Current Surgical Risk Scores Overestimate Risk in Minimally Invasive Aortic Valve Replacement

Artificial Intelligence and Big Data Technologies in the Construction of Surgical Risk Prediction Model for Patients with Coronary Artery Bypass Grafting

The objective of this work was to predict the risk of mortality rate in patients with coronary artery bypass grafting (CABG) based on the risk prediction model of CABG using artificial intelligence (AI) and big data technologies. The clinical data of 2,364 patients undergoing CABG in our hospital from January 2019 to August 2021 were collected in this work. Based on AI and big data technology, business requirement analysis, system requirement analysis, complication prediction module, big data mining technology, and model building are carried out, respectively; the successful CABG risk prediction system includes case feature analysis service, risk warning service, and case retrieval service. The commonly used precision, recall, and F1-score were adopted to evaluate the quality of the gradient-boosted tree (GBT) model. The analysis proved that the GBT model was the best in terms of precision, F1-score, and area under the receiver operating characteristic curve (ROC). According to the CABG risk prediction model, 1,382 patients had a score of <0, 463 patients had a score of 0 ≤ score ≤ 2, 252 patients had a score of 2 < score ≤ 5, and 267 patients had a score of >5, which were stratified into four groups: A, B, C, and D. The actual number of in-hospital deaths was 25, and the in-hospital mortality rate was 1.05%. The mortality rate predicted by the CABG risk prediction model was 2.67 ± 1.82% (95% confidential interval (CI) (2.87-2.98)), which was higher than the actual value. The CABG risk prediction model showed the credible results only in group B with AUC = 0.763 > 0.7. In group B, 3 patients actually died, the actual mortality rate was 0.33%, and the predicted mortality rate was 0.96 ± 0.78 (95% CI (0.82-0.87)), which overestimated the mortality rate of patients in group B. It successfully constructed a CABG risk prediction model based on the AI and big data technologies, which would overestimate the mortality of patients with intermediate risk, and it is suitable for different types of heart diseases through continuous research and development and innovation, and provides clinical guidance value.

Sternal-Sparing Surgical Options in Combined Aortic Valve and Coronary Artery Disease: Proof of Concept

OBJECTIVE

The standard management of concomitant aortic valve (AV) and coronary artery disease has been coronary artery bypass and AV replacement (AVR). With the advent of minimally invasive options, many isolated lesions have been successfully managed using a sternal-sparing approach. In our institution, patients with isolated AV disease are offered minimally invasive surgical or transcatheter AVR, and those with isolated coronary artery disease are routinely managed with robotic totally endoscopic coronary artery bypass or percutaneous coronary intervention. Various combinations of these techniques can be used when a sternal-sparing posture is desired because of patient risk or preference. The aim of this study was to review the outcomes in patients with combined AV and coronary disease who were managed using sternal-sparing approaches.

METHODS

We reviewed the records of 10 patients in our minimally invasive surgical database who presented with concomitant AV and coronary artery disease and underwent combined sternal-sparing management of these 2 lesions using various combinations of minimally invasive approaches.

RESULTS

Four patients had totally endoscopic coronary artery bypass and minimally invasive AVR at the same time, 2 patients underwent transcatheter AVR followed by totally endoscopic coronary artery bypass, and 4 patients underwent minimally invasive AVR with percutaneous coronary intervention. There was no 30-day mortality. The duration of postoperative surgical hospital stay was 3.1 ± 0.9 days.

CONCLUSIONS

Sternal-sparing approaches in combined AV and coronary artery disease are feasible with patient-specific treatment selection of minimally invasive techniques.

Machine learning using institution-specific multi-modal electronic health records improves mortality risk prediction for cardiac surgery patients

Background

The Society of Thoracic Surgeons risk scores are widely used to assess risk of morbidity and mortality in specific cardiac surgeries but may not perform optimally in all patients. In a cohort of patients undergoing cardiac surgery, we developed a data-driven, institution-specific machine learning-based model inferred from multi-modal electronic health records and compared the performance with the Society of Thoracic Surgeons models.

Methods

All adult patients undergoing cardiac surgery between 2011 and 2016 were included. Routine electronic health record administrative, demographic, clinical, hemodynamic, laboratory, pharmacological, and procedural data features were extracted. The outcome was postoperative mortality. The database was randomly split into training (development) and test (evaluation) cohorts. Models developed using 4 classification algorithms were compared using 6 evaluation metrics. The performance of the final model was compared with the Society of Thoracic Surgeons models for 7 index surgical procedures.

Results

A total of 6392 patients were included and described by 4016 features. Overall mortality was 3.0% (n = 193). The XGBoost algorithm using only features with no missing data (336 features) yielded the best-performing predictor. When applied to the test set, the predictor performed well (F-measure = 0.775; precision = 0.756; recall = 0.795; accuracy = 0.986; area under the receiver operating characteristic curve = 0.978; area under the precision-recall curve = 0.804). eXtreme Gradient Boosting consistently demonstrated improved performance over the Society of Thoracic Surgeons models when evaluated on index procedures within the test set.

Conclusions

Machine learning models using institution-specific multi-modal electronic health records may improve performance in predicting mortality for individual patients undergoing cardiac surgery compared with the standard-of-care, population-derived Society of Thoracic Surgeons models. Institution-specific models may provide insights complementary to population-derived risk predictions to aid patient-level decision making.

Assessment of TVT and STS Risk Score Performances in Patients Undergoing Transcatheter Aortic Valve Replacement

Background

The Society of Thoracic Surgeons (STS) score has been used to risk stratify patients undergoing transcatheter aortic valve replacement (TAVR). The Transcatheter Valve Therapy (TVT) score was developed to predict in-hospital mortality in high/prohibitive-risk patients. Its performance in low and intermediate-risk patients is unknown. We sought to compare TVT and STS scores' ability to predict clinical outcomes in all-surgical-risk patients undergoing TAVR.

Methods

Consecutive patients undergoing TAVR from 2012-2020 within a large health care system were retrospectively reviewed and stratified by STS risk score. Predictive abilities of TVT and STS scores were compared using observed-to-expected mortality ratios (O:E) and area under the receiver operating characteristics curves (AUCs) for 30-day and 1-year mortality.

Results

We assessed a total of 3270 patients (mean age 79 ± 9 years, 45% female), including 191 (5.8%) low-risk, 1093 (33.4%) intermediate-risk, 1584 (48.4%) high-risk, and 402 (5.8%) inoperable. Mean TVT and STS scores were 3.5% ± 2.0% and 6.1% ± 4.3%, respectively. Observed 30-day and 1-year mortality were 2.8% (92/3270; O:E TVT 0.8 ± 0.16 vs STS 0.46 ± 0.09), and 13.2% (432/3270), respectively. In the all-comers population, both TVT and STS risk scores showed poor prediction of 30-day (AUC: TVT 0.68 [0.62-0.74] vs STS 0.64 [0.58-0.70]), and 1-year (AUC: TVT 0.65 [0.62-0.58] vs STS 0.65 [0.62-0.58]) mortality. After stratifying by surgical risk, discrimination of the TVT and STS scores remained poor in all categories at 30 days and 1 year.

Conclusions

An updated TAVR risk score with improved predictive ability across all-surgical-risk categories should be developed based on a larger national registry.

Does valve morphological type impact TAVR outcomes?

Appropriate treatment for the bicuspid aortic valve demands attention to detail across the spectrum of bicuspid morphological types. Transcatheter aortic valve replacement outcomes, while encouraging, require in-depth evaluation before generalization to improve the precision of care.

Hypertensive and Hypotensive Emergencies in Interventional Radiology

Hyper- and hypotensive emergencies represent some of the most severe clinical issues that can occur during or around an interventional radiology procedure. While some patients are known to be more predisposed to cardiovascular collapse, nearly all patients are at risk for such an outcome. This is particularly true of patients undergoing moderate sedation, with the possibility of cardiovascular compromise occurring not just due to the underlying pathology for which the patient is being treated, but as a complication of sedation itself. Understanding the underlying cause of hyper- or hypotension is paramount to performing an appropriate and timely intervention. While the underlying cause is being corrected-if possible-the changes in blood pressure themselves may need to be intervened upon to maintain cardiovascular stability in these patients. Interventional radiologists must be familiar with measures taken to correct hyper- or hypotensive emergencies, including the most commonly used medications to treat these disorders. This article discusses the most common etiologies of such clinical scenarios, and the most common interventions performed for these settings.

Comparison between the age, creatinine and ejection fraction II score and the European System for Cardiac Operative Risk Evaluation II: which score for which patient?

OBJECTIVES

Each surgical risk prediction model requires a validation analysis within a large 'real-life' sample. The aim of this study is to validate the age, creatinine and ejection fraction (ACEF) II risk score compared with the European System for Cardiac Operative Risk Evaluation (EuroSCORE) II.

METHODS

All patients operated on at 8 Italian cardiac surgery centres in the period 2009-2019 with available data for the calculation of EuroSCORE II and ACEF II were included in the study. Mortality was recorded and receiver operating characteristic curves were plotted for the overall study population and for different patient subgroups according to the type of surgery.

RESULTS

A total of 14 804 patients were enrolled [median age of 70 (62-77) years, 35.4% female], and among these, 3.1% underwent emergency surgery. Thirty-day mortality was 2.84% (n = 420). In the total population, the area under the curve with EurosCORE II was significantly higher than that recorded with ACEF II [0.792, 95% confidence interval (CI) 0.79-0.8 vs 0.73, 95% CI 0.73-0.74; P < 0.001]. This finding was also confirmed in the patient subgroups undergoing isolated valve surgery (EuroSCORE II versus ACEF II: 0.80, 95% CI 0.79-0.814 vs 0.74, 95% CI 0.724-0.754; P = 0.045) or isolated aortic surgery (0.754, 95% CI 0.70-0.79 vs 0.53, 95% CI 0.48-0.58; P = 0.002). In contrast, the 2 scores did not differ significantly in patients undergoing isolated bypass surgery (0.8, 95% CI 0.78-0.81 vs 0.77, 95% CI 0.75-0.78; P = 1).

CONCLUSIONS

In both the overall population and patient subgroups, EuroSCORE II proved to be more accurate than ACEF II. However, in patients undergoing bypass surgery, ACEF II proved to be an easy and simple to use risk score, demonstrating comparable risk prediction performance with the more complex EuroSCORE II.