Artificial intelligence to enhance clinical value across the spectrum of cardiovascular healthcare.

Artificial intelligence (AI) is increasingly being utilized in healthcare. This article provides clinicians and researchers with a step-wise foundation for high-value AI that can be applied to a variety of different data modalities. The aim is to improve the transparency and application of AI methods, with the potential to benefit patients in routine cardiovascular care. Following a clear research hypothesis, an AI-based workflow begins with data selection and pre-processing prior to analysis, with the type of data (structured, semi-structured, or unstructured) determining what type of pre-processing steps and machine-learning algorithms are required. Algorithmic and data validation should be performed to ensure the robustness of the chosen methodology, followed by an objective evaluation of performance. Seven case studies are provided to highlight the wide variety of data modalities and clinical questions that can benefit from modern AI techniques, with a focus on applying them to cardiovascular disease management. Despite the growing use of AI, further education for healthcare workers, researchers, and the public are needed to aid understanding of how AI works and to close the existing gap in knowledge. In addition, issues regarding data access, sharing, and security must be addressed to ensure full engagement by patients and the public. The application of AI within healthcare provides an opportunity for clinicians to deliver a more personalized approach to medical care by accounting for confounders, interactions, and the rising prevalence of multi-morbidity.

Identification and Mapping Real-World Data Sources for Heart Failure, Acute Coronary Syndrome, and Atrial Fibrillation.

BACKGROUND: Transparent and robust real-world evidence sources are increasingly important for global health, including cardiovascular (CV) diseases. We aimed to identify global real-world data (RWD) sources for heart failure (HF), acute coronary syndrome (ACS), and atrial fibrillation (AF). METHODS: We conducted a systematic review of publications with RWD pertaining to HF, ACS, and AF (2010-2018), generating a list of unique data sources. Metadata were extracted based on the source type (e.g., electronic health records, genomics, and clinical data), study design, population size, clinical characteristics, follow-up duration, outcomes, and assessment of data availability for future studies and linkage. RESULTS: Overall, 11,889 publications were retrieved for HF, 10,729 for ACS, and 6,262 for AF. From these, 322 (HF), 287 (ACS), and 220 (AF) data sources were selected for detailed review. The majority of data sources had near complete data on demographic variables (HF: 94%, ACS: 99%, and AF: 100%) and considerable data on comorbidities (HF: 77%, ACS: 93%, and AF: 97%). The least reported data categories were drug codes (HF, ACS, and AF: 10%) and caregiver involvement (HF: 6%, ACS: 1%, and AF: 1%). Only a minority of data sources provided information on access to data for other researchers (11%) or whether data could be linked to other data sources to maximize clinical impact (20%). The list and metadata for the RWD sources are publicly available at www.escardio.org/bigdata. CONCLUSIONS: This review has created a comprehensive resource of CV data sources, providing new avenues to improve future real-world research and to achieve better patient outcomes.

Redefining ß-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis.

BACKGROUND: Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of ß-blocker efficacy in patients with sinus rhythm and atrial fibrillation. METHODS: Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of ß blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). FINDINGS: 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from ß blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of ß blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with ß blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p<0·0001 vs random), and cluster membership was externally validated across the nine independent trials. INTERPRETATION: An artificial intelligence-based clustering approach was able to distinguish prognostic response from ß blockers in patients with heart failure and reduced LVEF. This included patients in sinus rhythm with suboptimal efficacy, as well as a cluster of patients with atrial fibrillation where ß blockers did reduce mortality. FUNDING: Medical Research Council, UK, and EU/EFPIA Innovative Medicines Initiative BigData@Heart.

Improving the diagnosis of heart failure in patients with atrial fibrillation.

OBJECTIVE: To improve the echocardiographic assessment of heart failure in patients with atrial fibrillation (AF) by comparing conventional averaging of consecutive beats with an index-beat approach, whereby measurements are taken after two cycles with similar R-R interval. METHODS: Transthoracic echocardiography was performed using a standardised and blinded protocol in patients enrolled in the RATE-AF (RAte control Therapy Evaluation in permanent Atrial Fibrillation) randomised trial. We compared reproducibility of the index-beat and conventional consecutive-beat methods to calculate left ventricular ejection fraction (LVEF), global longitudinal strain (GLS) and E/e' (mitral E wave max/average diastolic tissue Doppler velocity), and assessed intraoperator/interoperator variability, time efficiency and validity against natriuretic peptides. RESULTS: 160 patients were included, 46% of whom were women, with a median age of 75 years (IQR 69-82) and a median heart rate of 100 beats per minute (IQR 86-112). The index-beat had the lowest within-beat coefficient of variation for LVEF (32%, vs 51% for 5 consecutive beats and 53% for 10 consecutive beats), GLS (26%, vs 43% and 42%) and E/e' (25%, vs 41% and 41%). Intraoperator (n=50) and interoperator (n=18) reproducibility were both superior for index-beats and this method was quicker to perform (p<0.001): 35.4 s to measure E/e' (95% CI 33.1 to 37.8) compared with 44.7 s for 5-beat (95% CI 41.8 to 47.5) and 98.1 s for 10-beat (95% CI 91.7 to 104.4) analyses. Using a single index-beat did not compromise the association of LVEF, GLS or E/e' with natriuretic peptide levels. CONCLUSIONS: Compared with averaging of multiple beats in patients with AF, the index-beat approach improves reproducibility and saves time without a negative impact on validity, potentially improving the diagnosis and classification of heart failure in patients with AF.

Effect of Digoxin vs Bisoprolol for Heart Rate Control in Atrial Fibrillation on Patient-Reported Quality of Life: The RATE-AF Randomized Clinical Trial.

Importance: There is little evidence to support selection of heart rate control therapy in patients with permanent atrial fibrillation, in particular those with coexisting heart failure. Objective: To compare low-dose digoxin with bisoprolol (a ß-blocker). Design, Setting, and Participants: Randomized, open-label, blinded end-point clinical trial including 160 patients aged 60 years or older with permanent atrial fibrillation (defined as no plan to restore sinus rhythm) and dyspnea classified as New York Heart Association class II or higher. Patients were recruited from 3 hospitals and primary care practices in England from 2016 through 2018; last follow-up occurred in October 2019. Interventions: Digoxin (n = 80; dose range, 62.5-250 µg/d; mean dose, 161 µg/d) or bisoprolol (n = 80; dose range, 1.25-15 mg/d; mean dose, 3.2 mg/d). Main Outcomes and Measures: The primary end point was patient-reported quality of life using the 36-Item Short Form Health Survey physical component summary score (SF-36 PCS) at 6 months (higher scores are better; range, 0-100), with a minimal clinically important difference of 0.5 SD. There were 17 secondary end points (including resting heart rate, modified European Heart Rhythm Association [EHRA] symptom classification, and N-terminal pro-brain natriuretic peptide [NT-proBNP] level) at 6 months, 20 end points at 12 months, and adverse event (AE) reporting. Results: Among 160 patients (mean age, 76 [SD, 8] years; 74 [46%] women; mean baseline heart rate, 100/min [SD, 18/min]), 145 (91%) completed the trial and 150 (94%) were included in the analysis for the primary outcome. There was no significant difference in the primary outcome of normalized SF-36 PCS at 6 months (mean, 31.9 [SD, 11.7] for digoxin vs 29.7 [11.4] for bisoprolol; adjusted mean difference, 1.4 [95% CI, -1.1 to 3.8]; P = .28). Of the 17 secondary outcomes at 6 months, there were no significant between-group differences for 16 outcomes, including resting heart rate (a mean of 76.9/min [SD, 12.1/min] with digoxin vs a mean of 74.8/min [SD, 11.6/min] with bisoprolol; difference, 1.5/min [95% CI, -2.0 to 5.1/min]; P = .40). The modified EHRA class was significantly different between groups at 6 months; 53% of patients in the digoxin group reported a 2-class improvement vs 9% of patients in the bisoprolol group (adjusted odds ratio, 10.3 [95% CI, 4.0 to 26.6]; P < .001). At 12 months, 8 of 20 outcomes were significantly different (all favoring digoxin), with a median NT-proBNP level of 960 pg/mL (interquartile range, 626 to 1531 pg/mL) in the digoxin group vs 1250 pg/mL (interquartile range, 847 to 1890 pg/mL) in the bisoprolol group (ratio of geometric means, 0.77 [95% CI, 0.64 to 0.92]; P = .005). Adverse events were less common with digoxin; 20 patients (25%) in the digoxin group had at least 1 AE vs 51 patients (64%) in the bisoprolol group (P < .001). There were 29 treatment-related AEs and 16 serious AEs in the digoxin group vs 142 and 37, respectively, in the bisoprolol group. Conclusions and Relevance: Among patients with permanent atrial fibrillation and symptoms of heart failure treated with low-dose digoxin or bisoprolol, there was no statistically significant difference in quality of life at 6 months. These findings support potentially basing decisions about treatment on other end points. Trial Registration: ClinicalTrials.gov Identifier: NCT02391337 and clinicaltrialsregister.eu Identifier: 2015-005043-13.

Impact of Renal Impairment on Beta-Blocker Efficacy in Patients With Heart Failure.

BACKGROUND: Moderate and moderately severe renal impairment are common in patients with heart failure and reduced ejection fraction, but whether beta-blockers are effective is unclear, leading to underuse of life-saving therapy. OBJECTIVES: This study sought to investigate patient prognosis and the efficacy of beta-blockers according to renal function using estimated glomerular filtration rate (eGFR). METHODS: Analysis of 16,740 individual patients with left ventricular ejection fraction <50% from 10 double-blind, placebo-controlled trials was performed. The authors report all-cause mortality on an intention-to-treat basis, adjusted for baseline covariates and stratified by heart rhythm. RESULTS: Median eGFR at baseline was 63 (interquartile range: 50 to 77) ml/min/1.73 m2; 4,584 patients (27.4%) had eGFR 45 to 59 ml/min/1.73 m2, and 2,286 (13.7%) 30 to 44 ml/min/1.73 m2. Over a median follow-up of 1.3 years, eGFR was independently associated with mortality, with a 12% higher risk of death for every 10 ml/min/1.73 m2 lower eGFR (95% confidence interval [CI]: 10% to 15%; p < 0.001). In 13,861 patients in sinus rhythm, beta-blockers reduced mortality versus placebo; adjusted hazard ratio (HR): 0.73 for eGFR 45 to 59 ml/min/1.73 m2 (95% CI: 0.62 to 0.86; p < 0.001) and 0.71 for eGFR 30 to 44 ml/min/1.73 m2 (95% CI: 0.58 to 0.87; p = 0.001). The authors observed no deterioration in renal function over time in patients with moderate or moderately severe renal impairment, no difference in adverse events comparing beta-blockers with placebo, and higher mortality in patients with worsening renal function on follow-up. Due to exclusion criteria, there were insufficient patients with severe renal dysfunction (eGFR <30 ml/min/1.73 m2) to draw conclusions. In 2,879 patients with atrial fibrillation, there was no reduction in mortality with beta-blockers at any level of eGFR. CONCLUSIONS: Patients with heart failure, left ventricular ejection fraction <50% and sinus rhythm should receive beta-blocker therapy even with moderate or moderately severe renal dysfunction.