Prediction of Patient Length of Stay on the Intensive Care Unit Following Cardiac Surgery: A Logistic Regression Analysis Based on the Cardiac Operative Mortality Risk Calculator, EuroSCORE

Meadows, Katherine; Gibbens, Richard J.; Gerrard, Caroline; Vuylsteke, Alain

doi:10.1053/j.jvca.2018.03.007

Cited by 41 publications

(36 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, nine predictive models were established to predict the progression of urinary protein in patients with chronic kidney disease, and model selection was based on several currently and frequently adopted predictive model types. For the linear model, the logistic regression model (LR) [14, 15], the elastic network model (Elastic Net) [16–18], the lasso regression model (Lasso) [19], and the ridge regression model (Ridge) were selected [20–22]. The neural network model (NN) [23] was chosen because it is an important class of nonlinear prediction models [24] and has been reported to predict CKD.…”

Section: Methodsmentioning

confidence: 99%

Comparison and development of machine learning tools in the prediction of chronic kidney disease progression

et al. 2019

View full text Add to dashboard Cite

Background Urinary protein quantification is critical for assessing the severity of chronic kidney disease (CKD). However, the current procedure for determining the severity of CKD is completed through evaluating 24-h urinary protein, which is inconvenient during follow-up. Objective To quickly predict the severity of CKD using more easily available demographic and blood biochemical features during follow-up, we developed and compared several predictive models using statistical, machine learning and neural network approaches. Methods The clinical and blood biochemical results from 551 patients with proteinuria were collected. Thirteen blood-derived tests and 5 demographic features were used as non-urinary clinical variables to predict the 24-h urinary protein outcome response. Nine predictive models were established and compared, including logistic regression, Elastic Net, lasso regression, ridge regression, support vector machine, random forest, XGBoost, neural network and k-nearest neighbor. The AU-ROC, sensitivity (recall), specificity, accuracy, log-loss and precision of each of the models were evaluated. The effect sizes of each variable were analysed and ranked. Results The linear models including Elastic Net, lasso regression, ridge regression and logistic regression showed the highest overall predictive power, with an average AUC and a precision above 0.87 and 0.8, respectively. Logistic regression ranked first, reaching an AUC of 0.873, with a sensitivity and specificity of 0.83 and 0.82, respectively. The model with the highest sensitivity was Elastic Net (0.85), while XGBoost showed the highest specificity (0.83). In the effect size analyses, we identified that ALB, Scr, TG, LDL and EGFR had important impacts on the predictability of the models, while other predictors such as CRP, HDL and SNA were less important. Conclusions Blood-derived tests could be applied as non-urinary predictors during outpatient follow-up. Features in routine blood tests, including ALB, Scr, TG, LDL and EGFR levels, showed predictive ability for CKD severity. The developed online tool can facilitate the prediction of proteinuria progress during follow-up in clinical practice. Electronic supplementary material The online version of this article (10.1186/s12967-019-1860-0) contains supplementary material, which is available to authorized users.

show abstract

Section: Methodsmentioning

confidence: 99%

Comparison and development of machine learning tools in the prediction of chronic kidney disease progression

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Model selection was based on several currently and frequently adopted predictive model types. For example, the linear LR model [64] and SVM model have been widely adopted in many clinical applications, such as for CKD disease prediction [65]. The DT model [66] is based on a radial basis function neural network and support vector machine coupled with firefly algorithm techniques; the XGboost and MLP models have also been used in clinical research [65,67].…”

Section: Model Construction In the Training Setmentioning

confidence: 99%

Comparison and development of machine learning tools for the prediction of chronic obstructive pulmonary disease in the Chinese population

Zhang³

et al. 2020

J Transl Med

View full text Add to dashboard Cite

Background: Chronic obstructive pulmonary disease (COPD) is a major public health problem and cause of mortality worldwide. However, COPD in the early stage is usually not recognized and diagnosed. It is necessary to establish a risk model to predict COPD development. Methods: A total of 441 COPD patients and 192 control subjects were recruited, and 101 single-nucleotide polymorphisms (SNPs) were determined using the MassArray assay. With 5 clinical features as well as SNPs, 6 predictive models were established and evaluated in the training set and test set by the confusion matrix AU-ROC, AU-PRC, sensitivity (recall), specificity, accuracy, F1 score, MCC, PPV (precision) and NPV. The selected features were ranked.

show abstract

“…[9,12] In one of the most largest study, including 18,377 consecutive patients, the AUC for prolonged hospitalization at ICU for both ESA and ESL was 0.73. [13] Atashi et al [8] indicated the importance of identifying reliable risk scores for prediction of prolonged hospitalization at ICU in a large systematic review published in 2018. They pointed difficulties in defining new risk factors and possible advantages of validating currently available risk scores.…”

Section: Discussionmentioning

confidence: 99%

Perioperative risk assessment with Euroscore and Euroscore II in patients with coronary artery or valvular disease

Czub¹,

Cacko²,

Gawałko

et al. 2018

Medicine

View full text Add to dashboard Cite

Nowadays, both the European System for Cardiac Operative Risk Evaluation (EuroSCORE) logistic (ESL) and EuroSCORE II (ESII) models are used worldwide in predicting in-hospital mortality after cardiac operation. However, these scales are based on different populations and represent different medical approaches. The aim of the study was to assess the effectiveness of the ESL and the ESII risk scores in predicting in-hospital death and prolonged hospitalization in intensive care unit (ICU) after coronary artery bypass grafting (CABG), aortic valve replacement (AVR), and mitral valve replacement (MVR) by comparison of an estimated risk and a real-life observation at a reference cardiac surgery unit.This retrospective study was based on medical records of patients who underwent a CABG, AVR, or MVR at a reference cardiac surgery unit in a 2-year period. Primary endpoint was defined as in-hospital death. Secondary endpoint was a prolonged hospitalization at the ICU, defined as longer than 3 days.The study encompassed 586 patients [114 (23.1%) female, mean age 65.8 ± 10.5 years], including 493 patients undergoing CABG, 66 patients undergoing AVR, and 27 patients undergoing MVR. The ESL and ESII risk scores were higher in MVR subgroup (31.7% ± 30.5% and 15.3% ± 19.4%) and AVR subgroup (9.7% ± 11.6% and 3.2% ± 4.2%) than in CABG subgroup (6.9% ± 10.4% and 2.5% ± 4.1%; P < .001). Subgroups of patients were significantly different in terms of clinical, biochemical, and echocardiography factors. Primary endpoint occurred in 36 (6.1%) patients: 21 (4.3%), 7 (10.6%), and 8 (29.7%) in CABG, AVR, and MVR subgroups, respectively. The ESII underestimated the risk of mortality. Secondary endpoint occurred in 210 (35.8%) patients: 172 (34.9%), 22 (33.4%), and 16 (59.3%) in CABG, AVR, and MVR subgroups, respectively.In the study, the perioperative risk estimated with the ESL and the ESII risk scores was compared with a real-life outcome among over 500 patients. Regardless of the type of surgery, result in the ESL was better correlated with the risk of in-hospital death.

show abstract

Prediction of Patient Length of Stay on the Intensive Care Unit Following Cardiac Surgery: A Logistic Regression Analysis Based on the Cardiac Operative Mortality Risk Calculator, EuroSCORE

Abstract: This analysis of an extensive data set shows that patient LOS in ICU after cardiac surgery in a single center can be predicted accurately using the simple cardiac operative risk scoring tool EuroSCORE. Using such predictions has the potential to improve ICU resource management.

Cited by 41 publications

References 20 publications

Comparison and development of machine learning tools in the prediction of chronic kidney disease progression

Comparison and development of machine learning tools in the prediction of chronic kidney disease progression

Comparison and development of machine learning tools for the prediction of chronic obstructive pulmonary disease in the Chinese population

Perioperative risk assessment with Euroscore and Euroscore II in patients with coronary artery or valvular disease

Contact Info

Product

Resources

About