Predictive ability of EuroSCORE II integrating cardiactroponin T in patients undergoing OPCABG

The clinical applications of ensemble machine learning based on the Bagging strategy for in-hospital mortality of coronary artery bypass grafting surgery

BACKGROUND

Machine learning (ML) has excelled after being introduced into the medical field. Ensemble ML models were able to integrate the advantages of several different ML models. This study compares the ensemble ML model's and EuroSCORE II's performance predicting in-hospital mortality in patients undergoing coronary artery bypass grafting surgery.

METHODS

The study included 4,764 patients from three heart centers between January 2007 and December 2021. Of these, 3812 patients were assigned to the modeling group, and 952 patients were assigned to the internal test group. Patients from other two heart center (1733 and 415 cases, respectively) constituted the external test group. The modeling set data are trained using each of the three ML strategies (XGBoost, CatBoost, and LightGBM), and the new model (XCL model) is constructed by integrating these three models through an ensemble ML strategy. Performance of different models in the three test groups comparative assessments were performed by calibration, discriminant, decision curve analysis, net reclassification index (NRI), integrated discriminant improvement (IDI), and Bland-Altman analysis.

RESULTS

In terms of discrimination, the XCL model performed the best with an impressive AUC value of 0.9145 in the internal validation group. The XCL model continued to perform best in both external test groups. The NRI and IDI suggested that the ML model showed positive improvements in all three test groups compared to EuroSCORE II.

CONCLUSIONS

ML models, particularly the XCL model, outperformed EuroSCORE II in predicting in-hospital mortality for CABG patients, with better discrimination, calibration, and clinical utility.

Association of Preoperative Copeptin Levels with Risk of All-Cause Mortality in a Prospective Cohort of Adult Cardiac Surgery Patients

We aimed to investigate the association of preoperative copeptin, a new cardiovascular biomarker, with short- and long-term mortality in a cohort of adult patients undergoing cardiac surgery, including its potential as a prognostic marker for clinical outcome. Preoperative blood samples of the Bern Perioperative Biobank, a prospective cohort of adults undergoing cardiac surgery during 2019, were analyzed. The primary and secondary outcome measures were 30-day and 1-year all-cause mortality. Optimal copeptin thresholds were calculated with the Youden Index. Associations of copeptin levels with the two outcomes were examined with multivariable logistic regression models; their discriminatory capacity was assessed with the area under the receiver operating characteristic (AUROC). A total of 519 patients (78.4% male, median age 67 y (IQR: 60-73 y)) were included, with a median preoperative copeptin level of 7.6 pmol/L (IQR: 4.7-13.2 pmol/L). We identified an optimal threshold of 15.9 pmol/l (95%-CI: 7.7 to 46.5 pmol/L) for 30-day mortality and 15.9 pmol/L (95%-CI: 9.0 to 21.3 pmol/L) for 1-year all-cause mortality. Regression models featured an AUROC of 0.79 (95%-CI: 0.56 to 0.95) for adjusted log-transformed preoperative copeptin for 30-day mortality and an AUROC of 0.76 (95%-CI: 0.64 to 0.88) for 1-year mortality. In patients undergoing cardiac surgery, the baseline levels of copeptin emerged as a strong marker for 1-year all-cause death. Preoperative copeptin levels might possibly identify patients at risk for a complicated, long-term postoperative course, and therefore requiring a more rigorous postoperative observation and follow-up.

Postoperative acute kidney injury increases short- and long-term death risks in elderly patients (≥ 75 years old) undergoing coronary artery bypass graft surgery

PURPOSE

To explore the incidence of postoperative acute kidney injury (AKI) after coronary artery bypass grafting (CABG) in elderly Chinese patients (≥ 75 years old) and its impacts on the short- and long-term prognosis.

METHODS

A total of 493 patients aged 75-88 years old who underwent CABG from two medical centers between January 2006 and October 2021 were involved. Perioperative (preoperative and 7 days after operation) serum creatinine (Scr) levels were measured in all the enrolled patients. Univariate and multivariate logistic regression analyses were conducted to explore the independent risk factors of postoperative in-hospital mortality. Kaplan-Meier curves and COX model were used to test the risk factors of all-cause death during follow-up. Propensity score matching was used to balance differences between AKI and control groups. The primary outcome event was in-hospital death, and the secondary outcome was all-cause death during follow-up.

RESULTS

The 198 patients were diagnosed with postoperative AKI. Intra-aortic balloon pump (IABP), cardiopulmonary bypass, and postoperative AKI were independent risk factors of in-hospital death. Gender, New York Heart Association Classification, preoperative eGFR, last eGFR within 7 days after operation, postoperative AKI, and postoperative renal function all impacted long-term prognosis. After 1:1 matching, 190 patients were included in the AKI and control groups. Use of IABP, use of cardiopulmonary bypass, and occurrence of postoperative AKI were still independent risk factors of in-hospital death. Preoperative eGFR, last eGFR within 7 days after operation, postoperative AKI and postoperative renal function all impacted long-term prognosis.

CONCLUSION

The incidence of postoperative AKI in elderly patients undergoing CABG is high, and postoperative AKI is an independent risk factor of both short- and long-term postoperative prognosis.

CAROTIDSCORE.RU-The first Russian computer program for risk stratification of postoperative complications of carotid endarterectomy

GOAL

Presentation of the first Russian computer program (www.carotidscore.ru) for risk stratification of postoperative complications of carotid endarterectomy (CEE).

MATERIAL AND METHODS

The present study is based on the analysis of a multicenter Russian database that includes 25,812 patients after CEE operated on from 01/01/2010 to 04/01/2022. The following types of CEE were implemented: 6814 classical CEE with plastic reconstruction of the reconstruction zone with a patch; 18,998 eversion CEE.

RESULTS

In the hospital postoperative period, 0.18% developed a lethal outcome, 0.14%-myocardial infarction, 0.35%-stroke. The combined endpoint was 0.68%. For each factor present in patients, a predictive coefficient was calculated. The prognostic coefficient was a numerical indicator reflecting the strength of the influence of each factor on the development of postoperative complications. Based on this formula, predictive coefficients were calculated for each factor present in patients in our study. The total contribution of these factors was reflected in "%" and denoted the risk of postoperative complications with a minimum value of 0% and a maximum of 100%. On the basis of the obtained calculations, a computer program CarotidSCORE was created. Its graphical interface is based on the QT framework (https://www.qt.io), which has established itself as one of the best solutions for desktop applications. It is possible not only to calculate the probability of developing a complication, but also to save all data about the patient in JSON format (for the patient's personal card and his anamnesis). The CarotidSCORE program contains 47 patient parameters, including clinical-demographic, anamnestic and angiographic characteristics. It allows you to choose one of the four types of CEE, which will provide an accurate stratification of the risk of complications for each of them in person.

CONCLUSION

CarotidSCORE (www.carotidscore.ru) is able to determine the likelihood of postoperative complications in patients undergoing CEE.

Machine-learning predictions for acute kidney injuries after coronary artery bypass grafting: a real-life muticenter retrospective cohort study

BACKGROUND

Acute kidney injury (AKI) after coronary artery bypass grafting (CABG) surgery is associated with poor outcomes. The objective of this study was to apply a new machine learning (ML) method to establish prediction models of AKI after CABG.

METHODS

A total of 2,780 patients from two medical centers in East China who underwent primary isolated CABG were enrolled. The dataset was randomly divided for model training (80%) and model testing (20%). Four ML models based on LightGBM, Support vector machine (SVM), Softmax and random forest (RF) algorithms respectively were established in Python. A total of 2,051 patients from two other medical centers were assigned to an external validation group to verify the performances of the ML prediction models. The models were evaluated using the area under the receiver operating characteristics curve (AUC), Hosmer-Lemeshow goodness-of-fit statistic, Bland-Altman plots, and decision curve analysis. The outcome of the LightGBM model was interpreted using SHapley Additive exPlanations (SHAP).

RESULTS

The incidence of postoperative AKI in the modeling group was 13.4%. Similarly, the incidence of postoperative AKI of the two medical centers in the external validation group was 8.2% and 13.6% respectively. LightGBM performed the best in predicting, with an AUC of 0.8027 in internal validation group and 0.8798 and 0.7801 in the external validation group. The SHAP revealed the top 20 predictors of postoperative AKI ranked according to the importance, and the top three features on prediction were the serum creatinine in the first 24 h after operation, the last preoperative Scr level, and body surface area.

CONCLUSION

This study provides a LightGBM predictive model that can make accurate predictions for AKI after CABG surgery. The LightGBM model shows good predictive ability in both internal and external validation. It can help cardiac surgeons identify high-risk patients who may experience AKI after CABG surgery.

Predictive Ability of European Heart Surgery Risk Assessment System II (EuroSCORE II) and the Society of Thoracic Surgeons (STS) Score for in-Hospital and Medium-Term Mortality of Patients Undergoing Coronary Artery Bypass Grafting

Objective

To evaluate the powers of European Heart Surgery Risk Assessment System II (EuroSCORE II) and the Society of Thoracic Surgeons (STS) score in predicting in-hospital and medium-term mortality of patients undergoing coronary artery bypass grafting (CABG).

Methods

Totally 1628 Chinese patients were included between January 2000 and January 2018. Their perioperative clinical data were collected and the patients were closely followed up. According to the length of follow-up time, the total cohort was divided into 1-year, 2-year, 3-year, 4-year and 5-year groups. The in-hospital and medium-term risk prediction of EuroSCORE II and STS score were comparatively assessed by calibration, discrimination, decision curve analysis (DCA), net reclassification index (NRI), integrated discrimination improvement (IDI) and Bland-Altman analysis.

Results

About 36 (2.21%) patients died during hospitalization. Both EuroSCORE II and STS score performed extremely well in predicting in-hospital mortality (area under curve = 0.900 and 0.879, respectively). However, calibration and discrimination analyses showed gradual decrease when these two risk evaluation systems were used to predict mortality during the follow-up period. At the same time, the predictive ability of EuroSCORE II was better than STS score. DCA curves showed that the performances of the two evaluation systems were roughly equal between the threshold probability of 0% to 20%. The percentage of correct reclassification of EuroSCORE II was 21.64% higher than that of STS score in predicting 2-year postoperative mortality. The IDI index showed that the predictive capabilities of these two systems were roughly equivalent. Bland-Altman analysis showed no significant difference between the values of the two systems.

Conclusion

EuroSCORE II and STS score have excellent predictive powers in predicting in-hospital mortality of patients undergoing CABG. In particular, EuroSCORE II is superior in calibration and discrimination. The prediction efficiency of the two risk evaluation systems is still acceptable for two-year postoperative mortality, but decreases year by year.

Effects of peak time of myocardial injury biomarkers on mid-term outcomes of patients undergoing OPCABG

BACKGROUND

With the development of cardiac surgery techniques, myocardial injury is gradually reduced, but cannot be completely avoided. Myocardial injury biomarkers (MIBs) can quickly and specifically reflect the degree of myocardial injury. Due to various reasons, there is no consensus on the specific values of MIBs in evaluating postoperative prognosis. This retrospective study was aimed to investigate the impact of MIBs on the mid-term prognosis of patients undergoing off-pump coronary artery bypass grafting (OPCABG).

METHODS

Totally 564 patients undergoing OPCABG with normal courses were included. Cardiac troponin T (cTnT) and creatine kinase myocardial band (CK-MB) were assessed within 48 h before operation and at 6, 12, 24, 48, 72, 96 and 120 h after operation. Patients were grouped by peak values and peak time courses of MIBs. The profile of MIBs and clinical variables as well as their correlations with mid-term prognosis were analyzed by univariable and multivariable Cox regression models.

RESULT

Continuous assessment showed that MIBs increased first (12 h after surgery) and then decreased. The peak cTnT and peak CK-MB occurred within 24 h after operation in 76.8% and 67.7% of the patients respectively. No significant correlation was found between CK-MB and mid-term mortality. Delayed cTnT peak (peak cTnT elevated after 24 h after operation) was correlated with lower creatinine clearance rate (69.36 ± 21.67 vs. 82.18 ± 25.17 ml/min/1.73 m²), body mass index (24.35 ± 2.58 vs. 25.27 ± 3.26 kg/m²), less arterial grafts (1.24 ± 0.77 vs. 1.45 ± 0.86), higher EuroSCORE II (2.22 ± 1.12 vs.1.72 ± 0.91) and mid-term mortality (26.5 vs.7.9%). Age (HR: 1.067, CI: 1.006-1.133), left ventricular ejection fraction (HR: 0.950, CI: 0.910-0.993), New York Heart Association score (HR: 1.839, CI: 1.159-2.917), total venous grafting (HR: 2.833, CI: 1.054-7.614) and cTnT peak occurrence within 24 h (HR: 0.362, CI: 0.196-0.668) were independent predictors of mid-term mortality.

CONCLUSION

cTnT is a better indicator than CK-MB. The peak value and peak occurrence of cTnT are related to mid-term mortality in patients undergoing OPCABG, and the peak phases have stronger predictive ability.

TRIAL REGISTRATION

Chinese Clinical Trial Registry, ChiCTR2000033850. Registered 14 June 2020, http://www.chictr.org.cn/edit.aspx?pid=55162&htm=4 .