Search

Aims Our aim was to develop a machine learning (ML)-based risk stratification system to predict 1-, 2-, 3-, 4-, and 5-year all-cause mortality from pre-implant parameters of patients undergoing cardiac resynchronization therapy (CRT). Methods and results Multiple ML models were trained on a retrospective database of 1510 patients undergoing CRT implantation to predict 1- to 5-year all-cause mortality. Thirty-three pre-implant clinical features were selected to train the models. The best performing model [SEMMELWEIS-CRT score (perSonalizEd assessMent of estiMatEd risk of mortaLity With machinE learnIng in patientS undergoing CRT implantation)], along with pre-existing scores (Seattle Heart Failure Model, VALID-CRT, EAARN, ScREEN, and CRT-score), was tested on an independent cohort of 158 patients. There were 805 (53%) deaths in the training cohort and 80 (51%) deaths in the test cohort during the 5-year follow-up period. Among the trained classifiers, random forest demonstrated the best performance. For the prediction of 1-, 2-, 3-, 4-, and 5-year mortality, the areas under the receiver operating characteristic curves of the SEMMELWEIS-CRT score were 0.768 (95% CI: 0.674–0.861; P Conclusion The SEMMELWEIS-CRT score (available at semmelweiscrtscore.com) exhibited good discriminative capabilities for the prediction of all-cause death in CRT patients and outperformed the already existing risk scores. By capturing the non-linear association of predictors, the utilization of ML approaches may facilitate optimal candidate selection and prognostication of patients undergoing CRT implantation.

Background Cardiac Resynchronization Therapy (CRT) has well-known beneficial effects in patients with advanced heart failure, reduced ejection fraction and wide QRS complex. However, mortality rates still remain high in this patient population. Therefore, precise risk stratification would be essential, nonetheless, the currently available risk scores have several shortcomings which hamper their utilization in the everyday clinical practice. Purpose Accordingly, our objective was to design and validate a machine learning based risk stratification system to predict 2-year and 5-year mortality from pre-implant parameters of patients undergoing CRT implantation. Methods We trained two models separately to predict 2-year (model 1) and 5-year mortality (model 2). As training cohort of model 1 we used 1678 patients (67±10 years, 1251 [75%] males) undergoing CRT implantation. From this population, 1320 patients (66±10 years, 1005 [76%] males) also completed 5-year follow-up and they served as the training cohort for model 2. Forty-seven pre-implant parameters (demographics, cardiovascular risk factors and clinical characteristics) were used to train the models. Our models were designed in a way to tolerate missing values. Among non-linear classifiers, random forest demonstrated the best performance. We validated our models, along with the Seattle Heart Failure Model (SHFM), VALID-CRT risk score and EAARN score on an independent cohort of 136 patients (66±10 years, 110 [81%] males). Based on the predicted probability of survival, patients were split into quartiles and survival was plotted via Kaplan-Meier (KM) curves. Results There were 358 (21%) deaths in the 2-year, 697 (53%) deaths in the 5-year training cohort. In the validation cohort, there were 30 (22%) deaths at 2 years and 58 (43%) deaths at 5 years after CRT implantation. For the prediction of 2-year mortality, the Area Under the Receiver-Operating Characteristic Curve (AUC) for model 1 was 0.77 (95% CI: 0.67–0.87; p=0.002), for SHFM was 0.54 (95% CI: 0.39–0.69; p=0.006), for EAARN was 0.57 (95% CI: 0.46–0.68, p=0.002), and for VALID-CRT was 0.62 (95% CI: 0.52–0.71; p=0.002). To predict 5-year mortality, the AUC for model 2 was 0.85 (95% CI: 0.78–0.91; p=0.001), for SHFM was 0.62 (95% CI: 0.51–0.74; p=0.003), for EAARN was 0.61 (95% CI: 0.51–0.70, p=0.002), for VALID-CRT was 0.65 (95% CI: 0.56–0.74; p=0.002). The AUCs of the machine learning based models were significantly higher than the AUCs of the pre-existing scores (DeLong test, all p Conclusion Our results indicate that machine learning algorithms can outperform the already existing linear model based scores. By capturing the non-linear association of predictors, the utilization of these state-of-the-art approaches may facilitate optimal candidate selection and prognostication of patients undergoing CRT implantation.

Purpose: Previous studies on open irrigation radiofrequency (RF) ablation catheters focused on created lesion volumes, and did not evaluate lesion quality. The aim of this study was to evaluate the effect of different irrigation levels on lesion quality and volume. Methods: A canine thigh model was used. A skin incision was made and a cradle was created and filled with blood (37 °C, 250 ml/min). The ablation catheter was positioned parallel or perpendicular to the muscle surface using a constant force of 10 g. RF lesions were created using a gold tip electrode catheter with 12 irrigation holes and a PtIr catheter with 6 irrigation holes (Biotronik, Berlin, Germany). Both catheters were 7F and had a 3.5 mm electrode tip. RF current was delivered for 60 s at either 50 W (n=18) or 30 W (n=39). Temperature probes were inserted immediately below the surface and at 3.5 and 7 mm depth. Histological examination was performed to evaluate lesion size as well as the homogeneity of the RF lesions. Inhomogeneity was defined as a visual observed multi-band lesion pattern indicating different histological characteristics. Electrode temperature and lesion dimensions were measured. Results: Data of 57 lesions, created in a mongrel dog model, were analyzed. A total number of 47 inhomogeneous and 10 homogeneous lesions were detected. Inhomogeneous lesions were deeper (7.5±1.4 vs. 6.3±2.1 mm, p=0.036), however lesion volume was comparable between the two groups (652±487 vs. 834±396, p=0.249). Lesions with inhomogeneous histological characteristics had lower maximum electrode temperatures during ablation (42.9±6.6 vs. 48.2±8.6 °C, p=0.049). The mean temperatures at 3,5 and 7 mm below the muscle surface were lower for inhomogeneous lesions (41.2±8.1 vs. 50.2±21.4 °C, p=0.035; 38.7±4.0 vs. 45.2±19.0 °C, p=0.034). Inhomogeneity of the lesion was not associated with catheter type (p=0.164), orientation (p=0.185), amount of saline infusion (p=0.305) and power output (p=0.213). Conclusions: Our data suggest that lower catheter tip and tissue temperatures are associated with the development of inhomogeneous acute RF lesion formation. The presence of inhomogeneous RF lesions can theoretically be responsible for the differences between acute and chronic ablation success rates.

Background: Prior studies suggested that right to left ventricular interlead sensed electrical delay (RL-IED) is predictive of echocardiographic response to cardiac resynchronization therapy (CRT). There is limited data on the effect of RL-IED on all-cause mortality in patients receiving a CRT device. Methods: We evaluated patients with RL-IED data available measured during the implantation procedure. Data on all-cause mortality was assessed using follow-up data and the Hungarian National Healthcare Fund Death Registry. Kaplan-Meier survival analyses and multivariate Cox proportional hazards regression models were used to analyze mortality in patients stratified by the median RL-IED (106.5 ms). Results: From 2000 to 2011, 494 patients (44%) of 1122 CRT implantation had RL-IED measurements available. Patients with RL-IED >106.5 ms were younger, had wider QRS complexes, less often atrial fibrillation and more dilated left ventricles. Left ventricular ejection fraction was similar in both groups (RL-IED ≤ 106.5 ms 28.1±7.5% vs. RL-IED > 106.5 ms 27.7±6.8%, p>0.5). During the median follow-up of 28 months, 145 (29%) patients died, 80 patients (16%) with RL-IED > 106.5 ms, and 65 patients (13%) with RL-IED ≤ 106.5 ms. RL-IED > 106.5 ms was associated with a significant, 48% risk reduction in all-cause mortality (HR= 0.52, 95% CI: 0.31-0.88, p= 0.01) after adjustment for clinical covariates (Fig. 1). ![Figure][1] Figure 1 Conclusions: In CRT patients, a longer right to left ventricular interlead electrical delay at the time of device implantation is associated with a significant mortality benefit as compared to those with shorter interlead delay. Right to left ventricular interlead electrical delay is an independent predictor of all-cause mortality in CRT patients. [1]: pending:yes