Incremental value of machine learning for risk prediction in tetralogy of Fallot.

OBJECTIVE: Machine learning (ML) can facilitate prediction of major adverse cardiovascular events (MACEs) in repaired tetralogy of Fallot (rTOF). We sought to determine the incremental value of ML above expert clinical judgement for risk prediction in rTOF. METHODS: Adult congenital heart disease (ACHD) clinicians (≥10 years of experience) participated (one cardiac surgeon and four cardiologists (two paediatric and two adult cardiology trained) with expertise in heart failure (HF), electrophysiology, imaging and intervention). Clinicians identified 10 high-yield variables for 5-year MACE prediction (defined as a composite of mortality, resuscitated sudden death, sustained ventricular tachycardia and HF). Risk for MACE (low, moderate or high) was assigned by clinicians blinded to outcome for adults with rTOF identified from an institutional database (n=25 patient reviews conducted by five independent observers). A validated ML model identified 10 variables for risk prediction in the same population. RESULTS: Prediction by ML was similar to the aggregate score of all experts (area under the curve (AUC) 0.85 (95% CI 0.58 to 0.96) vs 0.92 (0.72 to 0.98), p=0.315). Experts with ≥20 years of experience had superior discriminative capacity compared with <20 years (AUC 0.98 (95% CI 0.86 to 0.99) vs 0.80 (0.56 to 0.93), p=0.027). In those with <20 years of experience, ML provided incremental value such that the combined (clinical+ML) AUC approached ≥20 years (AUC 0.85 (95% CI 0.61 to 0.95), p=0.055). CONCLUSIONS: Robust prediction of 5-year MACE in rTOF was achieved using either ML or a multidisciplinary team of ACHD experts. Risk prediction of some clinicians was enhanced by incorporation of ML suggesting that there may be incremental value for ML in select circumstances.

Machine Learning for Prediction of Adverse Cardiovascular Events in Adults With Repaired Tetralogy of Fallot Using Clinical and Cardiovascular Magnetic Resonance Imaging Variables.

BACKGROUND: Existing models for prediction of major adverse cardiovascular events (MACE) after repair of tetralogy of Fallot have been limited by modest predictive capacity and limited applicability to routine clinical practice. We hypothesized that an artificial intelligence model using an array of parameters would enhance 5-year MACE prediction in adults with repaired tetralogy of Fallot. METHODS: A machine learning algorithm was applied to 2 nonoverlapping, institutional databases of adults with repaired tetralogy of Fallot: (1) for model development, a prospectively constructed clinical and cardiovascular magnetic resonance registry; (2) for model validation, a retrospective database comprised of variables extracted from the electronic health record. The MACE composite outcome included mortality, resuscitated sudden death, sustained ventricular tachycardia and heart failure. Analysis was restricted to individuals with MACE or followed ≥5 years. A random forest model was trained using machine learning (n=57 variables). Repeated random sub-sampling validation was sequentially applied to the development dataset followed by application to the validation dataset. RESULTS: We identified 804 individuals (n=312 for development and n=492 for validation). Model prediction (area under the curve [95% CI]) for MACE in the validation dataset was strong (0.82 [0.74-0.89]) with superior performance to a conventional Cox multivariable model (0.63 [0.51-0.75]; P=0.003). Model performance did not change significantly with input restricted to the 10 strongest features (decreasing order of strength: right ventricular end-systolic volume indexed, right ventricular ejection fraction, age at cardiovascular magnetic resonance imaging, age at repair, absolute ventilatory anaerobic threshold, right ventricular end-diastolic volume indexed, ventilatory anaerobic threshold % predicted, peak aerobic capacity, left ventricular ejection fraction, and pulmonary regurgitation fraction; 0.81 [0.72-0.89]; P=0.232). Removing exercise parameters resulted in inferior model performance (0.75 [0.65-0.84]; P=0.002). CONCLUSIONS: In this single-center study, a machine learning-based prediction model comprised of readily available clinical and cardiovascular magnetic resonance imaging variables performed well in an independent validation cohort. Further study will determine the value of this model for risk stratification in adults with repared tetralogy of Fallot.

Novel mediators of aneurysm progression in bicuspid aortic valve disease.

Bicuspid aortic valve (BAV) disease is a congenital abnormality that is associated with ascending aortic aneurysm yet many of the molecular mechanisms remain unknown. To identify novel molecular mechanisms of aneurysm formation we completed microarray analysis of the proximal (severely dilated) and distal (less dilated) regions of the ascending aorta from five patients with BAV. We identified 180 differentially expressed genes, 40 of which were validated by RT-qPCR. Most genes had roles in inflammation and endothelial cell function including cytokines and growth factors, cell surface receptors and the Activator Protein 1 (AP-1) transcription factor family (FOS, FOSB and JUN) which was chosen for further study. AP-1 was differentially expressed within paired BAV aneurysmal samples (n = 8) but not Marfan patients (n = 5). FOS protein was significantly enriched in BAV aortas compared to normal aortas but unexpectedly, ERK1/2 activity, an upstream regulator of FOS was reduced. ERK1/2 activity was restored when BAV smooth muscle cells were cultured in vitro. An mRNA-miRNA network within paired patient samples identified AP-1 as a central hub of miRNA regulation. FOS knockdown in BAV SMCs increased expression of miR-27a, a stretch responsive miRNA. AP-1 and miR-27a were also dysregulated in a mouse model of aortic constriction. In summary, this study identified a central role for AP-1 signaling in BAV aortic dilatation by using paired mRNA-miRNA patient sample. Upstream analysis of AP-1 regulation showed that the ERK1/2 signaling pathway is dysregulated and thus represents a novel chain of mediators of aortic dilatation in BAV which should be considered in future studies.

Clinical utility of a screening protocol for blunt cerebrovascular injury using computed tomography angiography.

OBJECTIVE Blunt cerebrovascular injury (BCVI) occurs in approximately 1% of the blunt trauma population and may lead to stroke and death. Early vascular imaging in asymptomatic patients at high risk of having BCVI may lead to earlier diagnosis and possible stroke prevention. The objective of this study was to determine if the implementation of a formalized asymptomatic BCVI screening protocol with CT angiography (CTA) would lead to improved BCVI detection and stroke prevention. METHODS Patients with vascular imaging studies were identified from a prospective trauma registry at a single Level 1 trauma center between 2002 and 2008. Detection of BCVI and stroke rates were compared during the 3-year periods before and after implementation of a consensus-based asymptomatic BCVI screening protocol using CTA in 2005. RESULTS A total of 5480 patients with trauma were identified. The overall BCVI detection rate remained unchanged postprotocol compared with preprotocol (0.8% [24 of 3049 patients] vs 0.9% [23 of 2431 patients]; p = 0.53). However, postprotocol there was a trend toward a decreased risk of stroke secondary to BCVI on a trauma population basis (0.23% [7 of 3049 patients] vs 0.53% [13 of 2431 patients]; p = 0.06). Overall, 75% (35 of 47) of patients with BCVI were treated with antiplatelet agents, but no patient developed new or progressive intracranial hemorrhage despite 70% of these patients having concomitant traumatic brain injury. CONCLUSIONS The results of this study suggest that a CTA screening protocol for BCVI may be of clinical benefit with possible reduction in ischemic complications. The treatment of BCVI with antiplatelet agents appears to be safe.

History of Cardiovascular Surgery at Toronto General Hospital.

The Division of Cardiovascular Surgery at Toronto General Hospital has enjoyed an enviable history of academic achievement and clinical success. The foundations of this success are innovation, creativity and excellence in patient care, which continue to influence the current members of the division.

Dual antiplatelet therapy use by Canadian cardiac surgeons.

BACKGROUND: Dual antiplatelet therapy is the cornerstone treatment for patients with acute coronary syndrome. Recent Canadian Guidelines recommend the use of dual antiplatelet therapy for 1 year after coronary artery bypass grafting in patients with acute coronary syndrome, but considerable variability remains. METHODS: We performed a survey of 75 Canadian cardiac surgeons to assess the use of dual antiplatelet therapy. RESULTS: Whereas 58.6% of respondents indicated that the benefits of dual antiplatelet therapy were seen irrespective of how patients were managed after acute coronary syndrome, 36.2% believed that the benefits of dual antiplatelet therapy were limited to those treated medically or percutaneously. In regard to the timing of dual antiplatelet therapy administration, 57% of respondents indicated that dual antiplatelet therapy should be given upstream in the emergency department, whereas 36.2% responded that dual antiplatelet therapy should be given only once the coronary anatomy has been defined. The majority surveyed (81%) weighed bleeding risk as being more important than ischemic risk reduction. In stable patients after acute coronary syndrome, the majority of surgeons would wait approximately 4 days after the last dose of P2Y12 antagonist before coronary artery bypass grafting. Only 44.6% indicated that they routinely use dual antiplatelet therapy postrevascularization in the setting of acute coronary syndrome. Rather, most surgeons use dual antiplatelet therapy for select patients, such as those with a stented vessel without a bypass graft, endarterectomy, or off-pump coronary artery bypass grafting. CONCLUSIONS: Cardiac surgeons exhibit variation in their attitudes and practice patterns toward dual antiplatelet therapy after coronary artery bypass grafting, and in approximately half of cases, their practice does not adhere to current guideline recommendations. New trials focusing on coronary artery bypass grafting cases in their primary analysis and educational initiatives for surgeons that focus on guideline recommendations may be warranted.