Dataset Information

A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

ABSTRACT:

Background

The benefits of cardiac surgery are sometimes difficult to predict and the decision to operate on a given individual is complex. Machine Learning and Decision Curve Analysis (DCA) are recent methods developed to create and evaluate prediction models.

Methods and finding

We conducted a retrospective cohort study using a prospective collected database from December 2005 to December 2012, from a cardiac surgical center at University Hospital. The different models of prediction of mortality in-hospital after elective cardiac surgery, including EuroSCORE II, a logistic regression model and a machine learning model, were compared by ROC and DCA. Of the 6,520 patients having elective cardiac surgery with cardiopulmonary bypass, 6.3% died. Mean age was 63.4 years old (standard deviation 14.4), and mean EuroSCORE II was 3.7 (4.8) %. The area under ROC curve (IC95%) for the machine learning model (0.795 (0.755-0.834)) was significantly higher than EuroSCORE II or the logistic regression model (respectively, 0.737 (0.691-0.783) and 0.742 (0.698-0.785), p < 0.0001). Decision Curve Analysis showed that the machine learning model, in this monocentric study, has a greater benefit whatever the probability threshold.

Conclusions

According to ROC and DCA, machine learning model is more accurate in predicting mortality after elective cardiac surgery than EuroSCORE II. These results confirm the use of machine learning methods in the field of medical prediction.

SUBMITTER: Allyn J

PROVIDER: S-EPMC5218502 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

Allyn Jérôme J Allou Nicolas N Augustin Pascal P Philip Ivan I Martinet Olivier O Belghiti Myriem M Provenchere Sophie S Montravers Philippe P Ferdynus Cyril C

PloS one 20170106 1

<h4>Background</h4>The benefits of cardiac surgery are sometimes difficult to predict and the decision to operate on a given individual is complex. Machine Learning and Decision Curve Analysis (DCA) are recent methods developed to create and evaluate prediction models.<h4>Methods and finding</h4>We conducted a retrospective cohort study using a prospective collected database from December 2005 to December 2012, from a cardiac surgical center at University Hospital. The different models of predic ...[more]

PMID: 28060903

Similar Datasets

Project description:ObjectiveThis study aims to construct and validate several machine learning (ML) algorithms to predict long-term mortality and identify risk factors in unselected patients post-cardiac surgery.MethodsThe Medical Information Mart for Intensive Care (MIMIC-III) database was used to perform a retrospective administrative database study. Candidate predictors consisted of the demographics, comorbidity, vital signs, laboratory test results, scoring systems, and treatment information on the first day of ICU admission. Four-year mortality was set as the study outcome. We used the ML methods of logistic regression (LR), artificial neural network (NNET), naïve bayes (NB), gradient boosting machine (GBM), adapting boosting (Ada), random forest (RF), bagged trees (BT), and eXtreme Gradient Boosting (XGB). The prognostic capacity and clinical utility of these ML models were compared using the area under the receiver operating characteristic curves (AUC), calibration curves, and decision curve analysis (DCA).ResultsOf 7,368 patients in MIMIC-III included in the final cohort, a total of 1,337 (18.15%) patients died during a 4-year follow-up. Among 65 variables extracted from the database, a total of 25 predictors were selected using recursive feature elimination and included in the subsequent analysis. The Ada model performed best among eight models in both discriminatory ability with the highest AUC of 0.801 and goodness of fit (visualized by calibration curve). Moreover, the DCA shows that the net benefit of the RF, Ada, and BT models surpassed that of other ML models for almost all threshold probability values. Additionally, through the Ada technique, we determined that red blood cell distribution width (RDW), blood urea nitrogen (BUN), SAPS II, anion gap (AG), age, urine output, chloride, creatinine, congestive heart failure, and SOFA were the Top 10 predictors in the feature importance rankings.ConclusionsThe Ada model performs best in predicting 4-year mortality after cardiac surgery among the eight ML models, which might have significant application in the development of early warning systems for patients following operations.

Project description:ImportanceA variety of perioperative risk factors are associated with postoperative mortality risk. However, the relative contribution of routinely collected intraoperative clinical parameters to short-term and long-term mortality remains understudied.ObjectiveTo examine the performance of multiple machine learning models with data from different perioperative periods to predict 30-day, 1-year, and 5-year mortality and investigate factors that contribute to these predictions.Design, setting, and participantsIn this prognostic study using prospectively collected data, risk prediction models were developed for short-term and long-term mortality after cardiac surgery. Included participants were adult patients undergoing a first-time valve operation, coronary artery bypass grafting, or a combination of both between 1997 and 2017 in a single center, the University Medical Centre Groningen in the Netherlands. Mortality data were obtained in November 2017. Data analysis took place between February 2020 and August 2021.ExposureCardiac surgery.Main outcomes and measuresPostoperative mortality rates at 30 days, 1 year, and 5 years were the primary outcomes. The area under the receiver operating characteristic curve (AUROC) was used to assess discrimination. The contribution of all preoperative, intraoperative hemodynamic and temperature, and postoperative factors to mortality was investigated using Shapley additive explanations (SHAP) values.ResultsData from 9415 patients who underwent cardiac surgery (median [IQR] age, 68 [60-74] years; 2554 [27.1%] women) were included. Overall mortality rates at 30 days, 1 year, and 5 years were 268 patients (2.8%), 420 patients (4.5%), and 612 patients (6.5%), respectively. Models including preoperative, intraoperative, and postoperative data achieved AUROC values of 0.82 (95% CI, 0.78-0.86), 0.81 (95% CI, 0.77-0.85), and 0.80 (95% CI, 0.75-0.84) for 30-day, 1-year, and 5-year mortality, respectively. Models including only postoperative data performed similarly (30 days: 0.78 [95% CI, 0.73-0.82]; 1 year: 0.79 [95% CI, 0.74-0.83]; 5 years: 0.77 [95% CI, 0.73-0.82]). However, models based on all perioperative data provided less clinically usable predictions, with lower detection rates; for example, postoperative models identified a high-risk group with a 2.8-fold increase in risk for 5-year mortality (4.1 [95% CI, 3.3-5.1]) vs an increase of 11.3 (95% CI, 6.8-18.7) for the high-risk group identified by the full perioperative model. Postoperative markers associated with metabolic dysfunction and decreased kidney function were the main factors contributing to mortality risk.Conclusions and relevanceThis study found that the addition of continuous intraoperative hemodynamic and temperature data to postoperative data was not associated with improved machine learning-based identification of patients at increased risk of short-term and long-term mortality after cardiac operations.

Project description:Background Frailty predicts an increased risk of postoperative morbidity and mortality. Comparison of the predictive performance between two deficit accumulation models of frailty, the modified frailty index (mFI) and the revised-Risk Analysis Index (RAI-rev), is poorly understood. This study compared the predictive abilities of the above two frailty indices in predicting life-threatening morbidity and mortality among older patients following elective high-risk abdominal surgery. Methods This retrospective cohort study extracted perioperative data of older patients (age ≥65 years) undergoing elective high-risk abdominal surgery at a single institution between January 2018 and December 2020. Preoperative frailty was screened by mFI and RAI-rev scoring systems. The primary outcome was the composite of postoperative life-threatening morbidity and mortality during hospitalization. Multivariable logistic regression analyses were performed to investigate the association of the two frailty indices with the primary outcome. Receiver-operating characteristic (ROC) curve was employed to test the predictive performances of the two frailty instruments in predicting the composite primary outcome. The difference between the area under the curves (AUCs) was assessed by DeLong’s test. Results 1,132 older patients (mean age, 73.4 ± 6.2 years; 63.9% male) were included. Of these, 107 (9.5%) developed postoperative life-threatening morbidity and mortality. In multivariable logistic regression analyses, rising continuous frailty scores (mFI: adjusted OR 1.319 per 0.09-point increase in score, 95% CI 1.151–1.511, p < 0.001; RAI-rev: adjusted OR 1.052 per 1-point increase in score, 95% CI 1.018–1.087, p = 0.002) as well as dichotomized frailty measures (mFI ≥0.27: adjusted OR 2.059, 95% CI 1.328–3.193, p = 0.001; RAI-rev ≥45: adjusted OR 1.862, 95% CI 1.188–2.919, p = 0.007) were associated with increased odds of the primary outcome separately. ROC curve analysis showed that the discrimination of mFI and RAI-rev scores for the life-threatening morbidity and mortality was poor and comparable (AUC: 0.598 [95% CI 0.569–0.627] vs. 0.613 [95% CI 0.583–0.641]; DeLong’s test: Z = 0.375, p = 0.7075). Conclusion High mFI and RAI-rev scores were associated with an increased risk of life-threatening morbidity and mortality in older patients undergoing elective high-risk abdominal surgery. However, both frailty indices displayed poor discrimination for postoperative life-threatening morbidity and mortality.

Project description:ObjectivesPlasma troponins, measured by fourth-generation assays, are associated with increased mortality and morbidity after cardiac surgery. They also offer predictive information in addition to EuroSCORE, a widely used risk model after cardiac surgery. However, preoperatively measured troponin has provided no additional information to postoperative values. Whether these facts hold true also for the high-sensitivity fifth-generation troponin assay and the better calibrated risk model, EuroSCORE II, is unknown. We hypothesized that preoperative and/or postoperative high-sensitivity troponin T (hs-TnT) would increase the predictive value of EuroSCORE II.MethodsConsecutive coronary artery bypass grafting (CABG) and other cardiac surgical patients were prospectively enrolled in a university hospital. Plasma samples and EuroSCORE II variables were collected. The primary and secondary end-points were 180-day mortality and any major adverse event within 30 days, and 961-day mortality. The data were analysed by Kaplan-Meier survival curves, regression analyses, receiver operator characteristic curves and decision curve analysis.ResultsMortality rates in 180 days were 3.5% (15/428) in CABG and 6.4% (14/220) in other cardiac surgical patients. Survival curves differed only in patients with not only high postoperative hs-TnT value (>500 ng/l), but also high preoperative hs-TnT value (>14 ng/l), compared with patients with both hs-TnT values low. Adding hs-TnT to EuroSCORE II improved the prediction of 180-day mortality in other cardiac surgical patients (maximum net benefit of 1.5%), but not in CABG patients. Regarding major adverse events, adding hs-TnT to EuroSCORE II improved the prediction in both CABG patients and other cardiac surgical patients (maximum net benefits of 3 and 7%).ConclusionsElevated postoperative hs-TnT was predictive of mortality only when combined with elevated preoperative hs-TnT. Hs-TnT measurements added information to the EuroSCORE II regarding major adverse events in all cardiac surgical patients and regarding 180-day mortality in non-CABG patients.

Dataset Information

A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

Background

Methods and finding

Conclusions

Publications

A Comparison of a Machine Learning Model with EuroSCORE II in Predicting Mortality after Elective Cardiac Surgery: A Decision Curve Analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets