Identification of predictors for cardiac resynchronization therapy (CRT) implantation

Background
Cardiac resynchronization therapy (CRT) is a well-established therapeutic strategy in patients that fulfill the criteria of (i) symptomatic heart failure (HF) with reduced ejection fraction (LV-EF≤ 35%) and (ii) QRS prolongation (≥130ms). While evidence suggests that timely CRT-implantation in eligible patients facilitates reverse remodeling and survival, identification of patients that may benefit from CRT implantation in the future but do not yet fulfill criteria remains challenging.

Aim
The aim of the current study was to identify variables that could predict CRT implantation in the future.

Methods and Results:
We retrospectively analysed all patients that underwent ICD implantation in heart failure at our department between January 2015 to January 2024. We then assessed indication for CRT upgrade at the time of battery change.

We included 239 patients, 34 of whom had an indication for CRT upgrade due to heart failure, as specified in the 2021 Guidelines on cardiac pacing and CRT. Interestingly, none of the baseline criteria including medication, ECG-, laboratory- and echocardiographic-parameters were able to identify patients prone to indicated CRT upgrade. In contrast, we identified LVEDD- and QRS progression over time as predictors of future CRT implantation. A direct comparison of both parameters utilizing a C-statistic approach, demonstrated that left ventricular end-diastolic diameter (LVEDD) progression had a higher accuracy in predicting the necessity for cardiac resynchronization therapy (LVEDD AUC 0.77 [95%CI 0.51 - 1]) compared to QRS progression (AUC 0.70 [95% CI 0.49-0.91]). Moreover, the risk of worsening ejection fraction (EF; defined as a decrease of ≥ 5%), deterioration in NYHA functional class (≥ 1 point), and the indication for cardiac resynchronization therapy (CRT) implantation increased by 11% (95% CI 2%-21%) per 1mm increase in LVEDD and was independent of heart failure medication in an adjusted model (i.e. adjusted for age, sex and HF medication; p = 0.018). Of note, patients at risk of LVEDD progression and subsequent worsening heart failure had a higher EF (29±9% vs. 34±8% p = 0.026), a smaller left ventricular end-diastolic diameter (65±7mm vs. 54±6mm, p <0.001) and less symptoms (NYHA 2.5±0.8 vs. 1.7±1.0, p = 0.002) at baseline. Therefore, these patients may be at risk as falsely being perceived as milder forms of heart failure with subsequent less intense heart failure treatment and follow up.

Conclusion:
We demonstrate that LVEDD progression over time predicts future necessity of CRT implantation in patients with ICD implantation. Patients with LVEDD progression over time had a higher LV-EF, smaller LVEDD and less HF symptoms. We advocate to closely monitor HF patients for signs of progressive LVEDD and assess indication of early CRT implantation.