467 Other 10 Antonopoulos 51 2013 Greece European CAD/MI 462 132 Other 11 Rizk 52 2013 Qatar West Asian ACS/MI 142 122 Other 12 Wang CH 53 2013 China East Asian CAD 101 116 TaqMan 9 Wu/276 54 2013 China East Asian CHD 188 200 PCR-RFLP 9 Cheung 55 2014 China East Asian CHD 184 2,012 Other 11 Foucan 2 56 2014 France African CAD 54 146 TaqMan 8 Shaker 57 2014 Egypt African MI 60 60 PCR-RFLP 8 Li Yang 58 2014 China East Asian CAD 234 365 PCR-RFLP 8 Alehagen 59 2015 Sweden European ATHERO 105 371 TaqMan 6 Torres 60 2015 Portugal European ATHERO 43 263 Other 7 Zhang M 61 2015 China East Asian CAD 563 412 Other 11 Liu Yun 62 2015 China East Asian CAD 200 200 PCR-RFLP 7 Du SX 63 2016 China East Asian CAD 493 304 PCR-RFLP 9 Mofarrah 64 2016 Iran West Asia CAD 152 72 Other 8 Mohammadzadeh 65 2016 Iran West Asia CAD 100 100 PCR-RFLP 9 Suo SZ 66 2016 China East Asian CAD 128 130 PCR-RFLP 9 Zhang Min 67 2016 China East Asian MI 306 412 Other 9 Li SS 68 2017 China East Asian Stroke 385 418 PCR-RFLP 10 ACI atherothrombotic cerebral infarction, ACS Acute Coronary Syndrome, ATHERO Atherosclerosis, CAD coronary artery disease, CHD coronary heart disease, CVD cardiovascular disease, IHD ischemic heart disease, MI myocardial infarction The 70-117 references are listed in Additional file 4 size or quality score, we found that rs266729 polymorph- Subgroup analyses were stratified by population, geno- ism was associated with the increased risk of CVD under typing method, sample size, and quality score. Firstly, on allelic, dominant, and heterozygote models (all OR > 1 and the basis of population, rs2241766 polymorphism was P < 0.05), after pooled the ORs by the subgroups of sample associated with the increased risk of CVD under the five size ≥ 1000 or quality score ≤ 10 (Table 2). dominant models in East Asian and under allelic, reces- sive, and homozygote models in West Asian (all OR >1 Association between rs2241766 (+ 45 T > G) and P < 0.05). Secondly, on the basis of genotyping polymorphism and CVD method, the results that genotyping was done by PCR- The meta-analysis of the association between rs2241766 RFLP or other methods showed that rs2241766 poly- (+ 45 T > G) polymorphism and CVD included 40 morphism was associated with the increased risk of studies with 25,548 subjects (10,746 cases and 14,802 CVD under five genetic models (all OR > 1 and P < 0.05). controls). Using inverse-variance weighted random effect Thirdly, on the basis of sample size, rs2241766 poly- model (P < 0.10 or I ≥ 50%), we found that rs2241766 morphism was associated with the increased risk of CVD polymorphism was associated with the increased risk of under the five genetic models in the subgroup of sample CVD in the five genetic models (allelic, dominant, size ≤1000 (all OR > 1 and P < 0.05), but was associated recessive, heterozygote, and homozygote) (all OR > 1 and with the decreased risk of CVD in the subgroup of P < 0.05) (Table 3, Fig. 3). sample size ≥1000 under recessive model (GG VS Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 6 of 21 Fig. 1 Flow diagram showing details of results of databases searched exclusion and inclusion of studies/articles in the meta-analysis. CNKI: Chinese National Knowledge Infrastructure; CBM: Chinese BioMedical Literature on Disc GT + TT: OR = 0.696, 95% CI = 0.539–0.885, P = 0.003) association between rs1501299 polymorphism and CVD in and under homozygote model (GG VS TT: OR =0. the five genetic models (all P > 0.05) (Table 4). 669, 95% CI =0.519–0.862, P = 0.002). Finally, on the In the subgroup analysis, no significant association basis of quality score, when we pooled the ORs by was found between rs1501299 polymorphism and CVD the subgroups of quality score ≤ 10, we found that risk under the five genetic models in any subgroup (all rs2241766 polymorphism was associated with the in- P > 0.05) (Table 4). creased risk of CVD under the five genetic models (all OR >1 and P <0.05) (Table 3). Heterogeneity analysis In this meta-analysis, meta-regression was used to inves- tigate the source of heterogeneity by year, population, Association between rs1501299 (+ 276 G > T) genotyping method, sample size, and quality score. We polymorphism and CVD found that sample size (allelic model: P = 0.019; domin- The meta-analysis of the association between rs1501299 ant model: P = 0.032; recessive model: P <0.001; and (+ 276 G > T) polymorphism and CVD included 44 studies homozygote model: P < 0.001) and quality score (allelic with 37,371 subjects (12,852 cases and 24,519 controls). model: P = 0.035; dominant model: P = 0.032; recessive Using the inverse-variance weighted random effect model model: P < 0.001; and homozygote model: P <0.001) (P <0.10 or I ≥ 50%), we found that there was no contributed to the observed heterogeneity across all the h Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 7 of 21 Table 2 Overall and subgroup meta-analysis of the association between ADIPOQ rs266729, −11,377 C > G polymorphisms and CVD Categories n Sample size G VS C GG + GC VS CC GG VS GC + CC GC VS CC GG VS CC 2( 2( 2( 2( 2( Case/Control OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph Overall 29 10,506/18,515 1.079 0.051 65.8/0.000 1.129 0.011 64.5/0.000 0.989 0.898 48.5/0.002 1.141 0.005 59.9/0.000 1.037 0.692 53.4/0.000 (1.000, 1.165) (1.028, 1.239) (0.838, 1.168) (1.041, 1.250) (0.867, 1.239) Population European 17 6,355/11,666 1.022 0.564 37.6/0.060 1.071 0.158 40.8/0.041 0.879 0.224 40.0/0.045 1.102 0.062 43.5/0.029 0.908 0.360 36.9/0.064 (0.948, 1.102) (0.974, 1.178) (0.714, 1.082) (0.995, 1.220) (0.739, 1.116) East Asian 12 4,151/6,849 1.154 0.051 76.8/0.000 1.198 0.043 75.7/0.000 1.149 0.293 52.5/0.017 1.184 0.048 70.7/0.000 1.231 0.164 61.4/0.003 (1.000, 1.332) (1.006, 1.427) (0.887, 1.487) (1.002, 1.398) (0.919, 1.650) Genotyping PCR-RFLP 8 2,382/4,976 1.186 0.083 77.5/0.000 1.276 0.038 75.4/0.000 1.162 0.411 53.5/0.035 1.282 0.025 69.7/0.002 1.285 0.223 61.3/0.011 (0.978, 1.438) (1.014, 1.607) (0.813, 1.661) (1.032, 1.592) (0.859, 1.922) TaqMan 12 3,910/6,312 1.031 0.544 45.3/0.044 1.054 0.331 30.7/0.146 0.951 0.721 53.5/0.014 1.064 0.236 20.6/0.242 0.973 0.849 52.2/0.018 (0.935, 1.137) (0.948, 1.173) (0.720, 1.256) (0.960, 1.180) (0.735, 1.288) Others 9 4,214/7,227 1.045 0.493 63.6/0.005 1.095 0.289 68.7/0.001 0.923 0.545 39.6/0.103 1.121 0.201 68.7/0.001 0.949 0.713 44.9/0.069 (0.921, 1.186) (0.926, 1.296) (0.711, 1.197) (0.941, 1.336) (0.717, 1.255) Sample size < 1000 21 5,048/6,708 1.065 0.270 67.9/0.000 1.114 0.119 65.9/0.000 0.955 0.722 53.4/0.002 1.128 0.075 60.9/0.000 0.992 0.956 57.6/0.001 (0.952, 1.192) (0.973, 1.276) (0.744, 1.228) (0.988, 1.287) (0.759, 1.298) ≥ 1000 8 5,458/11,807 1.108 0.042 63.4/0.008 1.162 0.018 64.6/0.006 1.017 0.864 38.6/0.122 1.172 0.012 61.5/0.011 1.087 0.445 45.4/0.077 (1.004, 1.222) (1.026, 1.315) (0.835, 1.240) (1.035, 1.326) (0.877, 1.347) Quality score < 10 16 3,489/5,128 1.152 0.040 68.0/0.000 1.211 0.019 64.6/0.000 1.147 0.348 49.6/0.013 1.207 0.016 57.8/0.002 1.215 0.224 55.9/0.003 (1.007, 1.318) (1.032, 1.420) (0.861, 1.528) (1.036, 1.406) (0.888, 1.664) ≥ 10 13 7,017/13,387 1.019 0.646 55.5/0.008 1.062 0.271 61.3/0.002 0.883 0.155 31.5/0.131 1.089 0.135 61.8/0.002 0.915 0.327 33.5/0.115 (0.940, 1.105) (0.954, 1.182) (0.744, 1.048) (0.974, 1.219) (0.767, 1.093) n study numbers, Bold values represent statistically significant findings Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 8 of 21 Fig. 2 Forest plots of the association between rs266729 polymorphism and CVD risk. (a) dominant model; (b) heterozygote model studies of the association between rs2241766 polymor- Discussion phisms and CVD risk. However, in the meta-analysis of the In this meta-analysis, we collected up-to-date informa- associations between rs266729/rs1501299 polymorphisms tion (July 1st, 2017) to investigate the association and CVD risk, we did not identify the source of heterogen- between ADIPOQ SNPs and the risk of CVD. Our re- eity (all P > 0.05) (Additional file 2:Table S4). sults demonstrate that rs266729 and rs2241766 variants of ADIPOQ are associated with the increased risk of CVD, but rs1501299 is not associated with CVD risk. Publication bias and sensitivity analysis In view of the association between rs266729 and CVD Publication bias was measured by funnel plots and quanti- risk, Yang et al. (2012) [60], Zhou et al. (2012) [61], and fied by Begg’sand Egger’s tests. No publication bias was Zhang et al. (2012) [50] performed meta-analyses. Yang found among the studies regarding the association between et al. reported that rs266729 is associated with the in- rs266729 polymorphisms and CVD risk (all P > 0.05). creased risk of CAD in allelic and dominant models Publication biases were found in analyses of the associa- [60]. Zhou et al. found the same association in overall tions between rs2241766 polymorphisms and CVD risk (al- population, Europeans, and East Asian in allelic, domin- lelic model: P = 0.001, P = 0.031; dominant model: Egger Begg ant and heterozygote models [61]. Zhang et al. also re- P = 0.001, P = 0.003; and heterozygote mode: P Egger Begg Egger vealed that rs266729 is associated with the increased risk = 0.003, P = 0.003), and between rs1501299 polymor- Begg of CAD in overall population and East Asian in allelic phisms and CVD risk (recessive model: P = 0.031, P Egger Begg model [50]. Our results further identified that rs266729 = 0.035) (Table 5 and Additional file 3:Figures S1,S2,and is associated with the increased risk of CVD in overall S3). Sensitivity analyses showed that this meta-analysis population and East Asian in dominant and heterozy- was relatively stable and credible (Figs. 4, 5,and 6). gote models. In addition, our results revealed that the significant association in studies on the basis of PCR- TSA RFLP method, indicating that different genotyping In the TSA of rs266729 and CVD, the Z-curve crossed trial method may result in different statistical results. sequential monitoring boundary and the sample size The association between rs2241766 and CVD risk also reached RIS in dominant and heterozygote models (Fig. 7). has been the subject of meta-analysis [60–64]. These In allelic, recessive, and homozygote models, the sample studies are inconsistent. Yang et al. found no significant size also reached RIS, although the Z-curve did not cross association between rs2241766 and CAD risk [60]. Zhang trial sequential monitoring boundary (Fig. 7). In the TSA et al. found no overall significant risk association between of rs2241766/rs1501299 and CVD, the sample size reached CHD and rs2241766 in Han Chinese population [62]. RIS in the five genetic models (Figs. 8 and 9). Thus, con- Zhou et al. reported that rs2241766 is associated with the crete conclusions were reached and further studies were decreased risk of CVD in recessive and homozygote not required. models, and the decreased risk of CVD in East Asian in Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 9 of 21 Table 3 Overall and subgroup meta-analysis of the association between ADIPOQ rs2241766, +45 T > G polymorphisms and CVD Categories n Sample G VS T GG + GT VS TT GG VS GT + TT GT VS TT GG VS TT size 2( 2( 2( 2( 2( Case/ OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph Control Overall 40 10,746/14,802 1.216 < 72.4/0.000 1.229 < 65.6/0.000 1.286 0.011 49.7/0.000 1.172 0.001 53.3/0.000 1.361 0.005 57.7/0.000 (1.102, 1.343) 0.001 (1.103, 1.369) 0.001 (1.061, 1.560) (1.063, 1.292) (1.095, 1.690) Population European 12 4,452/7,255 1.067 0.398 60.4/0.003 1.105 0.238 58.3/0.006 0.779 0.106 0.0/0.663 1.123 0.157 53.0/0.015 0.792 0.132 0.0/0.585 (0.918, 1.242) (0.937, 1.303) (0.576, 1.055) (0.956, 1.319) (0.584, 1.073) East Asian 20 5,305/6,505 1.194 0.004 70.5/0.000 1.225 0.007 67.3/0.000 1.315 0.010 43.1/0.024 1.180 0.018 58.1/0.001 1.431 0.005 58.6/0.001 (1.057, 1.348) (1.057, 1.420) (1.068, 1.618) (1.029, 1.353) (1.112, 1.842) West Asian 5 660/719 1.550 0.049 80.8/0.000 1.392 0.145 71.3/0.007 2.715 0.002 50.2/0.091 1.099 0.591 43.2/0.134 2.767 0.006 59.2/0.044 (1.002, 2.396) (0.893, 2.170) (1.452, 5.079) (0.779, 1.549) (1.347, 5.683) African 3 329/323 2.200 0.088 75.3/0.017 2.148 0.066 65.2/0.056 2.010 0.511 50.8/0.154 1.919 0.051 45.1/0.162 2.295 0.463 55.2/0.135 (0.890, 5.437) (0.952, 4.844) (0.251, 16.080) (0.998, 3.688) (0.250, 21.058) Genotyping PCR-RFLP 20 4,814/6,319 1.242 0.009 77.1/0.000 1.279 0.012 74.4/0.000 1.335 0.027 40.5/0.035 1.221 0.027 67.0/0.000 1.442 0.022 57.1/0.001 (1.055, 1.462) (1.057, 1.548) (1.034, 1.722) (1.023, 1.458) (1.054, 1.975) TaqMan 7 2,616/3,715 1.087 0.400 64.8/0.009 1.118 0.262 53.9/0.043 0.872 0.614 56.9/0.041 1.123 0.172 34.9/0.162 0.896 0.708 62.2/0.021 (0.895, 1.320) (0.920, 1.357) (0.513, 1.482) (0.951, 1.326) (0.506, 1.588) Other 13 3,316/4,768 1.263 0.005 68.7/0.000 1.238 0.009 51.4/0.016 1.453 0.038 56.7/0.006 1.150 0.044 27.6/0.166 1.522 0.024 57.7/0.005 (1.075, 1.485) (1.056, 1.452) (1.021, 2.066) (1.004, 1.317) (1.056, 2.193) Sample size < 1000 34 7,651/6,381 1.298 < 66.6/0.000 1.317 < 61.1/0.000 1.512 < 25.5/0.096 1.239 < 51.4/0.000 1.620 < 35.9/0.024 (1.164, 1.448) 0.001 (1.163, 1.492) 0.001 (1.264, 1.809) 0.001 (1.102, 1.393) 0.001 (1.324, 1.981) 0.001 ≥ 1000 6 3,095/8,421 0.920 0.097 23.9/0.255 0.945 0.344 25.5/0.243 0.690 0.003 0.0/0.758 0.981 0.728 11.3/0.343 0.669 0.002 0.0/0.661 (0.834, 1.015) (0.841, 1.062) (0.539, 0.885) (0.879, 1.094) (0.519, 0.862) Quality score < 10 26 5,467/4,951 1.366 < 77.0/0.000 1.404 < 72.8/0.000 1.529 0.001 46.0/0.008 1.314 0.001 64.4/0.000 1.692 < 58.4/0.000 (1.176, 1.586) 0.001 (1.183, 1.667) 0.001 (1.202, 1.944) (1.121, 1.539) (1.274, 2.248) 0.001 ≥ 10 14 5,279/9,851 1.036 0.455 37.3/0.079 1.038 0.376 0.0/0.575 0.978 0.887 49.9/0.017 1.043 0.343 0.0/0.818 0.985 0.925 47.5/0.025 (0.944, 1.139) (0.955, 1.128) (0.719, 1.331) (0.956, 1.137) (0.725, 1.340) n study numbers; Bold values represent statistically significant findings Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 10 of 21 Fig. 3 Forest plots of the association between rs2241766 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 11 of 21 Table 4 Overall and subgroup meta-analysis of the association between ADIPOQ rs1501299, +276 G > T polymorphism and CVD Categories n Sample size T VS G TT + TG VS GG TT VS TG + GG TG VS GG TT VS GG 2( 2( 2( 2( 2( Case/Control OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph Overall 44 12,852/24,519 0.956 0.189 64.7/0.000 0.967 0.431 60.6/0.000 0.899 0.086 42.0/0.002 0.987 0.737 49.2/0.000 0.886 0.104 55.5/0.000 (0.893, 1.023) (0.890, 1.051) (0.797, 1.015) (0.913, 1.066) (0.766, 1.025) Population European 18 7,002/11,337 0.957 0.146 16.2/0.260 0.967 0.380 16.1/0.262 0.851 0.066 35.2/0.070 0.988 0.773 21.3/0.201 0.854 0.068 30.2/0.110 (0.901, 1.016) (0.896, 1.043) (0.717, 1.011) (0.909, 1.073) (0.722, 1.012) East Asian 20 5,107/12,291 0.966 0.594 77.5/0.000 0.977 0.776 74.4/0.000 0.945 0.572 52.2/0.004 0.988 0.867 64.0/0.000 0.940 0.638 69.0/0.000 (0.849, 1.098) (0.834, 1.145) (0.778, 1.149) (0.858, 1.138) (0.726, 1.217) West Asian 4 479/645 0.973 0.897 79.6/0.002 0.960 0.880 77.8/0.004 0.999 0.997 41.9/0.160 0.952 0.843 70.2/0.018 0.986 0.970 62.1/0.048 (0.643, 1.473) (0.564, 1.635) (0.578, 1.727) (0.587, 1.546) (0.477, 2.040) African 2 264/246 0.848 0.278 11.8/0.287 0.856 0.428 0.0/0.490 0.724 0.258 7.8/0.298 0.927 0.719 0.0/0.725 0.700 0.266 14.7/0.279 (0.629, 1.143) (0.583, 1.257) (0.415, 1.266) (0.614, 1.400) (0.374, 1.312) Genotyping PCR-RFLP 14 3,359/5,817 0.970 0.688 74.2/0.000 0.997 0.978 70.6/0.000 0.881 0.329 55.3/0.006 1.051 0.861 58.4/0.003 0.901 0.535 68.4/0.000 (0.833, 1.128) (0.825, 1.206) (0.684, 1.136) (0.858, 1.202) (0.648, 1.253) TaqMan 13 3,666/6,001 0.977 0.701 61.3/0.002 0.987 0.859 58.0/0.005 0.970 0.771 30.1/0.144 1.001 0.994 47.8/0.028 0.956 0.718 46.8/0.032 (0.869, 1.099) (0.854, 1.140) (0.791, 1.189) (0.874, 1.146) (0.749, 1.221) Others 17 5,827/12,701 0.930 0.159 59.7/0.001 0.935 0.287 55.1/0.003 0.866 0.140 40.9/0.041 0.959 0.484 46.3/0.019 0.841 0.117 49.6/0.011 (0.841, 1.029) (0.827, 1.058) (0.715, 1.048) (0.852, 1.079) (0.678, 1.044) Sample size < 1000 36 8,167/9,201 0.945 0.191 64.8/0.000 0.959 0.438 60.2/0.000 0.876 0.079 41.8/0.005 0.985 0.758 47.8/0.001 0.865 0.116 56.0/0.000 (0.868, 1.029) (0.864, 1.065) (0.756, 1.016) (0.895, 1.085) (0.722, 1.036) ≥ 1000 8 4,685/15,318 0.984 0.784 68.6/0.002 0.985 0.831 66.7/0.004 0.968 0.762 44.8/0.080 0.987 0.853 59.7/0.015 0.955 0.711 56.3/0.025 (0.877, 1.104) (0.855, 1.134) (0.784, 1.195) (0.863, 1.129) (0.748, 1.219) Quality score < 10 24 4,690/5,424 0.954 0.438 69.1/0.000 0.976 0.752 65.3/0.000 0.879 0.189 41.7/0.018 1.002 0.981 52.8/0.001 0.876 0.294 59.3/0.000 (0.848, 1.074) (0.842, 1.132) (0.725, 1.065) (0.876, 1.145) (0.683, 1.122) ≥ 10 20 8,162/19,095 0.959 0.298 60.0/0.000 0.963 0.442 55.3/0.002 0.915 0.273 44.7/0.017 0.976 0.599 46.3/0.013 0.902 0.250 52.2/0.004 (0.886, 1.038/) (0.875, 1.060) (0.782, 1.072) (0.890, 1.070) (0.756, 1.075) Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 12 of 21 Table 5 Publication bias assessment of this meta-analysis publication bias, sample size, or insufficient statistical power. In addition, evidences have showed that studies SNPs Genetic model Egger’s test Begg’s test which deviate from HWE in controls may reflect the pres- t-value P z-value P ence of genotyping errors, population stratification, and rs266729 Allelic model 0.60 0.552 0.47 0.639 selection bias in the controls (or without representation of Dominant model 0.77 0.451 0.62 0.536 studied sample). Thus, including those studies may de- Recessive model −0.67 0.507 0.92 0.358 crease the quality of a meta-analysis or generate inconsist- Heterozygote model 0.79 0.435 0.81 0.420 ent results [67]. Homozygote model −0.45 0.658 0.73 0.464 Heterogeneity across all the studies of the associations should be noted because it may potentially affect the rs2241766 Allelic model 3.52 0.001 2.16 0.031 strengths of the present meta-analysis. We, thus, used Dominant model 3.63 0.001 2.99 0.003 random effect model. Our results showed that sample Recessive model 0.72 0.476 0.40 0.687 size and quality score are the factors of heterogeneity Heterozygote model 3.17 0.003 2.97 0.003 across all studies of association between rs2241766 poly- Homozygote model 0.88 0.383 0.33 0.744 morphisms and CVD, but no factors contribute the het- rs1501299 Allelic model −0.80 0.427 0.96 0.337 erogeneity across all studies of association between rs266729/rs1501299 polymorphisms and CVD. However, Dominant model 0.09 0.930 0.13 0.895 heterogeneity was still high in the subgroup analysis of Recessive model −2.24 0.031 2.11 0.035 the two factors. For these reasons, heterogeneity might Heterozygote model 0.60 0.549 0.11 0.911 be explained by other confounding factors, such as Homozygote model −1.45 0.155 1.49 0.137 gene-gene interaction and gene-environment interaction. Our meta-analysis has some limitations. Firstly, signifi- allelic, dominant, recessive, and homozygote models [61]. cant publication bias was found in the analysis of Zhou et al. performed a meta-analysis of the association rs2241766 (under allelic, dominant, and heterozygote between rs2241766 and CVD risk in allelic model, and models) and rs1501299 (under recessive model). they found that rs2241766 is associated with the increased Secondly, our meta-analysis mainly included Europeans risk of CVD [63]. In our meta-analysis, we found that and Asians with only few other races, thus limiting our rs2241766 is associated with the increased risk of CVD in power to generalize our findings in other races. Finally, overall population and East Asian in all the five genetic our results might be affected by the potential weaknesses models, and in West Asian in allelic, recessive, and homo- of genetic association studies, such as phenotype zygote models. Our findings is in agreement with the re- misclassifications, genotyping error, population stratifica- sults of Zhou et al., but is in disagreement with the results tion, gene-environment or gene-gene interactions, and of Yang et al., Zhang et al., and Zhou et al. selective reporting biases [68, 69]. With regard to the association between 1,501,299 and Despite the limitations highlighted above, our meta- CVD, the results are also conflicting [50, 60, 61]. Zhou et analysis also had some strength. Firstly, we searched al. revealed no significant association between rs1501299 extensively and investigated more studies and more polymorphism with CAD susceptibility [61]. Qi et al. re- participants than any other meta-analyses performed on ported the extremely large decrease in CVD risk associated the association between ADIPOQ variant and CVD, with rs1501299 polymorphism in diabetic patients [24]. which give our study more statistical power to draw Zhang et al. reported only the weak protective effect of the valid conclusion on this issue. Secondly, sensitivity ana- rs1501299 variant against CVD in general study subjects lysis showed that the results of our meta-analysis are [50]. The meta-analysis by Zhao et al. revealed that stable and robust. Thirdly, the evidence of our results rs1501299 polymorphism may play a protective role for are sufficient to reach concrete conclusions, which were CAD among patients with T2DM [22]. In comparison, our proved by TSA for the first time. We strongly believe results revealed no significant association. our findings will help settle some of the controversies Different genetic admixture and environmental factors surrounding the ADIPOQ-CVD association research. among human populations, which tend to explain ethnic background, strongly modulate the effects of ADIPOQ Conclusions polymorphisms on adiponectin levels [65, 66]. Studies Our meta-analysis found significant increased CVD risk is have reported that low levels of adiponectin (hypoadipoec- associated with rs266729 and rs2241766, but not tinemia) correlate with the risk of CVD, and high levels of associated with rs1501299. Investigating gene–gene and adiponectin protect against this disease [6–11]. These gene–environment interactions is needed to give more conflicting results of associations between the ADIPOQ insight into the genetic association between ADIPOQ polymorphism and CVD risk may be due to differences in variants and CVD. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 13 of 21 Fig. 4 Sensitivity analyses of the association between rs266729 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 14 of 21 Fig. 5 Sensitivity analyses of the association between rs2241766 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 15 of 21 Fig. 6 Sensitivity analyses of the association between rs1501299 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 16 of 21 Fig. 7 Trial sequential analysis of the association between rs266729 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 17 of 21 Fig. 8 Trial sequential analysis of the association between rs2241766 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 18 of 21 Fig. 9 Trial sequential analysis of the association between rs1501299 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 19 of 21 Additional files 5. Sun LY, Lee EW, Zahra A, Park JH. Risk factors of cardiovascular disease and their related socio-economical, environmental and health behavioral factors: focused on low-middle income countries- a narrative review article. Additional file 1: Summary of three SNPs characteristics. (DOCX 48 kb) Iran J Public Health. 2015;44:435–44. Additional file 2: Meta-regression results of the association between 6. Stojanovic S, Ilic MD, Ilic S, Petrovic D, Djukic S. The significance of the SNPs and CVD risk. (DOCX 59 kb) adiponectin as a biomarker in metabolic syndrome and/or coronary artery disease. Vojnosanit Pregl. 2015;72:779–84. Additional file 3: Funnel plots of three SNPs for publication bias. (DOCX 1250 kb) 7. Zhang HL, Jin X. Relationship between serum adiponectin and osteoprotegerin levels and coronary heart disease severity. Genet Mol Res. 2015;14:11023–9. Additional file 4: Additional references. (DOCX 22 kb) 8. Md Sayed AS, Zhao Z, Guo L, Li F, Deng X, Deng H, Xia K, Yang T. Serum lectin-like oxidized-low density lipoprotein receptor-1 and adiponectin Abbreviations levels are associated with coronary artery disease accompanied with CAD: Coronary artery disease; CHD: Coronary heart disease; metabolic syndrome. Iran Red Crescent Med J. 2014;16:e12106. CVD: Cardiovascular disease; HWE: Hardy–Weinberg equilibrium; 9. Di Chiara T, Argano C, Scaglione A, Duro G, Corrao S, Scaglione R, Licata G. MI: Myocardial infarction; RIS: Required information size; SNPs: Single- Hypoadiponectinemia, cardiometabolic comorbidities and left ventricular nucleotide polymorphisms; T2DM: Type 2 diabetic mellitus; TSA: Trial hypertrophy. Intern Emerg Med. 2015;10:33–40. sequential analysis 10. Obata Y, Yamada Y, Kyo M, Takahi Y, Saisho K, Tamba S, Yamamoto K, Katsuragi K, Matsuzawa Y. Serum adiponectin levels predict the risk of Acknowledgements coronary heart disease in Japanese patients with type 2 diabetes. J Diabetes Investig. 2013;4:475–82. The authors would like to thank staff and students of the Meta-analysis Group of the Department of Epidemiology and Biostatistics, School of Public 11. Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB. Plasma Health of Jilin University for their contributions to this work. adiponectin levels and risk of myocardial infarction in men. JAMA. 2004;291:1730–7. 12. Scherer PE, Williams S, Fogliano M, Baldini G, Lodish HF. A novel serum Funding protein similar to C1q, produced exclusively in adipocytes. J Biol Chem. This work was supported by The National Natural Science Foundation of China 1995;270:26746–9. (Grant 81573230), the Ministry of Science and Technology of the People’s 13. Shinoda Y, Yamaguchi M, Ogata N, Akune T, Kubota N, Yamauchi T, Republic of China (grant number: 2015DFA31580), and the Science and Terauchi Y, Kadowaki T, Takeuchi Y, Fukumoto S, et al. Regulation of bone Technology Commission of Jilin Province (grant number: 20150101130JC). formation by adiponectin through autocrine/paracrine and endocrine pathways. J Cell Biochem. 2006;99:196–208. Availability of data and materials 14. Krause MP, Liu Y, Vu V, Chan L, Xu A, Riddell MC, Sweeney G, Hawke Please contact author for data requests. TJ. Adiponectin is expressed by skeletal muscle fibers and influences muscle phenotype and function. Am J Physiol Cell Physiol. Authors’ contributions 2008;295:C203–12. Conception and design: JSK, SQ, YC, and YL. Provision of study materials: JSK, 15. Pineiro R, Iglesias MJ, Gallego R, Raghay K, Eiras S, Rubio J, Dieguez C, SQ, RL, and CK; Collection and assembly of data: JSK, SQ, and RL. Data Gualillo O, Gonzalez-Juanatey JR, Lago F. Adiponectin is synthesized analysis and interpretation: JSK, SQ, and RL. Manuscript writing: JSK and SQ. and secreted by human and murine cardiomyocytes. FEBS Lett. Revised the language/article: All authors. Final approval of manuscript: All 2005;579:5163–9. authors. 16. Fujita K, Maeda N, Sonoda M, Ohashi K, Hibuse T, Nishizawa H, Nishida M, Hiuge A, Kurata A, Kihara S, et al. Adiponectin protects against angiotensin Ethics approval and consent to participate II-induced cardiac fibrosis through activation of PPAR-alpha. Arterioscler Not applicable. Thromb Vasc Biol. 2008;28:863–70. 17. Ouchi N, Kihara S, Arita Y, Maeda K, Kuriyama H, Okamoto Y, Hotta K, Competing interests Nishida M, Takahashi M, Nakamura T, et al. Novel modulator for endothelial The authors declared that they have no competing interest. adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation. 1999;100:2473–6. 18. Shibata R, Sato K, Pimentel DR, Takemura Y, Kihara S, Ohashi K, Funahashi T, Publisher’sNote Ouchi N, Walsh K. Adiponectin protects against myocardial ischemia- Springer Nature remains neutral with regard to jurisdictional claims in reperfusion injury through AMPK- and COX-2-dependent mechanisms. Nat published maps and institutional affiliations. Med. 2005;11:1096–103. 19. Kato H, Kashiwagi H, Shiraga M, Tadokoro S, Kamae T, Ujiie H, Honda S, Author details Miyata S, Ijiri Y, Yamamoto J, et al. Adiponectin acts as an endogenous Department of Epidemiology and Biostatistics, School of Public Health of antithrombotic factor. Arterioscler Thromb Vasc Biol. 2006;26:224–30. Jilin University, 1163 Xinmin Street, Changchun 130021, China. The Cardiovascular Center, the First Hospital of Jilin University, Changchun 20. Ouchi N, Kihara S, Arita Y, Nishida M, Matsuyama A, Okamoto Y, Ishigami M, Kuriyama H, Kishida K, Nishizawa H, et al. Adipocyte-derived plasma protein, 130021, China. adiponectin, suppresses lipid accumulation and class a scavenger receptor expression in human monocyte-derived macrophages. Circulation. Received: 22 February 2018 Accepted: 4 May 2018 2001;103:1057–63. 21. Takahashi M, Arita Y, Yamagata K, Matsukawa Y, Okutomi K, Horie M, Shimomura I, Hotta K, Kuriyama H, Kihara S, et al. Genomic structure and References mutations in adipose-specific gene, adiponectin. Int J Obes Relat Metab 1. Roth GA, Huffman MD, Moran AE, Feigin V, Mensah GA, Naghavi M, Murray Disord. 2000;24:861–8. CJ. Global and regional patterns in cardiovascular mortality from 22. Zhao N, Li N, Zhang S, Ma Q, Ma C, Yang X, Yin J, Zhang R, Li J, Yang X, Cui T. 1990 to 2013. Circulation. 2015;132:1667–78. Associations between two common single nucleotide polymorphisms 2. Bhatnagar A. Environmental determinants of cardiovascular disease. (rs2241766 and rs1501299) of ADIPOQ gene and coronary artery disease in Circ Res. 2017;121:162–80. type 2 diabetic patients: a systematic review and meta-analysis. Oncotarget. 3. Roberts R, Stewart AF. Genes and coronary artery disease: where are we? J 2017;8:51994. Am Coll Cardiol. 2012;60:1715–21. 23. Katakami N, Kaneto H, Matsuoka TA, Takahara M, Maeda N, Shimizu I, Ohno 4. Ji F, Ning F, Duan H, Kaprio J, Zhang D, Zhang D, Wang S, Qiao Q, Sun J, K, Osonoi T, Kawai K, Ishibashi F, et al. Adiponectin G276T gene Liang J, et al. Genetic and environmental influences on cardiovascular polymorphism is associated with cardiovascular disease in Japanese disease risk factors: a study of Chinese twin children and adolescents. patients with type 2 diabetes. Atherosclerosis. 2012;220:437–42. Twin Res Hum Genet. 2014;17:72–9. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 20 of 21 24. Qi L, Doria A, Manson JE, Meigs JB, Hunter D, Mantzoros CS, Hu FB. 44. Oliveira CS, Saddi-Rosa P, Crispim F, Canani LH, Gerchman F, Giuffrida FM, Adiponectin genetic variability, plasma adiponectin, and cardiovascular risk Vieira JG, Velho G, Reis AF. Association of ADIPOQ variants, total and high in patients with type 2 diabetes. Diabetes. 2006;55:1512–6. molecular weight adiponectin levels with coronary artery disease in diabetic 25. Kaftan AN, Hussain MK. Association of adiponectin gene polymorphism and non-diabetic Brazilian subjects. J Diabetes Complicat. 2012;26:94–8. rs266729 with type two diabetes mellitus in Iraqi population. A pilot study. 45. Mofarrah M, Ziaee S, Pilehvar-Soltanahmadi Y, Zarghami F, Boroumand M, Gene. 2015;570:95–9. Zarghami N. Association of KALRN, ADIPOQ, and FTO gene polymorphism in type 2 diabetic patients with coronary artery disease: possible 26. Tsai MK, Wang HM, Shiang JC, Chen IH, Wang CC, Shiao YF, Liu WS, Lin TJ, predisposing markers. Coron Artery Dis. 2016;27:490–6. Chen TM, Chen YH. Sequence variants of ADIPOQ and association with type 2 diabetes mellitus in Taiwan Chinese Han population. 46. Chang YC, Jiang JY, Jiang YD, Chiang FT, Hwang JJ, Lien WP, Chuang LM. ScientificWorldJournal. 2014;2014:650393. Interaction of ADIPOQ genetic polymorphism with blood pressure and 27. Cheong MY, Bang OS, Cha MH, Park YK, Kim SH, Kim YJ. Association of the plasma cholesterol level on the risk of coronary artery disease. Circ J. adiponectin gene variations with risk of ischemic stroke in a Korean 2009;73:1934–8. population. Yonsei Med J. 2011;52:20–5. 47. Bacci S, Menzaghi C, Ercolino T, Ma X, Rauseo A, Salvemini L, Vigna C, 28. Liu F, He Z, Deng S, Zhang H, Li N, Xu J. Association of adiponectin gene Fanelli R, Di Mario U, Doria A, Trischitta V. The +276 G/T single nucleotide polymorphisms with the risk of ischemic stroke in a Chinese Han polymorphism of the adiponectin gene is associated with coronary artery population. Mol Biol Rep. 2011;38:1983–8. disease in type 2 diabetic patients. Diabetes Care. 2004;27:2015–20. 29. Chiodini BD, Specchia C, Gori F, Barlera S, D'OrazioA,Pietri S,Crociati L,Nicolucci 48. Esteghamati A, Mansournia N, Nakhjavani M, Mansournia MA, Nikzamir A, A, Franciosi M, Signorini S, et al. Adiponectin gene polymorphisms and their Abbasi M. Association of +45(T/G) and +276(G/T) polymorphisms in the adiponectin gene with coronary artery disease in a population of Iranian effect on the risk of myocardial infarction and type 2 diabetes: an association studyinanItalian population.TherAdv Cardiovasc Dis. 2010;4:223–30. patients with type 2 diabetes. Mol Biol Rep. 2012;39:3791–7. 49. Foucan L, Maimaitiming S, Larifla L, Hedreville S, Deloumeaux J, 30. Shaker OG, Ismail MF. Association of genetic variants of MTHFR, ENPP1, and Joannes MO, Blanchet-Deverly A, Velayoudom-Céphise FL, Aubert R, ADIPOQ with myocardial infarction in Egyptian patients. Cell Biochem Salamon R, et al. Adiponectin gene variants, adiponectin isoforms and Biophys. 2014;69:265–74. cardiometabolic risk in type 2 diabetic patients. Journal of Diabetes 31. Gu CY, Li QX, Zhu Y, Wang MY, Shi TY, Yang YY, Wang JC, Jin L, Wei QY, Ye Investigation. 2014;5:192–8. DW. Genetic variations of the ADIPOQgene and risk of prostate cancer in Chinese Han men. Asian J Androl. 2014;16:878–83. 50. Zhang H, Mo X, Hao Y, Gu D. Association between polymorphisms in the 32. Zhang G, Gu C, Zhu Y, Luo L, Dong D, Wan F, Zhang H, Shi G, Sun L, Ye D. adiponectin gene and cardiovascular disease: a meta-analysis. BMC Med ADIPOQ polymorphism rs182052 is associated with clear cell renal cell Genet. 2012;13:40. carcinoma. Cancer Sci. 2015;106:687–91. 51. Chanock SJ, Manolio T, Boehnke M, Boerwinkle E, Hunter DJ, Thomas G, Hirschhorn JN, Abecasis G, Altshuler D, Bailey-Wilson JE, et al. Replicating 33. Chung HF, Long KZ, Hsu CC, Mamun AA, Chiu YF, Tu HP, Chen PS, Jhang genotype-phenotype associations. Nature. 2007;447:655–60. HR, Hwang SJ, Huang MC. Adiponectin gene (ADIPOQ) polymorphisms 52. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in correlate with the progression of nephropathy in Taiwanese male patients meta-analyses. BMJ. 2003;327:557–60. with type 2 diabetes. Diabetes Res Clin Pract. 2014;105:261–70. 34. Elshamaa MF, Sabry SM, El-Sonbaty MM, Elghoroury EA, Emara N, Raafat M, 53. Stuck AE, Rubenstein LZ, Wieland D. Bias in meta-analysis detected by a Kandil D, Elsaaid G. Adiponectin: an adipocyte-derived hormone, and its simple, graphical test. Asymmetry detected in funnel plot was probably due gene encoding in children with chronic kidney disease. BMC Res Notes. to true heterogeneity. BMJ. 1998;316:469. author reply 470-461 2012;5:174. 54. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis 35. Yoshihara K, Yahata T, Kashima K, Mikami T, Tanaka K. Association of single detected by a simple, graphical test. BMJ. 1997;315:629–34. nucleotide polymorphisms in adiponectin and its receptor genes with 55. Turner RM, Bird SM, Higgins JP. The impact of study size on meta-analyses: polycystic ovary syndrome. J Reprod Med. 2009;54:669–74. examination of underpowered studies in Cochrane reviews. PLoS One. 2013;8:e59202. 36. Machado JS, Palei AC, Amaral LM, Bueno AC, Antonini SR, Duarte G, 56. Brok J, Thorlund K, Wetterslev J, Gluud C. Apparently conclusive meta- Tanus-Santos JE, Sandrim VC, Cavalli RC. Polymorphisms of the analyses may be inconclusive–trial sequential analysis adjustment of adiponectin gene in gestational hypertension and pre-eclampsia. J Hum random error risk due to repetitive testing of accumulating data in Hypertens. 2014;28:128–32. apparently conclusive neonatal meta-analyses. Int J Epidemiol. 37. Kanu JS, Gu Y, Zhi S, Yu M, Lu Y, Cong Y, Liu Y, Li Y, Yu Y, Cheng Y, 2009;38:287–98. Liu Y. Single nucleotide polymorphism rs3774261 in the AdipoQ gene is associated with the risk of coronary heart disease (CHD) in 57. Higgins JP, Whitehead A, Simmonds M. Sequential methods for random- northeast Han Chinese population: a case-control study. Lipids effects meta-analysis. Stat Med. 2011;30:903–21. Health Dis. 2016;15:6. 58. Wang F, Qin Z, Si S, Tang J, Xu L, Xu H, Li R, Han P, Yang H: Lack of 38. Mohammadzadeh G, Ghaffari MA, Heibar H, Bazyar M. Association of two association between NAT2 polymorphism and prostate cancer risk: a meta- common single nucleotide polymorphisms (+45T/G and +276G/T) of analysis and trial sequential analysis. Oncotarget 2017. ADIPOQ gene with coronary artery disease in type 2 diabetic patients. Iran 59. Zhu M, Wen X, Liu X, Wang Y, Liang C, Tu J: Association between 8q24 Biomed J. 2016;20:152–60. rs6983267 polymorphism and cancer susceptibility: a meta-analysis involving 170,737 subjects. Oncotarget 2017. 39. Du SX, Lu LL, Liu Y, Dong QJ, Xuan SY, Xin YN. Association of Adiponectin Gene Polymorphisms with the risk of coronary artery disease in patients 60. Yang Y, Zhang F, Ding R, Wang Y, Lei H, Hu D. Association of ADIPOQ gene with nonalcoholic fatty liver disease in a Chinese Han population. Hepat polymorphisms and coronary artery disease risk: a meta-analysis based on Mon. 2016;16:e37388. 12 465 subjects. Thromb Res. 2012;130:58–64. 40. Zhang M, Peng Y, Lyu S. Association between genetic variants in the 61. Zhou L, Xi B, Wei Y, Pan H, Yang W, Shen W, Li Y, Cai J, Tang H. Association adiponectin gene and premature myocardial infarction. Chinese Journal of between adiponectin gene polymorphisms and coronary artery disease Cardiology. 2016;44:577–82. across different populations. Thromb Res. 2012;130:52–7. 41. Stenvinkel P, Marchlewska A, Pecoits-Filho R, Heimburger O, Zhang Z, Hoff 62. Zhang BC, Li WM, Xu YW. A meta-analysis of the association of adiponectin C, Holmes C, Axelsson J, Arvidsson S, Schalling M, et al. Adiponectin in renal gene polymorphisms with coronary heart disease in Chinese Han disease: relationship to phenotype and genetic variation in the gene population. Clin Endocrinol. 2012;76:358–64. encoding adiponectin. Kidney Int. 2004;65:274–81. 63. Zhou D, Jin Y, Yao F, Duan Z, Wang Q, Liu J. Association between the 42. Pischon T, Pai JK, Manson JE, Hu FB, Rexrode KM, Hunter D, Rimm EB. Single adiponectin +45T>G genotype and risk of cardiovascular disease: a meta- nucleotide polymorphisms at the adiponectin locus and risk of coronary analysis. Heart Lung Circ. 2014;23:159–65. heart disease in men and women. Obesity (Silver Spring). 2007;15:2051–60. 64. Sun K, Li Y, Wei C, Tong Y, Zheng H, Guo Y. Recessive protective effect of 43. Jung CH, Rhee EJ, Kim SY, Shin HS, Kim BJ, Sung KC, Kim BS, Lee WY, Kang ADIPOQ rs1501299 on cardiovascular diseases with type 2 diabetes: a JH, Oh KW, et al. Associations between two single nucleotide meta-analysis. Mol Cell Endocrinol. 2012;349:162–9. polymorphisms of adiponectin gene and coronary artery diseases. 65. Gu HF. Biomarkers of adiponectin: plasma protein variation and genomic Endocr J. 2006;53:671–7. DNA polymorphisms. Biomark Insights. 2009;4:123–33. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 21 of 21 66. Enns JE, Taylor CG, Zahradka P. Variations in Adipokine genes AdipoQ, Lep, and LepR are associated with risk for obesity-related metabolic disease: the modulatory role of gene-nutrient interactions. J Obes. 2011;2011:168659. 67. Boccia S, De Feo E, Galli P, Gianfagna F, Amore R, Ricciardi G. A systematic review evaluating the methodological aspects of meta-analyses of genetic association studies in cancer research. Eur J Epidemiol. 2010;25:765–75. 68. Ioannidis JP, Trikalinos TA. Early extreme contradictory estimates may appear in published research: the Proteus phenomenon in molecular genetics research and randomized trials. J Clin Epidemiol. 2005;58:543–9. 69. Daly AK, Day CP. Candidate gene case-control association studies: advantages and potential pitfalls. Br J Clin Pharmacol. 2001;52:489–99. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Lipids in Health and Disease Springer Journals

Associations between three common single nucleotide polymorphisms (rs266729, rs2241766, and rs1501299) of ADIPOQ and cardiovascular disease: a meta-analysis

Free
21 pages

Loading next page...
 
/lp/springer_journal/associations-between-three-common-single-nucleotide-polymorphisms-EXkSDvmIU0
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Life Sciences; Lipidology; Medical Biochemistry; Clinical Nutrition
eISSN
1476-511X
D.O.I.
10.1186/s12944-018-0767-8
Publisher site
See Article on Publisher Site

Abstract

Background: Inconsistencies have existed in research findings on the association between cardiovascular disease (CVD) and single nucleotide polymorphisms (SNPs) of ADIPOQ, triggering this up-to-date meta-analysis. Methods: We searched for relevant studies in PubMed, EMBASE, Cochrane Library, CNKI, CBM, VIP, and WanFang databases up to 1st July 2017. We included 19,106 cases and 31,629 controls from 65 published articles in this meta-analysis. STATA 12.0 software was used for all statistical analyses. Results: Our results showed that rs266729 polymorphism was associated with the increased risk of CVD in dominant model or in heterozygote model; rs2241766 polymorphism was associated with the increased risk of CVD in the genetic models (allelic, dominant, recessive, heterozygote, and homozygote). In subgroup analysis, significant associations were found in different subgroups with the three SNPs. Meta-regression and subgroup analysis showed that heterogeneity might be explained by other confounding factors. Sensitivity analysis revealed that the results of our meta-analysis were stable and robust. In addition, the results of trial sequential analysis showed that evidences of our results are sufficient to reach concrete conclusions. Conclusions: In conclusion, our meta-analysis found significant increased CVD risk is associated with rs266729 and rs2241766, but not associated with rs1501299. Keywords: ADIPOQ, Single nucleotide polymorphisms, Cardiovascular disease, Association, Meta-analysis Background Adiponectin is involved in CVD: low levels of adipo- Cardiovascular disease (CVD) is the primary cause of nectin (hypoadipoectinemia) positively correlate with the death worldwide, leading to 32% of all deaths worldwide risk of CVD, and higher levels of adiponectin protect in 2013 [1]. Epidemiological and biological evidences against this disease [6–11]. Adiponectin is synthesized demonstrate that multiple environmental and genetic and secreted by adipose tissue [12], osteoblasts [13], factors are implicated in CVD, although the etiology of skeletal muscle [14], and cardiomyocytes [15]. This CVD has not been fully elucidated [2–5]. Identifying protein, as one of the most abundant adipocytokines in CVD-relative risk factors is critical in control of the blood, has anti-atherogenic, cardioprotective, anti- development and progress of CVD. inflammatory, and antithrombotic properties [16–20]. Adiponectin is encoded by ADIPOQ which is located in chromosome 3q27 [21], and adiponectin levels are influenced by single-nucleotide polymorphisms (SNPs) * Correspondence: ywliu@jlu.edu.cn † in ADIPOQ [22]. SNPs in ADIPOQ have been found to Joseph Sam Kanu and Shuang Qiu contributed equally to this work. Department of Epidemiology and Biostatistics, School of Public Health of be associated with CVD [23, 24], diabetes [25, 26], stroke Jilin University, 1163 Xinmin Street, Changchun 130021, China [27, 28], myocardial infarction [29, 30], cancer [31, 32], Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 2 of 21 kidney disease [33, 34], and even gynecological condi- ‘ischemic heart disease’ or ‘angina’ or ‘myocardial infarc- tions [35, 36]. Previous studies have shown the associ- tion (MI)’ or ‘stroke’ or ‘atherosclerosis’ or ‘arteriosclerosis’ ation between SNPs in ADIPOQ (rs3774261, rs1063537, or ‘coronary stenosis’ combined with ‘ADIPOQ’ or ‘APM1’ rs2082940, rs2241766, rs266729, and rs1501299) and or ‘ACDC’ or ‘adiponectin gene’ and ‘polymorphisms’ or CVD/subclinical CVD [30, 37, 38]. The three common ‘variants’ or ‘variations’. Joseph Sam Kanu and Shuang SNPs of ADIPOQ (rs266729, rs2241766, and rs1501299) Qiu independently performed the literature search for were most widely studied. However, findings from previ- potential articles included in this meta-analysis. All articles ous studies on the three SNPs in relation to CVD risk retrieved were first organized in reference manager are inconsistent and inconclusive. software (Endnote 6). For rs266729 (− 11,377 C/G) in ADIPOQ, Du et al. [39] and Zhang et al. [40] found that the SNP is associ- Inclusion and exclusion criteria ated with CVD risk; Stenvinkel et al. [41] revealed that A study included in this meta-analysis was based on the rs266729 is associated with the decreased risk of CVD; following criteria: 1) the study has sufficient data to Zhang et al. [40], Cheong et al. [27], and Chiodini et al. allow association between CVD risk and ADIPOQ SNP [29] found that there is no significant association be- to be assessed; 2) the study included original data (inde- tween rs266729 and CVD. For rs2241766 (+ 45 T/G), pendence among studies); 3) evaluation of the ADIPOQ Pischon et al. [42] and Jung et al. [43] identified no asso- polymorphisms (rs266729, rs2241766, and rs1501299) ciation between rs2241766 and the risk of coronary and CVD risk; 4) the language of the study was English artery disease (CAD) in patients with type 2 diabetic or Chinese; and 5) observed genotype frequencies in mellitus (T2DM); Du et al. [39], Oliveira et al. [44], and controls must be consistent with Hardy–Weinberg Mofarrah et al. [45] found that there is a significant equilibrium (HWE). We excluded a study based on: 1) association between rs2241766 polymorphism and CAD the study contained overlapping data; 2) the study with risk; Chang et al. [46] revealed that rs2241766 is associated missing information (particularly genotype distributions), with the decreased risk of CVD. Moreover, for rs1501299 after corresponding author, who was contacted by us with (+ 276 G / T), Bacci et al. [47] and Esteghamati et al. [48] email, failed to provide the required information; and 3) revealed that rs1501299 is associated with the decreased genome scans investigating linkages with no detailed risk of CAD; Mohammadzadeh et al. [38], however, genotype distributions between cases and controls. Where reported that there is an association between rs1501299 there was a disagreement on the selection of a study, the and CAD risk; Foucan et al. [49] found that there is no issue was resolved by discussion or consensus with significant association between rs1501299 and CAD in pa- the third investigator (Ri Li). For articles with missing tients with T2DM. Thus, those results are inconsistent. data, we emailed the corresponding authors for the Meta-analysis performed by Zhang et al. in 2012 required data. revealed that associations between the SNPs (rs2241766, rs1501299, and rs266729) in ADIPOQ and CVD were Assessment of study quality significant but weak [50]. Since that data, several more We used the NATURE-published guidelines proposed studies have emerged to investigate the association by the NCI-NHGRI Working Group on Replication in between SNPs in ADIPOQ and susceptibility to CVD Association Studies for assessing the quality of each [37, 38, 45]. In this study, we further collected references study included in this meta-analysis [51]. These guide- and updated meta-analysis of association between SNPs lines have a checklist of 53 conditions for authors, jour- (rs2241766, rs1501299, and rs266729) in ADIPOQ and nal editors, and referees to interpret data and results of CVD in order to get a more precise and reliable assess- genome-wide or other genotype–phenotype association ment of the association. studies clearly and unambiguously. We used the first set of 34 conditions in assessing the quality of each study. Methods We allocated a score of 1 point for each condition a Search strategy study met, and no point (0 score) if the condition or We performed an extensive literature search in PubMed, requirement is lacking. Each study was given a total EMBASE, Cochrane Library, CNKI, CBM, VIP, and Quality Score – the sum of all points each study WanFang databases for published articles on the associ- obtained. Study quality assessment was independently ation between ADIPOQ polymorphisms and CVD risk up carried out by Joseph Sam Kanu and Shuang Qiu. to July 1st, 2017. The literature search was done without any language or population restrictions imposed. During Data extraction the literature search, we used various combinations of Joseph Sam Kanu and Shuang Qiu extracted data from keywords, such as ‘coronary heart disease (CHD)’ or each study independently. We summarized the information ‘coronary artery disease’ or ‘cardiovascular disease’ or extracted from each article in Table 1. The characteristics Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 3 of 21 of articles included first author, year of publication, country threshold for statistical significance [57]. In present in which the study was done, study population (ethnicity), meta-analysis, we used trial sequential analysis software numbers of cases and controls, genotyping method, SNPs (TSA, version 0.9; Copenhagen Trial Unit, Copenhagen, investigated, genotype frequency of cases and controls, and Denmark, 2011) by setting an overall type I error of 5%, outcome (Table 1; Additional file 1:TablesS1,S2,and S3). a statistical test power of 80%, and a relative risk reduc- tion of 20% [58, 59]. Statistical analysis If the Z-curve crosses trial sequential monitoring HWE was evaluated for each study using Goodness of fit boundary or RIS has been reached, a sufficient level of Chi-square test in control groups, and P < 0.05 was evidence has been reached and further studies are un- considered as a significant deviation from HWE. The needed; otherwise, additional studies are needed to reach strength of association between the three ADIPOQ poly- a sufficient conclusion. morphisms and CVD susceptibility was assessed using odds ratios (OR) and 95% confidence intervals (95% CI). Results The associations were measured based on five different Overall results genetic models: allelic model (rs266729: G versus C; This meta-analysis included 68 studies from 65 arti- rs2241766: G versus T; rs1501299: T versus G), domin- cles after literature search and critical screening, as ant model (rs266729: GG + GC versus CC; rs2241766: described in methods (Fig. 1). Meta-analysis of the GG + GT versus TT; rs1501299: TT + TG versus GG), rs266729 (− 11,377 C > G), rs2241766 (+ 45 T > G), recessive model (rs266729: GG versus GC + CC; and rs1501299 (+ 276 G > T) variants included 29, 40, rs2241766: GG versus GT + TT; rs1501299: TT versus and 44 studies, respectively. We summarize the char- TG + GG), heterozygote model (rs266729: GC versus acteristics of each primary study in Table 1. Detailed CC; rs2241766: GT versus TT; rs1501299: TG versus characteristics of those studies are further presented in GG), and homozygote model (rs266729: GG versus CC; Additional file 1: Tables S1, S2, and S3, respectively. Over- rs2241766: GG versus TT; rs1501299: TT versus GG). all, this meta-analysis included a total of 50,735 subjects Heterogeneity were evaluated by the Chi-square test (19,106 cases and 31,629 controls). based Q-statistic, and quantified by I -statistic [52]. If there was no substantial statistical heterogeneity (P >0.10, Meta-analysis results I ≤ 50%), data were pooled by fixed-effect model (Mantel Association between rs266729 (− 11,377 C > G) and Haenszel methods); otherwise, the heterogeneity was polymorphism and CVD evaluated by random-effect model (DerSimonian and The meta-analysis of the association between rs266729 Laird methods). Meta-regression analysis was performed (− 11,377 C > G) polymorphism and CVD included 29 to detect main sources of heterogeneity. In addition, sub- studies with 29,021 subjects (10,506 cases and 18,515 group analyses were stratified by population (European, controls). Significant heterogeneity among studies was East Asian, West Asian, and African), genotyping method observed (P < 0.10 or I ≥ 50%). Thus, we selected (PCR-RFLP, TaqMan, and Others), sample size (< 1000 random-effect model, and found that rs266729 poly- and ≥ 1000), and quality score (< 10 and ≥ 10). Sensitivity morphism was associated with the increased risk of analysis was performed to examine stability of our results CVD in dominant model (GG + GC VS CC: OR = 1.129, by omitting each study in each turn. Publication bias was 95% CI = 1.028–1.239, P = 0.011) and in heterozygote measured by funnel plots [53], and quantified by the model (GC VS CC: OR = 1.141, 95% CI = 1.041–1.250, Begg’s and Egger’s tests [54](P <0.05 considered statisti- P = 0.005) (Table 2, Fig. 2). cally significant publication bias). STATA 12.0 software Based on population, genotyping method, sample size, (StataCorp. 2011. Stata Statistical Software: Release 12. and quality score, we performed subgroup analyses. On College Station, TX: StataCorp LP) was used for all statis- the basis of population, rs266729 polymorphism was as- tical analyses. P-value < 0.05 was considered statistically sociated with the increased risk of CVD under dominant significant, except where other-wise specified. A separate model (GG + GC VS CC: OR = 1.198, 95% CI = 1.006–1. analysis was performed for each SNPs included in the 427, P = 0.043) and under heterozygote model (GC VS meta-analysis. CC: OR = 1.184, 95% CI = 1.002–1.398, P = 0.048) in East Asian. On the basis of genotyping methods, a significant Trial sequential analysis (TSA) risk association between rs266729 polymorphism and Traditional meta-analysis may result in type I and type CVD was found when genotyping was performed using II errors owing to dispersed data and repeated signifi- PCR-RFLP method under dominant model (GG + GC cance testing [55, 56]. To reduce the risk of type I error, VS CC: OR = 1.276, 95% CI = 1.014–1.607, P = 0.038) TSA was used to estimate required information size and under heterozygote model (GC VS CC: OR = 1.282, (RIS) and confirm statistical reliability with an adjusted 95% CI = 1.032–1.592, P = 0.025). On the basis of sample Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 4 of 21 Table 1 Characteristics of included studies Study ID Year Country Population Outcome Sample size Genotyping Quality Method Score Cases Controls Lacquemant Swiss 1 2004 Switzerland European CAD 107 181 Other 9 Lacquemant French 2 2004 France European CAD 55 134 Other 9 Bacci 3 2004 Italy European CAD 142 234 Other 8 Ohashi 4 2004 Japan East Asian CAD 383 368 TaqMan 7 Stenvinkel 5 2004 America European CVD 63 141 Other 6 Filippi 6 2005 Italy European CAD 580 466 Other 9 Ru Y 7 2005 China East Asian CHD 131 136 TaqMan 6 Qi1 8 2005 America European CVD 239 640 TaqMan 10 Qi2 9 2006 America European CVD 285 704 TaqMan 10 Wang JN 10 2006 China East Asian CHD 120 131 PCR-RFLP 7 Hegener 1 11 2006 America European MI 341 341 TaqMan 11 Hegener 2 12 2006 America European Stroke 259 259 TaqMan 11 Jung 13 2006 Korea East Asian CAD 88 68 TaqMan 8 Gable 1 14 2007 UK European CVD 266 2,727 PCR-RFLP 11 Gable 2 15 2007 UK European MI 530 564 PCR-RFLP 12 Pischon 16 2007 America European CHD 1,036 2,071 TaqMan 11 Lu F 17 2007 China East Asian CHD 135 131 PCR-RFLP 7 Hoefle 18 2007 Austria European CHD 277 125 TaqMan 7 Yamada 19 2008 Japan East Asian ACI 313 971 Other 9 Oguri 20 2009 Japan East Asian MI 773 1,114 Other 10 Chang 21 2009 China East Asian CAD 600 718 PCR-RFLP 9 Zhang XL 22 2009 China East Asian CHD 205 135 PCR-RFLP 8 Zhong C 23 2010 China East Asian CAD 198 237 TaqMan 10 Foucan 1 24 2010 France African CAD 57 159 TaqMan 7 Xu L 25 2010 China East Asian CHD 153 73 PCR-RFLP 8 Chiodini 26 2010 Italy European MI 503 503 TaqMan 10 Persson 27 2010 Sweden European MI 244 244 TaqMan 9 Chen XL 28 2010 China East Asian Stroke 357 345 TaqMan 8 Luo SX 29 2010 China East Asian CHD 221 100 PCR-RFLP 8 Caterina 30 2011 Italy European MI 1,864 1,864 Other 13 Al-Daghri 31 2011 Saudi A. West Asian CAD 123 295 PCR-RFLP 8 Prior 32 2011 UK European CHD 85 298 PCR-RFLP 7 Leu 33 2011 China East Asian Stroke 80 3,330 Other 10 Liu F 34 2011 China East Asian Stroke 302 338 PCR-RFLP 9 Rodriguez 35 2011 Spain European CVD 119 555 TaqMan 9 Chen F 36 2011 China East Asian CHD 93 102 PCR-RFLP 8 Maimaitiyiming 37 2011 China East Asian CHD 196 124 PCR-RFLP 8 Hu HH 38 2011 China East Asian CHD 150 152 Other 8 Zhang YM 39 2011 China East Asian CHD 149 167 PCR-RFLP 8 Zhou NN 40 2011 China East Asian CAD 358 65 PCR-RFLP 8 Sabouri 41 2011 UK European CAD 329 106 PCR-RFLP 8 Boumaiza 42 2011 Tunisia African CAD 212 104 PCR-RFLP 10 Chengang 43 2012 China East Asian CAD 267 250 PCR-RFLP 8 Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 5 of 21 Table 1 Characteristics of included studies (Continued) Study ID Year Country Population Outcome Sample size Genotyping Quality Method Score Cases Controls Esteghamati 44 2012 Iran West Asia CAD 114 127 PCR-RFLP 10 Gui 45 2012 China East Asian CAD 438 443 TaqMan 10 Katakami 46 2012 Japan East Asian CVD 213 2,424 Other 12 Oliveira 47 2012 Brazil European CAD 450 153 Other 10 Shi KL 48 2012 China East Asian CAD 396 292 Other 8 Zhang HF 49 2012 China East Asian ATHERO 394 118 PCR-RFLP 8 Nannan 50 2012 China East Asian CAD 213 467 Other 10 Antonopoulos 51 2013 Greece European CAD/MI 462 132 Other 11 Rizk 52 2013 Qatar West Asian ACS/MI 142 122 Other 12 Wang CH 53 2013 China East Asian CAD 101 116 TaqMan 9 Wu/276 54 2013 China East Asian CHD 188 200 PCR-RFLP 9 Cheung 55 2014 China East Asian CHD 184 2,012 Other 11 Foucan 2 56 2014 France African CAD 54 146 TaqMan 8 Shaker 57 2014 Egypt African MI 60 60 PCR-RFLP 8 Li Yang 58 2014 China East Asian CAD 234 365 PCR-RFLP 8 Alehagen 59 2015 Sweden European ATHERO 105 371 TaqMan 6 Torres 60 2015 Portugal European ATHERO 43 263 Other 7 Zhang M 61 2015 China East Asian CAD 563 412 Other 11 Liu Yun 62 2015 China East Asian CAD 200 200 PCR-RFLP 7 Du SX 63 2016 China East Asian CAD 493 304 PCR-RFLP 9 Mofarrah 64 2016 Iran West Asia CAD 152 72 Other 8 Mohammadzadeh 65 2016 Iran West Asia CAD 100 100 PCR-RFLP 9 Suo SZ 66 2016 China East Asian CAD 128 130 PCR-RFLP 9 Zhang Min 67 2016 China East Asian MI 306 412 Other 9 Li SS 68 2017 China East Asian Stroke 385 418 PCR-RFLP 10 ACI atherothrombotic cerebral infarction, ACS Acute Coronary Syndrome, ATHERO Atherosclerosis, CAD coronary artery disease, CHD coronary heart disease, CVD cardiovascular disease, IHD ischemic heart disease, MI myocardial infarction The 70-117 references are listed in Additional file 4 size or quality score, we found that rs266729 polymorph- Subgroup analyses were stratified by population, geno- ism was associated with the increased risk of CVD under typing method, sample size, and quality score. Firstly, on allelic, dominant, and heterozygote models (all OR > 1 and the basis of population, rs2241766 polymorphism was P < 0.05), after pooled the ORs by the subgroups of sample associated with the increased risk of CVD under the five size ≥ 1000 or quality score ≤ 10 (Table 2). dominant models in East Asian and under allelic, reces- sive, and homozygote models in West Asian (all OR >1 Association between rs2241766 (+ 45 T > G) and P < 0.05). Secondly, on the basis of genotyping polymorphism and CVD method, the results that genotyping was done by PCR- The meta-analysis of the association between rs2241766 RFLP or other methods showed that rs2241766 poly- (+ 45 T > G) polymorphism and CVD included 40 morphism was associated with the increased risk of studies with 25,548 subjects (10,746 cases and 14,802 CVD under five genetic models (all OR > 1 and P < 0.05). controls). Using inverse-variance weighted random effect Thirdly, on the basis of sample size, rs2241766 poly- model (P < 0.10 or I ≥ 50%), we found that rs2241766 morphism was associated with the increased risk of CVD polymorphism was associated with the increased risk of under the five genetic models in the subgroup of sample CVD in the five genetic models (allelic, dominant, size ≤1000 (all OR > 1 and P < 0.05), but was associated recessive, heterozygote, and homozygote) (all OR > 1 and with the decreased risk of CVD in the subgroup of P < 0.05) (Table 3, Fig. 3). sample size ≥1000 under recessive model (GG VS Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 6 of 21 Fig. 1 Flow diagram showing details of results of databases searched exclusion and inclusion of studies/articles in the meta-analysis. CNKI: Chinese National Knowledge Infrastructure; CBM: Chinese BioMedical Literature on Disc GT + TT: OR = 0.696, 95% CI = 0.539–0.885, P = 0.003) association between rs1501299 polymorphism and CVD in and under homozygote model (GG VS TT: OR =0. the five genetic models (all P > 0.05) (Table 4). 669, 95% CI =0.519–0.862, P = 0.002). Finally, on the In the subgroup analysis, no significant association basis of quality score, when we pooled the ORs by was found between rs1501299 polymorphism and CVD the subgroups of quality score ≤ 10, we found that risk under the five genetic models in any subgroup (all rs2241766 polymorphism was associated with the in- P > 0.05) (Table 4). creased risk of CVD under the five genetic models (all OR >1 and P <0.05) (Table 3). Heterogeneity analysis In this meta-analysis, meta-regression was used to inves- tigate the source of heterogeneity by year, population, Association between rs1501299 (+ 276 G > T) genotyping method, sample size, and quality score. We polymorphism and CVD found that sample size (allelic model: P = 0.019; domin- The meta-analysis of the association between rs1501299 ant model: P = 0.032; recessive model: P <0.001; and (+ 276 G > T) polymorphism and CVD included 44 studies homozygote model: P < 0.001) and quality score (allelic with 37,371 subjects (12,852 cases and 24,519 controls). model: P = 0.035; dominant model: P = 0.032; recessive Using the inverse-variance weighted random effect model model: P < 0.001; and homozygote model: P <0.001) (P <0.10 or I ≥ 50%), we found that there was no contributed to the observed heterogeneity across all the h Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 7 of 21 Table 2 Overall and subgroup meta-analysis of the association between ADIPOQ rs266729, −11,377 C > G polymorphisms and CVD Categories n Sample size G VS C GG + GC VS CC GG VS GC + CC GC VS CC GG VS CC 2( 2( 2( 2( 2( Case/Control OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph Overall 29 10,506/18,515 1.079 0.051 65.8/0.000 1.129 0.011 64.5/0.000 0.989 0.898 48.5/0.002 1.141 0.005 59.9/0.000 1.037 0.692 53.4/0.000 (1.000, 1.165) (1.028, 1.239) (0.838, 1.168) (1.041, 1.250) (0.867, 1.239) Population European 17 6,355/11,666 1.022 0.564 37.6/0.060 1.071 0.158 40.8/0.041 0.879 0.224 40.0/0.045 1.102 0.062 43.5/0.029 0.908 0.360 36.9/0.064 (0.948, 1.102) (0.974, 1.178) (0.714, 1.082) (0.995, 1.220) (0.739, 1.116) East Asian 12 4,151/6,849 1.154 0.051 76.8/0.000 1.198 0.043 75.7/0.000 1.149 0.293 52.5/0.017 1.184 0.048 70.7/0.000 1.231 0.164 61.4/0.003 (1.000, 1.332) (1.006, 1.427) (0.887, 1.487) (1.002, 1.398) (0.919, 1.650) Genotyping PCR-RFLP 8 2,382/4,976 1.186 0.083 77.5/0.000 1.276 0.038 75.4/0.000 1.162 0.411 53.5/0.035 1.282 0.025 69.7/0.002 1.285 0.223 61.3/0.011 (0.978, 1.438) (1.014, 1.607) (0.813, 1.661) (1.032, 1.592) (0.859, 1.922) TaqMan 12 3,910/6,312 1.031 0.544 45.3/0.044 1.054 0.331 30.7/0.146 0.951 0.721 53.5/0.014 1.064 0.236 20.6/0.242 0.973 0.849 52.2/0.018 (0.935, 1.137) (0.948, 1.173) (0.720, 1.256) (0.960, 1.180) (0.735, 1.288) Others 9 4,214/7,227 1.045 0.493 63.6/0.005 1.095 0.289 68.7/0.001 0.923 0.545 39.6/0.103 1.121 0.201 68.7/0.001 0.949 0.713 44.9/0.069 (0.921, 1.186) (0.926, 1.296) (0.711, 1.197) (0.941, 1.336) (0.717, 1.255) Sample size < 1000 21 5,048/6,708 1.065 0.270 67.9/0.000 1.114 0.119 65.9/0.000 0.955 0.722 53.4/0.002 1.128 0.075 60.9/0.000 0.992 0.956 57.6/0.001 (0.952, 1.192) (0.973, 1.276) (0.744, 1.228) (0.988, 1.287) (0.759, 1.298) ≥ 1000 8 5,458/11,807 1.108 0.042 63.4/0.008 1.162 0.018 64.6/0.006 1.017 0.864 38.6/0.122 1.172 0.012 61.5/0.011 1.087 0.445 45.4/0.077 (1.004, 1.222) (1.026, 1.315) (0.835, 1.240) (1.035, 1.326) (0.877, 1.347) Quality score < 10 16 3,489/5,128 1.152 0.040 68.0/0.000 1.211 0.019 64.6/0.000 1.147 0.348 49.6/0.013 1.207 0.016 57.8/0.002 1.215 0.224 55.9/0.003 (1.007, 1.318) (1.032, 1.420) (0.861, 1.528) (1.036, 1.406) (0.888, 1.664) ≥ 10 13 7,017/13,387 1.019 0.646 55.5/0.008 1.062 0.271 61.3/0.002 0.883 0.155 31.5/0.131 1.089 0.135 61.8/0.002 0.915 0.327 33.5/0.115 (0.940, 1.105) (0.954, 1.182) (0.744, 1.048) (0.974, 1.219) (0.767, 1.093) n study numbers, Bold values represent statistically significant findings Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 8 of 21 Fig. 2 Forest plots of the association between rs266729 polymorphism and CVD risk. (a) dominant model; (b) heterozygote model studies of the association between rs2241766 polymor- Discussion phisms and CVD risk. However, in the meta-analysis of the In this meta-analysis, we collected up-to-date informa- associations between rs266729/rs1501299 polymorphisms tion (July 1st, 2017) to investigate the association and CVD risk, we did not identify the source of heterogen- between ADIPOQ SNPs and the risk of CVD. Our re- eity (all P > 0.05) (Additional file 2:Table S4). sults demonstrate that rs266729 and rs2241766 variants of ADIPOQ are associated with the increased risk of CVD, but rs1501299 is not associated with CVD risk. Publication bias and sensitivity analysis In view of the association between rs266729 and CVD Publication bias was measured by funnel plots and quanti- risk, Yang et al. (2012) [60], Zhou et al. (2012) [61], and fied by Begg’sand Egger’s tests. No publication bias was Zhang et al. (2012) [50] performed meta-analyses. Yang found among the studies regarding the association between et al. reported that rs266729 is associated with the in- rs266729 polymorphisms and CVD risk (all P > 0.05). creased risk of CAD in allelic and dominant models Publication biases were found in analyses of the associa- [60]. Zhou et al. found the same association in overall tions between rs2241766 polymorphisms and CVD risk (al- population, Europeans, and East Asian in allelic, domin- lelic model: P = 0.001, P = 0.031; dominant model: Egger Begg ant and heterozygote models [61]. Zhang et al. also re- P = 0.001, P = 0.003; and heterozygote mode: P Egger Begg Egger vealed that rs266729 is associated with the increased risk = 0.003, P = 0.003), and between rs1501299 polymor- Begg of CAD in overall population and East Asian in allelic phisms and CVD risk (recessive model: P = 0.031, P Egger Begg model [50]. Our results further identified that rs266729 = 0.035) (Table 5 and Additional file 3:Figures S1,S2,and is associated with the increased risk of CVD in overall S3). Sensitivity analyses showed that this meta-analysis population and East Asian in dominant and heterozy- was relatively stable and credible (Figs. 4, 5,and 6). gote models. In addition, our results revealed that the significant association in studies on the basis of PCR- TSA RFLP method, indicating that different genotyping In the TSA of rs266729 and CVD, the Z-curve crossed trial method may result in different statistical results. sequential monitoring boundary and the sample size The association between rs2241766 and CVD risk also reached RIS in dominant and heterozygote models (Fig. 7). has been the subject of meta-analysis [60–64]. These In allelic, recessive, and homozygote models, the sample studies are inconsistent. Yang et al. found no significant size also reached RIS, although the Z-curve did not cross association between rs2241766 and CAD risk [60]. Zhang trial sequential monitoring boundary (Fig. 7). In the TSA et al. found no overall significant risk association between of rs2241766/rs1501299 and CVD, the sample size reached CHD and rs2241766 in Han Chinese population [62]. RIS in the five genetic models (Figs. 8 and 9). Thus, con- Zhou et al. reported that rs2241766 is associated with the crete conclusions were reached and further studies were decreased risk of CVD in recessive and homozygote not required. models, and the decreased risk of CVD in East Asian in Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 9 of 21 Table 3 Overall and subgroup meta-analysis of the association between ADIPOQ rs2241766, +45 T > G polymorphisms and CVD Categories n Sample G VS T GG + GT VS TT GG VS GT + TT GT VS TT GG VS TT size 2( 2( 2( 2( 2( Case/ OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph OR (95% CI) PI %)/Ph Control Overall 40 10,746/14,802 1.216 < 72.4/0.000 1.229 < 65.6/0.000 1.286 0.011 49.7/0.000 1.172 0.001 53.3/0.000 1.361 0.005 57.7/0.000 (1.102, 1.343) 0.001 (1.103, 1.369) 0.001 (1.061, 1.560) (1.063, 1.292) (1.095, 1.690) Population European 12 4,452/7,255 1.067 0.398 60.4/0.003 1.105 0.238 58.3/0.006 0.779 0.106 0.0/0.663 1.123 0.157 53.0/0.015 0.792 0.132 0.0/0.585 (0.918, 1.242) (0.937, 1.303) (0.576, 1.055) (0.956, 1.319) (0.584, 1.073) East Asian 20 5,305/6,505 1.194 0.004 70.5/0.000 1.225 0.007 67.3/0.000 1.315 0.010 43.1/0.024 1.180 0.018 58.1/0.001 1.431 0.005 58.6/0.001 (1.057, 1.348) (1.057, 1.420) (1.068, 1.618) (1.029, 1.353) (1.112, 1.842) West Asian 5 660/719 1.550 0.049 80.8/0.000 1.392 0.145 71.3/0.007 2.715 0.002 50.2/0.091 1.099 0.591 43.2/0.134 2.767 0.006 59.2/0.044 (1.002, 2.396) (0.893, 2.170) (1.452, 5.079) (0.779, 1.549) (1.347, 5.683) African 3 329/323 2.200 0.088 75.3/0.017 2.148 0.066 65.2/0.056 2.010 0.511 50.8/0.154 1.919 0.051 45.1/0.162 2.295 0.463 55.2/0.135 (0.890, 5.437) (0.952, 4.844) (0.251, 16.080) (0.998, 3.688) (0.250, 21.058) Genotyping PCR-RFLP 20 4,814/6,319 1.242 0.009 77.1/0.000 1.279 0.012 74.4/0.000 1.335 0.027 40.5/0.035 1.221 0.027 67.0/0.000 1.442 0.022 57.1/0.001 (1.055, 1.462) (1.057, 1.548) (1.034, 1.722) (1.023, 1.458) (1.054, 1.975) TaqMan 7 2,616/3,715 1.087 0.400 64.8/0.009 1.118 0.262 53.9/0.043 0.872 0.614 56.9/0.041 1.123 0.172 34.9/0.162 0.896 0.708 62.2/0.021 (0.895, 1.320) (0.920, 1.357) (0.513, 1.482) (0.951, 1.326) (0.506, 1.588) Other 13 3,316/4,768 1.263 0.005 68.7/0.000 1.238 0.009 51.4/0.016 1.453 0.038 56.7/0.006 1.150 0.044 27.6/0.166 1.522 0.024 57.7/0.005 (1.075, 1.485) (1.056, 1.452) (1.021, 2.066) (1.004, 1.317) (1.056, 2.193) Sample size < 1000 34 7,651/6,381 1.298 < 66.6/0.000 1.317 < 61.1/0.000 1.512 < 25.5/0.096 1.239 < 51.4/0.000 1.620 < 35.9/0.024 (1.164, 1.448) 0.001 (1.163, 1.492) 0.001 (1.264, 1.809) 0.001 (1.102, 1.393) 0.001 (1.324, 1.981) 0.001 ≥ 1000 6 3,095/8,421 0.920 0.097 23.9/0.255 0.945 0.344 25.5/0.243 0.690 0.003 0.0/0.758 0.981 0.728 11.3/0.343 0.669 0.002 0.0/0.661 (0.834, 1.015) (0.841, 1.062) (0.539, 0.885) (0.879, 1.094) (0.519, 0.862) Quality score < 10 26 5,467/4,951 1.366 < 77.0/0.000 1.404 < 72.8/0.000 1.529 0.001 46.0/0.008 1.314 0.001 64.4/0.000 1.692 < 58.4/0.000 (1.176, 1.586) 0.001 (1.183, 1.667) 0.001 (1.202, 1.944) (1.121, 1.539) (1.274, 2.248) 0.001 ≥ 10 14 5,279/9,851 1.036 0.455 37.3/0.079 1.038 0.376 0.0/0.575 0.978 0.887 49.9/0.017 1.043 0.343 0.0/0.818 0.985 0.925 47.5/0.025 (0.944, 1.139) (0.955, 1.128) (0.719, 1.331) (0.956, 1.137) (0.725, 1.340) n study numbers; Bold values represent statistically significant findings Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 10 of 21 Fig. 3 Forest plots of the association between rs2241766 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 11 of 21 Table 4 Overall and subgroup meta-analysis of the association between ADIPOQ rs1501299, +276 G > T polymorphism and CVD Categories n Sample size T VS G TT + TG VS GG TT VS TG + GG TG VS GG TT VS GG 2( 2( 2( 2( 2( Case/Control OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph OR (95%CI) PI %)/Ph Overall 44 12,852/24,519 0.956 0.189 64.7/0.000 0.967 0.431 60.6/0.000 0.899 0.086 42.0/0.002 0.987 0.737 49.2/0.000 0.886 0.104 55.5/0.000 (0.893, 1.023) (0.890, 1.051) (0.797, 1.015) (0.913, 1.066) (0.766, 1.025) Population European 18 7,002/11,337 0.957 0.146 16.2/0.260 0.967 0.380 16.1/0.262 0.851 0.066 35.2/0.070 0.988 0.773 21.3/0.201 0.854 0.068 30.2/0.110 (0.901, 1.016) (0.896, 1.043) (0.717, 1.011) (0.909, 1.073) (0.722, 1.012) East Asian 20 5,107/12,291 0.966 0.594 77.5/0.000 0.977 0.776 74.4/0.000 0.945 0.572 52.2/0.004 0.988 0.867 64.0/0.000 0.940 0.638 69.0/0.000 (0.849, 1.098) (0.834, 1.145) (0.778, 1.149) (0.858, 1.138) (0.726, 1.217) West Asian 4 479/645 0.973 0.897 79.6/0.002 0.960 0.880 77.8/0.004 0.999 0.997 41.9/0.160 0.952 0.843 70.2/0.018 0.986 0.970 62.1/0.048 (0.643, 1.473) (0.564, 1.635) (0.578, 1.727) (0.587, 1.546) (0.477, 2.040) African 2 264/246 0.848 0.278 11.8/0.287 0.856 0.428 0.0/0.490 0.724 0.258 7.8/0.298 0.927 0.719 0.0/0.725 0.700 0.266 14.7/0.279 (0.629, 1.143) (0.583, 1.257) (0.415, 1.266) (0.614, 1.400) (0.374, 1.312) Genotyping PCR-RFLP 14 3,359/5,817 0.970 0.688 74.2/0.000 0.997 0.978 70.6/0.000 0.881 0.329 55.3/0.006 1.051 0.861 58.4/0.003 0.901 0.535 68.4/0.000 (0.833, 1.128) (0.825, 1.206) (0.684, 1.136) (0.858, 1.202) (0.648, 1.253) TaqMan 13 3,666/6,001 0.977 0.701 61.3/0.002 0.987 0.859 58.0/0.005 0.970 0.771 30.1/0.144 1.001 0.994 47.8/0.028 0.956 0.718 46.8/0.032 (0.869, 1.099) (0.854, 1.140) (0.791, 1.189) (0.874, 1.146) (0.749, 1.221) Others 17 5,827/12,701 0.930 0.159 59.7/0.001 0.935 0.287 55.1/0.003 0.866 0.140 40.9/0.041 0.959 0.484 46.3/0.019 0.841 0.117 49.6/0.011 (0.841, 1.029) (0.827, 1.058) (0.715, 1.048) (0.852, 1.079) (0.678, 1.044) Sample size < 1000 36 8,167/9,201 0.945 0.191 64.8/0.000 0.959 0.438 60.2/0.000 0.876 0.079 41.8/0.005 0.985 0.758 47.8/0.001 0.865 0.116 56.0/0.000 (0.868, 1.029) (0.864, 1.065) (0.756, 1.016) (0.895, 1.085) (0.722, 1.036) ≥ 1000 8 4,685/15,318 0.984 0.784 68.6/0.002 0.985 0.831 66.7/0.004 0.968 0.762 44.8/0.080 0.987 0.853 59.7/0.015 0.955 0.711 56.3/0.025 (0.877, 1.104) (0.855, 1.134) (0.784, 1.195) (0.863, 1.129) (0.748, 1.219) Quality score < 10 24 4,690/5,424 0.954 0.438 69.1/0.000 0.976 0.752 65.3/0.000 0.879 0.189 41.7/0.018 1.002 0.981 52.8/0.001 0.876 0.294 59.3/0.000 (0.848, 1.074) (0.842, 1.132) (0.725, 1.065) (0.876, 1.145) (0.683, 1.122) ≥ 10 20 8,162/19,095 0.959 0.298 60.0/0.000 0.963 0.442 55.3/0.002 0.915 0.273 44.7/0.017 0.976 0.599 46.3/0.013 0.902 0.250 52.2/0.004 (0.886, 1.038/) (0.875, 1.060) (0.782, 1.072) (0.890, 1.070) (0.756, 1.075) Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 12 of 21 Table 5 Publication bias assessment of this meta-analysis publication bias, sample size, or insufficient statistical power. In addition, evidences have showed that studies SNPs Genetic model Egger’s test Begg’s test which deviate from HWE in controls may reflect the pres- t-value P z-value P ence of genotyping errors, population stratification, and rs266729 Allelic model 0.60 0.552 0.47 0.639 selection bias in the controls (or without representation of Dominant model 0.77 0.451 0.62 0.536 studied sample). Thus, including those studies may de- Recessive model −0.67 0.507 0.92 0.358 crease the quality of a meta-analysis or generate inconsist- Heterozygote model 0.79 0.435 0.81 0.420 ent results [67]. Homozygote model −0.45 0.658 0.73 0.464 Heterogeneity across all the studies of the associations should be noted because it may potentially affect the rs2241766 Allelic model 3.52 0.001 2.16 0.031 strengths of the present meta-analysis. We, thus, used Dominant model 3.63 0.001 2.99 0.003 random effect model. Our results showed that sample Recessive model 0.72 0.476 0.40 0.687 size and quality score are the factors of heterogeneity Heterozygote model 3.17 0.003 2.97 0.003 across all studies of association between rs2241766 poly- Homozygote model 0.88 0.383 0.33 0.744 morphisms and CVD, but no factors contribute the het- rs1501299 Allelic model −0.80 0.427 0.96 0.337 erogeneity across all studies of association between rs266729/rs1501299 polymorphisms and CVD. However, Dominant model 0.09 0.930 0.13 0.895 heterogeneity was still high in the subgroup analysis of Recessive model −2.24 0.031 2.11 0.035 the two factors. For these reasons, heterogeneity might Heterozygote model 0.60 0.549 0.11 0.911 be explained by other confounding factors, such as Homozygote model −1.45 0.155 1.49 0.137 gene-gene interaction and gene-environment interaction. Our meta-analysis has some limitations. Firstly, signifi- allelic, dominant, recessive, and homozygote models [61]. cant publication bias was found in the analysis of Zhou et al. performed a meta-analysis of the association rs2241766 (under allelic, dominant, and heterozygote between rs2241766 and CVD risk in allelic model, and models) and rs1501299 (under recessive model). they found that rs2241766 is associated with the increased Secondly, our meta-analysis mainly included Europeans risk of CVD [63]. In our meta-analysis, we found that and Asians with only few other races, thus limiting our rs2241766 is associated with the increased risk of CVD in power to generalize our findings in other races. Finally, overall population and East Asian in all the five genetic our results might be affected by the potential weaknesses models, and in West Asian in allelic, recessive, and homo- of genetic association studies, such as phenotype zygote models. Our findings is in agreement with the re- misclassifications, genotyping error, population stratifica- sults of Zhou et al., but is in disagreement with the results tion, gene-environment or gene-gene interactions, and of Yang et al., Zhang et al., and Zhou et al. selective reporting biases [68, 69]. With regard to the association between 1,501,299 and Despite the limitations highlighted above, our meta- CVD, the results are also conflicting [50, 60, 61]. Zhou et analysis also had some strength. Firstly, we searched al. revealed no significant association between rs1501299 extensively and investigated more studies and more polymorphism with CAD susceptibility [61]. Qi et al. re- participants than any other meta-analyses performed on ported the extremely large decrease in CVD risk associated the association between ADIPOQ variant and CVD, with rs1501299 polymorphism in diabetic patients [24]. which give our study more statistical power to draw Zhang et al. reported only the weak protective effect of the valid conclusion on this issue. Secondly, sensitivity ana- rs1501299 variant against CVD in general study subjects lysis showed that the results of our meta-analysis are [50]. The meta-analysis by Zhao et al. revealed that stable and robust. Thirdly, the evidence of our results rs1501299 polymorphism may play a protective role for are sufficient to reach concrete conclusions, which were CAD among patients with T2DM [22]. In comparison, our proved by TSA for the first time. We strongly believe results revealed no significant association. our findings will help settle some of the controversies Different genetic admixture and environmental factors surrounding the ADIPOQ-CVD association research. among human populations, which tend to explain ethnic background, strongly modulate the effects of ADIPOQ Conclusions polymorphisms on adiponectin levels [65, 66]. Studies Our meta-analysis found significant increased CVD risk is have reported that low levels of adiponectin (hypoadipoec- associated with rs266729 and rs2241766, but not tinemia) correlate with the risk of CVD, and high levels of associated with rs1501299. Investigating gene–gene and adiponectin protect against this disease [6–11]. These gene–environment interactions is needed to give more conflicting results of associations between the ADIPOQ insight into the genetic association between ADIPOQ polymorphism and CVD risk may be due to differences in variants and CVD. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 13 of 21 Fig. 4 Sensitivity analyses of the association between rs266729 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 14 of 21 Fig. 5 Sensitivity analyses of the association between rs2241766 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 15 of 21 Fig. 6 Sensitivity analyses of the association between rs1501299 polymorphism and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 16 of 21 Fig. 7 Trial sequential analysis of the association between rs266729 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 17 of 21 Fig. 8 Trial sequential analysis of the association between rs2241766 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 18 of 21 Fig. 9 Trial sequential analysis of the association between rs1501299 and CVD risk. (a) allelic model; (b) dominant model; (c) recessive model; (d) heterozygote model; (e) homozygote model Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 19 of 21 Additional files 5. Sun LY, Lee EW, Zahra A, Park JH. Risk factors of cardiovascular disease and their related socio-economical, environmental and health behavioral factors: focused on low-middle income countries- a narrative review article. Additional file 1: Summary of three SNPs characteristics. (DOCX 48 kb) Iran J Public Health. 2015;44:435–44. Additional file 2: Meta-regression results of the association between 6. Stojanovic S, Ilic MD, Ilic S, Petrovic D, Djukic S. The significance of the SNPs and CVD risk. (DOCX 59 kb) adiponectin as a biomarker in metabolic syndrome and/or coronary artery disease. Vojnosanit Pregl. 2015;72:779–84. Additional file 3: Funnel plots of three SNPs for publication bias. (DOCX 1250 kb) 7. Zhang HL, Jin X. Relationship between serum adiponectin and osteoprotegerin levels and coronary heart disease severity. Genet Mol Res. 2015;14:11023–9. Additional file 4: Additional references. (DOCX 22 kb) 8. Md Sayed AS, Zhao Z, Guo L, Li F, Deng X, Deng H, Xia K, Yang T. Serum lectin-like oxidized-low density lipoprotein receptor-1 and adiponectin Abbreviations levels are associated with coronary artery disease accompanied with CAD: Coronary artery disease; CHD: Coronary heart disease; metabolic syndrome. Iran Red Crescent Med J. 2014;16:e12106. CVD: Cardiovascular disease; HWE: Hardy–Weinberg equilibrium; 9. Di Chiara T, Argano C, Scaglione A, Duro G, Corrao S, Scaglione R, Licata G. MI: Myocardial infarction; RIS: Required information size; SNPs: Single- Hypoadiponectinemia, cardiometabolic comorbidities and left ventricular nucleotide polymorphisms; T2DM: Type 2 diabetic mellitus; TSA: Trial hypertrophy. Intern Emerg Med. 2015;10:33–40. sequential analysis 10. Obata Y, Yamada Y, Kyo M, Takahi Y, Saisho K, Tamba S, Yamamoto K, Katsuragi K, Matsuzawa Y. Serum adiponectin levels predict the risk of Acknowledgements coronary heart disease in Japanese patients with type 2 diabetes. J Diabetes Investig. 2013;4:475–82. The authors would like to thank staff and students of the Meta-analysis Group of the Department of Epidemiology and Biostatistics, School of Public 11. Pischon T, Girman CJ, Hotamisligil GS, Rifai N, Hu FB, Rimm EB. Plasma Health of Jilin University for their contributions to this work. adiponectin levels and risk of myocardial infarction in men. JAMA. 2004;291:1730–7. 12. Scherer PE, Williams S, Fogliano M, Baldini G, Lodish HF. A novel serum Funding protein similar to C1q, produced exclusively in adipocytes. J Biol Chem. This work was supported by The National Natural Science Foundation of China 1995;270:26746–9. (Grant 81573230), the Ministry of Science and Technology of the People’s 13. Shinoda Y, Yamaguchi M, Ogata N, Akune T, Kubota N, Yamauchi T, Republic of China (grant number: 2015DFA31580), and the Science and Terauchi Y, Kadowaki T, Takeuchi Y, Fukumoto S, et al. Regulation of bone Technology Commission of Jilin Province (grant number: 20150101130JC). formation by adiponectin through autocrine/paracrine and endocrine pathways. J Cell Biochem. 2006;99:196–208. Availability of data and materials 14. Krause MP, Liu Y, Vu V, Chan L, Xu A, Riddell MC, Sweeney G, Hawke Please contact author for data requests. TJ. Adiponectin is expressed by skeletal muscle fibers and influences muscle phenotype and function. Am J Physiol Cell Physiol. Authors’ contributions 2008;295:C203–12. Conception and design: JSK, SQ, YC, and YL. Provision of study materials: JSK, 15. Pineiro R, Iglesias MJ, Gallego R, Raghay K, Eiras S, Rubio J, Dieguez C, SQ, RL, and CK; Collection and assembly of data: JSK, SQ, and RL. Data Gualillo O, Gonzalez-Juanatey JR, Lago F. Adiponectin is synthesized analysis and interpretation: JSK, SQ, and RL. Manuscript writing: JSK and SQ. and secreted by human and murine cardiomyocytes. FEBS Lett. Revised the language/article: All authors. Final approval of manuscript: All 2005;579:5163–9. authors. 16. Fujita K, Maeda N, Sonoda M, Ohashi K, Hibuse T, Nishizawa H, Nishida M, Hiuge A, Kurata A, Kihara S, et al. Adiponectin protects against angiotensin Ethics approval and consent to participate II-induced cardiac fibrosis through activation of PPAR-alpha. Arterioscler Not applicable. Thromb Vasc Biol. 2008;28:863–70. 17. Ouchi N, Kihara S, Arita Y, Maeda K, Kuriyama H, Okamoto Y, Hotta K, Competing interests Nishida M, Takahashi M, Nakamura T, et al. Novel modulator for endothelial The authors declared that they have no competing interest. adhesion molecules: adipocyte-derived plasma protein adiponectin. Circulation. 1999;100:2473–6. 18. Shibata R, Sato K, Pimentel DR, Takemura Y, Kihara S, Ohashi K, Funahashi T, Publisher’sNote Ouchi N, Walsh K. Adiponectin protects against myocardial ischemia- Springer Nature remains neutral with regard to jurisdictional claims in reperfusion injury through AMPK- and COX-2-dependent mechanisms. Nat published maps and institutional affiliations. Med. 2005;11:1096–103. 19. Kato H, Kashiwagi H, Shiraga M, Tadokoro S, Kamae T, Ujiie H, Honda S, Author details Miyata S, Ijiri Y, Yamamoto J, et al. Adiponectin acts as an endogenous Department of Epidemiology and Biostatistics, School of Public Health of antithrombotic factor. Arterioscler Thromb Vasc Biol. 2006;26:224–30. Jilin University, 1163 Xinmin Street, Changchun 130021, China. The Cardiovascular Center, the First Hospital of Jilin University, Changchun 20. Ouchi N, Kihara S, Arita Y, Nishida M, Matsuyama A, Okamoto Y, Ishigami M, Kuriyama H, Kishida K, Nishizawa H, et al. Adipocyte-derived plasma protein, 130021, China. adiponectin, suppresses lipid accumulation and class a scavenger receptor expression in human monocyte-derived macrophages. Circulation. Received: 22 February 2018 Accepted: 4 May 2018 2001;103:1057–63. 21. Takahashi M, Arita Y, Yamagata K, Matsukawa Y, Okutomi K, Horie M, Shimomura I, Hotta K, Kuriyama H, Kihara S, et al. Genomic structure and References mutations in adipose-specific gene, adiponectin. Int J Obes Relat Metab 1. Roth GA, Huffman MD, Moran AE, Feigin V, Mensah GA, Naghavi M, Murray Disord. 2000;24:861–8. CJ. Global and regional patterns in cardiovascular mortality from 22. Zhao N, Li N, Zhang S, Ma Q, Ma C, Yang X, Yin J, Zhang R, Li J, Yang X, Cui T. 1990 to 2013. Circulation. 2015;132:1667–78. Associations between two common single nucleotide polymorphisms 2. Bhatnagar A. Environmental determinants of cardiovascular disease. (rs2241766 and rs1501299) of ADIPOQ gene and coronary artery disease in Circ Res. 2017;121:162–80. type 2 diabetic patients: a systematic review and meta-analysis. Oncotarget. 3. Roberts R, Stewart AF. Genes and coronary artery disease: where are we? J 2017;8:51994. Am Coll Cardiol. 2012;60:1715–21. 23. Katakami N, Kaneto H, Matsuoka TA, Takahara M, Maeda N, Shimizu I, Ohno 4. Ji F, Ning F, Duan H, Kaprio J, Zhang D, Zhang D, Wang S, Qiao Q, Sun J, K, Osonoi T, Kawai K, Ishibashi F, et al. Adiponectin G276T gene Liang J, et al. Genetic and environmental influences on cardiovascular polymorphism is associated with cardiovascular disease in Japanese disease risk factors: a study of Chinese twin children and adolescents. patients with type 2 diabetes. Atherosclerosis. 2012;220:437–42. Twin Res Hum Genet. 2014;17:72–9. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 20 of 21 24. Qi L, Doria A, Manson JE, Meigs JB, Hunter D, Mantzoros CS, Hu FB. 44. Oliveira CS, Saddi-Rosa P, Crispim F, Canani LH, Gerchman F, Giuffrida FM, Adiponectin genetic variability, plasma adiponectin, and cardiovascular risk Vieira JG, Velho G, Reis AF. Association of ADIPOQ variants, total and high in patients with type 2 diabetes. Diabetes. 2006;55:1512–6. molecular weight adiponectin levels with coronary artery disease in diabetic 25. Kaftan AN, Hussain MK. Association of adiponectin gene polymorphism and non-diabetic Brazilian subjects. J Diabetes Complicat. 2012;26:94–8. rs266729 with type two diabetes mellitus in Iraqi population. A pilot study. 45. Mofarrah M, Ziaee S, Pilehvar-Soltanahmadi Y, Zarghami F, Boroumand M, Gene. 2015;570:95–9. Zarghami N. Association of KALRN, ADIPOQ, and FTO gene polymorphism in type 2 diabetic patients with coronary artery disease: possible 26. Tsai MK, Wang HM, Shiang JC, Chen IH, Wang CC, Shiao YF, Liu WS, Lin TJ, predisposing markers. Coron Artery Dis. 2016;27:490–6. Chen TM, Chen YH. Sequence variants of ADIPOQ and association with type 2 diabetes mellitus in Taiwan Chinese Han population. 46. Chang YC, Jiang JY, Jiang YD, Chiang FT, Hwang JJ, Lien WP, Chuang LM. ScientificWorldJournal. 2014;2014:650393. Interaction of ADIPOQ genetic polymorphism with blood pressure and 27. Cheong MY, Bang OS, Cha MH, Park YK, Kim SH, Kim YJ. Association of the plasma cholesterol level on the risk of coronary artery disease. Circ J. adiponectin gene variations with risk of ischemic stroke in a Korean 2009;73:1934–8. population. Yonsei Med J. 2011;52:20–5. 47. Bacci S, Menzaghi C, Ercolino T, Ma X, Rauseo A, Salvemini L, Vigna C, 28. Liu F, He Z, Deng S, Zhang H, Li N, Xu J. Association of adiponectin gene Fanelli R, Di Mario U, Doria A, Trischitta V. The +276 G/T single nucleotide polymorphisms with the risk of ischemic stroke in a Chinese Han polymorphism of the adiponectin gene is associated with coronary artery population. Mol Biol Rep. 2011;38:1983–8. disease in type 2 diabetic patients. Diabetes Care. 2004;27:2015–20. 29. Chiodini BD, Specchia C, Gori F, Barlera S, D'OrazioA,Pietri S,Crociati L,Nicolucci 48. Esteghamati A, Mansournia N, Nakhjavani M, Mansournia MA, Nikzamir A, A, Franciosi M, Signorini S, et al. Adiponectin gene polymorphisms and their Abbasi M. Association of +45(T/G) and +276(G/T) polymorphisms in the adiponectin gene with coronary artery disease in a population of Iranian effect on the risk of myocardial infarction and type 2 diabetes: an association studyinanItalian population.TherAdv Cardiovasc Dis. 2010;4:223–30. patients with type 2 diabetes. Mol Biol Rep. 2012;39:3791–7. 49. Foucan L, Maimaitiming S, Larifla L, Hedreville S, Deloumeaux J, 30. Shaker OG, Ismail MF. Association of genetic variants of MTHFR, ENPP1, and Joannes MO, Blanchet-Deverly A, Velayoudom-Céphise FL, Aubert R, ADIPOQ with myocardial infarction in Egyptian patients. Cell Biochem Salamon R, et al. Adiponectin gene variants, adiponectin isoforms and Biophys. 2014;69:265–74. cardiometabolic risk in type 2 diabetic patients. Journal of Diabetes 31. Gu CY, Li QX, Zhu Y, Wang MY, Shi TY, Yang YY, Wang JC, Jin L, Wei QY, Ye Investigation. 2014;5:192–8. DW. Genetic variations of the ADIPOQgene and risk of prostate cancer in Chinese Han men. Asian J Androl. 2014;16:878–83. 50. Zhang H, Mo X, Hao Y, Gu D. Association between polymorphisms in the 32. Zhang G, Gu C, Zhu Y, Luo L, Dong D, Wan F, Zhang H, Shi G, Sun L, Ye D. adiponectin gene and cardiovascular disease: a meta-analysis. BMC Med ADIPOQ polymorphism rs182052 is associated with clear cell renal cell Genet. 2012;13:40. carcinoma. Cancer Sci. 2015;106:687–91. 51. Chanock SJ, Manolio T, Boehnke M, Boerwinkle E, Hunter DJ, Thomas G, Hirschhorn JN, Abecasis G, Altshuler D, Bailey-Wilson JE, et al. Replicating 33. Chung HF, Long KZ, Hsu CC, Mamun AA, Chiu YF, Tu HP, Chen PS, Jhang genotype-phenotype associations. Nature. 2007;447:655–60. HR, Hwang SJ, Huang MC. Adiponectin gene (ADIPOQ) polymorphisms 52. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in correlate with the progression of nephropathy in Taiwanese male patients meta-analyses. BMJ. 2003;327:557–60. with type 2 diabetes. Diabetes Res Clin Pract. 2014;105:261–70. 34. Elshamaa MF, Sabry SM, El-Sonbaty MM, Elghoroury EA, Emara N, Raafat M, 53. Stuck AE, Rubenstein LZ, Wieland D. Bias in meta-analysis detected by a Kandil D, Elsaaid G. Adiponectin: an adipocyte-derived hormone, and its simple, graphical test. Asymmetry detected in funnel plot was probably due gene encoding in children with chronic kidney disease. BMC Res Notes. to true heterogeneity. BMJ. 1998;316:469. author reply 470-461 2012;5:174. 54. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis 35. Yoshihara K, Yahata T, Kashima K, Mikami T, Tanaka K. Association of single detected by a simple, graphical test. BMJ. 1997;315:629–34. nucleotide polymorphisms in adiponectin and its receptor genes with 55. Turner RM, Bird SM, Higgins JP. The impact of study size on meta-analyses: polycystic ovary syndrome. J Reprod Med. 2009;54:669–74. examination of underpowered studies in Cochrane reviews. PLoS One. 2013;8:e59202. 36. Machado JS, Palei AC, Amaral LM, Bueno AC, Antonini SR, Duarte G, 56. Brok J, Thorlund K, Wetterslev J, Gluud C. Apparently conclusive meta- Tanus-Santos JE, Sandrim VC, Cavalli RC. Polymorphisms of the analyses may be inconclusive–trial sequential analysis adjustment of adiponectin gene in gestational hypertension and pre-eclampsia. J Hum random error risk due to repetitive testing of accumulating data in Hypertens. 2014;28:128–32. apparently conclusive neonatal meta-analyses. Int J Epidemiol. 37. Kanu JS, Gu Y, Zhi S, Yu M, Lu Y, Cong Y, Liu Y, Li Y, Yu Y, Cheng Y, 2009;38:287–98. Liu Y. Single nucleotide polymorphism rs3774261 in the AdipoQ gene is associated with the risk of coronary heart disease (CHD) in 57. Higgins JP, Whitehead A, Simmonds M. Sequential methods for random- northeast Han Chinese population: a case-control study. Lipids effects meta-analysis. Stat Med. 2011;30:903–21. Health Dis. 2016;15:6. 58. Wang F, Qin Z, Si S, Tang J, Xu L, Xu H, Li R, Han P, Yang H: Lack of 38. Mohammadzadeh G, Ghaffari MA, Heibar H, Bazyar M. Association of two association between NAT2 polymorphism and prostate cancer risk: a meta- common single nucleotide polymorphisms (+45T/G and +276G/T) of analysis and trial sequential analysis. Oncotarget 2017. ADIPOQ gene with coronary artery disease in type 2 diabetic patients. Iran 59. Zhu M, Wen X, Liu X, Wang Y, Liang C, Tu J: Association between 8q24 Biomed J. 2016;20:152–60. rs6983267 polymorphism and cancer susceptibility: a meta-analysis involving 170,737 subjects. Oncotarget 2017. 39. Du SX, Lu LL, Liu Y, Dong QJ, Xuan SY, Xin YN. Association of Adiponectin Gene Polymorphisms with the risk of coronary artery disease in patients 60. Yang Y, Zhang F, Ding R, Wang Y, Lei H, Hu D. Association of ADIPOQ gene with nonalcoholic fatty liver disease in a Chinese Han population. Hepat polymorphisms and coronary artery disease risk: a meta-analysis based on Mon. 2016;16:e37388. 12 465 subjects. Thromb Res. 2012;130:58–64. 40. Zhang M, Peng Y, Lyu S. Association between genetic variants in the 61. Zhou L, Xi B, Wei Y, Pan H, Yang W, Shen W, Li Y, Cai J, Tang H. Association adiponectin gene and premature myocardial infarction. Chinese Journal of between adiponectin gene polymorphisms and coronary artery disease Cardiology. 2016;44:577–82. across different populations. Thromb Res. 2012;130:52–7. 41. Stenvinkel P, Marchlewska A, Pecoits-Filho R, Heimburger O, Zhang Z, Hoff 62. Zhang BC, Li WM, Xu YW. A meta-analysis of the association of adiponectin C, Holmes C, Axelsson J, Arvidsson S, Schalling M, et al. Adiponectin in renal gene polymorphisms with coronary heart disease in Chinese Han disease: relationship to phenotype and genetic variation in the gene population. Clin Endocrinol. 2012;76:358–64. encoding adiponectin. Kidney Int. 2004;65:274–81. 63. Zhou D, Jin Y, Yao F, Duan Z, Wang Q, Liu J. Association between the 42. Pischon T, Pai JK, Manson JE, Hu FB, Rexrode KM, Hunter D, Rimm EB. Single adiponectin +45T>G genotype and risk of cardiovascular disease: a meta- nucleotide polymorphisms at the adiponectin locus and risk of coronary analysis. Heart Lung Circ. 2014;23:159–65. heart disease in men and women. Obesity (Silver Spring). 2007;15:2051–60. 64. Sun K, Li Y, Wei C, Tong Y, Zheng H, Guo Y. Recessive protective effect of 43. Jung CH, Rhee EJ, Kim SY, Shin HS, Kim BJ, Sung KC, Kim BS, Lee WY, Kang ADIPOQ rs1501299 on cardiovascular diseases with type 2 diabetes: a JH, Oh KW, et al. Associations between two single nucleotide meta-analysis. Mol Cell Endocrinol. 2012;349:162–9. polymorphisms of adiponectin gene and coronary artery diseases. 65. Gu HF. Biomarkers of adiponectin: plasma protein variation and genomic Endocr J. 2006;53:671–7. DNA polymorphisms. Biomark Insights. 2009;4:123–33. Kanu et al. Lipids in Health and Disease (2018) 17:126 Page 21 of 21 66. Enns JE, Taylor CG, Zahradka P. Variations in Adipokine genes AdipoQ, Lep, and LepR are associated with risk for obesity-related metabolic disease: the modulatory role of gene-nutrient interactions. J Obes. 2011;2011:168659. 67. Boccia S, De Feo E, Galli P, Gianfagna F, Amore R, Ricciardi G. A systematic review evaluating the methodological aspects of meta-analyses of genetic association studies in cancer research. Eur J Epidemiol. 2010;25:765–75. 68. Ioannidis JP, Trikalinos TA. Early extreme contradictory estimates may appear in published research: the Proteus phenomenon in molecular genetics research and randomized trials. J Clin Epidemiol. 2005;58:543–9. 69. Daly AK, Day CP. Candidate gene case-control association studies: advantages and potential pitfalls. Br J Clin Pharmacol. 2001;52:489–99.

Journal

Lipids in Health and DiseaseSpringer Journals

Published: May 28, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off