GENETIC TESTING AND MOLECULAR BIOMARKERS Volume 18, Number 6, 2014 ª Mary Ann Liebert, Inc. Pp. 377–382 DOI: 10.1089/gtmb.2013.0501
ORIGINAL ARTICLES
FGB Gene - 148C > T Polymorphism Is Associated with Increased Risk of Ischemic Stroke in a Chinese Population: A Meta-Analysis Based on 18 Case–Control Studies Li-Jun Zhang, He-Hua Li, Sheng-Bo Tao, Bin Yuan, Hai-Qing Yan, Li Chang, and Jian-Hua Zhao
Background: Several published articles investigated the relationship between a polymorphism - 148C > T in the b-fibrinogen gene (FGB) and risk of ischemic stroke, and did not reach the same conclusion. To shed light on these inconclusive findings, we performed a meta-analysis of studies relating the FGB genetic polymorphism ( - 148C > T) to the risk of ischemic stroke. Methods: We identified articles by searching PubMed, EMBASE, Chinese National Knowledge Infrastructure databases (CNKI), and Wanfang database in China and by reviewing the references of retrieved articles. We included studies that reported odds ratio (OR) with 95% confidence interval (CI) for the association between the FGB - 148C > T polymorphism and stroke risk. Data from eligible studies were extracted for meta-analysis. Stroke risk associated with FGB - 148C > T polymorphism was estimated by pooled ORs and 95% CIs. The software Review Manager (version 5.2) was utilized for meta-analysis. Publication bias was tested by funnel plot. Results: Eighteen independent case–control studies containing 2159 ischemic stroke patients and 3222 control subjects were included. Our results showed that - 148C > T polymorphism in the FBG gene was associated with increased risk of ischemic stroke ([TT + CT] vs. CC: OR = 1.40, 95% CI [1.20–1.45], p < 0.0001; T vs. C: OR = 1.35, 95% CI [1.18–1.56], p < 0.0001, respectively] by a meta-analysis. Conclusion: The results of our meta-analysis suggested that the -148C > T polymorphism in the FGB gene is a susceptibility marker of ischemic stroke.
Introduction
I
schemic stroke, after ischemic heart disease, is the commonest cause of mortality worldwide and caused about 5.7 million deaths in 2005 (WHO, 2000). China is the world’s most populous country and stroke is an enormous healthcare burden, representing the second leading cause of death and occurring in *2 million people each year (Wei et al., 2010). Prospective studies with large samples have suggested that the plasma fibrinogen level is an independent risk factor for ischemic stroke (Wilhelmsen et al., 1984; Maresca et al., 1999). Elevated plasma fibrinogen levels can be affected by genetic factors. Fibrinogen is encoded by three separate genes located in a 50-kb cluster on the long arm of chromosome 4, which encode for the a, b, and g chains (Kant et al., 1985). The rate limiting step in fibrinogen formation is the synthesis of the b-polypeptide chain regulated by a b-fibrinogen promoter (Roy et al., 1990). Polymorphisms of the b-fibrinogen gene (FGB), including the b 2148 C > T polymorphism, have been shown to be related to the elevation of the plasma fibrinogen level (Xu et al., 2008; Guo et al., 2009; Yuan et al., 2009). Several studies have suggested that
the FGB 2148C > T polymorphism is associated with increased risk of ischemic stroke (Fu et al., 2005; Gao et al., 2006; Liang et al., 2006; Ma et al., 2006; Chen et al., 2007). However, the relatively small sample size of a single study may not have enough power to detect slight effects of 2148C > T polymorphism on stroke; meta-analysis may provide more credible evidence by systematically summarizing the existing data. Although a previous meta-analysis (Chen et al., 2007) has reported on the relationship between FBG gene polymorphism and stroke, it involved a small sample size and had lower power. In addition, there are several new articles that have been published since then. Therefore, we have collected almost all published case– control studies to perform a meta-analysis to further explore the relationship between FGB gene - 148C > T polymorphism and ischemic stroke. Materials and Methods Literature search and selection
We carried out a publication search in PubMed, EMBASE, Chinese National Knowledge Infrastructure [CNKI], and
Department of Neurology, First Affiliated Hospital of Xinxiang Medical College, Xinxiang, People’s Republic of China.
377
378
Wanfang database in China, with the following search terms: [‘‘fibrinogen’’ or ‘‘b-fibrinogen’’ or ‘‘FGB’’] and [‘‘cerebral infarction’’ or‘‘stroke’’ or ‘‘brain infarction’’ or ‘‘cerebrovascular disease’’] and [‘‘SNP’’ or ‘‘polymorphism’’ or ‘‘mutation’’ or ‘‘genetics’’] by two independent investigators. Publication language was restricted to English and Chinese, and the subjects were limited to Chinese in our search. By means of online retrieval and literature review, references obtained using the above-mentioned databases were reviewed again to ensure that no relevant studies were missed. Selection criteria
Inclusion criteria were as follows: (1) independently published case–control or cohort studies on the relationship between FBG gene polymorphism and stroke; (2) with comprehensive statistical indicators directly or indirectly: odds ratio (OR) or relative risk values and 95% confidence interval (CI); and (3) similar themes and methods, that is, case–control or cohort studies about the relationship of the FGB gene polymorphism and stroke. Articles were excluded if relevant data were not available or there was heterogeneity of gene polymorphism in the control population. For the heterogeneity test method, we utilized the Q-test and I2 test of the RevMan 5.2 software. Data extraction
Two reviewers independently evaluated the research design, enrolled patients, observation results of the literature, and selected trials according to the above-mentioned inclusion criteria. Inconsistencies were resolved through discussion. We used the Cochrane Handbook 5.2 quality evaluation criteria to assess the methodological quality of included studies such as study subjects and impact factors, the source of the cases and controls, matching age and gender. To determine the quality of data by the quality of ultimately determined literature, the useless ones were excluded, such as studies that have been reported repeatedly and those with poor quality or less information and having special selection of laboratory sample; relevant data were extracted from included articles. Because there is C or T allele in the
FIG. 1. Flow diagram of study identification.
ZHANG ET AL.
2148C > T site, the genotype in this locus of the population was identified as CC, CT, or TT. For each study that met our criteria, the following information was collected: first author, year of publication, country of origin, number of cases and controls, genotype distribution, genotyping methods, and allele frequency. Statistical analysis
For each study, we first examined whether the genotype distribution in controls was consistent with the Hardy– Weinberg equilibrium (HWE) by a chi-square test. Metaanalysis was performed using RevMan 5.2 software provided by the Cochrane Collaboration. We used the Q-test and I2 test to examine the heterogeneity between each study. We used OR for efficacy analysis statistics. Using the heterogeneity test, if p > 0.05, we selected the fixed-effects model, and if p < 0.05, we selected the random-effects model to merge the OR. p < 0.05 was considered as a significant difference. Analysis of sensitivity includes the difference of point estimation and CIs of the combined effects value of different models to observe whether it changes the result; poor quality articles were excluded or reanalyzed according to the quality evaluation criteria to determine whether it changed the findings. To test the publication bias, we used the RevMan 5.2 statistical software to make the funnel plot. Results Literature search
As shown in Figure 1, 476 studies were preliminarily detected, which includes 414 Chinese articles and 62 English articles; 404 articles were excluded because of duplicate publication and nonclinical-based research literature. Seventy-two studies appeared to be potentially relevant for inclusion in our study. Fifty-one studies were further excluded because they did not detect the 2148C > T genotype. Therefore, 21 full-text articles were reviewed. Three studies were further excluded for no control population. Therefore, a total of 18 articles met the inclusion criteria (Liu et al., 2002, 2004; Lv et al., 2003; Zhang et al., 2003; Zhao et al., 2003; Qian et al., 2004; Fu et al., 2005, 2006; Pan et al., 2005; Song et al., 2005;
FGB AND STROKE
379
Xu et al., 2005; Lu et al., 2007; Liu and Zhao, 2008; Yuan et al., 2009, 2010; Cui et al., 2010; He et al., 2010; Guo et al., 2009). Study characteristics
The characteristics of included studies are summarized in Table 1. The 18 included studies were published between 2000 and 2010 and comprised a total of 2159 ischemic stroke cases and 3222 control subjects. All the subjects included in these studies were Chinese. A classic polymerase chain reaction assay was performed in all of these 18 studies. The genotyping method in all these studies was polymerase chain reaction–restriction fragment length polymorphism (PCRRFLP), and the genotype distributions among the controls of all studies were in agreement with HWE.
Results of the meta-analysis are shown in Figures 2 and 3. These two figures showed the result of meta-analysis of studies on the correlation between ischemic stroke and FGB - 148C > T polymorphism in 18 case–control studies. The systematic reviews of the included studies can be seen from Figures 2 and 3, which include the number of case and control groups, weight, OR value, and 95% CI. The heterogeneity test of the various studies revealed heterogeneous results ( p = 0.04, I2 = 40%; p = 0.01, I2 = 48%); therefore, we used the random-effects model in the analysis. Overall, the association of FGB - 148 TT/CT genotype with a higher risk of ischemic stroke was observed (OR = 1.35, 95% CI [1.18–1.56]; p < 0.0001) (Fig. 2). In addition, we found the - 148 T allele carriers to be associated with the increased risk of ischemic stroke (OR = 1.40, 95% CI [1.20–1.65]; p < 0.0001) (Fig. 3).
Meta-analysis
The association between 2148C > T polymorphism and susceptibility to stroke was analyzed in 18 independent studies with 2159 stroke patients and 3222 control subjects.
Test of sensitivity
For the sensitivity analysis, we deleted one single study from the overall pooled analysis each time to check the
Table 1. The Characteristics of Included Studies
Author
Year
Country
Genotyping methods
Cui
2010
China
PCR-RFLP
Fu
2006
China
PCR-RFLP
He
2010
China
PCR-RFLP
Liu
2004
China
PCR-RFLP
Liu
2008
China
PCR-RFLP
Liu
2002
China
PCR-RFLP
Lu
2007
China
PCR-RFLP
Lv
2003
China
PCR-RFLP
Ma
2006
China
PCR-RFLP
Pan
2005
China
PCR-RFLP
Qian
2004
China
PCR-RFLP
Song
2005
China
PCR-RFLP
Song
2006
China
PCR-RFLP
Xu
2005
China
PCR-RFLP
Yuan
2010
China
PCR-RFLP
Yuan
2009
China
PCR-RFLP
Zhang
2003
China
PCR-RFLP
Zhao
2003
China
PCR-RFLP
- 148C > T (n) Groups
n
CC
CT
TT
Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control
92 98 132 171 92 80 90 102 220 140 96 273 148 130 151 113 151 101 69 60 102 90 88 80 135 120 90 60 160 162 144 1277 48 52 151 113
36 56 75 102 36 56 43 64 105 84 62 177 72 79 70 62 85 70 41 41 66 43 38 47 53 65 52 47 101 112 64 820 25 39 70 62
48 37 50 58 48 37 39 34 85 49 30 86 63 45 65 47 63 30 26 18 30 39 42 27 56 43 32 17 47 42 46 387 18 11 65 47
8 5 7 11 8 5 8 4 30 7 4 10 13 6 16 4 3 1 2 1 6 8 8 6 26 11 6 2 12 8 4 70 5 2 16 4
PCR-RFLP, polymerase chain reaction–restriction fragment length polymorphism.
380
ZHANG ET AL.
FIG. 2. Forest plot of cerebral infarction and 2148C > T polymorphism (TT + CT vs. CC), the horizontal lines correspond to the study-specific odds ratio (OR) and 95% confidence interval (CI), respectively. The area of the squares reflects the study-specific weight. The diamond represents the pooled results of OR and 95% CI.
influence of the removed data set to the overall ORs. The pooled ORs and 95% CIs were not significantly altered when any part of the study was omitted, which indicated that any single study had little impact on the overall ORs.
statistical evidence for bias either ( p = 0.13; p = 0.16, respectively). It indicates that there is no publication bias, and the result of the study is credible. Discussion
Publication bias
RevMan 5.2 software was used to analyze the publication bias; the funnel plot (Fig. 4) showed that the points are evenly distributed and symmetrical, and most of the points are within the 95% CI. Also, the shape of funnel plots showed no obvious asymmetry and the result of Egger’s test did not show
FIG. 3. Forest plot of cerebral infarction and 2148 C > T polymorphism (T vs. C), the horizontal lines correspond to the study-specific OR and 95% CI, respectively. The area of the squares reflects the study-specific weight. The diamond represents the pooled results of OR and 95% CI. In this analysis, the fixed-effects model was used.
In this meta-analysis, we found that the polymorphism 2148C > T in the FGB gene was significantly associated with ischemic stroke in a Chinese population by the pooled results from 18 published studies. The results demonstrated that the 2148T carriers had an increased 35% risk for developing ischemic stroke.
FGB AND STROKE
381
FIG. 4. Begg’s funnel plot for publication bias tests. Each point represents a separate study for the indicated association. LogOR represents natural logarithm of OR. Vertical line represents the mean effects size. (A) (TT + CT) versus CC. (B) T versus C.
It is well known that elevated plasma fibrinogen levels can be affected by environmental and genetic factors. Polymorphisms of the FGB, especially involved in the ratelimiting steps of the formation of the b-chain, have been shown to be closely related to elevation of the plasma fibrinogen level and ischemic stroke (Liu et al., 2002; Fu et al., 2005; Gao et al., 2006; Liang et al., 2006; Ma et al., 2006; Chen et al., 2007; Xu et al., 2008; Guo et al., 2009; Yuan et al., 2009, 2010). In the present study, we combined the results of 18 studies to pool analyze the relationship between 2148C > T polymorphism and ischemic stroke. The result showed that there was 40% increased risk of stroke for the variant genotypes (CT + TT) compared with the wild CC homozygotes. Thus, the allele T might be a genetic risk factor to increase the susceptibility to stroke at the protein and genetic levels. The characteristic of meta-analysis is to combine comparable studies to increase the sample size and statistical power and draw a more compelling result. However, meta-analysis confounds factors such as publication bias, method of sampling, different genetic backgrounds of subjects, different protocols, and quality of analysis. In the present study, we did not find publication bias, all the subjects were Chinese, and the genotypes in all studies were detected with genetic DNA from blood samples using PCR-RFLP genotyping methods. All the studies checked genotypes for quality control. Genotype distribution of controls in all studies was consistent with HWE. In addition, sensitivity analysis also showed that omission of any single study did not have significant impact on the combined ORs. This made the results of this meta-study more reliable to some extent. However, there remained some limitations in this metaanalysis. Although the genotyping methods used in all the studies were the same, other clinical factors such as age, sex, and different chemotherapies in each study might lead to bias. Determining whether or not these factors influence the results of this meta-analysis would need further investigation.
Conclusion
Our study suggested that 2148C > T polymorphism in the FGB gene was associated with a significantly increased risk of ischemic stroke. Larger well-designed epidemiological studies with ethnically diverse populations and functional evaluations are warranted to confirm our findings. Acknowledgment
This work was supported financially by the Project of Science & Technology of Xinxiang Medical College (2007YJAO2). Author Disclosure Statement
No competing financial interests exist. References
Chen XC, Xu MT, Zhou W, et al. (2007) A meta-analysis of relationship between beta-fibrinogen gene - 148C/T polymorphism and susceptibility to cerebral infarction in Han Chinese. Chin Med J [Engl] 120:1198–1202. Cui GF, Meng Z, Xiu ZS, et al. (2010) Relationship between beta-fibrinogen-455 G/AC/T polymorphisms and atrial fibrillation with cerebral infarction. Int J Cerebrovasc Dis 18:654–656. Fu Y, Ni PH, Liu JR, et al. (2006) Study on haplotypes of five b-fibrinogen genes in Han population. J Shanghai Jiaotong Univ [Med Sci] 26:758–760. Fu Y, Wei X, Ni PH, et al. (2005) The relationship between the five beta-fibrinogen gene polymorphisms and cerebral infarction. Zhonghua Nei Ke Za Zhi 44:914–917. Gao X, Yang H, ZhiPing T (2006) Association studies of genetic polymorphism, environmental factors and their interaction in ischemic stroke. Neurosci Lett 398:172–177. Guo X, Zhang D, Zhang X (2009) Fibrinogen gene polymorphism [Bbeta-148C/T] in Uygur patients with cerebral infarction. Neurol Res 31:381–384. He YP, Meng Z, Cui GF, et al. (2010) Study of the relationship of b-fibrinogen - 148C > T gene polymorphism in patients
382
with artial fibrillation and ischemic cardiol and cerebral vascular disease. Shi Yong Xin Nao Fei Xue Guan Bing Za Zhi 18:301–304. Kant JA, Fornace AJ, Jr., Saxe D, et al. (1985) Evolution and organization of the fibrinogen locus on chromosome 4:Gene duplication accompanied by transposition and inversion. Proc Natl Acad Sci U S A 82:2344–2348. Liang L, Sun C, Liao XP, et al. (2006) A related analysis for alpha, beta fibrinogen gene haplotypes and nucleotide polymorphisms associated with the ischemic stroke in Hainan Han population. Zhong hua Yi Xue Yi Chuan Xue Za Zhi 23:316–319. Liu CF, Qian JJ, Zhao KR (2004) Study on the b-fibrinogen gene - 455A/G and - 148C/T polymorphism and cerebral infarction. Chin J Neurol 37:90–91. Liu R, Li J, Mu H, et al. (2002) The relationship of betafibrinogen gene polymorphisms and ischaemic cardiocerebral vascular disease. Zhonghua Xue Ye Xue Za Zhi 23:453–456. Liu Z, Zhao B (2008) The association of beta fibrinogen -148C > T gene and plasminogen activator 1–675 4G/5G gene polymorphism with cerebral infarction. Nerv Dis Ment Health 8:94–97. Lu SJ, Zhao B, Chen YS (2007) Study on the relation between beta fibrinogen - 148C > T gene polymorphism and cerebral infarction. Chin J Nerv Ment Dis 33:425–428. Lv B, Zhao B, Xing YQ, et al. (2003) The study on the polymorphism of beta fibrinogen gene - 148C > T in the patients with cerebral infarction. J Clin Neurol 16:202–205. Ma AJ, Pan XD, Zhang CS, et al. (2006) A linkage between beta-fibrinogen gene - 148C/T polymorphism and cerebral infarction. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 23: 202–204. Maresca G, Di Blasio A, Marchioli R, Di Minno G (1999) Measuring plasma fibrinogen to predict stroke and myocardial infarction: an update. Arterioscler Thromb Vasc Biol 19:1368–1377. Pan XD, Ma AJ, Zou ZQ, et al. (2005) Study of beta fibrinogen gene - 148C > T polymorphism and plasma fibrinogen levels in young patients with acute cerebral infarction. J Clin Neurol 18:40–44. Qian JJ, Liu CF, Yue KR (2004) The relationship between bfibrinogen gene - 455A/G and - 148C/T polymorphism and cerebral infarction with carotid atherosclerosis. Jiangsu Med J 30:377–379. Roy SN, Mukhopadhyay G, Redman CM (1990) Regulation of fibrinogen assembly. Transfection of Hep G2 cells with B
ZHANG ET AL.
beta cDNA specifically enhances synthesis of the three component chains of fibrinogen. J Biol Chem 265:6389– 6393. Song YQ, Zou HL, Song XJ, Wang W (2005) Fibrinogen gene polymorphism [beta-148C > T] in patients with cerebral infarction. Cerebrovasc Dis Foreign Med Sci 13:895–897. Wei JW, Heeley EL, Jan S, et al. (2010) Variations and determinants of hospital costs for acute stroke in China. PLoS One 5:e13041. WHO (2000) The World Health Report 2000. Health Systems: Improving Performance. Geneva: World Health Organization. Wilhelmsen L, Sva¨rdsudd K, Korsan-Bengtsen K, et al. (1984) Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 311:501–505. Xu F, Ni PH, Ying YY, Hu YQ (2005) Fibrinogen B b gene -148 C > T polymorphism in stroke. J Diagn Concepts Pract 4:38–41. Xu X, Li J, Sheng W, Liu L (2008) Meta-analysis of genetic studies from journals published in China of ischemic stroke in the Han Chinese population. Cerebrovasc Dis: 26:48–62. Yuan XD, Wang SJ, Xu YR, et al. (2009) Relationship between multi-locus fibrinogen polymorphisms and fibrinogen concentration, molecular reactivity and cerebral infarction. Zhonghua Xue Ye Xue Za Zhi 30:582–587. Yuan XD, Wang SJ, Xu YR, et al. (2010) The relationship of muti-locus gene polymorphisms, functional expression of fibrinogen B b-Chain and the type of cerebral infarction. Zhong Guo Mian Yi Xue Za Zhi 26:356–358. Zhang XH, Xu G, Zhao XY, et al. (2003) Study of fibrinogen B b gene - 148 C > T gene polymorphism in patients with coronary heart disease and stroke. Chin J Emerg Med 12: 683–685. Zhao B, Lv B, Xing YQ, et al. (2003) The beta fibrinogen gene - 148C > T polymorphism and its association with plasma fibrinogen’s function and level in cerebral infarction. Chin J Neurol 36:450–454.
Address correspondence to: Jian-Hua Zhao, MS Department of Neurology First Affiliated Hospital of Xinxiang Medical College Xinxiang 453100 Henan People’s Republic of China E-mail: [email protected]
ORIGINAL ARTICLES
FGB Gene - 148C > T Polymorphism Is Associated with Increased Risk of Ischemic Stroke in a Chinese Population: A Meta-Analysis Based on 18 Case–Control Studies Li-Jun Zhang, He-Hua Li, Sheng-Bo Tao, Bin Yuan, Hai-Qing Yan, Li Chang, and Jian-Hua Zhao
Background: Several published articles investigated the relationship between a polymorphism - 148C > T in the b-fibrinogen gene (FGB) and risk of ischemic stroke, and did not reach the same conclusion. To shed light on these inconclusive findings, we performed a meta-analysis of studies relating the FGB genetic polymorphism ( - 148C > T) to the risk of ischemic stroke. Methods: We identified articles by searching PubMed, EMBASE, Chinese National Knowledge Infrastructure databases (CNKI), and Wanfang database in China and by reviewing the references of retrieved articles. We included studies that reported odds ratio (OR) with 95% confidence interval (CI) for the association between the FGB - 148C > T polymorphism and stroke risk. Data from eligible studies were extracted for meta-analysis. Stroke risk associated with FGB - 148C > T polymorphism was estimated by pooled ORs and 95% CIs. The software Review Manager (version 5.2) was utilized for meta-analysis. Publication bias was tested by funnel plot. Results: Eighteen independent case–control studies containing 2159 ischemic stroke patients and 3222 control subjects were included. Our results showed that - 148C > T polymorphism in the FBG gene was associated with increased risk of ischemic stroke ([TT + CT] vs. CC: OR = 1.40, 95% CI [1.20–1.45], p < 0.0001; T vs. C: OR = 1.35, 95% CI [1.18–1.56], p < 0.0001, respectively] by a meta-analysis. Conclusion: The results of our meta-analysis suggested that the -148C > T polymorphism in the FGB gene is a susceptibility marker of ischemic stroke.
Introduction
I
schemic stroke, after ischemic heart disease, is the commonest cause of mortality worldwide and caused about 5.7 million deaths in 2005 (WHO, 2000). China is the world’s most populous country and stroke is an enormous healthcare burden, representing the second leading cause of death and occurring in *2 million people each year (Wei et al., 2010). Prospective studies with large samples have suggested that the plasma fibrinogen level is an independent risk factor for ischemic stroke (Wilhelmsen et al., 1984; Maresca et al., 1999). Elevated plasma fibrinogen levels can be affected by genetic factors. Fibrinogen is encoded by three separate genes located in a 50-kb cluster on the long arm of chromosome 4, which encode for the a, b, and g chains (Kant et al., 1985). The rate limiting step in fibrinogen formation is the synthesis of the b-polypeptide chain regulated by a b-fibrinogen promoter (Roy et al., 1990). Polymorphisms of the b-fibrinogen gene (FGB), including the b 2148 C > T polymorphism, have been shown to be related to the elevation of the plasma fibrinogen level (Xu et al., 2008; Guo et al., 2009; Yuan et al., 2009). Several studies have suggested that
the FGB 2148C > T polymorphism is associated with increased risk of ischemic stroke (Fu et al., 2005; Gao et al., 2006; Liang et al., 2006; Ma et al., 2006; Chen et al., 2007). However, the relatively small sample size of a single study may not have enough power to detect slight effects of 2148C > T polymorphism on stroke; meta-analysis may provide more credible evidence by systematically summarizing the existing data. Although a previous meta-analysis (Chen et al., 2007) has reported on the relationship between FBG gene polymorphism and stroke, it involved a small sample size and had lower power. In addition, there are several new articles that have been published since then. Therefore, we have collected almost all published case– control studies to perform a meta-analysis to further explore the relationship between FGB gene - 148C > T polymorphism and ischemic stroke. Materials and Methods Literature search and selection
We carried out a publication search in PubMed, EMBASE, Chinese National Knowledge Infrastructure [CNKI], and
Department of Neurology, First Affiliated Hospital of Xinxiang Medical College, Xinxiang, People’s Republic of China.
377
378
Wanfang database in China, with the following search terms: [‘‘fibrinogen’’ or ‘‘b-fibrinogen’’ or ‘‘FGB’’] and [‘‘cerebral infarction’’ or‘‘stroke’’ or ‘‘brain infarction’’ or ‘‘cerebrovascular disease’’] and [‘‘SNP’’ or ‘‘polymorphism’’ or ‘‘mutation’’ or ‘‘genetics’’] by two independent investigators. Publication language was restricted to English and Chinese, and the subjects were limited to Chinese in our search. By means of online retrieval and literature review, references obtained using the above-mentioned databases were reviewed again to ensure that no relevant studies were missed. Selection criteria
Inclusion criteria were as follows: (1) independently published case–control or cohort studies on the relationship between FBG gene polymorphism and stroke; (2) with comprehensive statistical indicators directly or indirectly: odds ratio (OR) or relative risk values and 95% confidence interval (CI); and (3) similar themes and methods, that is, case–control or cohort studies about the relationship of the FGB gene polymorphism and stroke. Articles were excluded if relevant data were not available or there was heterogeneity of gene polymorphism in the control population. For the heterogeneity test method, we utilized the Q-test and I2 test of the RevMan 5.2 software. Data extraction
Two reviewers independently evaluated the research design, enrolled patients, observation results of the literature, and selected trials according to the above-mentioned inclusion criteria. Inconsistencies were resolved through discussion. We used the Cochrane Handbook 5.2 quality evaluation criteria to assess the methodological quality of included studies such as study subjects and impact factors, the source of the cases and controls, matching age and gender. To determine the quality of data by the quality of ultimately determined literature, the useless ones were excluded, such as studies that have been reported repeatedly and those with poor quality or less information and having special selection of laboratory sample; relevant data were extracted from included articles. Because there is C or T allele in the
FIG. 1. Flow diagram of study identification.
ZHANG ET AL.
2148C > T site, the genotype in this locus of the population was identified as CC, CT, or TT. For each study that met our criteria, the following information was collected: first author, year of publication, country of origin, number of cases and controls, genotype distribution, genotyping methods, and allele frequency. Statistical analysis
For each study, we first examined whether the genotype distribution in controls was consistent with the Hardy– Weinberg equilibrium (HWE) by a chi-square test. Metaanalysis was performed using RevMan 5.2 software provided by the Cochrane Collaboration. We used the Q-test and I2 test to examine the heterogeneity between each study. We used OR for efficacy analysis statistics. Using the heterogeneity test, if p > 0.05, we selected the fixed-effects model, and if p < 0.05, we selected the random-effects model to merge the OR. p < 0.05 was considered as a significant difference. Analysis of sensitivity includes the difference of point estimation and CIs of the combined effects value of different models to observe whether it changes the result; poor quality articles were excluded or reanalyzed according to the quality evaluation criteria to determine whether it changed the findings. To test the publication bias, we used the RevMan 5.2 statistical software to make the funnel plot. Results Literature search
As shown in Figure 1, 476 studies were preliminarily detected, which includes 414 Chinese articles and 62 English articles; 404 articles were excluded because of duplicate publication and nonclinical-based research literature. Seventy-two studies appeared to be potentially relevant for inclusion in our study. Fifty-one studies were further excluded because they did not detect the 2148C > T genotype. Therefore, 21 full-text articles were reviewed. Three studies were further excluded for no control population. Therefore, a total of 18 articles met the inclusion criteria (Liu et al., 2002, 2004; Lv et al., 2003; Zhang et al., 2003; Zhao et al., 2003; Qian et al., 2004; Fu et al., 2005, 2006; Pan et al., 2005; Song et al., 2005;
FGB AND STROKE
379
Xu et al., 2005; Lu et al., 2007; Liu and Zhao, 2008; Yuan et al., 2009, 2010; Cui et al., 2010; He et al., 2010; Guo et al., 2009). Study characteristics
The characteristics of included studies are summarized in Table 1. The 18 included studies were published between 2000 and 2010 and comprised a total of 2159 ischemic stroke cases and 3222 control subjects. All the subjects included in these studies were Chinese. A classic polymerase chain reaction assay was performed in all of these 18 studies. The genotyping method in all these studies was polymerase chain reaction–restriction fragment length polymorphism (PCRRFLP), and the genotype distributions among the controls of all studies were in agreement with HWE.
Results of the meta-analysis are shown in Figures 2 and 3. These two figures showed the result of meta-analysis of studies on the correlation between ischemic stroke and FGB - 148C > T polymorphism in 18 case–control studies. The systematic reviews of the included studies can be seen from Figures 2 and 3, which include the number of case and control groups, weight, OR value, and 95% CI. The heterogeneity test of the various studies revealed heterogeneous results ( p = 0.04, I2 = 40%; p = 0.01, I2 = 48%); therefore, we used the random-effects model in the analysis. Overall, the association of FGB - 148 TT/CT genotype with a higher risk of ischemic stroke was observed (OR = 1.35, 95% CI [1.18–1.56]; p < 0.0001) (Fig. 2). In addition, we found the - 148 T allele carriers to be associated with the increased risk of ischemic stroke (OR = 1.40, 95% CI [1.20–1.65]; p < 0.0001) (Fig. 3).
Meta-analysis
The association between 2148C > T polymorphism and susceptibility to stroke was analyzed in 18 independent studies with 2159 stroke patients and 3222 control subjects.
Test of sensitivity
For the sensitivity analysis, we deleted one single study from the overall pooled analysis each time to check the
Table 1. The Characteristics of Included Studies
Author
Year
Country
Genotyping methods
Cui
2010
China
PCR-RFLP
Fu
2006
China
PCR-RFLP
He
2010
China
PCR-RFLP
Liu
2004
China
PCR-RFLP
Liu
2008
China
PCR-RFLP
Liu
2002
China
PCR-RFLP
Lu
2007
China
PCR-RFLP
Lv
2003
China
PCR-RFLP
Ma
2006
China
PCR-RFLP
Pan
2005
China
PCR-RFLP
Qian
2004
China
PCR-RFLP
Song
2005
China
PCR-RFLP
Song
2006
China
PCR-RFLP
Xu
2005
China
PCR-RFLP
Yuan
2010
China
PCR-RFLP
Yuan
2009
China
PCR-RFLP
Zhang
2003
China
PCR-RFLP
Zhao
2003
China
PCR-RFLP
- 148C > T (n) Groups
n
CC
CT
TT
Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control Case Control
92 98 132 171 92 80 90 102 220 140 96 273 148 130 151 113 151 101 69 60 102 90 88 80 135 120 90 60 160 162 144 1277 48 52 151 113
36 56 75 102 36 56 43 64 105 84 62 177 72 79 70 62 85 70 41 41 66 43 38 47 53 65 52 47 101 112 64 820 25 39 70 62
48 37 50 58 48 37 39 34 85 49 30 86 63 45 65 47 63 30 26 18 30 39 42 27 56 43 32 17 47 42 46 387 18 11 65 47
8 5 7 11 8 5 8 4 30 7 4 10 13 6 16 4 3 1 2 1 6 8 8 6 26 11 6 2 12 8 4 70 5 2 16 4
PCR-RFLP, polymerase chain reaction–restriction fragment length polymorphism.
380
ZHANG ET AL.
FIG. 2. Forest plot of cerebral infarction and 2148C > T polymorphism (TT + CT vs. CC), the horizontal lines correspond to the study-specific odds ratio (OR) and 95% confidence interval (CI), respectively. The area of the squares reflects the study-specific weight. The diamond represents the pooled results of OR and 95% CI.
influence of the removed data set to the overall ORs. The pooled ORs and 95% CIs were not significantly altered when any part of the study was omitted, which indicated that any single study had little impact on the overall ORs.
statistical evidence for bias either ( p = 0.13; p = 0.16, respectively). It indicates that there is no publication bias, and the result of the study is credible. Discussion
Publication bias
RevMan 5.2 software was used to analyze the publication bias; the funnel plot (Fig. 4) showed that the points are evenly distributed and symmetrical, and most of the points are within the 95% CI. Also, the shape of funnel plots showed no obvious asymmetry and the result of Egger’s test did not show
FIG. 3. Forest plot of cerebral infarction and 2148 C > T polymorphism (T vs. C), the horizontal lines correspond to the study-specific OR and 95% CI, respectively. The area of the squares reflects the study-specific weight. The diamond represents the pooled results of OR and 95% CI. In this analysis, the fixed-effects model was used.
In this meta-analysis, we found that the polymorphism 2148C > T in the FGB gene was significantly associated with ischemic stroke in a Chinese population by the pooled results from 18 published studies. The results demonstrated that the 2148T carriers had an increased 35% risk for developing ischemic stroke.
FGB AND STROKE
381
FIG. 4. Begg’s funnel plot for publication bias tests. Each point represents a separate study for the indicated association. LogOR represents natural logarithm of OR. Vertical line represents the mean effects size. (A) (TT + CT) versus CC. (B) T versus C.
It is well known that elevated plasma fibrinogen levels can be affected by environmental and genetic factors. Polymorphisms of the FGB, especially involved in the ratelimiting steps of the formation of the b-chain, have been shown to be closely related to elevation of the plasma fibrinogen level and ischemic stroke (Liu et al., 2002; Fu et al., 2005; Gao et al., 2006; Liang et al., 2006; Ma et al., 2006; Chen et al., 2007; Xu et al., 2008; Guo et al., 2009; Yuan et al., 2009, 2010). In the present study, we combined the results of 18 studies to pool analyze the relationship between 2148C > T polymorphism and ischemic stroke. The result showed that there was 40% increased risk of stroke for the variant genotypes (CT + TT) compared with the wild CC homozygotes. Thus, the allele T might be a genetic risk factor to increase the susceptibility to stroke at the protein and genetic levels. The characteristic of meta-analysis is to combine comparable studies to increase the sample size and statistical power and draw a more compelling result. However, meta-analysis confounds factors such as publication bias, method of sampling, different genetic backgrounds of subjects, different protocols, and quality of analysis. In the present study, we did not find publication bias, all the subjects were Chinese, and the genotypes in all studies were detected with genetic DNA from blood samples using PCR-RFLP genotyping methods. All the studies checked genotypes for quality control. Genotype distribution of controls in all studies was consistent with HWE. In addition, sensitivity analysis also showed that omission of any single study did not have significant impact on the combined ORs. This made the results of this meta-study more reliable to some extent. However, there remained some limitations in this metaanalysis. Although the genotyping methods used in all the studies were the same, other clinical factors such as age, sex, and different chemotherapies in each study might lead to bias. Determining whether or not these factors influence the results of this meta-analysis would need further investigation.
Conclusion
Our study suggested that 2148C > T polymorphism in the FGB gene was associated with a significantly increased risk of ischemic stroke. Larger well-designed epidemiological studies with ethnically diverse populations and functional evaluations are warranted to confirm our findings. Acknowledgment
This work was supported financially by the Project of Science & Technology of Xinxiang Medical College (2007YJAO2). Author Disclosure Statement
No competing financial interests exist. References
Chen XC, Xu MT, Zhou W, et al. (2007) A meta-analysis of relationship between beta-fibrinogen gene - 148C/T polymorphism and susceptibility to cerebral infarction in Han Chinese. Chin Med J [Engl] 120:1198–1202. Cui GF, Meng Z, Xiu ZS, et al. (2010) Relationship between beta-fibrinogen-455 G/AC/T polymorphisms and atrial fibrillation with cerebral infarction. Int J Cerebrovasc Dis 18:654–656. Fu Y, Ni PH, Liu JR, et al. (2006) Study on haplotypes of five b-fibrinogen genes in Han population. J Shanghai Jiaotong Univ [Med Sci] 26:758–760. Fu Y, Wei X, Ni PH, et al. (2005) The relationship between the five beta-fibrinogen gene polymorphisms and cerebral infarction. Zhonghua Nei Ke Za Zhi 44:914–917. Gao X, Yang H, ZhiPing T (2006) Association studies of genetic polymorphism, environmental factors and their interaction in ischemic stroke. Neurosci Lett 398:172–177. Guo X, Zhang D, Zhang X (2009) Fibrinogen gene polymorphism [Bbeta-148C/T] in Uygur patients with cerebral infarction. Neurol Res 31:381–384. He YP, Meng Z, Cui GF, et al. (2010) Study of the relationship of b-fibrinogen - 148C > T gene polymorphism in patients
382
with artial fibrillation and ischemic cardiol and cerebral vascular disease. Shi Yong Xin Nao Fei Xue Guan Bing Za Zhi 18:301–304. Kant JA, Fornace AJ, Jr., Saxe D, et al. (1985) Evolution and organization of the fibrinogen locus on chromosome 4:Gene duplication accompanied by transposition and inversion. Proc Natl Acad Sci U S A 82:2344–2348. Liang L, Sun C, Liao XP, et al. (2006) A related analysis for alpha, beta fibrinogen gene haplotypes and nucleotide polymorphisms associated with the ischemic stroke in Hainan Han population. Zhong hua Yi Xue Yi Chuan Xue Za Zhi 23:316–319. Liu CF, Qian JJ, Zhao KR (2004) Study on the b-fibrinogen gene - 455A/G and - 148C/T polymorphism and cerebral infarction. Chin J Neurol 37:90–91. Liu R, Li J, Mu H, et al. (2002) The relationship of betafibrinogen gene polymorphisms and ischaemic cardiocerebral vascular disease. Zhonghua Xue Ye Xue Za Zhi 23:453–456. Liu Z, Zhao B (2008) The association of beta fibrinogen -148C > T gene and plasminogen activator 1–675 4G/5G gene polymorphism with cerebral infarction. Nerv Dis Ment Health 8:94–97. Lu SJ, Zhao B, Chen YS (2007) Study on the relation between beta fibrinogen - 148C > T gene polymorphism and cerebral infarction. Chin J Nerv Ment Dis 33:425–428. Lv B, Zhao B, Xing YQ, et al. (2003) The study on the polymorphism of beta fibrinogen gene - 148C > T in the patients with cerebral infarction. J Clin Neurol 16:202–205. Ma AJ, Pan XD, Zhang CS, et al. (2006) A linkage between beta-fibrinogen gene - 148C/T polymorphism and cerebral infarction. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 23: 202–204. Maresca G, Di Blasio A, Marchioli R, Di Minno G (1999) Measuring plasma fibrinogen to predict stroke and myocardial infarction: an update. Arterioscler Thromb Vasc Biol 19:1368–1377. Pan XD, Ma AJ, Zou ZQ, et al. (2005) Study of beta fibrinogen gene - 148C > T polymorphism and plasma fibrinogen levels in young patients with acute cerebral infarction. J Clin Neurol 18:40–44. Qian JJ, Liu CF, Yue KR (2004) The relationship between bfibrinogen gene - 455A/G and - 148C/T polymorphism and cerebral infarction with carotid atherosclerosis. Jiangsu Med J 30:377–379. Roy SN, Mukhopadhyay G, Redman CM (1990) Regulation of fibrinogen assembly. Transfection of Hep G2 cells with B
ZHANG ET AL.
beta cDNA specifically enhances synthesis of the three component chains of fibrinogen. J Biol Chem 265:6389– 6393. Song YQ, Zou HL, Song XJ, Wang W (2005) Fibrinogen gene polymorphism [beta-148C > T] in patients with cerebral infarction. Cerebrovasc Dis Foreign Med Sci 13:895–897. Wei JW, Heeley EL, Jan S, et al. (2010) Variations and determinants of hospital costs for acute stroke in China. PLoS One 5:e13041. WHO (2000) The World Health Report 2000. Health Systems: Improving Performance. Geneva: World Health Organization. Wilhelmsen L, Sva¨rdsudd K, Korsan-Bengtsen K, et al. (1984) Fibrinogen as a risk factor for stroke and myocardial infarction. N Engl J Med 311:501–505. Xu F, Ni PH, Ying YY, Hu YQ (2005) Fibrinogen B b gene -148 C > T polymorphism in stroke. J Diagn Concepts Pract 4:38–41. Xu X, Li J, Sheng W, Liu L (2008) Meta-analysis of genetic studies from journals published in China of ischemic stroke in the Han Chinese population. Cerebrovasc Dis: 26:48–62. Yuan XD, Wang SJ, Xu YR, et al. (2009) Relationship between multi-locus fibrinogen polymorphisms and fibrinogen concentration, molecular reactivity and cerebral infarction. Zhonghua Xue Ye Xue Za Zhi 30:582–587. Yuan XD, Wang SJ, Xu YR, et al. (2010) The relationship of muti-locus gene polymorphisms, functional expression of fibrinogen B b-Chain and the type of cerebral infarction. Zhong Guo Mian Yi Xue Za Zhi 26:356–358. Zhang XH, Xu G, Zhao XY, et al. (2003) Study of fibrinogen B b gene - 148 C > T gene polymorphism in patients with coronary heart disease and stroke. Chin J Emerg Med 12: 683–685. Zhao B, Lv B, Xing YQ, et al. (2003) The beta fibrinogen gene - 148C > T polymorphism and its association with plasma fibrinogen’s function and level in cerebral infarction. Chin J Neurol 36:450–454.
Address correspondence to: Jian-Hua Zhao, MS Department of Neurology First Affiliated Hospital of Xinxiang Medical College Xinxiang 453100 Henan People’s Republic of China E-mail: [email protected]
