Prediction Of Clinical Outcomes In Patients With Atrial Fibrillation After Catheter Ablation Based On Acoustic Cardiography

Background:Atrial fibrillation is a common persistent arrhythmia with a high rate of disability,mortality,and recurrence.The number of patients with atrial fibrillation in China is huge,reaching tens of millions.With the acceleration of the aging process of the population in China,it is expected that the number of patients with atrial fibrillation will increase significantly in the future.The treatment of atrial fibrillation has undergone significant changes in the past few decades,and catheter ablation is gradually becoming the first line treatment for atrial fibrillation.The problem of recurrence after catheter ablation has received considerable clinical attention.The recurrence rate after single catheter ablation for paroxysmal atrial fibrillation is 20～40%,and the recurrence rate for patients with persistent atrial fibrillation is 30～50%.Previous studies have confirmed that atrial fibrillation type,left atrial size,BNP,and LVEF are risk factors for recurrence of atrial fibrillation,but their prediction of recurrence after catheter ablation of atrial fibrillation is not effective.The“ABC”comprehensive treatment management scheme proposed in the 2020 ESC Guidelines for the Management of Atrial Fibrillation can significantly reduce the risk of all cause death/stroke/massive hemorrhage/cardiovascular death,and reduce the incidence of cardiovascular events and health related costs.m Health is an important component of the comprehensive management system for atrial fibrillation.Currently,wearable single lead ECG monitors such as bracelets and watches,which are widely used,have played a positive role in atrial fibrillation screening.However,electrocardiographic monitoring is often insufficient to reflect cardiac function.Under the recommendation of the"ABC"management system,technologies that can be easily used and more comprehensively reflect changes in cardiac function will have more significant clinical significance for the comprehensive management of atrial fibrillation and cardiovascular comorbidities.Portable acoustic cardiography（ACG）is a new technology that reflects the electromechanical synchronization and hemodynamics of the heart by synchronously recording cardiac auscultation signals and electrocardiograms.Compared to a single electrical signal,it can more comprehensively reflect changes in cardiac function.The electromechanical parameters detected by ACG can be used as auxiliary and alternative monitoring indicators for echocardiography in clinical practice.The core parameter,Electromechanical Activation Time（EMAT）,is defined as the time from the beginning of the QRS wave of the electrocardiogram to the first peak value of the phonocardiogram S1,reflecting the process from electrical activation to mechanical movement of the myocardium.EMAT%is defined as the proportion of EMAT to the entire cardiac cycle.Previous studies have shown that EMAT and EMAT%have a good correlation with LVEF and BNP,and can be used to evaluate cardiac function.Some parameters S3 and SDI have a correlation with postoperative recurrence of atrial fibrillation.Therefore,it is relatively simple,repeatable,and non-invasive ACG detection for the diagnosis of atrial fibrillation,cardiac function evaluation,and prediction of short-term prognosis after catheter ablation of atrial fibrillation has strong feasibility and innovation.With its portability and accuracy,ACG is expected to become an important tool for follow-up and evaluation of cardiac function in patients with atrial fibrillation.This study aims to explore the diagnostic value of ACG for atrial fibrillation and the accuracy of cardiac function assessment,and to explore the relationship between EMAT,EMAT%and clinical outcomes after catheter ablation of atrial fibrillation.Part I:Preliminary clinical application of portable ACG in diagnosis of atrial fibrillation and evaluation of cardiac functionObjectives:To explore the clinical value of portable ACG in the diagnosis of atrial fibrillation and evaluation of cardiac function,and to discuss its sensitivity,specificity,and consistency.Materials and Methods:From May 2021 to September 2022,193 inpatients with cardiac arrhythmias were enrolled in the Department of Cardiology,the First Affiliated Hospital of the Naval Military Medical University.All patients received standard 12-lead electrocardiogram,echocardiography,and BNP and ACG.Portable ACG was placed at the apex of the heart or in the pulmonary valve auscultation area to synchronously record heart sounds and ECG signals.The default monitoring duration for a single time was 1 minute,and each site was repeatedly measured three times.The automatic recognition algorithm obtained the average of relevant parameters.The clinical value of ACG in the diagnosis of atrial fibrillation was reflected through the authenticity,reliability,and benefit indicators of diagnostic tests.The correlation between EMAT,EMAT%and LVEF,BNP was analyzed through one-way linear regression,and the ROC curve was plotted to reflect the value of ACG in evaluating cardiac function.Results:1.193 patients were enrolled,including 122 male patients（63.21%）and 71 female patients（36.79%）.The enrolled age was between 19 and 81 years old,with a mean age of59.07±13.66 years.Clinical diagnosis of the selected cases:116 cases of atrial fibrillation（60.10%）,2 cases of atrial flutter（1.03%）,1 case of atrial premature beats（0.52%）,46cases of paroxysmal supraventricular tachycardia（23.83%）,18 cases of ventricular premature beats（9.33%）,8 cases of preexcitation syndrome（4.15%）,and 2 cases of ventricular tachycardia（1.03%）.Among them,73 patients had BNP>100 pg·ml^-1（37.82%）.2.Atrial fibrillation in ACG is characterized by irregular intervals between P waves and R peaks of the ECG signal.Automatic algorithms identify R peaks and obtain absolutely uneven heart rate line graphs corresponding to each cardiac cycle.Sinus rhythm in ACG is characterized by a regular pattern of P wave and R peak intervals,with each RR interval corresponding to two heart sound peaks,the first heart sound and the second heart sound.The heart rate curve corresponding to each cardiac cycle is neat.3.The sensitivity and specificity of the portable ACG recorder for diagnosis of atrial fibrillation were 91.23%,90.44%,80.00%,96.06%,8.77%,9.56%,and 90.67%,respectively.The area under the ROC curve（AUC）for ACG diagnosis of atrial fibrillation was 0.908,and the standard error was 0.035.Its 95%confidence interval was（0.857-0.959）,and the Jordan index was 0.816.4.Single factor linear regression analysis showed that EMAT,EMAT%,and LVEF had a linear regression relationship（β=-0.182,P=0.001;β=-1.291,P=0.001）,EMAT,EMAT%has linear regression with BNP（β=-1.283,P=0.034;β=16.543,P=0.001）.Further comparing the ROC curves of EMAT,EMAT%,and BNP in the diagnosis of left ventricular systolic dysfunction（LVEF<50%）,the results showed that EMAT%[0.948（95%CI 0.889～1）]had a slightly better effect than EMAT[0.888（95%CI 0.759～1）]and BNP[0.875（95%CI 0.782～0.968）]Conclusions:1.Atrial fibrillation in ACG is characterized by irregular P wave and R peak interval,and the heart rate curve corresponding to each ECG cycle obtained by automatic algorithm recognition of R peak is absolutely uneven.EMAT and EMAT%in atrial fibrillation group were significantly higher than those in non-atrial fibrillation group,which proved that the cardio-mechanical function of patients with atrial fibrillation was damaged.2.Portable ACG recorder can accurately,quickly,and conveniently diagnose atrial fibrillation with high sensitivity and specificity.3.EMAT and EMAT%can be used to evaluate cardiac function,with EMAT%having a stronger correlation with LVEF and BNP.Part II:Prediction of clinical outcomes in patients with atrial fibrillation after catheter ablation based on acoustic cardiographyObjectives:To explore the correlation between EMAT,EMAT%,and clinical outcomes after catheter ablation for atrial fibrillation.Materials and Methods:From May 2021 to September 2022,110 patients who underwent catheter ablation of atrial fibrillation in the Department of Cardiology of the First Affiliated Hospital of Naval Military Medical University were enrolled as the study subjects,and baseline data of patients were collected,such as age,gender,height,weight,BMI,Cr,AST,BNP,hypertension history,diabetes history,coronary heart disease history,echocardiography,etc;Acquire ACG images of patients,and use AI automatic analysis software to record EMAT,EMAT%,HR,and other indicators.Follow up for at least 6 months after catheter ablation.Follow up with echocardiography,BNP,and 24-hour dynamic electrocardiogram to determine whether there is recurrence.Single factor and multiple factor linear regression analysis was used to determine whether EMAT and EMAT%are independent predictors of cardiac function after catheter ablation of atrial fibrillation;Single factor and multiple factor logistic regression were used to analyze whether EMAT and EMAT%were predictive factors for recurrence after catheter ablation of atrial fibrillation,and their predictive value was evaluated by plotting corresponding ROC curves.Results:1.There were 61 patients（55.45%）in the paroxysmal atrial fibrillation group,49patients（44.56%）in the persistent atrial fibrillation group.BNP in the persistent atrial fibrillation group[194.01（128.35,307.93）(pg·ml^-1)vs 67.17（33.09,133.36）(pg·ml^-1)],left atrial size[（74.59±23.92）ml vs（55.58±19.85）ml)],surgical time[（150.16±57.86）min vs（121.44±54.33）min]EMAT%[（11.61±2.94）%vs（9.91±2.30）%]was significantly higher in the paroxysmal atrial fibrillation group（P<0.05）,there was no statistically significant difference in other indicators between the two groups.2.The preoperative BNP was significantly higher than that at 6 months after surgery[113.75（49.81,208.85）(pg·ml^﹣1)vs 75.70（36.00,141.56）(pg·ml^-1),P=0.046].There was no statistically significant difference between preoperative LVEF and postoperative LVEF at 6 months[（60.60±7.51）%vs（61.57±7.45）%,P=0.376].3.Left atrial volume,EMAT,EMAT%,and HR are independent predictors of BNP after catheter ablation of atrial fibrillation（β=3.998,95%CI 1.784～6.212,P=0.001;β=-2.583,95%CI-39.502～-19.664,P=0.001;β=280.301,95%CI 190.453～370.149,P=0.001;β=-38.196,95%CI-52.214～-24.178,P=0.001）4.There were 96 patients（87.27%）in the non recurrence group and 14 patients（12.73%）in the recurrence group.The proportion of persistent atrial fibrillation in the non recurrent group（40.62%vs 71.43%,P=0.030）,diabetes（14.6%vs 42.9%,P=0.028）,left atrial size[（60.36±19.89）ml vs（87.64±32.38）ml,P=0.001],EMAT[（86.58±13.12）ms vs（100.50±28.22）ms,P=0.003],EMAT%[（10.37±2.18）%vs（12.70±4.70）%,There was a statistically significant difference compared to the recurrence group（P=0.002）.5.Left atrial size（OR=1.031,95%CI 1.004～1.059,P=0.031）and EMAT（OR=1.045,95%CI 1.005～1.087,P=0.026）were independent risk factors for recurrence after catheter ablation of atrial fibrillation（P<0.05）.6.The AUC predicted by diabetes,LDL-C,left atrial size and EMAT for recurrence after catheter ablation of atrial fibrillation were 0.650（95%CI 0.478～0.821）,0.692（95%CI0.542～0.842）,0.770（95%CI 0.625～0.915）,0.640（95%CI 0.625～0.915）,and the optimal thresholds were 0.50,2.73mmol/L,77.5ml,87.5ms,respectively.The AUC predicted by the combination of the four indicators for recurrence after catheter ablation of atrial fibrillation was 0.851.Conclusions:1.The EMAT%in patients with persistent atrial fibrillation is longer than that in patients with paroxysmal atrial fibrillation.The EMAT and EMAT%in patients with recurrent atrial fibrillation after catheter ablation are higher than those without recurrence.2.EMAT and EMAT%are independent predictors of BNP after catheter ablation of atrial fibrillation.3.EMAT is an independent risk factor for recurrence of atrial fibrillation after catheter ablation,and has certain predictive value for recurrence of atrial fibrillation.Its combination with diabetes history,LDL-C,and left atrial size can better predict recurrence of atrial fibrillation after catheter ablation.