Dataset Information

Machine Learning-Based Risk Model for Predicting Early Mortality After Surgery for Infective Endocarditis.

ABSTRACT: Background The early mortality after surgery for infective endocarditis is high. Although risk models help identify patients at high risk, most current scoring systems are inaccurate or inconvenient. The objective of this study was to construct an accurate and easy-to-use prediction model to identify patients at high risk of early mortality after surgery for infective endocarditis. Methods and Results A total of 476 consecutive patients with infective endocarditis who underwent surgery at 2 centers were included. The development cohort consisted of 276 patients. Eight variables were selected from 89 potential predictors as input of the XGBoost model to train the prediction model, including platelet count, serum albumin, current heart failure, urine occult blood ≥(++), diastolic dysfunction, multiple valve involvement, tricuspid valve involvement, and vegetation >10 mm. The completed prediction model was tested in 2 separate cohorts for internal and external validation. The internal test cohort consisted of 125 patients independent of the development cohort, and the external test cohort consisted of 75 patients from another center. In the internal test cohort, the area under the curve was 0.813 (95% CI, 0.670-0.933) and in the external test cohort the area under the curve was 0.812 (95% CI, 0.606-0.956). The area under the curve was significantly higher than that of other ensemble learning models, logistic regression model, and European System for Cardiac Operative Risk Evaluation II (all, P<0.01). This model was used to develop an online, open-access calculator (http://42.240.140.58:1808/). Conclusions We constructed and validated an accurate and robust machine learning-based risk model to predict early mortality after surgery for infective endocarditis, which may help clinical decision-making and improve outcomes.

SUBMITTER: Luo L

PROVIDER: S-EPMC9238722 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundDespite novel improvements in the diagnosis and treatment of infective endocarditis (IE), there has been no significant improvement in the survival rate of IE, which indicates that many details still need to be optimized in the preoperative assessment. We sought to evaluate preoperative serum albumin as a biomarker for predicting early mortality after IE surgery.MethodsBetween October 2013 and June 2019, patients with a definite diagnosis of IE were enrolled in this study. Patients' albumin levels at admission were used as the preoperative albumin levels. Restricted cubic spline and multivariate logistic regression analyses were performed to evaluate the relationship between albumin and early mortality. Receiver operating characteristic curve analyses were performed to assess the role of albumin in predicting early mortality and compare the predictive capacity of traditional models with models that included albumin.ResultsOf the 276 IE patients, 20 (7.2%) died in hospital or within 30 days of surgery. Hypoalbuminemia (an albumin level <3.5 g/dL) was present in 109 (39.5%) patients. The multivariate logistic regression analysis showed that preoperative albumin was inversely associated with early mortality [adjusted odds ratio (OR) =0.22 per 1 g/dL, 95% confidence interval (CI): 0.07-0.65, P=0.006] after full adjustment. Preoperative albumin had value in predicting early mortality [area under the curve (AUC) =0.72, 95% CI: 0.61-0.84; P<0.01]. After adding albumin to the European System for Cardiac Operative Risk Evaluation (EuroSCORE) and Charlson score, the predictive ability of the model was further improved (EuroSCORE II: AUC =0.55; 95% CI: 0.42-0.67 to AUC =0.72; 95% CI: 0.61-0.84; Charlson score: AUC =0.73; 95% CI: 0.64-0.83 to AUC =0.78; 95% CI: 0.68-0.88).ConclusionsPreoperative serum albumin is inversely associated with early mortality after IE surgery, and is a promising prognostic indicator in preoperative risk stratification assessments of IE patients.

Project description:To systematically review early surgery and the optimal timing of surgery in patients with infective endocarditis (IE), a search for foreign and domestic articles on cohort studies about the association between early surgery and infective endocarditis published from inception to January 2015 was conducted in the PubMed, EMBASE, Chinese Biomedical Literature (CBM), Wanfang and Chinese National Knowledge Infrastructure (CNKI) databases. The studies were screened according to the inclusion and exclusion criteria, the data were extracted and the quality of the method of the included studies was assessed. Then, the meta-analysis was performed using the Stata 12.0 software. Sixteen cohort studies, including 8141 participants were finally included. The results of the meta-analysis revealed that, compared with non-early surgery, early surgery in IE lowers the incidence of in-hospital mortality [odds ratio (OR) = 0.57, 95% confidence interval (CI) (0.42, 0.77); P = 0.000, I(2) = 73.1%] and long-term mortality [OR = 0.57, 95% CI (0.43, 0.77); P = 0.001, I(2) = 67.4%]. Further, performing operation within 2 weeks had a more favourable effect on long-term mortality [OR = 0.63, 95% CI (0.41, 0.97); P = 0.192, I(2) = 39.4%] than non-early surgery. In different kinds of IE, we found that early surgery for native valve endocarditis (NVE) had a lower in-hospital [OR = 0.46, 95% CI (0.31, 0.69); P = 0.001, I(2) = 73.0%] and long-term [OR = 0.57, 95% CI (0.40, 0.81); P = 0.001, I(2) = 68.9%] mortality than the non-early surgery group. However, for prosthetic valve endocarditis (PVE), in-hospital mortality did not differ significantly [OR = 0.83, 95% CI (0.65, 1.06); P = 0.413, I(2) = 0.0%] between early and non-early surgery. We concluded that early surgery was associated with lower in-hospital and long-term mortality compared with non-early surgical treatment for IE, especially in NVE. However, the optimal timing of surgery remains unclear. Additional larger prospective clinical trials will be required to clarify the optimal timing for surgical intervention and determine its efficacy in PVE.

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.

Dataset Information

Machine Learning-Based Risk Model for Predicting Early Mortality After Surgery for Infective Endocarditis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets