Non‐Invasive Methods for Identifying Electrical Responders to Conduction System Pacing Cardiac Resynchronization Therapy
Jack W. Samways, Nadine Ali, Ahran D. Arnold, Ji‐Jian Chow, Matthew Shun‐Shin, Jagdeep Singh Mohal, Akriti Naraen, Keenan Saleh, Nandita Kaza, Chetanya Sharma, Pavel Leinveber, Pavel Jurák, Daniel Keene, Yixiu Liang, Nicholas S. Peters, Prapa Kanagaratnam, James P. Howard, Norman Qureshi, Fu Siong Ng, Nick W. Linton, David C. Lefroy, Darrel P. Francis, P. Boon Lim, Mark A. Tanner, Amal Muthumala, Frits W. Prinzen, Judith M. Lubrecht, Karol Čurila, Graham Cole, Zachary WhinnettABSTRACT
Background
Conduction system pacing (CSP) is an emerging alternative modality of cardiac resynchronization therapy (CRT). However, not all patients obtain electrical resynchronization with CSP, there is a clinical need for reliable non‐invasive predictors of response. We aimed to assess the ability of several non‐invasive modalities to predict electrical resynchronization with CSP‐CRT.
Methods
Patients with standard heart failure CRT indications (excluding right‐bundle branch block) underwent CSP‐CRT using His‐bundle and left‐bundle branch pacing. Electrical resynchronization was defined as a >10 ms reduction in left ventricular activation time or conversion to a physiological left ventricular activation pattern on non‐invasive multi‐electrode mapping. We assessed whether 12‐lead ECG morphology, vector electrocardiogram (VCG) derived QRS area, Ultra High Frequency (UHF)‐ECG parameters or MRI scar burden predicted electrical resynchronization.
Results
Thirty four patients were analyzed (mean age 69±10 years; 82% male; LVEF 30±6%; QRS duration 161±23 ms; 32% ischemic cardiomyopathy). Electrical resynchronization was achieved in 24/34 (71%; 95%CI 53–85). Strauss criteria positivity on 12‐lead ECG yielded a positive predictive value (PPV) of 83% (95%CI 61–95%) and negative predictive value (NPV) of 55% (95%CI 23–83%). VCG QRS area >100µVs demonstrated PPV 100% (95% CI 81–100%) and NPV 73% (95%CI 85–100%). UHF‐ECG e‐DYS >50 ms gave PPV 93% (95%CI 66–100%) and NPV 47% (23%–72%); >60 ms gave PPV 90% (95%CI 66–100%) and NPV 38% (95%CI 18–62%). MRI scar burden <15% resulted in PPV 88% (95%CI 64–99%) and NPV 56% (95%CI 21–86%). Lines of propagation discontinuity on multi‐electrode mapping showed PPV 100% (95%CI 85–100%) and NPV 83% (95%CI 52–98%).
Conclusions
Multiple non‐invasive markers, including QRS morphology, VCG QRS area, UHF‐ECG e‐DYS, MRI scar burden showed potential to identify patients who are likely to electrically resynchronize with CSP‐CRT, although their NPVs were generally modest. Non‐invasive propagation mapping had the highest PPV and NPV.