Machine-learning approaches for risk prediction in transcatheter aortic valve implantation: Systematic review and meta-analysis.

OBJECTIVES: With the expanding integration of artificial intelligence (AI) and machine learning (ML) into the structural heart domain, numerous ML models have emerged for the prediction of adverse outcomes after transcatheter aortic valve implantation (TAVI). We aim to identify, describe, and critically appraise ML prediction models for adverse outcomes after TAVI. Key objectives consisted in summarizing model performance, evaluating adherence to reporting guidelines, and transparency. METHODS: We searched PubMed, SCOPUS, and Embase through August 2023. We selected published machine learning models predicting TAVI outcomes. Two reviewers independently screened articles, extracted data, and assessed the study quality according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Outcomes included summary C-statistics and model risk of bias assessed with the Prediction Model Risk of Bias Assessment Tool. C-statistics were pooled using a random-effects model. RESULTS: Twenty-one studies (118,153 patients) employing various ML algorithms (76 models) were included in the systematic review. Predictive ability of models varied: 11.8% inadequate (C-statistic <0.60), 26.3% adequate (C-statistic 0.60-0.70), 31.6% acceptable (C-statistic 0.70-0.80), and 30.3% demonstrated excellent (C-statistic >0.80) performance. Meta-analyses revealed excellent predictive performance for early mortality (C-statistic: 0.81; 95% confidence interval [CI], 0.65-0.91), acceptable performance for 1-year mortality (C-statistic: 0.76; 95% CI, 0.67-0.84), and acceptable performance for predicting permanent pacemaker implantation (C-statistic: 0.75; 95% CI, 0.51-0.90). CONCLUSIONS: ML models for TAVI outcomes exhibit adequate-to-excellent performance, suggesting potential clinical utility. We identified concerns in methodology and transparency, emphasizing the need for improved scientific reporting standards.

Outcomes of Valve-in-Valve Transcatheter Aortic Valve Replacement.

Structural valve degeneration is increasingly seen given the higher rates of bioprosthetic heart valve use for surgical and transcatheter aortic valve replacement (TAVR). Valve-in-valve TAVR (VIV-TAVR) is an attractive alternate for patients who are otherwise at high risk for reoperative surgery. We compared patients who underwent VIV-TAVR and native valve TAVR through a retrospective analysis of our institutional transcatheter valve therapy (TVT) database from 2013 to 2022. Patients who underwent either a native valve TAVR or VIV-TAVR were included. VIV-TAVR was defined as TAVR in patients who underwent a previous surgical aortic valve replacement. Kaplan-Meier survival analysis was used to obtain survival estimates. A Cox proportional hazards regression model was used for the multivariable analysis of mortality. A total of 3,532 patients underwent TAVR, of whom 198 (5.6%) underwent VIV-TAVR. Patients in the VIV-TAVR cohort were younger than patients who underwent native valve TAVR (79.5 vs 84 years, p <0.001), with comparable number of women and a higher Society of Thoracic Surgeons risk score (6.28 vs 4.46, p <0.001). The VIV-TAVR cohort had a higher incidence of major vascular complications (2.5% vs 0.8%, p = 0.008) but lower incidence of permanent pacemaker placement (2.5% vs 8.1%, p = 0.004). The incidence of stroke was comparable between the groups (VIV-TAVR 2.5% vs native TAVR 2.4%, p = 0.911). The 30-day readmission rates (VIV-TAVR 7.1% vs native TAVR 9%, p = 0.348), as well as in-hospital (VIV-TAVR 2% vs native TAVR 1.4%, p = 0.46), and overall (VIV-TAVR 26.3% vs native TAVR 30.8%, p = 0.18) mortality at a follow-up of 1.8 years (0.83 to 3.5) were comparable between the groups. The survival estimates were also comparable between the groups (log-rank p = 0.27). On multivariable Cox regression analysis, VIV-TAVR was associated with decreased hazards of death (hazard ratio 0.68 [0.5 to 0.9], p = 0.02). In conclusion, VIV-TAVR is a feasible and safe strategy for high-risk patients with bioprosthetic valve failure. There may be potentially higher short-term morbidity with VIV-TAVR, with no overt impact on survival.

Readmission-related outcomes of surgical versus transcatheter aortic valve replacement in patients aged 65 or older with bicuspid aortic valve.

BACKGROUND: There continues to be debate regarding the superiority of transcatheter (TAVR) over surgical aortic valve replacement (SAVR) in patients with bicuspid aortic valves (BAV). We aimed to compare outcomes during readmissions in elderly patients with BAV who underwent SAVR or TAVR. METHODS: Patients 65 years or older with BAV who underwent TAVR or isolated SAVR were identified using the National Readmission Database from 2012 through 2018. We compared outcomes during readmissions within 90 days after discharge from the index surgery. Propensity score matching was performed to adjust the baseline differences. RESULTS: During the study period, 8555 and 1081 elderly patients with BAV underwent SAVR and TAVR, respectively. The number of patients who underwent TAVR went up by 179 % from 2012 to 2018. Propensity score matching yielded 573 patients in each group. A total of 111 (19.4 %) in the SAVR group and 125 (21.8 %) in the TAVR group were readmitted within 90 days after the index surgery (p = .31). The mortality during the readmissions within 90 days was equivalent between the two groups (0.9 % in the SAVR group vs. 3.2 % in the TAVR group, p = .22). However, the median hospital cost was approximately doubled in the TAVR group during the readmission (18,250 dollars vs. 9310 dollars in the SAVR group, p < .001). CONCLUSIONS: Readmission within 90 days was common in both groups. While the mortality during the readmissions after the surgery was equivalent between the two groups, hospital cost was significantly more expensive in the TAVR group.

Long-Term Outcomes of Valve-in-Valve Transcatheter Aortic Valve Implantation Versus Redo Surgical Aortic Valve Replacement: Meta-Analysis of Kaplan-Meier-Derived Data.

Valve-in-valve (ViV) transcatheter aortic valve implantation (ViV-TAVI) in patients with failed bioprostheses arose as an alternative to redo surgical aortic valve replacement (SAVR). To evaluate all-cause mortality in ViV-TAVI versus redo-SAVR, we performed a study-level meta-analysis of reconstructed time-to-event data from Kaplan-Meier curves of nonrandomized studies published by August 2023. A total of 16 studies met our eligibility criteria, with a total of 4,373 patients (2,204 patients underwent ViV-TAVI and 2,169 patients underwent redo-SAVR). Pooling all the studies, ViV-TAVI showed a lower risk of all-cause mortality in the first 6 months (hazard ratio [HR] 0.58, 95% confidence interval [CI] 0.46 to 0.73, p <0.001), with an HR reversal after this time point favoring redo-SAVR (HR 1.92, 95% CI 1.58 to 2.33, p <0.001). Pooling only the matched populations (which represented 64.6% of the overall population), ViV-TAVI showed a lower risk of all-cause mortality in the first 6 months (HR 0.56, 95% CI 0.43 to 0.73, p <0.001], with a reversal after 6 months favoring redo-SAVR (HR 1.55, 95% CI 1.25 to 1.93, p <0.001). The meta-regression analyses revealed a modulating effect of the following covariates: age, coronary artery disease, history of coronary artery bypass graft surgery, and implanted valves <25 mm. In conclusion, ViV-TAVI is associated with better survival immediately after the procedure than redo-SAVR; however, this primary advantage reverses over time, and redo-SAVR seems to offer better survival at a later stage. Because these results are pooled data from observational studies, they should be interpreted with caution, and randomized controlled trials are warranted.

Self-Expanding Transcatheter Aortic Valves Optimize Transvalvular Hemodynamics Independent of Intra- Versus Supra-Annular Design.

This study sought to characterize transvalvular hemodynamics during the first 30 days after transcatheter aortic valve implantation (TAVI) across various transcatheter heart valves (THVs), while adjusting for annular dimensions. This was an observational study of TAVIs from September 2021 to October 2022. The primary outcome was mean transvalvular pressure gradient (TVPG), measured using transthoracic echocardiography at day 0, day 1, and day 30 post-TAVI, and were compared across 3 THV, including the self-expandable intra-annular Portico (Abbott Vascular, Santa Clara, California) valve, the balloon-expandable SAPIEN 3 Ultra (Edwards Lifesciences, Irvine, California), and the self-expandable supra-annular Evolut Pro+ (Medtronic, Minneapolis, Minnesota). A total of 560 patients who underwent TAVI were identified, of which 106 (18.9%) received a Portico THV, 176 (31.4%) received a SAPIEN THV, and 278 (49.6%) received an Evolut THV. For Portico THV, the TVPG on day 0 increased from 6.0 (4.7 to 9.0) to 7.0 (6.0 to 10.0) by day 30 (p = 0.009). For SAPIEN THV, the TVPG on day 0 increased from 6.5 (5.0 to 8.0) to 12.0 (9.0 to 15.0) by day 30 (p <0.001). For Evolut THV, the TVPG on day 0 increased from 6.0 (5.0 to 9.0) to 7.2 (5.0 to 10.0) by day 30 (p = 0.001). Adjusting for time and annular diameter in a multivariable mixed effects model, the SAPIEN group had a significantly greater increase in TVPG over time than the Evolut reference group (p <0.001), while there was no difference in the change of TVPG over time for the Portico group vs. the Evolut group (p = 0.874). In conclusion, compared with balloon-expandable valves, self-expanding THV may optimize transvalvular hemodynamics across all annular diameters, independent of their supra-annular and intra-annular design.

Valve-in-valve transcatheter aortic valve replacement versus isolated redo surgical aortic valve replacement.

OBJECTIVE: To compare outcomes of patients undergoing valve-in-valve transcatheter aortic valve replacement (ViV TAVR) versus redo surgical aortic valve replacement (SAVR). METHODS: This was a retrospective study using institutional databases of transcatheter (2013-2022) and surgical (2011-2022) aortic valve replacements. Patients who underwent ViV TAVR were compared with patients who underwent redo isolated SAVR. Clinical and echocardiographic outcomes were analyzed. Kaplan-Meier survival estimation and Cox regression were performed. Cumulative incidence functions were generated for heart failure readmissions. RESULTS: A total of 4200 TAVRs and 2306 isolated SAVRs were performed. Of these, there were 198 patients who underwent ViV TAVR and 147 patients who underwent redo SAVR. Operative mortality was 2% in each group, but observed to expected operative mortality in the redo SAVR group was higher than in the ViV TAVR group (1.2 vs 0.32). Those who underwent redo SAVR were more likely to require transfusions and reoperation for bleeding, to have new-onset renal failure requiring dialysis, and to require a permanent pacemaker postoperatively than those in the ViV group. Mean gradient was significantly lower in the redo SAVR group than in the ViV group at 30 days and 1 year. Kaplan-Meier survival estimates at 1 year were comparable, and on multivariable Cox regression, ViV TAVR was not significantly associated with an increased hazard of death compared with redo SAVR (hazard ratio, 1.39; 95% CI, 0.65-2.99; P = .40). Competing-risk cumulative incidence estimates for heart-failure readmissions were higher in the ViV cohort. CONCLUSIONS: ViV TAVR and redo SAVR were associated with comparable mortality. Patients who underwent redo SAVR had lower postoperative mean gradients and greater freedom from heart failure readmissions, but they also had more postoperative complications than the VIV group, despite their lower baseline risk profiles.

Acute Coronary Syndrome After Transcatheter Aortic Valve Implantation (Results from Over 40,000 Patients).

Acute coronary syndrome (ACS) encompasses a broad category of presentations from unstable angina to ST-elevation myocardial infarctions. Most patients undergo coronary angiography upon presentation for diagnosis and treatment. However, the ACS management strategy after transcatheter aortic valve implantation (TAVI) may be complicated because of challenging coronary access. The National Readmission Database was reviewed to identify all patients who were readmitted with ACS within 90 days after TAVI between 2012 and 2018. Their outcomes were described between patients who were readmitted with ACS (ACS group) and without (non-ACS group). A total of 44,653 patients were readmitted within 90 days after TAVI. Among them, 1,416 patients (3.2%) were readmitted with ACS. The ACS group had a higher prevalence of men, diabetes, hypertension, congestive heart failure, peripheral vascular disease, and a history of percutaneous coronary intervention (PCI). In the ACS group, 101 patients (7.1%) developed cardiogenic shock, whereas 120 patients (8.5%) developed ventricular arrhythmias. Overall, 141 patients (9.9%) in the ACS group died during readmissions (vs 3.0% in the non-ACS group, p <0.001). Among the ACS group, PCI was performed in 33 (5.9%), whereas coronary bypass grafting was performed in 12 (0.82%). The factors associated with ACS readmission included a history of diabetes, congestive heart failure, chronic kidney disease, and PCI, and nonelective TAVI. Coronary artery bypass grafting was an independent factor related to in-hospital mortality during ACS readmission (odds ratio 11.9, 95% confidence interval 2.18 to 65.4, p = 0.004), whereas PCI was not (odds ratio 0.19, 95% confidence interval 0.03 to 1.44, p = 0.11). In conclusion, patients readmitted with ACS have significantly higher mortality compared with those readmitted without ACS. History of PCI is an independent factor associated with ACS after TAVI.

Outcomes of Transcatheter Aortic Valve Replacement in Patients With Concomitant Aortic Regurgitation.

BACKGROUND: This study sought to evaluate outcomes of transcatheter aortic valve replacement (TAVR) in patients with moderate or greater aortic regurgitation (AR). METHODS: This was an observational study using an institutional database of TAVRs from November 2012 to April 2022. The study compared outcomes of TAVR in patients with isolated aortic stenosis (AS) vs patients with AS and concomitant AR (moderate or greater). Those patients with trace or mild AR were excluded. Clinical and echocardiographic outcomes were compared, with end points established by the Valve Academic Research Consortium 3. Kaplan-Meier survival estimation and Cox regression for mortality were performed. Competing-risk cumulative incidence estimates for heart failure readmissions were also compared. RESULTS: Of 3295 patients, 605 (53.4%) had severe AS with no AR and 529 (46.6%) had severe AS with moderate or severe AR. There were no significant differences in in-hospital mortality, length of stay, stroke, myocardial infarction, permanent pacemaker requirement, transfusion requirement, minor or major vascular complications, or 30-day readmissions between the 2 groups (P > .05). There were also no significant differences in annular dissection or rupture, coronary obstruction, or device embolization. Mean gradient and paravalvular leak rates at 30 days and 1 year were similar between the groups. Survival estimates were comparable, and, on multivariable Cox regression, mixed aortic valvular disease was not associated with an increased hazard of death as compared with isolated AS (hazard ratio, 1.01; 95% CI, 0.81-1.25; P = .962). Cumulative incidence estimates for heart failure readmissions were comparable between groups. CONCLUSIONS: TAVR can be safely performed in patients with mixed valvular disease, with outcomes comparable to those in isolated AS.

Coronary Artery Bypass With Multiarterial Grafting vs Percutaneous Coronary Intervention.

BACKGROUND: Data comparing patients who undergo multiarterial grafting during coronary artery bypass grafting (CABG) vs percutaneous coronary intervention (PCI) in patients with multivessel coronary disease are scarce. This study addresses the relevance of using multiple arterial conduits vs PCI for appropriate patients. METHODS: This retrospective study included all patients with coronary artery disease who underwent CABG with multiple arterial conduits or PCI. Propensity score matching was performed for baseline characteristics. Kaplan-Meier estimates, cumulative incidence, and freedom from major adverse cardiac and cerebrovascular events (MACCE) curves were performed. RESULTS: The total patient population consisted of 3648 patients from 2011 to 2018 divided into 902 CABG patients and 2746 PCI patients. Patients were propensity matched (PCI, n = 838; CABG, n = 838). In the CABG cohort the left internal mammary artery was used in 837 patients (99.9%), the right internal mammary artery in 770 patients (92%), and radial arteries in 108 patients (12.9%). Patients in the PCI cohort had significantly higher 30-day mortality (24 [2.9%] vs 7 [0.8%], P < .01). Survival over follow-up (median, 4.9 years; range, 3.3-6.8) was better for the CABG cohort (730 [87.1%] vs 625 [74.6%], P < .01). Patients in the CABG cohort had greater freedom from MACCE (607 [72.4%] vs 339 [40.5%], P < .01). Cox multivariable regression showed that patients who underwent CABG had a significantly reduced risk of mortality (hazard ratio, 0.49; 95% confidence interval, 0.39-0.61; P < .01) and of MACCE (hazard ratio, 0.33; 95% confidence interval, 0.28-0.38; P < .01). CONCLUSIONS: Patients with coronary artery disease who undergo CABG with multiple arterial conduits have significantly fewer major adverse events, improved survival, and reduced hospital readmissions.

Readmission with infective endocarditis within 90 days following transcatheter aortic valve implantation.

INTRODUCTION: Transcatheter aortic valve implantation (TAVI) continues to be the most common modality of treating aortic stenosis in the United States. While infective endocarditis (IE) and its outcomes have been well documented after surgical aortic valve replacement, the incidence and outcomes of early IE after TAVI have not been well described. METHODS: All patients who underwent TAVI from 2012 through 2018 were identified using the National Readmission Database. Among them, patients who underwent TAVI at the index admission and readmitted within 90 days were included. Patients who died or had IE during the index admission were excluded. Clinical outcomes were compared between patients readmitted with IE (IE group) and those without (non-IE group). RESULTS: A total of 168,283 patients were readmitted to a hospital within 90 days after TAVI. The median age of the IE group and non-IE group were 81 and 82 years old, respectively (p = 0.21). Of those, 525 (0.3%) were readmitted with IE. The median time from TAVI to readmission was 20 days. During readmissions, 11.6% of the IE group died while only 3.15% of the non-IE group experienced death (p < 0.001). The most common causative organism of IE was enterococcus (22.1%). Multivariable analysis revealed that congestive heart failure, cerebrovascular disease, dialysis, concomitant valve disease, Medicaid, and discharge to a facility were independently associated with readmission with IE within 90 days. CONCLUSION: The incidence of readmission with IE is low after TAVI. However, the mortality was markedly high during readmissions. Surgical intervention was rarely performed for IE during the first admission. Enterococcus was the most common organism observed in IE after TAVI. DISCLOSURE: IS receives institutional research support from Abbott, Atricure, cryolife, and Medtronic. None related to this manuscript. CLINICAL TRIAL REGISTRATION: Not applicable.

Current Status of Transcatheter Tricuspid Valve Therapies.

Tricuspid regurgitation is a complex disease that carries a poor prognosis, and surgical repair is associated with high mortality. In light of the success of other transcatheter-based valve interventions, transcatheter tricuspid therapy has recently seen exponential use both clinically and in innovation. Given the rapid development of many tricuspid systems and multiple on-going clinical trials, the aim of this review is to highlight the current state of transcatheter tricuspid therapeutics and to provide an up-to-date view of their clinical use, outcomes and future directions.

Outcomes of Surgical Aortic Valve Replacement After Transcatheter Aortic Valve Implantation.

Few studies have reported their experience in explantation of a transcatheter heart valve. We found 2,100 patients who underwent transcatheter aortic valve implantation (TAVI) from 2013 through 2021. Of 2,100, a total of 17 patients underwent surgical aortic valve replacement after TAVI, including surgical bailout. The mean age was 78.3 years. Co-morbidities were very frequent, including coronary artery disease (70.6%), atrial fibrillation (52.9%), cerebrovascular disease (47.1%), and pulmonary hypertension (41.2%). A history of cardiac surgery was observed in 6 patients (35.3%). The mean predicted risk of mortality at the time of TAVI was 7.7%. Surgical bailout was the most common indication of valve explantation (n = 8, 47.1%), followed by infective endocarditis (n = 4, 23.5%) and paravalvular leak (n = 2, 11.8%). The valve-in-valve TAVI was not feasible because of endocarditis, paravalvular leak, and history of valve-in-valve TAVI. Overall, 13 (76.5%) were performed urgently or emergently, and 10 (58.9%) required aortic root reconstruction. The mean cardiopulmonary bypass time was 158.5 minutes. In-hospital mortality was 41.2%. Transcatheter heart valve explantation continues to be rare; however, these data will continue to be informative as TAVI explantations will become more common with time.

Outcomes of transcatheter aortic valve replacement at teaching versus nonteaching hospitals.

INTRODUCTION: Prior studies have demonstrated that outcomes of invasive cardiac interventions may vary by hospital teaching status and volume. As transcatheter aortic valve replacement (TAVR) rapidly expands from teaching to nonteaching hospitals across the country, the clinical impact of hospital teaching status has not been clearly established. This study aimed to compare TAVR outcomes between nonteaching and teaching hospitals. METHODS: An observational study was conducted using the Nationwide Readmission Database (NRD). Patients undergoing TAVR from 2011 to 2018 were included. Data was analyzed using multivariable logistic regression to determine outcomes of 30-day mortality and readmission between nonteaching and teaching hospitals. RESULTS: A total of 235,321 patients underwent TAVR during the study period. Patients undergoing TAVR at teaching hospitals presented with a higher frequency of baseline comorbidities compared to nonteaching hospitals. Postprocedure complications such as myocardial infarction, arrhythmia, pneumonia, acute kidney injury, sepsis, stroke, and hemorrhage occurred more often at teaching centers (p < 0.001); translating to a higher rate of in-hospital mortality (2.27% vs. 1.99%, p = 0.006) and hospital cost ($48,300 vs. $44,900, p < 0.001) in teaching versus nonteaching hospitals. After adjusting for baseline characteristics and postoperative morbidity, in-hospital mortality (p = 0.095) and readmission rate (p = 0.420) on multivariable analysis were not statistically different between centers. CONCLUSION: With the evolution and expansion of TAVR to nonteaching centers, mortality, and readmission rates are not significantly different between nonteaching and teaching hospitals. Higher unadjusted in-hospital mortality at teaching centers suggest these centers more often treat high risk patients with associated increased complications.

Impact of Transcatheter Aortic Valve Implantation on Severe Gastrointestinal Bleeding in Patients With Aortic Stenosis.

To investigate the effectiveness of transcatheter aortic valve implantation (TAVI) for resolving gastrointestinal bleeding (GIB) in patients with severe GIB and severe aortic stenosis. This was an observational study of consecutive TAVI procedures from 2011 to 2018, identified through a prospectively maintained institutional database. Patients with severe GIB defined as abnormal hemoglobin/hematocrit and overt bleeding or positive fecal occult blood test. Groups were compared for differences in baseline clinical and procedural characteristics and post-TAVI outcomes. A total of 1,192 patients who underwent TAVI were identified, of which 164 patients (13.8%) were found to have had severe GIB. Society of Thoracic Surgeons-Predicted Risk of Mortality scores were higher in the severe GIB group than in the non-GIB group (8.8 ± 5.3 vs 7.6 ± 4.5, p = 0.002). The 30-day mortality was similar across each group (p >0.05), with overall 30-day mortality being 3.2% (38 of 1,192) for the entire cohort. Of the 164 TAVIs with severe GIB, 130 (79.3%) had resolution of their GIB after their TAVI. Patients without resolution of their GIB had higher aortic valve mean pressure gradients after their TAVI (15.0 ± 5.3 vs 9.0 ± 4.3). In conclusion, TAVI was associated with substantial reductions in severe GIB. Despite higher baseline risk, patients with severe GIB achieved similar outcomes, including 30-day mortality, compared with patients without GIB. TAVI may be a safe and effective treatment for patients with severe aortic stenosis and severe GIB, whereas persistently high transvalvular pressure gradients after TAVI may be related to the likelihood of recurrent GIB.

Long-term Outcomes of Percutaneous Coronary Intervention in Patients with Prior Coronary Artery Bypass Grafting.

BACKGROUND Patients with a prior coronary artery bypass graft (CABG) may have a need for repeat revascularization, which is typically attempted first via percutaneous coronary intervention (PCI) of either a bypass graft or native vessel. Long-term outcomes of native vessel compared to graft PCI after CABG have not yet been explored in a large institution study. METHODS Patients with history of prior CABG who underwent PCI at our institution during 2010-2018 were included. Baseline characteristics and long-term outcomes of up to 5 years were compared between native vessel and bypass graft PCI groups. Cox regression was used to adjust for significant covariates in estimation of risk and calculation of hazard ratios. RESULTS During the study, 4,251 patients with a prior CABG underwent PCI. Native vessel PCI represented 67.1% (n=2,851) of the cohort. After adjusting for significant covariates, bypass graft PCI compared to native vessel PCI had a higher risk of overall mortality (HR 1.15; 95% CI, 1.04-1.29; p<0.05), all-cause readmission (HR 1.16; 95% CI, 1.1-1.3; p<0.05), readmission for PCI (HR 1.25; 95% CI, 1.13-1.38; p<0.05), readmission for heart failure (HR 1.16; 95% CI, 1.06-1.26; p<0.05), and composite of myocardial infarction and revascularization (HR 1.23; 95% CI, 1.12-1.35; p<0.05). CONCLUSIONS Among patients with prior CABG, bypass graft PCI compared to native vessel PCI was associated with higher risk of adverse long-term outcomes.

Transfemoral Versus Subclavian Access for Transcatheter Aortic Valve Replacement.

Objective: This study sought to compare outcomes of transcatheter aortic valve replacement (TAVR) performed through subclavian access with those performed through transfemoral access. Methods: This was an observational study utilizing an institutional TAVR database from 2010 to 2018. All patients undergoing a TAVR via a transfemoral (TF-TAVR) or subclavian (SC-TAVR) approach were included in the study. The groups were analyzed for differences in operative mortality and postoperative outcomes. Multivariable Cox analysis was performed to identify variables associated with long-term survival after TAVR. Results: Of the 1,095 patients identified, 133 patients underwent SC-TAVR and 962 patients underwent TF-TAVR. Patients who underwent SC-TAVR were younger, more likely to have chronic lung disease and peripheral vascular disease, had higher Society of Thoracic Surgeons predicted risk of mortality scores, and were more likely to have self-expanding valves placed (P < 0.05). Operative mortality was similar between the TF-TAVR (2.7%) and SC-TAVR (3.8%) groups. There were no significant differences in stroke, length of stay, 30-day readmission, blood transfusions, acute kidney injury, need for permanent pacemaker, paravalvular leak, or major vascular complications between the groups (P > 0.05). The unadjusted Kaplan-Meier survival estimate for TF-TAVR was significantly higher than for SC-TAVR (P = 0.009, log-rank). However, on multivariable Cox analysis, subclavian access was not significantly associated with an increased hazard of death as compared with transfemoral access (P = 0.21). Conclusions: Outcomes of SC-TAVR are comparable to those of TF-TAVR. Subclavian access may be a favorable alternative approach when TF-TAVR is contraindicated.

Long term mortality and readmissions after transcatheter aortic valve replacement.

BACKGROUND: Readmissions following transcatheter aortic valve replacement (TAVR) are common but detailed analysis of cardiac and non-cardiac inpatient readmissions beyond thirty days to different levels of care are limited. METHODS: Our study population was 1,037 consecutive patients who underwent TAVR between 2011-2017 within a multi-hospital quaternary health system. A retrospective chart review was performed and readmissions were adjudicated and classified based on primary readmission diagnosis (cardiac versus noncardiac) and level of care [intensive care unit (ICU) admission vs. non-ICU admission]. Incidence, causes, and outcomes of readmissions to up to three years post procedure were evaluated. RESULTS: Of the 1,017 patients who survived their index hospitalization, there were readmissions due to noncardiac causes in 350 (34.4%) and cardiac causes in 208 (20.5%) during a mean 1.96 years of follow-up. The most common non-cardiac causes of readmission were sepsis/infection (14.3%), gastrointestinal (8.3%), and respiratory (4.8%), whereas heart failure (14.0%) and arrhythmias (4.6%) were the most common cardiac causes of readmission. A total of 191 (18.8%) patients were readmitted to the ICU and 372 patients (36.6%) were non-ICU readmissions. The risk of a noncardiac readmission was highest in the period immediately following TAVR (~4.5% per month) with an early high hazard phase that gradually declined over months. However, the risk of cardiac readmission remained stable at ~1% per month throughout. TAVR patients that were readmitted for any cause had markedly increased mortality; this was especially true for patients readmitted to an ICU. CONCLUSIONS: In TAVR patients who survived their index hospitalization, non-cardiac readmissions were more prevalent than cardiac. The risk of readmission and subsequent mortality was highest immediately post-procedure and declined thereafter. Readmission to ICU portends the highest risk of subsequent death in this cohort. Patient baseline co-morbidities are an important consideration for TAVR patients and play a significant role in readmissions and outcomes.

Urgent transcatheter aortic valve replacement may be performed with acceptable long-term outcomes.

BACKGROUND: The wide availability of transcatheter aortic valve replacement (TAVR) and broadening of its indications to most patients with aortic stenosis may increase its utilization in the urgent setting. However, a comparison of long-term outcomes of patients undergoing urgent TAVR when compared to elective TAVR have not been well studied. METHODS: All patients that underwent TAVR from 2011 to 2018 were included. Primary outcomes included operative (30-day), 1-, and 5-year survival and readmissions. RESULTS: The total patient population undergoing TAVR was divided into urgent (n = 247) and elective (n = 946) cohorts. Thirty days mortality (6.5% vs. 2.3%; p = .001), acute kidney injury (2.8% vs. 0.6%; p = .003), and length of stay (12 vs. 3 days; p < .001) were higher for the urgent cohort. There was no significant difference between cohorts for 30-day all-cause (14.6% vs. 10.8%; p = .097) readmissions. Freedom from readmission for heart failure at 1-year (73.6% vs. 83.4%; p < .001) was lower for the urgent cohort. One- (79.0% vs. 87.1%; p < .001) and five-year (39.6% vs. 43.5%; p = .005) survival was lower for the urgent cohort. This difference was eliminated after risk adjustment (hazard ratio [HR]: 1.3; p = .158 and HR: 1.1; p = .639, respectively). CONCLUSION: Unadjusted survival was significantly worse for the urgent cohort up to 1 year. This trend continued for 5-year survival, however, after risk adjustment there was no significant difference in survival between cohorts. Although urgent TAVR is associated with increased periprocedural risk due to more comorbid disease, outcomes and long-term survival are encouraging and support the consideration of urgent TAVR as a viable alternative for this patient population.

Longitudinal Outcomes of Nonagenarians Undergoing Transcatheter Aortic Valve Replacement.

BACKGROUND: Transcatheter aortic valve replacement (TAVR) has emerged as the preferred alternative to surgical aortic valve replacement in elderly patients. However, the long-term outcomes of nonagenarians undergoing TAVR are unknown. METHODS: Octogenarian and nonagenarian patients undergoing TAVR from 2011 to 2018 were identified from a prospectively maintained institutional database. Cox proportional hazards regression was used for baseline-adjusted outcome comparison and risk prediction. Survival was compared with age and gender-matched population from the Social Security Actuarial Life Table. RESULTS: A total of 649 (54.4%) octogenarians and 157 (13.2%) nonagenarians underwent TAVR. Nonagenarians had a lower body mass index (P < .001), smaller BSA (P < .001), and a lower prevalence of chronic obstructive pulmonary disease (P = .023) but a higher Society of Thoracic Surgeons score (P < .001). The majority of nonagenarians and octogenarians were treated using self-expandable valves (60.3% vs 60.9%; P = .888) via transfemoral access (86.0% vs 81.0%; P = .148). At 30 days, 1 year, and 4 years, there was no difference in survival (95.5%, 80.3%, and 51.2% vs 96.9%, 87.4, and 57.6%, respectively) (adjusted hazard ratio [HR], 0.8; P = .205) and hospital readmissions for cardiac causes (7.9%, 25.7%, and 53.7% vs 10.3%, 27.9%, and 52.0%, respectively) (adjusted HR, 0.9; P = .488). Further, nonagenarians had a survival comparable to an age-matched and sex-matched U.S. population (P = .540). Post-TAVR paravalvular leak (HRs: 3.23 [P = .042] vs 2.66 [P = .032]) and anemia (HRs: 0.64 [P = .002] vs 0.80 [P = .004]) were associated with worse outcomes at 1 year. CONCLUSIONS: TAVR can be performed safely in nonagenarians, with comparable outcomes to younger patients approximating natural life expectancy. This age paradox should strengthen the role of TAVR in well selected nonagenarians by the heart team.

Percutaneous coronary intervention versus coronary artery bypass grafting in patients with reduced ejection fraction.

OBJECTIVE: The aim of this study was to evaluate comparative outcomes for percutaneous coronary intervention (PCI) versus coronary artery bypass grafting (CABG) in patients with reduced ejection fraction. METHODS: All patients from the University of Pittsburgh Medical Center from 2011 to 2018 who had reduced preoperative ejection fraction (<50%) and underwent CABG or PCI for coronary revascularization were included in this study. Patients were risk-adjusted with propensity matching (1:1) and primary outcomes included long-term survival, readmission, and major adverse cardiac and cerebrovascular events (MACCE). RESULTS: A total of 2000 patients were included in the current study, consisting of CABG (n = 1553) and PCI (n = 447) cohorts with a mean ejection fraction of 35% ± 9.53%. Propensity matching yielded a 1:1 match with 324 patients in each cohort, controlling for all baseline characteristics. Thirty-day mortality was similar for PCI versus CABG (6.2% vs 4.9%; P = .49). Overall mortality over the study follow-up period (median, 3.23 years; range, 1.83-4.98 years) was significantly higher for the PCI cohort (37.4% vs 21.3%; P < .001). Total hospital readmissions (24.1% vs 12.9%; P = .001), cardiac readmissions (20.4% vs 11.1%; P = .001), myocardial infarction event (7.7% vs 1.8%; P = .001), MACCE (41.4% vs 23.8%; P < .001), and repeat revascularization (6.5% vs 2.6%; P = .02) occurred more frequently in the PCI cohort. Freedom from MACCE at 1 year (74.4% vs 87.0%; P < .001) and 5 years (54.5% vs 74.0%; P < .001) was significantly lower for the PCI cohort. On multivariable cox regression analysis, CABG (hazard ratio, 0.57; 95% confidence interval, 0.44-0.73; P < .001) was significantly associated with improved survival. Prior liver disease, dialysis, diabetes, and peripheral artery disease were the most significant predictors of mortality. The cumulative incidence of hospital readmission was lower for the CABG cohort (hazard ratio, 0.51; 95% confidence interval, 0.37-0.71; P < .001). Multivariable cox regression for MACCE (hazard ratio, 0.48; 95% confidence interval, 0.39-0.58; P < .001) showed significantly fewer events for the CABG cohort. CONCLUSIONS: Patients with reduced ejection fraction who underwent CABG had significantly improved survival, lower MACCE, and fewer repeat revascularization procedures compared with patients who underwent PCI.