Artificial Intelligence Trumps TAVI2-SCORE and CoreValve Score in Predicting 1-Year Mortality Post-Transcatheter Aortic Valve Replacement

Agasthi, Pradyumna; Ashraf, Hasan; Pujari, Sai Harika; Girardo, Marlene; Tseng, Andrew; Mookadam, Farouk; Venepally, Nithin R; Buras, Matthew R.; Khetarpal, Banveet Kaur; Allam, Mohamed; Eleid, Mackram F.; Greason, Kevin L.; Beohar, Nirat; Siegel, Robert J.; Sweeney, John P.; Fortuin, F. David; Holmes, David R.; Arsanjani, Reza

doi:10.1016/j.carrev.2020.08.010

Cited by 31 publications

(31 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although our cohort size in the current study is slightly smaller due to excluding patients with missing values, the accuracy of this single parameter model using AugMAP1 almost approaches the same level of our previously published GBM machine learning model developed from the same database (5). Actually, among all patients with available AugMAP1 measurements in our TAVR database (n=971), the AUC of AugMAP1 was 0.703 (95%CI: 0.653 -0.743) in contrast to 0.611(95%CI: 0.554 -0.665) for STS risk score (p=0.009).…”

Section: An Overlooked Outcome Predictor: Augmented Blood Pressuresupporting

confidence: 67%

“…Given the underlying comorbidities, the clinical outcomes after TAVR procedure have gained significant attention that involves both conventional and machine learning research approaches (5)(6)(7)(8)(9)(10)(11)(12). There are newer studies focused on the relationship of blood pressure and valvuloarterial impedance (Zva), which estimates the overall left ventricular afterload, and its effect on TAVR prognosis (8)(9)(10)13).…”

Section: Background/introductionmentioning

confidence: 99%

See 1 more Smart Citation

A novel echo-parameter supersedes the Society of Thoracic Surgeons risk score in predicting post transcatheter aortic valve replacement mortality

Chao

Agasthi

Seri

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Background/Introduction In the era transcatheter aortic valve replacement (TAVR), predicting post-procedural outcome of patients is one of the most important goals in structural heart research. We proposed new parameters (augmented blood pressure) derived from blood pressure and aortic valve gradient measurements and hypothesized that they can predict post-TAVR 1-year mortality. Materials and Methods Patients in the Mayo Clinic National Cardiovascular Diseases Registry (NCDR)-TAVR database who underwent TAVR between January 1, 2014 and June 30, 2017 were identified to retrieve baseline demographics, STS risk score ECG, cardiac computed tomography, echocardiographic and mortality data. Augmented blood pressure parameters and valvulo-arterial impedance were evaluated by Cox regression. After logistic model generation, receiver operating curve analysis was used to assess the model performance against STS risk score. Results The final cohort contains 883 patients. The mean age is 81.3+/-8.5 years old, 58.2% are male. The mean STS risk score is 8.1+/-5.1. The median follow-up duration is 353 days and one-year all-cause mortality rates is 13.3%. Multivariate Cox regression showed that augmented SBP and augmented MAP parameters are independently predictors of 1-year mortality (all p<0.0001). A single-parameter model based on augmented MAP1 supersedes STS risk score in prediction mortality (AUC 0.697 vs. 0.591, p=0.0055). Conclusion Augmented mean arterial pressure provides a simple but effective approach for clinicians to quickly estimate the clinical outcome of TAVR patients and should be incorporated in the assessment of TAVR candidacy.

show abstract

Section: An Overlooked Outcome Predictor: Augmented Blood Pressuresupporting

confidence: 67%

Section: Background/introductionmentioning

confidence: 99%

A novel echo-parameter supersedes the Society of Thoracic Surgeons risk score in predicting post transcatheter aortic valve replacement mortality

Chao

Agasthi

Seri

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…[26][27][28][29] Studies evaluating the relationship with mid-long term and first one-year mortality are relatively few in number and include specific patient groups. [30,31] Considering the results of our study, it was observed that there was no significant relationship between predictive scoring after orthopedic surgery and first year mortality. Although the findings of our study do not clarify this, it should be considered that this relationship may be significant in specific orthopedic surgery patient populations with high comorbidity.…”

Section: Discussionmentioning

confidence: 63%

The Relationship Between SOFA, qSOFA, SIRS Criteria Scoring Systems and 1-Year Mortality in Patients Followed Up in Level 3 Intensive Care Unit after Orthopedic Surgery

Eren

Öztürk

Bilgic

et al. 2021

Journal of Contemporary Medicine

View full text Add to dashboard Cite

Predictive scoring systems are designed to predict patients' discharge status, patient mortality, Intensive Care Unit(ICU) mortality and length of hospital stay. These scoring systems, which are aimed to standardize and form a common language in terms of evaluating patient general health situation, are used in the evaluation of patients in many ICU. In this study, we aimed to evaluate the ability of the scoring systems that can be used to predict the mortality of the patients who will be followed up in the ICU after orthopedic surgery and predict the mortality that occurs in the first year outside the hospital. We established our hypothesis that there will be no difference between the predictive power of predictive scoring systems, which can be used as a marker of mortality, since relatively short-term hospitalizations occur in the ICU after orthopedic surgery. Material and Method: Our study was designed as a single center retrospective. Between January 2017 and August 2018, all patients undergoing level 3 intensive care follow-up were identified in the automation system. After the evaluation of 146 patients' files and automation system data, patients who were in compliance with the inclusion and exclusion criteria were identified. Result: A total of 40 patients were included in the study. The mean age was 75.69 (SD = 8.66 years), 50% male and 50% female. Fifteen patients (37.5%) died during their first year follow-up. When the patients were evaluated as survivors and non-survivors in the first year, it was observed that there was a significant difference between the groups in terms of quantitative variables in terms of early postoperative GCS and in terms of early postoperative oxygen saturation. There was no significant difference between the two groups in terms of qualitative variables (p>0.005). Conclusion:An appropriate mortality predictor score should be selected to standardize follow-up in intensive care unit.As a result of our study, there was no significant relationship between groups SOFA, qSOFA scores and SIRS criteria met suggested that these scoring systems were not suitable for predicting 1-year mortality in our patient group. We believe that specific criteria should be established with studies with larger series and more criteria.

show abstract

“…Clinical Epidemiology 2022:14 16 outcomes, 10,11 short-term prognosis 12,13 and long-term mortality 14 after TAVR. Evidences have shown that machine learning outperformed traditional linear regression models in these classification tasks.…”

Section: Dovepressmentioning

confidence: 99%

“…Machine learning has drawn significant attention in clinical prediction models in recent years, due to its ability to effectively model linear and non-linear relationships and interactions. 9 These techniques have been applied to predict in-hospital outcomes, 10,11 short-term prognoses 12,13 and long-term mortality 14 after TAVR. Despite the great performance of traditional machine learning methods (naive Bayes, random forest, support vector machine, etc) in predicting category outcomes (eg dead or alive), it is challenging for them to deal with time-to-event outcomes as in survival analysis.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning in Prediction of Late Major Bleeding After Transcatheter Aortic Valve Replacement

Jia¹,

Luosang²,

Li³

et al. 2022

CLEP

View full text Add to dashboard Cite

Late major bleeding is one of the main complications after transcatheter aortic valve replacement (TAVR). We aimed to develop a risk prediction model based on deep learning to predict major or life-threatening bleeding complications (MLBCs) after TAVR. Patients and Methods: This was a retrospective study including TAVR patients from West China Hospital of Sichuan University Transcatheter Aortic Valve Replacement Registry (ChiCTR2000033419) between April 17, 2012 and May 27, 2020. A deep learning-based model named BLeNet was developed with 56 features covering baseline, procedural, and post-procedural characteristics. The model was validated with the bootstrap method and evaluated using Harrell's concordance index (c-index), receiver operating characteristics (ROC) curve, calibration curve, and Kaplan-Meier estimate. Captum interpretation library was applied to identify feature importance. The BLeNet model was compared with the traditional Cox proportional hazard (Cox-PH) model and the random survival forest model in the metrics mentioned above. Results: The BLeNet model outperformed the Cox-PH and random survival forest models significantly in discrimination [optimism-corrected c-index of BLeNet vs Cox-PH vs random survival forest: 0.81 (95% CI: 0.79-0.92) vs 0.72 (95% CI: 0.63-0.77) vs 0.70 (95% CI: 0.61-0.74)] and calibration (integrated calibration index of BLeNet vs Cox-PH vs random survival forest: 0.007 vs 0.015 vs 0.019). In Kaplan-Meier analysis, BLeNet model had great performance in stratifying high-and low-bleeding risk patients (p < 0.0001). Conclusion:Deep learning is a feasible way to build prediction models concerning TAVR prognosis. A dedicated bleeding risk prediction model was developed for TAVR patients to facilitate well-informed clinical decisions.

show abstract

Artificial Intelligence Trumps TAVI2-SCORE and CoreValve Score in Predicting 1-Year Mortality Post-Transcatheter Aortic Valve Replacement

Cited by 31 publications

References 38 publications

A novel echo-parameter supersedes the Society of Thoracic Surgeons risk score in predicting post transcatheter aortic valve replacement mortality

A novel echo-parameter supersedes the Society of Thoracic Surgeons risk score in predicting post transcatheter aortic valve replacement mortality

The Relationship Between SOFA, qSOFA, SIRS Criteria Scoring Systems and 1-Year Mortality in Patients Followed Up in Level 3 Intensive Care Unit after Orthopedic Surgery

Deep Learning in Prediction of Late Major Bleeding After Transcatheter Aortic Valve Replacement

Contact Info

Product

Resources

About