Parsimonious machine learning models to predict resource use in cardiac surgery across a statewide collaborative.

Objective: We sought to several develop parsimonious machine learning models to predict resource utilization and clinical outcomes following cardiac operations using only preoperative factors. Methods: All patients undergoing coronary artery bypass grafting and/or valve operations were identified in the 2015-2021 University of California Cardiac Surgery Consortium repository. The primary end point of the study was length of stay (LOS). Secondary endpoints included 30-day mortality, acute kidney injury, reoperation, postoperative blood transfusion and duration of intensive care unit admission (ICU LOS). Linear regression, gradient boosted machines, random forest, extreme gradient boosting predictive models were developed. The coefficient of determination and area under the receiver operating characteristic (AUC) were used to compare models. Important predictors of increased resource use were identified using SHapley summary plots. Results: Compared with all other modeling strategies, gradient boosted machines demonstrated the greatest performance in the prediction of LOS (coefficient of determination, 0.42), ICU LOS (coefficient of determination, 0.23) and 30-day mortality (AUC, 0.69). Advancing age, reduced hematocrit, and multiple-valve procedures were associated with increased LOS and ICU LOS. Furthermore, the gradient boosted machine model best predicted acute kidney injury (AUC, 0.76), whereas random forest exhibited greatest discrimination in the prediction of postoperative transfusion (AUC, 0.73). We observed no difference in performance between modeling strategies for reoperation (AUC, 0.80). Conclusions: Our findings affirm the utility of machine learning in the estimation of resource use and clinical outcomes following cardiac operations. We identified several risk factors associated with increased resource use, which may be used to guide case scheduling in times of limited hospital capacity.

Remote simulator training of coronary artery bypass grafting during the coronavirus disease 2019 pandemic.

OBJECTIVES: The coronavirus disease 2019 pandemic presents in-person exposure risk during surgical education. We aimed to validate the feasibility of fully remote faculty-supervised surgical training sessions focused on coronary artery bypass grafting using a synthetic simulator and online videochat software. METHODS: This observational study organized 24 sessions of 2-hour remote training. Each session involved 3 trainees, 1 faculty member, and 1 host. A total of 70 trainees and 24 faculty members were enrolled. The participants joined the remote sessions via online videochat and performed focused training in coronary artery anastomosis using a commercially available simulator. A survey was conducted to validate the feasibility of the remote sessions. Performance improvement of the trainees who repeatedly participated (n = 13) were analyzed comparing initial and final scores of various performance indicators. RESULTS: All trainees and faculty members were satisfied with the efficacy of the remote session. Additionally, most trainees (79%) and faculty members (95%) agreed that the remote training sessions were equivalent to conventional onsite training seminars. A significant improvement between initial and last sessions was observed in the scoring components of near side (3.4 ± 1.0 vs 4.1 ± 0.9; P = .02), far side (3.3 ± 0.8 vs 3.9 ± 0.8; P = .03), external appearance (3.5 ± 0.8 vs 4.2 ± 0.7; P = .01), and internal appearance (2.8 ± 0.9 vs 4.0 ± 0.9; P = .004). CONCLUSIONS: Faculty-supervised remote surgical training sessions were executed with satisfactory results. This methodology may have important implications for surgical education during the coronavirus disease 2019 pandemic.

Coronary artery bypass graft surgery outcomes in the United States: Impact of the coronavirus disease 2019 (COVID-19) pandemic.

OBJECTIVE: There has been a substantial decline in patients presenting for emergent and routine cardiovascular care in the United States after the onset of the coronavirus disease 2019 (COVID-19) pandemic. We sought to assess the risk of adverse clinical outcomes among patients undergoing coronary artery bypass graft (CABG) surgery during the 2020 COVID-19 pandemic period and compare the risks with those undergoing CABG before the pandemic in the year 2019. METHODS: A retrospective cross-sectional analysis of the TriNetX Research Network database was performed. Patients undergoing CABG between January 20, 2019, and September 15, 2019, contributed to the 2019 cohort, and those undergoing CABG between January 20, 2020, and September 15, 2020, contributed to the 2020 cohort. Propensity-score matching was performed, and the odds of mortality, acute kidney injury, stroke, acute respiratory distress syndrome, and mechanical ventilation occurring by 30 days were evaluated. RESULTS: The number of patients undergoing CABG in 2020 declined by 35.5% from 5534 patients in 2019 to 3569 patients in 2020. After propensity-score matching, 3569 patient pairs were identified in the 2019 and the 2020 cohorts. Compared with those undergoing CABG in 2019, the odds of mortality by 30 days were 0.96 (95% confidence interval [CI], 0.69-1.33; P = .80) in those undergoing CABG in 2020. The odds for stroke (odds ratio [OR], 1.201; 95% CI, 0.96-1.39), acute kidney injury (OR, 0.76; 95% CI, 0.59-1.08), acute respiratory distress syndrome (OR, 1.01; 95% CI, 0.60-2.42), and mechanical ventilation (OR, 1.11; 95% CI, 0.94-1.30) were similar between the 2 cohorts. CONCLUSIONS: The number of patients undergoing CABG in 2020 has substantially declined compared with 2019. Similar odds of adverse clinical outcomes were seen among patients undergoing CABG in the setting of COVID-19 compared with those in 2019.

Incidence of acute myocardial infarction-related cardiogenic shock during corona virus disease 19 (COVID-19) pandemic.

AIMS: The hospitalization of patients with MI has decreased during global lockdown due to the COVID-19 pandemic. Whether this decrease is associated with more severe MI, e.g. MI-CS, is unknown. We aimed to examine the association of Corona virus disease (COVID-19) pandemic and incidence of acute myocardial infarction with cardiogenic shock (MI-CS). METHODS: On March 11, 2020, the Danish government announced national lock-down. Using Danish nationwide registries, we identified patients hospitalized with MI-CS. Incidence rates (IR) and incidence rate ratios (IRR) were used to compare MI-CS before and after March 11 in 2015-2019 and in 2020. RESULTS: We identified 11,769 patients with MI of whom 696 (5.9%) had cardiogenic shock in 2015-2019. In 2020, 2132 MI patients were identified of whom 119 had cardiogenic shock (5.6%). The IR per 100,000 person years before March 11 in 2015-2019 was 9.2 (95% CI: 8.3-10.2) and after 8.9 (95% CI: 8.0-9.9). In 2020, the IR was 7.5 (95% CI: 5.8-9.7) before March 11 and 7.7 (95% CI: 6.0-9.9) after. The IRRs comparing the 2020-period with the 2015-2019 period before and after March 11 (lockdown) were 0.81 (95% CI: 0.59-1.12) and 0.87 (95% CI: 0.57-1.32), respectively. The IRR comparing the 2020-period during and before lockdown was 1.02 (95% CI: 0.74-1.41). No difference in 7-day mortality or in-hospital management was observed between study periods. CONCLUSION: We could not identify a significant association of the national lockdown on the incidence of MI-CS, along with similar in-hospital management and mortality in patients with MI-CS.