| Objective: The purpose of this study was to analyze the relationship between the concentration of P-selectin and the polymorphism of PS gene-2123C>G(rs1800807)and-1817T>C(rs1800808)in atrial fibrillation with thrombosis.Analyze the relationship between PS gene SNP and P-selectin concentration;Screen the related risk factors of atrial fibrillation thrombosis events,evaluate the diagnostic value of risk factors in diseases,and construct nomograms to predict the occurrence probability of diseases.Methods: Part one:Using the method of meta-analysis,we determined the key wrods,searched the major databases,drew the forest map with Review Manager 5.4 software,and the odds ratio(OR)and its 95% confidence interval(95% CI)were calculated by random effect model,to analyze the correlation between different genetic models of PS gene and the concentration of P-selectin in atrial fibrillation and atrial fibrillation with thrombosis.Part two: Collect clinical samples,there were 291 cases in thrombus group,534 cases in atrial fibrillation group and 981 cases in control group.The genotypes of P-selectin were identified by polymerase chain reaction-restriction fragment length polymorphism(PCR-RFLP),and the concentration of P-selectin was determined by enzyme-linked immunosorbent assay(ELISA).The distribution of genotypes and alleles of PS gene-2123C>G(rs1800807)and-1817 T > C(rs1800808)in each group were compared,and the concentration expression of different binding genotypes were compared.Part three: Screened the risk factors of atrial fibrillation and atrial fibrillation thrombus events by Logistic regression,ROC curve is used to judge the diagnostic value of risk factors for diseases,nomogram is constructed to predict the risk of atrial fibrillation and atrial fibrillation thrombus events,and DCA curve and clinical impact curve are used to evaluate the profit value of the model.Results: Part one:1)Meta-analysis included 25 studies.The results of forest map showed that: 1)In comparison between atrial fibrillation thrombus group and atrial fibrillation group,the concentration of P-selectin(SMD=1.57,95%CI: 0.46-2.69,P=0.006),allele genetic model of-2123 C > G locus(OR=1.41,95%CI: 1.10-1.81,P=0.007),homozygous genetic model(OR=2.20,95%CI: 1.03-4.70,P=0.04),-1817T>C locus recessive genes mode(OR=1.98,95%CI: 1.22-3.22,P=0.006)may be a risk factor for thrombosis in patients with atrial fibrillation.2)In comparison between atrial fibrillation group and control group,the genetic model of heterozygote at-1817T>C locus(OR=1.57,95%CI: 1.11-2.22,P=0.01)and the concentration of P-selectin(SMD=1.01,95%CI: 0.64-1.39,P<0.00001)may be a risk factor for atrial fibrillation thrombosis in healthy patients.Part two: 1)Through collecting clinical samples for research,the results show that in atrial fibrillation thrombus group,atrial fibrillation group and normal control group,the proportion of CC genotype was 18.2%,18.9% and 34.9%,the proportion of CG genotype was 48.1%,50.4% and 47.7%,the proportion of GG genotype was 33.7%,30.7% and 17.4% at-2123C>G locus;the proportion of CC genotype was 16.5%,8.4% and 18.2%,the proportion of CT genotype was 43.6%,42.0%and 48.9%,the proportion of TT genotype was 39.9%,49.6% and 32.8% at-1817T>C locus.2)Logistic regression showed that the OR of the risk of thrombosis is 2.147(95%CI: 1.390-3.316)when the-1817T>C locus is a recessive model in patients with atrial fibrillation;the risk of atrial fibrillation thrombosis when healthy people carry G allele and GG genotype of-2123 C > G locus is 1.943 times(95%CI:1.611-2.343)and 3.698 times(95%CI:2.526-5.415)of C allele and CC genotype,the OR of the risk of atrial fibrillation with thrombosis is 2.405(95% CI: 1.793-3.227)when-2123C>G locus is a recessive model;The OR of the risk of atrial fibrillation with thrombosis in healthy people is 1.357(95% CI: 1.036-1.777)when-1817T>C locus is a dominant model;3)The risk of thrombotic events of CC and GC haplotype carriers in patients with atrial fibrillation was1.378(95% CI: 1.047-1.814)and 1.373(95% CI: 1.059-1.781).The risk of atrial fibrillation thrombosis in healthy people with GT haplotype was 2.219(95% CI:1.817-2.710).4)The frequency of binding genotype CG/TT(-2123C>G/-1817T>C)in atrial fibrillation thrombus group was higher(21.6%).The frequency of combined genotype CG/CT(-2123C>G/-1817T>C)was higher(23.0%)in atrial fibrillation group.The frequency of CG/CT(-2123C>G/-1817T>C)in normal control group was higher(25.1%)..In the comparison of the three groups,the expression of P selectin concentration in different binding genotypes was different(P < 0.05).Part three: 1)Body mass index(BMI),fibrinogen(FIB),C-reactive protein(CRP),platelet(PLT),CC genotype of-1817 T > C locus are positively correlated with the risk of thrombosis in patients with atrial fibrillation.Smoking,Drinking,Age,DBP,BMI,CRP,PLT,the concentration of P-selectin,GG genotype at-2123C>G locus are positively correlated with the risk of atrial fibrillation in healthy people,.2)According to the AUC of ROC curve,the diagnostic value of the indexes included in atrial fibrillation thrombosis is: PLT(0.724),FIB(0.597),CRP(0.565),concentration of P-selectin(0.558),BMI(0.532),SBP(0.525).3)The prediction nomogram model of atrial fibrillation thrombus events and the prediction model of atrial fibrillation events nomogram have good fitting degree,good discrimination and accuracy.Conclusion: 1)P-selectin-2123C>G,-1817 T > C locus have gene polymorphism,which is related to atrial fibrillation thrombosis events..2)The expression of p-selectin concentration is related to the genotype of P-selectin,and the increased level of p-selectin may increase the risk of atrial fibrillation and atrial fibrillation thrombosis.3)Haplotypes of PS gene are CC,CT,GC and GT.GC and CC haplotype increases the risk of thrombosis in patients with atrial fibrillation,while GT haplotype may increase the risk of atrial fibrillation with thrombosis in healthy people.4)The nomogram model of atrial fibrillation thrombus events and atrial fibrillation events can be used as an auxiliary system of personalized treatment and applied in clinic. |