Validation of a Machine Learning Algorithm to Identify Pulmonary Vein Isolation during Ablation Procedures for the Treatment of Atrial Fibrillation: Results of the PVISION Study

BACKGROUND AND AIMS

Pulmonary Vein Isolation (PVI) is the cornerstone of ablation for atrial fibrillation. Confirmation of PVI can be challenging due to far field electrograms and sometimes requires additional pacing maneuvers or mapping. This prospective multicenter study assessed the agreement between a previously trained automated algorithm designed to determine vein isolation status with expert opinion in real-world clinical setting.

METHOD

Consecutive patients scheduled for PVI were recruited at four centers. The ECGenius electrophysiology recording system (CathVision ApS, Denmark) was connected in parallel with the lab's existing system. Electrograms from a circular mapping catheter were annotated during sinus rhythm at baseline pre-ablation, time of isolation, and post-ablation. The ground truth for isolation status was based on operator opinion. The algorithm was applied to the collected PV signals off-line and compared to expert opinion. The primary endpoint was a sensitivity and specificity exceeding 80%.

RESULTS

Overall, 498 electrograms (248 at baseline and 250 at PVI) with 5,473 individual PV beats from 89 patients (32 females, 62 ±12 years) were analyzed. The algorithm performance reached an area under curve (AUC) of 92% and met the primary study endpoint with a sensitivity and specificity of 86% and 87%, respectively (p = 0.005; p = 0.004). The algorithm had an accuracy of 87% in classifying the time of isolation.

CONCLUSION

This study validated an automated algorithm using machine learning (ML) to assess the isolation status of pulmonary veins in patients undergoing PVI with different ablation modalities. The algorithm reached an AUC of 92% with both sensitivity and specificity exceeding the primary study endpoints.

Determinants of myocardial injury following transcatheter aortic valve implantation: a pre-specified substudy from the POPular TAVI trial

Background

Myocardial injury is frequently observed in patients undergoing transcatheter aortic valve implantation (TAVI) and has been linked to worse prognosis [1,2]. Yet, knowledge concerning the underlying mechanisms and preventive strategies is scarce.

Purpose

To identify clinical determinants and the effect of periprocedural antithrombotic strategies on markers of myocardial injury after TAVI.

Methods

The POPular TAVI trial was a prospective, open label, multicentre randomized controlled trial, investigating the addition of clopidogrel to aspirin (cohort A) or oral anticoagulation (OAC) (cohort B) in patients undergoing TAVI [3] Patients randomised to clopidogrel received a 300mg loading dose before TAVI, followed by a 75mg maintenance dose once daily. In patients using OAC, this was continued during TAVI with an international normalized ratio aimed at 2.0. All OAC patients used a vitamin-K antagonist. Blood samples were taken at baseline, 6, 24, 48, and 72 hours following TAVI to determine myocardial injury using Creatine Kinase-MB (CK-MB) and high-sensitive cardiac troponin T (hs-cTnT) according to the VARC-2 criteria. Also, baseline and procedural variables were collected in detail. A linear mixed effects model was used for pair-wise analysis of the changes in enzyme levels at different time points between groups. Regression analysis was performed using the logistic regression model. Statistical analyses were performed using R (version 3.4.1).

Results

In total, 131 patients undergoing transfemoral TAVI were included at two study sites, of whom 63 (48%) received clopidogrel and 68 (52%) did not. Almost half of the patients (45%) were on OAC. The rise in CK-MB (mean peak 23.4±13.3 U/l) and hs-cTnT (mean peak of 0.23±0.33 ug/) was maximal at 6 and 24 hours, respectively. The CK-MB and hs-cTnT levels did not differ between the clopidogrel and no clopidogrel group at any time point (figure 1). Myocardial injury occurred in 18 (30.1%) patients receiving OAC versus 39 (54.2%) patients not receiving OAC (p=0.007). The course of hs-cTnT reached higher levels in patients with chronic kidney disease (p<0.001) and in patients with a preserved left ventricular ejection fraction (LVEF) (p=0.008). Also, the use of a controlled mechanical expanding prosthesis was associated with a higher rise of hs-cTnT (p=0.007). (Figure 2) In multivariable analysis, predictors of a maximal increase in hs-cTnT were a preserved LVEF (OR 1.15, 95% CI 1.02–1.30) and chronic kidney disease (OR 1.13, 95% CI 1.01–1.28). Other procedural factors, like balloon dilation and rapid ventricular pacing, were not associated with myocardial injury.

Conclusions

The addition of clopidogrel to aspirin or OAC during TAVI was not associated with a reduction in myocardial injury. Instead, OAC therapy, as compared to aspirin, was associated with a reduction in rise and fall of hs-cTnT. Also, patients with a preserved LVEF or chronic kidney disease observed higher levels of hs-cTnT.

Funding Acknowledgement

Type of funding sources: Public grant(s) – National budget only. Main funding source(s): ZonMWSt. Antonius Research Fund

Transcatheter aortic valve implantation amid the COVID-19 pandemic: a nationwide analysis of the first COVID-19 wave in the Netherlands

Introduction

The coronavirus disease 2019 (COVID-19) pandemic has put tremendous pressure on healthcare systems. Most transcatheter aortic valve implantation (TAVI) centres have adopted different triage systems and procedural strategies to serve highest-risk patients first and to minimise the burden on hospital logistics and personnel. We therefore assessed the impact of the COVID-19 pandemic on patient selection, type of anaesthesia and outcomes after TAVI.

Methods

We used data from the Netherlands Heart Registration to examine all patients who underwent TAVI between March 2020 and July 2020 (COVID cohort), and between March 2019 and July 2019 (pre-COVID cohort). We compared patient characteristics, procedural characteristics and clinical outcomes.

Results

We examined 2131 patients who underwent TAVI (1020 patients in COVID cohort, 1111 patients in pre-COVID cohort). EuroSCORE II was comparable between cohorts (COVID 4.5 ± 4.0 vs pre-COVID 4.6 ± 4.2, p = 0.356). The number of TAVI procedures under general anaesthesia was lower in the COVID cohort (35.2% vs 46.5%, p < 0.001). Incidences of stroke (COVID 2.7% vs pre-COVID 1.7%, p = 0.134), major vascular complications (2.3% vs 3.4%, p = 0.170) and permanent pacemaker implantation (10.0% vs 9.4%, p = 0.634) did not differ between cohorts. Thirty-day and 150-day mortality were comparable (2.8% vs 2.2%, p = 0.359 and 5.2% vs 5.2%, p = 0.993, respectively).

Conclusions

During the COVID-19 pandemic, patient characteristics and outcomes after TAVI were not different than before the pandemic. This highlights the fact that TAVI procedures can be safely performed during the COVID-19 pandemic, without an increased risk of complications or mortality.

Supplementary Information

The online version of this article (10.1007/s12471-022-01704-9) contains supplementary material, which is available to authorized users.

Major threats to early safety after transcatheter aortic valve implantation in a contemporary cohort of real-world patients

INTRODUCTION

Despite considerable advances in the last decade, major adverse events remain a concern after transcatheter aortic valve implantation (TAVI). The aim of this study was to provide a detailed overview of their underlying causes and contributing factors in order to identify key domains for quality improvement.

METHODS

This observational, prospective registry included all patients undergoing TAVI between 31 December 2015 and 1 January 2020 at the St. Antonius Hospital in Nieuwegein and the University Medical Centre in Utrecht. Outcomes of interest were all-cause mortality, stroke, major bleeding, life-threatening or disabling bleeding, major vascular complications, myocardial infarction, severe acute kidney injury and conduction disturbances requiring permanent pacemaker implantation within 30 days after TAVI, according to the Valve Academic Research Consortium‑2 criteria.

RESULTS

Of the 1250 patients who underwent TAVI in the evaluated period, 146 (11.7%) developed a major complication. In 54 (4.3%) patients a thromboembolic event occurred, leading to stroke in 36 (2.9%), myocardial infarction in 13 (1.0%) and lower limb ischaemia in 11 (0.9%). Major bleeding occurred in 65 (5.2%) patients, most frequently consisting of acute cardiac tamponade (n = 25; 2.0%) and major access-site bleeding (n = 21; 1.7%). Most complications occurred within 1 day of the procedure. Within 30 days a total of 54 (4.3%) patients died, the cause being directly TAVI-related in 30 (2.4%). Of the patients who died from causes that were not directly TAVI-related, 14 (1.1%) had multiple hospital-acquired complications.

CONCLUSION

A variety of underlying mechanisms and causes form a wide spectrum of major threats affecting early safety in 11.7% of patients undergoing TAVI in a contemporary cohort of real-world patients.

Transcatheter paravalvular leak closure is an effective alternative to surgical repair with respect to 5-year outcomes

Introduction

Paravalvular leakage (PVL) is a relatively common complication of heart valve replacement, associated with heart failure and increased mortality. Transcatheter PVL closure may be a promising alternative to surgical repair, especially in high-risk patients.

Purpose

Assessment of safety and efficacy of transcatheter PVL closure compared to surgical repair.

Methods

This is a retrospective single-centre study including all consecutive patients who underwent either transcatheter PVL closure between January 2013 and December 2020, or surgical repair between March 2015 and December 2020. Primary endpoints were 5-year all-cause mortality and the composite of 5-year cardiovascular mortality and rehospitalization for the underlying condition. Secondary endpoints were technical success and individual patient success at one year according to the PVL Academic Research Consortium.

Results

Of the 129 patients included, 85 went for transcatheter repair and 44 went for surgical repair. As compared to surgical repair, patients undergoing transcatheter PVL closure were older (71 years vs. 64,5 years; p≤0,01) and more symptomatic (NYHA class III & IV; 76,5% vs. 59,1%; p=0,04). At 5 years, transcatheter PVL closure was comparable to surgery in terms of the primary composite endpoint (HR: 1,20; 95% CI: 0,68–2,13; p=0,54), all-cause mortality (HR: 1,70; 95% CI: 0,82–3,50; p=0,15) and rehospitalization for the underlying condition (HR: 1,12; 95% CI: 0,54–2,89; p=0,780). Rates of technical success (92,9% vs. 95,5%; p=0,58) and individual patient success at one year (70,6% vs. 77,3%; p=0,87) were similar between transcatheter PVL closure and surgery respectively. Transcatheter PVL closure was associated with shorter in-hospital stay (7 days vs. 14 days; p≤0,01).

Conclusion

These findings support the use of transcatheter closure of PVL, especially in high-risk patients. Long term survival remains temperate in these challenging patients.

Funding Acknowledgement

Type of funding sources: None.

Transcatheter tricuspid valve repair: early experience in the Netherlands

Background

Symptomatic tricuspid regurgitation (TR) is increasingly prevalent and impairs quality of life and survival, despite medical treatment. Transcatheter tricuspid valve repair (TTVR) has recently become available as a treatment option for patients not eligible for tricuspid valve surgery. In this study we describe the early experience with TTVR in the Netherlands.

Methods

All consecutive patients scheduled for TTVR in two tertiary hospitals were included in the current study. Patients were symptomatic and had severe functional TR. TTVR was performed either with the MitraClip (off-label use) or dedicated TriClip delivery system and device. Procedural success was defined as achievement of clip implantation, TR reduction ≥ 1 grade and no need for re-do surgical or transcatheter intervention. Clinical improvement was evaluated after 4 weeks.

Results

Twenty-one patients (median age 78 years, 33% male, 95% New York Heart Association class ≥ 3, 100% history of atrial fibrillation) underwent TTVR. Procedural success was achieved in 16 patients, of whom 15 reported symptomatic improvement (New York Heart Association class 1 or 2). There was no in-hospital mortality and no major complications occurred. Baseline glomerular filtration rate and TR coaptation gap size were associated with procedural success.

Conclusion

The current study showed that TTVR seems a promising treatment option for patients with severe functional TR deemed high risk for surgery. Successful TR reduction is most likely in patients with limited coaptation gap size and strongly determines clinical benefit. Adequate patient selection and timing of treatment seem essential for an optimal patient outcome.

Supplementary Information

The online version of this article (10.1007/s12471-021-01613-3) contains supplementary material, which is available to authorized users.

Update and, internal and temporal-validation of the FRANCE-2 and ACC-TAVI early-mortality prediction models for Transcatheter Aortic Valve Implantation (TAVI) using data from the Netherlands heart registration (NHR)

Background

The predictive performance of the models FRANCE-2 and ACC-TAVI for early-mortality after Transcatheter Aortic Valve Implantation (TAVI) can decline over time and can be enhanced by updating them on new populations. We aim to update and internally and temporally validate these models using a recent TAVI-cohort from the Netherlands Heart Registration (NHR).

Methods

We used data of TAVI-patients treated in 2013-2017. For each original-model, the best update-method (model-intercept, model-recalibration, or model-revision) was selected by a closed-testing procedure. We internally validated both updated models with 1000 bootstrap samples. We also updated the models on the 2013-2016 dataset and temporally validated them on the 2017-dataset. Performance measures were the Area-Under ROC-curve (AU-ROC), Brier-score, and calibration graphs.

Results

We included 6177 TAVI-patients, with 4.5% observed early-mortality. The selected update-method for FRANCE-2 was model-intercept-update. Internal validation showed an AU-ROC of 0.63 (95%CI 0.62-0.66) and Brier-score of 0.04 (0.04-0.05). Calibration graphs show that it overestimates early-mortality. In temporal-validation, the AU-ROC was 0.61 (0.53-0.67).The selected update-method for ACC-TAVI was model-revision. In internal-validation, the AU-ROC was 0.63 (0.63-0.66) and Brier-score was 0.04 (0.04-0.05). The updated ACC-TAVI calibrates well up to a probability of 20%, and subsequently underestimates early-mortality. In temporal-validation the AU-ROC was 0.65 (0.58-0.72).

Conclusion

Internal-validation of the updated models FRANCE-2 and ACC-TAVI with data from the NHR demonstrated improved performance, which was better than in external-validation studies and comparable to the original studies. In temporal-validation, ACC-TAVI outperformed FRANCE-2 because it suffered less from changes over time.

The Coronary Sinus Reducer; 5-year Dutch experience

Background

Refractory angina is a growing and major health-care problem affecting millions of patients with coronary artery disease worldwide. The Coronary Sinus Reducer (CSR) is a device that may be considered for the relief of symptoms of refractory angina. It causes increased venous pressure leading to a dilatation of arterioles and reduced arterial vascular resistance in the sub-endocardium. This study describes the 5‑year Dutch experience regarding safety and efficacy of the CSR.

Methods

One hundred and thirty-two patients with refractory angina were treated with the CSR. The primary efficacy endpoint of the study was Canadian Cardiovascular Society (CCS) class improvement between baseline and 6‑month follow-up. The primary safety endpoint was successful CSR implantation in the absence of any device-related events.

Results

Eighty-five patients (67%) showed improvement of at least 1 CCS class and 43 patients (34%) of at least 2 classes. Mean CCS class improved from 3.17 ± 0.61 to 2.12 ± 1.07 after implantation (P < 0.001). The CSR was successfully implanted in 99% of the patients and only minor complications during implantation were reported.

Conclusion

The CSR is a simple, safe, and effective option for most patients with refractory angina. However, approximately thirty percent of the patients showed no benefit after implantation. Future studies should focus on the exact underlying mechanisms of action and reasons for non-response to better identify patients that could benefit most from this therapy.

Electronic supplementary material

The online version of this article (10.1007/s12471-020-01525-8) contains supplementary material, which is available to authorized users.

Direct comparison of predictive performance of PRECISE-DAPT versus PARIS versus CREDO-Kyoto: a subanalysis of the ReCre8 trial

Background

Multiple scores have been proposed to guide risk stratification after percutaneous coronary intervention. This study assessed the performance of the PRECISE-DAPT, PARIS and CREDO-Kyoto risk scores to predict post-discharge ischaemic or bleeding events.

Methods

A total of 1491 patients treated with latest-generation drug-eluting stent implantation were evaluated. Risk scores for post-discharge ischaemic or bleeding events were calculated and directly compared. Prognostic performance of both risk scores was assessed with calibration, Harrell’s c‑statistics net reclassification index and decision curve analyses.

Results

Post-discharge ischaemic events occurred in 56 patients (3.8%) and post-discharge bleeding events in 34 patients (2.3%) within the first year after the invasive procedure. C‑statistics for the PARIS ischaemic risk score was marginal (0.59, 95% confidence interval (CI) 0.51–0.68), whereas the CREDO-Kyoto ischaemic risk score was moderate (0.68, 95% CI 0.60–0.75). With regard to post-discharge bleeding events, CREDO-Kyoto displayed moderate discrimination (c-statistic 0.67, 95% CI 0.56–0.77), whereas PRECISE-DAPT (0.59, 95% CI 0.48–0.69) and PARIS (0.55, 95% CI 0.44–0.65) had a marginal discriminative capacity. Net reclassification index and decision curve analysis favoured CREDO-Kyoto-derived bleeding risk assessment.

Conclusion

In this contemporary all-comer population, PARIS and PRECISE-DAPT risk scores were not resilient to independent testing for post-discharge bleeding events. CREDO-Kyoto-derived risk stratification was associated with a moderate predictive capability for post-discharge ischaemic or bleeding events. Future studies are warranted to improve risk stratification with more focus on robustness and rigorous testing.

Electronic supplementary material

The online version of this article (10.1007/s12471-020-01486-y) contains supplementary material, which is available to authorized users.

Incidence, angiographic and clinical predictors, and impact of stent thrombosis: a 6-year survey of 6,545 consecutive patients

Objective

We sought to determine the incidence, angiographic predictors, and impact of stent thrombosis (ST).

Background

Given the high mortality after ST, this study emphasises the importance of ongoing efforts to identify angiographic predictors of ST.

Methods

All consecutive patients with angiographically confirmed ST between 2010 and 2016 were 1:4 matched for (1) percutaneous coronary intervention (PCI) indication and (2) index date ±6 weeks to randomly selected controls. Index PCI angiograms were reassessed by two independent cardiologists. A multivariable conditional logistic regression model was built to identify independent predictors of ST.

Results

Of 6,545 consecutive patients undergoing PCI, 55 patients [0.84%, 95% confidence interval (CI) 0.63–1.10%] presented with definite ST. Multivariable logistic regression identified dual antiplatelet therapy (DAPT) non-use as the strongest predictor of ST (odds ratio (OR) 10.9, 95% CI 2.47–48.5, p < 0.001), followed by: stent underexpansion (OR 5.70, 95% CI 2.39–13.6, p < 0.001), lesion complexity B2/C (OR 4.32, 95% CI 1.43–13.1, p = 0.010), uncovered edge dissection (OR 4.16, 95% CI 1.47–11.8, p = 0.007), diabetes mellitus (OR 3.23, 95% CI 1.25–8.36, p = 0.016), and residual coronary artery disease at the stent edge (OR 3.02, 95% CI 1.02–8.92, p = 0.045). ST was associated with increased rates of mortality as analysed by Kaplan-Meier estimates (27.3 vs 11.3%, p_log-rank < 0.001) and adjusted Cox proportional-hazard regression (hazard ratio 2.29, 95% CI 1.03–5.10, p = 0.042).

Conclusions

ST remains a serious complication following PCI with a high rate of mortality. DAPT non-use was associated with the highest risk of ST, followed by various angiographic parameters and high lesion complexity.

Electronic supplementary material

The online version of this article (10.1007/s12471-019-1253-2) contains supplementary material, which is available to authorized users.

[Non-incarcerated inguinal hernia in children: operation within 7 days not necessary]

OBJECTIVE

To assess the necessity to operate on non-incarcerated inguinal hernia in children within 7 days of diagnosis.

DESIGN

Retrospective.

METHOD

Data on 360 children, 0-10 years old (104 girls and 256 boys) who were operated on for inguinal hernia between 1 January 1993-31 December 2001 at the St. Elisabeth Hospital in Tilburg, the Netherlands, were collected from the medical records. These data included sex, age, interval between diagnosis and repair, recurrence, incarceration, length of hospitalisation and complications.

RESULTS

In the group of 113 children 0-1 years old, 137 inguinal hernias were repaired, ofwhich 16 were incarcerated on presentation. The interval between diagnosis and repair was known in 93 of 121 cases: 37 hernias were repaired within 7 days and 56 at a later stage. In the latter group, there was one case of secondary incarceration (1.8%; 95% CI: 0-5.4). The number needed to treat was 56. In the group of 247 children 1-10 years old, 269 inguinal hernias were repaired, of which 8 were primarily incarcerated. The interval between diagnosis and repair was known in 208 of 261 cases: 34 hernias were repaired within 7 days and 174 at a later stage. In the latter group, 3 hernias incarcerated secondarily (1.7%; 95% CI: 0-3.7). The number needed to treat was 58. In the group of non-incarcerated hernias 1 complication occurred, in the group of incarcerated hernias none. The mean length of hospitalisation of children with non-incarcerated hernia was 0.85 days, and of children with incarcerated hernia 2.4 days.

CONCLUSION

In children with a non-incarcerated inguinal hernia who are waiting for an operation, the risk of secondary incarceration and complications is 2% which we do not think is enough reason to carry out an elective hernia-repair procedure within 7 days.

[Grief work an support experienced during abortion and stillbirth]

We investigated by means of a semi-structured, oral interview how women cope with a miscarriage or stillbirth. The need for psychosocial support and the support which was experienced were also studied. The results stress that the mourning process after a miscarriage or stillbirth proceeds in one sense like the 'conventional' mourning process, but differs in some aspects. Women with one or more children show less intense emotions than childless women. The experiences with the professional help and the environmental support differ depending on the moment: before, during or after the bereavement. Negative experiences usually refer to a too businesslike attitude of doctors and nurses in charge, lack of information, shortcomings in the aftercare, clumsy reactions of the environment and unfulfilled needs for contact with other women who had a miscarriage or stillbirth.

The generality of mental addition models: simple and complex addition in a decision-production task

Recently, Widaman, Geary, Cormier, and Little (in press) offered a general componential model for simple and complex addition. In the present study, the generalizability of this model to very complex addition problems and to production task performance was tested. We investigated addition-production performance using a new research paradigm, the decision-production task. The paradigm was used to assess the reaction times of 16 adults to 100 single-digit addition problems (Experiment 1) and to very complex three-digit addition problems (Experiment 2). Generally, the results give evidence for the basic assumptions of the Widaman et al. model. Some of the results, however, demonstrate that short-term memory load factors and strategy choices also have important effects on complex addition-production performance. Furthermore, the results of both experiments indicate that, in production tasks, the nature of the elementary search/compute process can be characterized as a memory-retrieval process and that the decision-production paradigm is a valuable alternative to the classical verbal production paradigm.