Comparison of the additive, logistic european system for cardiac operative risk (EuroSCORE) with the EuroSCORE 2 to predict mortality in high-risk cardiac surgery

External validation of the ARCH score in patients undergoing aortic arch reconstruction under circulatory arrest

BACKGROUND

Aortic arch surgery with hypothermic circulatory arrest (HCA) carries a higher risk of morbidity and mortality compared to routine cardiac surgical procedures. The newly developed ARCH (arch reconstruction under circulatory arrest with hypothermia) score has not been externally validated. We sought to externally validate this score in our local population.

METHODS

All consecutive open aortic arch surgeries with HCA performed between 2014 and 2023 were included. Univariable and multivariable analyses were performed. Model discrimination was assessed by the C-statistic with 95% confidence intervals as part of the receiver operating characteristic (ROC) curve analysis. Model performance was visualized by a calibration plot and quantified by the Brier score.

RESULTS

A total of 760 patients (38.3% females) were included. The mean age was 61 (±13.6) years, with 56.4% of patients' age >60 years. The procedures were carried out mostly emergently or urgently (59.6%). Total arch replacement was performed in 32.5% of the patients, and aortic root procedures were carried out in 74.6%. In-hospital death occurred in 64 patients (8.4%), and stroke occurred in 5.4%. The C-statistic revealed a low discriminatory ability for predicting in-hospital mortality (area under the ROC curve, 0.62; 95% confidence interval, 0.54-0.69; P = .002); however, model calibration was found to be excellent (Brier score of 0.07).

CONCLUSIONS

The ARCH score for in-hospital mortality showed low discriminatory ability in our local population, although with excellent ability for prediction of mortality.

Markers of Tissue Perfusion as Predictors of Adverse Outcomes in Patients with Left Ventricular Dysfunction Undergoing Coronary Artery Bypass Surgery

BACKGROUND

Cardiac surgery patients may be exposed to tissue hypoperfusion and anaerobic metabolism.

OBJECTIVE

To verify whether the biomarkers of tissue hypoperfusion have predictive value for prolonged intensive care unit (ICU) stay in patients with left ventricular dysfunction who underwent coronary artery bypass surgery.

METHODS

After approval by the institution's Ethics Committee and the signing of informed consent, 87 patients with left ventricular dysfunction (ejection fraction < 50%) undergoing coronary artery bypass surgery were enrolled. Hemodynamic and metabolic biomarkers were collected at five time points: after anesthesia, at the end of the surgery, at ICU admission, and at six and twelve hours after. An analysis of variance for repeated measures followed by a Bonferroni post hoc test was used for repeated, continuous variables (hemodynamic and metabolic variables) to determine differences between the two groups over the course of the study period. The level of statistical significance adopted was 5%.

RESULTS

Thirty-eight patients (43.7%) who presented adverse outcomes were older, higher Euro score (p<0.001), and elevated ΔpCO2 as analyzed 12 hours after ICU admission (p<0.01), while increased arterial lactate concentration at 6 hours postoperatively was found to be a negative predictive factor (p<0.01).

CONCLUSIONS

Euro SCORE, six-hour postoperative arterial lactate, 12-hour postoperative ΔPCO2, and eRQ are independent predictors of adverse outcomes in patients with left ventricular dysfunction after cardiac surgery.

Validation of the European system for cardiac operative risk evaluation II in a large Taiwan cardiac surgical centre

BACKGROUND

The European System for Cardiac Operative Risk Evaluation (EuroSCORE II) is a well-established scoring system for predicting mortality in cardiac surgery. This system was derived predominantly from a European patient cohort; however, no validation of this system has been conducted in Taiwan. We sought to assess the performance of EuroSCORE II at a tertiary centre.

METHODS

The 2161 adult patients receiving cardiac surgery between 2017 and 2020 in our institution were included.

RESULTS

Overall, the in-hospital mortality rate was 7.89%. The performance of EuroSCORE II was assessed using the area under the receiver operator curve (AUC) for discrimination and the Hosmer-Lemeshow (H-L) test for calibration. Data were analysed for type of surgery, risk stratification, and status of the operation. EuroSCORE II had good discriminative power (AUC=0.854, 95% Confidence Interval (CI): 0.822-0.885) and good calibration (χ2=5.19, p=0.82) for all types of surgery except ventricular assist devices (AUC=0.618, 95% CI: 0.497-0.738). EuroSCORE II also showed good calibration for most types of surgery except coronary artery bypass surgery (CABG) combined procedure (P=0.033), heart transplantation (HT) (P=0.017), and urgent operation (P=0.041). EuroSCORE II significantly underestimated the risk for CABG combined procedure and urgent operations, and overestimated the risk for HT.

CONCLUSION

EuroSCORE II had satisfactory discrimination and calibration power to predict surgical mortality in Taiwan. However, the model is poorly calibrated for CABG combined procedure, HT, urgent operation, and, likely, lower- and higher-risk patients.

Cardiac surgery risk prediction using ensemble machine learning to incorporate legacy risk scores: A benchmarking study

Objective

The introduction of new clinical risk scores (e.g. European System for Cardiac Operative Risk Evaluation (EuroSCORE) II) superseding original scores (e.g. EuroSCORE I) with different variable sets typically result in disparate datasets due to high levels of missingness for new score variables prior to time of adoption. Little is known about the use of ensemble learning to incorporate disparate data from legacy scores. We tested the hypothesised that Homogenenous and Heterogeneous Machine Learning (ML) ensembles will have better performance than ensembles of Dynamic Model Averaging (DMA) for combining knowledge from EuroSCORE I legacy data with EuroSCORE II data to predict cardiac surgery risk.

Methods

Using the National Adult Cardiac Surgery Audit dataset, we trained 12 different base learner models, based on two different variable sets from either EuroSCORE I (LogES) or EuroScore II (ES II), partitioned by the time of score adoption (1996-2016 or 2012-2016) and evaluated on holdout set (2017-2019). These base learner models were ensembled using nine different combinations of six ML algorithms to produce homogeneous or heterogeneous ensembles. Performance was assessed using a consensus metric.

Results

Xgboost homogenous ensemble (HE) was the highest performing model (clinical effectiveness metric (CEM) 0.725) with area under the curve (AUC) (0.8327; 95% confidence interval (CI) 0.8323-0.8329) followed by Random Forest HE (CEM 0.723; AUC 0.8325; 95%CI 0.8320-0.8326). Across different heterogenous ensembles, significantly better performance was obtained by combining siloed datasets across time (CEM 0.720) than building ensembles of either 1996-2011 (t-test adjusted, p = 1.67×10-6) or 2012-2019 (t-test adjusted, p = 1.35×10-193) datasets alone.

Conclusions

Both homogenous and heterogenous ML ensembles performed significantly better than DMA ensemble of Bayesian Update models. Time-dependent ensemble combination of variables, having differing qualities according to time of score adoption, enabled previously siloed data to be combined, leading to increased power, clinical interpretability of variables and usage of data.

Intraoperative hemoadsorption in high-risk patients with infective endocarditis

Background

Postoperative sepsis is an important cause of morbidity and mortality in patients with infective endocarditis undergoing surgical therapy. Blood purification using hemoadsorption therapy shows promising results in the treatment of sepsis. In this study, the clinical effects of intraoperative hemoadsorption in high-risk patients with infective endocarditis were evaluated.

Methods

Eligible candidates were high-risk patients with infective endocarditis undergoing cardiac surgery between January 2014 and December 2019. Patients with intraoperative hemoadsorption (hemoadsorption) were compared to patients without hemoadsorption (control). The endpoints were the incidence of postoperative sepsis, sepsis-associated death and in-hospital mortality. Additionally, postoperative vasopressor need, systemic vascular resistance indices and Sequential Organ Failure Assessment (SOFA) scores were compared.

Results

After propensity score matching, 70 high-risk patients were included. Postoperative sepsis occurred in 14 patients in the hemoadsorption group and in 16 patients in the control group, p = 0.629. Four patients died due to postoperative sepsis in the hemoadsorption group, while 11 postoperative septic patients died in the control group, p = 0.041. In-hospital mortality was 34% in the hemoadsorption group versus 43% in the control group, p = 0.461. On ICU-admission and the first postoperative day, the cumulative vasopressor need was 0.17 versus 0.25 μg/kgBW/min, p = 0.123 and 0.06 versus 0.11 μg/kgBW/min, p = 0.037, and the systemic vascular resistance index was 1448 versus 941 dyn·s·cm-5, p = 0.013 and 1156 versus 858 dyn·s·cm-5, p = 0.110 in the hemoadsorption versus control group, respectively. Postoperative course of SOFA score normalized significantly (p = 0.01) faster in the hemoadsorption group.

Conclusions

In high-risk cardiac surgical patients with infective endocarditis, intraoperative hemoadsorption significantly reduced sepsis-associated mortality. Furthermore, intraoperative hemoadsorption resulted in significant faster recovery of hemodynamics and organ function. Intraoperative hemoadsorption seems to attenuate the severity of postoperative sepsis.