Use The 18-lead Surface Electrocardiograms To Compare The Value Of Six Algorithms That Differentiating The Origin Of Left And Right Outflow Tract Ventricular Premature Contractions

Objective:The purpose of this study is to systematically compare the value of six algorithms that differentiating the origin of left and right outflow tract ventricular premature contractions（OT-VPCs）,in order to provide reference value of the methodological choice of surface electrocardiograms localization algorithms before these OT-VPCs patients receive radiofrequency ablation surgery for the clinicians in the future.Methods:This study retrospectively analyzed the data of the OT-VPCs patients who successfully received radiofrequency ablation surgery in cardiovascular internal medicine department in Ningbo First Hospital from July 20,2015 to March 20,2017.The 18-lead surface electrocardiograms were mainly collected and arranged.Firstly,the gold standard was the site of successful ablation during surgery,according to the reported optimal cut-off points of these algorithms,we respectively located the origin of ventricular premature contractions（PVCs）in 105patients.Meanwhile,with the above obtained origin compared with the gold standard target one by one,the kappa values of the six algorithms were obtained by the Kappa test.;Next,in the applicable patients of the six algorithms respectively,the ROC（Receptive operating characteristic,ROC）curve was used to find the these algorithms’actual optimal cut-off points that differentiating the origin of left and right OT-VPCs,and we also initially compared the maximal area under the ROC curve of the six algorithms at the actual optimal cut-off point.;Finally,we selected 83patients to whom the six algorithms were all applied among the 105 patients so that we could further enhance the comparability of the statistical indicators of the six algorithms.When we found the six algorithms’actual optimal cut-off points in the 83 patients at the same way,we compared kappa value,the sensitivity,specificity,Yuden index,the area under the ROC curve,and the difference of the maximal area under the ROC curve of the six algorithms.In the end,combining the obtained statistical data and simplicity in practical work,we could analyze and compare the diagnostic efficiency of the six algorithms.Results:There were a total of 105 patients who were enrolled in this study,with an average age of 50.66±15.37 years,38（36%）males,and 73（70%）patients with VPCs in the right ventricular outflow tract.Firstly,when we evaluated the consistency between the judgment results of 6algorithms and the results of the gold standard,kappa value of the R wave duration and amplitude index,transitional zone index,V₂ transition ratio,V₂ S/V₃ R index,V₄/V₈ index and V_3R/V₇ index respectively was 0.569,0.641,0.110,0.691,0.275,0.159;Next,in the applicable patients of the six algorithms respectively,the maximal area under the ROC curve of the R wave duration and amplitude index,transitional zone index,V₂ transition ratio,V₂ S/V₃ R index,V₄/V₈ index and V_3R/V₇ index was 0.940、0.862、0.807、0.952、0.878、0.729 at the actual optimal cut-off point;Finally,in the 83 patients to whom the six algorithms were all applied,when we used actual optimal cut-off point（>1.25）of the V₂ S/V₃ R index to locate the right-sided locations,the sensitivity,specificity,Yuden index,and the area under the ROC curve of the V₂ S/V₃ R index respectively was 82%、95%、0.775、0.936.As for another five algorithms as a whole,the V₂S/V₃R index had the higher value of sensitivity,specific,Yuden index,and the area under the ROC curve.In the process of pairwise comparison of the maximal area under the ROC curve among six algorithms,the difference of the area under the ROC curve between V₂ S/V₃ R index and the R wave duration and amplitude index had no statistical significance while the difference between V₂ S/V₃ R index and the other four algorithms respectively had statistical significance.According to the comparison of the above statistical indicators and simplicity of six algorithms in practical work,we could infer that the V₂S/V₃R index might be the best way to differentiate the origin of left and right OT-VPCs.Conclusions:V₂ S/V₃ R index may be the simplest and most effective body surface electrocardiograms standard for differentiating the origin of left and right OT-VPCs,but it needs further study and verification in the future multi-center big data.