Prevalence, mortality, cost, and disparities in transcatheter mitral valve repair and replacement in cancer patients: Artificial intelligence and propensity score national 5-year analysis of 7495 procedures

Konstantinos Marmagkiolis; Dominique J Monlezun; Jaime Caballero; Mehmet Cilingiroglu; Matthew N Brown; Vlasis Ninios; Ali Abdelrahman; Cezar A Iliescu

doi:10.1016/j.ijcard.2024.132091

Prevalence, mortality, cost, and disparities in transcatheter mitral valve repair and replacement in cancer patients: Artificial intelligence and propensity score national 5-year analysis of 7495 procedures

Int J Cardiol. 2024 Apr 23:132091. doi: 10.1016/j.ijcard.2024.132091. Online ahead of print.

Authors

Konstantinos Marmagkiolis¹, Dominique J Monlezun², Jaime Caballero³, Mehmet Cilingiroglu², Matthew N Brown², Vlasis Ninios⁴, Ali Abdelrahman², Cezar A Iliescu²

Affiliations

¹ University of Texas Houston, MD Anderson Cancer Center, Houston, TX, United States of America; University of South Florida, Tampa, FL, United States of America; Tampa General Hospital, Tampa, FL, United States of America. Electronic address: c.marmagiolis@gmail.com.
² University of Texas Houston, MD Anderson Cancer Center, Houston, TX, United States of America.
³ University of South Florida, Tampa, FL, United States of America; Tampa General Hospital, Tampa, FL, United States of America.
⁴ 2nd Cardiology Department, Interbalkan Medical Center, 55535 Thessaloniki, Greece.

PMID: 38663811
DOI: 10.1016/j.ijcard.2024.132091

Abstract

Introduction: We conducted the first comprehensive evaluation of the therapeutic value and safety profile of transcatheter mitral edge-to-edge repair (TEER) and transcatheter mitral valve replacement (TMVR) in individuals concurrently afflicted with cancer.

Methods: Utilizing the National Inpatient Sample (NIS) dataset, we analyzed all adult hospitalizations between 2016 and 2020 (n = 148,755,036). The inclusion criteria for this retrospectively analyzed prospective cohort study were all adult hospitalizations (age 18 years and older). Regression and machine learning analyses in addition to model optimization were conducted using ML-PSr (Machine Learning-augmented Propensity Score adjusted multivariable regression) and BAyesian Machine learning-augmented Propensity Score (BAM-PS) multivariable regression.

Results: Of all adult hospitalizations, there were 5790 (0.004%) TMVRs and 1705 (0.001%) TEERs. Of the total TMVRs, 160 (2.76%) were done in active cancer. Of the total TEERs, 30 (1.76%) were done in active cancer. After the comparable rates of TEER/TMVR in active cancer in 2016, the prevalence of TEER/TMVR was significantly less in active cancer from 2017 to 2020 (2.61% versus 7.28% p < 0.001). From 2017 to 2020, active cancer significantly decreased the odds of receiving TEER or TMVR (OR 0.28, 95%CI 0.13-0.68, p = 0.008). In patients with active cancer who underwent TMVR/TEER, there were no significant differences in socio-economic disparities, mortality or total hospitalization costs.

Conclusion: The presence of malignancy does not contribute to increased mortality, length of stay or procedural costs in TMVR or TEER. Whereas the prevalence of TMVR has increased in patients with active cancer, the utilization of TEER in the context of active cancer is declining despite a growing patient population.

Keywords: Cardio-oncology; Interventional cardiology; Valvular heart disease.