Machine Learning on 50,000 Manuscripts Shows Increased Clinical Research by Academic Cardiac Surgeons.

INTRODUCTION: Academic cardiac surgeons are productive researchers and innovators. We sought to perform a comprehensive machine learning (ML)-based characterization of cardiac surgery research over the past 40 y to identify trends in research pursuits. METHODS: US-based academic websites were queried for surgeon profiles. Publications since 1980 were obtained from Web of Science, and publication classifications (e.g., "human", "animal") were collected through the National Institutes of Health iCite tool. Publications were deemed "basic or translational" if >50% of their classification was under "animal" or "molecular or cell", and "clinical" if otherwise. ML-based clustering was performed on publication titles and Medical Subject Heading terms to identify research topics. RESULTS: A total of 944 cardiac surgeons accounted for 48,031 unique publications. Average citations per year have decreased since 1980 (P < 0.001). The percentage of basic or translational publications by cardiac surgeons has decreased over time (P < 0.001), comprising of only 8% of publications in 2022. Adult cardiac surgeons, those who received an F32, K08, or R01, and those with a PhD were more likely to publish basic or translational research. Top areas of basic or translational research were myocardial reperfusion, aortic aneurysms or remodeling, and transplant immunology. Major areas of clinical research included aortic disease, aortic valve disease, and mechanical circulatory support. Collaboration analysis revealed that 55% of publications were single-center, and the yearly percentage of these publications has decreased over time (P < 0.001). CONCLUSIONS: Cardiac surgeons are performing less basic or translational research relative to clinical research than ever before. The majority of publications over the past 40 y did not involve cross-center collaboration. Continued support for clinical research is needed, while also encouraging collaborative basic or translational science to foster innovation in patient care.

Prophylactic left atrial appendage ligation during coronary artery bypass grafting: A propensity score-matched analysis.

OBJECTIVE: The study objective was to evaluate short-term outcomes and statewide practice patterns of prophylactic left atrial appendage ligation in patients undergoing coronary artery bypass grafting without preoperative atrial fibrillation. METHODS: Adult patients who underwent on-pump coronary artery bypass grafting (2017-2023) within a regional collaborative were identified. Patients with a history of atrial fibrillation, previous cardiac surgery, or nondevice-based left atrial appendage ligation were excluded. Patients were stratified by left atrial appendage ligation status and were propensity score matched. Univariable analysis was used to compare short-term clinical outcomes. RESULTS: Of 16,547 patients examined, 442 underwent prophylactic left atrial appendage ligation. The propensity score-matched cohort (439 in the prophylactic left atrial appendage ligation group, 439 in the no prophylactic left atrial appendage ligation group) was compared and had no significant differences in preoperative Congestive heart failure, Hypertension, Age ≥ 75 (doubled), Diabetes mellitus, prior Stroke, Transient ischemic attack or Thromboembolism (doubled), Vascular disease, Age 65 to 74, Sex category (female)(CHA2DS2-VASc) scores or operative variables. The prophylactic left atrial appendage ligation group had longer crossclamp time (82 vs 76 minutes, P = .001), intensive care unit hours (72 vs 66, P = .001), and length of stay (6.0 vs 6.0 days, P = .010); increased postoperative atrial fibrillation (35% vs 24%, P < .001); and more discharges on anticoagulation (17% vs 8.2%, P < .001). There were no significant differences in postoperative stroke (1.1% vs 2.1%, P = .423), readmission (13% vs 9.6%, P = .118), operative mortality (2.5% vs 1.6%, P = .480), or readmission for thrombotic or bleeding complications (0.7% vs 1.1%, P = .724). Hospitalization costs were significantly higher for the prophylactic left atrial appendage ligation group ($43,478 vs $40,645, P < .001). The rate of prophylactic left atrial appendage ligation during coronary artery bypass grafting increased from 1.61% (2017) to 5.65% (2023) (P < .001). CONCLUSIONS: Despite higher rates of postoperative atrial fibrillation, discharge on anticoagulation, and hospitalization costs in patients undergoing prophylactic left atrial appendage ligation during coronary artery bypass grafting, there was no difference in short-term clinical end points including stroke and operative mortality.

Cardiac Surgical Unit-Advanced Life Support-certified centers are associated with improved failure to rescue after cardiac arrest: A propensity score-matched analysis.

OBJECTIVE: The impact of Cardiac Surgical Unit-Advanced Life Support (CSU-ALS) training on failure to rescue after cardiac arrest (FTR-CA) is unknown. We hypothesized that institutional CSU-ALS certification would be associated with lower FTR-CA. METHODS: Patients undergoing Society of Thoracic Surgeons index operations from 2020 to 2023 from a regional collaborative were analyzed. Each institution was surveyed regarding its status as a CSU-ALS-certified center. Patients stratified by CSU-ALS certification were 1:1 propensity score matched with subsequent multivariable model reviewing associations with FTR-CA. RESULTS: A total of 12,209 patients were included in the study period across 15 institutions. Eight centers reported CSU-ALS certification. After propensity score matching, 2 patient cohorts were formed (n = 3557). Patients at CSU-ALS centers had greater rates of intensive care unit readmission (3.9% vs 2.3%, P < .01) and total operating room time (340 minutes vs 323 minutes, P < .01). Hospital readmission was less likely in the CSU-ALS centers (9.0% vs 10.1%, P < .01). There was no difference in the rate of postoperative cardiac arrest (1.8% vs 2.2%, P = .24) or operative mortality (2.5% vs 2.9%, P = .30). After risk adjustment, CSU-ALS centers (odds ratio, 0.30; 95% confidence interval, 0.12-0.72, P < .01) and greater-volume centers (odds ratio, 0.15; confidence interval, 0.03-0.74, P = .02) had reduced odds of FTR-CA. CONCLUSIONS: Centers with CSU-ALS certification are associated with a lower risk-adjusted likelihood of FTR-CA. This highlights the importance of well-trained staff and treatment algorithms in the care of patients postcardiac surgery.

Does Timing Matter? The Effect of Intensive Care Unit Arrival Timing on Elective Cardiac Surgery.

BACKGROUND: Due to staffing changes at scheduled intervals and decreases in essential staff in the evenings, late intensive care unit (ICU) arrivals may be at risk for suboptimal outcomes. Utilizing a regional collaborative, we sought to determine the effect of ICU arrival timing on outcomes in elective isolated coronary artery bypass. METHODS: Adults undergoing elective, isolated coronary artery bypass from 17 hospitals between 2013 and 2023 were identified. Patients with missing predicted risk of mortality or missing ICU arrival time were excluded. Late ICU arrival time was defined as between 6:00 pm and 6:00 am. Hierarchical logistic regression with appropriate predicted risk scores was utilized for outcome risk adjustment. RESULTS: We identified 11,638 patients, with 972 (8.4%) experiencing late ICU arrival. Late ICU arrival patients had higher predicted risk of morbidity or mortality (8.2%; [interquartile range {IQR}, 5.6%, 12.0%] vs 7.7% [IQR, 5.5%, 11.5%], P = .048) compared with early ICU arrival patients with longer median cardiopulmonary bypass times (96 minutes [IQR, 78, 119] vs 93 [IQR, 73, 116], P < .001). Late ICU arrival patients experienced more unadjusted complications including prolonged ventilation (7.7% vs 4.2%, P < .001) and operative mortality (2.0% vs 1.1%, P = .02), although no difference in failure-to-rescue (11.0% vs 10.4%, P = .84). Logistic regression with risk adjustment demonstrated late ICU arrival as a predictor of prolonged ventilation (odds ratio, 1.49 [95% CI, 1.12-1.99], P = .006). CONCLUSIONS: After adjustment, late ICU arrivals experienced higher rates of prolonged ventilation, although this did not translate to failure-to-rescue.

Mitral valve repair in a regional quality collaborative: Respect or resect?

Objective: Mitral valve repair is the gold standard for treatment of mitral regurgitation, but the optimal technique remains debated. By using a regional collaborative, we sought to determine the change in repair technique over time. Methods: We identified all patients undergoing isolated mitral valve repair from 2012 to 2022 for degenerative mitral disease. Those with endocarditis, transcatheter repair, or tricuspid intervention were excluded. Continuous variables were analyzed via Wilcoxon rank sum, and categorical variables were analyzed via chi-square testing. Results: We identified 1653 patients who underwent mitral valve repair, with 875 (59.2%) undergoing a no resection repair. Over the last decade, there was no significant trend in the proportion of repair techniques across the region (P = .96). Those undergoing no resection repairs were more likely to have undergone prior cardiac surgery (5.0% vs 2.2%, P = .002) or minimally invasive approaches (61.4% vs 24.7%, P < .001) with similar predicted risk of mortality (median 0.6% vs 0.6%, P = .75). Intraoperatively, no resection repairs were associated with longer bypass times (140 [117-167] minutes vs 122 [91-159] minutes, P < .001). Operative mortality was similar between both groups (1.1% vs 1.0%, P = .82), as were other postoperative outcomes. Anterior leaflet prolapse (odds ratio, 11.16 [6.34-19.65], P < .001) and minimally invasive approach (odds ratio, 6.40 [5.06-8.10], P < .001) were most predictive of no resection repair. Conclusions: Despite minor differences in operative times, statewide over the past decade there remains a diverse mix of both classic "resect" and newer "respect" strategies with comparable short-term outcomes and no major timewise trends. These data may suggest that both approaches are equivocal.

Operating Room Extubation for Patients Undergoing Cardiac Surgery: A National Society of Thoracic Surgeons Database Analysis.

BACKGROUND: The utility of operating room extubation (ORE) after cardiac surgery over fast-track extubation (FTE) within 6 hours remains contested. We hypothesized ORE would be associated with equivalent rates of morbidity and mortality, relative to FTE. METHODS: Patients undergoing nonemergent cardiac surgery were identified in The Society of Thoracic Surgeons Adult Cardiac Surgery Database between July 2017 and December 2022. Only procedures with The Society of Thoracic Surgeons risk models were included. Risk-adjusted outcomes of ORE and FTE were compared by observed-to-expected ratios with 95% CIs aggregated over all procedure types, and ORE vs FTE adjusted odds ratios (ORs) specific to each procedure type using multivariable logistic regression. Analyzed outcomes were operative mortality, prolonged length of stay, composite reoperation for bleeding and reintubation, and composite morbidity and mortality. RESULTS: The study population of 669,099 patients across 1069 hospitals included 36,298 ORE patients in 296 hospitals. Risk-adjusted analyses found that ORE was associated with statistically similar or better results across each of the 4 outcomes and procedure subtypes. Notably, rates of postoperative mortality were significantly lower in ORE patients undergoing coronary artery bypass grafting (OR, 0.54; 95% CI, 0.46-0.65), aortic valve replacement (OR, 0.43; 95% CI, 0.24-0.77), and mitral valve replacement (OR, 0.48; 95% CI, 0.26-0.89). CONCLUSIONS: Extubation in the OR was safe and effective in a selected patient population and may be associated with superior outcomes in coronary artery bypass, aortic valve replacement, and mitral valve replacement. These national data appear to confirm institutional experiences regarding the potential benefit of OR extubation. Further refinement of optimal populations may justify randomized investigation.

Venoarterial Extracorporeal Life Support Use in Acute Pulmonary Embolism Shows Favorable Outcomes.

BACKGROUND: Differences in outcomes by indication for venoarterial extracorporeal life support (VA-ECLS) are poorly described. We hypothesized that patients on VA-ECLS for acute pulmonary embolism (PE) have fewer complications and better survival than patients on VA-ECLS for other indications. METHODS: All patients ≥18 years on VA-ECLS from the Extracorporeal Life Support Organization global registry (2010-2019) were evaluated (n = 29,842). After excluding patients aged >79 years (n = 729) and those with incomplete indication data (n = 2530), patients were stratified by VA-ECLS indication for PE vs all other indications. The association between being discharged alive and each type of complication with VA-ECLS indication was assessed. RESULTS: Of 26,583 patients included in the analysis, 978 (3.7%) were on VA-ECLS for a primary diagnosis of acute PE. Acute PE patients were younger (53.1 vs 56.7 years, P < .001) and were more likely to be women (52.1% vs 32.3%, P < .001). Patients who underwent VA-ECLS for acute PE were 78% more likely to be discharged alive vs patients supported with VA-ECLS for other reasons (52.8% vs 40.4%; P < .001). Acute PE patients had fewer cardiovascular and renal complications (26.6% vs 38.0% and 31.1% vs 39.4%, respectively; adjusted P < .001). Acute PE patients had higher odds of having clots and mechanical complications (8.7% vs 7.9% and 16.7% vs 14.6%, respectively; adjusted P < .001). CONCLUSIONS: Patients undergoing VA-ECLS for acute PE have higher odds of survival to hospital discharge compared with those supported for other indications. Additionally, VA-ECLS in this population is associated with fewer cardiovascular and renal complications but higher mechanical complications.

Impact of Frailty in Patients Undergoing Minimally Invasive Mitral Valve Surgery.

OBJECTIVE: Psoas muscle size is a reliable marker of sarcopenia and frailty that correlates with adverse outcomes after cardiac surgery. However, its use in mitral and minimally invasive cardiac surgery is lacking. We sought to determine whether frailty, as measured by psoas muscle index, increases surgical risk for minimally invasive mitral valve surgery. METHODS: Patients undergoing isolated minimally invasive mitral surgery via right minithoracotomy were identified. Patients who underwent maze, tricuspid intervention, and those who were emergent were excluded. Total psoas muscle area was calculated using the average cross-sectional area at the L3 vertebra on computed tomography scan and indexed to body surface area. Sarcopenia was defined as <25th gender-specific percentile. Patients were stratified by sarcopenia status and outcomes compared. RESULTS: Of 287 total patients, 192 patients met inclusion criteria. Sarcopenic patients were 6 years older (66 vs 60 years, P = 0.01), had lower preoperative albumin levels (4.0 vs 4.3 g/dL, P < 0.001), and had higher Society of Thoracic Surgeons risk of morbidity/mortality (13.1% vs 9.0%, P = 0.003). Operative major morbidity or mortality was 6.4% versus 5.5% (P = 0.824), while the 1-year mortality rate was 2.1% versus 0% (P = 0.08). After risk adjustment, psoas index did not predict operative morbidity or mortality. However, sarcopenia was associated with higher odds of readmission (odds ratio = 0.74, P = 0.02). CONCLUSIONS: Contrary to other cardiac operations, for patients undergoing isolated minimally invasive mitral valve surgery, sarcopenia was not associated with increased perioperative risk except for higher readmission rates. Minimally invasive surgical approaches should be strongly considered as the approach of choice in frail patients.

Extracorporeal Life Support Organization Center of Excellence recognition is associated with improved failure to rescue after cardiac arrest.

OBJECTIVE: The influence of Extracorporeal Life Support Organization (ELSO) center of excellence (CoE) recognition on failure to rescue after cardiac surgery is unknown. We hypothesized that ELSO CoE would be associated with improved failure to rescue. METHODS: Patients undergoing a Society of Thoracic Surgeons index operation in a regional collaborative (2011-2021) were included. Patients were stratified by whether or not their operation was performed at an ELSO CoE. Hierarchical logistic regression analyzed the association between ELSO CoE recognition and failure to rescue. RESULTS: A total of 43,641 patients were included across 17 centers. In total, 807 developed cardiac arrest with 444 (55%) experiencing failure to rescue after cardiac arrest. Three centers received ELSO CoE recognition, and accounted for 4238 patients (9.71%). Before adjustment, operative mortality was equivalent between ELSO CoE and non-ELSO CoE centers (2.08% vs 2.36%; P = .25), as was the rate of any complication (34.5% vs 33.8%; P = .35) and cardiac arrest (1.49% vs 1.89%; P = .07). After adjustment, patients undergoing surgery at an ELSO CoE facility were observed to have 44% decreased odds of failure to rescue after cardiac arrest, relative to patients at non-ELSO CoE facility (odds ratio, 0.56; 95% CI, 0.316-0.993; P = .047). CONCLUSIONS: ELSO CoE status is associated with improved failure to rescue following cardiac arrest for patients undergoing cardiac surgery. These findings highlight the important role that comprehensive quality programs serve in improving perioperative outcomes in cardiac surgery.

Renal recovery in cardiac surgery patients requiring postoperative renal-replacement therapy.

OBJECTIVE: Renal failure after cardiac surgery is associated with increased morbidity and mortality. There is a lack of data examining the rate of renal recovery after patients have started dialysis following cardiac surgery. We aimed to determine the frequency of and time to renal recovery of patients requiring dialysis after cardiac surgery. METHODS: All patients who developed new-onset renal failure requiring dialysis following cardiac surgery at our institution from 2011 to 2022 were included. Renal recovery, time to renal recovery, and mortality at 1 year were merged with patients' Society of Thoracic Surgeons Adult Cardiac Surgery Database files. Kaplan-Meier analysis was used to predict time to renal recovery; we censored patients who died or were lost to follow up. Cox regression was used for risk-adjustment. RESULTS: A total of 312 patients were included in the final analysis. Mortality during index hospital admission was 33% (n = 105), and mortality at 1 year was 45% (n = 141). Of those surviving at 1 year, 69% (n = 118) remained renally recovered. Median renal recovery time was 56 (37-74) days. Accounting for mortality as a competing risk, 51% of patients were predicted to achieve renal recovery. Increasing age (hazard ratio, 0.98; 0.514-0.94, P < .026) and increasing total packed red blood cells (hazard ratio, 0.0958; 0.937-0.987, P < .001) received were found to be significant negative predictors of renal recovery in the Fine-Gray model for subhazard distribution. CONCLUSIONS: More than two-thirds of patients with renal failure who survived the perioperative period had renal recovery within 1 year after surgery. Recovery was driven primarily by postoperative complications rather than comorbidities and intraoperative factors, suggesting renal failure in the postoperative cardiac surgery patient surviving to discharge is unlikely to be permanent.