Phenogrouping heart failure with preserved or mildly reduced ejection fraction using electronic health record data.

BACKGROUND: Heart failure (HF) with preserved or mildly reduced ejection fraction includes a heterogenous group of patients. Reclassification into distinct phenogroups to enable targeted interventions is a priority. This study aimed to identify distinct phenogroups, and compare phenogroup characteristics and outcomes, from electronic health record data. METHODS: 2,187 patients admitted to five UK hospitals with a diagnosis of HF and a left ventricular ejection fraction ≥ 40% were identified from the NIHR Health Informatics Collaborative database. Partition-based, model-based, and density-based machine learning clustering techniques were applied. Cox Proportional Hazards and Fine-Gray competing risks models were used to compare outcomes (all-cause mortality and hospitalisation for HF) across phenogroups. RESULTS: Three phenogroups were identified: (1) Younger, predominantly female patients with high prevalence of cardiometabolic and coronary disease; (2) More frail patients, with higher rates of lung disease and atrial fibrillation; (3) Patients characterised by systemic inflammation and high rates of diabetes and renal dysfunction. Survival profiles were distinct, with an increasing risk of all-cause mortality from phenogroups 1 to 3 (p < 0.001). Phenogroup membership significantly improved survival prediction compared to conventional factors. Phenogroups were not predictive of hospitalisation for HF. CONCLUSIONS: Applying unsupervised machine learning to routinely collected electronic health record data identified phenogroups with distinct clinical characteristics and unique survival profiles.

Eligibility of Asian and European registry patients for phase III trials in heart failure with reduced ejection fraction.

AIMS: Traditional approaches to designing clinical trials for heart failure (HF) have historically relied on expertise and past practices. However, the evolving landscape of healthcare, marked by the advent of novel data science applications and increased data availability, offers a compelling opportunity to transition towards a data-driven paradigm in trial design. This research aims to evaluate the scope and determinants of disparities between clinical trials and registries by leveraging natural language processing for the analysis of trial eligibility criteria. The findings contribute to the establishment of a robust design framework for guiding future HF trials. METHODS AND RESULTS: Interventional phase III trials registered for HF on ClinicalTrials.gov as of the end of 2021 were identified. Natural language processing was used to extract and structure the eligibility criteria for quantitative analysis. The most common criteria for HF with reduced ejection fraction (HFrEF) were applied to estimate patient eligibility as a proportion of registry patients in the ASIAN-HF (N = 4868) and BIOSTAT-CHF registries (N = 2545). Of the 375 phase III trials for HF, 163 HFrEF trials were identified. In these trials, the most frequently encountered inclusion criteria were New York Heart Association (NYHA) functional class (69%), worsening HF (23%), and natriuretic peptides (18%), whereas the most frequent comorbidity-based exclusion criteria were acute coronary syndrome (64%), renal disease (55%), and valvular heart disease (47%). On average, 20% of registry patients were eligible for HFrEF trials. Eligibility distributions did not differ (P = 0.18) between Asian [median eligibility 0.20, interquartile range (IQR) 0.08-0.43] and European registry populations (median 0.17, IQR 0.06-0.39). With time, HFrEF trials became more restrictive, where patient eligibility declined from 0.40 in 1985-2005 to 0.19 in 2016-2022 (P = 0.03). When frequency among trials is taken into consideration, the eligibility criteria that were most restrictive were prior myocardial infarction, NYHA class, age, and prior HF hospitalization. CONCLUSIONS: Based on 14 trial criteria, only one-fifth of registry patients were eligible for phase III HFrEF trials. Overall eligibility rates did not differ between the Asian and European patient cohorts.

Prognostic significance of troponin in patients with malignancy (NIHR Health Informatics Collaborative TROP-MALIGNANCY study).

BACKGROUND: Cardiac troponin is commonly raised in patients presenting with malignancy. The prognostic significance of raised troponin in these patients is unclear. OBJECTIVES: We sought to investigate the relation between troponin and mortality in a large, well characterised cohort of patients with a routinely measured troponin and a primary diagnosis of malignancy. METHODS: We used the National Institute for Health Research (NIHR) Health Informatics Collaborative data of 5571 patients, who had troponin levels measured at 5 UK cardiac centres between 2010 and 2017 and had a primary diagnosis of malignancy. Patients were classified into solid tumour or haematological malignancy subgroups. Peak troponin levels were standardised as a multiple of each laboratory's 99th -percentile upper limit of normal (xULN). RESULTS: 4649 patients were diagnosed with solid tumours and 922 patients with haematological malignancies. Raised troponin was an independent predictor of mortality in all patients (Troponin > 10 vs. <1 adjusted HR 2.01, 95% CI 1.73 to 2.34), in solid tumours (HR 1.84, 95% CI 1.55 to 2.19), and in haematological malignancy (HR 2.72, 95% CI 1.99 to 3.72). There was a significant trend in increasing mortality risk across troponin categories in all three subgroups (p < 0.001). CONCLUSION: Raised troponin level is associated with increased mortality in patients with a primary diagnosis of malignancy regardless of cancer subtype. Mortality risk is stable for patients with a troponin level below the ULN but increases as troponin level increases above the ULN in the absence of acute coronary syndrome.

Consumer wearable devices for evaluation of heart rate control using digoxin versus beta-blockers: the RATE-AF randomized trial.

Consumer-grade wearable technology has the potential to support clinical research and patient management. Here, we report results from the RATE-AF trial wearables study, which was designed to compare heart rate in older, multimorbid patients with permanent atrial fibrillation and heart failure who were randomized to treatment with either digoxin or beta-blockers. Heart rate (n = 143,379,796) and physical activity (n = 23,704,307) intervals were obtained from 53 participants (mean age 75.6 years (s.d. 8.4), 40% women) using a wrist-worn wearable linked to a smartphone for 20 weeks. Heart rates in participants treated with digoxin versus beta-blockers were not significantly different (regression coefficient 1.22 (95% confidence interval (CI) -2.82 to 5.27; P = 0.55); adjusted 0.66 (95% CI -3.45 to 4.77; P = 0.75)). No difference in heart rate was observed between the two groups of patients after accounting for physical activity (P = 0.74) or patients with high activity levels (≥30,000 steps per week; P = 0.97). Using a convolutional neural network designed to account for missing data, we found that wearable device data could predict New York Heart Association functional class 5 months after baseline assessment similarly to standard clinical measures of electrocardiographic heart rate and 6-minute walk test (F1 score 0.56 (95% CI 0.41 to 0.70) versus 0.55 (95% CI 0.41 to 0.68); P = 0.88 for comparison). The results of this study indicate that digoxin and beta-blockers have equivalent effects on heart rate in atrial fibrillation at rest and on exertion, and suggest that dynamic monitoring of individuals with arrhythmia using wearable technology could be an alternative to in-person assessment. ClinicalTrials.gov identifier: NCT02391337 .

Evaluation of the 2021 ESC recommendations for family screening in hereditary transthyretin cardiac amyloidosis.

AIMS: The 2021 European Society of Cardiology (ESC) screening recommendations for individuals carrying a pathogenic transthyretin amyloidosis variant (ATTRv) are based on expert opinion. We aimed to (i) determine the penetrance of ATTRv cardiomyopathy (ATTRv-CM) at baseline; (ii) examine the value of serial evaluation; and (iii) establish the yield of first-line diagnostic tests (i.e. electrocardiogram, echocardiogram, and laboratory tests) as per 2021 ESC position statement. METHODS AND RESULTS: We included 159 relatives (median age 55.6 [43.2-65.9] years, 52% male) at risk for ATTRv-CM from 10 centres. The primary endpoint, ATTRv-CM diagnosis, was defined as the presence of (i) cardiac tracer uptake in bone scintigraphy; or (ii) transthyretin-positive cardiac biopsy. The secondary endpoint was a composite of heart failure (New York Heart Association class ≥II) and pacemaker-requiring conduction disorders. At baseline, 40/159 (25%) relatives were diagnosed with ATTRv-CM. Of those, 20 (50%) met the secondary endpoint. Indication to screen (≤10 years prior to predicted disease onset and absence of extracardiac amyloidosis) had an excellent negative predictive value (97%). Other pre-screening predictors for ATTRv-CM were infrequently identified variants and male sex. Importantly, 13% of relatives with ATTRv-CM did not show any signs of cardiac involvement on first-line diagnostic tests. The yield of serial evaluation (n = 41 relatives; follow-up 3.1 [2.2-5.2] years) at 3-year interval was 9.4%. CONCLUSIONS: Screening according to the 2021 ESC position statement performs well in daily clinical practice. Clinicians should adhere to repeating bone scintigraphy after 3 years, as progressing to ATTRv-CM without signs of ATTRv-CM on first-line diagnostic tests or symptoms is common.

Using Artificial Intelligence to Predict Heart Failure Risk from Single-lead Electrocardiographic Signals: A Multinational Assessment.

Importance: Despite the availability of disease-modifying therapies, scalable strategies for heart failure (HF) risk stratification remain elusive. Portable devices capable of recording single-lead electrocardiograms (ECGs) can enable large-scale community-based risk assessment. Objective: To evaluate an artificial intelligence (AI) algorithm to predict HF risk from noisy single-lead ECGs. Design: Multicohort study. Setting: Retrospective cohort of individuals with outpatient ECGs in the integrated Yale New Haven Health System (YNHHS) and prospective population-based cohorts of UK Biobank (UKB) and Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Participants: Individuals without HF at baseline. Exposures: AI-ECG-defined risk of left ventricular systolic dysfunction (LVSD). Main Outcomes and Measures: Among individuals with ECGs, we isolated lead I ECGs and deployed a noise-adapted AI-ECG model trained to identify LVSD. We evaluated the association of the model probability with new-onset HF, defined as the first HF hospitalization. We compared the discrimination of AI-ECG against the pooled cohort equations to prevent HF (PCP-HF) score for new-onset HF using Harrel's C-statistic, integrated discrimination improvement (IDI), and net reclassification improvement (NRI). Results: There were 194,340 YNHHS patients (age 56 years [IQR, 41-69], 112,082 women [58%]), 42,741 UKB participants (65 years [59-71], 21,795 women [52%]), and 13,454 ELSA-Brasil participants (56 years [41-69], 7,348 women [55%]) with baseline ECGs. A total of 3,929 developed HF in YNHHS over 4.5 years (2.6-6.6), 46 in UKB over 3.1 years (2.1-4.5), and 31 in ELSA-Brasil over 4.2 years (3.7-4.5). A positive AI-ECG screen was associated with a 3- to 7-fold higher risk for HF, and each 0.1 increment in the model probability portended a 27-65% higher hazard across cohorts, independent of age, sex, comorbidities, and competing risk of death. AI-ECG's discrimination for new-onset HF was 0.725 in YNHHS, 0.792 in UKB, and 0.833 in ELSA-Brasil. Across cohorts, incorporating AI-ECG predictions in addition to PCP-HF resulted in improved Harrel's C-statistic (Δ=0.112-0.114), with an IDI of 0.078-0.238 and an NRI of 20.1%-48.8% for AI-ECG vs. PCP-HF. Conclusions and Relevance: Across multinational cohorts, a noise-adapted AI model with lead I ECGs as the sole input defined HF risk, representing a scalable portable and wearable device-based HF risk-stratification strategy.

Digital twins: reimagining the future of cardiovascular risk prediction and personalised care.

The rapid evolution of highly adaptable and reusable artificial intelligence models facilitates the implementation of digital twinning and has the potential to redefine cardiovascular risk prevention. Digital twinning combines vast amounts of data from diverse sources to construct virtual models of an individual. Emerging artificial intelligence models, called generalist AI, enable the processing of different types of data, including data from electronic health records, laboratory results, medical texts, imaging, genomics, or graphs. Among their unprecedented capabilities are an easy adaptation of a model to previously unseen medical tasks and the ability to reason and explain output using precise medical language derived from scientific literature, medical guidelines, or knowledge graphs. The proposed combination of a digital twinning approach with generalist AI is a path to accelerate the implementation of precision medicine and enhance early recognition and prevention of cardiovascular disease. This proposed strategy may extend to other domains to advance predictive, preventive, and precision medicine and also boost health research discoveries.

Changes in the investigation and management of suspected myocardial infarction and injury during COVID-19: a multi-centre study using routinely collected healthcare data.

Objective: The COVID-19 pandemic was associated with a reduction in the incidence of myocardial infarction (MI) diagnosis, in part because patients were less likely to present to hospital. Whether changes in clinical decision making with respect to the investigation and management of patients with suspected MI also contributed to this phenomenon is unknown. Methods: Multicentre retrospective cohort study in three UK centres contributing data to the National Institute for Health Research Health Informatics Collaborative. Patients presenting to the Emergency Department (ED) of these centres between 1st January 2020 and 1st September 2020 were included. Three time epochs within this period were defined based on the course of the first wave of the COVID-19 pandemic: pre-pandemic (epoch 1), lockdown (epoch 2), post-lockdown (epoch 3). Results: During the study period, 10,670 unique patients attended the ED with chest pain or dyspnoea, of whom 6,928 were admitted. Despite fewer total ED attendances in epoch 2, patient presentations with dyspnoea were increased (p < 0.001), with greater likelihood of troponin testing in both chest pain (p = 0.001) and dyspnoea (p < 0.001). There was a dramatic reduction in elective and emergency cardiac procedures (both p < 0.001), and greater overall mortality of patients (p < 0.001), compared to the pre-pandemic period. Positive COVID-19 and/or troponin test results were associated with increased mortality (p < 0.001), though the temporal risk profile differed. Conclusions: The first wave of the COVID-19 pandemic was associated with significant changes not just in presentation, but also the investigation, management, and outcomes of patients presenting with suspected myocardial injury or MI.

Serial cardiac biomarkers, pulmonary artery pressures and traditional parameters of fluid status in relation to prognosis in patients with chronic heart failure: Design and rationale of the BioMEMS study.

AIMS: Heart failure (HF), a global pandemic affecting millions of individuals, calls for adequate predictive guidance for improved therapy. Congestion, a key factor in HF-related hospitalizations, further underscores the need for timely interventions. Proactive monitoring of intracardiac pressures, guided by pulmonary artery (PA) pressure, offers opportunities for efficient early-stage intervention, since haemodynamic congestion precedes clinical symptoms. METHODS: The BioMEMS study, a substudy of the MONITOR-HF trial, proposes a multifaceted approach integrating blood biobank data with traditional and novel HF parameters. Two additional blood samples from 340 active participants in the MONITOR-HF trial were collected at baseline, 3-, 6-, and 12-month visits and stored for the BioMEMS biobank. The main aims are to identify the relationship between temporal biomarker patterns and PA pressures derived from the CardioMEMS-HF system, and to identify the biomarker profile(s) associated with the risk of HF events and cardiovascular death. CONCLUSION: Since the prognostic value of single baseline measurements of biomarkers like N-terminal pro-B-type natriuretic peptide is limited, with the BioMEMS study we advocate a dynamic, serial approach to better capture HF progression. We will substantiate this by relating repeated biomarker measurements to PA pressures. This design rationale presents a comprehensive review on cardiac biomarkers in HF, and aims to contribute valuable insights into personalized HF therapy and patient risk assessment, advancing our ability to address the evolving nature of HF effectively.

Contemporary guideline-directed medical therapy in de novo, chronic, and worsening heart failure patients: First data from the TITRATE-HF study.

AIMS: Despite clear guideline recommendations for initiating four drug classes in all patients with heart failure (HF) with reduced ejection fraction (HFrEF) and the availability of rapid titration schemes, information on real-world implementation lags behind. Closely following the 2021 ESC HF guidelines and 2023 focused update, the TITRATE-HF study started to prospectively investigate the use, sequencing, and titration of guideline-directed medical therapy (GDMT) in HF patients, including the identification of implementation barriers. METHODS AND RESULTS: TITRATE-HF is an ongoing long-term HF registry conducted in the Netherlands. Overall, 4288 patients from 48 hospitals were included. Among these patients, 1732 presented with de novo, 2240 with chronic, and 316 with worsening HF. The median age was 71 years (interquartile range [IQR] 63-78), 29% were female, and median ejection fraction was 35% (IQR 25-40). In total, 44% of chronic and worsening HFrEF patients were prescribed quadruple therapy. However, only 1% of HFrEF patients achieved target dose for all drug classes. In addition, quadruple therapy was more often prescribed to patients treated in a dedicated HF outpatient clinic as compared to a general cardiology outpatient clinic. In each GDMT drug class, 19% to 36% of non-use in HFrEF patients was related to side-effects, intolerances, or contraindications. In the de novo HF cohort, 49% of patients already used one or more GDMT drug classes for other indications than HF. CONCLUSION: This first analysis of the TITRATE-HF study reports relatively high use of GDMT in a contemporary HF cohort, while still showing room for improvement regarding quadruple therapy. Importantly, the use and dose of GDMT were suboptimal, with the reasons often remaining unclear. This underscores the urgency for further optimization of GDMT and implementation strategies within HF management.

Scalable Risk Stratification for Heart Failure Using Artificial Intelligence applied to 12-lead Electrocardiographic Images: A Multinational Study.

Background: Current risk stratification strategies for heart failure (HF) risk require either specific blood-based biomarkers or comprehensive clinical evaluation. In this study, we evaluated the use of artificial intelligence (AI) applied to images of electrocardiograms (ECGs) to predict HF risk. Methods: Across multinational longitudinal cohorts in the integrated Yale New Haven Health System (YNHHS) and in population-based UK Biobank (UKB) and Brazilian Longitudinal Study of Adult Health (ELSA-Brasil), we identified individuals without HF at baseline. Incident HF was defined based on the first occurrence of an HF hospitalization. We evaluated an AI-ECG model that defines the cross-sectional probability of left ventricular dysfunction from a single image of a 12-lead ECG and its association with incident HF. We accounted for the competing risk of death using the Fine-Gray subdistribution model and evaluated the discrimination using Harrel's c-statistic. The pooled cohort equations to prevent HF (PCP-HF) were used as a comparator for estimating incident HF risk. Results: Among 231,285 individuals at YNHHS, 4472 had a primary HF hospitalization over 4.5 years (IQR 2.5-6.6) of follow-up. In UKB and ELSA-Brasil, among 42,741 and 13,454 people, 46 and 31 developed HF over a follow-up of 3.1 (2.1-4.5) and 4.2 (3.7-4.5) years, respectively. A positive AI-ECG screen portended a 4-fold higher risk of incident HF among YNHHS patients (age-, sex-adjusted HR [aHR] 3.88 [95% CI, 3.63-4.14]). In UKB and ELSA-Brasil, a positive-screen ECG portended 13- and 24-fold higher hazard of incident HF, respectively (aHR: UKBB, 12.85 [6.87-24.02]; ELSA-Brasil, 23.50 [11.09-49.81]). The association was consistent after accounting for comorbidities and the competing risk of death. Higher model output probabilities were progressively associated with a higher risk for HF. The model's discrimination for incident HF was 0.718 in YNHHS, 0.769 in UKB, and 0.810 in ELSA-Brasil. Across cohorts, incorporating model probability with PCP-HF yielded a significant improvement in discrimination over PCP-HF alone. Conclusions: An AI model applied to images of 12-lead ECGs can identify those at elevated risk of HF across multinational cohorts. As a digital biomarker of HF risk that requires just an ECG image, this AI-ECG approach can enable scalable and efficient screening for HF risk.

Explainable machine learning using echocardiography to improve risk prediction in patients with chronic coronary syndrome.

Aims: The European Society of Cardiology guidelines recommend risk stratification with limited clinical parameters such as left ventricular (LV) function in patients with chronic coronary syndrome (CCS). Machine learning (ML) methods enable an analysis of complex datasets including transthoracic echocardiography (TTE) studies. We aimed to evaluate the accuracy of ML using clinical and TTE data to predict all-cause 5-year mortality in patients with CCS and to compare its performance with traditional risk stratification scores. Methods and results: Data of consecutive patients with CCS were retrospectively collected if they attended the outpatient clinic of Amsterdam UMC location AMC between 2015 and 2017 and had a TTE assessment of the LV function. An eXtreme Gradient Boosting (XGBoost) model was trained to predict all-cause 5-year mortality. The performance of this ML model was evaluated using data from the Amsterdam UMC location VUmc and compared with the reference standard of traditional risk scores. A total of 1253 patients (775 training set and 478 testing set) were included, of which 176 patients (105 training set and 71 testing set) died during the 5-year follow-up period. The ML model demonstrated a superior performance [area under the receiver operating characteristic curve (AUC) 0.79] compared with traditional risk stratification tools (AUC 0.62-0.76) and showed good external performance. The most important TTE risk predictors included in the ML model were LV dysfunction and significant tricuspid regurgitation. Conclusion: This study demonstrates that an explainable ML model using TTE and clinical data can accurately identify high-risk CCS patients, with a prognostic value superior to traditional risk scores.

Characterization of Non-Ischemic Dilated Cardiomyopathy in a Native Tanzanian Cohort: MOYO Study.

Background: Non-ischemic dilated cardiomyopathy (NIDCM) is a common cause of heart failure with progressive tendency. The disease occurs in one in every 2,500 individuals in the developed world, with high morbidity and mortality. However, detailed data on the role of NIDCM in heart failure in Tanzania is lacking. Aim: To characterize NIDCM in a Tanzanian cohort with respect to demographics, clinical profile, imaging findings and management. Methods: Characterization of non-ischemic dilated cardioMyOpathY in a native Tanzanian cOhort (MOYO) is a prospective cohort study of NIDCM patients seen at the Jakaya Kikwete Cardiac Institute. Patients aged ≥18 years with a clinical diagnosis of heart failure, an ejection fraction of ≤45% on echocardiography and no evidence of ischemia were enrolled. Clinical data, echocardiography, electrocardiography (ECG), coronary angiography and stress ECG information were collected from February 2020 to March 2022. Results: Of 402 patients, n = 220 (54.7%) were males with a median (IQR) age of 55.0 (41.0, 66.0) years. Causes of NIDCM were presumably hypertensive n = 218 (54.2%), idiopathic n = 116 (28.9%), PPCM n = 45 (11.2%), alcoholic n = 10 (2.5%) and other causes n = 13 (3.2%). The most common presenting symptoms were dyspnea n = 342 (85.1%), with the majority of patients presenting with New York Heart Association (NYHA) Class III n = 195 (48.5%). The mean (SD) left ventricular ejection fraction (LVEF) was 29.4% (±7.7), and severe systolic dysfunction (LVEF <30%) was common n = 208 (51.7%). Compared with other forms of DCM, idiopathic DCM patients were significantly younger, had more advanced NYHA class (p < 0.001) and presented more often with left bundle branch block on ECG (p = 0.0042). There was suboptimal use of novel guidelines recommended medications ARNI n = 10 (2.5%) and SGLT2 2-inhibitors n = 2 (0.5%). Conclusions: In our Tanzanian cohort, the majority of patients with NIDCM have an identified underlying cause, and they present at late stages of the disease. Patients with idiopathic DCM are younger with more severe disease compared to other forms of NIDCM.

Prioritising genetic findings for drug target identification and validation.

The decreasing costs of high-throughput genetic sequencing and increasing abundance of sequenced genome data have paved the way for the use of genetic data in identifying and validating potential drug targets. However, the number of identified potential drug targets is often prohibitively large to experimentally evaluate in wet lab experiments, highlighting the need for systematic approaches for target prioritisation. In this review, we discuss principles of genetically guided drug development, specifically addressing loss-of-function analysis, colocalization and Mendelian randomisation (MR), and the contexts in which each may be most suitable. We subsequently present a range of biomedical resources which can be used to annotate and prioritise disease-associated proteins identified by these studies including 1) ontologies to map genes, proteins, and disease, 2) resources for determining the druggability of a potential target, 3) tissue and cell expression of the gene encoding the potential target, and 4) key biological pathways involving the potential target. We illustrate these concepts through a worked example, identifying a prioritised set of plasma proteins associated with non-alcoholic fatty liver disease (NAFLD). We identified five proteins with strong genetic support for involvement with NAFLD: CYB5A, NT5C, NCAN, TGFBI and DAPK2. All of the identified proteins were expressed in both liver and adipose tissues, with TGFBI and DAPK2 being potentially druggable. In conclusion, the current review provides an overview of genetic evidence for drug target identification, and how biomedical databases can be used to provide actionable prioritisation, fully informing downstream experimental validation.

Digital Solutions to Optimize Guideline-Directed Medical Therapy Prescriptions in Heart Failure Patients: Current Applications and Future Directions.

PURPOSEOF REVIEW: Guideline-directed medical therapy (GDMT) underuse is common in heart failure (HF) patients. Digital solutions have the potential to support medical professionals to optimize GDMT prescriptions in a growing HF population. We aimed to review current literature on the effectiveness of digital solutions on optimization of GDMT prescriptions in patients with HF. RECENT FINDINGS: We report on the efficacy, characteristics of the study, and population of published digital solutions for GDMT optimization. The following digital solutions are discussed: teleconsultation, telemonitoring, cardiac implantable electronic devices, clinical decision support embedded within electronic health records, and multifaceted interventions. Effect of digital solutions is reported in dedicated studies, retrospective studies, or larger studies with another focus that also commented on GDMT use. Overall, we see more studies on digital solutions that report a significant increase in GDMT use. However, there is a large heterogeneity in study design, outcomes used, and populations studied, which hampers comparison of the different digital solutions. Barriers, facilitators, study designs, and future directions are discussed. There remains a need for well-designed evaluation studies to determine safety and effectiveness of digital solutions for GDMT optimization in patients with HF. Based on this review, measuring and controlling vital signs in telemedicine studies should be encouraged, professionals should be actively alerted about suboptimal GDMT, the researchers should consider employing multifaceted digital solutions to optimize effectiveness, and use study designs that fit the unique sociotechnical aspects of digital solutions. Future directions are expected to include artificial intelligence solutions to handle larger datasets and relieve medical professional's workload.

Soluble Suppression of Tumorigenicity-2 Predicts Mortality and Right Heart Failure in Patients With a Left Ventricular Assist Device.

BACKGROUND: Soluble suppression of tumorigenicity-2 (sST2) predicts mortality in patients with heart failure. The predictive value of sST2 in patients with a left ventricular assist device remains unknown. Therefore, we studied the relationship between sST2 and outcome after left ventricular assist device implantation. METHODS AND RESULTS: sST2 levels of patients with a left ventricular assist device implanted between January 2015 and December 2022 were included in this observational study. The median follow-up was 25 months, during which 1573 postoperative sST2 levels were measured in 199 patients, with a median of 29 ng/mL. Survival of patients with normal and elevated preoperative levels was compared using Kaplan-Meier analysis, which did not differ significantly (P=0.22) between both groups. The relationship between postoperative sST2, survival, and right heart failure was evaluated using a joint model, which showed a significant relationship between the absolute sST2 level and mortality, with a hazard ratio (HR) of 1.20 (95% CI, 1.10-1.130; P<0.01) and an HR of 1.22 (95% CI, 1.07-1.39; P=0.01) for right heart failure, both per 10-unit sST2 increase. The sST2 instantaneous change was not predictive for survival or right heart failure (P=0.99 and P=0.94, respectively). Multivariate joint model analysis showed a significant relationship between sST2 with mortality adjusted for NT-proBNP (N-terminal pro-B-type natriuretic peptide), with an HR of 1.19 (95% CI, 1.00-1.42; P=0.05), whereas the HR of right heart failure was not significant (1.22 [95% CI, 0.94-1.59]; P=0.14), both per 10-unit sST2 increase. CONCLUSIONS: Time-dependent postoperative sST2 predicts all-cause mortality after left ventricular assist device implantation after adjustment for NT-proBNP. Future research is warranted into possible target interventions and the optimal monitoring frequency.

Acute Coronary Syndrome Subphenotypes Based on Repeated Biomarker Measurements in Relation to Long-Term Mortality Risk.

BACKGROUND: We aimed to identify patients with subphenotypes of postacute coronary syndrome (ACS) using repeated measurements of high-sensitivity cardiac troponin T, N-terminal pro-B-type natriuretic peptide, high-sensitivity C-reactive protein, and growth differentiation factor 15 in the year after the index admission, and to investigate their association with long-term mortality risk. METHODS AND RESULTS: BIOMArCS (BIOMarker Study to Identify the Acute Risk of a Coronary Syndrome) was an observational study of patients with ACS, who underwent high-frequency blood sampling for 1 year. Biomarkers were measured in a median of 16 repeated samples per individual. Cluster analysis was performed to identify biomarker-based subphenotypes in 723 patients without a repeat ACS in the first year. Patients with a repeat ACS (N=36) were considered a separate cluster. Differences in all-cause death were evaluated using accelerated failure time models (median follow-up, 9.1 years; 141 deaths). Three biomarker-based clusters were identified: cluster 1 showed low and stable biomarker concentrations, cluster 2 had elevated concentrations that subsequently decreased, and cluster 3 showed persistently elevated concentrations. The temporal biomarker patterns of patients in cluster 3 were similar to those with a repeat ACS during the first year. Clusters 1 and 2 had a similar and favorable long-term mortality risk. Cluster 3 had the highest mortality risk. The adjusted survival time ratio was 0.64 (95% CI, 0.44-0.93; P=0.018) compared with cluster 1, and 0.71 (95% CI, 0.39-1.32; P=0.281) compared with patients with a repeat ACS. CONCLUSIONS: Patients with subphenotypes of post-ACS with different all-cause mortality risks during long-term follow-up can be identified on the basis of repeatedly measured cardiovascular biomarkers. Patients with persistently elevated biomarkers have the worst outcomes, regardless of whether they experienced a repeat ACS in the first year.

Rationale and design of the THIRST Alert feasibility study: a pragmatic, single-centre, parallel-group randomised controlled trial of an interruptive alert for oral fluid restriction in patients treated with intravenous furosemide.

INTRODUCTION: Acute heart failure (HF) is a major cause of unplanned hospitalisation characterised by excess body water. A restriction in oral fluid intake is commonly imposed on patients as an adjunct to pharmacological therapy with loop diuretics, but there is a lack of evidence from traditional randomised controlled trials (RCTs) to support the safety and effectiveness of this intervention in the acute setting.This study aims to explore the feasibility of using computer alerts within the electronic health record (EHR) system to invite clinical care teams to enrol patients into a pragmatic RCT at the time of clinical decision-making. It will additionally assess the effectiveness of using an alert to help address the clinical research question of whether oral fluid restriction is a safe and effective adjunct to pharmacological therapy for patients admitted with fluid overload. METHODS AND ANALYSIS: THIRST (Randomised Controlled Trial within the electronic Health record of an Interruptive alert displaying a fluid Restriction Suggestion in patients with the treatable Trait of congestion) Alert is a single-centre, parallel-group, open-label pragmatic RCT embedded in the EHR system that will be conducted as a feasibility study at an National Health Service (NHS) hospital in London. The clinical care team will be invited to enrol suitable patients in the study using a point-of-care alert with a target sample size of 50 patients. Enrolled patients will then be randomised to either restricted or unrestricted oral fluid intake. Two primary outcomes will be explored (1) the proportion of eligible patients enrolled in the study and (2) the mean difference in oral fluid intake between randomised groups. A series of secondary outcomes are specified to evaluate the effectiveness of the alert, adherence to the randomised treatment allocation and the quality of data generated from routine care, relevant to the outcomes of interest. ETHICS AND DISSEMINATION: This study was approved by Riverside Research Ethics Committee (Ref: 22/LO/0889) and will be published on completion. TRIAL REGISTRATION NUMBER: NCT05869656.

Embedding routine health care data in clinical trials: with great power comes great responsibility.

Randomised clinical trials (RCTs) are vital for medical progress. Unfortunately, 'traditional' RCTs are expensive and inherently slow. Moreover, their generalisability has been questioned. There is considerable overlap in routine health care data (RHCD) and trial-specific data. Therefore, integration of RHCD in an RCT has great potential, as it would reduce the effort and costs required to collect data, thereby overcoming some of the major downsides of a traditional RCT. However, use of RHCD comes with other challenges, such as privacy issues, as well as technical and practical barriers. Here, we give a current overview of related initiatives on national cardiovascular registries (Netherlands Heart Registration, Heart4Data), showcasing the interrelationships between and the relevance of the different registries for the practicing physician. We then discuss the benefits and limitations of RHCD use in the setting of a pragmatic RCT from a cardiovascular perspective, illustrated by a case study in heart failure.

Artificial intelligence: revolutionizing cardiology with large language models.

Natural language processing techniques are having an increasing impact on clinical care from patient, clinician, administrator, and research perspective. Among others are automated generation of clinical notes and discharge letters, medical term coding for billing, medical chatbots both for patients and clinicians, data enrichment in the identification of disease symptoms or diagnosis, cohort selection for clinical trial, and auditing purposes. In the review, an overview of the history in natural language processing techniques developed with brief technical background is presented. Subsequently, the review will discuss implementation strategies of natural language processing tools, thereby specifically focusing on large language models, and conclude with future opportunities in the application of such techniques in the field of cardiology.