MMMartha M.O. McGilvray

Martha M.O. McGilvray

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Martha M.O. McGilvray has a total of 57 co-authors. Here are the top co-authors by number of shared publications:

  • Christian Zemlin

    Division of Cardiothoracic Surgery, Washington University School of Medicine, Barnes-Jewish Hospital, Saint Louis, MO, USA

    4 shared publications
  • Ralph J. Damiano

    Department of Surgery, Division of Cardiothoracic Surgery (M.R.S., R.V., T.-A.Y., M.M.O.M., C.W.Z., R.B.S., R.J.D.), Washington University in St. Louis, MO

    4 shared publications
  • Amit Pawale

    Division of Cardiothoracic Surgery Washington University School of Medicine in St. Louis St. Louis Missouri USA

    2 shared publications
  • Brian P. Cupps

    Division of Cardiothoracic Surgery, Department of Surgery, School of Medicine, Washington University in St. Louis

    2 shared publications
  • Christian W. Zemlin

    Division of Cardiothoracic Surgery, Department of Surgery, Washington University School of Medicine, Barnes-Jewish Hospital, St Louis, Mo

    2 shared publications
  • Hailey M. Shepherd

    Surgery, Washington University School of Medicine, St. Louis, MO, USA

    2 shared publications

Latest Publications

  • Open Access

    Predicting Near-term Mortality in Heart Failure: External Validation of Electronic Health Record-Based Deep Learning Model

    by Martha M.O. McGilvray, Amit A. Pawale, Sophia R. Pyeatte, Hailey M. Shepherd, Adam Wilcox, Jeffrey T. Heaton, Michael K. Pasque

    Background The dire consequences of heart failure (HF) patient non-response to guideline directed medical therapy often fuel early, non-selective referral for surgical intervention (ventricular assist device [VAD] or transplant). The high-risk associated with these interventions mandates precision in directing them only toward those patients who would otherwise suffer severe near-term deterioration. We previously reported a 52,265-patient deep learning model that predicted 1-year severe decompensation/death in HF inpatients, with a C-statistic of 0.91. We now present external model validation. Few groups applying deep learning to large-scale datasets have achieved external validation using equally large-scale independent datasets, yet proof of generalization is essential to practical applicability. Methods Our previous study used standard electronic health record (EHR) data to build ensemble deep learning models employing time-series and densely connected networks. The positive-class included both all-cause mortality and referral for HF surgical intervention within 1 year. In the current study, we assessed generalization of model architecture in an external validation test set from the Veterans Cardiac Health and Artificial Intelligence Model Predictions (V-CHAMPS) challenge, a synthetic national governmental sample using a distinct EHR system. While V-CHAMPS is a robust dataset, variables that capture VAD/transplant referral were not readily extracted, limiting the positive-class to mortality only. Results A total of 380,441 distinct admissions from 75,086 HF patients contributed >720 million EHR datapoints. 23% of observations fit positive-class criteria. The model C-statistic in the external-validation cohort was 0.79. Conclusions Despite being developed in a single-center dataset with a more precise positive-class, our model architecture maintained relative accuracy when applied to a national sample in an unrelated EHR system. This supports clinical relevancy of the deep-learning model and adaptability with retraining to disparate contexts. This broad applicability suggests considerable potential of EHR-based deep learning models to assist HF clinicians in improving the usage of advanced surgical therapy.

  • Sinus Rhythm Atrial Electrocardiographic Imaging in Patients With Mitral Regurgitation: Clues to the Substrate for Atrial Fibrillation

    by Matthew R. Schill, Ramya Vijayakumar, Tari-Ann Yates, Martha M.O. McGilvray, Christian W. Zemlin, Richard B. Schuessler, Yoram Rudy, Ralph J. Damiano


  • Open Access

    Applying Deep Learning in Heart Failure: Hospital Readmission is Not Like Other Health Quality Metrics

    by Hailey M. Shepherd, Jeffrey T. Heaton, Theodore Marghitu, Yun Z. Bai, Melanie P. Subramanian, Sophia H. Roberts, Martha M.O. McGilvray, Amit A. Pawale, Gregory A. Ewald, Brian P. Cupps, et al.

    Background Early identification of heart failure patients at increased risk for near-term adverse outcomes would assist clinicians in efficient resource allocation and improved care. Deep learning can improve identification of these patients. Methods This retrospective study examined adult heart failure patients admitted to a tertiary care institution between January 2009 and December 2018. A deep learning model was constructed with a dense input layer, three long short-term memory (LSTM) layers, and a dense hidden layer to cohesively extract features from time-series and non-time-series EHR data. Primary outcomes were all-cause hospital readmission or death within 30 days after hospital discharge. Results Among a final subset of 49,675 heart failure patients, we identified 171,563 hospital admissions described by 330 million EHR data points. There were 22,111 (13%) admissions followed by adverse 30-day outcomes, including 19,122 readmissions (87%) and mortality in 3,330 patients (15%). Our final deep learning model achieved an area under the receiver-operator characteristic curve (AUC) of 0.613 and precision-recall (PR) AUC of 0.38. Conclusions This EHR-based deep learning model developed from a decade of heart failure care achieved marginal clinical accuracy in predicting very early hospital readmission or death despite previous accurate prediction of 1-year mortality in this large study cohort. These findings suggest that factors unavailable in standard EHR data play pivotal roles in influencing early hospital readmission. What is new?We developed an EHR-based deep learning model trained by 330 million data points from one of the largest cohorts of heart failure patients to date. Despite this model’s highly accurate prediction of long-term outcomes, such as mortality and disease progression, our findings suggest that EHR data alone offers limited predictive power for predicting the short-term outcomes of 30-day hospital readmission or death. What are the clinical implications?Our study supports the notion that hospital readmission, in contrast to other health outcomes, is uniquely driven by additional factors beyond traditional EHR variables. Once identified, incorporation of these determinants into future deep learning models could allow for accurate heart failure risk-stratification at hospital discharge to facilitate more efficient allocation of limited resources to the most vulnerable patients.

  • Minimally Invasive Mitral Valve Surgery With Concomitant Cox Maze Procedure Is as Effective as a Median Sternotomy With Decreased Morbidity

    by Tari-Ann Yates, Martha McGilvray, Connor Vinyard, Laurie Sinn, Nicholas Razo, June He, Harold G. Roberts, Matthew R. Schill, Christian Zemlin, Ralph J. Damiano

    Objective: A right minithoracotomy (RMT) is a minimally invasive surgical approach that has been increasingly performed for the concomitant Cox maze IV procedure (CMP) and mitral valve surgery (MVS). Little is known regarding whether long-term rhythm and survival outcomes are affected by the RMT as compared with the traditional median sternotomy (MS) approach. Methods: Between April 2004 and April 2021, 377 patients underwent the concomitant CMP and MVS, of whom 38% had RMT. Propensity score matching yielded 116 pairs. Freedom from atrial tachyarrhythmias (ATA) was assessed with prolonged monitoring annually for 8 years. Survival, rhythm, and perioperative outcomes were compared. Results: The unmatched RMT cohort had a greater freedom from ATA recurrence at 1 year (99% vs 90%, P = 0.001) and 3 years (94% vs 86%, P = 0.045). The matched RMT cohort had longer cardiopulmonary bypass (median: 215 [199 to 253] vs 170 [136 to 198] min, P < 0.001) and aortic cross-clamp (110 [98 to 124] vs 86 [71 to 102] min, P < 0.001) times but shorter intensive care time (48 [24 to 95] vs 71 [26 to 144] h, P = 0.001) and length of stay (8 [6 to 11] vs 10 [7 to 14] h, P < 0.001). More pacemakers (18% vs 4%, P < 0.001) and postoperative transfusions (57% vs 41%, P = 0.014) occurred in the MS cohort. The 30-day mortality ( P = 0.651) and 8-year survival ( P = 0.072) was not significantly different between the cohorts. Conclusions: Early 1-year and 3-year freedom from ATA recurrence was better in the RMT cohort compared with the MS cohort. Despite longer operative times, the RMT cohort had shorter lengths of stay, fewer postoperative transfusions, and fewer pacemakers placed.


  • Open Access

    An experimental model of chronic severe mitral regurgitation

    by Martha M.O. McGilvray, Tari-Ann E. Yates, Lynn Pauls, Meghan O. Kelly, Nicholas Razo, Stacie McElligott, Glenn J. Foster, Jie Zheng, Jonathan K. Zoller, Christian Zemlin, et al.


  • Open Access

    The Cox-Maze procedure: What lesions and why

    by Martha M.O. McGilvray, Lauren Barron, Tari-Ann E. Yates, Christian W. Zemlin, Ralph J. Damiano


  • Efficacy of a Novel Bipolar Radiofrequency Clamp: An Acute Porcine Model

    by Tari-Ann Yates, Martha McGilvray, Nick Razo, Stacie McElligott, Spencer J. Melby, Christian Zemlin, Ralph J. Damiano

    Objective: Expert consensus guidelines recommend surgical ablation (SA) for patients with symptomatic atrial fibrillation (AF), but less than half of patients with AF undergoing cardiac procedures receive concomitant SA. Complete isolation of the left atrial posterior wall (LAPW) has been shown to be the most critical part of the Cox maze procedure. The purpose of this study was to investigate the performance of a novel radiofrequency (RF) bipolar device, EnCompass™ (AtriCure, Inc., Mason, OH, USA), designed to isolate the LAPW in a single application. Methods: Five adult pigs underwent SA in a beating heart model. After a single ablation, the heart was arrested, explanted, and stained with triphenyl-tetrazolium-chloride for histological assessment. Each lesion was sectioned, and the ablation depth, muscle, and fat thickness were determined. The lesion width, energy delivery, and ablation times were compared with those from a reference RF clamp (Synergy™, AtriCure). Results: Transmurality was documented in 100% of lesions (5 of 5) and cross sections (160 of 160). Electrical isolation was documented in every instance. There was no evidence of clot, charring, or pulmonary vein stenosis. Compared with the reference clamp, the lesions created by the EnCompass™ clamp were 1.5 times wider on average. The average energy delivered was 5 times higher over a duration that was 4.5 times longer due to the increased volume of tissue ablated. Conclusions: The EnCompass™ clamp reproducibly created transmural isolation of the LAPW with a single application. This may allow for simplification of the SA strategy and increased adoption of AF treatment during concomitant surgery.


  • Electronic Health Record-Based Deep Learning Prediction of Death or Severe Decompensation in Heart Failure Patients

    by Martha M.O. McGilvray, Jeffrey Heaton, Aixia Guo, M. Faraz Masood, Brian P. Cupps, Marci Damiano, Michael K. Pasque, Randi Foraker


  • Efficacy of the stand‐alone Cox‐Maze IV procedure in patients with longstanding persistent atrial fibrillation

    by Martha M.O. McGilvray MSt, Nadia H. Bakir Bs, Meghan O. Kelly, Samuel C. Perez, Laurie A. Sinn, Richard B. Schuessler, Christian W. Zemlin, Hersh S. Maniar, Spencer J. Melby, Ralph J. Damiano

    Introduction: Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and results in significant morbidity and mortality. The Cox‐Maze IV procedure (CMP‐IV) has been shown to have excellent efficacy in returning patients to sinus rhythm, but there have been few reports of late follow‐up in sizable cohorts of patients with longstanding persistent AF, the most difficult type of AF to treat.Methods and Results: Between May 2003 and March 2020, 174 consecutive patients underwent a stand‐alone CMP‐IV for longstanding persistent AF. Rhythm outcome was assessed postoperatively for up to 10 years, primarily via prolonged monitoring (Holter monitor, pacemaker interrogation, or implantable loop recorder). Fine‐Gray regression was used to investigate factors associated with atrial tachyarrhythmia (ATA) recurrence, with death as a competing risk. Median duration of preoperative AF was 7.8 years (interquartile range: 4.0–12.0 years), with 71% (124/174) having failed at least one prior catheter‐based ablation. There were no 30‐day mortalities. Freedom from ATAs was 94% (120/128), 83% (53/64), and 88% (35/40) at 1, 5, and 7 years, respectively. On regression analysis, preoperative AF duration and early postoperative ATAs were associated with late ATAs recurrence.Conclusion: Despite the majority of patients having a long‐duration of preoperative AF and having failed at least one catheter‐based ablation, the stand‐alone CMP‐IV had excellent late efficacy in patients with longstanding persistent AF, with low morbidity and no mortality. We recommend consideration of stand‐alone CMP‐IV for patients with longstanding persistent AF who have failed or are poor candidates for catheter ablation.


  • Open Access

    Efficacy of the Stand-Alone Cox-Maze IV Procedure in Patients with Longstanding Persistent Atrial Fibrillation

    by Martha McGilvray, Nadia Bakir, Meghan Kelly, Samuel Perez, Laurie Sinn, Richard Schuessler, Christian Zemlin, Hersch Maniar, Spencer Lelby, Ralph Damiano

    Introduction: Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and results in significant morbidity and mortality. The Cox-Maze IV procedure (CMP-IV) has been shown to have excellent efficacy in returning patients to sinus rhythm, but there have been few reports of late follow-up in sizable cohorts of patients with longstanding persistent AF, the most difficult type of AF to treat. Methods & Results: Between May 2003 and March 2020, 174 consecutive patients underwent a stand-alone CMP-IV for longstanding persistent AF. Rhythm outcome was assessed postoperatively for up to 10 years, primarily via prolonged monitoring (Holter monitor, pacemaker interrogation, or implantable loop recorder). Fine-Gray regression was used to investigate factors associated with atrial tachyarrhythmia (ATA) recurrence, with death as a competing risk. Median duration of preoperative AF was 7.8 years (interquartile range [IQR] 4.0-12.0 years), with 71% (124/174) having failed at least one prior catheter-based ablation. There were no 30-day mortalities. Freedom from ATAs was 94% (120/128), 83% (53/64), and 88% (35/40) at 1, 5, and 7 years, respectively. On regression analysis, preoperative AF duration and early postoperative ATAs were associated with late ATAs recurrence. Conclusion: Despite the majority of patients having a long-duration of preoperative AF and having failed at least one catheter-based ablation, the stand-alone CMP-IV had excellent late efficacy in patients with longstanding persistent AF, with low morbidity and no mortality. We recommend consideration of stand-alone CMP-IV for patients with longstanding persistent AF who have failed or are poor candidates for catheter ablation.