Prediction models for prolonged intensive care unit stay after cardiac surgery: systematic review and validation study

Risk Factors Associated With Prolonged Mechanical Ventilation and Length of Stay After Repair of Tetralogy of Fallot

BACKGROUND

This study examined preoperative, intraoperative, and postoperative data to identify factors that are associated with prolonged mechanical ventilation (PMV) and prolonged intensive care unit length of stay (ICU LOS) in tetralogy of Fallot (TOF) patients undergoing repair surgery.

METHODS

A retrospective study was carried out after approval from the institutional review board. All patients (age 0-52 years) who underwent TOF repair from January 2016 to September 2022 were included. Prolonged mechanical ventilation was defined as >24 h of ventilation, while prolonged ICU LOS was defined as ICU stay >3 days.

RESULTS

A total of 922 patients were included, among whom 288 (31.2%) were intubated for >24 h and 222 (24.1%) stayed in ICU for >3 days. Younger age (odds ratio [OR] = 2, 95% confidence interval [CI] 1.2-3.3, P = .007), lower weight (OR = 2.1, 95% CI 1.2-3.5, P = .003), and residual lesion (OR = 3.27, 95% CI 1.2-8.7, P = .017) were associated with PMV. Moreover, independent risk factors for prolonged ICU LOS are similar to PMV risk factors, including younger age (OR = 2.3, 95% CI 1.28-4.12, P = .005), lower weight (OR = 2.83, 95% CI 1.58-5, P < .001), underweight status (OR = 1.7, 95% CI 1.12-2.57, P = .012), and residual lesion (OR = 3.79, 95% CI 1.43-10.05, P = .007). Both aortic cross-clamp and cardiopulmonary bypass times did not exhibit clinically significant risk factors toward PMV and prolonged ICU LOS.

CONCLUSIONS

The risk factors for PMV and prolonged ICU LOS were residual lesion, younger age, and lower weight. Nutritional status contributed to the risk of prolonged ICU LOS, but not PMV. Consideration of these factors may provide optimal care to improve the outcome following TOF corrective surgery.

Surgical scheduling via optimization and machine learning with long-tailed data : Health care management science, in press

Using data from cardiovascular surgery patients with long and highly variable post-surgical lengths of stay (LOS), we develop a modeling framework to reduce recovery unit congestion. We estimate the LOS and its probability distribution using machine learning models, schedule procedures on a rolling basis using a variety of optimization models, and estimate performance with simulation. The machine learning models achieved only modest LOS prediction accuracy, despite access to a very rich set of patient characteristics. Compared to the current paper-based system used in the hospital, most optimization models failed to reduce congestion without increasing wait times for surgery. A conservative stochastic optimization with sufficient sampling to capture the long tail of the LOS distribution outperformed the current manual process and other stochastic and robust optimization approaches. These results highlight the perils of using oversimplified distributional models of LOS for scheduling procedures and the importance of using optimization methods well-suited to dealing with long-tailed behavior.

Prediction of in-hospital death following acute type A aortic dissection

Background

Our goal was to create a prediction model for in-hospital death in Chinese patients with acute type A aortic dissection (ATAAD).

Methods

A retrospective derivation cohort was made up of 340 patients with ATAAD from Tianjin, and the retrospective validation cohort was made up of 153 patients with ATAAD from Nanjing. For variable selection, we used least absolute shrinkage and selection operator analysis, and for risk scoring, we used logistic regression coefficients. We categorized the patients into low-, middle-, and high-risk groups and looked into the correlation with in-hospital fatalities. We established a risk classifier based on independent baseline data using a multivariable logistic model. The prediction performance was determined based on the receiver operating characteristic curve (ROC). Individualized clinical decision-making was conducted by weighing the net benefit in each patient by decision curve analysis (DCA).

Results

We created a risk prediction model using risk scores weighted by five preoperatively chosen variables [AUC: 0.7039 (95% CI, 0.643–0.765)]: serum creatinine (Scr), D-dimer, white blood cell (WBC) count, coronary heart disease (CHD), and blood urea nitrogen (BUN). Following that, we categorized the cohort's patients as low-, intermediate-, and high-risk groups. The intermediate- and high-risk groups significantly increased hospital death rates compared to the low-risk group [adjusted OR: 3.973 (95% CI, 1.496–10.552), P &lt; 0.01; 8.280 (95% CI, 3.054–22.448), P &lt; 0.01, respectively). The risk score classifier exhibited better prediction ability than the triple-risk categories classifier [AUC: 0.7039 (95% CI, 0.6425–0.7652) vs. 0.6605 (95% CI, 0.6013–0.7197); P = 0.0022]. The DCA showed relatively good performance for the model in terms of clinical application if the threshold probability in the clinical decision was more than 10%.

Conclusion

A risk classifier is an effective strategy for predicting in-hospital death in patients with ATAAD, but it might be affected by the small number of participants.

The Need to Prioritize Model-Updating Processes in Clinical Artificial Intelligence (AI) Models: Protocol for a Scoping Review

BACKGROUND

With an increase in the number of artificial intelligence (AI) and machine learning (ML) algorithms available for clinical settings, appropriate model updating and implementation of updates are imperative to ensure applicability, reproducibility, and patient safety.

OBJECTIVE

The objective of this scoping review was to evaluate and assess the model-updating practices of AI and ML clinical models that are used in direct patient-provider clinical decision-making.

METHODS

We used the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist and the PRISMA-P protocol guidance in addition to a modified CHARMS (Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies) checklist to conduct this scoping review. A comprehensive medical literature search of databases, including Embase, MEDLINE, PsycINFO, Cochrane, Scopus, and Web of Science, was conducted to identify AI and ML algorithms that would impact clinical decision-making at the level of direct patient care. Our primary end point is the rate at which model updating is recommended by published algorithms; we will also conduct an assessment of study quality and risk of bias in all publications reviewed. In addition, we will evaluate the rate at which published algorithms include ethnic and gender demographic distribution information in their training data as a secondary end point.

RESULTS

Our initial literature search yielded approximately 13,693 articles, with approximately 7810 articles to consider for full reviews among our team of 7 reviewers. We plan to complete the review process and disseminate the results by spring of 2023.

CONCLUSIONS

Although AI and ML applications in health care have the potential to improve patient care by reducing errors between measurement and model output, currently there exists more hype than hope because of the lack of proper external validation of these models. We expect to find that the AI and ML model-updating methods are proxies for model applicability and generalizability on implementation. Our findings will add to the field by determining the degree to which published models meet the criteria for clinical validity, real-life implementation, and best practices to optimize model development, and in so doing, reduce the overpromise and underachievement of the contemporary model development process.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

PRR1-10.2196/37685.

The reporting of prognostic prediction models for obstetric care was poor: a cross-sectional survey of 10-year publications

BACKGROUND

To investigate the reporting of prognostic prediction model studies in obstetric care through a cross-sectional survey design.

METHODS

PubMed was searched to identify prognostic prediction model studies in obstetric care published from January 2011 to December 2020. The quality of reporting was assessed by the TRIPOD checklist. The overall adherence by study and the adherence by item were calculated separately, and linear regression analysis was conducted to explore the association between overall adherence and prespecified study characteristics.

RESULTS

A total of 121 studies were included, while no study completely adhered to the TRIPOD. The results showed that the overall adherence was poor (median 46.4%), and no significant improvement was observed after the release of the TRIPOD (43.9 to 46.7%). Studies including both model development and external validation had higher reporting quality versus those including model development only (68.1% vs. 44.8%). Among the 37 items required by the TRIPOD, 10 items were reported adequately with an adherence rate over of 80%, and the remaining 27 items had an adherence rate ranging from 2.5 to 79.3%. In addition, 11 items had a report rate lower than 25.0% and even covered key methodological aspects, including blinding assessment of predictors (2.5%), methods for model-building procedures (4.5%) and predictor handling (13.5%), how to use the model (13.5%), and presentation of model performance (14.4%).

CONCLUSIONS

In a 10-year span, prognostic prediction studies in obstetric care continued to be poorly reported and did not improve even after the release of the TRIPOD checklist. Substantial efforts are warranted to improve the reporting of obstetric prognostic prediction models, particularly those that adhere to the TRIPOD checklist are highly desirable.

Preoperative red cell distribution width to lymphocyte ratio as biomarkers for prolonged intensive care unit stay among older patients undergoing cardiac surgery: a retrospective longitudinal study

Introduction: Our aim was to use the red cell distribution width-lymphocyte ratio (RLR) as a novel biomarker to predict prolonged intensive care unit (ICU) length of stay (LOS) among older patients undergoing cardiovascular surgery. Methods: This longitudinal study included older patients admitted to a tertiary cardiovascular surgery hospital between January 2017 and January 2022. Results: A total of 574 patients were studied, including 83 patients (14.5%) who had prolonged ICU LOS and 471 (85.5%) control subjects. After adjustment for the European System for Cardiac Operative Risk Evaluation 2, the RLR score showed a 10% increased risk of prolonged ICU LOS (odds ratio: 1.10; CI: 1.05-1.16; p = 0.01). Conclusion: Preoperative RLR can be used to predict the risk of long-term intensive care stay in older cardiac surgery patients.

Composite Outcomes in Clinical Prediction Modeling: Are We Trying to Predict Apples and Oranges?

Composite outcomes are widely used in clinical research. Existing literature has considered the pros and cons of composite outcomes in clinical trials, but their extensive use in clinical prediction has received much less attention. Clinical prediction assists decision-making by directing patients with higher risks of adverse outcomes toward interventions that provide the greatest benefits to those at the greatest risk. In this article, we summarize our existing understanding of the advantages and disadvantages of composite outcomes, consider how these relate to clinical prediction, and highlight the problem of key predictors having markedly different associations with individual components of the composite outcome. We suggest that a "composite outcome fallacy" may occur when a clinical prediction model is based on strong associations between key predictors and one component of a composite outcome (such as mortality) and used to direct patients toward intervention when these predictors actually have an inverse association with a more relevant component of the composite outcome (such as the use of a lifesaving intervention). We propose that clinical prediction scores using composite outcomes should report their accuracy for key components of the composite outcome and examine for inconsistencies among predictor variables.

Selective prediction-set models with coverage rate guarantees

The current approach to using machine learning (ML) algorithms in healthcare is to either require clinician oversight for every use case or use their predictions without any human oversight. We explore a middle ground that lets ML algorithms abstain from making a prediction to simultaneously improve their reliability and reduce the burden placed on human experts. To this end, we present a general penalized loss minimization framework for training selective prediction-set (SPS) models, which choose to either output a prediction set or abstain. The resulting models abstain when the outcome is difficult to predict accurately, such as on subjects who are too different from the training data, and achieve higher accuracy on those they do give predictions for. We then introduce a model-agnostic, statistical inference procedure for the coverage rate of an SPS model that ensembles individual models trained using K-fold cross-validation. We find that SPS ensembles attain prediction-set coverage rates closer to the nominal level and have narrower confidence intervals for its marginal coverage rate. We apply our method to train neural networks that abstain more for out-of-sample images on the MNIST digit prediction task and achieve higher predictive accuracy for ICU patients compared to existing approaches.

Predicting Prolonged Length of ICU Stay through Machine Learning

This study aimed to construct machine learning (ML) models for predicting prolonged length of stay (pLOS) in intensive care units (ICU) among general ICU patients. A multicenter database called eICU (Collaborative Research Database) was used for model derivation and internal validation, and the Medical Information Mart for Intensive Care (MIMIC) III database was used for external validation. We used four different ML methods (random forest, support vector machine, deep learning, and gradient boosting decision tree (GBDT)) to develop prediction models. The prediction performance of the four models were compared with the customized simplified acute physiology score (SAPS) II. The area under the receiver operation characteristic curve (AUROC), area under the precision-recall curve (AUPRC), estimated calibration index (ECI), and Brier score were used to measure performance. In internal validation, the GBDT model achieved the best overall performance (Brier score, 0.164), discrimination (AUROC, 0.742; AUPRC, 0.537), and calibration (ECI, 8.224). In external validation, the GBDT model also achieved the best overall performance (Brier score, 0.166), discrimination (AUROC, 0.747; AUPRC, 0.536), and calibration (ECI, 8.294). External validation showed that the calibration curve of the GBDT model was an optimal fit, and four ML models outperformed the customized SAPS II model. The GBDT-based pLOS-ICU prediction model had the best prediction performance among the five models on both internal and external datasets. Furthermore, it has the potential to assist ICU physicians to identify patients with pLOS-ICU risk and provide appropriate clinical interventions to improve patient outcomes.

Can soluble urokinase plasminogen receptor predict outcomes after cardiac surgery?

OBJECTIVES

Soluble urokinase plasminogen activator receptor (suPAR) is a biomarker that has been implicated in several cardiac pathologies and has been shown to be elevated in critically ill populations. We measured plasma suPAR in a cohort of cardiac surgical patients to evaluate its ability to predict prolonged intensive care unit (ICU) and hospital length of stay and development of complications following surgery. We compared suPAR against EuroSCORE II and C-reactive protein (CRP).

METHODS

Ninety patients undergoing cardiac surgery were recruited with samples taken preoperatively and on postoperative days 1, 2 and 3. suPAR was measured using enzyme-linked immunosorbent assay. Area under the receiver operator curve (AUROC) was used to test predictive capability of suPAR. Comparison was made with EuroSCORE II and CRP.

RESULTS

suPAR increased over time (P < 0.001) with higher levels in patients requiring prolonged ICU and hospital stay, and prolonged ventilation (P < 0.05). suPAR was predictive for prolonged ICU and hospital stay, and prolonged ventilation at all time points (AUROC 0.66-0.74). Interestingly, this association was also observed preoperatively, with preoperative suPAR predicting prolonged ICU (AUROC 0.66), and hospital stay (AUROC 0.67) and prolonged ventilation (AUROC 0.74). The predictive value of preoperative suPAR compared favourably to EuroSCORE II and CRP.

CONCLUSIONS

suPAR increases following cardiac surgery and levels are higher in those who require prolonged ICU stay, prolonged hospital stay and prolonged ventilation. Preoperative suPAR compares favourably to EuroSCORE II and CRP in the prediction of these outcomes. suPAR could be a useful biomarker in predicting outcome following cardiac surgery, helping inform clinical decision-making.

CLINICAL REGISTRATION

West of Scotland Research Ethics Committee Reference: 12/WS/0179 (AM01).

Predicting Intensive Care Unit Length of Stay After Acute Type A Aortic Dissection Surgery Using Machine Learning

Background: Patients with acute type A aortic dissection are usually transferred to the intensive care unit (ICU) after surgery. Prolonged ICU length of stay (ICU-LOS) is associated with higher level of care and higher mortality. We aimed to develop and validate machine learning models for predicting ICU-LOS after acute type A aortic dissection surgery.

Methods: A total of 353 patients with acute type A aortic dissection transferred to ICU after surgery from September 2016 to August 2019 were included. The patients were randomly divided into the training dataset (70%) and the validation dataset (30%). Eighty-four preoperative and intraoperative factors were collected for each patient. ICU-LOS was divided into four intervals (<4, 4–7, 7–10, and >10 days) according to interquartile range. Kendall correlation coefficient was used to identify factors associated with ICU-LOS. Five classic classifiers, Naive Bayes, Linear Regression, Decision Tree, Random Forest, and Gradient Boosting Decision Tree, were developed to predict ICU-LOS. Area under the curve (AUC) was used to evaluate the models' performance.

Results: The mean age of patients was 51.0 ± 10.9 years and 307 (87.0%) were males. Twelve predictors were identified for ICU-LOS, namely, D-dimer, serum creatinine, lactate dehydrogenase, cardiopulmonary bypass time, fasting blood glucose, white blood cell count, surgical time, aortic cross-clamping time, with Marfan's syndrome, without Marfan's syndrome, without aortic aneurysm, and platelet count. Random Forest yielded the highest performance, with an AUC of 0.991 (95% confidence interval [CI]: 0.978–1.000) and 0.837 (95% CI: 0.766–0.908) in the training and validation datasets, respectively.

Conclusions: Machine learning has the potential to predict ICU-LOS for acute type A aortic dissection. This tool could improve the management of ICU resources and patient-throughput planning, and allow better communication with patients and their families.

Prognostic Value of GDF-15 in Predicting Prolonged Intensive Care Stay following Cardiac Surgery: A Pilot Study

INTRODUCTION

Predicting intensive care unit length of stay and outcome following cardiac surgery is currently based on clinical parameters. Novel biomarkers could be employed to improve the prediction models.

MATERIALS AND METHODS

We performed a qualitative cytokine screening array to identify highly expressed biomarkers in preoperative blood samples of cardiac surgery patients. After identification of one highly expressed biomarker, growth differentiation factor 15 (GDF-15), a quantitative ELISA was undertaken. Preoperative levels of GDF-15 were compared in regard to duration of intensive care stay, cardiopulmonary bypass time, and indicators of organ dysfunction.

RESULTS

Preoperatively, GDF-15 was highly expressed in addition to several less highly expressed other biomarkers. After qualitative analysis, we could show that preoperatively raised levels of GDF-15 were positively associated with prolonged ICU stay exceeding 48 h (median 713 versus 1041 pg/ml, p = 0.003). It was also associated with prolonged mechanical ventilation and rates of severe sepsis but not with dialysis rates or cardiopulmonary bypass time. In univariate regression, raised GDF-15 levels were predictive of a prolonged ICU stay (OR 1.01, 95% confidence interval 1-1.02, and p = 0.029). On ROC curves, GDF-15 was found to predict prolonged ICU stay (AUC = 0.86, 95% confidence interval 0.71-0.99, and p = 0.003).

CONCLUSION

GDF-15 showed potential as predictor of prolonged intensive care stay following cardiac surgery, which might be valuable for risk stratification models.

Risk factors associated with longer stays in cardiovascular surgical intensive care unit after CABG

BACKGROUND/PURPOSE

Monitoring ICU length of stay (LOS) after CABG and examining its risk factors can guide initiatives on the improvement of care. But few have evaluated this issue to include personal and clinical factors, and demands of ICU care. This study applied Donabedian model to identify risk factors for longer ICU stays after CABG. Lifestyle, clinical factors during and after CABG, TISS were viewed as structure factors, and infection and organ failures during ICU did as process factors.

METHODS

This retrospective cohort study used data via medical records at a medical center. A stratified randomized sample of 230 adults from a cohort of 690 isolated CABGs was to reflect the rate of 34.7% longer than 3-day-ICU LOS. The sample comprised of longer-stay group (n = 150) and shorter-stay group (n = 80).

RESULT

Hierarchical logistic regression analysis revealed that potential signs of infection (3-day average WBC higher than 10,000/μL, OR: 3.41 and the body temperature higher than 38 °C, OR:5.67) and acute renal failure (OR: 8.97) remained as the most significant predicted factors of stay longer than 3 ICU days. Along with higher TISS score within 24 hours (OR:1.06), structure factors of female gender (OR:4.16) smoking(OR: 4.87), higher CCI before surgery(OR:1.49), bypass during CABG (OR:3.51) had higher odds of risk to stay longer.

CONCLUSION

Further quality improvement initiatives to shorten ICU stay after CABG may include the promotion of a smoking cessation program in clinical practice, and better management of the manpower allocation, infection control and renal failure.

Multiple arterial conduits for multi-vessel coronary artery bypass grafting in patients with mild to moderate left ventricular systolic dysfunction: a multicenter retrospective study

Background

Advantages of multiple arterial conduits for coronary artery bypass grafting (CABG) have been reported previously. We aimed to evaluate the mid-term outcomes of multiple arterial CABG (MABG) among patients with mild to moderate left ventricular systolic dysfunction (LVSD).

Methods

This multicenter study using propensity score matching took place from January 2013 to June 2019 in Jiangsu Province and Shanghai, China, with a mean and maximum follow-up of 3.3 and 6.8 years, respectively. We included patients with mild to moderate LVSD, undergoing primary, isolated multi-vessel CABG with left internal thoracic artery. The in-hospital and mid-term outcomes of MABG versus conventional left internal thoracic artery supplemented by saphenous vein grafts (single arterial CABG) were compared. The primary end points were death from all causes and death from cardiovascular causes. The secondary end points were stroke, myocardial infarction, repeat revascularization, and a composite of all mentioned outcomes, including death from all causes (major adverse events). Sternal wound infection was included with 6 months of follow-up after surgery.

Results

243 and 676 patients were formed in MABG and single arterial CABG cohorts after matching in a 1:3 ratio. In-hospital death was not significantly different (MABG 1.6% versus single arterial CABG 2.2%, p = 0.78). After a mean (±SD) follow-up time of 3.3 ± 1.8 years, MABG was associated with lower rates of major adverse events (HR, 0.64; 95% CI, 0.44–0.94; p = 0.019), myocardial infarction (HR, 0.39; 95% CI, 0.16–0.99; p = 0.045) and repeat revascularization (HR, 0.42; 95% CI, 0.18–0.97; p = 0.034). There was no difference in the rates of death, stroke, and sternal wound infection.

Conclusions

MABG was associated with reduced mid-term rates of major adverse events and cardiovascular events and may be the procedure of choice for patients with mild to moderate LVSD requiring CABG.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13019-021-01463-5.

Externally validated model predicting gait independence after stroke showed fair performance and improved after updating

OBJECTIVE

To externally validate recent prognostic models that predict independent gait following stroke.

STUDY DESIGN AND SETTING

A systematic search identified recent models (<10 years) that predicted independent gait in adult stroke patients, using easily obtainable predictors. Predictors from the original models were assigned proxies when required, and model performance was evaluated in the validation cohort (n = 957). Models were updated to determine if performance could be improved.

RESULTS

Three prognostic models met our criteria, all with high Risk of Bias. Validation data was only available for the Australian model. This model used National Institute of Health Stroke Scale (NIHSS) and age to predict independent gait, using Motor Assessment Scale (MAS) walking item. For validation, Scandinavian Stroke Scale (SSS) was a proxy for NIHSS, and Functional Independence Measure (FIM) locomotion item was a proxy for MAS. The Area Under the Curve was 0.77 (0.74-0.80) and had good calibration in the validation dataset. Adjustment of the intercept and regression coefficients slightly improved discrimination. By adding paretic leg strength, the model further improved (AUC 0.82).

CONCLUSION

External validation of the Australian model with proxies showed fair discrimination and good calibration. Updating the model by adding paretic leg strength further improved model performance.

Prediction of Prolonged Intensive Care Unit Length of Stay Following Cardiac Surgery

Intensive care unit (ICU) costs comprise a significant proportion of the total inpatient charges for cardiac surgery. No reliable method for predicting intensive care unit length of stay following cardiac surgery exists, making appropriate staffing and resource allocation challenging. We sought to develop a predictive model to anticipate prolonged ICU length of stay (LOS). All patients undergoing coronary artery bypass grafting (CABG) and/or valve surgery with a Society of Thoracic Surgeons (STS) predicted risk score were evaluated from an institutional STS database. Models were developed using 2014-2017 data; validation used 2018-2019 data. Prolonged ICU LOS was defined as requiring ICU care for at least three days postoperatively. Predictive models were created using lasso regression and relative utility compared. A total of 3283 patients were included with 1669 (50.8%) undergoing isolated CABG. Overall, 32% of patients had prolonged ICU LOS. Patients with comorbid conditions including severe COPD (53% vs 29%, P < 0.001), recent pneumonia (46% vs 31%, P < 0.001), dialysis-dependent renal failure (57% vs 31%, P < 0.001) or reoperative status (41% vs 31%, P < 0.001) were more likely to experience prolonged ICU stays. A prediction model utilizing preoperative and intraoperative variables correctly predicted prolonged ICU stay 76% of the time. A preoperative variable-only model exhibited 74% prediction accuracy. Excellent prediction of prolonged ICU stay can be achieved using STS data. Moreover, there is limited loss of predictive ability when restricting models to preoperative variables. This novel model can be applied to aid patient counseling, resource allocation, and staff utilization.

Predictors of Length of Stay in Intensive Care Unit after Coronary Artery Bypass Grafting: Development a Risk Scoring System

Introduction

To determine predictors of length of stay (LOS) in the intensive care unit (ICU) after coronary artery bypass grafting (CABG) and to develop a risk scoring system were the objectives of this study.

Methods

In this retrospective study, 1202 patients' medical records after CABG were evaluated by a research-made checklist. Tarone-Ware test was used to determine the predictors of patients' LOS in the ICU. Cox regression model was used to determine the risk factors and risk ratios associated with ICU LOS.

Results

The mean ICU LOS after CABG was 55.27±17.33 hours. Cox regression model showed that having more than two chest tubes (95% confidence interval [CI] 1.005-1.287, Relative Risk [RR]=1.138), occurrence of atelectasis (95% CI 1.000-3.007, RR=1.734), and occurrence of atrial fibrillation after CABG (95% CI 1.428-2.424, RR=1.861) were risk factors associated with longer ICU LOS. The discrimination power of this set of predictors was demonstrated with an area under the receiver operating characteristic curve and it was 0.69. A simple risk scoring system was developed based on three identified predictors that can raise ICU LOS.

Conclusion

The simple risk scoring system developed based on three identified predictors can help to plan more accurately a patient's LOS in hospital for CABG and can be useful in managing human and financial resources.

Supplementing prediction by EuroSCORE with social and patient-reported measures among patients undergoing cardiac surgery

OBJECTIVES

The risk of poor outcomes is traditionally attributed to biological and physiological processes in cardiac surgery. However, evidence exists that other factors, such as emotional, behavioral, social, and functional, are predictive of poor outcomes. Objectives were to evaluate the predictive value of several emotional, social, functional, and behavioral factors on four outcomes: death within 90 days, prolonged stay in intensive care, prolonged hospital admission, and readmission within 90 days following cardiac surgery.

METHODS

This prospective study included adults undergoing cardiac surgery 2013-2014, including information on register-based socioeconomic factors and self-reported health in a nested subsample. Logistic regression analyses to determine the association and incremental value of each candidate predictor variable were conducted. Multiple regression analyses were used to determine the incremental value of each candidate predictor variable, as well as discrimination and calibration based on the area under the curve (AUC) and Brier score.

RESULTS

Of 3217 patients, 3% died, 9% had prolonged intensive care stay, 51% had prolonged hospital admission, and 39% were readmitted to hospital. Patients living alone (odds ratio, 1.19; 95% confidence interval, 1.02-1.38), with lower educational levels (1.27; 1.04-1.54) and low health-related quality of life (1.43; 1.02-2.01) had prolonged hospital admission. Analyses revealed living alone as predictive of prolonged intensive care unit (ICU) stay (Brier, 0.08; AUC, 0.68), death (0.03; 0.71), and prolonged hospital admission (0.24; 0.62).

CONCLUSION

Living alone was found to supplement EuroSCORE in predicting death, prolonged hospital admission, and prolonged ICU stay following cardiac surgery. Low educational level and impaired health-related quality of life were, furthermore, predictive of prolonged hospital admission.

Derivation and Validation of a Clinical Model to Predict Intensive Care Unit Length of Stay After Cardiac Surgery

Background Across the globe, elective surgeries have been postponed to limit infectious exposure and preserve hospital capacity for coronavirus disease 2019 (COVID-19). However, the ramp down in cardiac surgery volumes may result in unintended harm to patients who are at high risk of mortality if their conditions are left untreated. To help optimize triage decisions, we derived and ambispectively validated a clinical score to predict intensive care unit length of stay after cardiac surgery. Methods and Results Following ethics approval, we derived and performed multicenter valida tion of clinical models to predict the likelihood of short (≤2 days) and prolonged intensive care unit length of stay (≥7 days) in patients aged ≥18 years, who underwent coronary artery bypass grafting and/or aortic, mitral, and tricuspid value surgery in Ontario, Canada. Multivariable logistic regression with backward variable selection was used, along with clinical judgment, in the modeling process. For the model that predicted short intensive care unit stay, the c-statistic was 0.78 in the derivation cohort and 0.71 in the validation cohort. For the model that predicted prolonged stay, c-statistic was 0.85 in the derivation and 0.78 in the validation cohort. The models, together termed the CardiOttawa LOS Score, demonstrated a high degree of accuracy during prospective testing. Conclusions Clinical judgment alone has been shown to be inaccurate in predicting postoperative intensive care unit length of stay. The CardiOttawa LOS Score performed well in prospective validation and will complement the clinician's gestalt in making more efficient resource allocation during the COVID-19 period and beyond.

Deep Learning for Improved Risk Prediction in Surgical Outcomes

The Norwood surgical procedure restores functional systemic circulation in neonatal patients with single ventricle congenital heart defects, but this complex procedure carries a high mortality rate. In this study we address the need to provide an accurate patient specific risk prediction for one-year postoperative mortality or cardiac transplantation and prolonged length of hospital stay with the purpose of assisting clinicians and patients' families in the preoperative decision making process. Currently available risk prediction models either do not provide patient specific risk factors or only predict in-hospital mortality rates. We apply machine learning models to predict and calculate individual patient risk for mortality and prolonged length of stay using the Pediatric Heart Network Single Ventricle Reconstruction trial dataset. We applied a Markov Chain Monte-Carlo simulation method to impute missing data and then fed the selected variables to multiple machine learning models. The individual risk of mortality or cardiac transplantation calculation produced by our deep neural network model demonstrated 89 ± 4% accuracy and 0.95 ± 0.02 area under the receiver operating characteristic curve (AUROC). The C-statistics results for prediction of prolonged length of stay were 85 ± 3% accuracy and AUROC 0.94 ± 0.04. These predictive models and calculator may help to inform clinical and organizational decision making.

Derivation and Validation of a Nomogram to Predict In-Hospital Complications in Children with Tetralogy of Fallot Repaired at an Older Age

Background

We aimed to develop and validate a prediction model for in‐hospital complications in children with tetralogy of Fallot repaired at an older age.

Methods and Results

A total of 513 pediatric patients from the Tianjin data set formed a derivation cohort, and 158 pediatric patients from the Hefei and Xiamen data sets formed validation cohorts. We applied least absolute shrinkage and selection operator analysis for variable selection and logistic regression coefficients for risk scoring. We classified patients into different risk categorizations by threshold analysis and investigated the association with in‐hospital complications using logistic regression. In‐hospital complications were defined as death, need for extensive pharmacologic support (vasoactive‐inotrope score of ≥20), and need for mechanical circulatory support. We developed a nomogram based on risk classifier and independent baseline variables using a multivariable logistic model. Based on risk scores weighted by 11 preoperative and 4 intraoperative selected variables, we classified patients as low, intermediate, and high risk in the derivation cohort. With reference to the low‐risk group, the intermediate‐ and high‐risk groups conferred significantly higher in‐hospital complication risks (adjusted odds ratio: 2.721 [95% CI, 1.267–5.841], P=0.0102; 9.297 [95% CI, 4.601–18.786], P<0.0001). A nomogram integrating the ARIAR‐Risk classifier (absolute and relative low risk, intermediate risk, and aggressive and refractory high risk) with age and mean blood pressure showed good discrimination and goodness‐of‐fit for derivation (area under the receiver operating characteristic curve: 0.785 [95% CI, 0.731–0.839]; Hosmer‐Lemeshow test, P=0.544) and external validation (area under the receiver operating characteristic curve: 0.759 [95% CI, 0.636–0.881]; Hosmer‐Lemeshow test, P=0.508).

Conclusions

A risk‐classifier–oriented nomogram is a reliable prediction model for in‐hospital complications in children with tetralogy of Fallot repaired at an older age, and strengthens risk/benefit–based decision‐making.

Methodological standards for the development and evaluation of clinical prediction rules: a review of the literature

Clinical prediction rules (CPRs) that predict the absolute risk of a clinical condition or future outcome for individual patients are abundant in the medical literature; however, systematic reviews have demonstrated shortcomings in the methodological quality and reporting of prediction studies. To maximise the potential and clinical usefulness of CPRs, they must be rigorously developed and validated, and their impact on clinical practice and patient outcomes must be evaluated. This review aims to present a comprehensive overview of the stages involved in the development, validation and evaluation of CPRs, and to describe in detail the methodological standards required at each stage, illustrated with examples where appropriate. Important features of the study design, statistical analysis, modelling strategy, data collection, performance assessment, CPR presentation and reporting are discussed, in addition to other, often overlooked aspects such as the acceptability, cost-effectiveness and longer-term implementation of CPRs, and their comparison with clinical judgement. Although the development and evaluation of a robust, clinically useful CPR is anything but straightforward, adherence to the plethora of methodological standards, recommendations and frameworks at each stage will assist in the development of a rigorous CPR that has the potential to contribute usefully to clinical practice and decision-making and have a positive impact on patient care.

Updating EuroSCORE by including emotional, behavioural, social and functional factors to the risk assessment of patients undergoing cardiac surgery: a study protocol

INTRODUCTION

Conventional risk assessment in cardiac surgery focus on medical and physiological factors and have been developed to predict mortality. Other relevant risk factors associated with increased risk of poor outcomes are not included. Adding non-medical variables as potential prognostic factors to risk assessments direct attention away from specific diagnoses towards a more holistic view of the patients and their predicament. The aim of this paper is to describe the method and analysis plan for the development and validation of a prognostic screening tool as a supplement to the European System for Cardiac Operative Risk Evaluation (EuroSCORE) to predict mortality, readmissions and prolonged length of admission in patients within 90 days after cardiac surgery, as individual outcomes.

METHODS AND ANALYSIS

The development of a prognostic screening tool with inclusion of emotional, behavioural, social and functional factors complementing risk assessment by EuroSCORE will adopt the methods recommended by the PROGnosis RESearch Strategy Group and report using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis statement. In the development stage, we will use data derived from three datasets comprising 1143, 3347 and 982 patients for a prospective cohort study of patients undergoing cardiac surgery, respectively. We will construct logistic regression models to predict mortality, prolonged length of admission and 90-day readmissions. In the validation stage, we will use data from a separate sample of 333 patients planned to undergo cardiac surgery to assess the performance of the developed prognostic model. We will produce validation plots showing the overall performance, area under the curve statistic for discrimination and the calibration slope and intercept.

ETHICS AND DISSEMINATION

The study will follow the requirements from the Ethical Committee System ensuring voluntary participation in accordance with the Helsinki declarations. Data will be filed in accordance with the requirements of the Danish Data Protection Agency.

Impaired right ventricular ejection fraction after cardiac surgery is associated with a complicated ICU stay

Background

Right ventricular (RV) dysfunction is a known risk factor for increased mortality in cardiac surgery. However, the association between RV performance and ICU morbidity is largely unknown.

Methods

We performed a single-centre, retrospective study including cardiac surgery patients equipped with a pulmonary artery catheter, enabling continuous right ventricular ejection fraction (RVEF) measurements. Primary endpoint of our study was ICU morbidity (as determined by ICU length of stay, duration of mechanical ventilation, usage of inotropic drugs and fluids, and kidney dysfunction) in relation to RVEF. Patients were divided into three groups according to their RVEF; < 20%, 20-30%, and > 30%.

Results

We included 1109 patients. Patients with a RVEF < 20% had a significantly longer stay in ICU, a longer duration of mechanical ventilation, higher fluid balance, a higher incidence of inotropic drug usage, and more increase in postoperative creatinine levels in comparison to the other subgroups. In a multivariate analysis, RVEF was independently associated with increased ICU length of stay (OR 0.934 CI 0.908-0.961, p < 0.001), prolonged duration of mechanical ventilation (OR 0.969, CI 0.942-0.998, p = 0.033), usage of inotropic drugs (OR 0.944, CI 0.917-0.971, p < 0.001), and increase in creatinine (OR 0.962, CI 0.934-0.991, p = 0.011).

Conclusions

A decreased RVEF is independently associated with a complicated ICU stay.

Accurate Prediction of Congenital Heart Surgical Length of Stay Incorporating a Procedure-Based Categorical Variable

OBJECTIVES

There is increasing demand for the limited resource of Cardiac ICU care. In this setting, there is an expectation to optimize hospital resource use without restricting care delivery. We developed methodology to predict extended cardiac ICU length of stay following surgery for congenital heart disease.

DESIGN

Retrospective analysis by multivariable logistic regression of important predictive factors for outcome of postoperative ICU length of stay greater than 7 days.

SETTING

Cardiac ICU at Boston Children's Hospital, a large, pediatric cardiac surgical referral center.

PATIENTS

All patients undergoing congenital heart surgery at Boston Children's Hospital from January 1, 2010, to December 31, 2015.

INTERVENTIONS

No study interventions.

MEASUREMENTS AND MAIN RESULTS

The patient population was identified. Clinical variables and Congenital Heart Surgical Stay categories were recorded based on surgical intervention performed. A model was built to predict the outcome postoperative ICU length of stay greater than 7 days at the time of surgical intervention. The development cohort included 4,029 cases categorized into five Congenital Heart Surgical Stay categories with a C statistic of 0.78 for the outcome ICU length of stay greater than 7 days. Explanatory value increased with inclusion of patient preoperative status as determined by age, ventilator dependence, and admission status (C statistic = 0.84). A second model was optimized with inclusion of intraoperative factors available at the time of postoperative ICU admission, including cardiopulmonary bypass time and chest left open (C statistic 0.87). Each model was tested in a validation cohort (n = 1,008) with equivalent C statistics.

CONCLUSIONS

Using a model comprised of basic patient characteristics, we developed a robust prediction tool for patients who will remain in the ICU longer than 7 days after cardiac surgery, at the time of postoperative ICU admission. This model may assist in patient counseling, case scheduling, and capacity management. Further examination in external settings is needed to establish generalizability.

Participatory design of probability-based decision support tools for in-hospital nurses

Objective

To describe nurses' preferences for the design of a probability-based clinical decision support (PB-CDS) tool for in-hospital clinical deterioration.

Methods

A convenience sample of bedside nurses, charge nurses, and rapid response nurses (n = 20) from adult and pediatric hospitals completed participatory design sessions with researchers in a simulation laboratory to elicit preferred design considerations for a PB-CDS tool. Following theme-based content analysis, we shared findings with user interface designers and created a low-fidelity prototype.

Results

Three major themes and several considerations for design elements of a PB-CDS tool surfaced from end users. Themes focused on "painting a picture" of the patient condition over time, promoting empowerment, and aligning probability information with what a nurse already believes about the patient. The most notable design element consideration included visualizing a temporal trend of the predicted probability of the outcome along with user-selected overlapping depictions of vital signs, laboratory values, and outcome-related treatments and interventions. Participants expressed that the prototype adequately operationalized requests from the design sessions.

Conclusions

Participatory design served as a valuable method in taking the first step toward developing PB-CDS tools for nurses. This information about preferred design elements of tools that support, rather than interrupt, nurses' cognitive workflows can benefit future studies in this field as well as nurses' practice.

Intermediate Care Unit After Cardiac Surgery: Impact on Length of Stay and Outcomes

INTRODUCTION AND OBJECTIVES

Current postoperative management of adult cardiac surgery often comprises transfer from the intensive care unit (ICU) to a conventional ward. Intermediate care units (IMCU) permit hospital resource optimization. We analyzed the impact of an IMCU on length of stay (both ICU and in-hospital) and outcomes (in-hospital mortality and 30-day readmissions) after adult cardiac surgery (IMCU-CS).

METHODS

From November 2012 to April 2015, 1324 consecutive patients were admitted to a university hospital for cardiac surgery. In May 2014, an IMCU-CS was established for postoperative care. For the purposes of this study, patients were classified into 2 groups, depending on the admission period: pre-IMCU-CS (November 2012-April 2014, n=674) and post-IMCU-CS (May 2014-April 2015, n=650).

RESULTS

There were no statistically significant differences in age, sex, risk factors, comorbidities, EuroSCORE 2, left ventricular ejection fraction, or the types of surgery (valvular in 53%, coronary in 26%, valvular plus coronary in 11.5%, and aorta in 1.8%). The ICU length of stay decreased from 4.9±11 to 2.9±6 days (mean±standard deviation; P<.001); 2 [1-4] to 1 [0-3] (median [Q1-Q3]); in-hospital length of stay decreased from 13.5±15 to 12.7±11 days (mean±standard deviation; P=.01); 9 [7-13] to 9 [7-11] (median [Q1-Q3]), in pre-IMCU-CS to post-IMCU-CS, respectively. There were no statistically significant differences in in-hospital mortality (4.9% vs 3.5%; P=.28) or 30-day readmission rate (4.3% vs 4.2%; P=.89).

CONCLUSIONS

After the establishment of an IMCU-CS for postoperative cardiac surgery, there was a reduction in ICU and in-hospital mean lengths of stay with no increase in in-hospital mortality or 30-day readmissions.

Factors Associated with the Need for, and the Impact of, Extracorporeal Membrane Oxygenation in Children with Congenital Heart Disease during Admissions for Cardiac Surgery

Introduction: This study aimed to determine factors associated with the need for extracorporeal membrane oxygenation (ECMO) in children with congenital heart disease (CHD) during admission for cardiac surgery (CS). A secondary aim was to determine how ECMO impacted length, cost, and mortality of the admission. Methods: Data from the Kids’ Inpatient Database (KIDS) were utilized. Admissions with CHD under 18 years of age with cardiac surgery were included. Need for ECMO in these admissions was then identified. Univariate analysis was conducted to compare characteristics between admissions with and without ECMO. Regression analyses were conducted to determine what factors were independently associated with ECMO and whether ECMO independently impacted admission characteristics. Results: A total of 46,176 admissions with CHD and CS were included in the final analysis. Of these, 798 (1.7%) required ECMO. Median age of ECMO admissions was 0.5 years. The following were associated with ECMO: decreased age, heart failure, acute kidney injury, arrhythmia, double outlet right ventricle, atrioventricular septal defect, transposition, Ebstein anomaly, hypoplastic left heart syndrome, common arterial trunk, tetralogy of Fallot, coronary anomaly, valvuloplasty, repair of total anomalous pulmonary venous connection, arterial switch, RV to PA conduit placement, and heart transplant (p < 0.01). ECMO independently increased length of stay by 17.8 days, cost of stay by approximately $415,917, and inpatient mortality 22-fold. Conclusion: Only a small proportion of CHD patients undergoing CS require ECMO, although these patients require increased resource utilization and have high mortality. Specific cardiac lesions, cardiac surgeries, and comorbidities are associated with increased need for ECMO.

A Method to Account for Variation in Congenital Heart Surgery Length of Stay

OBJECTIVES

We sought to develop a risk-adjustment methodology for length of stay in congenital heart surgery, as none exist.

DESIGN

Prospective cohort analysis combined with previously obtained retrospective cohort analysis of a Department of Cardiovascular Surgery clinical database.

PATIENTS

Patients discharged from Boston Children's Hospital between October 1, 2006, and May 31, 2014, that underwent a congenital heart surgery procedure(s) linked to one of 103 surgical procedure types.

MEASUREMENTS AND MAIN RESULTS

Six thousand two hundred nine discharges during the reporting period at Boston Children's Hospital comprised the cohort. Seven Surgical Length Categories were developed to group surgical procedure types. A multivariable model for outcome length of stay was built using a derivation cohort consisting of a 75% random sample, starting with Surgical Length Categories and considering additional a priori factors. Postoperative factors were then added to improve predictive performance. The remaining 25% of the cohort was used to validate the multivariable models. The coefficient of determination (R) was used to estimate the variability in length of stay explained by each factor. The Surgical Length Categories yielded an R of 42%. Model performance increased when the a priori factors preoperative status, noncardiac abnormality, genetic anomaly, preoperative catheterization during episode of care, weight less than 3 kg, and preoperative vasoactive support medication were introduced to the model (R = 60.8%). Model performance further improved when postoperative ventilation greater than 7 days, operating room time, postoperative catheterization during episode of care, postoperative reintubation, number of postoperative vasoactive support medications, postoperative ICU infection, and greater than or equal to one secondary surgical procedure were added (R = 76.7%). The validation cohort yielded an R of 76.5%.

CONCLUSIONS

We developed a statistically valid procedure-based categorical variable and multivariable model for length of stay of congenital heart surgeries. The Surgical Length Categories and important a priori and postoperative factors may be used to pursue a predictive tool for length of stay to inform scheduling and bed management practices.

Benefits of ultra-fast-track anesthesia in left ventricular assist device implantation: a retrospective, propensity score matched cohort study of a four-year single center experience

BACKGROUND

The use of left ventricular assist devices (LVADs) has gained significant importance for treatment of end-stage heart failure. Fast-track procedures are well established in cardiac surgery, whereas knowledge of their benefits after LVAD implantation is sparse. We hypothesized that ultra-fast-track anesthesia (UFTA) with in-theater extubation or at a maximum of 4 h. after surgery is feasible in Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) level 3 and 4 patients and might prevent postoperative complications.

METHODS

From March, 2010 to March, 2012, 53 LVADs (50 Heart Mate II and 3 Heart Ware) were implanted in patients in our department. UFTA was successfully performed (LVAD _ultra ) in 13 patients. After propensity score matching, we compared the LVAD _ultra group with a matched group (LVAD _match ) receiving conventional anesthesia management.

RESULTS

Patients in the LVAD _ultra group had significantly lower incidences of pneumonia (p = 0.031), delirium (p = 0.031) and right ventricular failure (RVF) (p = 0.031). They showed a significantly higher cardiac index in the first 12 h. (p = 0.017); a significantly lower central venous pressure during the first 24 h. postoperatively (p = 0.005) and a significantly shorter intensive care unit (ICU) stay (p = 0.016). Kaplan-Meier analysis after four years of follow-up showed no significant difference in survival.

CONCLUSION

In this pilot study, we demonstrated the feasibility of ultra-fast-track anesthesia in LVAD implantation in selected patients with INTERMACS level 3-4. Patients had a lower incidence of postoperative complications, better hemodynamic performance, shorter length of ICU stay and lower incidence of RVF after UFTA. Prospective randomized investigations should examine the preservation of right ventricular function in larger numbers and identify appropriate selection criteria.

Prediction Models and Their External Validation Studies for Mortality of Patients with Acute Kidney Injury: A Systematic Review

OBJECTIVES

To systematically review AKI outcome prediction models and their external validation studies, to describe the discrepancy of reported accuracy between the results of internal and external validations, and to identify variables frequently included in the prediction models.

METHODS

We searched the MEDLINE and Web of Science electronic databases (until January 2016). Studies were eligible if they derived a model to predict mortality of AKI patients or externally validated at least one of the prediction models, and presented area under the receiver-operator characteristic curves (AUROC) to assess model discrimination. Studies were excluded if they described only results of logistic regression without reporting a scoring system, or if a prediction model was generated from a specific cohort.

RESULTS

A total of 2204 potentially relevant articles were found and screened, of which 12 articles reporting original prediction models for hospital mortality in AKI patients and nine articles assessing external validation were selected. Among the 21 studies for AKI prediction models and their external validation, 12 were single-center (57%), and only three included more than 1,000 patients (14%). The definition of AKI was not uniform and none used recently published consensus criteria for AKI. Although good performance was reported in their internal validation, most of the prediction models had poor discrimination with an AUROC below 0.7 in the external validation studies. There were 10 common non-renal variables that were reported in more than three prediction models: mechanical ventilation, age, gender, hypotension, liver failure, oliguria, sepsis/septic shock, low albumin, consciousness and low platelet count.

CONCLUSIONS

Information in this systematic review should be useful for future prediction model derivation by providing potential candidate predictors, and for future external validation by listing up the published prediction models.

Factors Associated with a Short (<2 Days) or Long (>10 Days) Length of Stay after Colectomy: A Multivariate Analysis of over 400 Patients

A prospectively maintained database of 415 patients undergoing colectomy was evaluated. We performed a logistic regression analysis to identify factors associated with 1) length of stay (LOS) of 2 days or less and 2) LOS of 10 days or more. Investigated variables included demographics, American Society of Anesthesiology (ASA) score, diagnosis, operative procedure, approach and time, transfusion requirements, and occurrence of any complications. Factors associated with a LOS of two days or less included ASA [odds ratio (OR): 0.34, 95% confidence interval (CI): 0.208–0.576], use of transversus abdominis plane block (OR: 5.259, 95% CI: 2.825–9.791), and operative time (OR: 0.98, 95% CI: 0.974–0.986). Age >65 had an OR of 1.73, though this did not reach statistical significance. Factors associated with LOS >10 days included ASA (OR: 2.152, 95% CI: 1.245–3.721), anastomotic leak (OR: 2.163, 95% CI: 1.486–3.148), ileus (OR: 8.790, 95% CI: 4.501–17.165), and surgical site infection (OR: 5.846, 95% CI: 2.764–12.362). Cancer and transfusion status were associated but did not reach statistical significance. Although operative time was longer in left-sided resections, no differences in LOS were observed. In conclusion, numerous factors are associated with short or long LOS and may help stratify resource utilization after colectomy. Further study is needed to confirm our findings.

Prognosis of Prolonged Intensive Care Unit Stay After Aortic Valve Replacement for Severe Aortic Stenosis in Octogenarians

OBJECTIVES

Octogenarians considered for cardiac surgery encounter more complications than other patients. Postoperative complications raise the question of continuation of high-cost care for patients with limited life expectancy. Duration of hospitalization in intensive care after cardiac surgery may differ between octogenarians and other patients. The objectives were evaluating the mortality rate of octogenarians experiencing prolonged hospitalization in intensive care and defining the best cut-off for prolonged intensive care unit length of stay.

DESIGN

A single-center observational study.

SETTING

A postoperative surgical intensive care unit in a tertiary teaching hospital in Paris, France.

PARTICIPANTS

All consecutive patients older than 80 years considered for aortic valve replacement for aortic stenosis were included.

MEASUREMENTS AND MAIN RESULTS

Mortality rate was determined among patients experiencing prolonged stay in intensive care with organ failure and without organ failure. An ROC curve determined the optimal cut-off defining prolonged hospitalization in intensive care according to the occurrence of postoperative complications. Multivariate analysis determined risk factors for early death or prolonged intensive care stay. The optimal cut-off defining prolonged intensive care unit length of stay was 4 days. Low ventricular ejection fraction (odds ratio [OR] = 0.95; 95% confidence interval [CI] 0.96-0.83; p = 0.0016), coronary disease (OR = 2.34; 95% CI 1.19-4.85; p = 0.014), and need for catecholamine (OR = 2.79; 95% CI 1.33-5.88; p = 0.0068) were associated with eventful postoperative course. There was not a hospitalization duration beyond which the prognosis significantly worsened.

CONCLUSIONS

Prolonged length of stay in ICU without organ failure is not associated with increased mortality. No specific duration of hospitalization in intensive care was associated with increased mortality. Continuation of care should be discussed on an individual basis.

Prediction models for cardiovascular disease risk in the general population: systematic review

OBJECTIVE

 To provide an overview of prediction models for risk of cardiovascular disease (CVD) in the general population.

DESIGN

 Systematic review.

DATA SOURCES

 Medline and Embase until June 2013.

ELIGIBILITY CRITERIA FOR STUDY SELECTION

 Studies describing the development or external validation of a multivariable model for predicting CVD risk in the general population.

RESULTS

 9965 references were screened, of which 212 articles were included in the review, describing the development of 363 prediction models and 473 external validations. Most models were developed in Europe (n=167, 46%), predicted risk of fatal or non-fatal coronary heart disease (n=118, 33%) over a 10 year period (n=209, 58%). The most common predictors were smoking (n=325, 90%) and age (n=321, 88%), and most models were sex specific (n=250, 69%). Substantial heterogeneity in predictor and outcome definitions was observed between models, and important clinical and methodological information were often missing. The prediction horizon was not specified for 49 models (13%), and for 92 (25%) crucial information was missing to enable the model to be used for individual risk prediction. Only 132 developed models (36%) were externally validated and only 70 (19%) by independent investigators. Model performance was heterogeneous and measures such as discrimination and calibration were reported for only 65% and 58% of the external validations, respectively.

CONCLUSIONS

 There is an excess of models predicting incident CVD in the general population. The usefulness of most of the models remains unclear owing to methodological shortcomings, incomplete presentation, and lack of external validation and model impact studies. Rather than developing yet another similar CVD risk prediction model, in this era of large datasets, future research should focus on externally validating and comparing head-to-head promising CVD risk models that already exist, on tailoring or even combining these models to local settings, and investigating whether these models can be extended by addition of new predictors.