Lack of clinical predictors of optimal V-V delay in patients with cardiac resynchronization devices

Introduction Cardiac resynchronization therapy (CRT) is a well-established therapy for patients with moderate-to-severe heart failure (HF), left ventricular dysfunction with an ejection fraction ≤ 35% and a QRS on the surface electrocardiogram of ≥130 msec. Device optimization is often performed, adjusting the timing of RV and LV stimulation to produce a pacing sequence that yields the best global cardiac performance. However, no standard guidelines exist for optimization and many invasive and non-invasive techniques have been employed with mixed results. The aim of the present study was to determine whether there are any clinical predictors of the optimal V-V settings in patients implanted with CRT devices. Methods and results We prospectively evaluated 47 consecutive patients with HF who were referred to our device optimization clinic. The mean patient age was 64.9 ± 12.7 years. Patients were in both sinus rhythm (83%) and atrial fibrillation. Prior to device implant, 51% of patients had left bundle branch block (LBBB), 17% had intra-ventricular conduction delay (IVCD) and 21% were RV paced. Sixty-two percent were male, the mean QRS duration was 152 ± 29 ms, mean LVEF 26 ± 8% and 60% had a non-ischemic cardiomyopathy. Overall, 82% of patients required sequential pacing with 69% requiring LV pre-excitation to produce the best global cardiac function as determined by aortic velocity time integrals (VTI). In our cohort, none of the clinical characteristics evaluated, including etiology of the cardiomyopathy, QRS duration, LVEF, pre-implant rhythm or AV delay were predictive of an optimal simultaneous or sequential V-V setting. Conclusions None of the clinical variables tested in our analysis predicted optimal RV-LV settings. Our results suggest that individual optimization and programming of V-V settings is necessary. The inability to predict optimal settings likely reflects the unique characteristics of each patient and supports the need for individualized programming of each device.