Systematic review of factors influencing length of stay in ICU after adult cardiac surgery

Predicting hospital length of stay using machine learning on a large open health dataset

BACKGROUND

Governments worldwide are facing growing pressure to increase transparency, as citizens demand greater insight into decision-making processes and public spending. An example is the release of open healthcare data to researchers, as healthcare is one of the top economic sectors. Significant information systems development and computational experimentation are required to extract meaning and value from these datasets. We use a large open health dataset provided by the New York State Statewide Planning and Research Cooperative System (SPARCS) containing 2.3 million de-identified patient records. One of the fields in these records is a patient's length of stay (LoS) in a hospital, which is crucial in estimating healthcare costs and planning hospital capacity for future needs. Hence it would be very beneficial for hospitals to be able to predict the LoS early. The area of machine learning offers a potential solution, which is the focus of the current paper.

METHODS

We investigated multiple machine learning techniques including feature engineering, regression, and classification trees to predict the length of stay (LoS) of all the hospital procedures currently available in the dataset. Whereas many researchers focus on LoS prediction for a specific disease, a unique feature of our model is its ability to simultaneously handle 285 diagnosis codes from the Clinical Classification System (CCS). We focused on the interpretability and explainability of input features and the resulting models. We developed separate models for newborns and non-newborns.

RESULTS

The study yields promising results, demonstrating the effectiveness of machine learning in predicting LoS. The best R2 scores achieved are noteworthy: 0.82 for newborns using linear regression and 0.43 for non-newborns using catboost regression. Focusing on cardiovascular disease refines the predictive capability, achieving an improved R2 score of 0.62. The models not only demonstrate high performance but also provide understandable insights. For instance, birth-weight is employed for predicting LoS in newborns, while diagnostic-related group classification proves valuable for non-newborns.

CONCLUSION

Our study showcases the practical utility of machine learning models in predicting LoS during patient admittance. The emphasis on interpretability ensures that the models can be easily comprehended and replicated by other researchers. Healthcare stakeholders, including providers, administrators, and patients, stand to benefit significantly. The findings offer valuable insights for cost estimation and capacity planning, contributing to the overall enhancement of healthcare management and delivery.

Possible role of QRS duration in the right ventricle as a perioperative monitoring parameter for right ventricular function: a prospective cohort analysis in robotic mitral valve surgery

Background

The clinical importance of the right ventricle (RV) has recently been recognized; however, assessing its function during cardiac surgery remains challenging owing to its complex anatomy. A temporary transvenous pacing catheter is a useful tool in the small surgical field of minimally invasive cardiac surgery, and an electrocardiogram recorded through the catheter is composed of the direct electrophysiological activity of the RV. Therefore, we hypothesized that QRS duration in the RV (QRSRV) could be a useful monitoring parameter for perioperative RV function.

Methods

We conducted a prospective cohort analysis involving adult patients undergoing robotic mitral valve repair. A bipolar pacing catheter was inserted using x-ray fluoroscopy, and the QRSRV duration was assessed at four time points: preoperative baseline, during one-lung ventilation, after weaning from cardiopulmonary bypass, and before the end of surgery. At the same time points, right ventricular fractional area change (RVFAC) measured by transesophageal echocardiography and QRS duration at V5 lead of the body surface electrocardiogram (QRSV5) were also evaluated.

Results

In the 94 patients analyzed, QRSRV duration was significantly prolonged during robotic mitral valve repair (p = 0.0009), whereas no significant intraoperative changes in RVFAC were observed (p = 0.2). By contrast, QRSV5 duration was significantly shortened during surgery (p < 0.00001). Multilinear regression showed a significant correlation of QRSRV duration with RVFAC (p = 0.00006), but not with central venous pressure (p = 0.9), or left ventricular ejection fraction (p = 0.3). When patients were divided into two groups by postoperative QRSRV > 100 or ≤100 ms, 25 patients (26.6%) exhibited the prolonged QRSRV duration, and the mean increase in the postoperative QRSRV from preoperative baseline was 12 ms (p = 0.001), which was only 0.6 ms in patients with QRSRV ≤ 100 ms (p = 0.6). Cox regression analysis showed that prolonged postoperative QRSRV duration was the only significant parameter associated with a longer ICU stay after surgery (p = 0.02; hazard ratio, 0.55).

Conclusion

Our data suggest that QRSRV duration is a useful parameter for monitoring the RV during cardiac surgery, possibly better than a commonly used echocardiographic parameter, RVFAC. An electrophysiological assessment by QRSRV duration could be a practical tool for the complex anatomy of the RV, especially with limited modalities in perioperative settings.

First initiation of mobilization out of bed after cardiac surgery - an observational cross-sectional study in Sweden

BACKGROUND

Cardiac surgery is associated with a period of postoperative bed rest. Although early mobilization is a vital component of postoperative care, for preventing complications and enhancing physical recovery, there is limited data on routine practices and optimal strategies for early mobilization after cardiac surgery. The aim of the study was to define the timing for the first initiation of out of bed mobilization after cardiac surgery and to describe the type of mobilization performed.

METHODS

In this observational study, the first mobilization out of bed was studied in a subset of adult cardiac surgery patients (n = 290) from five of the eight university hospitals performing cardiothoracic surgery in Sweden. Over a five-week period, patients were evaluated for mobilization routines within the initial 24 h after cardiac surgery. Data on the timing of the first mobilization after the end of surgery, as well as the duration and type of mobilization, were documented. Additionally, information on patient characteristics, anesthesia, and surgery was collected.

RESULTS

A total of 277 patients (96%) were mobilized out of bed within the first 24 h, and 39% of these patients were mobilized within 6 h after surgery. The time to first mobilization after the end of surgery was 8.7 ± 5.5 h; median of 7.1 [4.5-13.1] hours, with no significant differences between coronary artery bypass grafting, valve surgery, aortic surgery or other procedures (p = 0.156). First mobilization session lasted 20 ± 41 min with median of 10 [1-11]. Various kinds of first-time mobilization, including sitting on the edge of the bed, standing, and sitting in a chair, were revealed. A moderate association was found between longer intubation time and later first mobilization (ρ = 0.487, p < 0.001). Additionally, there was a moderate correlation between the first timing of mobilization duration of the first mobilization session (ρ = 0.315, p < 0.001).

CONCLUSIONS

This study demonstrates a median time to first mobilization out of bed of 7 h after cardiac surgery. A moderate correlation was observed between earlier timing of mobilization and shorter duration of the mobilization session. Future research should explore reasons for delayed mobilization and investigate whether earlier mobilization correlates with clinical benefits.

TRIAL REGISTRATION

FoU in VGR (Id 275,357) and Clinical Trials (NCT04729634).

The Adherence to an Intraoperative Blood Product Transfusion Algorithm Is Associated With Reduced Blood Product Transfusions in Cardiac Surgical Patients Undergoing Coronary Artery Bypass Grafts and Aortic and/or Valve Replacement Surgery: A Single-Center, Observational Study

OBJECTIVE

To demonstrate the value of a viscoelastic-based intraoperative transfusion algorithm to reduce non-RBC product administration in adult cardiac surgical patients.

DESIGN

A prospective observational study.

SETTING

At a quaternary academic teaching hospital.

PARTICIPANTS

Cardiac surgical patients.

INTERVENTIONS

Viscoelastic-based intraoperative transfusion algorithm.

MEASUREMENTS AND MAIN RESULTS

The study authors compared intraoperative blood product transfusion rates in 184 cardiac surgical patients to 236 historic controls after implementing a viscoelastic-based algorithm. The authors found a non-significant reduction in transfusion of 23.8% for fresh frozen plasma (FFP) units (0.84 ± 1.4 v 0.64 ± 1.38; p = ns), 33.4% for platelet units (0.90 ± 1.39 v 0.60 ± 131; p = ns), and 15.8% for cryoprecipitate units (0.19 ± 0.54 v 0.16 ± 0.50; p = ns). They found a 43.9% reduction in red blood cell (RBC) units transfused (1.98 ± 2.24 v 0.55 ± 1.36; p = 0.008). There were no statistically significant differences in time to extubation (8.0 hours (4.0-21.0) v 8.0 (4.0-22.3), reoperation for bleeding (15 [12.3%] v 10 [10.6%]), intensive care unit length of stay (ICU LOS) (51.0 hours [28.0-100.5] v 53.5 [33.3-99.0]) or hospital LOS (9.0 days [6.0-15.0] v 10.0 [7.0-17.0]). Deviation from algorithm adherence was 32.7% (48/147). Packed RBC, FFP, platelets, cryoprecipitate, and cell saver were significantly reduced in the Algorithm Compliant Cohort compared with historic controls, whereas times to extubation, ICU LOS, and hospital LOS did not reach significance.

CONCLUSIONS

After the implementation of a viscoelastic-based algorithm, patients received fewer packed RBC, FFP, platelets, cryoprecipitate, and cell saver. Algorithm-compliant patients received fewer transfusions; however, reductions in times to extubation, ICU LOS, and hospital LOS were not statistically significant compared with historic controls.

A closed surgical intensive care unit organization improves cardiac surgical patient outcomes

Background

Intensive care unit (ICU) organization is a critical factor in optimizing patient outcomes. ICU organization can be divided into "OPEN" (O) and "CLOSED" (C) models, where the specialist or intensivist, respectively, assumes the role of primary physician. Recent studies support improved outcomes in closed ICUs, however, most of the available data is centered on ICUs generally or on subspecialty surgical patients in the setting of a subspecialized surgical intensive care unit (SICU). We examined the impact of closing a general SICU on patient outcomes following cardiac and ascending aortic surgery.

Methods

A retrospective cohort of patients following cardiac or ascending aortic surgery by median sternotomy was examined at a single academic medical center one year prior and one year after implementation of a closed SICU model. Patients were divided into "OPEN" (O; n=53) and "CLOSED" (C; n=73) cohorts.

Results

Cohorts were comparable in terms of age, race, and number of comorbid conditions. A significant difference in male gender (O: 60.4% vs. C: 76.7%, P=0.049), multiple procedure performed (O: 13.21% vs. C: 35.62%, P=0.019), and hospital readmission rates was detected (O: 39.6% vs. C: 9.6%, P=0.0003). Using a linear regression model, a closed model SICU organization decreased SICU length of stay (LOS). Using a multivariate logistic regression, being treated in a closed ICU decreased a patient's likelihood of having an ICU LOS greater than 48 hours.

Conclusions

Our study identified a decreased ICU LOS and hospital readmission in cardiac and ascending aortic patients in a closed general SICU despite increased procedure complexity. Further study is needed to clarify the effects on surgical complications and hospital charges.

Factors affecting duration of stay in the intensive care unit after coronary artery bypass surgery and its impact on in-hospital mortality: a retrospective study

BACKGROUND

Different risk factors affect the intensive care unit (ICU) stay after cardiac surgery. This study aimed to evaluate these risk factors.

PATIENTS AND METHODS

A retrospective analysis was conducted on clinical, operative, and outcome data from 1070 patients (mean age: 59 ± 9.8 years) who underwent isolated coronary bypass grafting CABG surgery with cardiopulmonary bypass. The outcome variable was prolonged length of stay LOS in the CICU stay (> 3 nights after CABG).

RESULTS

Univariate predictors of prolonged ICU stays included a left atrial diameter of > 4 cm (P < 0.001),chronic obstructive airway disease COPD (P = 0.005), hypertension (P = 0.006), diabetes mellitus (P = 0.009), having coronary stents (P = 0.006), B-blockers use before surgery (either because the surgery was done on urgent or emergency basis or the patients have contraindication to B-blockers use) (P = 0.005), receiving blood transfusion during surgery (P = 0.001), post-operative acute kidney injury (AKI) (P < 0.001), prolonged inotropic support of > 12 h (P < 0.001), and ventilation support of > 12 h (P < 0.001), post-operative sepsis or pneumonia (P < 0.001), post-operative stroke/TIA (P = 0.001), sternal wound infection (P = 0.002), and postoperative atrial fibrillation POAF (P < 0.001). Multivariate regression revealed that patients with anleft atrial LA diameter of > 4 cm (AOR 2.531, P = 0.003), patients who did not take B-blockers before surgery (AOR 1.1 P = 0.011), patients on ventilation support > 12 h (AOR 3.931, P =  < 0.001), patients who developed pneumonia (AOR 20.363, P =  < 0.001), and patients who developed post-operative atrial fibrillation (AOR 30.683, P =  < 0.001) were more likely to stay in the ICU for > 3 nights after CABG.

CONCLUSION

Our results showed that LA diameter > 4 cm, patients who did not take beta-blockers before surgery, on ventilation support > 12 h, developed pneumonia post-operatively, and developed POAF were more likely to have stays lasting > 3 nights. Efforts should be directed toward reducing these postoperative complications to shorten the duration of CICU stay, thereby reducing costs and improving bed availability.

Safety and effectiveness of neuromuscular electrical stimulation in cardiac surgery: A systematic review

BACKGROUND

Lack of mobilization and prolonged stay in the intensive care unit (ICU) are major factors resulting in the development of ICU-acquired muscle weakness (ICUAW). ICUAW is a type of skeletal muscle dysfunction and a common complication of patients after cardiac surgery, and may be a risk factor for prolonged duration of mechanical ventilation, associated with a higher risk of readmission and higher mortality. Early mobilization in the ICU after cardiac surgery has been found to be low with a significant trend to increase over ICU stay and is also associated with a reduced duration of mechanical ventilation and ICU length of stay. Neuromuscular electrical stimulation (NMES) is an alternative modality of exercise in patients with muscle weakness. A major advantage of NMES is that it can be applied even in sedated patients in the ICU, a fact that might enhance early mobilization in these patients.

AIM

To evaluate safety, feasibility and effectiveness of NMES on functional capacity and muscle strength in patients before and after cardiac surgery.

METHODS

We performed a search on Pubmed, Physiotherapy Evidence Database (PEDro), Embase and CINAHL databases, selecting papers published between December 2012 and April 2023 and identified published randomized controlled trials (RCTs) that included implementation of NMES in patients before after cardiac surgery. RCTs were assessed for methodological rigor and risk of bias via the PEDro. The primary outcomes were safety and functional capacity and the secondary outcomes were muscle strength and function.

RESULTS

Ten studies were included in our systematic review, resulting in 703 participants. Almost half of them performed NMES and the other half were included in the control group, treated with usual care. Nine studies investigated patients after cardiac surgery and 1 study before cardiac surgery. Functional capacity was assessed in 8 studies via 6MWT or other indices, and improved only in 1 study before and in 1 after cardiac surgery. Nine studies explored the effects of NMES on muscle strength and function and, most of them, found increase of muscle strength and improvement in muscle function after NMES. NMES was safe in all studies without any significant complication.

CONCLUSION

NMES is safe, feasible and has beneficial effects on muscle strength and function in patients after cardiac surgery, but has no significant effect on functional capacity.

Supervised machine learning algorithms to predict the duration and risk of long-term hospitalization in HIV-infected individuals: a retrospective study

Objective

The study aimed to use supervised machine learning models to predict the length and risk of prolonged hospitalization in PLWHs to help physicians timely clinical intervention and avoid waste of health resources.

Methods

Regression models were established based on RF, KNN, SVM, and XGB to predict the length of hospital stay using RMSE, MAE, MAPE, and R2, while classification models were established based on RF, KNN, SVM, NN, and XGB to predict risk of prolonged hospital stay using accuracy, PPV, NPV, specificity, sensitivity, and kappa, and visualization evaluation based on AUROC, AUPRC, calibration curves and decision curves of all models were used for internally validation.

Results

In regression models, XGB model performed best in the internal validation (RMSE = 16.81, MAE = 10.39, MAPE = 0.98, R2 = 0.47) to predict the length of hospital stay, while in classification models, NN model presented good fitting and stable features and performed best in testing sets, with excellent accuracy (0.7623), PPV (0.7853), NPV (0.7092), sensitivity (0.8754), specificity (0.5882), and kappa (0.4672), and further visualization evaluation indicated that the largest AUROC (0.9779), AUPRC (0.773) and well-performed calibration curve and decision curve in the internal validation.

Conclusion

This study showed that XGB model was effective in predicting the length of hospital stay, while NN model was effective in predicting the risk of prolonged hospitalization in PLWH. Based on predictive models, an intelligent medical prediction system may be developed to effectively predict the length of stay and risk of HIV patients according to their medical records, which helped reduce the waste of healthcare resources.

Diversifying cardiac intensive care unit models: Successful example of an operating surgeon-led unit

Objective

The intensivist-led cardiovascular intensive care unit model is the standard of care in cardiac surgery. This study examines whether a cardiovascular intensive care unit model that uses operating cardiac surgeons, cardiothoracic surgery residents, and advanced practice providers is associated with comparable outcomes.

Methods

This is a single-institution review of the first 400 cardiac surgery patients admitted to an operating surgeon-led cardiovascular intensive care unit from 2020 to 2022. Inclusion criteria are elective status and operations managed by both cardiovascular intensive care unit models (aortic operations, valve operations, coronary operations, septal myectomy). Patients from the surgeon-led cardiovascular intensive care unit were exact matched by operation type and 1:1 propensity score matched with controls from the traditional cardiovascular intensive care unit using a logistic regression model that included age, sex, preoperative mortality risk, incision type, and use of cardiopulmonary bypass and circulatory arrest. Primary outcome was total postoperative length of stay. Secondary outcomes included postoperative intensive care unit length of stay, 30-day mortality, 30-day Society of Thoracic Surgeons-defined morbidity (permanent stroke, renal failure, cardiac reoperation, prolonged intubation, deep sternal infection), packed red cell transfusions, and vasopressor use. Outcomes between the 2 groups were compared using chi-square, Fisher exact test, or 2-sample t test as appropriate.

Results

A total of 400 patients from the surgeon-led cardiovascular intensive care unit (mean age 61.2 ± 12.8 years, 131 female patients [33%], 346 patients [86.5%] with European System for Cardiac Operative Risk Evaluation II <2%) and their matched controls were included. The most common operations across both units were coronary artery bypass grafting (n = 318, 39.8%) and mitral valve repair or replacement (n = 238, 29.8%). Approximately half of the operations were performed via sternotomy (n = 462, 57.8%). There were 3 (0.2%) in-hospital deaths, and 47 patients (5.9%) had a 30-day complication. The total length of stay was significantly shorter for the surgeon-led cardiovascular intensive care unit patients (6.3 vs 7.0 days, P = .028), and intensive care unit length of stay trended in the same direction (2.5 vs 2.9 days, P = .16). Intensive care unit readmission rates, 30-day mortality, and 30-day morbidity were not significantly different between cardiovascular intensive care unit models. The surgeon-led cardiovascular intensive care unit was associated with fewer postoperative red blood cell transfusions in the cardiovascular intensive care unit (P = .002) and decreased vasopressor use (P = .001).

Conclusions

In its first 2 years, the surgeon-led cardiovascular intensive care unit demonstrated comparable outcomes to the traditional cardiovascular intensive care unit with significant improvements in total length of stay, postoperative transfusions in the cardiovascular intensive care unit, and vasopressor use. This early success exemplifies how an operating surgeon-led cardiovascular intensive care unit can provide similar outcomes to the standard-of-care model for patients undergoing elective cardiac surgery.

Machine learning-based prediction of hospital prolonged length of stay admission at emergency department: a Gradient Boosting algorithm analysis

Objective

This study aims to develop and compare different models to predict the Length of Stay (LoS) and the Prolonged Length of Stay (PLoS) of inpatients admitted through the emergency department (ED) in general patient settings. This aim is not only to promote any specific model but rather to suggest a decision-supporting tool (i.e., a prediction framework).

Methods

We analyzed a dataset of patients admitted through the ED to the "Sant"Orsola Malpighi University Hospital of Bologna, Italy, between January 1 and October 26, 2022. PLoS was defined as any hospitalization with LoS longer than 6 days. We deployed six classification algorithms for predicting PLoS: Random Forest (RF), Support Vector Machines (SVM), Gradient Boosting (GB), AdaBoost, K-Nearest Neighbors (KNN), and logistic regression (LoR). We evaluated the performance of these models with the Brier score, the area under the ROC curve (AUC), accuracy, sensitivity (recall), specificity, precision, and F1-score. We further developed eight regression models for LoS prediction: Linear Regression (LR), including the penalized linear models Least Absolute Shrinkage and Selection Operator (LASSO), Ridge and Elastic-net regression, Support vector regression, RF regression, KNN, and eXtreme Gradient Boosting (XGBoost) regression. The model performances were measured by their mean square error, mean absolute error, and mean relative error. The dataset was randomly split into a training set (70%) and a validation set (30%).

Results

A total of 12,858 eligible patients were included in our study, of whom 60.88% had a PloS. The GB classifier best predicted PloS (accuracy 75%, AUC 75.4%, Brier score 0.181), followed by LoR classifier (accuracy 75%, AUC 75.2%, Brier score 0.182). These models also showed to be adequately calibrated. Ridge and XGBoost regressions best predicted LoS, with the smallest total prediction error. The overall prediction error is between 6 and 7 days, meaning there is a 6-7 day mean difference between actual and predicted LoS.

Conclusion

Our results demonstrate the potential of machine learning-based methods to predict LoS and provide valuable insights into the risks behind prolonged hospitalizations. In addition to physicians' clinical expertise, the results of these models can be utilized as input to make informed decisions, such as predicting hospitalizations and enhancing the overall performance of a public healthcare system.

Three Logistic Predictive Models for the Prediction of Mortality and Major Pulmonary Complications after Cardiac Surgery

Background and Objectives: Pulmonary complications are a leading cause of morbidity after cardiac surgery. The aim of this study was to develop models to predict postoperative lung dysfunction and mortality. Materials and Methods: This was a single-center, observational, retrospective study. We retrospectively analyzed the data of 11,285 adult patients who underwent all types of cardiac surgery from 2003 to 2015. We developed logistic predictive models for in-hospital mortality, postoperative pulmonary complications occurring in the intensive care unit, and postoperative non-invasive mechanical ventilation when clinically indicated. Results: In the "preoperative model" predictors for mortality were advanced age (p < 0.001), New York Heart Association (NYHA) class (p < 0.001) and emergent surgery (p = 0.036); predictors for non-invasive mechanical ventilation were advanced age (p < 0.001), low ejection fraction (p = 0.023), higher body mass index (p < 0.001) and preoperative renal failure (p = 0.043); predictors for postoperative pulmonary complications were preoperative chronic obstructive pulmonary disease (p = 0.007), preoperative kidney injury (p < 0.001) and NYHA class (p = 0.033). In the "surgery model" predictors for mortality were intraoperative inotropes (p = 0.003) and intraoperative intra-aortic balloon pump (p < 0.001), which also predicted the incidence of postoperative pulmonary complications. There were no specific variables in the surgery model predicting the use of non-invasive mechanical ventilation. In the "intensive care unit model", predictors for mortality were postoperative kidney injury (p < 0.001), tracheostomy (p < 0.001), inotropes (p = 0.029) and PaO2/FiO2 ratio at discharge (p = 0.028); predictors for non-invasive mechanical ventilation were kidney injury (p < 0.001), inotropes (p < 0.001), blood transfusions (p < 0.001) and PaO2/FiO2 ratio at the discharge (p < 0.001). Conclusions: In this retrospective study, we identified the preoperative, intraoperative and postoperative characteristics associated with mortality and complications following cardiac surgery.

Concomitant trauma of brain and upper cervical spine: lessons in injury patterns and outcomes

PURPOSE

The literature on concomitant traumatic brain injury (TBI) and traumatic spinal injury is sparse and a few, if any, studies focus on concomitant TBI and associated upper cervical injury. The objective of this study was to fill this gap and to define demographics, patterns of injury, and clinical data of this specific population.

METHODS

Records of patients admitted at a single trauma centre with the main diagnosis of TBI and concomitant C0-C1-C2 injury (upper cervical spine) were identified and reviewed. Demographics, clinical, and radiological variables were analyzed and compared to those of patients with TBI and: (i) C3-C7 injury (lower cervical spine); (ii) any other part of the spine other than C1-C2 injury (non-upper cervical); (iii) T1-L5 injury (thoracolumbar).

RESULTS

1545 patients were admitted with TBI and an associated C1-C2 injury was found in 22 (1.4%). The mean age was 64 years, and 54.5% were females. Females had a higher rate of concomitant upper cervical injury (p = 0.046 vs non-upper cervical; p = 0.050 vs thoracolumbar). Patients with an upper cervical injury were significantly older (p = 0.034 vs lower cervical; p = 0.030 vs non-upper cervical). Patients older than 55 years old had higher odds of an upper cervical injury when compared to the other groups (OR = 2.75). The main mechanism of trauma was road accidents (RAs) (10/22; 45.5%) All pedestrian injuries occurred in the upper cervical injured group (p = 0.015). ICU length of stay was longer for patients with an upper cervical injury (p = 0.018). Four patients died in the upper cervical injury group (18.2%), and no death occurred in other comparator groups (p = 0.003).

CONCLUSIONS

The rate of concomitant cranial and upper cervical spine injury was 1.4%. Risk factors were female gender, age ≥ 55, and pedestrians. RAs were the most common mechanism of injury. There was an association between the upper cervical injury group and longer ICU stay as well as higher mortality rates. Increased understanding of the pattern of concomitant craniospinal injury can help guide comprehensive diagnosis, avoid missed injuries, and appropriate treatment.

The Rationalisation of Intra-Operative Imaging During Cardiac Surgery: A Systematic Review

INTRODUCTION

One critical complication of cardiac surgery is cerebrovascular accidents (CVAs). Ascending aorta atherosclerosis poses a significant risk of embolisation to distal vessels and to cerebral arteries. Epi-aortic ultrasonography (EUS) is thought to offer a safe, high-quality accurate visualisation of the diseased aorta to guide the surgeon on the best surgical approach to the planned procedure and potentially improve neurological outcomes post-cardiac surgery.

METHOD

The authors conducted a comprehensive search of PubMed, Scopus and Embase. Studies that reported on epi-aortic ultrasound use in cardiac surgery were included. Major exclusion criteria were: (1) abstracts, conference presentations, editorials, literature reviews; (2) case series with <5 participants; (3) epi-aortic ultrasound in trauma or other surgeries.

RESULTS

A total of 59 studies and 48,255 patients were included in this review. Out of the studies that reported patient co-morbidities prior to cardiac surgery, 31.6% had diabetes, 59.5% had hyperlipidaemia and 66.1% had a diagnosis of hypertension. Of those that reported significant ascending aorta atherosclerosis found on EUS, this ranged from 8.3% of patients to 95.2% with a mean percentage of 37.8%. Hospital mortality ranged from 7% to 13%; four studies reported zero deaths. Long-term mortality and stroke rate varied significantly with hospital duration.

CONCLUSION

Current data have shown EUS to have superiority over manual palpation and transoesophageal echocardiography in the prevention of CVAs following cardiac surgery. Yet, EUS has not been implemented as a routine standard of care. Extensive adoption of EUS in clinical practice is warranted to aid large, randomised trials before making prospective conclusions on the efficacy of this screening method.

Artificial intelligence for clinical decision support for monitoring patients in cardiovascular ICUs: A systematic review

Background

Artificial intelligence (AI) and machine learning (ML) models continue to evolve the clinical decision support systems (CDSS). However, challenges arise when it comes to the integration of AI/ML into clinical scenarios. In this systematic review, we followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA), the population, intervention, comparator, outcome, and study design (PICOS), and the medical AI life cycle guidelines to investigate studies and tools which address AI/ML-based approaches towards clinical decision support (CDS) for monitoring cardiovascular patients in intensive care units (ICUs). We further discuss recent advances, pitfalls, and future perspectives towards effective integration of AI into routine practices as were identified and elaborated over an extensive selection process for state-of-the-art manuscripts.

Methods

Studies with available English full text from PubMed and Google Scholar in the period from January 2018 to August 2022 were considered. The manuscripts were fetched through a combination of the search keywords including AI, ML, reinforcement learning (RL), deep learning, clinical decision support, and cardiovascular critical care and patients monitoring. The manuscripts were analyzed and filtered based on qualitative and quantitative criteria such as target population, proper study design, cross-validation, and risk of bias.

Results

More than 100 queries over two medical search engines and subjective literature research were developed which identified 89 studies. After extensive assessments of the studies both technically and medically, 21 studies were selected for the final qualitative assessment.

Discussion

Clinical time series and electronic health records (EHR) data were the most common input modalities, while methods such as gradient boosting, recurrent neural networks (RNNs) and RL were mostly used for the analysis. Seventy-five percent of the selected papers lacked validation against external datasets highlighting the generalizability issue. Also, interpretability of the AI decisions was identified as a central issue towards effective integration of AI in healthcare.

Increasing obesity is associated with lower postoperative bleeding in coronary bypass patients

OBJECTIVE

Despite inherent comorbidities, obese cardiac surgical patients paradoxically had shown lower morbidity and mortality, although the nature of this association is still unclear. Thus, the authors intended in this large registry-based study to investigate the impact of obesity on short- and long-term postoperative outcomes, focusing on bleeding and transfusion requirements.

DESIGN

Retrospective registry study.

SETTING

Three university hospitals.

PARTICIPANTS

A cohort of 12,330 prospectively compiled data from coronary bypass grafting patients undergoing surgery between 2007 to 2020 were retrieved from the Western Denmark Heart Registry.

INTERVENTIONS

The parameters were analyzed to assess the association between body mass index (BMI) and the selected outcome parameters.

MEASUREMENTS AND MAIN RESULTS

The crude data showed a clear statistically significant association in postoperative drainage from 637 (418-1108) mL in underweight patients with BMI <18.5 kg/m² to 427 (295-620) mL in severely obese patients with BMI ≥40 kg/m² (p < 0.0001, Kruskal-Wallis). Further, 50.0% of patients with BMI <18.5 received an average of 451 mL/m² in red blood cell transfusions, compared to 16.7% of patients with BMI >40 receiving 84 mL/m². The obese groups were less often submitted to reexploration due to bleeding, and fewer received perioperative hemostatics, inotropes, and vasoconstrictors. The crude data showed increasing 30-day and 6-month mortality with lower BMI, whereas the one-year mortality showed a V-shaped pattern, but BMI had no independent impact on mortality in logistic regression analysis.

CONCLUSION

Patients with high BMI may carry protection against postoperative bleeding after cardiac surgery, probably secondary to an inherent hypercoagulable state, whereas underweight patients carry a higher risk of bleeding and worse outcomes.

Risk factors associated with prolonged intensive care unit stay in post coronary artery bypass grafting patients with chronic kidney disease

Objective

Prolonged intensive care unit stay not only increases hospital cost but it also prevents hospital equipment to be used by other patients who need them. The aim of this study was to identify factors that affect the duration of intensive care unit stay in post coronary artery bypass grafting patients with chronic kidney disease.

Method

This is a single centered observational prospective study done on 191 post coronary artery bypass grafting patients from June 2018 to April 2019 at Cardiac Surgery Unit of Doctor's hospital and medical center, Lahore, Pakistan. Patients above 18 years with and without chronic kidney disease were included.

Results

Mean age of the patients was 57.83 years (± 9.862 SD. Logistic regression analysis shows that patients with post op arrhythmias had the strongest positive association with prolonged intensive care unit stay (OR:11; p value :<0.01), followed by recent myocardial infarction less than 90 days pre coronary artery bypass grafting (OR:5.93; p value:<0.01), shock (OR:3.93;p value:0.04) and acute kidney injury (OR :2.08;p value:0.04). 37.5% chronic kidney disease patients with recent myocardial infarction less than 90 days pre coronary artery bypass grafting and 51.4% patients of chronic kidney disease found with acute kidney injury, showed significant association with p values less than 0.05.

Conclusion

Post op arrhythmias, recent myocardial infarction, shock and acute kidney injury are independent risk factors causing prolonged intensive care unit stay in post coronary artery bypass grafting patients.

Assessing Risk Factors for Prolonged Intensive Care Unit Stay After Surgery for Adult Congenital Heart Disease: A Study From a Lower-Middle-Income Country

Background Prolonged post-surgery intensive care unit (ICU) stay for congenital heart disease (CHD) has been explored in the pediatric population. However, there is limited data for adult CHD (ACHD), also called grown-up congenital heart (GUCH) disease, especially in low-resource countries where intensive care beds are scarce. This study identifies factors associated with prolonged ICU stay following surgery for ACHD in Pakistan, a lower-middle-income country (LMIC). Methods This retrospective study included all adult patients (⩾18 years) who underwent cardiac surgery with cardiopulmonary bypass for their CHD from 2011-2016 at a tertiary-care private hospital in Pakistan. Prolonged ICU stay was defined as stay >6 days (75^th percentile). Regression analysis was used to explore risk factors of prolonged ICU stay. Results A total of 166 patients (53.6% males) with a mean age of 32.05 ± 12.11 years were included. Atrial septal defect repair was the most common surgery (42.2%). Most patients were categorized as Risk Adjustment for Congenital Heart Surgery 1 (RACHS-1) Category 1 (51.8%) and Category 2 (30.1%). Forty-three of 166 patients (25.9%) experienced prolonged ICU stay. Complications occurred in 38.6% of patients postoperatively, with the most common being acute kidney injury (29.5%). On multivariable logistic regression adjusted for age, gender, and RACHS-1 categories, intraoperative inotrope score, cardiopulmonary bypass time, aortic cross-clamp time duration of mechanical ventilation, and postoperative acute kidney injury (AKI) were associated with prolonged ICU stay. Conclusion Surgeons managing ACHD in LMICs must strive for shorter operative durations and the judicious use of intraoperative inotropes in addition to anticipating and promptly managing postoperative complications such as AKI, to minimize ICU stay in countries where intensive care beds are a scarce resource.

Predicting readmission to the cardiovascular intensive care unit using recurrent neural networks

If a patient can be discharged from an intensive care unit (ICU) is usually decided by the treating physicians based on their clinical experience. However, nowadays limited capacities and growing socioeconomic burden of our health systems increase the pressure to discharge patients as early as possible, which may lead to higher readmission rates and potentially fatal consequences for the patients. Therefore, here we present a long short-term memory-based deep learning model (LSTM) trained on time series data from Medical Information Mart for Intensive Care (MIMIC-III) dataset to assist physicians in making decisions if patients can be safely discharged from cardiovascular ICUs. To underline the strengths of our LSTM we compare its performance with a logistic regression model, a random forest, extra trees, a feedforward neural network and with an already known, more complex LSTM as well as an LSTM combined with a convolutional neural network. The results of our evaluation show that our LSTM outperforms most of the above models in terms of area under receiver operating characteristic curve. Moreover, our LSTM shows the best performance with respect to the area under precision-recall curve. The deep learning solution presented in this article can help physicians decide on patient discharge from the ICU. This may not only help to increase the quality of patient care, but may also help to reduce costs and to optimize ICU resources. Further, the presented LSTM-based approach may help to improve existing and develop new medical machine learning prediction models.

Derivation and validation of pragmatic clinical models to predict hospital length of stay after cardiac surgery in Ontario, Canada: a population-based cohort study

BACKGROUND

Cardiac surgery is resource intensive and often requires multidisciplinary involvement to facilitate discharge. To facilitate evidence-based resource planning, we derived and validated clinical models to predict postoperative hospital length of stay (LOS).

METHODS

We used linked, population-level databases with information on all Ontario residents and included patients aged 18 years or older who underwent coronary artery bypass grafting, valvular or thoracic aorta surgeries between October 2008 and September 2019. The primary outcome was hospital LOS. The models were derived by using patients who had surgery before Sept. 30, 2016, and validated after that date. To address the rightward skew in LOS data and to identify top-tier resource users, we used logistic regression to derive a model to predict the likelihood of LOS being more than the 98th percentile (> 30 d), and γ regression in the remainder to predict continuous LOS in days. We used backward stepwise variable selection for both models.

RESULTS

Among 105 193 patients, 2422 (2.3%) had an LOS of more than 30 days. Factors predicting prolonged LOS included age, female sex, procedure type and urgency, comorbidities including frailty, high-risk acute coronary syndrome, heart failure, reduced left ventricular ejection fraction and psychiatric and pulmonary circulatory disease. The C statistic was 0.92 for the prolonged LOS model and the mean absolute error was 2.4 days for the continuous LOS model.

INTERPRETATION

We derived and validated clinical models to identify top-tier resource users and predict continuous LOS with excellent accuracy. Our models could be used to benchmark clinical performance based on expected LOS, rationally allocate resources and support patient-centred operative decision-making.

Outcomes of Prolonged ICU Stay for Patients Undergoing Cardiac Surgery in Australia and New Zealand

OBJECTIVE

To determine the effect of intensive care unit (ICU) length of stay (LOS) on hospital mortality and non-home discharge for patients undergoing cardiac surgery over a 16-year period in Australia and New Zealand.

DESIGN

A retrospective, multicenter cohort study covering the period January 1, 2004 to December 31, 2019.

SETTING

One hundred one hospitals in Australia and New Zealand that submitted data to the Australia New Zealand Intensive Care Society Adult Patient Database.

PARTICIPANTS

Adult patients (aged >18) who underwent coronary artery bypass grafting, valve surgery, or combined valve + coronary artery surgery.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

The authors analyzed 252,948 cardiac surgical patients from 101 hospitals, with a median age of 68.3 years (IQR 60-75.5), of whom 74.2% (187,632 of 252,948) were male patients. A U-shaped relationship was observed between ICU LOS and hospital mortality, with significantly elevated mortality at short (<20 hours) and long (>5 days) ICU LOS, which persisted after adjustment for illness severity and across clinically important subgroups (odds ratio for mortality with ICU LOS >5 days = 3.21, 95% CI 2.88-3.58, p < 0.001).

CONCLUSIONS

Prolonged duration of ICU LOS after cardiac surgery is associated with increased hospital mortality in a U-shaped relationship. An ICU LOS >5 days should be considered a meaningful definition for prolonged ICU stay after cardiac surgery.

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.

Prognostic value of combined preoperative phase angle and handgrip strength in cardiac surgery

OBJECTIVES

Phase angle (PA) constitutes a bioelectrical impedance measurement, indicating cell membrane health and integrity, hydration, and nutritional status. Handgrip strength (HS) has been also associated with body composition, nutritional status, inflammation, and functional ability in several chronic diseases. Although their prognostic significance as independent biomarkers has been already investigated regarding the outcomes of a cardiac surgery, our study is the first one to assess the combined predictive value of preoperative PA and HS.

DESIGN AND METHODS

HS and PA measurements were performed preoperativelyin 195 patients undergoing cardiac surgery. The association ofthe combination of HS and PAwith all-cause mortality rates was the primary study outcome, while its association with the intensive care unit (ICU) length of stay (LOS) was the secondary one.

RESULTS

PA was positively correlated with HS (r = 0.446, p < 0.005) and negatively with EuroSCORE II (r = - 0.306 p < 0.005). The combination of PA < 5.15 and HS < 25.5 was associated with higher one-year all-cause mortality (OR = 9.28; 95% CI 2.50-34.45; p = 0.001) compared to patients with PA > 5.15 and HS > 25.5, respectively. Patients with combined lower values of PA and HS (PA < 5.15 and HS < 30.7) were at higher risk of prolonged ICU LOS (OR = 4.02; 95% CI 1.53-10.56; p = 0.005) compared to those with higher PA-HS (PA > 5.15-HS > 30.7). The combination of PA-HS was also significantly linked with EuroSCORE II.

CONCLUSION

The combination of low preoperative PA and HS values was significantly associated with higher risk of all-cause mortality at 12 months and prolonged ICU LOS; thereby it might serve as a clinically useful prognostic biomarker after cardiac surgery procedures.

Parsimonious machine learning models to predict resource use in cardiac surgery across a statewide collaborative

Objective

Methods

Results

Conclusions

Robotic-assisted versus conventional off-pump coronary surgery in women: A propensity-matched study

BACKGROUND

Off-pump coronary artery bypass (OPCAB) previously demonstrated its potential benefits in women; however, robotic-assisted OPCAB was scarcely studied.

OBJECTIVES

To investigate whether robotic-assisted OPCAB could further improve the outcomes in women and the potential impact of hybrid approaches with stents and completeness of revascularization on the late outcomes.

METHODS

Women who underwent robotic-assisted or conventional OPCAB (with sternotomy) between May 2005 and January 2021 at Lankenau Heart Institute were included. Propensity score matching was used to match 273 pairs on 27 characteristics.

RESULTS

In the intraoperative period, women who underwent robotic-assisted OPCAB presented longer operative times (6.00 vs. 5.38 h; p < 0.001), higher rates of extubation in the operating room (83.9% vs. 75.5%; p = 0.019) and lower rates of blood transfusion (13.2% vs. 32.2%; p < 0.001). In the postoperative period, women who underwent robotic-assisted OPCAB presented lower rates of new onset atrial fibrillation (16.8% vs. 25.6%; p = 0.016), need of blood transfusion (33.0% vs. 54.9%; p < 0.001), shorter intensive care unit (ICU) (46.1 vs. 49.8 h; p = 0.006) and hospital length of stay (5.0 vs. 6.0 days; p < 0.001). We observed no statistically significant differences in the rates of operative death between the groups (1.47% vs. 1.47%; p = 0.771). In the follow-up, we observed no differences in terms of overall survival regardless of hybrid procedures with stents and completeness of revascularization.

CONCLUSIONS

Robotic-assisted OPCAB in women is as safe as conventional OPCAB and may further improve outcomes. Hybrid coronary revascularization was a valuable adjunct in the robotic scenario and completeness of revascularization did not play a role in this setting.

A comparison of statistical methods for modeling count data with an application to hospital length of stay

Background

Hospital length of stay (LOS) is a key indicator of hospital care management efficiency, cost of care, and hospital planning. Hospital LOS is often used as a measure of a post-medical procedure outcome, as a guide to the benefit of a treatment of interest, or as an important risk factor for adverse events. Therefore, understanding hospital LOS variability is always an important healthcare focus. Hospital LOS data can be treated as count data, with discrete and non-negative values, typically right skewed, and often exhibiting excessive zeros. In this study, we compared the performance of the Poisson, negative binomial (NB), zero-inflated Poisson (ZIP), and zero-inflated negative binomial (ZINB) regression models using simulated and empirical data.

Methods

Data were generated under different simulation scenarios with varying sample sizes, proportions of zeros, and levels of overdispersion. Analysis of hospital LOS was conducted using empirical data from the Medical Information Mart for Intensive Care database.

Results

Results showed that Poisson and ZIP models performed poorly in overdispersed data. ZIP outperformed the rest of the regression models when the overdispersion is due to zero-inflation only. NB and ZINB regression models faced substantial convergence issues when incorrectly used to model equidispersed data. NB model provided the best fit in overdispersed data and outperformed the ZINB model in many simulation scenarios with combinations of zero-inflation and overdispersion, regardless of the sample size. In the empirical data analysis, we demonstrated that fitting incorrect models to overdispersed data leaded to incorrect regression coefficients estimates and overstated significance of some of the predictors.

Conclusions

Based on this study, we recommend to the researchers that they consider the ZIP models for count data with zero-inflation only and NB models for overdispersed data or data with combinations of zero-inflation and overdispersion. If the researcher believes there are two different data generating mechanisms producing zeros, then the ZINB regression model may provide greater flexibility when modeling the zero-inflation and overdispersion.

Comparison of Four Machine Learning Techniques for Prediction of Intensive Care Unit Length of Stay in Heart Transplantation Patients

Background

Post-operative heart transplantation patients often require admission to an intensive care unit (ICU). Early prediction of the ICU length of stay (ICU-LOS) of these patients is of great significance and can guide treatment while reducing the mortality rate among patients. However, conventional linear models have tended to perform worse than non-linear models.

Materials and Methods

We collected the clinical data of 365 patients from Wuhan Union Hospital who underwent heart transplantation surgery between April 2017 and August 2020. The patients were randomly divided into training data (N = 256) and test data (N = 109) groups. 84 clinical features were collected for each patient. Features were validated using the Least Absolute Shrinkage and Selection Operator (LASSO) regression’s fivefold cross-validation method. We obtained Shapley Additive explanations (SHAP) values by executing package “shap” to interpret model predictions. Four machine learning models and logistic regression algorithms were developed. The area under the receiver operating characteristic curve (AUC-ROC) was used to compare the prediction performance of different models. Finally, for the convenience of clinicians, an online web-server was established and can be freely accessed via the website https://wuhanunion.shinyapps.io/PredictICUStay/.

Results

In this study, 365 consecutive patients undergoing heart transplantation surgery for moderate (NYHA grade 3) or severe (NYHA grade 4) heart failure were collected in Wuhan Union Hospital from 2017 to 2020. The median age of the recipient patients was 47.2 years, while the median age of the donors was 35.58 years. 330 (90.4%) of the donor patients were men, and the average surgery duration was 260.06 min. Among this cohort, 47 (12.9%) had renal complications, 25 (6.8%) had hepatic complications, 11 (3%) had undergone chest re-exploration and 19 (5.2%) had undergone extracorporeal membrane oxygenation (ECMO). The following six important clinical features were selected using LASSO regression, and according to the result of SHAP, the rank of importance was (1) the use of extracorporeal membrane oxygenation (ECMO); (2) donor age; (3) the use of an intra-aortic balloon pump (IABP); (4) length of surgery; (5) high creatinine (Cr); and (6) the use of continuous renal replacement therapy (CRRT). The eXtreme Gradient Boosting (XGBoost) algorithm presented significantly better predictive performance (AUC-ROC = 0.88) than other models [Accuracy: 0.87; sensitivity: 0.98; specificity: 0.51; positive predictive value (PPV): 0.86; negative predictive value (NPV): 0.93].

Conclusion

Using the XGBoost classifier with heart transplantation patients can provide an accurate prediction of ICU-LOS, which will not only improve the accuracy of clinical decision-making but also contribute to the allocation and management of medical resources; it is also a real-world example of precision medicine in hospitals.

Blood pressure variability during pediatric cardiac surgery is associated with acute kidney injury

BACKGROUND

Blood pressure variability (BPV), defined as the degree of variation between discrete blood pressure readings, is associated with poor outcomes in acute care settings. Acute kidney injury (AKI) is a common and serious postoperative complication of cardiac surgery with cardiopulmonary bypass (CPB) in children. No studies have yet assessed the association between intraoperative BPV during cardiac surgery with CPB and the development of AKI in children.

METHODS

A retrospective chart review of children undergoing cardiac surgery with CPB was performed. Intraoperative BPV was calculated using average real variability (ARV) and standard deviation (SD). Multiple regression models were used to examine the association between BPV and outcomes of AKI, hospital and intensive care unit (PICU) length of stay, and length of mechanical ventilation.

RESULTS

Among 231 patients (58% males, median age 8.6 months) reviewed, 51.5% developed AKI (47.9% Stage I, 41.2% Stage II, 10.9% Stage III). In adjusted models, systolic and diastolic ARV were associated with development of any stage AKI (OR 1.40, 95% CI 1.08-1.8 and OR 1.4, 95% CI 1.05-1.8, respectively). Greater diastolic SD was associated with longer PICU length of stay (β 0.94, 95% CI 0.62-1.2). When stratified by age, greater systolic ARV and SD were associated with AKI in infants ≤ 12 months, but there was no relationship in children > 12 months.

CONCLUSIONS

Greater BPV during cardiac surgery with CPB was associated with development of postoperative AKI in infants, suggesting that BPV is a potentially modifiable risk factor for AKI in this high-risk population.

Comparison of Warfarin Initiation at 3 mg Versus 5 mg for Anticoagulation of Patients with Mechanical Mitral Valve Replacement Surgery: A Prospective Randomized Trial

Background

The increased warfarin sensitivity observed after mechanical mitral valve replacement (MVR) operations dictates clinical discretion in warfarin dose initiation. Evidence is still lacking with regard to anticoagulation management of MVR patients.

Objective

This study aimed to compare initiating warfarin at the recommended dosing regimen versus empirically lowered doses intended to account for the variation in warfarin sensitivity.

Methods

A prospective, single-blind, randomized, comparative study was conducted in postoperative MVR patients. Patients were randomly assigned to either the 5 mg group (n = 25) or the 3 mg group (n = 25) and were initiated on a 5 or 3 mg warfarin dose, respectively. Time to target international normalized ratio (INR), time in therapeutic range, occurrence of bleeding/thromboembolic events, and cost of bridging with enoxaparin were assessed for both groups.

Results

Target INR was achieved earlier in the 5 mg group than in the 3 mg group (p = 0.033), with a mean ± SD of 5.3 ± 2.0 and 6.6 ± 2.0, respectively (95% confidence interval of the mean difference 1.022–1.890). Bleeding events did not differ significantly between the two groups. The cost of enoxaparin consumption per patient was significantly higher in the 3 mg group versus the 5 mg group (p = 0.002).

Conclusions

The initiation of warfarin at a 5 mg dose in MVR patients was more efficacious than the 3 mg dose in terms of time to reach the target INR. Moreover, the cost of enoxaparin bridging was significantly reduced with a 5 mg warfarin initiation dose. Bleeding events were comparable.

ClinicalTrials.gov ID

NCT04235569, 22 January 2020.

The effect of continuous intercostal nerve block vs. single shot on analgesic outcomes and hospital stays in minimally invasive direct coronary artery bypass surgery: a retrospective cohort study

BACKGROUND

Minimally invasive direct coronary artery bypass (MIDCAB) grafting surgery is accompanied by severe pain. Although continuous intercostal nerve block (CINB) has become one of the multimodal analgesic techniques in single port thoracoscopic surgery, its effects on MIDCAB are unclear. The purpose of this study was to compare the effects of CINB and single shot on analgesic outcomes and hospital stays in patients undergoing MIDCAB in a real-world setting.

METHODS

A retrospective cohort study was carried out at Peking University Third Hospital, China. Two hundred and sixteen patients undergoing MIDCAB were divided into two groups: a CINB group and a single block (SI) group. The primary outcome was postoperative maximal visual analog scale (VAS); secondary outcomes included the number of patients with maximal VAS ≤ 3, the demand for and consumed doses of pethidine and tramadol, and the length of intensive care unit (ICU) and hospital stays. The above data and the area under the VAS curve in the 70 h after extubation for the two subgroups (No. of grafts = 1) were also compared.

RESULTS

The maximum VAS was lower in the CINB group, and there were more cases with maximum VAS ≤ 3 in the CINB group: CINB 52 (40%) vs. SI 17 (20%), P = 0.002. The percentage of cases requiring tramadol and pethidine was less in CINB, P = 0.001. Among all patients, drug doses were significantly lower in the CINB group [tramadol: CINB 0 (0-100) mg vs. SI 100 (0-225) mg, P = 0.0001; pethidine: CINB 0 (0-25) mg vs. SI 25 (0-50) mg, P = 0.0004]. Further subgroup analysis showed that the area under the VAS curve in CINB was smaller: 28.05 in CINB vs. 30.41 in SI, P = 0.002. Finally, the length of ICU stay was shorter in CINB than in SI: 20.5 (11.3-26.0) h vs. 22.0 (19.0-45.0) h, P = 0.011.

CONCLUSIONS

CINB is associated with decreased demand for rescue analgesics and shorter length of ICU stay when compared to single shot intercostal nerve block. Additional randomized controlled trial (RCT) is needed to support these findings.

Application of machine learning missing data imputation techniques in clinical decision making: taking the discharge assessment of patients with spontaneous supratentorial intracerebral hemorrhage as an example

Background

There are often many missing values in medical data, which directly affect the accuracy of clinical decision making. Discharge assessment is an important part of clinical decision making. Taking the discharge assessment of patients with spontaneous supratentorial intracerebral hemorrhage as an example, this study adopted the missing data processing evaluation criteria more suitable for clinical decision making, aiming at systematically exploring the performance and applicability of single machine learning algorithms and ensemble learning (EL) under different data missing scenarios, as well as whether they had more advantages than traditional methods, so as to provide basis and reference for the selection of suitable missing data processing method in practical clinical decision making.

Methods

The whole process consisted of four main steps: (1) Based on the original complete data set, missing data was generated by simulation under different missing scenarios (missing mechanisms, missing proportions and ratios of missing proportions of each group). (2) Machine learning and traditional methods (eight methods in total) were applied to impute missing values. (3) The performances of imputation techniques were evaluated and compared by estimating the sensitivity, AUC and Kappa values of prediction models. (4) Statistical tests were used to evaluate whether the observed performance differences were statistically significant.

Results

The performances of missing data processing methods were different to a certain extent in different missing scenarios. On the whole, machine learning had better imputation performance than traditional methods, especially in scenarios with high missing proportions. Compared with single machine learning algorithms, the performance of EL was more prominent, followed by neural networks. Meanwhile, EL was most suitable for missing imputation under MAR (the ratio of missing proportion 2:1) mechanism, and its average sensitivity, AUC and Kappa values reached 0.908, 0.924 and 0.596 respectively.

Conclusions

In clinical decision making, the characteristics of missing data should be actively explored before formulating missing data processing strategies. The outstanding imputation performance of machine learning methods, especially EL, shed light on the development of missing data processing technology, and provided methodological support for clinical decision making in presence of incomplete data.

Robotic hybrid coronary revascularization versus conventional off-pump coronary bypass surgery in women with two-vessel disease

BACKGROUND

Hybrid coronary revascularization (HCR) treats coronary artery disease (CAD) by combining a minimally invasive surgical approach with the left internal mammary artery (LIMA) to the left anterior descending (LAD) artery and percutaneous coronary intervention (PCI) for non-LAD vessels. This study aimed to compare immediate and long-term outcomes between robotic HCR and off-pump coronary artery bypass (OPCAB) via sternotomy in women with two-vessel CAD.

METHODS AND RESULTS

We compared all robotic HCR (LIMA-to-LAD plus stent; n = 55) and OPCAB (LIMA-to-LAD plus saphenous vein graft; n = 54) performed at a single institution between May 2005 and January 2021. To adjust for the selection bias of receiving either HCR or OPCAB, we performed a propensity score analysis of 31 matched pairs. In the immediate postoperative period, no statistically significant difference was observed for operative mortality and HCR was associated with lower rates of blood transfusion (25.8% vs. 54.8%; p = .038), and shorter hospital length of stay (4.0 vs. 6.0 days; p = .009). After a mean follow-up of 7.0 ± 4.9 years, we observed no statistically significant differences between the groups for overall survival (hazard ratio [HR]: 0.48, 95% confidence interval [CI]: 0.09-2.64, p = .401), myocardial infarction (HR: 1.60, 95% CI: 0.14-17.64, p = .703), stroke (HR not assessable; almost zero events), target vessel revascularization (HR: 0.45, 95% CI: 0.08-2.47, p = .359), angina (HR: 0.64, 95% CI: 0.20-2.01, p = .444) and major adverse cardiac and cerebrovascular events (HR: 0.46, 95% CI: 0.14-1.52, p = .202).

CONCLUSIONS

Robotic HCR provides for women with two-vessel CAD a shorter postoperative recovery with fewer blood transfusions, with similar long-term outcomes when compared with conventional OPCAB via sternotomy.

Postoperative delirium after cardiac surgery of elderly patients as an independent risk factor for prolonged length of stay in intensive care unit and in hospital

Background

Postoperative delirium (POD) is a relevant and underdiagnosed complication after cardiac surgery that is associated with increased intensive care unit (ICU) and hospital length of stay (LOS). The aim of this subgroup study was to compare the frequency of tested POD versus the coded International Statistical Classification of Diseases and Related Health Problems (ICD) diagnosis of POD and to evaluate the influence of POD on LOS in ICU and hospital.

Methods

254 elective cardiac surgery patients (mean age, 70.5 ± 6.4 years) at the University Hospital Bonn between September 2018 and October 2019 were evaluated. The endpoint tested POD was considered positive, if one of the tests Confusion Assessment Method for ICU (CAM-ICU) or Confusion Assessment Method (CAM), 4 'A's Test (4AT) or Delirium Observation Scale (DOS) was positive on one day.

Results

POD occurred in 127 patients (50.0%). LOS in ICU and hospital were significantly different based on presence (ICU 165.0 ± 362.7 h; Hospital 26.5 ± 26.1 days) or absence (ICU 64.5 ± 79.4 h; Hospital 14.6 ± 6.7 days) of POD (p < 0.001). The multiple linear regression showed POD as an independent predictor for a prolonged LOS in ICU (48%; 95%CI 31–67%) and in hospital (64%; 95%CI 27–110%) (p < 0.001). The frequency of POD in the study participants that was coded with the ICD F05.0 and F05.8 by hospital staff was considerably lower than tests revealed by the study personnel.

Conclusion

Approximately 50% of elderly patients who underwent cardiac surgery developed POD, which is associated with an increased ICU and hospital LOS. Furthermore, POD is highly underdiagnosed in clinical routine.

Nurse-guided incentive spirometry use and postoperative pulmonary complications among cardiac surgery patients: A randomized controlled trial

AIMS

To assess the effect of nurse-guided use of incentive spirometer on postoperative oxygenation and pulmonary complications after coronary artery bypass graft surgery.

BACKGROUND

Deep breathing exercises have been shown to improve postoperative lung expansion and reduce pulmonary complications. An incentive spirometer is a deep breathing exercises device that imitates continuous sigh-like maximal inspiration.

DESIGN

Randomized control trial, two groups nonblinded parallel design.

METHODS

A total of n = 89 eligible patients were randomized to either control or intervention group. Patients in the intervention group received bihourly nurse-guided incentive spirometry for 48-h postextubation. The endpoints were: the number and duration of hypoxic events during the first 24-hr postsurgery, pneumonia and pulmonary function parameters. Data were collected May to September 2019.

RESULTS

Patients in the intervention group had a significantly lower mean number of hypoxic events with shorter duration and shorter length of stay in the hospital and the ICU. Patients in the intervention group also had greater postoperative forced expiratory volume in 1 second.

CONCLUSION

Nurse-guided use of the incentive spirometer reduces the risk of pulmonary complications and hospital length of stay after cardiac surgery.

Risk factors for prolonged intensive care unit stays in patients after cardiac surgery with cardiopulmonary bypass: A retrospective observational study

Objectives

Patients after cardiac surgery with cardiopulmonary bypass (CPB) require a stay in the ICU postoperatively. This study aimed to investigate the incidence of prolonged length of stay (LOS) in the ICU after cardiac surgery with CPB and identify associated risk factors.

Methods

The current investigation was an observational, retrospective study that included 395 ICU patients who underwent cardiac surgery with CPB at a tertiary hospital in Guangzhou from June 2015 to June 2017. Data were obtained from the hospital database. Binary logistic regression modeling was used to analyze risk factors for prolonged ICU LOS.

Results

Of 395 patients, 137 (34.7%) had a prolonged ICU LOS (>72.0 h), and the median ICU LOS was 50.9 h. Several variables were found associated with prolonged ICU LOS: duration of CPB, prolonged mechanical ventilation and non-invasive assisted ventilation use, PaO₂/FiO₂ ratios within 6 h after surgery, type of surgery, red blood cell infusion during surgery, postoperative atrial arrhythmia, postoperative ventricular arrhythmia (all P < 0.05).

Conclusions

These findings are clinically relevant for identifying patients with an estimated prolonged ICU LOS, enabling clinicians to facilitate earlier intervention to reduce the risk and prevent resulting delayed recovery.

Hospital Length of Stay Prediction Methods: A Systematic Review

OBJECTIVE

This systematic review sought to establish a picture of length of stay (LOS) prediction methods based on available hospital data and study protocols designed to measure their performance.

MATERIALS AND METHODS

An English literature search was done relative to hospital LOS prediction from 1972 to September 2019 according to the PRISMA guidelines. Articles were retrieved from PubMed, ScienceDirect, and arXiv databases. Information were extracted from the included papers according to a standardized assessment of population setting and study sample, data sources and input variables, LOS prediction methods, validation study design, and performance evaluation metrics.

RESULTS

Among 74 selected articles, 98.6% (73/74) used patients' data to predict LOS; 27.0% (20/74) used temporal data; and 21.6% (16/74) used the data about hospitals. Overall, regressions were the most popular prediction methods (64.9%, 48/74), followed by machine learning (20.3%, 15/74) and deep learning (17.6%, 13/74). Regarding validation design, 35.1% (26/74) did not use a test set, whereas 47.3% (35/74) used a separate test set, and 17.6% (13/74) used cross-validation. The most used performance metrics were R2 (47.3%, 35/74), mean squared (or absolute) error (24.4%, 18/74), and the accuracy (14.9%, 11/74). Over the last decade, machine learning and deep learning methods became more popular (P=0.016), and test sets and cross-validation got more and more used (P=0.014).

CONCLUSIONS

Methods to predict LOS are more and more elaborate and the assessment of their validity is increasingly rigorous. Reducing heterogeneity in how these methods are used and reported is key to transparency on their performance.

Forecasting the length-of-stay of pediatric patients in hospitals: a scoping review

BACKGROUND

Healthcare management faces complex challenges in allocating hospital resources, and predicting patients' length-of-stay (LOS) is critical in effectively managing those resources. This work aims to map approaches used to forecast the LOS of Pediatric Patients in Hospitals (LOS-P) and patients' populations and environments used to develop the models.

METHODS

Using the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) methodology, we performed a scoping review that identified 28 studies and analyzed them. The search was conducted on four databases (Science Direct, Scopus, Web of Science, and Medline). The identification of relevant studies was structured around three axes related to the research questions: (i) forecast models, (ii) hospital length-of-stay, and (iii) pediatric patients. Two authors carried out all stages to ensure the reliability of the review process. Articles that passed the initial screening had their data charted on a spreadsheet. Methods reported in the literature were classified according to the stage in which they are used in the modeling process: (i) pre-processing of data, (ii) variable selection, and (iii) cross-validation.

RESULTS

Forecasting models are most often applied to newborn patients and, consequently, in neonatal intensive care units. Regression analysis is the most widely used modeling approach; techniques associated with Machine Learning are still incipient and primarily used in emergency departments to model patients in specific situations.

CONCLUSIONS

The studies' main benefits include informing family members about the patient's expected discharge date and enabling hospital resources' allocation and planning. Main research gaps are associated with the lack of generalization of forecasting models and limited reported applicability in hospital management. This study also provides a practical guide to LOS-P forecasting methods and a future research agenda.

Hydroxocobalamin or Methylene Blue for Vasoplegic Syndrome in Adult Cardiothoracic Surgery

OBJECTIVE

To compare hydroxocobalamin and methylene blue for the treatment of vasopressor-refractory vasoplegic syndrome (VS) after adult cardiac surgery with cardiopulmonary bypass (CPB).

DESIGN

A retrospective, propensity-matched, cohort study was performed. The primary endpoints were the percentage change in vasopressor use at 30, 60, and 120 minutes, characterized as both norepinephrine equivalents and vasoactive inotropic score. Eligible patients who received methylene blue were matched 3:1 with patients who received hydroxocobalamin based on sequential organ failure assessment score, preoperative mechanical circulatory support, CPB duration, and use of pre-CPB vasopressors, angiotensin-converting enzyme inhibitors, or beta-blockers.

SETTING

A quaternary care academic medical center.

PARTICIPANTS

Adult patients who underwent cardiac surgery with CPB from July 2013 to June 2019.

INTERVENTIONS

Patients were included who received either hydroxocobalamin (5,000 mg) or methylene blue (median 1.2 mg/kg) for VS in the operating room during the index surgery or in the intensive care unit up to 24 hours after CPB separation.

MEASUREMENTS AND MAIN RESULTS

Of the 142 included patients, 120 received methylene blue and 22 received hydroxocobalamin. After matching, 66 patients in the methylene blue group were included in the analysis. Baseline demographics, surgical characteristics, and vasoactive medications were similar between groups. There were no significant between-group differences in percentage change in norepinephrine equivalents or vasoactive inotropic score at each timepoint.

CONCLUSIONS

In adult patients undergoing cardiothoracic surgery using CPB with VS, the ability to reduce vasopressor use was similar with hydroxocobalamin compared with methylene blue.

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.

Predicting Length of Stay of Coronary Artery Bypass Grafting Patients Using Machine Learning

BACKGROUND

There is a growing need to identify which bits of information are most valuable for healthcare providers. The aim of this study was to search for the highest impact variables in predicting postsurgery length of stay (LOS) for patients who undergo coronary artery bypass grafting (CABG).

MATERIALS AND METHODS

Using a single institution's Society of Thoracic Surgeons (STS) Registry data, 2121 patients with elective or urgent, isolated CABG were analyzed across 116 variables. Two machine learning techniques of random forest and artificial neural networks (ANNs) were used to search for the highest impact variables in predicting LOS, and results were compared against multiple linear regression. Out-of-sample validation of the models was performed on 105 patients.

RESULTS

Of the 10 highest impact variables identified in predicting LOS, four of the most impactful variables were duration intubated, last preoperative creatinine, age, and number of intraoperative packed red blood cell transfusions. The best performing model was an ANN using the ten highest impact variables (testing sample mean absolute error (MAE) = 1.685 d, R² = 0.232), which performed consistently in the out-of-sample validation (MAE = 1.612 d, R² = 0.150).

CONCLUSION

Using machine learning, this study identified several novel predictors of postsurgery LOS and reinforced certain known risk factors. Out of the entire STS database, only a few variables carry most of the predictive value for LOS in this population. With this knowledge, a simpler linear regression model has been shared and could be used elsewhere after further validation.

Association of frailty with days alive at home after cardiac surgery: a population-based cohort study

BACKGROUND

Frailty is a geriatric syndrome that leaves people vulnerable to adverse outcomes. In cardiac surgery, minimal data describe associations between frailty and patient-centred outcomes. Our objective was to estimate the association between frailty and days alive at home after cardiac surgery.

METHODS

We conducted a population-based cohort study using linked health administrative data in the Canadian province of Ontario. All individuals >65 yr at the time of cardiac surgery were assigned a frailty score using a validated frailty index. Days alive and at home in the 30 and 365 days after surgery were calculated. The unadjusted and adjusted associations between frailty and days alive at home were calculated.

RESULTS

We identified 61 389 patients from 2009 to 2015. Frailty was associated with reduced days at home within 30 days (adjusted ratio of means for every 10% increase in frailty=0.79; 95% confidence interval [CI], 0.78-0.81; P<0.0001) and 365 days (adjusted ratio of means for every 10% increase in frailty=0.92; 95% CI, 0.91-0.93; P<0.0001) of surgery. Results were consistent in sensitivity analyses (5.0 fewer days alive at home [95% CI, 4.8-5.2] within 30 days and 9.0 fewer days alive at home [95% CI, 8.7-9.2] within 365 days after surgery).

CONCLUSION

Frailty is associated with a reduction in days alive at home after major cardiac surgery. This information should be considered in prognostic discussions before surgery and in care planning for vulnerable older patient groups. Days alive at home may be a useful outcome for routine measurement in quality, reporting, and studies using routinely collected data.

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.

Predictive factors of a prolonged length of stay in a community Nursing-Led unit: A retrospective cohort study

AIMS AND OBJECTIVES

To describe the care provided to patients admitted into a community Nursing-Led inpatient unit and to identify factors predicting a length of stay exceeding an established threshold.

BACKGROUND

Few studies have been conducted to describe the care provided in a Nursing-Led unit. No studies have investigated factors affecting length of stay in these services.

DESIGN

Retrospective cohort study.

METHODS

Consecutive patients admitted to a community Nursing-Led unit between 2009-2015 were enrolled. Sociodemographic, medical and nursing care (diagnoses and activities) variables were collected from electronic health records. Descriptive analysis and a backward stepwise logistic regression model were applied. The study followed the STROBE guidelines.

RESULTS

The study enrolled 904 patients (mean age: 77.7 years). The most frequent nursing diagnoses were bathing self-care deficit and impaired physical mobility. The nursing activities most provided were enteral medication administration and vital signs measurement. Approximately 37% of the patients had a length of stay longer than the established threshold. Nine covariates, including being discharged to home, having an impaired memory nursing diagnosis or being treated for advanced wound care, were found to be independent predictors of prolonged length of stay. Variables related to medical conditions did not affect the length-of-stay threshold.

CONCLUSIONS

The length of stay in the community Nursing-Led unit was mainly predicted by conditions related to sociodemographic factors, nursing complexity and functional status. This result confirms that the medical and nursing needs of a community Nursing-Led unit population substantively differ from those of hospitalised acute patients.

RELEVANCE TO CLINICAL PRACTICE

The nursing complexity and related nursing care to be provided may be adopted as a criterion to establish the appropriate length of stay in the community Nursing-Led unit for each individual patient.

Clinical burden associated with postsurgical complications in major cardiac surgeries in Asia-Oceania countries: A systematic review and meta-analysis

BACKGROUND

Evidence on the burden of postsurgical complications is mainly from studies in western countries, and little is highlighted in the Asia-Oceania region. This study aimed to identify and compare the burden of postsurgical complications in major cardiac surgeries in Asia-Oceania countries.

METHODS

A systematic search was performed in PubMed, Embase, and CENTRAL between January 2000 and July 2018. Inclusion criteria were: (a) observational studies or randomized control trials; (b) studied in coronary artery bypass graft (CABG) and/or heart valve procedures; (c) measured postsurgical clinical outcomes; and (d) conducted in Asia-Oceania countries. Pooled effects were calculated using a random-effects model.

RESULTS

Of the 6032 articles screened, 472 studies with a total of 614 161 patients met the inclusion criteria. The pooled incidences (95% confidence interval) of hospital mortality and 30-day mortality were similar at 2.38% (2.16%-2.59%) and 2.33% (2.16%-2.50%), respectively. Length of stay (LOS) was 14.07 days (13.44-14.71 days). The incidence for atrial fibrillation (AF) and stroke/cerebrovascular accident (CVA) was 17.49% (15.99%-18.99%) and 1.64% (1.51%-1.78%), respectively. Below outcomes tended to be better in studies on CABG compared to heart valve procedures, including the incidence of hospital mortality (1.97%[1.75%-2.18%] vs 3.97% [3.29%-4.65%]), AF (16.47% [14.85%-18.10%] vs 21.98% [17.41%-26.54%]), stoke/CVA (1.51% [1n 37%-1.65%] vs 2.55% [2.07%-3.04%]), and mean LOS (days) (13.08 [12.51-13.65] vs 19.58 [16.72-22.45]). Similarly, all postsurgical complications tended to be higher in studies involving high-risk patients vs non-high-risk patients.

CONCLUSIONS

There are opportunities to improve clinical outcomes of patients with high surgical risks and those undertaking heart valve procedures, as they tend to have poorer survival and higher risk in developing postsurgical complications.