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.

Association of biomarkers and risk scores with subclinical left ventricular dysfunction in patients with type 2 diabetes mellitus.

BACKGROUND: Subclinical LV dysfunction (LVD) identifies heart failure (HF) risk in type 2 diabetes mellitus (T2DM). We sought the extent to which clinical scores (ARIC-HF, WATCH-DM), natriuretic peptides (NTpBNP) and troponin (hs-TnT) were associated with subclinical LV dysfunction (LVD). These associations could inform the ability of these tests to identify which patients should undergo echocardiography. METHODS: Participants with T2DM were prospectively recruited from three community-based populations. ARIC-HF risk at 4 years and WATCH-DM scores were calculated from clinical data. NTpBNP and hs-TnT were measured using an electro-chemiluminescence assay. All underwent a comprehensive echocardiogram. We calculated the sensitivity and specificity of clinical scores and biomarkers to identify abnormal global longitudinal strain (GLS ≥ -16%)), diastolic function (E/e' ≥ 14 or e' < 8 cm/s), left atrial volume index (LAV > 34 ml/m2) and LV hypertrophy (LV mass index > 88 g/m2 (F) > 102 g/m2(M)). RESULTS: Of 804 participants (median age 69 years [inter-quartile range (IQR) 65-73], 36% female), clinical scores suggested significant HF risk (median ARIC-HF 8% [IQR 4-12]; WATCH-DM 10 points [IQR 8-12]), and the median NTpBNP was 50 pg/mL [IQR 25-101] and hs-TnT 9.6 pg/mL [IQR 6.8-13.6]. Abnormal GLS was present in 126 (17%), elevated E/e' in 114 (15%), impaired e' in 629 (78%), increased LAV in 351 (44%) and LV hypertrophy in 113 (14%). After adjustments for age, body-mass index, and renal function, each standard deviation increase in NTpBNP was associated with a GLS increase of 0.32 (p < 0.001) and hs-TnT increase by 0.26 (p < 0.001). Similar trends were observed with ARIC-HF (standardised ß = 0.22, p < 0.001) and WATCH-DM (standardised ß = 0.22, p < 0.001) in univariable analyses. However, none of the risk assessment tools provided satisfactory discrimination for abnormal GLS (AUC 63%), diastolic indices (e' AUC 54-61%) or LV mass (AUC 59-67%). At a sensitivity of 90%, there was an unacceptably low (< 50%) specificity. CONCLUSION: Although risk assessment based on clinical scores or biomarkers would be desirable to stratify HF risk in people with T2DM, they show a weak relationship with subclinical LVD.

Implementation of more sensitive cardiac troponin T assay in a state-wide health service.

AIMS: Explore the impact of deploying high-sensitivity (hs) cardiac troponin T (cTnT) assay across a state-wide health service. METHODS AND RESULTS: Presentations to emergency departments of six tertiary hospitals between January 2008 and August 2019 were included; standard cTnT assay was superseded by hs-cTnT in June 2011 without changing the reference range (≥30 ng/L reported as elevated), despite cTnT level of 30 ng/L being equivalent to ∼44 ng/L with hs-cTnT. Clinical outcomes were captured using state-wide linked health records. Interrupted time series analyses were used adjusted for seasonality and multiple co-morbidities using propensity score matching allowing for correlation within hospitals. In total, 614,847 presentations had ≥1 troponin measurement. Clinical ordering of troponin decreased throughout the study with no increase in elevated measurements amongst those tested with hs-cTnT. Small but statistically significant changes in index myocardial infarction (MI) diagnosis (-0.36%/year, 95%CI [confidence interval]:-0.48, -0.24,p < 0.001) and invasive coronary angiography (0.12%/year,95%CI:0, 0.24,p = 0.02) were seen, with no impact on death/MI at 30 days or 3-year survival in episodes of care (EOCs) with elevated cTnT after hs-cTnT implementation. Length of stay (LOS) was shorter among those with an elevated hs-cTnT (-4.44 h/year, 95%CI:-5.27, -3.60, p < 0.001). Non-elevated cTnT EOCs demonstrated shorter total LOS and improved 3-year survival (adjusted hazard ratio:0.90, 95%CI:0.83, 0.97,p = 0.008) although death/MI at 30 days was unchanged using hs-cTnT. CONCLUSION: Widespread implementation of hs-cTnT without altering clinical thresholds reported to clinicians provided significantly shorter LOS without a clinically significant impact on clinical outcomes. A safer cohort with non-elevated cTnT was identified by hs-cTnT compared to the standard cTnT assay.

Use of novel non-invasive techniques and biomarkers to guide outpatient management of fluid overload and reduce hospital readmission: systematic review and meta-analysis.

AIMS: Fluid congestion is a leading cause of hospital admission, readmission, and mortality in heart failure (HF). We performed a systematic review and meta-analysis to determine the effectiveness of an advanced fluid management programme (AFMP). The AFMP was defined as an intervention providing tailored diuretic therapy guided by intravascular volume assessment, in hospitalized patients or after discharge. The AFMP group was compared with patients who received standard care treatment. The aim of this systematic review and meta-analysis was to determine the effectiveness of an AFMP in improving patient outcomes. METHODS AND RESULTS: A systematic review of randomized controlled trials, case-control studies, and crossover studies using the terms 'heart failure', 'fluid management', and 'readmission' was conducted in PubMed, CINAHL, and Scopus up until November 2020. Studies reporting the association of an AFMP on readmission and/or mortality were included in our meta-analyses. Risk of bias was assessed in non-randomized studies using the Newcastle-Ottawa Scale. From 232 retrieved studies, 12 were included in the data synthesis. The 6040 patients in the included studies had a mean age of 72 ± 4 years and mean left ventricular ejection fraction of 39 ± 8%, there were slightly more men (n = 3022) than women, and the follow-up period was a mean of 4.8 ± 3.1 months. Readmission data were available in 5362 patients; of these, 1629 were readmitted. Mortality data were available in 5787 patients; of these, 584 died. HF patients who had an AFMP in hospital and/or after discharge had lower odds of all-cause readmission (odds ratio-OR 0.64 [95% confidence interval-CI 0.44, 0.92], P = 0.02) with moderate heterogeneity (I2  = 46.5) and lower odds of all-cause mortality (OR 0.82 [95% CI 0.69, 0.98], P = 0.03) with low heterogeneity (I2  = 0). The use of an AFMP was equally effective in reducing readmission and mortality regardless of age and follow-up duration. Effective pre-discharge diuresis was associated with significantly lower readmission odds (OR 0.43 [95% CI 0.26, 0.71], P = 0.001) compared with a fluid management plan as part of post-discharge follow-up. CONCLUSIONS: An effective AFMP is associated with improving readmission and mortality in HF. Our results encourage attainment of optimal volume status at discharge and prescription of optimal diuretic dose. Ongoing support to maintain euvolaemia and effective collaboration between healthcare teams, along with effective patient education and engagement, may help to reduce adverse outcomes in HF patients.

Role of microvascular dysfunction in left ventricular dysfunction in type 2 diabetes mellitus.

BACKGROUND: Although microvascular disease (mVD) has been linked to poor cardiovascular outcomes in diabetes mellitus, the contribution of mVD to diabetic cardiomyopathy (DC) is unexplored. We investigated whether LV systolic and diastolic dysfunction is associated with mVD in T2DM. METHODS: We recruited 32 asymptomatic patients with T2DM (age 71 ±â€¯4 years, 31% females) from a community-based population. All underwent a comprehensive echocardiogram at baseline including assessment of global longitudinal strain (GLS) and diastolic function. Adenosine stress perfusion on cardiac magnetic resonance imaging (CMR) was performed in all patients. Coronary sinus flow (CSF) was measured offline at rest and peak stress with coronary flow reserve (CFR) calculated as the ratio of global stress and rest CSF. RESULTS: Resting CSF was reduced in 15 (47%) compared to 4 (13%) with adenosine-stress (p = 0.023). Overall, CFR was observed to be reduced in the cohort (2.38 [IQR 2.20]). Abnormal CFR was not associated with diabetes duration of ≥10 years or poor glycaemic control. CFR was not associated with abnormal GLS (OR 1.04 [95% CI 0.49, 2.20], p = 0.93). However, a modest negative correlation was observed with e' and CFR (r = -0.49, p = 0.004). CONCLUSION: This pilot study did not show correlation between subclinical systolic dysfunction and a novel MRI biomarker of microvascular disease. However, there was a weak correlation with myocardial relaxation. Confirmation of these findings in larger studies is indicated.

Evolution of Myocardial Dysfunction in Asymptomatic Patients at Risk of Heart Failure.

OBJECTIVES: The determinants of changes in systolic and diastolic parameters in patients age >65 years, at risk of heart failure (HF), and with and without asymptomatic type 2 diabetes mellitus (T2DM) was assessed by echocardiography. The association between metformin and myocardial function was also assessed. BACKGROUND: The increasing prevalence of T2DM will likely further fuel the epidemic of HF. Understanding the development or progression of left ventricular (LV) dysfunction may inform effective measures for HF prevention. METHODS: A total of 982 patients with at least one HF risk factor (hypertension, obesity, or T2DM) were recruited from 2 community-based populations and divided into 2 groups: T2DM (n = 431, age 71 ± 4 years) and non-T2DM (n = 551, age 71 ± 5 years). Associations of metformin therapy were evaluated in the T2DM group. All underwent a comprehensive echocardiogram, including global longitudinal strain (GLS) and diastolic function (transmitral flow [E], annular velocity [e']) at baseline and follow-up (median 19 months [interquartile range: 17 to 26 months]). Comparisons were facilitated by propensity matching. RESULTS: A reduction in GLS was observed in the T2DM group (baseline -17.8 ± 2.6% vs. follow-up -17.4 ± 2.8%; p = 0.003), but not in the non-T2DM group (-18.7 ± 2.7% vs. -18.6 ± 3.0%; p = 0.41). Estimated LV filling pressures increased in both the T2DM group (p = 0.001) and the non-T2DM group (p = 0.04). Metformin-treated patients with T2DM did not increase estimated LV filling pressure (E/e' baseline 8.9 ± 2.7 vs. follow-up 9.1 ± 2.7; p = 0.485) or change e' (7.6 ± 1.5 cm/s vs. 7.6 ± 1.8 cm/s; p = 0.88). After propensity matching, metformin was associated with a smaller change in e' (ß = 0.58 [95% CI: 0.13 to 1.03]; p = 0.013) and E/e' (ß = -0.96 [95% CI: -1.66 to -0.26]; p = 0.007) but was not associated with a change in GLS (p = 0.46). CONCLUSIONS: Over 2 years, there is a worsening of GLS and LV filling pressures in asymptomatic diabetic patients with HF risk factors. Metformin use is associated with less deterioration of LV filling pressures and myocardial relaxation but had no association with systolic function.

Metformin treatment in heart failure with preserved ejection fraction: a systematic review and meta-regression analysis.

BACKGROUND: Observational series suggest a mortality benefit from metformin in the heart failure (HF) population. However, the benefit of metformin in HF with preserved ejection fraction (HFpEF) has yet to be explored. We performed a systematic review and meta-analysis to identify whether variation in EF impacts mortality outcomes in HF patients treated with metformin. METHODS: MEDLINE and EMBASE were searched up to October 2019. Observational studies and randomised trials reporting mortality in HF patients and the proportion of patients with an EF > 50% at baseline were included. Other baseline variables were used to assess for heterogeneity in treatment outcomes between groups. Regression models were used to determine the interaction between metformin and subgroups on mortality. RESULTS: Four studies reported the proportion of patients with a preserved EF and were analysed. Metformin reduced mortality in both preserved or reduced EF after adjustment with HF therapies such as angiotensin converting enzyme inhibitors (ACEi) and beta-blockers (ß = - 0.2 [95% CI - 0.3 to - 0.1], p = 0.02). Significantly greater protective effects were seen with EF > 50% (p = 0.003). Metformin treatment with insulin, ACEi and beta-blocker therapy were also shown to have a reduction in mortality (insulin p = 0.002; ACEi p < 0.001; beta-blocker p = 0.017), whereas female gender was associated with worse outcomes (p < 0.001). CONCLUSIONS: Metformin treatment is associated with a reduction in mortality in patients with HFpEF.

Invasive management of acute coronary syndrome: Interaction with competing risks.

BACKGROUND: The aim of this study was to characterise the interaction between ACS- and non-ACS-risk on the benefits of invasive management in patients presenting with acute coronary syndrome (ACS). METHODS: Consecutive patients admitted to a tertiary hospital's Cardiac Care Unit in the months of July-December, 2003-2011 with troponin elevation (>30 ng/L) were included. "ACS-specific-risk" was estimated using the GRACE score and "non-ACS-risk" was estimated using the Charlson-Comorbidity-Index (CCI). Inverse-probability-of-treatment weighting was used to adjust for baseline differences between patients who did or did not receive invasive management. A multivariable flexible parametric model was used to characterise the time-varying hazard. RESULTS: In total, 3057 patients were included with a median follow-up of 9.0 years. Based on CCI, 1783 patients were classified as 'low-non-ACS risk' (CCI ≤ 1; invasive management 81%; 12-month mortality 5%), 820 as 'medium-non-ACS risk' (CCI 2-3; invasive management 68%; 12-month mortality 13%), and 468 as 'high-non-ACS risk' (CCI ≥ 4; invasive management 47%; 12-month mortality 29%). After adjustment, invasive management was associated with a significant reduction in one-year overall-mortality in the 'low-risk' and 'medium-risk' groups (HR = 0.38, 95%CI:0.26-0.56; HR = 0.46, 95%CI:0.32-0.67); but not in the 'high-risk' group (HR = 1.02, 95%CI:0.67-1.56). The absolute benefit of invasive management was greatest with higher baseline ACS-risk, with a non-linear interaction between ACS- and non-ACS-risk. CONCLUSIONS: There is a complex interaction between ACS- and non-ACS-risk on the benefit of invasive management. These results highlight the need to develop robust methods to objectively quantify risk attributable to non-ACS comorbidities in order to make informed decisions regarding the use of invasive management in individuals with numerous comorbidities.

Has invasive management for acute coronary syndromes become more 'risk-appropriate': pooled results of five Australian registries.

Background: Despite being recommended in acute coronary syndrome (ACS) guidelines, the use of invasive management within specific risk groups continues to be debated. This study examines the change in the use of invasive management in ACS by patient risk and the associated change in mortality within Australia over the last 17 years. Methods: Pooled cohorts derived from five ACS registries (ACACIA, CONCORDANCE, GRACE, Snapshot-ACS, and Predict) spanned from 1999 to 2015. After excluding patients without a final diagnosis of ACS (n = 4460), enrolled outside Australia (n = 1477) and without an enrolling year (n = 4), 15 912 patients were analysed. Data was stratified across three time periods (1999-2004, 2005-2009, and 2010-2015) using clinical risk characteristics (age, ACS diagnosis, biomarker elevation, and GRACE score) to monitor change in practice. Results: Over the 17-year period, the use of invasive management increased (4073/6863 (59.3%) cases [1999-2009] vs. 6670/8706 (76.6%) cases [2010-2015]). Invasive management accounted for improvements in mortality in intermediate- and high-risk groups (intermediate risk: 14% (95% CI 1-66%) [1999-2009] vs. 49% (95% CI 2-59%) [2010-2015]; high risk: 24% (95% CI 6-42%) [1999-2009] vs. 48% (95% CI 19-76%) [2010-2015]). Patients receiving no angiography compared with interventional management had worse outcomes (1999-2004 1.55 HR [95% CI 1.36-1.80], P < 0.0001 vs. 2010-2015 1.90 HR [95% CI 1.45-2.51], P < 0.0001). Conclusions: Clinical practice in ACS has changed over the last 17 years with positive outcomes seen with invasive management among high-risk patients. Unfortunately, a considerable burden of mortality remains in patients managed medically, highlighting a need for more focused strategies that improve care and outcomes in this group.

Rational clinical evaluation of suspected acute coronary syndromes: The value of more information.

OBJECTIVE: Many meta-analyses have provided synthesised likelihood ratio data to aid clinical decision-making. However, much less has been published on how to safely combine clinical information in practice. We aimed to explore the benefits and risks of pooling clinical information during the ED assessment of suspected acute coronary syndrome. METHODS: Clinical information on 1776 patients was collected within a randomised trial conducted across five South Australian EDs between July 2011 and March 2013. Bayes theorem was used to calculate patient-specific post-test probabilities using age- and gender-specific pre-test probabilities and likelihood ratios corresponding to the presence or absence of 18 clinical factors. Model performance was assessed as the presence of adverse cardiac outcomes among patients theoretically discharged at a post-test probability less than 1%. RESULTS: Bayes theorem-based models containing high-sensitivity troponin T (hs-troponin) outperformed models excluding hs-troponin, as well as models utilising TIMI and GRACE scores. In models containing hs-troponin, a plateau in improving discharge safety was observed after the inclusion of four clinical factors. Models with fewer clinical factors better approximated the true event rate, tended to be safer and resulted in a smaller standard deviation in post-test probability estimates. CONCLUSIONS: We showed that there is a definable point where additional information becomes uninformative and may actually lead to less certainty. This evidence supports the concept that clinical decision-making in the assessment of suspected acute coronary syndrome should be focused on obtaining the least amount of information that provides the highest benefit for informing the decisions of admission or discharge.