Machine learning-derived major adverse event prediction of patients undergoing transvenous lead extraction: Using the ESC EHRA EORP European lead extraction ConTRolled ELECTRa registry.

BACKGROUND: Transvenous lead extraction (TLE) remains a high-risk procedure. OBJECTIVE: The purpose of this study was to develop a machine learning (ML)-based risk stratification system to predict the risk of major adverse events (MAEs) after TLE. A MAE was defined as procedure-related major complication and procedure-related death. METHODS: We designed and evaluated an ML-based risk stratification system trained using the European Lead Extraction ConTRolled (ELECTRa) registry to predict the risk of MAEs in 3555 patients undergoing TLE and tested this on an independent registry of 1171 patients. ML models were developed, including a self-normalizing neural network (SNN), stepwise logistic regression model ("stepwise model"), support vector machines, and random forest model. These were compared with the ELECTRa Registry Outcome Score (EROS) for MAEs. RESULTS: There were 53 MAEs (1.7%) in the training cohort and 24 (2.4%) in the test cohort. Thirty-two clinically important features were used to train the models. ML techniques were similar to EROS by balanced accuracy (stepwise model: 0.74 vs EROS: 0.70) and superior by area under the curve (support vector machines: 0.764 vs EROS: 0.677). The SNN provided a finite risk for MAE and accurately identified MAE in 14 of 169 "high (>80%) risk" patients (8.3%) and no MAEs in all 198 "low (<20%) risk" patients (100%). CONCLUSION: ML models incrementally improved risk prediction for identifying those at risk of MAEs. The SNN has the additional advantage of providing a personalized finite risk assessment for patients. This may aid patient decision making and allow better preoperative risk assessment and resource allocation.

Risk stratification of patients undergoing transvenous lead extraction with the ELECTRa Registry Outcome Score (EROS): an ESC EHRA EORP European lead extraction ConTRolled ELECTRa registry analysis.

AIMS: Transvenous lead extraction is associated with a significant risk of complications and identifying patients at highest risk pre-procedurally will enable interventions to be planned accordingly. We developed the ELECTRa Registry Outcome Score (EROS) and applied it to the ELECTRa registry to determine if it could appropriately risk-stratify patients. METHODS AND RESULTS: EROS was devised to risk-stratify patients into low risk (EROS 1), intermediate risk (EROS 2), and high risk (EROS 3). This was applied to the ESC EORP European Lead Extraction ConTRolled ELECTRa registry; 57.5% EROS 1, 31.8% EROS 2, and 10.7% EROS 3. Patients with EROS 3 or 2 were significantly more likely to require powered sheaths and a femoral approach to complete procedures. Patients with EROS 3 were more likely to suffer procedure-related major complications including deaths (5.1 vs. 1.3%; P < 0.0001), both intra-procedural (3.5 vs. 0.8%; P = 0.0001) and post-procedural (1.6 vs. 0.5%; P = 0.0192). They were more likely to suffer post-procedural deaths (0.8 vs. 0.2%; P 0.0449), cardiac avulsion or tear (3.8 vs. 0.5%; P < 0.0001), and cardiovascular lesions requiring pericardiocentesis, chest tube, or surgical repair (4.6 vs. 1.0%; P < 0.0001). EROS 3 was associated with procedure-related major complications including deaths [odds ratio (OR) 3.333, 95% confidence interval (CI) 1.879-5.914; P < 0.0001] and all-cause in-hospital major complications including deaths (OR 2.339, 95% CI 1.439-3.803; P = 0.0006). CONCLUSION: EROS successfully identified patients who were at increased risk of significant procedural complications that require urgent surgical intervention.

The effect of centre volume and procedure location on major complications and mortality from transvenous lead extraction: an ESC EHRA EORP European Lead Extraction ConTRolled ELECTRa registry subanalysis.

AIMS: Transvenous lead extraction (TLE) should ideally be undertaken by experienced operators in a setting that allows urgent surgical intervention. In this analysis of the ELECTRa registry, we sought to determine whether there was a significant difference in procedure complications and mortality depending on centre volume and extraction location. METHODS AND RESULTS: Analysis of the ESC EORP European Lead Extraction ConTRolled ELECTRa registry was conducted. Low-volume (LoV) centres were defined as <30 procedures/year, and high-volume (HiV) centres as ≥30 procedures/year. Three thousand, two hundred, and forty-nine patients underwent TLE by a primary operator cardiologist; 17.1% in LoV centres and 82.9% in HiV centres. Procedures performed by primary operator cardiologists in LoV centres were less likely to be successful (93.5% vs. 97.1%; P < 0.0001) and more likely to be complicated by procedure-related deaths (1.1% vs. 0.4%; P = 0.0417). Transvenous lead extraction undertaken by primary operator cardiologists in LoV centres were associated with increased procedure-related major complications including death (odds ratio 1.858, 95% confidence interval 1.007-3.427; P = 0.0475). Transvenous lead extraction locations varied; 52.0% operating room, 9.5% hybrid theatre and 38.5% catheterization laboratory. Rates of procedure-related major complications, including death occurring in a high-risk environment (combining operating room and hybrid theatre), were similar to those undertaken in the catheterization laboratory (1.7% vs. 1.6%; P = 0.9297). CONCLUSION: Primary operator cardiologists in LoV centres are more likely to have extractions complicated by procedure-related deaths. There was no significant difference in procedure complications between different extraction settings. These findings support the need for TLE to be performed in experienced centres with appropriate personnel present.

Clinical impact of antithrombotic therapy in transvenous lead extraction complications: a sub-analysis from the ESC-EORP EHRA ELECTRa (European Lead Extraction ConTRolled) Registry.

AIMS: A sub-analysis of the ESC-EHRA European Lead Extraction ConTRolled (ELECTRa) Registry to evaluate the clinical impact of antithrombotic (AT) on transvenous lead extraction (TLE) safety and efficacy. METHODS AND RESULTS: ELECTRa outcomes were compared between patients without AT therapy (No AT Group) and with different pre-operative AT regimens, including antiplatelets (AP), anticoagulants (AC), or both (AP + AC). Out of 3510 pts, 2398 (68%) were under AT pre-operatively. AT patients were older with more comorbidities (P < 0.0001). AT subgroups, defined as AP, AC, or AP + AC, were 1096 (31.2%), 985 (28%), and 317 (9%), respectively. Regarding AP patients, 1413 (40%) were under AP, 1292 (91%) with a single AP, interrupted in 26% about 3.8 ± 3.7 days before TLE. In total, 1302 (37%) patients were under AC, 881 vitamin K antagonist (68%), 221 (17%) direct oral anticoagulants, 155 (12%) low weight molecular heparin, and 45 (3.5%) unfractionated heparin. AC was 'interrupted without bridging' in 696 (54%) and 'interrupted with bridging' in 504 (39%) about 3.3 ± 2.3 days before TLE, and 'continued' in 87 (7%). TLE success rate was high in all subgroups. Only overall in-hospital death (1.4%), but not the procedure-related one, was higher in the AT subgroups (P = 0.0500). Age >65 years and New York Heart Association Class III/IV, but not AT regimens, were independent predictors of death for any cause. Haematomas were more frequent in AT subgroups, especially in AC 'continued' (P = 0.025), whereas pulmonary embolism in the No-AT (P < 0.01). CONCLUSIONS: AT minimization is safe in patients undergoing TLE. AT does not seem to predict death but identifies a subset of fragile patients with a worse in-hospital TLE outcome.