Critical appraisal of artificial intelligence-based prediction models for cardiovascular disease.

The medical field has seen a rapid increase in the development of artificial intelligence (AI)-based prediction models. With the introduction of such AI-based prediction model tools and software in cardiovascular patient care, the cardiovascular researcher and healthcare professional are challenged to understand the opportunities as well as the limitations of the AI-based predictions. In this article, we present 12 critical questions for cardiovascular health professionals to ask when confronted with an AI-based prediction model. We aim to support medical professionals to distinguish the AI-based prediction models that can add value to patient care from the AI that does not.

Insights in a restricted temporary pacemaker strategy in a lean transcatheter aortic valve implantation program.

OBJECTIVES: To study the safety and feasibility of a restrictive temporary-RV-pacemaker use and to evaluate the need for temporary pacemaker insertion for failed left ventricular (LV) pacing ability (no ventricular capture) or occurrence of high-degree AV-blocks mandating continuous pacing. BACKGROUND: Ventricular pacing remains an essential part of contemporary transcatheter aortic valve implantation (TAVI). A temporary-right-ventricle (RV)-pacemaker lead is the standard approach for transient pacing during TAVI but requires central venous access. METHODS: An observational registry including 672 patients who underwent TAVI between June 2018 and December 2020. Patients received pacing on the wire when necessary, unless there was a high-anticipated risk for conduction disturbances post-TAVI, based on the baseline-ECG. The follow-up period was 30 days. RESULTS: A temporary-RV-pacemaker lead (RVP-cohort) was inserted in 45 patients, pacing on the wire (LVP-cohort) in 488 patients, and no pacing (NoP-cohort) in 139 patients. A bailout temporary pacemaker was implanted in 14 patients (10.1%) in the NoP-cohort and in 24 patients (4.9%) in the LVP-cohort. One patient in the LVP-cohort needed an RV-pacemaker for incomplete ventricular capture. Procedure time was significantly longer in the RVP-cohort (68 min [IQR 52-88.] vs. 55 min [IQR 44-72] in NoP-cohort and 55 min [IQR 43-71] in the LVP-cohort [p < 0.005]). Procedural high-degree AV-block occurred most often in the RVP-cohort (45% vs. 14% in the LVP and 16% in the NoP-cohort [p ≤ 0.001]). Need for new PPI occurred in 47% in the RVP-cohort, versus 20% in the NoP-cohort and 11% in the LVP-cohort (p ≤ 0.001). CONCLUSION: A restricted RV-pacemaker strategy is safe and shortens procedure time. The majority of TAVI-procedures do not require a temporary-RV-pacemaker.

Prophylactic permanent pacemaker strategy in patients with right bundle branch block undergoing transcatheter aortic valve replacement.

OBJECTIVES: To report on the experience with a selective prophylactic permanent pacemaker (PPx-PPM) implantation strategy in patients with pre-existing right bundle branch block (RBBB) undergoing transcatheter aortic valve replacement (TAVR). BACKGROUND: Pre-existing RBBB is an independent predictor for PPM after TAVR and has been linked to increased mortality. METHODS: Hospital patient flow and longer-term clinical endpoints were compared for TAVR patients with pre-existing RBBB treated in a period with and without selective PPx-PPM strategy (2013-2020). RESULTS: A total of 260 patients were included: 170 in the early period without PPx-PPM strategy and 90 patients in the late period with selective PPx-PPM strategy. A PPx-PPM was implanted in 44% of patients in the late period. Overall, 69% versus 80% of all patients in the early versus late period ended up with a PPM (p = .06). Streamlined transfemoral TAVR was routinely used from 2017-in this series of patients, both TAVR procedural time and hospital length of stay (LoS) were significantly shorter in the late versus early period (mean procedural time: 70 vs. 83 min and LoS ≥5 days: 15% vs. 40%; p < .05). No difference in all-cause/cardiovascular mortality was observed between both strategies, whereas cardiac rehospitalization was significantly higher for patients treated in the early versus late period (hazard ratio 2.33 [1.04-5.26]; p = .042)-this mainly due to (sub)acute PPM-implantation early after discharge. CONCLUSIONS: Selective prophylactic PPM implantation in TAVR candidates with pre-existing RBBB results in shorter TAVR procedural time and hospital LoS and prevents early cardiac rehospitalization related to complete heart block.

Vascular complications with a plug-based vascular closure device after transcatheter aortic valve replacement: Predictors and bail-outs.

BACKGROUND: The MANTA vascular closure device (VCD) is dedicated to large bore access closure and associated with favorable results in selected study populations. Anatomical predictors for access site complications are lacking. AIM: To evaluate MANTA in a real-world population and identify predictors for vascular complications. METHODS: All patients undergoing transfemoral transcatheter aortic valve replacement (TAVR) between January 2016 and May 2020 with MANTA closure were included. Baseline characteristics were collected, pre-procedural computed tomography and post-deployment femoral angiograms were analyzed for anatomical differences. The primary endpoint was a composite of access site related major and minor vascular complications at 30 days follow-up according to the VARC-2 definitions. Secondary endpoints included bleeding, time to hemostasis, procedural length and incomplete arteriotomy closure or arterial occlusion by angiography. A Cox proportional hazards model was used to compare all-cause mortality for patients with and without an access site complication. RESULTS: The 512 patients underwent TAVR with MANTA access closure. Median age was 80 (IQR 75-85), 53% was male, median BMI was 26.4 kg/m2 (IQR 23.4-29.7). Access site related major- or minor vascular complication occurred in 20 (4%) and 23 (4%) of patients respectively. Median time to hemostasis was 42 s (IQR 28-98). Post deployment angiogram showed an occlusion in 24 patients (5%), incomplete closure in 60 patients (12%) or both in three patients (1%). Of these 87 patients, 36 (41%) had a vascular complication. Femoral artery diameter (OR 0.70 [0.53-0.93]), low- (OR 3.47 [1.21-10.00]) and high (OR 2.43 [1.16-5.10]) arteriotomies were independent predictors for vascular complications. CONCLUSION: In this contemporary TAVR population, access-site related complications occurred in 8% of patients and were mainly due to percutaneous closure device failure. Small artery diameter and off-target punctures were independent predictors.

Validation of novel 3-dimensional quantitative coronary angiography based software to calculate fractional flow reserve post stenting.

OBJECTIVES: To validate novel dedicated 3D-QCA based on the software to calculate post PCI vessel-FFR (vFFR) in a consecutive series of patients, to assess the diagnostic accuracy, and to assess inter-observer variability. BACKGROUND: Low post percutaneous coronary intervention (PCI) fractional flow reserve (FFR) predicts future adverse cardiac events. However, FFR assessment requires the insertion of a pressure wire in combination with the use of a hyperemic agent. METHODS: FAST POST study is an observational, retrospective, single-center cohort study. One hundred patients presenting with stable angina or non ST-elevation myocardial infarction, who underwent post PCI FFR assessment using a dedicated microcatheter were included. Two orthogonal angiographic projections were acquired to create a 3D reconstruction of the coronary artery using the CAAS workstation 8.0. vFFR was subsequently calculated using the aortic root pressure. RESULTS: Mean age was 65±12 years and 70% were male. Mean microcatheter based FFR and vFFR were 0.91±0.07 and 0.91±0.06, respectively. A good linear correlation was found between FFR and vFFR (r = 0.88; p <.001). vFFR had a higher accuracy in the identification of patients with FFR values <0.90, AUC 0.98 (95% CI: 0.96-1.00) as compared with 3D-QCA AUC 0.62 (95% CI: 0.94-0.74). Assessment of vFFR had a low inter-observer variability (r = 0.95; p <.001). CONCLUSION: 3D-QCA derived post PCI vFFR correlates well with invasively measured microcatheter based FFR and has a high diagnostic accuracy to detect FFR <0.90 with low inter-observer variability.

Correlation between 3D-QCA based FFR and quantitative lumen assessment by IVUS for left main coronary artery stenoses.

OBJECTIVES: We aimed to evaluate the feasibility of using three dimensional-quantitative coronary angiography (3D-QCA) based fractional flow reserve (FFR) (vessel fractional flow reserve [vFFR], CAAS8.1, Pie Medical Imaging) and to correlate vFFR values with intravascular ultrasound (IVUS) for the evaluation of intermediate left main coronary artery (LMCA) stenosis. BACKGROUND: 3D-QCA derived FFR indices have been recently developed for less invasive functional lesion assessment. However, LMCA lesions were vastly under-represented in first validation studies. METHODS: This observational single-center cohort study enrolled consecutive patients with stable angina, unstable angina, or non-ST-segment elevation myocardial infarction and nonostial, intermediate grade LMCA stenoses who underwent IVUS evaluation. vFFR was computed based on two angiograms with optimal LMCA stenosis projection and correlated with IVUS-derived minimal lumen area (MLA). RESULTS: A total of 256 patients with intermediate grade LMCA stenosis evaluated with IVUS were screened for eligibility; 147 patients met the clinical inclusion criteria and had a complete IVUS LMCA footage available, of them, 63 patients (63 lesions) underwent 3D-QCA and vFFR analyses. The main reason for screening failure was insufficient quality of the angiogram (51 patients,60.7%). Mean age was 65 ± 11 years, 75% were male. Overall, mean MLA within LMCA was 8.77 ± 3.17 mm2 , while mean vFFR was 0.87 ± 0.09. A correlation was observed between vFFR and LMCA MLA (r = .792, p = .001). The diagnostic accuracy of vFFR ≤0.8 in identifying lesions with MLA < 6.0 mm2 (sensitivity 98%, specificity 71.4%, area under the curve (AUC) 0.95, 95% confidence interval (CI) 0.89-1.00, p = .001) was good. CONCLUSIONS: In patients with good quality angiographic visualization of LMCA and available complete LMCA IVUS footage, 3D-QCA based vFFR assessment of LMCA disease correlates well to LMCA MLA as assessed by IVUS.

Dedicated plug based closure for large bore access -The MARVEL prospective registry.

OBJECTIVES: To study safety and performance of the MANTA Vascular closure device (VCD) under real world conditions in 10 centers. BACKGROUND: The MANTA is a novel plug-based device for large bore arteriotomy closure. METHODS: We included all eligible patients who underwent transfemoral large bore percutaneous procedures. Exclusion criteria were per operator's discretion and included severe calcification or marked tortuosity of the access vessel, presence of marked obesity/cachexia or a systolic blood pressure above 180 mmHg. The primary performance endpoint was time to hemostasis. Primary and secondary safety endpoints were major and minor access site related vascular complications up to 30 days, respectively. Vascular complications were adjudicated by an independent clinical event committee according to VARC-2 criteria. We performed multivariable logistic regression to estimate the effect of baseline and procedural characteristics on any and major vascular complications. RESULTS: Between February 2018 and July 2019 500 patients were enrolled undergoing Transcatheter aortic valve replacement (TAVR, N = 496), Balloon aortic valvuloplasty (BAV, N = 2), Mechanical circulatory support (MCS, N = 1) or Endovascular aneurysm repair (EVAR, N = 1). Mean age was 80.8 ± 6.6 years with a median STS-score of 2.7 [IQR 2.0-4.3] %. MANTA access site complications were major in 20 (4%) and minor in 28 patients (5.6%). Median time to hemostasis was 50 [IQR 20-120] sec. Severe femoral artery calcification, scar presence in groin, longer procedure duration, female gender and history of hypertension were independent predictors for vascular complications. CONCLUSION: In this study, MANTA appeared to be a safe and effective device for large bore access closure under real-world conditions.

Two decades after coronary radiation therapy: A single center longitudinal clinical study.

OBJECTIVES: The aim of this study was to evaluate the very long-term clinical outcome after radioactive stent (RS) implantation and intracoronary ß radiation brachytherapy (IRBT). BACKGROUND: Radioactive stents (RS) and intracoronary ß radiation brachytherapy (IRBT) were introduced to prevent restenosis after percutaneous coronary intervention (PCI). Both techniques were associated with a higher incidence of major adverse cardiac events (MACE) in the short and intermediate-term follow up as compared to conventional PCI. METHODS: One hundred and thirty-three patients received radioactive stents (32 P) and 301 patients were treated with IRBT adjunctive to PCI. These groups were propensity matched to respectively 266 and 602 control patients who were treated with routine PCI during the same inclusion period. Endpoints were all-cause mortality and MACE, defined as all-cause death, any myocardial infarction or any revascularization. RESULTS: Median follow-up duration was 17 years. All-cause mortality rates were similar in all groups. Adjusted hazard ratios for MACE and mortality in the RS cohort were 1.55 (95% CI 1.20-2.00) and 0.92 (95% CI 0.63-1.34), respectively. Adjusted hazard ratios for MACE and all-cause mortality in the IRBT cohort were 1.41 (95% CI 1.18-1.67) and 0.95 (95% CI 0.74-1.21), respectively. The difference in MACE rates was predominantly driven by coronary revascularizations in both groups, with a higher MI rate in the IRBT group as well. CONCLUSIONS: Coronary radiation therapy was associated with early increased MACE rates, but the difference in MACE rates decreased beyond 2 years, resulting in a comparable long-term clinical outcome. Importantly, no excess in mortality was observed.

Serial invasive imaging follow-up of the first clinical experience with the Magmaris magnesium bioresorbable scaffold.

OBJECTIVES: To assess the performance of the commercially available Magmaris sirolimus-eluting bioresorbable scaffold (BRS) with invasive imaging at different time points. BACKGROUND: Coronary BRS with a magnesium backbone have been recently studied as an alternative to polymeric scaffolds, providing enhanced vessel support and a faster resorption rate. We aimed to assess the performance of the commercially available Magmaris sirolimus-eluting BRS at different time points. METHODS: A prospective, single-center, nonrandomized study was performed at the Thoraxcenter, Erasmus Medical Center, Rotterdam, The Netherlands. Six patients with stable de novo coronary artery lesions underwent single-vessel revascularization with the Magmaris sirolimus-eluting BRS. Invasive follow-up including intravascular imaging using optical coherence tomography (OCT) was performed at different time points. RESULTS: At a median of 8 months (range 4-12 months) target lesion failure occurred in one patient. Angiography revealed a late lumen loss of 0.59 ± 0.39 mm, a percentage diameter stenosis of 39.65 ± 15.81%, and a binary restenosis rate of 33.3%. OCT showed a significant reduction in both minimal lumen area (MLA) and scaffold area at the site of the MLA by 43.44 ± 28.62 and 38.20 ± 25.74%, respectively. A fast and heterogeneous scaffold degradation process was found with a significant reduction of patent struts at 4-5 months. CONCLUSIONS: Our findings show that the latest iteration of magnesium BRS suffers from premature dismantling, resulting in a higher than expected decrease in MLA.

Long-term outcome in patients treated with first- versus second-generation drug-eluting stents for the treatment of unprotected left main coronary artery stenosis.

OBJECTIVE AND BACKGROUND: The study aim is to provide long-term clinical outcome after percutaneous coronary intervention (PCI) for unprotected left main coronary arteries (ULMCA) stenosis with the first-generation (1st -gen) drug-eluting stents (DES) in comparison to 2nd -gen DES, since this is largely unknown. METHODS: Between May 2002, and December 2014, a consecutive series of 656 all-comer patients underwent a PCI for ULMCA stenosis at the Erasmus Medical Center. A total of 235 patients were treated with 1st -gen DES, while a total of 421 patients were treated with 2nd -gen DES. RESULTS: Overall, the population consisted of 73% males and 58% presented with an acute coronary syndrome. Median follow-up time was 1,361 days (range from 0 to 5,031). At 5 years, the cumulative incidence of major adverse clinical events (the primary composite endpoint of all-cause death, any myocardial infarction or target lesion revascularization; MACE) did not differ between 1st - and 2nd -gen DES (36.8 vs. 38.6%, respectively, Log Rank p = .79, adjusted hazard ratio [HR] = 1.28 [95% confidence interval (CI) 0.94-1.74]). No difference was found in the individual endpoints of all-cause mortality (29.5 vs. 29% respectively, p = .88, adjusted HR = 1.19 [95% CI, 0.84-1.68]), target vessel myocardial infarction (5.0 vs. 8.4%, p = 0.17, adjusted HR = 1.75 [95% CI, 0.78-3.96]) and target lesion revascularization (8.1 vs. 9.8%, p = .94, adjusted HR = 1.16 [95% CI, 0.59-2.29]) between the 1st - and 2nd -gen DES cohorts, respectively. CONCLUSIONS: In this large cohort of consecutive patients treated for ULMCA stenosis, no significant differences were found in the safety and efficacy of 1st versus 2nd -gen DES at 5 years follow-up.

Automatic Detection of the Aortic Annular Plane and Coronary Ostia from Multidetector Computed Tomography.

Anatomic landmark detection is crucial during preoperative planning of transcatheter aortic valve implantation (TAVI) to select the proper device size and assess the risk of complications. The detection is currently a time-consuming manual process influenced by the image quality and subject to operator variability. In this work, we propose a novel automatic method to detect the relevant aortic landmarks from MDCT images using deep learning techniques. We trained three convolutional neural networks (CNNs) with 344 multidetector computed tomography (MDCT) acquisitions to detect five anatomical landmarks relevant for TAVI planning: the three basal attachment points of the aortic valve leaflets and the left and right coronary ostia. The detection strategy used these three CNN models to analyse a single MDCT image and yield three segmentation volumes as output. These segmentation volumes were averaged into one final segmentation volume, and the final predicted landmarks were obtained during a postprocessing step. Finally, we constructed the aortic annular plane, defined by the three predicted hinge points, and measured the distances from this plane to the predicted coronary ostia (i.e., coronary height). The methodology was validated on 100 patients. The automatic landmark detection was able to detect all the landmarks and showed high accuracy as the median distance between the ground truth and predictions is lower than the interobserver variations (1.5 mm [1.1-2.1], 2.0 mm [1.3-2.8] with a paired difference -0.5 ± 1.3 mm and p value <0.001). Furthermore, a high correlation is observed between predicted and manually measured coronary heights (for both R 2 = 0.8). The image analysis time per patient was below one second. The proposed method is accurate, fast, and reproducible. Embedding this tool based on deep learning in the preoperative planning routine may have an impact in the TAVI environments by reducing the time and cost and improving accuracy.

Incidence, determinants and consequences of delirium in older patients after transcatheter aortic valve implantation.

BACKGROUND: delirium is an event leading to negative health outcomes and increased mortality in patients. The aim of this study is to investigate the incidence, determinants and consequences of post-operative delirium (POD) in older patients undergoing transcatheter aortic valve implantation (TAVI). METHODS: The TAVI Care and Cure program is a prospective, observational registry in patients referred for TAVI at Erasmus University Medical Centre. The presence of delirium was evaluated by daily clinical assessment by a geriatrician pre- and up to 3 days post-TAVI. Mortality data were obtained from the Dutch Civil Registry. RESULTS: A total of 543 patients underwent TAVI between January 2014 and December 2017. Overall, the incidence of POD was 14% (75/543 patients) but declined from 18% in 2014 to 7% in 2017 (P = 0.009). Patients who developed POD were older (81.9 ± 5.8 versus 78.6 ± 8.3 years, P < 0.001), had higher prevalence of renal dysfunction and prior stroke (54% versus 40%, P = 0.02; 31% versus 18%, P = 0.01) and were more often frail (32% versus 25%, P = 0.02). From a procedural perspective, general anesthesia (odds ratios (OR), 2.31; 95% CI, 1.40-3.83; P = 0.001), non-transfemoral access (OR, 2.37; 95% CI, 1.20-4.70; P = 0.01) and longer procedural time (OR, 1.01; 95% CI, 1.01-1.02; P < 0.001) were significantly associated with POD. One-year survival rate was 68% among patients who had suffered a POD and was 85% in patients without a POD (hazard ratio's 1.8 (95% CI 1.01-3.10), P = 0.045). CONCLUSION: POD frequently occurs after TAVI and is associated with increased mortality. It might be speculated that patient selection and the minimalistic approach of TAVI may reduce the frequency of delirium.

Impact of frailty on health-related quality of life 1 year after transcatheter aortic valve implantation.

BACKGROUND: Transcatheter aortic valve implantation (TAVI) brings symptom relief and improvement in health-related quality of life (HRQoL) in the majority of patients treated for symptomatic, severe aortic stenosis. However, there is a substantial group of patients that do not benefit from TAVI. The aim of this study is to investigate the impact of frailty on HRQoL 1 year after TAVI. METHODS: The TAVI Care & Cure Program is an ongoing, prospective, observational study including patients referred for TAVI to our institution. A comprehensive geriatric assessment was performed to evaluate existence of frailty using the Erasmus Frailty Score (EFS). HRQoL was assessed using the EQ-5D-5 L at baseline and 1 year after TAVI. RESULTS: 239 patients underwent TAVI and completed HRQoL assessment 1 year after TAVI. Seventy (29.3%) patients were classified as frail (EFS ≥ 3). In non-frail patients, the EQ-5D-5 L index did not change (0.71(± 0.22) to 0.68(± 0.33) points, P = 0.22); in frail patients, the EQ-5D-5 L index decreased from 0.55(±0.26) to 0.44 points (±0.33) (P = 0.022). Frailty was an independent predictor of deteriorated HRQoL 1 year after TAVI (OR 2.24, 95% CI 1.07-4.70, P = 0.003). In frail patients, the absence of peripheral artery disease (OR 0.17, 95% 0.05-0.50, P = 0.001) and renal dysfunction (OR 0.13, 95% CI 0.04-0.41, P = <0.001) at baseline was associated with improved HRQoL 1 year after TAVI. CONCLUSION: Frailty is associated with deterioration of HRQoL 1 year after TAVI. Notably, HRQoL did improve in frail patients with no peripheral arterial disease or renal impairment at baseline.

Fractional flow reserve guided percutaneous coronary intervention optimization directed by high-definition intravascular ultrasound versus standard of care: Rationale and study design of the prospective randomized FFR-REACT trial.

BACKGROUND: Post percutaneous coronary intervention (PCI) fractional flow reserve (FFR) is a significant predictor of major adverse cardiac events (MACE). The rationale for low post procedural FFR values often remains elusive based on angiographic findings alone, warranting further assessment using an FFR pullback or additional intravascular imaging. It is currently unknown if additional interventions intended to improve the PCI, decrease MACE rates. STUDY DESIGN: The FFR REACT trial is a prospective, single-center randomized controlled trial in which 290 patients with a post PCI FFR <0.90 will be randomized (1:1) to either standard of care (no additional intervention) or intravascular ultrasound (IVUS)-directed optimization of the FFR (treatment arm). Eligible patients are those treated with angiographically successful PCI for (un)stable angina or non-ST elevation myocardial infarction (MI). Assuming 45% of patients will have a post PCI FFR <0.90, approximately 640 patients undergoing PCI will need to be enrolled. Patients with a post PCI FFR ≥ 0.90 will be enrolled in a prospective registry. The primary end point is defined as a composite of cardiac death, target vessel MI and clinically driven target vessel revascularisation (target vessel failure) at 1 year. Secondary end points will consist of individual components of the primary end point, procedural success, stent thrombosis and correlations on clinical outcome, changes in post PCI Pd/Pa and FFR and IVUS derived dimensions. All patients will be followed for 3 years. CONCLUSION: The FFR-REACT trial is designed to explore the potential benefit of HD-IVUS-guided PCI optimization in patients with a post PCI FFR <0.90 (Dutch trial register: NTR6711).

References for left main stem dimensions: A cross sectional intravascular ultrasound analysis.

BACKGROUND: Angiographic assessment of left main coronary artery (LMCA) lesions remains challenging and limited data is available on reference diameters and length of nonobstructive LMCA dimensions. Our aim was to provide insights in the dimensions of nonobstructive LMCA and to find a possible correlation with gender and patient habitus. METHODS: This retrospective single center study was performed in a consecutive cohort of patients who underwent Intravascular Ultrasound (IVUS) guided percutaneous coronary interventions of the left coronary system including complete pullbacks of a non-obstructive LMCA (n = 254). RESULTS: Mean LMCA length as measured with IVUS was 7.37 ± 4.2 mm and mean lumen area (LA) was 15.63 ± 4.76 mm2 corresponding to a mean lumen diameter (LD) of 4.41 ± 0.67 mm. An IVUS derived mean LD of >4 mm was present in 71.7%, >4.5 mm in 43% and > 5 mm in 19% of patients. LMCA mean LA was significantly smaller in women as compared to men (14.1 ± 4.1 mm2 and 16.2 ± 4.8 mm2 , P < 0.01). Multivariable analysis identified weight of the patient as the sole significant predictor for LMCA length while height of the patient and LMCA length were predictors of LMCA mean LA. Correlation coefficients of determination for all independent predictors were low (R2 < 0.1 for all). CONCLUSION: This study demonstrated that the mean LD of a non-obstructive LMCA is 4 mm or greater in the majority of patients, with a mean LMCA length of 7.4 mm. Women have smaller luminal dimension than men. No clinically relevant predictors were found for both LMCA length and mean LA.

Impact of baseline cigarette smoking status on clinical outcome after transcatheter aortic valve replacement.

OBJECTIVES: To explore the prevalence of smoking, and its association with clinical and mortality outcome among patients undergoing transcatheter aortic valve replacement (TAVR). BACKGROUND: Less data exist regarding the effect of baseline smoking status on clinical and mortality outcome among patients undergoing TAVR. METHODS: Consecutive patients who underwent TAVR at two high volume Dutch centers were included. Smoking status was prospectively questioned by a structured interview at admission. Primary endpoint was 1-year all-cause mortality after TAVR. RESULTS: A total of 913 consecutive patients (80.1 ± 7.6 years; logistic EuroSCORE: 16.5 ± 9.9%) who underwent TAVR for severe aortic valve stenosis were included. There were 47% (n = 432) males, and 57% (n = 522) never-smokers, and 35% (n = 317) prior-smokers, and 8% (n = 74) current-smokers. Smokers (i.e., prior-smokers or current-smokers) were younger compared to never-smokers (78.9 ± 7.9 and 76.4 ± 8.0 vs. 81.3 ± 7.1, P < 0.000, respectively). Median follow-up time was 365 (interquartile range [IQR]: 280-365) days. Overall, prior-smoking was not associated with all-cause mortality at 1-year following TAVR (hazard ratio [HR] 0.83, 95% confidence interval [CI] 0.55-1.23). After stratification according to sex, male prior-smokers showed better 1-year survival after TAVR than male never-smokers (12% vs. 20%; P = 0.018, respectively, HR 0.52, 95% CI 0.29-0.89), while this reversed effect was not observed among female prior-smokers versus female never-smokers after TAVR (HR 1.70, 95% CI 0.95-3.05). CONCLUSIONS: Overall, baseline prior-smokers had similar 1-year mortality outcome after TAVR compared with baseline never-smokers. However, there was a reversed association between baseline prior-smoking status and 1-year mortality after TAVR among males, which could partially be explained due to the favorable baseline characteristics.

Enabling Automated Device Size Selection for Transcatheter Aortic Valve Implantation.

The number of transcatheter aortic valve implantation (TAVI) procedures is expected to increase significantly in the coming years. Improving efficiency will become essential for experienced operators performing large TAVI volumes, while new operators will require training and may benefit from accurate support. In this work, we present a fast deep learning method that can predict aortic annulus perimeter and area automatically from aortic annular plane images. We propose a method combining two deep convolutional neural networks followed by a postprocessing step. The models were trained with 355 patients using modern deep learning techniques, and the method was evaluated on another 118 patients. The method was validated against an interoperator variability study of the same 118 patients. The differences between the manually obtained aortic annulus measurements and the automatic predictions were similar to the differences between two independent observers (paired diff. of 3.3 ± 16.8 mm2 vs. 1.3 ± 21.1 mm2 for the area and a paired diff. of 0.6 ± 1.7 mm vs. 0.2 ± 2.5 mm for the perimeter). The area and perimeter were used to retrieve the suggested prosthesis sizes for the Edwards Sapien 3 and the Medtronic Evolut device retrospectively. The automatically obtained device size selections accorded well with the device sizes selected by operator 1. The total analysis time from aortic annular plane to prosthesis size was below one second. This study showed that automated TAVI device size selection using the proposed method is fast, accurate, and reproducible. Comparison with the interobserver variability has shown the reliability of the strategy, and embedding this tool based on deep learning in the preoperative planning routine has the potential to increase the efficiency while ensuring accuracy.

Prevalence and consequences of noncardiac incidental findings on preprocedural imaging in the workup for transcatheter aortic valve implantation, renal sympathetic denervation, or MitraClip implantation.

BACKGROUND: Dedicated data on the prevalence of incidental findings (IF) stratified according to overall clinical relevance and their subsequent correlation to outcome are lacking. The aim of the present study was to describe the prevalence and consequences of noncardiac IF on computed tomography or magnetic resonance imaging in the workup for interventional cardiovascular procedures. METHODS: A total of 916 patients underwent preprocedural computed tomography or magnetic resonance imaging in the workup for transcatheter aortic valve implantation (TAVI), renal sympathetic denervation (RDN), or MitraClip implantation. RESULTS: IF were found in 395 of 916 patients (43.1%), with an average of 1.8 IF per patient. Classifying the IF resulted in 155 patients with minor, 171 patients with moderate, and 69 patients with major IF. The intended procedure was delayed or canceled in only 15 of 916 (1.6%) of the patients because of the presence of potential malignant IF. In patients that did undergo the intended procedure (n = 774), the presence of a moderate or major IF (23.8%) did not impact 1-year mortality compared to no or minor IF (adjusted HR 0.90, 95% CI 0.56-1.44, P value = .65). These findings were consistent among patients referred for TAVI, RDN, or MitraClip. CONCLUSIONS: IF are frequent in patients referred for cardiovascular procedures. IF did not result in a delay or cancellation of the intended procedure in the vast majority of cases, irrespective of their clinical relevance. The presence of a major or moderate IF did not significantly impact 1-year mortality.

Association of stent-induced changes in coronary geometry with late stent failure: Insights from three-dimensional quantitative coronary angiographic analysis.

BACKGROUND: The relationship between vessel angulation and large changes in vessel geometry after stent implantation and the occurrence of stent failure still remains unclear. We sought to investigate the association of the change in the coronary bending angle after stenting and the risk for late stent failure by three-dimensional quantitative coronary angiography (3D QCA). METHODS: The bending angle in coronary lesions that presented with late stent failure and those without stent failure was computed during the cardiac cycle, before and after stenting using a recently developed 3D QCA software. RESULTS: A total of 40 lesions with stent failure (cases) were successfully matched to 47 lesions without stent failure (controls).The mean duration to follow-up coronary angiography was 1,011 days in cases and 1,109 days in the control group (P = 0.14). In stent failure, the systolic bending angle after stenting was smaller (14.45° [12.18, 17.68] versus 18.20° [14.00, 20.30], P = 0.01), while the stent-induced change in systolic bending angle was significantly larger (4.15° [1.13, 7.20] versus 1.80° [-1.90, 4.40], P = 0.004). Multivariable logistic regression analysis suggested that systolic bending angle after stenting (odds ratio: 0.88; 95% CI: 0.79-0.99; P = 0.03), and decrease in systolic bending angle after stenting (odds ratio: 1.13; 95% CI: 1.02-1.26; P = 0.03) were predictors of stent failure. CONCLUSIONS: Our study suggests that a change in the natural tortuous course of the coronaries by stent implantation with the decrease in coronary bending angle is a potentially major contributor in stent failure.

Complete filter-based cerebral embolic protection with transcatheter aortic valve replacement.

OBJECTIVES: To evaluate the value of left vertebral artery filter protection in addition to the current filter-based embolic protection technology to achieve complete cerebral protection during TAVR. BACKGROUND: The occurrence of cerebrovascular events after transcatheter aortic valve replacement (TAVR) has fueled concern for its potential application in younger patients with longer life expectancy. Transcatheter cerebral embolic protection (TCEP) devices may limit periprocedural cerebrovascular events by preventing macro and micro-embolization to the brain. Conventional filter-based TCEP devices cover three extracranial contributories to the brain, yet leave the left vertebral artery unprotected. METHODS: Patients underwent TAVR with complete TCEP. A dual-filter system was deployed in the brachiocephalic trunk and left common carotid artery with an additional single filter in the left vertebral artery. After TAVR all filters were retrieved and sent for histopathological evaluation by an experienced pathologist. RESULTS: Eleven patients received a dual-filter system and nine of them received an additional left vertebral filter. In the remaining two patients, the left vertebral filter could not be deployed. No periprocedural strokes occurred. We found debris in all filters, consisting of thrombus, tissue derived debris, and foreign body material. The left vertebral filter contained debris in an equal amount of patients as the Sentinel filters. The size of the captured particles was similar between all filters. CONCLUSIONS: The left vertebral artery is an important entry route for embolic material to the brain during TAVR. Selective filter protection of the left vertebral artery revealed embolic debris in all patients. The clinical value of complete filter-based TCEP during TAVR warrants further research.