Non-progressors to cardiac resynchronization therapy show long-term mortality benefit compared to progressors

Introduction Cardiac resynchronization therapy (CRT) reduces morbidity and mortality in selected patients with symptomatic heart failure with reduced ejection fraction (HFrEF) and prolonged QRS. However, approximately one-third of the implanted patients fail to show clinical improvement or reverse remodeling. The criteria by response have been recently changed, those patients with a minimal improvement in left ventricular ejection fraction (LVEF) are defined as "non-progressors" rather than as "non-responders", selecting from those patients in whom the progression could not be modified. Data are scarce regarding the long-term outcome of this patient population. Purpose We aimed to evaluate the long-term outcome of CRT patients by their response, comparing non-progressors with responders and progressors. Methods Altogether 1019 patients undergoing CRT implantation between 2000-2020 in our center were registered and analyzed retrospectively. Patients were divided into 4 groups according to their response status, which was defined on the basis of LVEF change in 12 months after CRT implantation as follows: super-responders ≥20% (n=113), responders 6-19% (n=448), non-progressors 0-5% (n=244), and progressors <0% (n=214). The primary endpoint was the composite of all-cause mortality, heart transplantation, or left ventricular assist device implantation. Results During the median follow-up time of 4.7 years 547 (54%) patients died, 35 (31%) super-responder, 223 (50%) responder, 133 (55%) non-progressor, and 156 (73%) progressor. The mean change in LVEF occurred as follows: super-responders 24.5% ± 4.1%, responders 11.5% ± 3.8%, non-progressors 2.8% ± 1.8%, and progressors -6.6% ± 4.5% (p<0.0001). Non-progressors were more commonly male, had worse NYHA functional status, worse renal function, higher baseline LVEF and serum creatinine levels, and more frequently had an ischemic etiology. Univariate Cox regression analysis revealed that non-progressors had similar outcome to responders (HR 1.17; 95%CI 0.94-1.45; p=0.15) and superior outcome to progressors (HR 0.60; 95%CI 0.48-0.76; p<0.0001), which was also confirmed by multivariate analysis after adjustment for age, gender, ischemic etiology, LVEF, and serum creatinine levels: nonprogressors vs. responders (HR 1.25; 95%CI 0.98-1.58; p=0.07) and nonprogressors vs. progressors (HR 0.62; 95%CI 0.47-0.80; p<0.0001). Conclusions In our current study, non-progressors to CRT had a similar long-term outcome to responders and superior outcome to progressors. These findings suggest that non-progressor patients would have continued to adversely remodel without CRT and CRT is beneficial for these patients by moderating the remodeling process rather than improving it. Our results support the fact that categorizing patients as "responders" or "non-responders" is inappropriate and patients could be grouped as super-responders, responders, non-progressors, and progressors.