Assessing calibration and bias of a deployed machine learning malnutrition prediction model within a large healthcare system.

Malnutrition is a frequently underdiagnosed condition leading to increased morbidity, mortality, and healthcare costs. The Mount Sinai Health System (MSHS) deployed a machine learning model (MUST-Plus) to detect malnutrition upon hospital admission. However, in diverse patient groups, a poorly calibrated model may lead to misdiagnosis, exacerbating health care disparities. We explored the model's calibration across different variables and methods to improve calibration. Data from adult patients admitted to five MSHS hospitals from January 1, 2021 - December 31, 2022, were analyzed. We compared MUST-Plus prediction to the registered dietitian's formal assessment. Hierarchical calibration was assessed and compared between the recalibration sample (N = 49,562) of patients admitted between January 1, 2021 - December 31, 2022, and the hold-out sample (N = 17,278) of patients admitted between January 1, 2023 - September 30, 2023. Statistical differences in calibration metrics were tested using bootstrapping with replacement. Before recalibration, the overall model calibration intercept was -1.17 (95% CI: -1.20, -1.14), slope was 1.37 (95% CI: 1.34, 1.40), and Brier score was 0.26 (95% CI: 0.25, 0.26). Both weak and moderate measures of calibration were significantly different between White and Black patients and between male and female patients. Logistic recalibration significantly improved calibration of the model across race and gender in the hold-out sample. The original MUST-Plus model showed significant differences in calibration between White vs. Black patients. It also overestimated malnutrition in females compared to males. Logistic recalibration effectively reduced miscalibration across all patient subgroups. Continual monitoring and timely recalibration can improve model accuracy.

Transcatheter or Surgical Replacement for Failed Bioprosthetic Aortic Valves.

Importance: The use of valve-in-valve (ViV) transcatheter aortic valve replacement (TAVR) has been rapidly expanding as an alternative treatment to redo surgical aortic valve replacement (SAVR) for failed bioprosthetic valves despite limited long-term data. Objective: To assess mortality and morbidity in patients undergoing intervention for failed bioprosthetic SAVR. Design, Setting, and Participants: This was a retrospective population-based cohort analysis conducted between January 1, 2015, and December 31, 2020, with a median (IQR) follow-up time of 2.3 (1.1-4.0) years. A total of 1771 patients with a history of bioprosthetic SAVR who underwent ViV-TAVR or redo SAVR in California, New York, and New Jersey were included. Data were obtained from the California Department of Health Care Access and Information, the New York Statewide Planning and Research Cooperative System, and the New Jersey Discharge Data Collection System. Exclusion criteria included undergoing TAVR or redo SAVR within 5 years from initial SAVR, as well as infective endocarditis, concomitant surgical procedures, and out-of-state residency. Propensity matching yielded 375 patient pairs. Data were analyzed from January to December 2023. Interventions: ViV-TAVR vs redo SAVR. Main Outcomes and Measurements: The primary outcome was all-cause mortality. Secondary outcomes were stroke, heart failure hospitalization, reoperation, major bleeding, acute kidney failure, new pacemaker insertion, and infective endocarditis. Results: From 2015 through 2020, the proportion of patients undergoing ViV-TAVR vs redo SAVR increased from 159 of 451 (35.3%) to 498 or 797 (62.5%). Of 1771 participants, 653 (36.9%) were female, and the mean (SD) age was 74.4 (11.3) years. Periprocedural mortality and stroke rates were similar between propensity-matched groups. The ViV-TAVR group had lower periprocedural rates of major bleeding (2.4% vs 5.1%; P = .05), acute kidney failure (1.3% vs 7.2%; P < .001), and new pacemaker implantations (3.5% vs 10.9%; P < .001). The 5-year all-cause mortality rate was 23.4% (95% CI, 15.7-34.1) in the ViV-TAVR group and 13.3% (95% CI, 9.2-18.9) in the redo SAVR group. In a landmark analysis, no difference in mortality was observed up to 2 years (hazard ratio, 1.03; 95% CI, 0.59-1.78), but after 2 years, ViV-TAVR was associated with higher mortality (hazard ratio, 2.97; 95% CI, 1.18-7.47) as well as with a higher incidence of heart failure hospitalization (hazard ratio, 3.81; 95% CI, 1.57-9.22). There were no differences in 5-year incidence of stroke, reoperation, major bleeding, or infective endocarditis. Conclusions and Relevance: Compared with redo SAVR, ViV-TAVR was associated with a lower incidence of periprocedural complications and a similar incidence of all-cause mortality through 2 years' follow-up. However, ViV-TAVR was associated with higher rates of late mortality and heart failure hospitalization. These findings may be influenced by residual confounding and require adjudication in a randomized clinical trial.

Maternal Characteristics and Rates of Unexpected Complications in Term Newborns by Hospital.

Importance: The Joint Commission Unexpected Complications in Term Newborns measure characterizes newborn morbidity potentially associated with quality of labor and delivery care. Infant exclusions isolate relatively low-risk births, but unexpected newborn complications (UNCs) are not adjusted for maternal factors that may be associated with outcomes independently of hospital quality. Objective: To investigate the association between maternal characteristics and hospital UNC rates. Design, Setting, and Participants: This cohort study was conducted using linked 2016 to 2018 New York City birth and hospital discharge datasets among 254 259 neonates at low risk (singleton, ≥37 weeks, birthweight ≥2500 g, and without preexisting fetal conditions) at 39 hospitals. Logistic regression was used to calculate unadjusted hospital-specific UNC rates and replicated analyses adjusting for maternal covariates. Hospitals were categorized into UNC quintiles; changes in quintile ranking with maternal adjustment were examined. Data analyses were performed from December 2022 to July 2023. Main Outcomes and Measures: UNCs were classified according to Joint Commission International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) criteria. Maternal preadmission comorbidities, obstetric factors, social characteristics, and hospital characteristics were ascertained. Results: Among 254 259 singleton births at 37 weeks or later who were at low risk (125 245 female [49.3%] and 129 014 male [50.7%]; 71 768 births [28.2%] to Hispanic, 47 226 births [18.7%] to non-Hispanic Asian, 42 682 births [16.8%] to non-Hispanic Black, and 89 845 births [35.3%] to non-Hispanic White mothers and 2738 births [1.0%] to mothers with another race or ethnicity), 148 393 births (58.4%) were covered by Medicaid and 101 633 births (40.0%) were covered by commercial insurance. The 2016 to 2018 cumulative UNC incidence in New York City hospitals was 37.1 UNCs per 1000 births. Infants of mothers with preadmission risk factors had increased UNC risk; for example, among mothers with vs without preeclampsia, there were 104.4 and 35.8 UNCs per 1000 births, respectively. Among hospitals, unadjusted UNC rates ranged from 15.6 to 215.5 UNCs per 1000 births and adjusted UNC rates ranged from 15.6 to 194.0 UNCs per 1000 births (median [IQR] change from adjustment, 1.4 [-4.7 to 1.0] UNCs/1000 births). The median (IQR) change per 1000 births for adjusted vs unadjusted rates showed that hospitals with low (<601 deliveries/year; -2.8 [-7.0 to -1.6] UNCs) to medium (601 to <954 deliveries/year; -3.9 [-7.1 to -1.9] UNCs) delivery volume, public ownership (-3.6 [-6.2 to -2.3] UNCs), or high proportions of Medicaid-insured (eg, ≥90.72%; -3.7 [-5.3 to -1.9] UNCs), Black (eg, ≥32.83%; -5.3 [-9.1 to -2.2] UNCs), or Hispanic (eg, ≥6.25%; -3.7 [-5.3 to -1.9] UNCs) patients had significantly decreased UNC rates after adjustment, while rates increased or did not change in hospitals with the highest delivery volume, private ownership, or births to predominantly White or privately insured individuals. Among all 39 hospitals, 7 hospitals (17.9%) shifted 1 quintile comparing risk-adjusted with unadjusted quintile rankings. Conclusions and Relevance: In this study, adjustment for maternal case mix was associated with small overall changes in hospital UNC rates. These changes were associated with performance assessment for some hospitals, and these results suggest that profiling on this measure should consider the implications of small changes in rates for hospitals with higher-risk obstetric populations.

Permanent Pacemaker Implantation and Long-Term Outcomes of Patients Undergoing Concomitant Mitral and Tricuspid Valve Surgery.

BACKGROUND: Tricuspid valve annuloplasty (TA) during mitral valve repair (MVr) is associated with increased risk of permanent pacemaker (PPM) implantation, but the magnitude of risk and long-term clinical consequences have not been firmly established. OBJECTIVES: This study assesses the incidence rates of PPM implantation after isolated MVr and following MVr with TA as well as the associated long-term clinical consequences of PPM implantation. METHODS: State-mandated hospital discharge databases of New York and California were queried for patients undergoing MVr (isolated or with concomitant TA) between 2004 and 2019. Patients were stratified by whether or not they received a PPM within 90 days of index surgery. After weighting by propensity score, survival, heart failure hospitalizations (HFHs), endocarditis, stroke, and reoperation were compared between patients with or without PPM. RESULTS: A total of 32,736 patients underwent isolated MVr (n = 28,003) or MVr + TA (n = 4,733). Annual MVr + TA volumes increased throughout the study period (P < 0.001, trend), and PPM rates decreased (P < 0.001, trend). The incidence of PPM implantation <90 days after surgery was 7.7% for MVr and 14.0% for MVr + TA. In 90-day conditional landmark-weighted analyses, PPMs were associated with reduced long-term survival among MVr (HR: 1.96; 95% CI: 1.75-2.19; P < 0.001) and MVr + TA recipients (HR: 1.65; 95% CI: 1.28-2.14; P < 0.001). In both surgical groups, PPMs were also associated with an increased risk of HFH (HR: 1.56; 95% CI: 1.27-1.90; P < 0.001) and endocarditis (HR: 1.95; 95% CI: 1.52-2.51; P < 0.001), but not with stroke or reoperation. CONCLUSIONS: Compared to isolated MVr, adding TA to MVr was associated with a higher risk of 90-day PPM implantation. In both surgical groups, PPM implantation was associated with an increase in mortality, HFH, and endocarditis.

Choice of revascularization strategy for ischemic cardiomyopathy due to multivessel coronary disease.

OBJECTIVE: Limited comparative data guide the decision between coronary artery bypass grafting and percutaneous coronary intervention for multivessel revascularization in ischemic cardiomyopathy. The study objective was to compare the long-term outcomes of coronary artery bypass grafting and percutaneous coronary intervention for ischemic cardiomyopathy. METHODS: Clinical registries from the New Jersey Department of Health linked to administrative databases were used to compare all-cause mortality, repeat revascularization, heart failure readmissions, myocardial infarction, and stroke using Cox proportional hazards and propensity matching with competing risk analysis in 5988 patients with ejection fraction 35% or less who underwent coronary artery bypass grafting (3673, 61.3%) or percutaneous coronary intervention (2315, 38.6%) for multivessel coronary disease between 2007 and 2018. Median follow-up time was 5.2 years (range, 0-13 years); the last follow-up date was December 31, 2020. RESULTS: After controlling for completeness of revascularization, at 13 years, mortality was 57% (95% CI, 51-63) after percutaneous coronary intervention and 60% (95% CI, 53-66) after coronary artery bypass grafting (hazard ratio [HR], 1.10; 95% CI, 0.93-1.31; P = .28); risk of repeat revascularization was 18% for percutaneous coronary intervention versus 14% for coronary artery bypass grafting (HR, 1.62; 95% CI, 1.17-2.25; P = .003); risk of readmission for heart failure was 16% after percutaneous coronary intervention and coronary artery bypass grafting (HR, 1.13,95% CI, 0.84-1.51, weighted P = .10); risk of myocardial infarction was 10% versus 6%, respectively (HR, 1.91; 95% CI, 1.18-3.09; P = .007); and stroke risk was 3% versus 4%, respectively (HR, 0.79; 95% CI, 0.41-1.53; P = .52). Rate of complete revascularization was lower after percutaneous coronary intervention than after coronary artery bypass grafting and associated with higher mortality after percutaneous coronary intervention (HR, 1.35; 95% CI, 1.20-1.52; P < .001). CONCLUSIONS: Coronary bypass was associated with similar mortality, stroke, and heart failure readmissions, and reduced repeat revascularization compared with percutaneous coronary intervention in patients with ischemic cardiomyopathy if similar rates of complete revascularization were achieved. These findings support consensus recommendations for coronary artery bypass grafting and medical therapy in patients with multivessel coronary disease and left ventricular dysfunction.

Minimally invasive procedures for right side infective endocarditis: A targeted literature review.

INTRODUCTION: Right-side infective endocarditis (RSIE) is caused by microorganisms and develops into intracardiac and extracardiac complications with high in-hospital and 1-year mortality. Treatments involve antibiotic and surgical intervention. However, those presenting with extremes e.g. heart failure, or septic shock who are not ideal candidates for conventional medical therapy might benefit from minimally invasive procedures. OBJECTIVE: This review summarizes existing observational studies that reported minimally invasive procedures to debulk vegetation due to infective endocarditis either on valve or cardiac implantable electronic devices. METHODS: A targeted literature review was conducted to identify studies published in PubMed/MEDLINE, EMBASE, and Cochrane Central Database from January 1, 2015 to June 5, 2023. The efficacy and/or effectiveness of minimally invasive procedural interventions to debulk vegetation due to RSIE were summarized following PRISMA guidelines. RESULTS: A total of 11 studies with 208 RSIE patients were included. There were 9 studies that assessed the effectiveness of the AngioVac system and 2 assessed the Penumbra system. Overall procedure success rate was 87.9%. Among 8 studies that reported index hospitalization, 4 studies reported no death, while the other 4 studies reported 10 deaths. CONCLUSIONS: This study demonstrates that multiple systems can provide minimally invasive procedure options for patients with RSIE with high procedural success. However, there are mixed results regarding complications and mortality rates. Further large cohort studies or randomized clinical trials are warranted to assess and/or compare the efficacy and safety of these systems.

Temporal Trend in Revascularization for Patients With Ischemic Cardiomyopathy and Multivessel Coronary Artery Disease.

BACKGROUND: Current guidelines recommend revascularization in patients with ischemic cardiomyopathy (ICM). However, there is limited information about the trends and outcomes of coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI) in ischemic cardiomyopathy patients with multivessel coronary artery disease. METHODS AND RESULTS: Using New Jersey state mandatory registries, 8083 patients with ischemic cardiomyopathy with CABG or PCI revascularization for multivessel coronary artery disease from 2007 to 2018 were included in the analysis. Joinpoint regression and multivariable logistic regression analyses were performed to assess the annual percentage change in trends and predictors of the 30-day mortality rate, respectively. A decline in CABG procedures was observed from 2007 to 2011 (annual percentage change, -11.5%; P=0.003), followed by stabilization. The PCI trend remained unchanged from 2007 to 2010 and then increased significantly (annual percentage change, 3.2%; P=0.02). In the subsample of patients with proximal left anterior descending artery plus circumflex and right coronary artery, CABG was a predominant procedure until 2011, and the proportion of both procedures did not differ thereafter. In the subsample of patients with left anterior descending artery and any other artery stenosis, PCI remained dominant from 2007 to 2018, while in patients with left main and any other artery stenosis, CABG remained dominant from 2007 to 2018 (P<0.001). The 30-day risk-adjusted mortality rate was higher after PCI versus CABG for each year, but after adjustment for completeness of revascularization, there was no difference between groups. CONCLUSIONS: The patterns of revascularization procedures for patients with ischemic cardiomyopathy with multivessel coronary artery disease have changed over the years, as evidenced by the changes in CABG and PCI trends. CABG and PCI had comparable 30-day risk-adjusted mortality risks.

Incidence, Outcomes, and Opportunity for Left Ventricular Assist Device Weaning for Myocardial Recovery.

BACKGROUND: Myocardial recovery occurs in patients with advanced heart failure on left ventricular assist device (LVAD) support, but there is the premise that it is rare with uncertain results. OBJECTIVES: The goal of this study was to investigate the incidence and consequence of LVAD explant after myocardial recovery. METHODS: Using the United Network for Organ Sharing registry, LVAD implants in the United States between 2005 and 2020 were tracked until death, transplantation, or explant for myocardial recovery. The cohort undergoing explant was followed up for heart failure relapse (defined as relisting followed by delisting due to death, being too ill, or transplantation; or second durable LVAD implant). RESULTS: Of 15,728 LVAD implants, 126 patients underwent explant for recovery, which only occurred in 55 (38%) of 145 implanting centers. The crude cumulative incidence was 0.7% at 2 years, whereas the incidence reached 4.7% among designated centers in the selected young nonischemic cohort. Of 126 explanted patients, 76 (60%) were subsequently delisted for sustained recovery. Heart failure relapsing had a relatively higher hazard in the early phase, with a 30-day incidence of 6% (7 of 126) but tapered following with the freedom rate of 72.5% at 4 years. CONCLUSIONS: In the United States, LVAD explant for myocardial recovery was underutilized, leading to a very low incidence at the national level despite a realistic rate being achieved in designated centers for selected patients. With follow-up extending up to 4 years after explant, more than one-half were successfully removed and stayed off the waitlist, and approximately 70% were free from heart failure relapse events.

Surgery and transcatheter intervention for degenerative mitral regurgitation in the United States.

OBJECTIVES: We evaluated practice trends and 3-year outcomes of transcatheter edge-to-edge repair (TEER) and surgical repair for degenerative mitral regurgitation in the United States. METHODS: From the Centers for Medicare and Medicaid Services data (2012-2019), 53,117 mitral valve interventions (surgery or TEER) were performed for degenerative mitral regurgitation, identified by excluding rheumatic and congenital disease, endocarditis, myocardial infarction, cardiomyopathy, and concomitant or prior coronary revascularizations. Median follow-up was 2.9 years (interquartile range, 1.2-5.1 years). End points were 3-year survival, stroke, mitral reinterventions, and heart failure readmissions. RESULTS: Volume of total annual mitral interventions did not significantly change (P = .18) between 2012 and 2019. However, surgical cases decreased by one-third, whereas TEER increased. Among 27,170 patients (52.5% men; mean age, 73.5 years) who underwent TEER (n = 7755) or surgical repair (n = 19,415), surgical patients were younger (71.8 vs 80.8 years; P < .001), with less comorbidity and frailty. In 4532 patient pairs matched for age, frailty, and comorbidity, 3-year survival after TEER was 65.9% (95% CI, 64.3%-67.6%) and 85.7% (95% CI, 84.5%-86.9%) after surgery (P < .001). Three years after TEER or surgery, stroke rates were 1.8% (95% CI, 1.5%-2.2%) and 2.0% (95% CI, 1.6%-2.4%) (P = .49); heart failure readmission rates were 17.8% (95% CI, 16.7%-18.9%) and 11.2% (95% CI, 10.3%-12.2%) (P < .001); and mitral reintervention rates were 6.1% (95% CI, 5.5%-6.9%) and 1.3% (95% CI, 1.0%-1.7%) (P < .001), respectively. CONCLUSIONS: Among Medicare beneficiaries with degenerative mitral regurgitation, an increase in TEER utilization was associated with worse survival, increased heart failure readmissions, and more mitral reinterventions. Randomized trials are needed to better inform treatment choice.

Three reasons why parental burnout is more prevalent in individualistic countries: a mediation study in 36 countries.

PURPOSE: The prevalence of parental burnout, a condition that has severe consequences for both parents and children, varies dramatically across countries and is highest in Western countries characterized by high individualism. METHOD: In this study, we examined the mediators of the relationship between individualism measured at the country level and parental burnout measured at the individual level in 36 countries (16,059 parents). RESULTS: The results revealed three mediating mechanisms, that is, self-discrepancies between socially prescribed and actual parental selves, high agency and self-directed socialization goals, and low parental task sharing, by which individualism leads to an increased risk of burnout among parents. CONCLUSION: The results confirm that the three mediators under consideration are all involved, and that mediation was higher for self-discrepancies between socially prescribed and actual parental selves, then parental task sharing, and lastly self-directed socialization goals. The results provide some important indications of how to prevent parental burnout at the societal level in Western countries.

Predictive value of the G8 screening tool for postoperative complications in older adults undergoing cancer surgery: A systematic review and meta-analysis.

INTRODUCTION: Older adults with cancer who are being considered for cancer surgery are heterogenous, with variation in their physical, mental, and social baselines and risk of postoperative complications. Due in part to the complex nature of this population, the optimal preoperative evaluation method is not clearly defined. In this study we investigated whether geriatric-8 (G8), a screening tool for older patients with cancer that determines the need for a full geriatric assessment, is suitable for assessing the risk of postoperative complications in this population being considered for surgery. MATERIALS AND METHODS: Studies that enrolled older patients undergoing cancer surgery and compared prevalence of postoperative complications in G8 "high" (≥15) patients and G8 "low" (<15) patients were identified using PubMed and EMBASE. A meta-analysis was conducted to calculate the risk ratio of postoperative complication rate. Postoperative mortality was systematically reviewed. RESULTS: Eleven studies published between 2017 and 2022 were included in our analysis with a total of 2,691 older patients who underwent various types of cancer surgery and were characterized by their G8 scores: 1,255 G8 high (≥15) patients and 1,436 G8 low (<15) patients. G8 low patients had a significantly higher prevalence of postoperative complications than G8 high patients (risk ratio [95% confidence interval]: 1.56 [1.18-2.07], p = 0.002, I2 = 79%). DISCUSSION: G8 can be an effective and efficient preoperative tool to assess risk of postoperative complications in older adults undergoing cancer surgery and identify potential need for further evaluation of an individual's risk with a comprehensive geriatric assessment.

Total artificial heart implantation as a bridge to transplantation in the United States.

OBJECTIVES: Although the SynCardia total artificial heart (SynCardia Systems, LLC) was approved for use as a bridge to transplantation in 2004 in the United States, most centers do not adopt the total artificial heart as a standard bridging strategy for patients with biventricular failure. This study was designed to characterize the current use and outcomes of patients placed on total artificial heart support. METHODS: The United Network of Organ Sharing Standard Transplant Research File was queried to identify total artificial heart implantation in the United States between 2005 and 2018. Multivariable Cox regression models were used for risk prediction. RESULTS: A total of 471 patients (mean age, 49 years; standard deviation, 13 years; 88% were male) underwent total artificial heart implantation. Of 161 transplant centers, 11 centers had cumulative volume of 10 or more implants. The 6-month cumulative incidence of mortality on the total artificial heart was 24.6%. The 6-month cumulative incidence of transplant was 49.0%. The 1-year mortality post-transplantation was 20.0%. Cumulative center volume less than 10 implants was predictive of both mortality on the total artificial heart (hazard ratio, 2.2, 95% confidence interval, 1.5-3.1, P < .001) and post-transplant mortality after a total artificial heart bridge (hazard ratio, 1.5, 95% confidence interval, 1.0-2.2, P = .039). CONCLUSIONS: Total artificial heart use is low, but the total artificial heart can be an option for biventricular bridge to transplant with acceptable bridge to transplant and post-transplant survival, especially in higher-volume centers. The observation of inferior outcomes in lower-volume centers raises questions as to whether targeted training, center certifications, and minimum volume requirements could improve outcomes for patients requiring the total artificial heart.

Limited cumulative experience with ex vivo lung perfusion is associated with inferior outcomes after lung transplantation.

OBJECTIVE: Ex vivo lung perfusion (EVLP) allows for prolonged preservation and evaluation/resuscitation of donor lungs. We evaluated the influence of center experience with EVLP on lung transplant outcomes. METHODS: We identified 9708 isolated, first-time adult lung transplants from the United Network for Organ Sharing database (March 1, 2018-March 1, 2022), 553 (5.7%) involved using donor lungs after EVLP. Using the total volume of EVLP lung transplants per center during the study period, centers were dichotomized into low- (1-15 cases) and high-volume (>15 cases) EVLP centers. RESULTS: Forty-one centers performed EVLP lung transplants, including 26 low-volume and 15 high-volume centers (median volume, 3 vs 23 cases; P < .001). Recipients at low-volume centers (n = 109) had similar baseline comorbidities compared with high-volume centers (n = 444). Low-volume centers used numerically more donation after circulatory death donors (37.6 vs 28.4%; P = .06) and more donors with Pao2/Fio2 ratio <300 (24.8 vs 9.7%; P < .001). After EVLP lung transplants, low-volume centers had worse 1-year survival (77.8% vs 87.5%; P = .007), with an adjusted hazard ratio of 1.63 (95% CI, 1.06-2.50, adjusting for recipient age, sex, diagnosis, lung allocation score, donation after circulatory death donor, donor Pao2/Fio2 ratio, and total annual lung transplant volume per center). When compared to non-EVLP lung transplants, 1-year survival of EVLP lung transplants was significantly worse at low-volume centers (adjusted hazard ratio, 2.09; 95% CI, 1.47-2.97) but similar at high-volume centers (adjusted hazard ratio, 1.14; 95% CI, 0.82-1.58). CONCLUSIONS: The use of EVLP in lung transplantation remains limited. Increasing cumulative EVLP experience is associated with improved outcomes of lung transplantation using EVLP-perfused allografts.

Racial Disparities in Failure to Rescue after Pediatric Heart Surgeries in the US.

OBJECTIVE: To identify the trend in failure to rescue (FTR) and risk factors contributing to racial disparities in FTR after pediatric heart surgery using contemporary nationwide data. STUDY DESIGN: We identified 85 267 congenital heart surgeries in patients <18 years of age from 2009 to 2019 using the Kid's Inpatient Database. The primary outcome was FTR. A mixed-effect logistic regression model with hospital random intercept was used to identify independent predictors of FTR. RESULTS: Among 36 753 surgeries with postoperative complications, the FTR was 7.3%. The FTR decreased from 7.4% in 2009 to 6.3% in 2019 (P = .02). FTR was higher among Black than White children for all years. The FTR was higher among girls (7.2%) vs boys (6.6%), children aged <1 (9.6%) vs 12-17 years (2.4%), and those of Black (8.5%) vs White race (5.9%) (all P < .05). Black race was associated with a higher FTR odds (OR, 1.40; 95% CI, 1.20-1.65) after adjusting for demographics, medical complexity, nonelective admission, and hospital surgical volume. Higher hospital volume was associated with a lower odds of FTR for all racial groups, but fewer Black (19.7%) vs White (31%) children underwent surgery at high surgical volume hospitals (P < .001). If Black children were operated on in the same hospitals as White children, the racial differences in FTR would decrease by 47.3%. CONCLUSIONS: Racial disparities exist in FTR after pediatric heart surgery in the US. The racial differences in the location of care may account for almost half the disparities in FTR.

Bicuspid aortic stenosis: National three-year outcomes of transcatheter versus surgical aortic valve replacement among Medicare beneficiaries.

OBJECTIVE: Randomized trials of transcatheter versus surgical aortic valve replacements have excluded bicuspid anatomy. We compared 3-year outcomes of transcatheter aortic valve replacement versus surgical aortic valve replacement in patients aged more than 65 years with bicuspid aortic stenosis. METHODS: The Centers for Medicare and Medicaid data were used to identify 6450 patients undergoing isolated surgical aortic valve replacement (n = 3771) or transcatheter aortic valve replacement (n = 2679) for bicuspid aortic stenosis (2012-2019). Propensity score matching with 21 baseline characteristics including frailty created 797 pairs. RESULTS: Unmatched patients undergoing transcatheter aortic valve replacement were older than patients undergoing surgical aortic valve replacement (78 vs 70 years), with more comorbidities and frailty (all P < .001). After matching, transcatheter aortic valve replacement was associated with a similar mortality risk compared with surgical aortic valve replacement within the first 6 months (hazard ratio [HR], 1.08, 95% CI, 0.67-1.69) but a higher mortality risk between 6 months and 3 years (HR, 2.16, 95% CI, 1.22-3.83). Additionally, transcatheter aortic valve replacement was associated with a lower risk of heart failure readmissions before 6 months (HR, 0.51, 95% CI, 0.31-0.87) but a higher risk between 6 months and 3 years (HR, 4.78, 95% CI, 2.21-10.36). The 3-year risks of aortic valve reintervention (HR, 1.03, 95% CI, 0.30-3.56) and stroke (HR, 1.21, 95% CI, 0.75-1.96) were similar. CONCLUSIONS: Among matched Medicare beneficiaries undergoing transcatheter aortic valve replacement or surgical aortic valve replacement for bicuspid aortic stenosis, 3-year mortality was higher after transcatheter aortic valve replacement. However, transcatheter aortic valve replacement was associated with a similar risk of mortality and a lower risk of heart failure readmissions during the first 6 months after the intervention. Randomized comparative data are needed to best inform treatment choice.

Development of a prediction model of postpartum hospital use using an equity-focused approach.

BACKGROUND: Racial inequities in maternal morbidity and mortality persist into the postpartum period, leading to a higher rate of postpartum hospital use among Black and Hispanic people. Delivery hospitalizations provide an opportunity to screen and identify people at high risk to prevent adverse postpartum outcomes. Current models do not adequately incorporate social and structural determinants of health, and some include race, which may result in biased risk stratification. OBJECTIVE: This study aimed to develop a risk prediction model of postpartum hospital use while incorporating social and structural determinants of health and using an equity approach. STUDY DESIGN: We conducted a retrospective cohort study using 2016-2018 linked birth certificate and hospital discharge data for live-born infants in New York City. We included deliveries from 2016 to 2017 in model development, randomly assigning 70%/30% of deliveries as training/test data. We used deliveries in 2018 for temporal model validation. We defined "Composite postpartum hospital use" as at least 1 readmission or emergency department visit within 30 days of the delivery discharge. We categorized diagnosis at first hospital use into 14 categories based on International Classification of Diseases-Tenth Revision diagnosis codes. We tested 72 candidate variables, including social determinants of health, demographics, comorbidities, obstetrical complications, and severe maternal morbidity. Structural determinants of health were the Index of Concentration at the Extremes, which is an indicator of racial-economic segregation at the zip code level, and publicly available indices of the neighborhood built/natural and social/economic environment of the Child Opportunity Index. We used 4 statistical and machine learning algorithms to predict "Composite postpartum hospital use", and an ensemble approach to predict "Cause-specific postpartum hospital use". We simulated the impact of each risk stratification method paired with an effective intervention on race-ethnic equity in postpartum hospital use. RESULTS: The overall incidence of postpartum hospital use was 5.7%; the incidences among Black, Hispanic, and White people were 8.8%, 7.4%, and 3.3%, respectively. The most common diagnoses for hospital use were general perinatal complications (17.5%), hypertension/eclampsia (12.0%), nongynecologic infections (10.7%), and wound infections (8.4%). Logistic regression with least absolute shrinkage and selection operator selection retained 22 predictor variables and achieved an area under the receiver operating curve of 0.69 in the training, 0.69 in test, and 0.69 in validation data. Other machine learning algorithms performed similarly. Selected social and structural determinants of health features included the Index of Concentration at the Extremes, insurance payor, depressive symptoms, and trimester entering prenatal care. The "Cause-specific postpartum hospital use" model selected 6 of the 14 outcome diagnoses (acute cardiovascular disease, gastrointestinal disease, hypertension/eclampsia, psychiatric disease, sepsis, and wound infection), achieving an area under the receiver operating curve of 0.75 in training, 0.77 in test, and 0.75 in validation data using a cross-validation approach. Models had slightly lower performance in Black and Hispanic subgroups. When simulating use of the risk stratification models with a postpartum intervention, identifying high-risk individuals with the "Composite postpartum hospital use" model resulted in the greatest reduction in racial-ethnic disparities in postpartum hospital use, compared with the "Cause-specific postpartum hospital use" model or a standard approach to identifying high-risk individuals with common pregnancy complications. CONCLUSION: The "Composite postpartum hospital use" prediction model incorporating social and structural determinants of health can be used at delivery discharge to identify persons at risk for postpartum hospital use.

Lung transplantation outcomes in patients from socioeconomically distressed communities.

BACKGROUND: Previous studies have demonstrated racial and gender disparities in lung allocation, but contemporary data regarding socioeconomic disparities in post-transplant outcomes are lacking. We evaluated the impact of a composite socioeconomic disadvantage index on post-transplant outcomes. METHODS: The Scientific Registry of Transplant Recipients identified 27,763 adult patients undergoing isolated primary lung transplantation between 2005 and 2020. Zip code-level socioeconomic distress was characterized using the Distressed Communities Index (DCI: 0-no distress, 100-severe distress) based on education level, poverty, unemployment, housing vacancies, median income, and business growth, and patients were stratified into high (DCI ≥60) or low (DCI <60) distressed groups. RESULTS: Recipients from high-distress communities (n = 8006, 28.8%) were younger (59years [interquartile range {IQR} 50-64] vs 61years [IQR 52-66]), less often white (73 vs 85%), less likely to have a college degree (45 vs 59%), and more likely to have public insurance (57 vs 49%, all p < 0.001) compared to those from low-distress communities. Additionally, high-distress recipients were more likely to have group A diagnoses (32 vs 27%) and undergo bilateral lung transplants (72.4 vs 69.3%, all p < 0.001). Post-transplant survival at 5years was 55.7% (95% confidence interval [CI]: 54.4-56.9) in high-distress recipients and 58.2% (95% CI: 57.4-58.9) in low-distress recipients (p = 0.003). After adjustment, high distress level was independently associated with an increased risk of 5-year mortality (hazard ratio:1.09, 95% CI:1.04-1.15). CONCLUSIONS: Recipients from distressed communities are at increased mortality risk following lung transplantation. Efforts should be focused on increased resource allocation and further study to better understand factors which may mitigate this disparity.

Socioeconomic disparities in midterm outcomes after repair for degenerative mitral regurgitation.

OBJECTIVE: The influence of socioeconomic disparities on survival after mitral repair is poorly defined. We examined the association between socioeconomic disadvantage and midterm outcomes of repair in Medicare beneficiaries with degenerative mitral regurgitation. METHODS: US Centers for Medicare and Medicaid Services data were used to identify 10,322 patients undergoing isolated first-time repair for degenerative mitral regurgitation between 2012 and 2019. Zip code-level socioeconomic disadvantage was dichotomized with the Distressed Communities Index, which incorporates education level, poverty, unemployment, housing security, median income, and business growth; those with Distressed Communities Index score ≥80 were classified as distressed. The primary outcome was survival, censored at 3 years. Secondary outcomes included cumulative incidences of heart failure readmission, mitral reintervention, and stroke. RESULTS: Of the 10,322 patients undergoing degenerative mitral repair, 9.7% (n = 1003) came from distressed communities. Patients from distressed communities underwent surgery at lower volume centers (11 vs 16 cases/year) and traveled further for surgical care (40 vs 17 miles) (both P values < .001). At 3 years, unadjusted survival (85.4%; 95% CI, 82.9%-87.5% vs 89.7%; 95% CI, 89.0%-90.4%) and cumulative incidence of heart failure readmission (11.5%; 95% CI, 9.6%-13.7% vs 7.4%; 95% CI, 6.9%-8.0%) were worse in patients from distressed communities (all P values < .001), whereas mitral reintervention rates were similar (2.7%; 95% CI, 1.8%-4.0% vs 2.8%; 95% CI, 2.5%-3.2%; P = .75). After adjustment, community distress was independently associated with 3-year mortality (hazard ratio, 1.21; 95% CI, 1.01-1.46) and heart failure readmissions (hazard ratio, 1.28; 95% CI, 1.04-1.58). CONCLUSIONS: Community-level socioeconomic distress is associated with worse outcomes in degenerative mitral repair among Medicare beneficiaries.

Multiple arterial versus single arterial grafting in patients with diabetes undergoing coronary artery bypass surgery.

Background: Data on long-term outcomes in patients with diabetes receiving multiarterial grafting (MAG) versus single-artery grafting (SAG) are limited. Objectives: The objective of this study is to compare long-term outcomes between MAG and SAG for coronary artery bypass graft (CABG) surgery in patients with diabetes. Methods: Patients with diabetes who underwent isolated CABG surgeries between 2000 to 2016 were identified using the New Jersey mandatory state clinical registry linked with death records and hospital discharge data (last follow-up December 31, 2019). Patients who underwent CABG for single-vessel disease, with only venous conduits, patients with previous heart surgeries, or hemodynamically unstable were excluded. Patients undergoing MAG and SAG were matched by propensity score. Cox proportional hazard models were used to investigate long-term survival and competing risk analysis was used for secondary outcomes. Results: Of 24,944 patients, 2955 underwent MAG, and 21,989 underwent SAG CABG. Patients receiving MAG were younger, predominantly men, with a lower prevalence of hypertension, peripheral vascular disease, congestive heart failure, chronic lung disease, and renal failure. MAG was associated with lower long-term mortality compared with SAG in 2882 propensity score-matched pairs (hazard ratio [HR], 0.75; 95% CI, 0.68-0.83); lower risks of repeat revascularization (subdistribution HR, 0.86; 95% CI, 0.76-0.97); and composite outcome defined as death from any cause, stroke, postoperative myocardial infarction, and/or repeat revascularization (HR, 0.76; 95% CI, 0.71-0.82). These results were confirmed in subgroup analyses of women, men, age <70 years, and age ≥70 years patients with diabetes. MAG was also associated with lower mortality compared with SAG in patients with diabetes taking insulin in the entire cohort (Video Abstract). Conclusions: Patients with diabetes benefit from receiving MAG over SAG and demonstrated improved long-term survival, and lower hazards of secondary and composite outcomes. Coordinated efforts are needed to offer MAG to patients with diabetes.

Left atrial appendage closure during mitral repair in patients without atrial fibrillation.

OBJECTIVE: Routine left atrial appendage closure during mitral repair in patients without atrial fibrillation (AF) is controversial. We aimed to compare the incidence of stroke after mitral repair in patients without recent AF according to left atrial appendage closure. METHODS: An institutional registry identified 764 consecutive patients without recent AF, endocarditis, prior appendage closure, or stroke undergoing isolated robotic mitral repair between 2005 and 2020. Left atrial appendages were closed via left atriotomy using a double-layer continuous suture in 5.3% (15 out of 284) patients before 2014, versus 86.7% (416 out of 480) after 2014. The cumulative incidence of stroke (including transient ischemic attack) was determined using statewide hospital data. Median follow-up was 4.5 years (range, 0-16.6 years). RESULTS: Patients undergoing left atrial appendage closure were older (63 vs 57.5 years, P < .001), with higher prevalence of remote AF requiring cryomaze (9%, n = 40 vs 1%, n = 3, P < .001). After appendage closure there were fewer reoperations for bleeding (0.7% [n = 3] vs 3% [n = 10]; P = .02), and more AF (31.8% [n = 137] vs 25.2% [n = 84]; P = .047). Two-year freedom from >2+ mitral regurgitation was 97%. Six strokes and 1 transient ischemic attack occurred after appendage closure compared with 14 and 5 in patients without (P = .002), associated with a significant difference in 8-year cumulative incidence of stroke/transient ischemic attack (hazard ratio, 0.3; 95% CI, 0.14-0.85; P = .02). This difference persisted in the sensitivity analysis, excluding patients undergoing concomitant cryomaze procedures. CONCLUSIONS: Routine left atrial appendage closure during mitral repair in patients without recent AF appears safe and was associated with a lower risk of subsequent stroke/transient ischemic attack.