Academic Appointments


All Publications


  • National Evaluation of the Updated ISHLT Early Graft Dysfunction Framework in Adult Heart Transplantation: Diagnostic Migration, Attrition, and Prognostic Implications. Transplantation Leon, M., Ruaengsri, C., He, H., Pelletier, G., Flores, M. Q., Bethencourt, D. M., Shudo, Y. 2026

    Abstract

    The International Society for Heart and Lung Transplantation recently published a revised consensus on graft dysfunction after heart transplantation, introducing substantial modifications, including a shift toward more stringent conjunctive physiologic criteria. How these changes affect disease incidence, diagnostic migration, and prognostic interpretation when applied to a national registry remains unknown.Using adult heart transplant data from the United Network for Organ Sharing (September 2023 to June 2025), we applied the previous primary graft dysfunction (PGD) and updated early graft dysfunction (EGD) criteria with registry-adapted operationalization, quantified diagnostic migration between frameworks, and compared post-transplant outcomes across reclassified groups using Kaplan-Meier analyses.Among 2482 recipients, left ventricular dysfunction was identified in 334 (13.5%) under the previous PGD framework versus 11 (0.4%) under the updated EGD framework, representing a 96.7% relative reduction. Only 8 cases retained identical severity grading across both frameworks. Most discordant cases failed to meet hemodynamic criteria despite a left ventricular ejection fraction ≤40%. Notably, 184 recipients previously classified as severe PGD-left ventricular (LV) no longer met EGD-LV criteria yet had significantly higher mortality than recipients without graft dysfunction (20.1% versus 6.2%; P < 0.001). Right ventricular (RV) dysfunction incidence changed modestly (PGD-RV: 32 [1.3%]; EGD-RV: 28 [1.1%]).The updated EGD framework reclassified most previous PGD-LV cases as non-EGD, yet many reclassified recipients remained at elevated risk for graft failure and mortality. These findings reveal a substantial gap between the updated consensus criteria and current registry data, driven largely by failure to satisfy concurrent physiologic criteria required for EGD-LV classification.

    View details for DOI 10.1097/TP.0000000000005937

    View details for PubMedID 42905652

  • Impact of social determinants of health on perioperative and interstage outcomes in congenital heart surgery a systematic review FRONTIERS IN CARDIOVASCULAR MEDICINE Sakowitz, S., Nelson, J. S., Robinson, J., Rossi, C. S., Shahian, D. M., Cooke, D. T., Jacobs, J. P., Aughtman, S., Schultz, M., Nathan, M., Beaty, C., Shafer, M., Kozik, D., Pelletier, G. J. 2026; 13
  • Heart-lung transplantation using temperature- and airway pressure-controlled donor organ preservation system. JTCVS techniques Duda, M. M., Elde, S., Pelletier, G. J., Leon, M., Verdi, K. G., Ruaengsri, C., Woo, Y. J., Shudo, Y. 2026; 38: 102404

    View details for DOI 10.1016/j.xjtc.2026.102404

    View details for PubMedID 42577010

    View details for PubMedCentralID PMC13453641

  • Blinded two-phase evaluation of large language models in complex cardiac surgery: task-specific performance and human-AI collaboration. Frontiers in digital health Leon, M., Feng, R., Quiroz Flores, M., Pelletier, G., Bethencourt, D., Shibata, M., He, H., Ruaengsri, C. 2026; 8: 1769467

    Abstract

    Large language models (LLMs) have demonstrated strong performance on standardized medical benchmarks. However, their potential in complex surgical decision-making is largely uncharacterized. Critically, human-LLM collaboration regarding the extent to which clinicians can effectively recognize and integrate model-generated reasoning has emerged as an unaddressed question. To address these gaps, we developed a two-phase evaluation framework to simultaneously assess LLM performance and human-LLM collaboration in cardiac surgery.A panel of senior cardiac surgeons independently developed 15 high-fidelity cardiac surgery scenarios, each paired with a clinically relevant open-ended reasoning task, expert-curated reference answers, and a 10-dimensional weighted evaluation framework. Five representative LLMs (O1, O3-mini-high, DeepSeek-R1, GPT-4, and Llama3-OpenBioLLM-70B) were prompted using a multi-agent strategy. A separate group of senior surgeons conducted a blinded two-phase evaluation to assess model performance and evaluator judgment shifts: in the first round, they rated LLMs independently; in the second, they were shown the reference answers and invited to revise their ratings, with changes being optional.LLM performance varied across scenarios, but relative rankings remained stable. Median normalized scores were highest for O1 (0.896), followed by O3-mini-high (0.854), DeepSeek-R1 (0.792), GPT-4 (0.667), and Llama3-OpenBioLLM-70B (0.521). Across evaluation dimensions, scenario comprehension scored highest (0.920), while patient safety (0.507), hallucination avoidance (0.549), and clinical efficiency (0.597) were lowest across models. Second-round normalized scores declined for four LLMs, with 7.57% of ratings revised from affirmative to negative and only 2.59% from negative to affirmative. Among the five highest-weighted evaluation dimensions, 10.16% of second-round ratings were revised from affirmative to negative.Reasoning-optimized LLMs outperformed all other models. However, all models exhibited clinical limitations, including poor performance in core evaluation dimensions and scenarios requiring complex, longitudinal reasoning tasks. Overacceptance was the dominant collaboration imbalance, reflecting that clinicians over-accepted model reasoning that appears clinically sound yet is incorrect or potentially harmful. These findings suggest that these LLMs are not yet ready for safe use in complex surgical settings due to both performance limitations and human-LLM collaboration imbalance.

    View details for DOI 10.3389/fdgth.2026.1769467

    View details for PubMedID 42294049

    View details for PubMedCentralID PMC13260534

  • Heart-lung transplantation using temperature- and airway pressure-controlled donor organ preservation system JTCVS Techniques Duda, M. M., Elde, S., Pelletier, G. J., Leon, M., Verdi, K. G., Ruaengsri, C., Woo, Y., Shudo, Y. 2026
  • Histologic changes in systemically loaded right ventricles at time of transplantation. Frontiers in cardiovascular medicine Nilkant, R., Choi, P. S., Mullis, D. M., Chang, R., Sharir, A., Reddy, S., Pelletier, G. J., Ma, M. 2025; 12: 1640561

    Abstract

    In a subset of complex congenital heart disease patients, the right ventricle (RV) is connected to the higher-resistance systemic circulation, often leading to RV dysfunction. This study characterizes the systemic RV histology at time of heart transplantation in two groups - patients with hypoplastic left heart syndrome with Fontan palliation (HLHS-F) and those with dextro-transposition of the great arteries post atrial switch operation (d-TGA-AS). We sought to better understand histological differences between the systemic RV in the single ventricle vs. biventricular circulations.We procured RV tissue samples at the mid-cavity free walls from nine explanted recipient hearts: six HLHS-F and three d-TGA-AS. RV and LV muscle samples from two organ donors whose hearts were unused for non-cardiac reasons served as controls. Tissue sections were stained with Masson's Trichrome and Hematoxylin and Eosin. Given the small cohort size and heterogeneity, analyses were descriptive. Continuous variables are reported as median (range).The d-TGA-AS population was older than the HLHS-F population (median: 42 years; range: 38-44 vs. median: 24 years; range: 12-32, respectively). RV fibrosis in the d-TGA-AS population was greater at 28% (7-35) vs. the HLHS-F population which was 4% (2-24) and donor controls (median: 2% range: 0-4). In contrast, RV wall thickness was greater in HLHS-F (median: 12,303 µm; range: 9,976-15,745) than in d-TGA-AS (median: 9,063 µm; range: 8,316-10,322) and donors (median 7,984 µm; range: 2,582-13,386). Donor LV thickness (median: 17,056 µm; range: 16,688-17,423) exceeded all RV groups.The primary histologic finding for the d-TGA-AS group was fibrosis, while the HLHS-F group showed predominantly hypertrophy. The temporal presentation of the patients was different, with the HLHS-F patients presenting earlier for transplant than the d-TGA-AS. These observations suggest that different histologic changes may occur in response to longstanding systemic pressures in these two anatomic subgroups of patients with systemic RV.

    View details for DOI 10.3389/fcvm.2025.1640561

    View details for PubMedID 41189996

    View details for PubMedCentralID PMC12580193

  • Harnessing the Power of ChatGPT in Cardiovascular Medicine: Innovations, Challenges, and Future Directions. Journal of clinical medicine Leon, M., Ruaengsri, C., Pelletier, G., Bethencourt, D., Shibata, M., Flores, M. Q., Shudo, Y. 2024; 13 (21)

    Abstract

    Cardiovascular diseases remain the leading cause of morbidity and mortality globally, posing significant challenges to public health. The rapid evolution of artificial intelligence (AI), particularly with large language models such as ChatGPT, has introduced transformative possibilities in cardiovascular medicine. This review examines ChatGPT's broad applications in enhancing clinical decision-making-covering symptom analysis, risk assessment, and differential diagnosis; advancing medical education for both healthcare professionals and patients; and supporting research and academic communication. Key challenges associated with ChatGPT, including potential inaccuracies, ethical considerations, data privacy concerns, and inherent biases, are discussed. Future directions emphasize improving training data quality, developing specialized models, refining AI technology, and establishing regulatory frameworks to enhance ChatGPT's clinical utility and mitigate associated risks. As cardiovascular medicine embraces AI, ChatGPT stands out as a powerful tool with substantial potential to improve therapeutic outcomes, elevate care quality, and advance research innovation. Fully understanding and harnessing this potential is essential for the future of cardiovascular health.

    View details for DOI 10.3390/jcm13216543

    View details for PubMedID 39518681

  • An Exploration of Myths, Barriers and Strategies for Improving Diversity Among STS Members. The Annals of thoracic surgery Backhus, L. M., Kpodonu, J. n., Romano, J. C., Pelletier, G. n., Preventza, O. n., Cooke, D. T. 2019

    Abstract

    Diversity within health care organizations has many proven benefits, yet women and other groups remain under-represented in cardiothoracic surgery. We sought to explore responses from a Society of Thoracic Surgeons (STS) survey to identify myths and barriers for informing organizational strategies in the STS and cardiothoracic surgery. We performed a qualitative review of narrative survey responses within three domains surrounding diversity in cardiothoracic surgery: Myths, Barriers, and Strategies for improvement. Common diversity myths included: diversity as a pipeline problem (24%), diversity equated to exclusivity (21%), and diversity not supporting meritocracy (18%). The most frequent barrier code was perceived prejudice (22%). Suggested strategies towards improvement were: culture change prioritizing diversity (22%) and training the leaders (14%). Notably, 15% of response codes reflected the belief that disparities do not exist thus the issue should not be prioritized by the organization. The results do not necessarily reflect the beliefs of the majority of STS membership, nonetheless they provide important insight critical to guide any efforts towards eliminating disparities within cardiothoracic surgery and improving the care of our patients.

    View details for DOI 10.1016/j.athoracsur.2019.09.007

    View details for PubMedID 31593654