Performance of contemporary surgical risk scores for mitral valve surgery

Wang, Tom Kai Ming; Harmos, Sophie; Gamble, Greg; Ramanathan, Tharumenthiran; Ruygrok, Peter

doi:10.1111/jocs.13107

Cited by 11 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our random forest model has a similar performance to EuroSCORE and EuroSCORE II in one study (AUC=0.72 for both) [50]. Even though our GBM model has the lowest performance out of these three in our study, it still outperforms the commonly used risk models in most aforementioned studies, apart from EuroSCORE, EuroSCORE II, STS [50] and Initial Parsonnet [43] in two studies.…”

Section: Discussionmentioning

confidence: 54%

Section: Discussionmentioning

confidence: 74%

“…Looking at the AUC, our adaboost model outperforms all aforementioned studies, apart from Parsonnet score in one study [43] and STS score another study [50], where these scores have a similar performance with the adaboost model (AUC=0.73). Our random forest model has a similar performance to EuroSCORE and EuroSCORE II in one study (AUC=0.72 for both) [50]. Even though our GBM model has the lowest performance out of these three in our study, it still outperforms the commonly used risk models in most aforementioned studies, apart from EuroSCORE, EuroSCORE II, STS [50] and Initial Parsonnet [43] in two studies.…”

Section: Discussionmentioning

confidence: 74%

“…All of these models were developed using logistic regression. In spite of these scores being initially developed to predominantly predict postoperative mortality, some studies have been carried out to assess the use of these scoring systems to predict combinations of postoperative complications [32, [43][44][45][46][47][48][49][50].…”

Section: Discussionmentioning

confidence: 99%

See 3 more Smart Citations

Evaluation of Random Forest and Ensemble Methods at Predicting Complications Following Cardiac Surgery

Lapp

Bouamrane

Kavanagh

et al. 2019

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Cardiac patients undergoing surgery face increased risk of postoperative complications, due to a combination of factors, including: higher risk surgery, their age at time of surgery and the presence of co-morbid conditions. They will therefore require high levels of care and clinical resources throughout their perioperative journey (i.e. before, during and after surgery). Although surgical mortality rates in the UK have remained low, postoperative complications on the other hand are common and can have a significant impact on patients' quality of life, increase hospital length of stay and healthcare costs. In this study we used and compared several machine learning methods -random forest, adaboost, gradient boosting model and stacking -to predict severe postoperative complications after cardiac surgery based on preoperative variables obtained from a surgical database of a large acute care hospital in Scotland. Our results show that adaboost has the best overall performance (AUC=0.731), however random forest (Sensitivity = 0.852, negative predictive value = 0.923) and gradient boosting model (Sensitivity = 0.875 and negative predictive value = 0.920) have the best performance at predicting severe postoperative complications based on sensitivity and negative predictive value.

show abstract

Section: Discussionmentioning

confidence: 54%

Section: Discussionmentioning

confidence: 74%

Section: Discussionmentioning

confidence: 74%

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Random Forest and Ensemble Methods at Predicting Complications Following Cardiac Surgery

Lapp

Bouamrane

Kavanagh

et al. 2019

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

show abstract

“…Advances in techniques utilizing annuloplasty rings and neochords have helped to make mitral valve repair the surgical procedure of choice for patients with degenerative disease resulting in mitral regurgitation . However, current risk‐scoring systems have not assessed the feasibility and outcomes of mitral valve repair in patients with native valve infective endocarditis (IE) …”

mentioning

confidence: 99%

Mitral valve repair for patients with infectious endocarditis-Is it the procedure of choice for all patients?

Lazar

2018

J Card Surg

View full text Add to dashboard Cite

Advances in techniques utilizing annuloplasty rings and neochords have helped to make mitral valve repair the surgical procedure of choice for patients with degenerative disease resulting in mitral regurgitation. 1 However, current risk-scoring systems have not assessed the feasibility and outcomes of mitral valve repair in patients with native valve infective endocarditis (IE). 2 In this edition of the Journal, Harky et al have sought to analyze the role of mitral valve repair versus replacement in patients with native valve IE by performing a meta-analysis to assess short-and long-term outcomes in 8978 patients. 3 Their meta-analysis is unique in that it reports not only outcomes for morbidity and mortality, but

show abstract

EuroSCORE II for Risk Evaluation and Predicting Cardiac Intensive Care Length of Stay in Indian Patients Undergoing Adult Cardiac Surgery

Kate,

Shouche,

Singh

et al. 2023

Intensive Care Res

View full text Add to dashboard Cite

Background Higher EuroSCORE II values are usually associated with increased postoperative morbidity and longer durations of Cardiac intensive care unit (CICU) stay following cardiac surgery. Aim/Purpose The aim is to investigate the predictive performance of EuroSCORE II for the Indian population and its relationship with ICU length of stay. Methods Prospective, observational study in 250 adult cardiac patients undergoing CABG under general Anaesthesia. Preoperatively values of EuroSCORE II was calculated by an online calculator available on www.euroscore.org. These patients were followed up after surgery for 30 days to note mortality and length of CICU stay. Results Of 250 cases studied, 39 (15.6%) had EuroSCORE II less than 1, 163 (65.2%) had EuroSCORE II between 1 and 3, 36 (14.4%) had EuroSCORE II between 3.1 and 5.0 and 12 (4.8%) had EuroSCORE II more than 5 in the study group. Mean EuroSCORE II and length of stay in CICU after surgery was 2.2 ± 1.4 and 4.2 ± 2.5 days respectively. The area under the curve (AUC) for EuroSCORE II as a sole predictor of mortality in the study group based on receiver operating characteristic curve (ROC) analysis was 0.919 (95% CI 0.86–0.97). Based on ROC analysis, AUC is significantly higher for predicting mortality (p < 0.001). Conclusion EuroSCORE II in Indian cardiac patients undergoing cardiac surgery is lesser than in European patients (mean 2.2 vs 3.7). Incidence of mortality is higher in patients with higher EuroSCORE II. As per our study patients with higher EuroSCORE II tend to stay longer in ICU.

show abstract

Performance of contemporary surgical risk scores for mitral valve surgery

Abstract: All scores discriminated mortality and most morbidities after MVR, although EuroSCORE over-estimated operative mortality. The STS Score was the best overall predictor of mortality and morbidity in the MVR cohort.

Cited by 11 publications

References 20 publications

Evaluation of Random Forest and Ensemble Methods at Predicting Complications Following Cardiac Surgery

Evaluation of Random Forest and Ensemble Methods at Predicting Complications Following Cardiac Surgery

Mitral valve repair for patients with infectious endocarditis-Is it the procedure of choice for all patients?

EuroSCORE II for Risk Evaluation and Predicting Cardiac Intensive Care Length of Stay in Indian Patients Undergoing Adult Cardiac Surgery

Contact Info

Product

Resources

About