Eleid, Mackram F.

Introduction: Ischemic stroke (IS) causes substantial morbidity and mortality in patients undergoing percutaneous coronary intervention (PCI) with a 5 yr incidence ~ 3%. We sought the test the accuracy of Machine learning (ML) algorithms in predicting IS in patients undergoing PCI. Methods: Mayo Clinic CathPCI registry data were retrospectively analyzed from Jan 2003 - June 2018 including 21,872 patients who underwent PCI. The cohort was randomly divided into a training sample (75%, n=16404) and a unique test sample (25%, n=5468) prior to model generation. The risk prediction model was generated utilizing a random forest algorithm (RF model) on 188 unique variables to predict the risk of IS at 6-month, 1, 2, and 5-year post PCI. Conventional risk factors for stroke were used for logistic regression. The receiver operating characteristic (ROC) curve and area under the curve for the RF and logistic regression models were compared for the test cohort. Results: The mean age was 66.9 ± 12.4 years, and 71% were male. Patient demographics and outcomes are shown in Table 1 . The ROC area under the curve for the RF model was superior compared to the logistic regression model in predicting IS at 6 months, 1,2 and 5 yrs for the test cohort ( Figure 1 .) Conclusions: The RF model accurately predicts short and long term risk of IS and outperforms logistic regression analysis in patients undergoing PCI.

AbstractObjective/BackgroundSmall balloon expandable valves have higher echocardiographic transvalvular gradients and rates of prosthesis‐patient mismatch (PPM) compared to larger valves. However, the impact of these echocardiographic findings on clinical outcomes is unknown. We sought to determine the clinical outcomes of 20 mm SAPIEN 3 (S3 BEV) compared to larger S3 BEV in relation to echocardiographic hemodynamics.MethodsUsing the STS/ACC transcatheter valve registry, we performed a propensity‐matched comparison of patients undergoing treatment of native aortic valve stenosis using transfemoral, balloon‐expandable implantation of 20 mm and ≥ 23 mm S3 BEVs. Baseline and procedure characteristics, echocardiographic variables and survival were analyzed. Multivariable logistic regression was used to identify predictors of 1‐year mortality.ResultsAfter propensity matching of the 20 mm and ≥ 23 mm SAPIEN 3 valves, 3,931 pairs with comparable baseline characteristics were identified. Small valves were associated with significantly higher echocardiographic gradients at discharge (15.7 ± 7.1 mmHg vs. 11.7 ± 5.5 mmHg, p < 0.0001) and severe PPM rates (21.5% vs. 9.7%, p < 0.0001). There was no significant difference in 1‐year all‐cause mortality (20 mm: 13.0% vs. ≥23 mm: 12.7%, p = 0.72) or other major adverse event rates and outcomes between the two cohorts. Based on a multivariable analysis, elevated discharge mean gradient (>20 mmHg), severe PPM and the use of 20 mm versus ≥23 mm were not independent predictors of 1‐year mortality.ConclusionSAPIEN 3 20 mm valves were associated with higher echocardiographic gradients, and severe PPM rates compared to larger valves but these factors were not associated with significant differences in 1‐year all‐cause mortality or rehospitalization.

Abstract Background Bioprosthetic tricuspid valve stenosis (TS) is an uncommon and frequently under-diagnosed condition. Although the resulting right heart failure symptoms are well-known, the associated thrombogenic potential is under-recognized. Case summary A 44-year-old woman with bioprosthetic tricuspid valve (TV) replacement in 2001 was referred for urgent consultation due to acute worsening of dyspnoea and severe swelling and pain in her left arm and neck. She was diagnosed with atrial fibrillation 6 months before the presentation and was found to have right atrial (RA) thrombus with pulmonary embolism and extensive retrograde venous extension 1 month prior. Review of studies done at her local institution revealed 10 mmHg mean gradient (MG) across the bioprosthetic TV that was only reported as mild–moderate TS. Echocardiography done at our instruction confirmed suspicion of severe TS with calcified immobile leaflets. Computed tomography showed persistent RA thrombus and therefore surgical replacement of the TV was undertaken. Subsequently, patient’s dyspnoea rapidly improved. Discussion Progressive dyspnoea and symptoms of right heart failure in a patient with a history of bioprosthetic TV replacement should be investigated for prosthetic valve dysfunction. Due to its rarity, TS diagnosis can be overlooked on routine echocardiography. In our patient, despite a measured MG of 10 mmHg, the presence of critical TS was not initially recognized. As TS is associated with increased thrombogenic potential and given the rare occurrence of in situ RA thrombosis, physicians must have a high index of suspicion for TS in the appropriate clinical context.

Background: Identifying patients at a high risk of mortality post percutaneous coronary intervention (PCI) is of vital clinical importance. We investigated the utility of machine learning algorithms to predict short and intermediate-term risk of all-cause mortality in patients undergoing PCI. Methods: Patient-level demographics, clinical, electrocardiographic ,echocardiographic and angiographic data from January 2006 to December 2017 were extracted from the Mayo Clinic CathPCI registry and clinical records. For patients with multiple PCI events, data collected at the time of the index PCI was used for analysis. Patients who underwent bailout coronary artery bypass graft surgery (CABG) prior to discharge were excluded. 306 variables were incorporated into random forest machine learning model (RF) to predict all-cause mortality at 6 months and 1 year after PCI. Ten-fold cross-validation repeated five times was used to optimize the hyperparameters and estimate its external performance. The National Cardiovascular Data Registry (NCDR) based logistic regression model was used for comparison. The area under receiver operator characteristic curves (AUC) was calculated to assess the ability of the models to predict all-cause mortality. Results: A total of 17356 unique patients were included for the final analysis after excluding 165 patients who underwent CABG surgery during the index hospitalization. The mean age was 66.9 ± 12.5 years;71% were male. Indications for PCI were ST-elevation myocardial infarction (9.4%), non-ST elevation myocardial infarction (12.9%), unstable angina (17.7%), and stable angina (52.8%) in the cohort. In-hospital, 6-month & 1 year mortality rates were 1.9%,4.2% & 5.8% respectively. The RF model was superior to the NCDR model in predicting inhospital, 6-month, 1 year mortality (p<0.0001) ( Figure 1 ). Conclusion: Machine learning is superior to NCDR model in predicting short and intermediate risk of all-cause mortality post PCI.

Background The temporal incidence of high‐grade atrioventricular block (HAVB) after transcatheter aortic valve replacement (TAVR) is uncertain. As a result, periprocedural monitoring and pacing strategies remain controversial. This study aimed to describe the temporal incidence of initial episode of HAVB stratified by pre‐ and post‐TAVR conduction and identify predictors of delayed events. Methods and Results Consecutive patients undergoing TAVR at a single center between February 2012 and June 2019 were retrospectively assessed for HAVB within 30 days. Patients with prior aortic valve replacement, permanent pacemaker (PPM), or conversion to surgical replacement were excluded. Multivariable logistic regression was performed to assess predictors of delayed HAVB (initial event >24 hours post‐TAVR). A total of 953 patients were included in this study. HAVB occurred in 153 (16.1%). After exclusion of those with prophylactic PPM placed post‐TAVR, the incidence of delayed HAVB was 33/882 (3.7%). Variables independently associated with delayed HAVB included baseline first‐degree atrioventricular block or right bundle‐branch block, self‐expanding valve, and new left bundle‐branch block. Forty patients had intraprocedural transient HAVB, including 16 who developed HAVB recurrence and 6 who had PPM implantation without recurrence. PPM was placed for HAVB in 130 (13.6%) (self‐expanding valve, 23.7% versus balloon‐expandable valve, 11.9%; P <0.001). Eight (0.8%) patients died by 30 days, including 1 unexplained without PPM present. Conclusions Delayed HAVB occurs with higher frequency in patients with baseline first‐degree atrioventricular block or right bundle‐branch block, new left bundle‐branch block, and self‐expanding valve. These findings provide insight into optimal monitoring and pacing strategies based on periprocedural ECG findings.

Background Transcatheter mitral valve repair (TMVr) is currently offered at selected centers that meet certain operator and institutional requirements. We sought to explore the hypothesis that the availability of TMVr is associated with improved outcomes of MV surgery. Methods and Results We used the Nationwide Readmissions Database to identify patients who underwent MV surgery at centers with or without TMVr capabilities between January 1 and December 31, 2017. The primary end point was in‐hospital mortality. Secondary end points were postoperative complications, resource use, and 30‐day readmissions. A total of 24 477 patients from 595 centers (446 TMVr, 149 non‐TMVr) were included. There were modest but statistically significant differences in the prevalence of comorbidities between the groups. Patients at non‐TMVr centers had higher unadjusted in‐hospital mortality than those at TMVr centers (5.6% versus 3.6%, P <0.001). They also had higher rates of postoperative complications, longer hospitalizations, higher cost, and fewer home discharges but similar 30‐day readmission rates. After propensity matching, mortality remained higher at non‐TMVr centers (5.5% versus 4.0%, P <0.001). Rates of postoperative complications, prolonged hospitalizations, and nonhome discharges also remained higher. Postoperative mortality was consistently higher at non‐TMVr centers in multiple risk‐adjustment analyses incrementally accounting for differences in risk factors, surgical volume, availability of surgical repair, and excluding concomitant procedures. In the most comprehensive model, surgery at non‐TMVr centers was associated with higher odds of death (odds ratio, 1.41; 95% CI, 1.14–1.73; P =0.002). Conclusions Mitral valve surgery at TMVr centers is associated with improved in‐hospital outcomes compared with non‐TMVr centers.

Abstract Aims Beat-to-beat variability in cycle length is well-known in atrial fibrillation (Afib); whether this also translates to variability in annulus size remains unknown. Defining annulus maximal size in Afib is critical for accurate selection of percutaneous devices given the frequent association with mitral and tricuspid valve diseases. Methods and results Images were obtained from 170 patients undergoing 3D echocardiography [100 (50 sinus rhythm (SR) and 50 Afib) for mitral annulus (MA) and 70 (35 SR and 35 Afib) for tricuspid annulus (TA)]. Images were analysed for differences in annular dynamics with a commercially available software. Number of cardiac cycles analysed was 567 in mitral valve and 346 in tricuspid valve. Median absolute difference in maximal MA area over four to six cycles was 1.8 cm2 (range 0.5–5.2 cm2) in Afib vs. 0.8 cm2 (range 0.1–2.9 cm2) in SR, P &lt; 0.001. Maximal MA area was observed within 30–70% of the R-R interval in 81% of cardiac cycles in SR and in 73% of cycles in Afib. Median absolute difference in maximal TA area over four to six cycles was 1.4 cm2 (range 0.5–3.6 cm2) in Afib vs. 0.7 cm2 (range 0.3–1.7 cm2) in SR, P &lt; 0.001. Maximal TA area was observed within 60–100% of the R-R interval in 81% of cardiac cycles in SR, but only in 49% of cycles in Afib. Conclusion MA and TA reach maximal size within a broad time interval centred around end-systole and end-diastole, respectively, with significant beat-to-beat variability. Afib leads to a larger beat-to-beat variability in both timing of occurrence and values of annulus size than in SR.

AbstractBackgroundChange in left atrial pressure (LAP) has been shown to be associated with symptom improvement post‐MitraClip; however, the association between acute procedural changes in transmitral diastolic mean gradient (MG) compared to LAP and symptom improvement is not well established.Methods164 consecutive patients undergoing MitraClip at Mayo Clinic between June 2014 and May 2018 were included. Preclip and postclip MG and LAP were recorded. Baseline demographics, clinical, and echocardiographic outcomes, including 30‐day New York Heart Association (NYHA) functional status were obtained from patient charts.ResultsMedian age was 81.5 years (IQR: 76.3, 87), 34% were female and 94.5% had NYHA class III and IV functional status at baseline. At baseline, median MG was 4 mmHg (IQR: 3, 5) and LAP was 19 mmHg (IQR: 16, 23.5). Following MitraClip deployment, the median MG was 4 mmHg (IQR: 3, 6) and the median LAP was 17 mmHg (IQR: 14, 21), 69.5% of patients had less than moderate MR. There was no statistically significant association between change in MG and NYHA functional class at 30 days (OR = 0.95, 95% CI: 0.76–1.20). However, a reduction in LAP following MitraClip deployment was significantly associated with improvement in NYHA functional status at 30 days following adjustments for age and sex (aOR 3.36, 95% CI: 1.34–8.65). There was no significant correlation between change in mean LAP and change in MG (p = .98).ConclusionUnlike change in left atrial pressure, change in MG post‐MitraClip was not associated with patient reported outcomes at 30 days and did not correlate with change in left atrial pressure. Long‐term follow up is needed to evaluate the impact of LA pressure on symptoms.