MFMichael E. Farkouh

Michael E. Farkouh

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Michael E. Farkouh has a total of 410 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Machine Learning With Genetic and Clinical Data to Predict Ischemic Outcomes After PCI

    by Caroline W. Grant, Brenden S. Ingraham, Ryan J. Lennon, Anvi Raina, Shulan Tian, Gurukripa Kowlgi, Larisa H. Cavallari, Craig R. Lee, Amber Beitelshees, Dominick J. Angiolillo, et al.

    Ischemic events after contemporary percutaneous coronary intervention (PCI) are uncommon but carry high morbidity/mortality. Identifying patients at highest risk is critical to guide dual antiplatelet therapy (DAPT) intensity while minimizing bleeding. Current machine learning (ML) model‐developed risk scores do not incorporate pharmacogenetic data. To develop and externally validate ML models that integrate clinical, demographic, and CYP2C19 genetic information to predict 1‐year ischemic outcomes post‐PCI to refine clinical decision making. We analyzed 8317 patients from TAILOR‐PCI trial ( n = 4572) and the Precision PCI registry ( n = 3745). The outcome was a composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis at 1 year. Boruta feature selection identified 11 predictors. Multiple ML algorithms were trained in TAILOR‐PCI and validated in Precision PCI using cross‐validation and synthetic minority oversampling (SMOTE). Model performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The best external performance was achieved with a support vector machine (SVM, polynomial kernel) (AUC 0.667; sensitivity 0.871; specificity 0.282), while XGBoost provided a more balanced profile (AUC 0.619; sensitivity 0.442; specificity 0.688). Variable importance for the SVM polynomial model demonstrated that all 11 included Boruta feature selected predictors had relatively high importance (> 75). ML models trained on large clinical trial and real‐world registry datasets can help identify the small subset of patients at high ischemic risk after PCI. This distinction is of relevance because ischemic events are rare, and most patients may safely de‐escalate DAPT to reduce bleeding risk while maintaining ischemic protection through a multimodal approach to risk stratification. Trial Registration: TAILOR‐PCI URL: https://clinicaltrials.gov/ct2/show/NCT01742117

  • Open Access

    Reassessing alcohol consumption and cardiovascular disease by addressing bias in observational data: results from the multi-ethnic study of atherosclerosis

    by Shantanu Srivatsa, Elizabeth K Farkouh, Tim Stockwell, James R Pike, James M Clay, Ganga S Bey, Khurram Nasir, Matthew Budoff, Timothy Naimi, Michael E Farkouh

    Aims: Alcohol and its relationship with cardiovascular disease (CVD) remain contentious. Several observational studies have found cardioprotective associations, while Mendelian randomization studies have found null or harmful effects of moderate drinking. Biases, including abstainer misclassification and sick-quitter effects, may obscure true associations. It is widely understood that former drinkers are often sick quitters. However, removal of former drinkers from analyses addresses bias in the abstainer group but fails to account for induced survivorship among current drinkers. This study sought to determine whether adjusting for biases altered the association between alcohol and CVD by comparing conventional models with those using (i) occasional drinkers as the reference group instead of lifetime abstainers and (ii) ‘intention-to-treat’ reallocation of former drinkers based on past drinking patterns. Methods and results: The Multi-Ethnic Study of Atherosclerosis is a prospective cohort study of 6814 participants aged 45–84 years between 2000 and 2002 without baseline CVD. A total of 6755 participants were included after excluding those with missing data. Alcohol consumption was self-reported and classified as lifetime abstainer, former, or current drinker [occasional (≤1 drink/week), light (2–7 drinks/week), moderate (8–14 drinks/week), heavy (>14 drinks/week)]. Outcomes included coronary artery calcium (CAC; 0, 1–100, 101–300, and >300) and clinical events [myocardial infarction (MI), stroke, cardiovascular mortality, composite major adverse cardiovascular events (MACEs), and heart failure]. Of 6755 participants, 20.6% were lifetime abstainers, 24.1% former drinkers, and 55.3% current drinkers. Adjusting for biases altered the observed relationship between alcohol and CVD, showing increased harm with light drinking, decreased protection with moderate drinking, and a dose–response relationship between alcohol and CAC, particularly CAC >300 in adjusted models. Protective associations between moderate drinking and MI [hazard ratio (HR): 0.64, 95% confidence interval (CI): 0.38–1.06] and cardiovascular mortality (HR: 0.81, 95% CI: 0.53–1.25) were attenuated and became non-significant. Light drinking shifted from null to increased risk for MI (HR: 1.46, 95% CI: 1.11–1.92), stroke (HR: 1.54, 95% CI: 1.17–2.03), MACE (HR: 1.33, 95% CI: 1.12–1.57), and heart failure (HR: 1.42, 95% CI: 1.10–1.84). Conclusion: Associations between alcohol and CVD in observational studies are sensitive to methodologic variation. Results from models correcting for abstainer/sick-quitter biases may help explain discrepant results from previous studies.


  • Patterns and Outcomes of Completeness of Revascularization in Patients With Diabetes and Non–ST-Segment–Elevation Myocardial Infarction in Ontario, Canada

    by Lucas C. Godoy, Michael E. Farkouh, Peter C. Austin, Baiju R. Shah, Feng Qiu, Maneesh Sud, José C. Nicolau, Louai Razzouk, Stephen E. Fremes, Patrick R. Lawler, et al.

    BACKGROUND: It remains unknown how often, and to which extent, patients with diabetes and non–ST-segment–elevation myocardial infarction are revascularized. This study aimed to identify practice patterns and clinical outcomes in patients with diabetes and non–ST-segment–elevation myocardial infarction according to completeness of revascularization. METHODS: All patients with diabetes and multivessel disease hospitalized for non–ST-segment–elevation myocardial infarction in Ontario, Canada, between April 2009 and March 2020 and undergoing coronary angiography were included. Patients with previous coronary artery bypass grafting (CABG) at any time, percutaneous coronary intervention (PCI) in the previous 90 days, or an ST-segment–elevation myocardial infarction in the previous 90 days were excluded, as were patients with hemodynamic instability at hospital admission. Patients were classified into 4 groups, from the most to the least complete revascularization strategy: CABG, complete revascularization with PCI, incomplete PCI, or no revascularization. The primary outcome was all-cause death at 5 years. Cox regression adjusted for multiple baseline characteristics was used to compare outcomes between groups. RESULTS: We included 14 511 patients (mean age: 68.7±11.5 years; 69.6% male patients); 4525 (31.2%) were treated with CABG, 3008 (20.7%) with complete PCI, 3624 (25.0%) with incomplete PCI, and 3354 (23.1%) did not receive any revascularization procedure. Adjusted 5-year risks of all-cause death after CABG, complete PCI, incomplete PCI, and no revascularization were progressively increased: 25.9%, 29.8%, 32.2%, and 39.4% respectively. Compared with no revascularization, a 46% reduction in all-cause death was observed after CABG (hazard ratio, 0.54 [95% CI, 0.50–0.58]), 35% after complete PCI (hazard ratio, 0.65 [95% CI, 0.60–0.71]), and 27% after incomplete PCI (hazard ratio, 0.73 [95% CI, 0.68–0.79]), over a median follow-up of 5.8 years (interquartile range, 3.3–8.6). CONCLUSIONS: Almost a quarter of the patients with diabetes and non–ST-segment–elevation myocardial infarction did not receive any revascularization procedure. There was an incremental mortality reduction associated with CABG or complete revascularization with PCI, compared with incomplete revascularization with PCI or no revascularization.

  • Open Access

    Cognitive Outcomes in Randomized Controlled Trials of Coronary Artery Bypass Graft Surgery From 2005 to 2025: A Systematic Review

    by Shantanu Srivatsa, Sajeevan Sujanthan, Sydni Paleczny, David A. Ripsman, Stephanie Niño de Rivera, Arnaldo Dimagli, Foad Taghdiri, Pai‐Shan Cheng, Wendy Lou, Michael E. Farkouh, et al.

    Background: Cognitive decline after coronary artery bypass graft (CABG) is common and affects morbidity, mortality, and quality of life. We systematically reviewed randomized CABG trial control arms to characterize cognitive assessments, testing frequency, attrition, and ability to detect perioperative change. Methods: We searched MEDLINE, Embase, Cochrane Library, and PsycINFO for randomized controlled trials of CABG surgery that included at least one arm of patients solely undergoing CABG and that reported at least one objective cognitive assessment, from January 2005 to February 2025. Trials with mixed cardiac surgery or only subjective measures were excluded. We summarized task frequency, cognitive domains, and attrition. For tasks assessed preoperatively and postoperatively in ≥3 trials, we reported control group means and SDs. Risk of bias was assessed using the Cochrane Risk of Bias tool among 6 bias domains. This study was supported by NIH–R01NS123639. Results: Of 3494 screened studies, 2284 were CABG trials, and only 71 (3.1%) reported cognitive evaluation. These involved 15 925 patients (79% men; mean age, 64.2 years; median follow‐up, 90 days) and used 145 unique cognitive tasks, with the Trail Making Test Part B (40 of 71; 56.3%) and Part A (38 of 71; 53.5%) being the most frequently administered. Among 7 tasks with sufficient data, none detected preoperative to postoperative changes. Attrition rates averaged 18.9%, with a broad range of 0 to 62%. Conclusions: Cognitive assessment is uncommon in CABG trials, and commonly used tests rarely detect change. Heterogeneity precluded meta‐analysis, and high attrition raises concerns about selection and survivorship bias. To evaluate cognitive impact after CABG, trials need standardized, sensitive assessment strategies resilient to attrition and feasible for broad deployment.


  • The limitations of biomarkers in addressing bias in alcohol and cardiovascular disease research

    by Elizabeth K Farkouh, Shantanu Srivatsa, Michael E Farkouh

    This commentary refers to ‘Urinary tartaric acid as a biomarker of wine consumption and cardiovascular risk: the PREDIMED trial’, by I. Domínguez-López et al., https://doi.org/10.1093/eurheartj/ehae804.


  • Open Access

    Outcome Prediction After Tetralogy of Fallot Repair: A Prospective Clinical and Cardiovascular Magnetic Resonance Study

    by Rachel M. Wald, George Tomlinson, Christopher A. Caldarone, Nagib Dahdah, Frederic Dallaire, Christian Drolet, Michael E. Farkouh, Jasmine Grewal, Camille Hancock‐Friesen, Edward J. Hickey, et al.

    Background: Identification of individuals at risk for major adverse cardiovascular events is essential for contemporary management of patients with repaired tetralogy of Fallot. We sought to identify clinical and cardiovascular magnetic resonance imaging (CMR) predictors of adverse clinical outcomes in repaired tetralogy of Fallot. Methods: Children and adults prospectively enrolled in the CORRELATE (Comprehensive Outcomes Registry Late After Tetralogy of Fallot Repair) registry followed in North American, European, and Asian centers were studied. All patients had at least moderate pulmonary regurgitation and CMR at enrollment. Time‐to‐event analyses were performed from CMR completion to primary outcome, defined as mortality, resuscitated sudden death, sustained ventricular arrhythmia, or heart failure admission. Principal component analysis was used to create distinct CMR scores that collectively captured 80% of the variance among 10 CMR measures (systolic function, biventricular volumes/mass, and biatrial areas). Results: In 720 patients (55% male, median age 30.3±14 years, 78% adult) with mean follow‐up 5.7±1.8 years, the primary outcome occurred in 38 patients (5.2%) at a rate of 0.9/100 patient‐years. A well‐calibrated risk scoring system was created for prediction of the primary outcome at 5 years based on 5 predictors: age, diabetes, right ventricular systolic pressure, and 2 CMR principal component scores (predominantly reflecting atrial areas in the first principal component score and ventricular volumes in the second principal component score) (c‐statistic for the composite risk score 0.79 [95% Cl, 0.71–0.88]). Conclusions: Clinical and imaging characteristics can contribute to risk prediction in repaired tetralogy of Fallot. Further study will be required to evaluate the utility of a risk scoring system for identification of individuals who may benefit from enhanced surveillance, intensified medical therapy, and/or optimally timed intervention.


  • Open Access

    Body Mass Index, Comorbidities, and Ambulatory Care Visits: The REGARDS Study

    by Emily B. Levitan, Aowen Zhu, Vera Bittner, Todd M. Brown, Michael E. Farkouh, Mariam Girguis, Lei Huang, Elizabeth A. Jackson, Suzanne E. Judd, Shannon Rhodes, et al.

    Background: Individuals with obesity have more ambulatory care usage than individuals with normal weight and overweight. There is limited information on whether this is consistent across provider specialties and whether comorbidities explain the associations. Methods and Results: Among REGARDS (Reasons for Geographic and Racial Differences in Stroke) cohort study participants with Medicare fee‐for‐service coverage (n=9648), we identified ambulatory visits over 5 years. We used marginalized zero‐inflated Poisson models to calculate ratios of means by body mass index (BMI), adjusted for demographics and health behaviors, and inverse odds weighting to evaluate mediation by diabetes, hypertension, dyslipidemia, stroke, coronary heart disease, atrial fibrillation, heart failure, chronic kidney disease, depressive symptoms, cancer, arthritis, and sleep apnea. The mean age of participants was 71.7±7.3 years, 35.1% were Black individuals and 64.9% White individuals, and 51.1% were women. Participants had a mean of 37.8 total, 16.0 primary care, 3.4 cardiology, 1.9 orthopedics, 0.9 pulmonology, and 0.4 endocrinology visits. Compared with individuals with BMI 18.5 to 2 (n=2613), participants with BMI ≥35 kg/m 2 (n=1259) had 23% (95% CI, 21%–24%) more ambulatory visits. Participants with BMI ≥35 kg/m 2 had 26% more primary care, 20% more cardiology, 74% more orthopedics, 62% more pulmonology, and 85% more endocrinology visits. Comorbidities partly explained associations with overall, primary care, and orthopedics visits (39%, 38%, and 15%, respectively) and largely explained associations with cardiology, pulmonology, and endocrinology visits. Conclusions: Understanding which specialty visits are associated with higher BMI can help with workforce planning and allocation of resources.


  • Open Access

    Efficacy of Automated Digital Counseling for Chronic Heart Failure (ODYSSEE-vCHAT-CHF): A Randomized Controlled Pilot Trial (Preprint)

    by Janice Montbriand, Heather Ross, Chris Chan, Valeria E Rac, Ella Huszti, Joanne Arcand, Jilliane Code, George Tomlinson, Juan Duero Posada, Andrew Gentlin, et al.

    BACKGROUND Chronic Heart Failure (CHF) is an increasing health burden associated with psychological distress, decreased quality of life, and high one year hospitalization rates (40%). CHF patients benefit from appropriate self-care and there is an increasing need for automated scalable programs to motivate and educate these patients. OBJECTIVE A Virtual Community Promoting Health Literacy, Self-care and Peer Support for Heart Failure (vCHAT-CHF Trial) used an automated scalable digital program: prOmoting health with DigitallY based counSeling of Self-care bEhavior and quality of life (ODYSSEE) to improve adherence and self-care in CHF patients, thereby improving health and health-related quality of life vs. enhanced usual care (eUC). The primary outcome was a composite index of all-cause hospitalization and emergency room (ER) visits. Secondary outcomes included changes in self-reported physical and psychosocial well-being. METHODS The vCHAT-CHF Trial was a 2-armed single-blind randomized control pilot trial. This was an evidence-based online multimedia digital health platform, based on cognitive behavioral and motivational interviewing techniques. ODYSSEE-vCHAT employed weekly topics on an interactive digital health platform, as well as weekly webcast presentations on various self-care themes and moderated online chat rooms, in addition to usual care. RESULTS 61 CHF patients recruited from three tertiary hospitals in Canada were enrolled in the program for a median of 7 months. 44 (72%) of participants completed the endpoint assessment questionnaires (n= 19 (63%) in ODYSSEE-vChat vs. 25 (81%) in enhanced Usual Care (eUC)). Primary and secondary outcomes were investigated through linear and/or logistic multivariate models as appropriate, comparing program groups (e.g. ODYSSEE-vCHAT vs. eUC) by both intention-to-treat and treatment-received methods. ODYSSEE-vCHAT was not associated with a composite of unplanned hospitalizations and/or ER visits in the current trial (23% usual care vs. 23% ODYSSEE, P = .78). However, multivariate logistic regression demonstrated that individuals in ODYSSEE-vCHAT (vs. eUC) were more likely (P = .003, OR = 6.0) to show clinically relevant change in the Kansas City Cardiomyopathy Questionnaire (KCCQ) as defined by a change score ≥ 5 from baseline to end of trial. Multivariate linear regression showed that ODYSSEE-vCHATwas also associated with positive change in Self-Efficacy for Managing Chronic Disease (SEMCD6; P = .03, B = 4.38). CONCLUSIONS In summary, the vCHAT-CHF pilot trial showed that ODYSSEE-vCHATwas associated with clinically relevant change on the KCCQ and an increase in self-efficacy in management of chronic disease vs eUC, but failed to show a difference in number of hospitalizations and unplanned ER visits. This trial illustrated the potential of evidence-based automated digital health programs in self-management of CHF. Future research with a larger sample size may illuminate further relationships. CLINICALTRIAL ClinicalTrials.gov NCT04966104.