OBJECTIVE:To assess the role of cytotoxin-associated gene-A (CagA) positive strains of Helicobacterpylori (Hp)in ischemic stroke (IS)subtypes.DATA SOURCES:A computer-based online search of PubMed,EMBASE,the Cochrane Collaboration database,the CNKI database and the VIP database,from January 1997 to July 2010,was performed to find relevant studies.DATA SELECTION:Case-control studies relevant to CagA with IS and IS subtypes were selected.Data regarding related factors in the case group and control group were acquired using the same approach.All patients had been diagnosed as exhibiting IS using skull CT or MRI,and were etiologically typed according to the 1993 TOAST diagnosis criteria.Two investigators independently performed the same search and study selection.Meta-analyses were then performed for the selected studies using RevMan 5.0 software (Cochrane Collaboration) after strict screening.Heterogeneity tests,sensitivity analyses and publication bias assessments were then conducted.MAIN OUTCOME MEASURES:Relationship of CagA with IS and IS subtypes.RESULTS:Eight studies were selected,involving data from 879 patients with IS,and 849 healthy controls.Five out of eight of the selected studies were related to large artery atherosclerosis (461patients with IS and 497 health controls).The results of our meta-analysis revealed a significant association between prior infection with CagA-positive strains and increased risk of IS (odds ratio (OR) = 2.31,95% confidence interval (CI):1.89 2.82,P＜0.01).In addition,we found an association between infection with CagA-negative strains and IS (OR = 0.57,95%C/:0.47 0.70,P＜0.01).CagA positive and negative strains were found to correlate with large artery atherosclerosis (CagA-positive strains:OR = 2.87,95%CI:2.19-3.77,P ＜ 0.01;CagA-negative strains:OR = 0.51,95%CI:0.39-0.67,P＜ 0.01).Because of the diversity of etiological factors in the case-control study,we conducted further analyses after correcting for confounding factors,and the overall effects were recalculated.The results revealed significant relationships between CagA-positive strains and IS (OR = 2.36,95%CI:1.84 3.02,P ＜ 0.01),and between CagA-positive strains and large artery atherosclerosis (OR = 3.10,95%CI:2.29-4.19,P ＜ 0.01 ).A heterogeneity test of CagA-positive strains in IS and its subtypes revealed good homogeneity ((P) = 0%;(P)2 = 0%) and we adopted a fixed-effects model to calculate OR.Sensitivity analysis confirmed that the results of the meta-analysis were reliable.However,the funnel plot suggested that the experimental results may be affected by bias,possibly resulting from a lack of published studies reporting negative outcomes in the meta-analysis.CONCLUSION:Infection with CagA-positive strains is a risk factor for IS,especially the large artery atherosclerosis subtype.However,the evidence from case-control studies is weak,and more prospective studies are required to conclusively determine whether infection by CagA-positive strains should be considered a novel risk factor for IS and its subtypes.