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 < 0.001), 0.793 (95% CI: 0.718–0.867; P < 0.001), 0.785 (95% CI: 0.711–0.859; P < 0.001), 0.776 (95% CI: 0.703–0.849; P < 0.001), and 0.803 (95% CI: 0.733–0.872; P < 0.001), respectively. The discriminative ability of our model was superior to other evaluated scores. 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.
Introduction Our pilot study aimed to evaluate the role of local impedance drop in lesion formation during pulmonary vein isolation with a novel contact force sensing ablation catheter that records local impedance as well and to find a local impedance cut-off value that predicts successful lesion formation. Materials and methods After completing point-by-point radiofrequency pulmonary vein isolation, the success of the applications was evaluated by pacing along the ablation line at 10 mA, 2 ms pulse width. Lesions were considered successful if loss of local capture was achieved. Results Out of 645 applications, 561 were successful and 84 were unsuccessful. Compared to the unsuccessful ablation points, the successful applications were shorter (p = 0.0429) and had a larger local impedance drop (p<0.0001). There was no difference between successful and unsuccessful applications in terms of mean contact force (p = 0.8571), force-time integral (p = 0.0699) and contact force range (p = 0.0519). The optimal cut-point for the local impedance drop indicating successful lesion formation was 21.80 Ohms on the anterior wall [AUC = 0.80 (0.75–0.86), p<0.0001], and 18.30 Ohms on the posterior wall [AUC = 0.77 (0.72–0.83), p<0.0001]. A local impedance drop larger than 21.80 Ohms on the anterior wall and 18.30 Ohms on the posterior wall was associated with an increased probability of effective lesion creation [OR = 11.21, 95%CI 4.22–29.81, p<0.0001; and OR = 7.91, 95%CI 3.77–16.57, p<0.0001, respectively]. Conclusion The measurement of the local impedance may predict optimal lesion formation. A local impedance drop > 21.80 Ohms on the anterior wall and > 18.30 Ohms on the posterior wall significantly increases the probability of creating a successful lesion.
IntroductionHigh-power short-duration (HPSD) radiofrequency ablation has been proposed to produce rapid and effective lesions for pulmonary vein isolation (PVI). We aimed to evaluate the procedural characteristics and the first-pass isolation (FPI) rate of HPSD and very high-power short-duration (vHPSD) ablation compared to the low-power long-duration (LPLD) ablation technique.MethodsOne hundred fifty-six patients with atrial fibrillation (AF) were enrolled and assigned to LPLD, HPSD, or vHPSD PVI. The energy setting was 30, 50, and 90 W in the LPLD, HPSD, and vHPSD groups, respectively. In the vHPSD group, 90 W/4 s energy delivery was used in the QMODE+ setting. In the other groups, ablation index-guided applications were delivered with 30 W (LPLD) or 50 W (HPSD).ResultsBilateral PVI was achieved in all cases. Compared to the LPLD group, the HPSD and vHPSD groups had shorter procedure time [85 (75–101) min, 79 (65–91) min, and 70 (53–83) min], left atrial dwelling time [61 (55–70) min, 53 (41–56) min, and 45 (34–52) min], total RF time [1,567 (1,366–1,761) s, 1,398 (1,021–1,711) s, and 336 (247–386) s], but higher bilateral FPI rate (57, 78, and 80%) (all p-values < 0.01). The use of HPSD (OR = 2.72, 95% CI 1.15–6.44, p = 0.023) and vHPSD (OR = 2.90, 95% CI 1.24–6.44, p = 0.014) ablation techniques were associated with a higher probability of bilateral FPI. The 9-month AF-recurrence rate was lower in case of HPSD and vHPSD compared to LPLD ablation (10, 8, and 36%, p = 0.0001). Moreover, the presence of FPI was associated with a lower AF-recurrence rate at 9-month (OR = 0.09, 95% CI 0.04–0.24, p = 0.0001).ConclusionOur prospective, observational cohort study showed that both HPSD and vHPSD RF ablation shortens procedure and RF time and results in a higher rate of FPI compared to LPLD ablation. Moreover, the use of HPSD and vHPSD ablation increased the acute and mid-term success rate. No safety concerns were raised for HPSD or vHPSD ablation in our study.
AimsThe low lymphocyte counts and high neutrophil leucocyte fractions have been associated with poor prognosis in chronic heart failure. We hypothesized that the baseline ratio of the neutrophil leucocytes to the lymphocytes (NL ratio) would predict the outcome of chronic heart failure patients undergoing cardiac resynchronization therapy (CRT).Methods and resultsThe qualitative blood counts and the serum levels of N-terminal of the prohormone brain natriuretic peptide (NT-proBNP) of 122 chronic heart failure patients and 122 healthy controls were analysed prospectively in this observational study. The 2-year mortality was considered as primary endpoint and the 6-month reverse remodelling (≥15% decrease in the end-systolic volume) as secondary endpoint. Multivariable regression analyses were applied and net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were calculated. The NL ratio was elevated in chronic heart failure patients when compared with the healthy controls [2.93 (2.12–4.05) vs. 2.21 (1.64–2.81), P < 0.0001]. The baseline NL ratio exceeding 2.95 predicted the lack of the 6-month reverse remodelling [n = 63, odds ratio = 0.38 (0.17–0.85), P = 0.01; NRI = 0.49 (0.14–0.83), P = 0.005; IDI = 0.04 (0.00–0.07), P = 0.02] and the 2-year mortality [n = 29, hazard ratio = 2.44 (1.04–5.71), P = 0.03; NRI = 0.63 (0.24–1.01), P = 0.001; IDI = 0.04 (0.00–0.08), P = 0.02] independently of the NT-proBNP levels or other factors.ConclusionThe NL ratio is elevated in chronic heart failure and predicts outcome after CRT. According to the reclassification analysis, 4% of the patients would have been better categorized in the prediction models by combining the NT-proBNP with the NL ratio. Thus, a single blood count measurement could facilitate the optimal patient selection for the CRT.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.