Predictive Value Of LAF Combined With Serum ST2 On Recurrence Of Paroxysmal Atrial Fibrillation (Non-valvular)

Objective: To explore the value of LAF combined with serum ST2 as pedictors of paroxysmal atrial fibrillation(non-valvular)recurrence and provide more clinical evidence for stratified comprehensive management of patients with paroxysmal atrial fibrillation patientsMethods:In the period of December 1,2017 to October 1,2018,100 patients hospitalized in cardiovascular internal medicine ward of Affiliated Hospital of Hebei University were diagnosed as non-valvular symptomal initial paroxysmal atrial fibrillation by ECG and/or Holter(24-hour dynamic electrocardiogram),and valvular heart disease were excluded by echocardiography.Record the general information of the patients including gender,age,history of smoking and drinking,previous medical history(diabetes,hypertension,cerebrovascular disease,hyperthyroidism,heart valve disease,recent history of surgery and trauma,etc.).The results of laboratory examination such as serum ST2,cystatin C,blood routine,NT-proBNP,etc.and echocardiographic parameters such as LAD,E peak,LVEF,and lLAF were examined.Within the 6 months,all 100 patients were followed up and record recurrence of atrial fibrillation,which don't take any antiarrhythmic drugs.6 months later,serum LAF and ST2 of all 100 patients were examined and the relationship between LAF combined with serum ST2 and the occurrence of nonvalvular paroxysmal atrial fibrillation was analyzed,and further described to determine the value of LAF combined with serum ST2 as pedictors of non-valvular paroxysmal atrial fibrillation recurrence.Results:(1)The median LAF and interquartile interval of the patients with recurrent nvaf were 0.076 and 0.050(mean =0.726,95%CI(0.585,0.866),lower than that of the patients without recurrent nvaf(mean =0.169,95%CI(0.156,0.183)),and there were significant differences in the two groups(P<0.05).(2)K-m univariate analysis showed that the recurrence of para-atrial fibrillation was related to ST2 level in the LAF serum,LAD NIR BNP ejection fraction and serum cystatin C level,but was not related to gender and age,previous history of hypertension and diabetes.(3)After COX multivariate regression model,excluding confounders,LAF(Exp(B)=0.483,95%CI(0.316,0.71),P=0.001<0.05)and LA(Exp(B)=11.837,95%CI(2.127,65.869),P=0.005<0.05)were independent risk factors for recurrence of paroxysmal atrial fibrillation.(4)The area under the ROC curve of LAF inpredicting the recurrence of paroxysmal atrial fibrillation(non-valvular)was 0.904(95%CI0.846~0.963,P<0.05),the optimal boundary value was 0.727 N,the sensitivity was 88.6%,and the specificity was 84.1%.The area under the ROC curve of serum ST2 in predicting the recurrence of paroxysmal atrial fibrillation(non-valvular)was 0.835(95% CI 0.754~0.915,P<0.05),the optimal threshold was 209.66ng/ml,the sensitivity was 51.4%,and the specificity was 68.0%.The area under the ROC curve of LAF combined with serum ST2 in predicting paroxysmal atrial fibrillation(non-valvular)recurrence was 0.940(95%CI0.891-0.989,P<0.05),the sensitivity was 88.6%,and the specificity was 95.2%..Conclusion:(1)ST2 and LAF can be used as predictors of AF recurrence.(2)The decrease of LAF,the increase of serum ST2 level,the increase of LA,the increase of NIR,the increase of BNP,the decrease of ejection fraction and the increase of serum cystatin C level were the risk factors for recurrence of AF.(3)The optimal threshold of LAF for predicting recurrence of paroxysmal atrial fibrillation(non-valvular)was 0.727 N,with a sensitivity of88.6% and a specificity of 84.1%.The optimal threshold of serum ST2 for predicting recurrence of paroxysmal atrial fibrillation(non-valvular)was 209.66ng/ml,with a sensitivity of 51.4% and a specificity of 68.0%.The combination of the two can improve the predictive value of af recurrence,,with a sensitivity of 88.6% and a specificity of 95.2%.