Long-term comparative outcomes of hemodynamic sensor-guided CRT versus conventional CRT: cardiac and functional performance insights

Abstract

In cardiac resynchronization therapy (CRT), integration of a hemodynamic sensor into the device allows automatic individualized optimization of atrioventricular (AV) and interventricular (VV) timing based on ventricular contractility with the purpose of improving patient outcomes by dynamic adjustment to cardiac function. We aimed to compare the effectiveness of CRT using a sensor-based system (SonR, Microport) to standard CRT in terms of functional and cardiac remodeling outcomes.

Methods: Consecutive patients (P) with heart failure, symptomatic New York Heart Association (NYHA) class II-IV, with Left Ventricular Ejection Fraction (LVEF) ≤35% after 3 months of Guideline-directed medical therapy and a prolonged QRS submitted to CRT implantation at our center between 2015 and 2022 were included. P with an existing pacemaker or Implantable Cardioverter Defibrillator (ICD) who develop a clinical indication for CRT were also included. A paired-sample T-test analysis was performed to evaluate pre- and post-therapy metrics in two groups: SonR P and non-SonR P. Primary endpoints included changes in NYHA class, LVEF, and left ventricular end-systolic volumes (LVESV). The effect sizes were analysed using Cohen’s and Hedge’s correction.

Results: Out of 161 P (62±11.3 years, 117 [72.7%] male), 48 (29.8%) were implanted with a SonR CRT system. The mean follow-up time was 55±29 months. For the SonR group, a significant improvement in LVEF was observed, with an increase of 11.34 percentage points from baseline to one-year post-intervention (mean difference = 11.34, 95% CI [7.2, 15.5], p < 0.001). In the non SonR group, there was a mean difference of 12.25 (95% CI [9.8, 14.7], p < 0.001). Both groups showed a statistically significative difference in LVESV (mean difference = 33.8, p =0.029 in the SonR group and mean difference = 36.8, p < 0.001 in the non-SonR group). The functional NYHA class had a greater improvement in the SonR group (mean difference = 0.8, 95% CI [0.6, 1.0], p < 0.001) than in the non SonR group (mean difference = 0.6, 95% CI [0.5, 0.7], p < 0.001). The SonR group had a greater effect size (Cohen’s d = 1.211, 95% CI [0.792, 1.622] vs Cohen’s d = 0.868, 95% CI [0.627, 1.105] in the non-SonR group), indicating a more pronounced impact on functional status.

Conclusion: The sensor-based CRT was associated with similar improvement in LVEF and LVESV compared to the standard CRT. However, the SonR group exhibited a greater improvement in functional class, reflected by a larger effect size. These findings suggest that automatic individualized optimization of AV and VV intervals based on ventricular contractility may add benefit in enhancing patients' functional status.