Dataset Information

Pre-operative Machine Learning for Heart Transplant Patients Bridged with Temporary Mechanical Circulatory Support.

ABSTRACT: Background: Existing prediction models for post-transplant mortality in patients bridged to heart transplantation with temporary mechanical circulatory support (tMCS) perform poorly. A more reliable model would allow clinicians to provide better pre-operative risk assessment and develop more targeted therapies for high-risk patients. Methods: We identified adult patients in the United Network for Organ Sharing database undergoing isolated heart transplantation between 01/2009 and 12/2017 who were supported with tMCS at the time of transplant. We constructed a machine learning model using extreme gradient boosting (XGBoost) with a 70:30 train:test split to predict 1-year post-operative mortality. All pre-transplant variables available in the UNOS database were included to train the model. Shapley Additive Explanations was used to identify and interpret the most important features for XGBoost predictions. Results: A total of 1584 patients were included, with a median age of 56 (interquartile range: 46-62) and 74% male. Actual 1-year mortality was 12.1%. Out of 498 available variables, 43 were selected for the final model. The area under the receiver operator characteristics curve (AUC) for the XGBoost model was 0.71 (95% CI: 0.62-0.78). The most important variables predictive of 1-year mortality included recipient functional status, age, pulmonary capillary wedge pressure (PCWP), cardiac output, ECMO usage, and serum creatinine. Conclusions: An interpretable machine learning model trained on a large clinical database demonstrated good performance in predicting 1-year mortality for patients bridged to heart transplantation with tMCS. Machine learning may be used to enhance clinician judgement in the care of markedly high-risk transplant recipients.

SUBMITTER: Shou BL

PROVIDER: S-EPMC9500687 | biostudies-literature | 2022 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Pre-operative Machine Learning for Heart Transplant Patients Bridged with Temporary Mechanical Circulatory Support.

Shou Benjamin L BL Chatterjee Devina D Russel Joseph W JW Zhou Alice L AL Florissi Isabella S IS Lewis Tabatha T Verma Arjun A Benharash Peyman P Choi Chun Woo CW

Journal of cardiovascular development and disease 20220919 9

<b>Background:</b> Existing prediction models for post-transplant mortality in patients bridged to heart transplantation with temporary mechanical circulatory support (tMCS) perform poorly. A more reliable model would allow clinicians to provide better pre-operative risk assessment and develop more targeted therapies for high-risk patients. <b>Methods:</b> We identified adult patients in the United Network for Organ Sharing database undergoing isolated heart transplantation between 01/2009 and 1 ...[more]

PMID: 36135456

Similar Datasets

Project description:BackgroundThe 2018 U.S. heart allocation system offers an accelerated pathway for heart transplantation to the most urgent patients.ObjectivesThis study sought to determine whether the new allocation system resulted in lower likelihood of candidate recovery.MethodsAdult patients waitlisted for heart transplantation with temporary mechanical circulatory support at the time of initial listing between 2010 and 2020 in the United Network for Organ Sharing registry were included. Competing events of heart transplantation, waitlist death or delisting for deteriorating condition, and delisting for improved condition (candidate recovery) were analyzed in the new vs old heart allocation system.ResultsA total of 688 patients were waitlisted with venoarterial extracorporeal membrane oxygenation or a surgical nondischargeable biventricular assist device (status 1 or old 1A). Overall, 2,237 patients were waitlisted with an intra-aortic balloon pump, a percutaneous left ventricular assist device (LVAD), or a surgical nondischargeable LVAD (status 2 or old 1A). Patients waitlisted with venoarterial extracorporeal membrane oxygenation or a nondischargeable biventricular assist device had significantly shorter median waitlist times (5 vs 31 days), higher incidence for cardiac transplantation (81.5% vs 43.0%), and lower incidence of candidate recovery (1.5% vs 7.9%) in the new vs old heart allocation system (all P < 0.05). Patients waitlisted with an intra-aortic balloon pump or percutaneous or a nondischargeable LVAD also had significantly shorter median waitlist times (8 vs 35 days), higher incidence of transplantation (88.9% vs 64.9%), and lower incidence of candidate recovery (0.2% vs 1.6%) in the new vs old heart allocation system (all P < 0.05).ConclusionsCurrent practice of the new allocation system may not offer select temporary mechanical circulatory support patients the opportunity and adequate time to recover to the point of waitlist removal. Further research will determine which patients would benefit from urgent transplantation vs recovery strategy.

Dataset Information

Pre-operative Machine Learning for Heart Transplant Patients Bridged with Temporary Mechanical Circulatory Support.

Publications

Pre-operative Machine Learning for Heart Transplant Patients Bridged with Temporary Mechanical Circulatory Support.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets