Enhancing transvenous lead extraction risk prediction: Integrating imaging biomarkers into machine learning models.

BACKGROUND: Machine learning (ML) models have been proposed to predict risk related to transvenous lead extraction (TLE). OBJECTIVE: The purpose of this study was to test whether integrating imaging data into an existing ML model increases its ability to predict major adverse events (MAEs; procedure-related major complications and procedure-related deaths) and lengthy procedures (≥100 minutes). METHODS: We hypothesized certain features-(1) lead angulation, (2) coil percentage inside the superior vena cava (SVC), and (3) number of overlapping leads in the SVC-detected from a pre-TLE plain anteroposterior chest radiograph (CXR) would improve prediction of MAE and long procedural times. A deep-learning convolutional neural network was developed to automatically detect these CXR features. RESULTS: A total of 1050 cases were included, with 24 MAEs (2.3%) . The neural network was able to detect (1) heart border with 100% accuracy; (2) coils with 98% accuracy; and (3) acute angle in the right ventricle and SVC with 91% and 70% accuracy, respectively. The following features significantly improved MAE prediction: (1) ≥50% coil within the SVC; (2) ≥2 overlapping leads in the SVC; and (3) acute lead angulation. Balanced accuracy (0.74-0.87), sensitivity (68%-83%), specificity (72%-91%), and area under the curve (AUC) (0.767-0.962) all improved with imaging biomarkers. Prediction of lengthy procedures also improved: balanced accuracy (0.76-0.86), sensitivity (75%-85%), specificity (63%-87%), and AUC (0.684-0.913). CONCLUSION: Risk prediction tools integrating imaging biomarkers significantly increases the ability of ML models to predict risk of MAE and long procedural time related to TLE.

Long-term survival following transvenous lead extraction: unpicking differences according to sex.

AIMS: Female sex is a recognized risk factor for procedure-related major complications including in-hospital mortality following transvenous lead extraction (TLE). Long-term outcomes following TLE stratified by sex are unclear. The purpose of this study was to evaluate factors influencing long-term survival in patients undergoing TLE according to sex. METHODS AND RESULTS: Clinical data from consecutive patients undergoing TLE in the reference centre between 2000 and 2019 were prospectively collected. The total cohort was divided into groups based on sex. We evaluated the association of demographic, clinical, device-related, and procedure-related factors on long-term mortality. A total of 1151 patients were included, with mean 66-month follow-up and mortality of 34.2% (n = 392). The majority of patients were male (n = 834, 72.4%) and 312 (37.4%) died. Males were more likely to die on follow-up [hazard ratio (HR) = 1.58 (1.23-2.02), P < 0.001]. Males had a higher mean age at explant (66.2 ± 13.9 vs. 61.3 ± 16.3 years, P < 0.001), greater mean co-morbidity burden (2.14 vs. 1.27, P < 0.001), and lower mean left ventricular ejection fraction (LVEF) (43.4 ± 14.0 vs. 50.8 ± 12.7, P = 0.001). For the female cohort, age > 75 years [HR = 3.45 (1.99-5.96), P < 0.001], estimated glomerular filtration rate < 60 [HR = 1.80 (1.03-3.11), P = 0.037], increasing co-morbidities (HR = 1.29 (1.06-1.56), P = 0.011), and LVEF per percentage increase [HR = 0.97 (0.95-0.99), P = 0.005] were all significant factors predicting mortality. The same factors influenced mortality in the male cohort; however, the HRs were lower. CONCLUSION: Female patients undergoing TLE have more favourable long-term outcomes than males with lower long-term mortality. Similar factors influenced mortality in both groups.

Left bundle branch area pacing reduces epicardial dispersion of repolarization compared with biventricular cardiac resynchronization therapy.

BACKGROUND: Biventricular endocardial pacing (BiV-endo) and left bundle branch area pacing (LBBAP) are novel methods of delivering cardiac resynchronization therapy. These techniques are associated with improved activation times and acute hemodynamic response compared with conventional biventricular epicardial pacing (BiV-epi); however, the effects on repolarization and arrhythmic risk are unknown. OBJECTIVE: The purpose of this study was to compare the effects of temporary BiV-epi, BiV-endo, and LBBAP on epicardial left ventricular (LV) repolarization using electrocardiographic imaging (ECGi). METHODS: Eleven patients indicated for cardiac resynchronization therapy underwent a temporary pacing protocol with ECGi. BiV-endo was delivered via endocardial stimulation of the LV lateral wall. LBBAP was delivered by pacing the LV septum. Epicardial LV repolarization time (LVRT-95; time taken for 95% of the LV to repolarize), LV RT dispersion, mean LV activation recovery interval (ARI), LV ARI dispersion, and RT gradients were calculated. RESULTS: The protocol was completed in 10 patients. During LBBAP, there were significant reductions in LVRT-95 (94.9 ± 17.4 ms vs 125.0 ± 29.4 ms; P = .03) and LV RT dispersion (29.4 ± 6.3 ms vs 40.8 ± 11.4 ms; P = .015) compared with BiV-epi. In contrast, there were no significant differences between baseline, BiV-epi, or BiV-endo. There was a nonsignificant reduction in mean RT gradients between LBBAP and baseline rhythm (0.74 ± 0.22 ms/mm vs 1.01 ± 0.31 ms/mm; P = .07). There were no significant differences in mean LV ARI or LV ARI dispersion between groups. CONCLUSION: Temporary LBBAP reduces epicardial dispersion of repolarization compared with conventional BiV-epi. Further study is required to determine whether these repolarization changes on ECGi translate into a reduced risk of ventricular arrhythmia in clinical practice.