Use of artificial intelligence-guided echocardiography to detect cardiac dysfunction and heart valve disease in rural and remote areas: Rationale and design of the AGILE-echo trial.

BACKGROUND: Transthoracic echocardiography (TTE) is essential in the diagnosis of cardiovascular diseases (CVD), including but not limited to heart failure (HF) and heart valve disease (HVD). However, its dependence on expert acquisition means that its accessibility in rural areas may be limited, leading to delayed management decisions and potential missed diagnoses. Artificial intelligence-guided (AI)-TTE offers a solution by permitting non-expert image acquisition. The impact of AI-TTE on the timing of diagnosis and early initiation of cardioprotection is undefined. METHODS: AGILE-Echo (use of Artificial intelligence-Guided echocardiography to assIst cardiovascuLar patient managEment) is a randomized-controlled trial conducted in 5 rural and remote areas around Australia. Adults with CV risk factors and exercise intolerance, or concerns regarding HVD are randomized into AI-TTE or usual care (UC). AI-TTE participants may have a cardiovascular problem excluded, identified (leading to AI-guided interventions) or unresolved (leading to conventional TTE). UC participants undergo usual management, including referral for standard TTE. The primary endpoint is a composite of HVD or HF diagnosis at 12-months. Subgroup analysis, stratified based on age range and sex, will be conducted. All statistical analyses will be conducted using R. RESULTS: Of the first 157 participants, 78 have been randomized into AI-TTE (median age 68 [IQR 17]) and 79 to UC (median age 65 [IQR 17], P = .034). HVD was the primary concern in 37 participants (23.6%) while 84.7% (n = 133) experienced exercise intolerance. The overall 10-year HF incidence risk was 13.4% and 20.0% (P = .089) for UC and AI-TTE arm respectively. Atrial remodeling, left ventricular remodeling and valvular regurgitation were the most common findings. Thirty-three patients (42.3%) showed no abnormalities. CONCLUSIONS: This randomized-controlled trial of AI-TTE will provide proof-of-concept for the role of AI-TTE in identifying pre-symptomatic HF or HVD when access to TTE is limited. Additionally, this could promote the usage of AI-TTE in rural or remote areas, ultimately improving health and quality of life of community dwelling adults with risks, signs or symptoms of cardiac dysfunction.

Absolute cardiovascular risk assessment using 'real world' clinic blood pressures compared to standardized unobserved and ambulatory methods: an observational study.

Clinic blood pressure (BP) is recommended for absolute cardiovascular disease (CVD) risk assessment. However, in 'real-world' settings, clinic BP measurement is unstandardised and less reliable compared to more rigorous methods but the impact for absolute CVD risk assessment is unknown. This study aimed to determine the difference in absolute CVD risk assessment using real-world clinic BP compared to standardised BP methods. Participants were patients (n = 226, 59 ± 15 years; 58% female) with hypertension referred to a BP clinic for assessment. 'Real-world' clinic BP was provided by the referring doctor. All participants had unobserved automated office BP (AOBP) and 24-h ambulatory BP monitoring (ABPM) measured at the clinic. Absolute CVD risk was calculated (Framingham) using systolic BP from the referring doctor (clinic BP), AOBP and ABPM, with agreement assessed by Kappa statistic. Clinic systolic BP was 18 mmHg than AOBP and daytime ABPM and 22 mmHg higher than 24-h ABPM (p < 0.001). Subsequently, absolute CVD risk scores using clinic BP were higher compared to AOBP, daytime ABPM and 24-h ABPM (10.4 ± 8.1%, 7.8 ± 6.4%, 7.8 ± 6.3%, and 7.3 ± 6.1%, respectively, P < 0.001). As a result, more participants were classified as high CVD risk using clinic BP (n = 89, 40%) compared with AOBP (n = 44, 20%) daytime ABPM (n = 38, 17%) and 24-h ABPM (n = 38, 17%) (p < 0.001) with weak agreement in risk classification (κ = 0.57[0.45-0.69], κ = 0.52[0.41-0.64] and κ = 0.55[0.43-0.66], respectively). Real-world clinic BP was higher and classified twice as many participants at high CVD risk compared to AOBP or ABPM. Given the challenges to high-quality BP measurement in clinic, more rigorous BP measurement methods are needed for absolute CVD risk assessment.

Prognostication and Interventional Guidance Using Acceleration-Ejection Time Ratio in Undifferentiated Paradoxical Low-Flow Low-Gradient Aortic Stenosis.

BACKGROUND: Studies in paradoxical low-flow low-gradient aortic stenosis (PLFAS) have demonstrated conflicting outcomes with variable survival advantage from aortic valve replacement (AVR). PLFAS is a heterogeneous composition of patients with uncertainty regarding true stenosis severity that continues to confound decision-making for AVR. OBJECTIVES: The purpose of this study was to investigate the utility of the Doppler acceleration (AT) to ejection (ET) time ratio (AT:ET) for prediction of prognosis and benefit from AVR in undifferentiated PLFAS. METHODS: Patients with echocardiographic findings of PLFAS (aortic valve area <1.0 cm2 or indexed aortic valve area <0.6 cm2/m2, mean gradient <40 mm Hg, indexed stroke volume <35 mL/m2, and left ventricular ejection fraction ≥50%) were identified and grouped according to an AT:ET cutoff of 0.35. The primary outcome was a 5-year composite of cardiac mortality or AVR. Secondary outcomes included the individual components of the primary endpoint and all-cause mortality at 5 years. Effect of AVR was analyzed in the AT:ET <0.35 and ≥0.35 groups. RESULTS: A total of 171 PLFAS patients (median age 77.0 years, 57% women) were followed for a median of 8.9 years. AT:ET ≥0.35 was an independent predictor of the primary outcome (HR: 4.77 [95% CI: 2.94-7.75]; P < 0.001) with incremental value over standard indices of stenosis severity (net reclassification improvement: 0.57 [95% CI: 0.14-0.84]). AT:ET ≥0.35 also remained predictive of increased cardiac death (HR: 2.91 [95% CI: 1.47-5.76]; P = 0.002) and AVR (HR: 8.45 [95% CI: 4.16-17.1]; P < 0.001), respectively, following competing risk analysis. No difference in all-cause mortality was observed. AVR in the AT:ET ≥0.35 group was associated with significant reductions in 5-year cardiac (HR: 0.09 [95% CI: 0.02-0.36]; P < 0.001) and all-cause mortality (HR: 0.16 [95% CI: 0.07-0.38]; P < 0.001). No improvement in survival from AVR was demonstrated in AT:ET <0.35 patients. CONCLUSIONS: AT:ET ≥0.35 in PLFAS predicts poorer outcomes and/or need for AVR. In undifferentiated PLFAS patients, AT:ET may have a potential role in improving patient selection for prognostic AVR.

Definition and Validation of Prognostic Phenotypes in Moderate Aortic Stenosis.

BACKGROUND: Adverse outcomes from moderate aortic stenosis (AS) may be caused by progression to severe AS or by the effects of comorbidities. In the absence of randomized trial evidence favoring aortic valve replacement (AVR) in patients with moderate AS, phenotyping patients according to risk may assist decision making. OBJECTIVES: This study sought to identify and validate clusters of moderate AS that may be used to guide patient management. METHODS: Unsupervised clustering algorithms were applied to demographics, comorbidities, and echocardiographic parameters in a training data set in patients with moderate AS (n = 2,469). External validation was obtained by assigning the defined clusters to an independent group with moderate AS (n = 1,358). The primary outcome, a composite of cardiac death, heart failure hospitalization, or aortic valve (AV) intervention after 5 years, was assessed between clusters in both data sets. RESULTS: Four distinct clusters-cardiovascular (CV)-comorbid, low-flow, calcified AV, and low-risk-with significant outcomes (log-rank P < 0.0001 in both data sets) were identified and replicated. The highest risk was in the CV-comorbid cluster (validation HR: 2.00 [95% CI: 1.54-2.59]; P < 0.001). The effect of AVR on cardiac death differed among the clusters. There was a significantly lower rate of outcomes after AVR in the calcified AV cluster (validation HR: 0.21 [95% CI: 0.08-0.57]; P = 0.002), but no significant effect on outcomes in the other 3 clusters. These analyses were limited by the low rate of AVR. CONCLUSIONS: Moderate AS has several phenotypes, and multiple comorbidities are the key drivers of adverse outcomes in patients with moderate AS. Outcomes of patients with noncalcified moderate AS were not altered by AVR in these groups. Careful attention to subgroups of moderate AS may be important to define treatable risk.

Mechanical dyssynchrony as a selection criterion for cardiac resynchronization therapy: Design of the AMEND-CRT trial.

AIMS: One third of patients do not improve after cardiac resynchronization therapy (CRT). Septal flash (SF) and apical rocking (ApRock) are deformation patterns observed on echocardiography in most patients eligible for CRT. These markers of mechanical dyssynchrony have been associated to improved outcome after CRT in observational studies and may be useful to better select patients. The aim of this trial is to investigate whether the current guideline criteria for selecting patients for CRT should be modified and include SF and ApRock to improve therapy success rate, reduce excessive costs and prevent exposure to device-related complications in patients who would not benefit from CRT. METHODS: The AMEND-CRT trial is a multicentre, randomized, parallel-group, double-blind, sham-controlled trial with a non-inferiority design. The trial will include 578 patients scheduled for CRT according to the 2021 ESC guidelines who satisfy all inclusion criteria. The randomization is performed 1:1 to an active control arm ('guideline arm') or an experimental arm ('echo arm'). All participants receive a device, but in the echo arm, CRT is activated only when SF or ApRock or both are present. The outcome of both arms will be compared after 1 year. The primary outcome measures are the average change in left ventricular end-systolic volume and patient outcome assessed using a modified Packer Clinical Composite Score. CONCLUSIONS: The findings of this trial will redefine the role of echocardiography in CRT and potentially determine which patients with heart failure and a prolonged QRS duration should receive CRT, especially in patients who currently have a class IIa or class IIb recommendation.

Racial Differences in Diabetic Cardiomyopathy: The ARISE-HF Trial.

BACKGROUND: Diabetic cardiomyopathy (DbCM) increases risk of overt heart failure in individuals with diabetes mellitus. Racial and ethnic differences in DbCM remain unexplored. OBJECTIVES: The authors sought to identify racial and ethnic differences among individuals with type 2 diabetes mellitus, structural heart disease, and impaired exercise capacity. METHODS: The ARISE-HF (Aldolase Reductase Inhibitor for Stabilization of Exercise Capacity in Heart Failure) trial is assessing the efficacy of an aldose reductase inhibitor for exercise capacity preservation in 691 persons with DbCM. Baseline characteristics, echocardiographic parameters, and functional capacity were analyzed and stratified by race and ethnicity. RESULTS: The mean age of the study participants was 67.4 years; 50% were women. Black and Hispanic patients had lower use of diabetes mellitus treatments. Black patients had poorer baseline ventricular function and more impaired global longitudinal strain. Overall, health status was preserved, based on Kansas City Cardiomyopathy Questionnaire scores, but reduced exercise capacity was present as evidenced by reduced Physical Activity Scale for the Elderly (PASE) scores. When stratified by race and ethnicity and compared with the entire cohort, Black patients had poorer health status, more reduced physical activity, and a greater impairment in exercise capacity during cardiopulmonary exercise testing, whereas Hispanic patients also displayed compromised cardiopulmonary exercise testing functional capacity. White patients demonstrated higher physical activity and functional capacity. CONCLUSIONS: Racial and ethnic differences exist in baseline characteristics of persons affected by DbCM, with Black and Hispanic study participants demonstrating higher risk features. These insights inform the need to address differences in the population with DbCM. (Safety and Efficacy of AT-001 in Patients With Diabetic Cardiomyopathy [ARISE-HF]; NCT04083339).

Lipidomic Risk Score to Enhance Cardiovascular Risk Stratification for Primary Prevention.

BACKGROUND: Accurate risk stratification is vital for primary prevention of cardiovascular disease (CVD). However, traditional tools such as the Framingham Risk Score (FRS) may underperform within the diverse intermediate-risk group, which includes individuals requiring distinct management strategies. OBJECTIVES: This study aimed to develop a lipidomic-enhanced risk score (LRS), specifically targeting risk prediction and reclassification within the intermediate group, benchmarked against the FRS. METHODS: The LRS was developed via a machine learning workflow using ridge regression on the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab; n = 10,339). It was externally validated with the Busselton Health Study (n = 4,492), and its predictive utility for coronary artery calcium scoring (CACS)-based outcomes was independently validated in the BioHEART cohort (n = 994). RESULTS: LRS significantly improved discrimination metrics for the intermediate-risk group in both AusDiab and Busselton Health Study cohorts (all P < 0.001), increasing the area under the curve for CVD events by 0.114 (95% CI: 0.1123-0.1157) and 0.077 (95% CI: 0.0755-0.0785), with a net reclassification improvement of 0.36 (95% CI: 0.21-0.51) and 0.33 (95% CI: 0.15-0.49), respectively. For CACS-based outcomes in BioHEART, LRS achieved a significant area under the curve improvement of 0.02 over the FRS (0.76 vs 0.74; P < 1.0 × 10-5). A simplified, clinically applicable version of LRS was also created that had comparable performance to the original LRS. CONCLUSIONS: LRS, augmenting the FRS, presents potential to improve intermediate-risk stratification and to predict atherosclerotic markers using a simple blood test, suitable for clinical application. This could facilitate the triage of individuals for noninvasive imaging such as CACS, fostering precision medicine in CVD prevention and management.

Association of Reduced Left Atrial Reserve With Exercise Intolerance and Outcome in Hypertension.

BACKGROUND: Hypertensive heart disease (HHD) is a leading contributor to heart failure with preserved ejection fraction (HFpEF). However, the mechanisms behind the transition to the symptomatic phase remain unclear. OBJECTIVES: We sought to find the association of the exercise response of left atrial (LA) mechanical function with functional capacity, symptoms, and outcome across the heart failure (HF) spectrum in hypertension. METHODS: Echocardiography (including LA reservoir peak atrial longitudinal strain [PALS] and peak atrial contractile strain [PACS] and LA stiffness index) was performed at rest and immediately postexercise in 139 patients with HHD-35 with stage A, 48 with stage B, and 56 with stage C HFpEF. Patients were followed for HF and atrial fibrillation. RESULTS: Exercise capacity was progressively worse from stage A through stage B to stage C and was accompanied by a gradual impairment of changes in PALS and PACS from rest to exercise, whereas LA stiffness reserve remained unchanged until stage C. Peak atrial longitudinal strain and PACS reserves were independently associated with exercise capacity (P = .017 and .008, respectively). Left atrial stiffness reserve and E/e' were the strongest associations of symptomatic HF. Over a median of 25 months, 35 patients developed HF and/or atrial fibrillation. Peak atrial longitudinal strain and PACS reserves were associated with the study end points after adjusting for age, diabetes, N-terminal pro-B type natriuretic peptide, LA volume index, resting E/e', and resting PALS/PACS. CONCLUSIONS: Impaired exercise reserve of LA strain and stiffness are associated with reduced functional capacity in hypertension, and LA strain reserve is independently associated with outcome. These parameters appear to be determinants of progression to overt HF in HHD; however, their contribution may differ depending on HF stage.

Comparison of Echocardiography and Multi-Planar Gated Acquisition Scans for Predicting Cancer-Treatment-Related Cardiovascular Dysfunction.

BACKGROUND: Current guidelines recommend using sequential cardiac imaging to monitor for cancer treatment-related cardiac dysfunction (CTRCD) in patients undergoing potentially cardiotoxic chemotherapy. Multiple different imaging cardiac modalities are available and there are few prospective head-to-head comparative studies to help guide treatment. OBJECTIVES: To perform an exploratory prospective cohort study of "real-world" CTRCD comparing multigated acquisition nuclear ventriculography (MUGA) at the referring cancer specialist's discretion with a novel echocardiographic strategy at an Australian tertiary hospital. METHOD: Patients were recruited from haematology and oncology outpatient clinics if they were scheduled for treatment with anthracyclines and/or trastuzumab. Patients underwent simultaneous MUGA-based cardiac imaging (conventional strategy) at a frequency according to evidenced-based guidelines in addition to researcher-conducted echocardiographic imaging. The echocardiographic imaging was performed in all patients at time points recommended by international society guidelines. Outcomes included adherence to guideline recommendations, concordance between MUGA and echocardiographic left ventricular ejection fraction (LVEF) measurements, and detection of cardiac dysfunction (defined as >5% LVEF decrement from baseline by three-dimensional [3D]-LVEF). A secondary end point was accuracy of global longitudinal strain in predicting cardiac dysfunction. RESULTS: In total, 35 patients were recruited, including 15 with breast cancer, 19 with haematological malignancy, and one with gastric cancer. MUGA and echocardiographic LVEF measurements correlated poorly with limits of agreement of 30% between 3D-LVEF and MUGA-LVEF and 37% for 3D-LVEF and MUGA-LVEF. Only one case (2.9%) of CTRCD was diagnosed by MUGA, compared with 12 (34.2%) cases by echocardiography. Four (4) patients had >10% decrement in 3D-LVEF that was not detected by MUGA. Global longitudinal strain at 2 months displayed significant ability to predict CTRCD (area under the curve, 0.75, 95% confidence interval, 0.55-0.94). CONCLUSIONS: The MUGA correlates poorly with echocardiographic assessment with substantial discrepancy between MUGA and echocardiography in CTRCD diagnosis. Echocardiographic and MUGA imaging strategies should not be considered equivalent for imaging cancer patients, and a single imaging modality should ideally be used per patient to prevent misdiagnosis by inter-modality variation These findings should be considered hypothesis-generating and require confirmation with larger studies.

Blood Pressure Lowering in Patients With Central Hypertension: A Randomized Clinical Trial.

BACKGROUND: Cuff blood pressure (BP) is recommended for guiding hypertension management. However, central BP has been proposed as a superior clinical measurement. This study aimed to determine whether controlling hypertension as measured by central BP was beneficial in reducing left ventricular mass index beyond control of standard cuff hypertension. METHODS: This multicenter, open-label, blinded-end point trial was conducted in individuals treated for uncomplicated hypertension with controlled cuff BP (<140/90 mm Hg) but elevated central BP (≥0.5 SD above age- and sex-specific normal values). Participants were randomized to 24-months intervention with spironolactone 25 mg/day (n=148) or usual care control (n=153). The primary outcome was change in left ventricular mass index measured by cardiac MRI. Cuff and central BPs were measured by clinic, 7-day home and 24-hour ambulatory BPs. RESULTS: At 24-months, there was a greater reduction in left ventricular mass index (-3.2 [95% CI, -5.0 to -1.3] g/m2; P=0.001) with intervention compared with control. Cuff and central BPs were lowered by a similar magnitude across all BP measurement modes (eg, clinic cuff systolic BP, -6.16 [-9.60 to -2.72] mm Hg and clinic central systolic BP, -4.96 [-8.06 to -1.86] mm Hg; P≥0.48 all). Secondary analyses found that changes in left ventricular mass index correlated to changes in BP, with the magnitude of effect nearly identical for BP measured by cuff (eg, 24-hour systolic BP, ß, 0.17 [0.02-0.31] g/m2) or centrally (24-hour systolic BP, ß, 0.16 [0.01-0.32] g/m2). CONCLUSIONS: Among individuals with central hypertension, spironolactone had beneficial effects in reducing LV mass. Secondary analyses showed that changes in LV mass were equally well associated with lower measured standard cuff BP and central BP. REGISTRATION: URL: https://www.anzctr.org.au/; Unique identifier: ACTRN12613000053729.

Noninvasive Techniques for Tracking Biological Aging of the Cardiovascular System: JACC Family Series.

Population aging is one of the most important demographic transformations of our time. Increasing the "health span"-the proportion of life spent in good health-is a global priority. Biological aging comprises molecular and cellular modifications over many years, which culminate in gradual physiological decline across multiple organ systems and predispose to age-related illnesses. Cardiovascular disease is a major cause of ill health and premature death in older people. The rate at which biological aging occurs varies across individuals of the same age and is influenced by a wide range of genetic and environmental exposures. The authors review the hallmarks of biological cardiovascular aging and their capture using imaging and other noninvasive techniques and examine how this information may be used to understand aging trajectories, with the aim of guiding individual- and population-level interventions to promote healthy aging.

Evolving Applications of Echocardiography in the Evaluation of Left Atrial and Right Ventricular Strain.

PURPOSE OF REVIEW: Speckle-tracking echocardiography (STE) can assess myocardial motion in non-LV chambers-including assessment of left atrial (LA) and right ventricular (RV) strain. This review seeks to highlight the diagnostic, prognostic, and clinical significance of these parameters in heart failure, atrial fibrillation (AF), diastolic dysfunction, pulmonary hypertension (PH), tricuspid regurgitation, and heart transplant recipients. RECENT FINDINGS: Impaired LA strain reflects worse LV diastolic function in individuals with and without HF, and this is associated with decreased exercise capacity. Initiating treatments targeting these functional aspects may enhance exercise capacity and potentially prevent heart failure (HF). Impaired LA strain also identifies patients with a high risk of AF, and this recognition may lead to preventive strategies. Impaired RV strain has significant clinical and prognostic implications across various clinical scenarios, including HF, PH, tricuspid regurgitation, or in heart transplant recipients. STE should not be limited to the assessment of deformation of the LV myocardium. The use of LA and RV strain is supported by a substantial evidence base, and these parameters should be used more widely.

Randomized Trial of a Selective Aldose Reductase Inhibitor in Patients With Diabetic Cardiomyopathy.

BACKGROUND: Progression to symptomatic heart failure is a complication of type 2 diabetes; heart failure onset in this setting is commonly preceded by deterioration in exercise capacity. OBJECTIVES: This study sought to determine whether AT-001, a highly selective aldose reductase inhibitor, can stabilize exercise capacity among individuals with diabetic cardiomyopathy (DbCM) and reduced peak oxygen uptake (Vo2). METHODS: A total of 691 individuals with DbCM meeting inclusion and exclusion criteria were randomized to receive placebo or ascending doses of AT-001 twice daily. Stratification at inclusion included region of enrollment, cardiopulmonary exercise test results, and use of sodium-glucose cotransporter 2 inhibitors or glucagon-like peptide-1 receptor agonists. The primary endpoint was proportional change in peak Vo2 from baseline to 15 months. Subgroup analyses included measures of disease severity and stratification variables. RESULTS: The mean age was 67.5 ± 7.2 years, and 50.4% of participants were women. By 15 months, peak Vo2 fell in the placebo-treated patients by -0.31 mL/kg/min (P = 0.005 compared to baseline), whereas in those receiving high-dose AT-001, peak Vo2 fell by -0.01 mL/kg/min (P = 0.21); the difference in peak Vo2 between placebo and high-dose AT-001 was 0.30 (P = 0.19). In prespecified subgroup analyses among those not receiving sodium-glucose cotransporter 2 inhibitors or glucagon-like peptide-1 receptor agonists at baseline, the difference between peak Vo2 in placebo vs high-dose AT-001 at 15 months was 0.62 mL/kg/min (P = 0.04; interaction P = 0.10). CONCLUSIONS: Among individuals with DbCM and impaired exercise capacity, treatment with AT-001 for 15 months did not result in significantly better exercise capacity compared with placebo. (Safety and Efficacy of AT-001 in Patients With Diabetic Cardiomyopathy [ARISE-HF]; NCT04083339).

Use of the energy waveform electrocardiogram to detect subclinical left ventricular dysfunction in patients with type 2 diabetes mellitus.

BACKGROUND: Recent guidelines propose N-terminal pro-B-type natriuretic peptide (NT-proBNP) for recognition of asymptomatic left ventricular (LV) dysfunction (Stage B Heart Failure, SBHF) in type 2 diabetes mellitus (T2DM). Wavelet Transform based signal-processing transforms electrocardiogram (ECG) waveforms into an energy distribution waveform (ew)ECG, providing frequency and energy features that machine learning can use as additional inputs to improve the identification of SBHF. Accordingly, we sought whether machine learning model based on ewECG features was superior to NT-proBNP, as well as a conventional screening tool-the Atherosclerosis Risk in Communities (ARIC) HF risk score, in SBHF screening among patients with T2DM. METHODS: Participants in two clinical trials of SBHF (defined as diastolic dysfunction [DD], reduced global longitudinal strain [GLS ≤ 18%] or LV hypertrophy [LVH]) in T2DM underwent 12-lead ECG with additional ewECG feature and echocardiography. Supervised machine learning was adopted to identify the optimal combination of ewECG extracted features for SBHF screening in 178 participants in one trial and tested in 97 participants in the other trial. The accuracy of the ewECG model in SBHF screening was compared with NT-proBNP and ARIC HF. RESULTS: SBHF was identified in 128 (72%) participants in the training dataset (median 72 years, 41% female) and 64 (66%) in the validation dataset (median 70 years, 43% female). Fifteen ewECG features showed an area under the curve (AUC) of 0.81 (95% CI 0.787-0.794) in identifying SBHF, significantly better than both NT-proBNP (AUC 0.56, 95% CI 0.44-0.68, p < 0.001) and ARIC HF (AUC 0.67, 95%CI 0.56-0.79, p = 0.002). ewECG features were also led to robust models screening for DD (AUC 0.74, 95% CI 0.73-0.74), reduced GLS (AUC 0.76, 95% CI 0.73-0.74) and LVH (AUC 0.90, 95% CI 0.88-0.89). CONCLUSIONS: Machine learning based modelling using additional ewECG extracted features are superior to NT-proBNP and ARIC HF in SBHF screening among patients with T2DM, providing an alternative HF screening strategy for asymptomatic patients and potentially act as a guidance tool to determine those who required echocardiogram to confirm diagnosis. Trial registration LEAVE-DM, ACTRN 12619001393145 and Vic-ELF, ACTRN 12617000116325.

A novel approach to accurately measuring the burden of hospitalisations for cardiovascular disease in people with diabetes: A pilot study.

AIM: To determine the reliability of hospital discharge codes for heart failure (HF), acute myocardial infarction (AMI) and stroke compared with adjudicated diagnosis, and to pilot a scalable approach to adjudicate records on a population-based sample. METHODS: A population-based sample of 685 people with diabetes admitted (1274 admissions) to one of three Australian hospitals during 2018-2020 were randomly selected for this study. All medical records were reviewed and adjudicated. RESULTS: Cardiovascular diseases were the most common primary reason for hospitalisation in people with diabetes, accounting for ~17% (215/1274) of all hospitalisations, with HF as the leading cause. ICD-10 codes substantially underestimated HF prevalence and had the lowest agreement with the adjudicated diagnosis of HF (Kappa = 0.81), compared with AMI and stroke (Kappa ≥ 0.91). While ICD-10 codes provided suboptimal sensitivity (72%) for HF, the performance was better for AMI (sensitivity 84%; specificity 100%) and stroke (sensitivity 85%; specificity 100%). A novel approach to screen possible HF cases only required adjudicating 8% (105/1274) of records, correctly identified 78/81 of HF admissions and yielded 96% sensitivity and 98% specificity. CONCLUSIONS: While ICD-10 codes appear reliable for AMI or stroke, a more complex diagnosis such as HF benefits from a two-stage process to screen for suspected HF cases that need adjudicating. The next step is to validate this novel approach on large multi-centre studies in diabetes.