About

Wilson Y Szeto, M.D., is the Julian Johnson Professor of Cardiothoracic Surgery II at the University of Pennsylvania. He is an active surgeon at Pennsylvania Hospital, Penn Presbyterian Medical Center, and the Hospital of the University of Pennsylvania in Philadelphia, PA. Dr. Szeto serves as the Surgical Director of Transcatheter Cardio-Aortic Therapies within the University of Pennsylvania Health System, and holds leadership roles including Co-Director of Penn Presbyterian Medical Center, Chief of Cardiovascular Surgery at Penn Presbyterian Medical Center, Vice Chief of Clinical Operations and Quality at the University of Pennsylvania, and Chief of the Division of Cardiac Surgery at the University of Pennsylvania Health System. His department is the Department of Surgery, specifically within the Division of Cardiac Surgery. Dr. Szeto's educational background includes a B.A. in Biology from the University of Virginia (1992) and an M.D. from the Medical College of Virginia/Virginia Commonwealth University (1996). His postgraduate training encompasses internships and residencies in General Surgery and Cardiothoracic Surgery at the University of Pennsylvania Medical Center, along with a Post-doctoral Research Fellowship in Surgical Research at the University of Pennsylvania School of Medicine. He is certified by the American Board of Surgery (recertified 2013-2024) and the American Board of Thoracic Surgery (recertified 2014-2026). His research and clinical focus are centered on cardiothoracic surgery, with particular expertise in transcatheter therapies and aortic surgery.

Research topics

• Medicine
• Internal medicine
• Surgery
• Cardiology
• Computer Science
• Anesthesia

Selected publications

• Outcomes of Hemiarch Versus Zone 2 Arch Replacement for DeBakey Type I Aortic Dissection

  European Journal of Cardio-Thoracic Surgery · 2026-01-01

  article

  OBJECTIVES: The objective of this study was to compare perioperative and long-term outcomes of hemiarch versus zone 2 arch replacement for DeBakey type I acute aortic dissection. METHODS: From 2002 to 2023, 743 patients underwent surgical repair for DeBakey type I acute aortic dissection with either hemiarch (N = 605, 81.4%) or zone 2 arch replacement (N = 138, 18.6%). 4:1 propensity-score matching yielded a cohort of 437 hemiarch patients (76.1%) and 137 zone 2 patients (23.9%). Key outcomes included 30-day mortality, 10-year survival, and distal reintervention. RESULTS: In the matched cohort, 30-day mortality was 11.9% for hemiarch and 8.0% for zone 2 arch replacement (P = .198). On multivariable analysis, zone 2 arch replacement was not associated with increased risk of 30-day mortality. Ten-year survival was similar between groups, at 56.2% (50.6%-62.6%) in the hemiarch group and 59.8% (46.4%-77.1%) in the zone 2 group. Distal reintervention was more common after zone 2 arch replacement (46.7% vs 15.8%), though the majority of reinterventions in the zone 2 group were thoracic endovascular aortic repair procedures (95.3%). In contrast, over half (53.6%) of distal reinterventions after hemiarch repair were open procedures. CONCLUSIONS: Zone 2 arch replacement resulted in comparable early and late survival to hemiarch replacement. Although distal reinterventions were more common after zone 2 arch replacement, these were predominantly low-risk thoracic endovascular aortic repair procedures. In appropriately selected patients, zone 2 arch replacement may enable more complete aortic repair without added perioperative risk.

  PublisherDOI

• Surgical aortic valve replacement with a novel bovine pericardial tissue valve: a real world comparison of short-term costs and outcomes in US hospitals

  Figshare · 2026-01-01

  articleOpen access

  Use of bioprosthetic tissue valves in surgical aortic valve replacement (SAVR) has increased in recent years. The effect of novel bovine pericardial tissue valves on short-term costs and utilization has not been established. We assessed INSPIRIS RESILIA aortic tissue valves (*Edwards Lifesciences, Irvine CA) during SAVR hospitalization and 90 days post-discharge compared to other tissue valves for all ages, and to mechanical valves for patients of shared decision-making age (50–65). Retrospective observational study using US hospital data. Adults admitted for first-time SAVR 2018–2021 with identifiable tissue or mechanical valves were analyzed. Outcomes were adjusted for demographics, comorbidities, procedure and hospital characteristics using generalized linear modeling. 20,125 patients were analyzed; majority were male, with elective procedures in academic hospitals. Compared to other tissue valves, length of stay and ICU stay were shorter with INSPIRIS (<i>p</i> &lt; 0.001). In-hospital mortality was not statistically different. Readmissions at 30 days were 7.1% with INSPIRIS, 9.0% other tissue valves; 10.4% vs. 13.0% at 90 days (both <i>p</i> &lt; 0.001). Reduced cost of readmission (-$1,013 at 90 days) partially offset the difference of $2,429 in initial hospitalization cost. INSPIRIS valves compared to mechanical valves (ages 50–65) showed similar patterns in utilization. In-hospital mortality was lower with INSPIRIS than mechanical valves (2.0% vs. 3.2%, <i>p</i> = 0.009), but did not differ when limited to isolated SAVR only. Initial hospitalization cost was higher for INSPIRIS ($7,079, <i>p</i> &lt; 0.001), with lower 90-day readmissions cost (-$899, <i>p</i> &lt; 0.05). Non-randomized analysis of real-world data. Patients undergoing SAVR with INSPIRIS valves had shorter length of stay and fewer readmissions compared to other tissue valves and to mechanical valves. Reduced cost of readmissions partially offset higher cost of initial SAVR admission. Further research is warranted to explore this association between valve used and utilization outcomes. This study looked at adults in the United States who had surgery to replace their diseased or worn-out aortic heart valve. Researchers compared outcomes for patients who received a newer type of valve named INSPIRIS RESILIA made from treated cow tissue to those who received other tissue valves or mechanical valves. Researchers wanted to know whether the INSPIRIS valve benefited patients in the short term within 90 days after surgery. The study used hospital billing data from a representative sample covering about a quarter of US hospitals. Researchers used advanced statistical methods and adjusted for differences in patient health, surgeon experience and hospital characteristics to ensure robust comparisons. However, some factors could not be measured due to limitations in the data. Patients who received the INSPIRIS valve generally discharged home, spent less time in the hospital and in the intensive care unit (ICU) compared to those with other tissue valves. They also had fewer hospital readmissions within 30 and 90 days after surgery. The study also looked at patients aged 50 to 65, a group where either tissue or mechanical valves may be used. Compared to mechanical valves, patients with INSPIRIS valves had lower in-hospital death rates, shorter ICU and hospital stays, and fewer hospital readmissions within 30 and 90 days. Overall, the INSPIRIS valve was associated with shorter hospital stays and fewer readmissions, which may benefit patients. The researchers propose that these observations should be studied further to see if these benefits are directly related to the INSPIRIS valve.

  PublisherDOI

• Sixty-Second Annual Meeting, The Society of Thoracic Surgeons

  The Annals of Thoracic Surgery · 2026-04-23

  article1st authorCorresponding
  PublisherDOI

• Fate of the Aortic Root After Reconstruction With Felt Neomedia for Acute Type A Aortic Dissection

  European Journal of Cardio-Thoracic Surgery · 2026-01-01

  article

  OBJECTIVES: We sought to characterize the durability of aortic root reconstruction with felt neomedia for acute type A aortic dissection (ATAAD). METHODS: Of 894 patients undergoing type A aortic dissection repair between 2010 and 2024 at a single institution, 537 underwent aortic root reconstruction with felt neomedia and aortic valve resuspension (median age 64 years, 37.2% female). Median follow-up was 4.3 years. Outcomes included survival, proximal aortic reintervention, greater than moderate aortic insufficiency, and root dilatation. Predictors of a composite outcome-defined as proximal aortic reintervention, greater than moderate aortic insufficiency, or root dilatation ≥ 50 mm-were evaluated. RESULTS: Preoperatively, 19.9% of patients had greater than moderate aortic insufficiency, and median root diameter was 39.0 mm. Thirty-day mortality was 12.8%, and 10-year survival was 52.3%. The cumulative incidence of proximal aortic reintervention at 1, 5, and 10 years was 0.4%, 2.4%, and 5.5%. During follow-up, 13 patients developed greater than moderate aortic insufficiency and 16 developed root dilatation ≥ 50 mm. Freedom from the composite outcome was 98.8% at 1 year and 91.9% at 5 years. On multivariable analysis, preoperative aortic insufficiency and baseline root diameter ≥ 45 mm were independent predictors of the composite outcome. CONCLUSIONS: Aortic root reconstruction with felt neomedia and aortic valve resuspension is a durable technique for ATAAD. Patients with preoperative aortic insufficiency or root dilatation are at increased risk for root-related events and warrant close surveillance.

  PublisherDOI

• Surgical aortic valve replacement with a novel bovine pericardial tissue valve: a real world comparison of short-term costs and outcomes in US hospitals

  Figshare · 2026-01-01

  articleOpen access

  Use of bioprosthetic tissue valves in surgical aortic valve replacement (SAVR) has increased in recent years. The effect of novel bovine pericardial tissue valves on short-term costs and utilization has not been established. We assessed INSPIRIS RESILIA aortic tissue valves (*Edwards Lifesciences, Irvine CA) during SAVR hospitalization and 90 days post-discharge compared to other tissue valves for all ages, and to mechanical valves for patients of shared decision-making age (50–65). Retrospective observational study using US hospital data. Adults admitted for first-time SAVR 2018–2021 with identifiable tissue or mechanical valves were analyzed. Outcomes were adjusted for demographics, comorbidities, procedure and hospital characteristics using generalized linear modeling. 20,125 patients were analyzed; majority were male, with elective procedures in academic hospitals. Compared to other tissue valves, length of stay and ICU stay were shorter with INSPIRIS (<i>p</i> &lt; 0.001). In-hospital mortality was not statistically different. Readmissions at 30 days were 7.1% with INSPIRIS, 9.0% other tissue valves; 10.4% vs. 13.0% at 90 days (both <i>p</i> &lt; 0.001). Reduced cost of readmission (-$1,013 at 90 days) partially offset the difference of $2,429 in initial hospitalization cost. INSPIRIS valves compared to mechanical valves (ages 50–65) showed similar patterns in utilization. In-hospital mortality was lower with INSPIRIS than mechanical valves (2.0% vs. 3.2%, <i>p</i> = 0.009), but did not differ when limited to isolated SAVR only. Initial hospitalization cost was higher for INSPIRIS ($7,079, <i>p</i> &lt; 0.001), with lower 90-day readmissions cost (-$899, <i>p</i> &lt; 0.05). Non-randomized analysis of real-world data. Patients undergoing SAVR with INSPIRIS valves had shorter length of stay and fewer readmissions compared to other tissue valves and to mechanical valves. Reduced cost of readmissions partially offset higher cost of initial SAVR admission. Further research is warranted to explore this association between valve used and utilization outcomes. This study looked at adults in the United States who had surgery to replace their diseased or worn-out aortic heart valve. Researchers compared outcomes for patients who received a newer type of valve named INSPIRIS RESILIA made from treated cow tissue to those who received other tissue valves or mechanical valves. Researchers wanted to know whether the INSPIRIS valve benefited patients in the short term within 90 days after surgery. The study used hospital billing data from a representative sample covering about a quarter of US hospitals. Researchers used advanced statistical methods and adjusted for differences in patient health, surgeon experience and hospital characteristics to ensure robust comparisons. However, some factors could not be measured due to limitations in the data. Patients who received the INSPIRIS valve generally discharged home, spent less time in the hospital and in the intensive care unit (ICU) compared to those with other tissue valves. They also had fewer hospital readmissions within 30 and 90 days after surgery. The study also looked at patients aged 50 to 65, a group where either tissue or mechanical valves may be used. Compared to mechanical valves, patients with INSPIRIS valves had lower in-hospital death rates, shorter ICU and hospital stays, and fewer hospital readmissions within 30 and 90 days. Overall, the INSPIRIS valve was associated with shorter hospital stays and fewer readmissions, which may benefit patients. The researchers propose that these observations should be studied further to see if these benefits are directly related to the INSPIRIS valve.

  PublisherDOI

• Impact of supra-aortic vessel false lumen communications on aortic remodeling: Analysis from the PERSEVERE trial

  JTCVS structural and endovascular. · 2025-10-09

  articleOpen accessSenior author

  Objective:The PERSEVERE (ProspEctive, Single ARm, Multi-center Clinical InveStigation to EValuatE the Safety and Effectiveness of AMDS in the TREatment of Acute DeBakey Type I Dissection) Trial is a US investigational device exemption study designed to evaluate the Ascyrus Medical Dissection hybrid prosthesis in the open repair of patients with acute DeBakey type I dissection with preoperative malperfusion.We sought to determine the impact of false lumen (FL) communications within the supra-aortic vessels (SAV) on aortic remodeling by analyzing total aortic diameter (TAD), FL diameter, and true lumen (TL) diameter.Methods: Ninety-three patients were enrolled from July 2022 to November 2023.Anatomic measurements were made by a core laboratory using computed tomography angiogram (CTA).For this study, analysis was limited to zones 2, 3, and 4. TAD was assessed on first postoperative CTA and at 1-year postoperatively, and growth was defined as >1.0 cm.In addition, the percent of patients with FL diameter larger than TL diameter were compared on the basis of the presence or absence of any FL communications in the SAV.Results: Fifty-five patients had a preoperative, first postoperative CTA, and 1-year CTA scan and were included in the analysis.SAV FL communications were present on preoperative computed tomography scan in 35% (19/55) and absent in 65% (36/ 55).In those with SAV FL communications, TAD growth occurred in 0%, 5.3%, and 21.1% in zones 2, 3, and 4, respectively.In those without SAV FL communications, TAD growth occurred in 2.8%, 2.8%, and 8.3% in zones 2, 3 and 4, respectively; the difference did not reach statistical significance.The FL was larger than TL in significantly more patients in zones 3 and 4 in those with compared with those without SAV FL communications (zone 3, 53% vs 18%; zone 4, 58% vs 18%; P < .05). Conclusions:The presence of preoperative SAV FL communications has a significant impact on aortic remodeling at 1-year postoperatively.The presence of these communications is associated with a trend toward negative remodeling in zones 2, 3, and 4, in addition to a significantly larger FL compared with TL diameter, likely representing ongoing FL pressurization.Techniques to better identify SAV FL communications preoperatively may lead to a more tailored approach to operative management and more targeted postoperative surveillance in these patients.(JTCVS Structural and Endovascular 2025;8:100081) Innominate artery false lumen communication as demonstrated on preoperative CT.

  PublisherOA PDFDOI

• Arterial Cannulation and Brain Perfusion Strategies for Acute Type A Dissection in North America

  The Annals of Thoracic Surgery · 2025-11-10 · 3 citations

  article
  PublisherDOI

• One-Year Results of Novel Aortic Arch Hybrid Prosthesis for Repair of Acute DeBakey Type I Dissection With Malperfusion: PERSEVERE Study

  The Annals of Thoracic Surgery · 2025-11-16 · 1 citations

  article1st authorCorresponding
  PublisherDOI

• Cardiac valve prosthesis choice for left-sided infective endocarditis in Medicare patients

  JTCVS Open · 2025-10-25

  articleOpen access

  Objective: To compare the outcomes between bioprosthetic and mechanical prosthesis choice for aortic valve and mitral valve infective endocarditis (IE) in patients >65 years old. Methods: Medicare Provider Analysis and Review files from 2011 to 2019 were queried for adult patients >65 years old with de novo IE who underwent isolated bioprosthetic aortic valve replacement (bAVR), mechanical aortic valve replacement (mAVR), bioprosthetic mitral valve replacement (bMVR), or mechanical mitral valve replacement (mMVR). Patients with preoperative ischemic or hemorrhagic stroke were excluded. Analyses were conducted separately for the aortic and mitral position: bAVR versus mAVR, and bMVR versus mMVR. Propensity score matching was used to account for measured confounders. The primary outcome was 5-year overall survival analyzed using restricted mean survival time; secondary outcomes were the 5-year cumulative incidences of valve reoperation, heart failure readmission, recurrent IE, bleeding, and ischemic stroke analyzed using Fine-Gray regression with death as a competing risk. Results: = .451). There were no significant differences in the cumulative incidences of secondary outcomes. Conclusions: This analysis of patients >65 years old with left-sided IE demonstrated that mechanical valves were associated with worse early mortality, no 5-year survival advantage over bioprosthetic valves, and provide no benefit in this population.

  PublisherDOI

• Abstract 4367379: Early Re-Initiation of Optimal Guideline-Directed Medical Therapy after Coronary Artery Bypass Grafting in Patients with Heart Failure with Reduced Ejection is Associated with Decreased Mortality

  Circulation · 2025-11-03

  article

  Introduction: Guideline-directed medical therapy (GDMT) improves survival in medical patients with heart failure with reduced ejection fraction (HFrEF). For cardiac surgical patients with HFrEF, GDMT is typically held perioperatively; re-introduction may be hampered by post-operative frailty and vasoplegia. This study sought to examine patterns of GDMT prescription after coronary artery bypass grafting (CABG) in patients with HFrEF and baseline optimal medical therapy, and the potential benefit of return to optimal dosing. Methods: The population included Medicare beneficiaries who underwent isolated CABG from 2016-2019, with a HFrEF diagnosis in the prior 6 months, on baseline optimal GDMT as characterized by the Heart Failure Collaboratory score, which standardizes across medication classes and doses. Prescription patterns were evaluated across patients who did and did not achieve optimal GDMT by 90 days post-discharge (defined as a score of ≥4 with both beta-blocker and renin-angiotensin-aldosterone inhibitor (RAASi) prescription). Via propensity score matching, a Cox proportional hazard model was applied to compare all-cause mortality and a composite of mortality and HF readmission between the optimal and non-optimal cohorts. Results: Of the 634 patients meeting inclusion criteria, 247 (39.0%) were re-initiated on optimal GDMT within 90 days of discharge. Mineralocorticoid receptor antagonists (MRAs) represented the class with the lowest uptake in both the non-optimal and optimal cohorts (17.1% and 70.0%) (Figure 1). Angiotensin receptor-neprilysin inhibitor (ARNI) use was scarce, representing 33.6% of RAASi in the optimal cohort and &lt;5% in the non-optimal cohort. Cohorts were similar at baseline, although optimal patients were more likely to have had surgery at larger hospitals (Table 1). After matching, the optimal group demonstrated significantly decreased hazard of mortality (8.1% vs. 17.0%; HR 0.45 [0.25-0.76], p=0.002) and composite of mortality and HF readmission at 2 years (19.0% vs. 30.4%; HR 0.57 [0.40-0.82], p=0.003) (Figure 2). Conclusion: Patients with HFrEF on optimal GDMT at baseline who undergo CABG and are re-initiated on an optimal regimen within 90 days of discharge had significantly decreased risk of mortality and heart failure readmission. These findings highlight a need to identify barriers to GDMT after cardiac surgery such that these patients may benefit long-term from the combined effects of surgical and medical optimization.

  PublisherDOI

Frequent coauthors

• Susheel Kodali

  Columbia University Irving Medical Center

  713 shared

• Martin B. Leon

  Columbia University Irving Medical Center

  712 shared

• Vinod H. Thourani

  Valve (United States)

  629 shared

• Mathew Williams

  NYU Langone Health

  525 shared

• Rebecca T. Hahn

  Cardiovascular Research Foundation

  489 shared

• Vasilis Babaliaros

  Morristown Medical Center

  436 shared

• Joseph E. Bavaria

  University of Pennsylvania

  397 shared

• Samir Kapadia

  Cleveland Clinic

  377 shared

Education

• B.A., Biology

  University of Virginia

  1992

• M.D.

  Medical College of Virginia/ Virginia Commonwealth University

  1996

Awards & honors

• American Board of Surgery (Recertification 2013-2024)
• American Board of Thoracic Surgery (Recertification 2014-202…