RATIONALE:Machine learning may be useful to characterize cardiovascular risk, predict outcomes, and identify biomarkers in population studies.
OBJECTIVE:To test the ability of random survival ...forests, a machine learning technique, to predict 6 cardiovascular outcomes in comparison to standard cardiovascular risk scores.
METHODS AND RESULTS:We included participants from the MESA (Multi-Ethnic Study of Atherosclerosis). Baseline measurements were used to predict cardiovascular outcomes over 12 years of follow-up. MESA was designed to study progression of subclinical disease to cardiovascular events where participants were initially free of cardiovascular disease. All 6814 participants from MESA, aged 45 to 84 years, from 4 ethnicities, and 6 centers across the United States were included. Seven-hundred thirty-five variables from imaging and noninvasive tests, questionnaires, and biomarker panels were obtained. We used the random survival forests technique to identify the top-20 predictors of each outcome. Imaging, electrocardiography, and serum biomarkers featured heavily on the top-20 lists as opposed to traditional cardiovascular risk factors. Age was the most important predictor for all-cause mortality. Fasting glucose levels and carotid ultrasonography measures were important predictors of stroke. Coronary Artery Calcium score was the most important predictor of coronary heart disease and all atherosclerotic cardiovascular disease combined outcomes. Left ventricular structure and function and cardiac troponin-T were among the top predictors for incident heart failure. Creatinine, age, and ankle-brachial index were among the top predictors of atrial fibrillation. TNF-α (tissue necrosis factor-α) and IL (interleukin)-2 soluble receptors and NT-proBNP (N-Terminal Pro-B-Type Natriuretic Peptide) levels were important across all outcomes. The random survival forests technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 10%–25%).
CONCLUSIONS:Machine learning in conjunction with deep phenotyping improves prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. These methods may lead to greater insights on subclinical disease markers without apriori assumptions of causality.
CLINICAL TRIAL REGISTRATION:URLhttp://www.clinicaltrials.gov. Unique identifierNCT00005487.
Introduction
Vasodilator stress cardiovascular magnetic resonance (CMR) is a powerful diagnostic modality, but data toward its use in patients with permanent pacemakers (PPMs) or implantable ...cardioverter‐defibrillators (ICDs) is limited.
Methods and Results
Patients with ICDs (>1% pacing) or PPMs who underwent regadenoson single photon emission computed tomography (SPECT) and all patients with ICDs or PPMs who underwent stress CMR were retrospectively identified. SPECT tests were analyzed for hemodynamic responses and new pacing requirements; CMR studies were examined for safety, device characteristics and programming, hemodynamic responses, and image quality. Changes from baseline were evaluated with the Related‐Samples Wilcoxon Signed Rank Test. Of 67 patients (median age 65 IQR 58–72 years, 31 46% female, 31 46% Black), 47 underwent SPECT and 20 CMR. With regadenoson SPECT, 89% of patients experienced tachycardic responses above resting heart rates (+19 13–32 beats per minute, p < .01). During stress CMR, 10 (50%) devices were asynchronously paced approximately 10 beats per minute above resting rates, and the remaining were temporarily deactivated. Those with asynchronous pacing had no changes in heart rates, whereas patients with deactivated devices had near uniform heart rate accelerations. Image quality was diagnostic in the majority of stress CMR sequences, with nonconditional ICDs contributing 40 of 57 (70%) of nondiagnostic segments.
Conclusion
This data supports the safety of vasodilator stress CMR with promising diagnostic quality images in patients with CMR conditional ICDs and PPMs. Despite a near uniform tachycardic response to regadenoson in the SPECT environment, high rates of asynchronous pacing during vasodilator stress CMR did not result in competitive pacing or adverse arrhythmic events. Further studies are needed to validate these findings and confirm the diagnostic and prognostic performance of stress CMR in these individuals.
Heart rate responses and cardiovascular magnetic resonance (CMR) image quality of vasodilator stress testing in patients with pacemakers or defibrillators. In a cohort of patients with pacemakers or defibrillators who underwent vasodilator stress testing, there was a near uniform tachycardic response to regadenoson agent in the nuclear environment (where device reprogramming is not required). In the CMR environment, heart rate response was either uniformly tachycardic (in those with temporary device deactivation) or unchanged (in those temporarily set to asynchronous pacing). Imaging quality was largely diagnostic in the CMR‐conditional devices, whereas nonconditional implantable cardioverter‐defibrillators typically had extensive artifact obscuring several myocardial segments from the relevant imaging sequences.
Patients with or receiving potentially cardiotoxic treatment for cancer are susceptible to developing decrements in left ventricular mass, diastolic function, or systolic function. They may also ...experience valvular heart disease, pericardial disease, or intracardiac masses. Cardiovascular magnetic resonance may be used to assess cardiac anatomy, structure, and function and to characterize myocardial tissue. This combination of features facilitates the diagnosis and management of disease processes in patients with or those who have survived cancer. This report outlines and describes prior research involving cardiovascular magnetic resonance for assessing cardiovascular disease in patients with or previously having received treatment for cancer.
Display omitted
The study sought to evaluate the effects of endurance exercise training (ET) on endothelial-dependent flow-mediated arterial dilation (FMD) and carotid artery stiffness, and their potential ...contributions to the training-related increase in peak exercise oxygen consumption (Vo2) in older patients with heart failure with preserved ejection fraction (HFPEF).
Elderly HFPEF patients have severely reduced peak Vo2, which improves with ET, however, the mechanisms of this improvement are unclear. FMD and arterial distensibility are critical components of the exercise response and are reduced with aging. However, it is unknown whether these improve with ET in elderly HFPEF or contribute to the training-related improvement in peak Vo2.
A total of 63 HFPEF patients (age 70 ± 7 years) were randomized to 16 weeks of ET (walking, arm and leg ergometry, n = 32) or attention control (CT) (n = 31). Peak Vo2, brachial artery FMD in response to cuff ischemia, carotid artery distensibility by high-resolution ultrasound, left ventricular function, and quality of life were measured at baseline and follow-up.
ET increased peak Vo2 (ET: 15.8 ± 3.3 ml/kg/min vs. CT: 13.8 ± 3.1 ml/kg/min, p = 0.0001) and quality of life. However, brachial artery FMD (ET: 3.8 ± 3.0% vs. CT: 4.3 ± 3.5%, p = 0.88), and carotid arterial distensibility (ET: 0.97 ± 0.56 vs. CT: 1.07 ± 0.34 × 10(-3) mm·mm Hg(-2); p = 0.65) were unchanged. Resting left ventricular systolic and diastolic function were unchanged by ET.
In elderly HFPEF patients, 16 weeks of ET improved peak Vo2 without altering endothelial function or arterial stiffness. This suggests that other mechanisms, such as enhanced skeletal muscle perfusion and/or oxygen utilization, may be responsible for the ET-mediated increase in peak Vo2 in older HFPEF patients. (Prospective Aerobic Reconditioning Intervention Study PARIS; NCT01113840).
...CMR has low intra- and interobserver variability and high test-retest reproducibility for measurement of LV function (3,4), characteristics that are crucial in clinical situations requiring ...accurate serial monitoring of LV function that occurs in cancer patients receiving potentially cardiotoxic therapy for cancer.
Purpose of Review
Although cancer treatments have increased overall survival rates, the cardiovascular consequences of cancer therapy place patients at an increased risk of adverse outcomes. This ...manuscript presents data accumulated to date regarding cardiovascular outcomes relating to the administration of 3-hydroxy-3-methylglutarylcoenzyme-A reductase inhibitor (or statin) therapy in individuals receiving potentially cardiotoxic cancer treatments.
Recent Findings
Retrospective observational studies in humans and randomized controlled trials in animals suggest that statins may reduce cancer-specific and all-cause mortality. Further, statins may attenuate cancer therapy-induced declines in left ventricular ejection fraction (LVEF) and increases in blood pressure.
Summary
Observational studies suggest a potential attenuation in LVEF decline in patients with cancer and primary or secondary indications to receive a statin for prevention of cardiovascular events. Large randomized clinical trials are warranted to understand the efficacy and potential impacts of statin class, dosage, and duration on cardiovascular outcomes in patients treated for cancer.
Purpose of Review
This review aims to summarize recent developments in identifying and quantifying both the presence and amount of myocardial fibrosis by imaging and biomarkers. Further, this review ...seeks to describe in general ways how this information may be used to identify hypertension and the transition to heart failure with preserved ejection fraction.
Recent Findings
Recent studies using cardiac magnetic resonance imaging highlight the progressive nature of fibrosis from normal individuals to those with hypertension to those with clinical heart failure. However, separating hypertensive patients from those with heart failure remains challenging. Recent studies involving echocardiography show the subclinical myocardial strain changes between hypertensive heart disease and heart failure. Lastly, recent studies highlight the potential use of biomarkers to identify those with hypertension at the greatest risk of developing heart failure.
Summary
In light of the heterogeneous nature between hypertension and heart failure with preserved ejection fraction, an integrated approach with cardiac imaging and biomarker analysis may enable clinicians and investigators to more accurately characterize, prevent, and treat heart failure in those with hypertension.
Cardiovascular magnetic resonance T1 mapping characteristics are elevated in adult cancer survivors; however, it remains unknown whether these elevations are related to age or presence of coincident ...cardiovascular comorbidities.
We performed blinded cardiovascular magnetic resonance analyses of left ventricular T1 and extracellular volume (ECV) fraction in 327 individuals (65% women, aged 64±12 years). Thirty-seven individuals had breast cancer or a hematologic malignancy but had not yet initiated their treatment, and 54 cancer survivors who received either anthracycline-based (n=37) or nonanthracycline-based (n=17) chemotherapy 2.8±1.3 years earlier were compared with 236 cancer-free participants. Multivariable analyses were performed to determine the association between T1/ECV measures and variables associated with myocardial fibrosis. Age-adjusted native T1 was elevated pre- (1058±7 ms) and post- (1040±7 ms) receipt of anthracycline chemotherapy versus comparators (965±3 ms; P<0.0001 for both). Age-adjusted ECV, a marker of myocardial fibrosis, was elevated in anthracycline-treated cancer participants (30.4±0.7%) compared with either pretreatment cancer (27.8±0.7%; P<0.01) or cancer-free comparators (26.9±0.2%; P<0.0001). T1 and ECV of nonanthracycline survivors were no different than pretreatment survivors (P=0.17 and P=0.16, respectively). Native T1 and ECV remained elevated in cancer survivors after accounting for demographics (including age), myocardial fibrosis risk factors, and left ventricular ejection fraction or myocardial mass index (P<0.0001 for all).
Three years after anthracycline-based chemotherapy, elevations in myocardial T1 and ECV occur independent of underlying cancer or cardiovascular comorbidities, suggesting that imaging biomarkers of interstitial fibrosis in cancer survivors are related to prior receipt of a potentially cardiotoxic cancer treatment regimen.