Clinical and procedural characteristics of acute hemodynamic responders undergoing triple-site ventricular pacing for advanced heart failure.

The advantages of triple-site ventricular pacing (Tri-V) compared to conventional biventricular site pacing (Bi-V) have been reported. We sought to identify the predictors of acute hemodynamic Tri-V responders. Acute hemodynamic studies were performed in 32 patients with advanced heart failure during Tri-V implantation. After the right ventricular (RV) and left ventricular (LV) leads were implanted for a conventional Bi-V system, an additional pacing lead was implanted in the RV outflow tract for Tri-V. The LV peak +dP/dt and tau were measured during AAI, Bi-V, and Tri-V pacing. A Tri-V responder was defined as a patient whose percentage of increase in the peak +dP/dt during Tri-V was >10% compared to of that during Bi-V. The baseline clinical variables and RV outflow tract lead location were analyzed to identify the characteristics of the Tri-V responders. Of the 32 patients, 10 (31%) were classified as Tri-V responders. The LV end-diastolic volume was greater (246 ± 48 vs 173 ± 53 ml, p <0.01), and the RV outflow tract lead was implanted at a greater outflow tract portion (p <0.05) in the Tri-V responders. Multivariate analysis revealed that only the baseline LV end-diastolic volume (per 50-ml greater) predicted the Tri-V response (odds ratio 2.87, 95% confidence interval 1.03 to 8.00, p <0.05). The area under the receiver operating characteristic curve for the LV end-diastolic volume was 0.84 (p <0.01) and an LV end-diastolic volume of >212 ml had a sensitivity of 80% and specificity of 77% to distinguish Tri-V responders. In conclusion, Tri-V provides greater hemodynamic effect for patients with a larger LV end-diastolic volume owing to its resynchronization effects on the LV anterior wall.