Eligibility and Cost-Utility Analysis of Dapagliflozin in Patients with Heart Failure Across the Whole Spectrum of Ejection Fraction in South Korea.

BACKGROUND: The DAPA-HF and DELIVER trials demonstrated the clinical benefits of dapagliflozin in heart failure (HF) patients across the entire ejection fraction (EF) spectrum. However, further investigation is needed for the real-world application of dapagliflozin in HF patients. This study examines the proportion of real-world HF patients eligible for dapagliflozin and evaluates the cost-effectiveness of adding dapagliflozin to current HF therapy. METHODS: Data from the nationwide prospective registry, the Korean Acute Heart Failure (KorAHF) registry, were used to determine dapagliflozin eligibility based on the enrollment criteria of the DAPA-HF/DELIVER trials. A cost-utility analysis was conducted using a Markov model to assess the cost-effectiveness of dapagliflozin by comparing it to the standard of care. RESULTS: Out of 5178 KorAHF patients, 48.7% met the enrollment criteria of the DAPA-HF/DELIVER trials, while 89.5% met the label criteria (US Food and Drug Administration, European Medicines Agency, and Korean Ministry of Food and Drug Safety). Eligibility was highest among HF patients with preserved EF (55.3% vs. HF with mildly reduced EF and HF with reduced EF 46.4%). Dapagliflozin proved to be cost-effective, with an incremental cost-effectiveness ratio (ICER) of 4557 US dollar (US$) per quality-adjusted life year, which falls below the US$18,182 willingness-to-pay threshold. The cost-effectiveness benefit was more pronounced in patients with a left ventricular EF (LVEF) ≤ 40% (ICER US$3279 for LVEF ≤ 40% vs. US$8383 for LVEF > 40%). CONCLUSIONS: Discrepancies in dapagliflozin eligibility were observed between real-world data and clinical trial results. The addition of dapagliflozin to HF therapy proved to be highly cost-effective across the entire EF spectrum.

Real-World Eligibility and Cost-Effectiveness Analysis of Empagliflozin for Heart Failure in Korea.

BACKGROUND: The US Food and Drug Administration (FDA) and European Medicines Agency (EMA) approved empagliflozin for reducing cardiovascular mortality and heart failure (HF) hospitalization in patients with both HF with reduced ejection fraction (HFrEF) and HF with preserved ejection fraction (HFpEF). However, limited data are available on the generalizability of empagliflozin to clinical practice. Therefore, we evaluated real-world eligibility and potential cost-effectiveness based on a nationwide prospective HF registry. METHODS: A total of 3,108 HFrEF and 2,070 HFpEF patients from the Korean Acute Heart Failure (KorAHF) registry were analyzed. Eligibility was estimated by inclusion and exclusion criteria of EMPagliflozin outcomE tRial in Patients With chrOnic heaRt Failure With Reduced Ejection Fraction (EMPEROR-Reduced) and EMPagliflozin outcomE tRial in Patients With chrOnic heaRt Failure With Preserved Ejection Fraction (EMPEROR-Preserved) trials and by FDA & EMA label criteria. The cost-utility analysis was done using a Markov model to project the lifetime medical cost and quality-adjusted life year (QALY). RESULTS: Among the KorAHF patients, 91.4% met FDA & EMA label criteria, while 44.7% met the clinical trial criteria. The incremental cost-effectiveness ratio of empagliflozin was calculated at US$6,764 per QALY in the overall population, which is far below a threshold of US$18,182 per QALY. The cost-effectiveness benefit was more evident in patients with HFrEF (US$5,012 per QALY) than HFpEF (US$8,971 per QALY). CONCLUSION: There is a large discrepancy in real-world eligibility for empagliflozin between FDA & EMA labels and clinical trial criteria. Empagliflozin is cost-effective in HF patients regardless of ejection fraction in South Korea health care setting. The efficacy and safety of empagliflozin in real-world HF patients should be further investigated for a broader range of clinical applications. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01389843.

Metformin treatment is associated with improved survival in diabetic patients hospitalized with acute heart failure: A prospective observational study using the Korean acute heart failure registry data.

AIMS: Although the hypothesis that metformin is beneficial for patients with diabetes and heart failure (HF) has been steadily raised, there is limited data on metformin use in patients with acute HF. We analyzed the association of metformin on all-cause mortality in hospitalized patients with type 2 diabetes and acute HF. METHODS: The Korean Acute Heart Failure registry prospectively enrolled patients hospitalized for acute HF from 2011 to 2014. Among this cohort, we analyzed patients with diabetes with baseline estimated glomerular filtration rate (eGFR) of 30 ml/min/1.73 m2 or more. We analyzed the all-cause mortality and re-hospitalization for HF within 1 year after discharge. Inverse probability treatment weighting method was used to adjust baseline differences on metformin treatment. RESULTS: The study analyzed data from 1,309 patients with HF and diabetes (mean age 69 years, 56 % male). Among them, 613 (47 %) patients were on metformin at admission. During the median follow-up period of 11 months, 132 (19 %) and 74 (12 %) patients not receiving and receiving metformin treatment died, respectively. The mortality rate was lower in metformin users than in non-users (hazard ratio 0.616 [0.464-0.819] P<0.001). After adjustment, metformin was significantly associated with a lower risk for the mortality (hazard ratio 0.677 [0.495-0.928] P=0.015). In subgroup analyses, this association remains significant irrespective of baseline kidney function (eGFR <60 or ≥60 ml/min/1.73 m2, P-for-interaction=0.176) or left ventricular ejection fraction (<40 %, 40-49 %, or ≥50 %, P-for-interaction=0.224). CONCLUSIONS: Metformin treatment at the time of admission was associated with a lower risk for 1-year mortality in patients with diabetes, hospitalized for acute HF.

In-hospital glycemic variability and all-cause mortality among patients hospitalized for acute heart failure.

BACKGROUND: High glycemic variability (GV) is a poor prognostic marker in cardiovascular diseases. We aimed to investigate the association of GV with all-cause mortality in patients with acute heart failure (HF). METHODS: The Korean Acute Heart Failure registry enrolled patients hospitalized for acute HF from 2011 to 2014. Blood glucose levels were measured at the time of admission, during hospitalization, and at discharge. We included those who had 3 or more blood glucose measurements in this study. Patients were divided into two groups based on the coefficient of variation (CoV) as an indicator of GV. Among survivors of the index hospitalization, we investigated all-cause mortality at 1 year after discharge. RESULTS: The study analyzed 2,617 patients (median age, 72 years; median left-ventricular ejection fraction, 36%; 53% male). During the median follow-up period of 11 months, 583 patients died. Kaplan-Meier curve analysis revealed that high GV (CoV > 21%) was associated with lower cumulative survival (log-rank P < 0.001). Multivariate Cox proportional analysis showed that high GV was associated with an increased risk of 1-year (HR 1.56, 95% CI 1.26-1.92) mortality. High GV significantly increased the risk of 1-year mortality in non-diabetic patients (HR 1.93, 95% CI 1.47-2.54) but not in diabetic patients (HR 1.19, 95% CI 0.86-1.65, P for interaction = 0.021). CONCLUSIONS: High in-hospital GV before discharge was associated with all-cause mortality within 1 year, especially in non-diabetic patients with acute HF.

Impact of Cardiac Troponin Elevation on Mortality of Patients with Acute Heart Failure: Insights from the Korea Acute Heart Failure (KorAHF) Registry.

We aimed to conduct the largest study evaluating the impact of cardiac troponin (TnI) status on mid- and long-term mortality in patients admitted for acute heart failure (AHF) as compared between patients with ischemic (IHF) vs. non-ischemic heart failure (non-IHF). Among 5625 patients from the Korea Acute Heart Failure (KorAHF) registry, 4396 eligible patients with TnI measurement were analyzed. The patients were included on admission with the diagnosis of AHF, and TnI level was measured on the day of admission. A TnI value of <0.05 ng/mL was considered normal. The patients were divided into four groups according to the etiology of heart failure and the status of TnI: non-IHF with normal TnI (n = 1009) vs. non-IHF with elevated TnI (n = 1665) vs. IHF with normal TnI (n = 258) vs. IHF with elevated TnI (n = 1464). The primary outcome was death from all causes according to the etiology (non-IHF vs. IHF) and TnI elevation during the entire follow-up period of 784 days (IQR 446−1116). Elevation of TnI was observed in 71.2% of all patients with AHF. Patients with IHF had higher all-cause mortality compared to those with non-IHF. Elevated TnI was associated with higher 90-day and post-90-day mortality in the non-IHF group. IHF as compared to non-IHF and elevation of TnI were independent predictors of mortality also in the adjustment analysis. In the IHF group, however, elevated TnI had a higher mortality with only 90-day follow-up (18.6% vs. 25.9%, log-rank p < 0.001), not in the post-90-day follow-up (31.1% vs. 32.5%, log-rank p = 0.799). In conclusion, elevated TnI in patients with heart failure is associated with increased all-cause mortality regardless of the etiology of HF. Elevation of TnI was associated to a higher post-90 day mortality in patients with non-IHF but not in patients with IHF.

Corrigendum: The Prescription Characteristics, Efficacy and Safety of Spironolactone in Real-World Patients With Acute Heart Failure Syndrome: A Prospective Nationwide Cohort Study.

[This corrects the article DOI: 10.3389/fcvm.2022.791446.].

Artificial intelligence assessment for early detection and prediction of renal impairment using electrocardiography.

PURPOSE: Although renal failure is a major healthcare burden globally and the cornerstone for preventing its irreversible progression is an early diagnosis, an adequate and noninvasive tool to screen renal impairment (RI) reliably and economically does not exist. We developed an interpretable deep learning model (DLM) using electrocardiography (ECG) and validated its performance. METHODS: This retrospective cohort study included two hospitals. We included 115,361 patients who had at least one ECG taken with an estimated glomerular filtration rate measurement within 30 min of the index ECG. A DLM was developed using 96,549 ECGs of 55,222 patients. The internal validation included 22,949 ECGs of 22,949 patients. Furthermore, we conducted an external validation with 37,190 ECGs of 37,190 patients from another hospital. The endpoint was to detect a moderate to severe RI (estimated glomerular filtration rate < 45 ml/min/1.73m2). RESULTS: The area under the receiver operating characteristic curve (AUC) of a DLM using a 12-lead ECG for detecting RI during the internal and external validation was 0.858 (95% confidence interval 0.851-0.866) and 0.906 (0.900-0.912), respectively. In the initial evaluation of 25,536 individuals without RI patients whose DLM was defined as having a higher risk had a significantly higher chance of developing RI than those in the low-risk group (17.2% vs. 2.4%, p < 0.001). The sensitivity map indicated that the DLM focused on the QRS complex and T-wave for detecting RI. CONCLUSION: The DLM demonstrated high performance for RI detection and prediction using 12-, 6-, single-lead ECGs.

Prognostic Impact and Predictors of New-Onset Atrial Fibrillation in Heart Failure.

Background: The prognostic impact and predictors of NOAF in HF patients are not fully elucidated. This study aims to determine whether new-onset atrial fibrillation (NOAF) affects patient outcome and investigate predictors of atrial fibrillation (AF) in acute heart failure (HF) patients using real-world data. Methods: The factors associated with NOAF in 2894 patients with sinus rhythm (SR) enrolled in the Korean Acute Heart Failure (KorAHF) registry were investigated. Survival was analyzed using AF as a time-dependent covariate. Relevant predictors of NOAF were analyzed using multivariate proportional hazards models. Results: Over 27.4 months, 187 patients developed AF. The median overall survival time was over 48 and 9.9 months for the SR and NOAF groups, respectively. Cox regression analysis with NOAF as a time-dependent covariate showed a higher risk of death among patients with NOAF. Multivariate Cox modeling showed that age, worsening HF, valvular heart disease (VHD), loop diuretics, lower heart rate, larger left atrium (LA) diameter, and elevated creatinine levels were independently associated with NOAF. Risk score indicated the number of independent predictors. The incidence of NOAF was 2.9%, 9.4%, and 21.8% in the low-risk, moderate-risk, and high-risk groups, respectively (p < 0.001). Conditional inference tree analysis identified worsening HF, heart rate, age, LA diameter, and VHD as discriminators. Conclusions: NOAF was associated with decreased survival in acute HF patients with SR. Age, worsening HF, VHD, loop diuretics, lower heart rate, larger LA diameter, and elevated creatinine could independently predict NOAF. This may be useful to risk-stratify HF patients at risk for AF.

The Prescription Characteristics, Efficacy and Safety of Spironolactone in Real-World Patients With Acute Heart Failure Syndrome: A Prospective Nationwide Cohort Study.

Background: Randomized clinical trials of spironolactone showed significant mortality reduction in patients with heart failure with reduced ejection fraction. However, its role in acute heart failure syndrome (AHFS) is largely unknown. Aim: To investigate the prescription characteristics, efficacy and safety of spironolactone in real-world patients with AHFS. Methods: 5,136 AHFS patients who survived to hospital discharge using a nationwide prospective registry in Korea were analyzed. The primary efficacy outcome was 3-year all-cause mortality. Results: Spironolactone was prescribed in 2,402 (46.8%) at discharge: <25 mg in 890 patients (37.1%), ≥25 mg, and <50 mg in 1,154 patients (48.0%), and ≥50 mg in 358 patients (14.9%). Patients treated with spironolactone had a lower proportion of chronic renal failure and renal replacement therapy during hospitalization and had lower serum creatinine level than those who did not. In overall patients, 3-year mortality was not different in both groups (35.9 vs. 34.5%, P = 0.279). The incidence of renal injury and hyperkalemia was 2.2% and 4.3%, respectively, at the first follow-up visit. The treatment effect of spironolactone on mortality was different across subpopulations according to LVEF. The use of spironolactone was associated with a significant reduction in 3-year morality in patients with LVEF ≤ 26% (33.8 vs. 44.3%, P < 0.001; adjusted HR 0.79, 95% CI 0.64-0.97, P = 0.023), but not in patients with LVEF > 26%. Conclusions: Although spironolactone was frequently used at lower doses in real-world practice, use of spironolactone significantly reduced 3-year mortality in patients with severely reduced LVEF with acceptable safety profile. However, our findings remain prone to various biases and further prospective randomized controlled studies are needed to confirm these findings.

Real-world eligibility for vericiguat in decompensated heart failure with reduced ejection fraction.

Aims In 2021, vericiguat was approved by the US Food and Drug Administration (FDA) and the European Commission (EC) for reducing cardiovascular mortality and heart failure (HF) hospitalizations in patients with HF with reduced ejection fraction (HFrEF) based on the Vericiguat Global Study in Subjects with Heart Failure with Reduced Ejection Fraction (VICTORIA) trial. However, there has been no report for characterizing the generalizability of vericiguat to real-world clinical practice. Methods and results The Korean Acute Heart Failure (KorAHF) registry is a multicentre prospective cohort study. A total of 5625 patients who were admitted for HF decompensation were consecutively enrolled. We excluded the patients without left ventricular ejection fraction (LVEF) quantification, patients with LVEF > 45%, patients with in-hospital death or urgent heart transplantation, and patients without natriuretic peptide measurement. Among a total of 3014 enrolled patients, there were 21.9% patients with lower systolic blood pressure (SBP) (<100 mmHg) and 20.1% patients without elevated natriuretic peptide. Regarding chronic kidney disease (CKD) status, 5.1% patients had CKD Stage V [estimated glomerular filtration rate (eGFR) <15 mL/min/1.73 m2 ] and 11.8% patients had CKD Stage IV (15 ≤ eGFR < 30 mL/min/1.73 m2 ). When we analysed these criteria sequentially, 21.9% were excluded from lower SBP, 15.9% were excluded from elevated natriuretic peptide, and 4.2% were excluded from advanced CKD Stage V (9.6% for CKD Stages IV and V). Among the KorAHF registry patients, we found two main reasons for not meeting the inclusion criteria of the VICTORIA trial such as low SBP and non-elevated natriuretic peptide. Conclusions Among the Korean hospitalized HFrEF patients, 94.9% met the FDA/EC label criteria, while 58% met the inclusion criteria of the VICTORIA trial. Our findings suggest the need for better strategies to integrate up-to-date HF treatment in a real-world HF population, especially decompensated HF patients with low SBP and non-elevated natriuretic peptide.

A dose-response relationship of renin-angiotensin system blockers and beta-blockers in patients with acute heart failure syndrome: a nationwide prospective cohort study.

AIMS: It remains unclear if patients with acute heart failure syndrome (AHFS) need to reach the maximally tolerated doses of renin-angiotensin system blockers (RASBs) or beta-blockers (BBs) to obtain a survival benefit. This study evaluated the dose-response relationship between RASBs or BBs and survival in AHFS patients. METHODS AND RESULTS: In total, 5331 patients in the Korean Acute Heart Failure registry were analysed based on the doses of RASBs and BBs at discharge. In AHFS patients, RASB use at discharge was associated with a significant reduction in all-cause mortality risk. This effect was dose-dependent for heart failure with reduced ejection fraction (HFrEF) but did not attain statistical significance for heart failure with preserved ejection fraction (HFpEF). BB use at discharge was associated with reduced all-cause mortality in HFrEF patients but not in HFpEF patients. In an additional analysis of 4613 patients with dosage information at the first post-discharge follow-up visit, a significantly higher mortality risk was associated with the maintenance or withdrawal of RASBs compared with up-titrating the dose in HFrEF patients. CONCLUSION: Using RASBs or BBs at discharge was associated with improved survival. A dose-response relationship between RASBs and all-cause mortality was evident in AHFS patients with a reduced ejection fraction but not BBs. It is important to initiate and up-titrate RASBs to the maximally tolerated dose in AHFS patients during the transition period, especially for patients with a reduced ejection fraction.

Determinants of left ventricular function improvement for cardiac resynchronization therapy candidates.

AIMS: A waiting period of more than 3 months is recommended for patients before undergoing cardiac resynchronization therapy (CRT). However, due to an anticipated high mortality rate, early implementation of CRT might be beneficial for some patients. We aimed to evaluate the rate and the probability of left ventricular (LV) function improvement and their predictors in patients with heart failure (HF) with indications for CRT. METHODS AND RESULTS: From March 2011 to February 2014, a total of 5625 hospitalized patients for acute HF were consecutively enrolled in 10 tertiary hospitals. Among them, we analysed 1792 patients (mean age 63.96 ± 15.42 years, female 63.1%) with left ventricular ejection fraction (LVEF) ≤ 35% at the baseline echocardiography and divided them into three groups: 144 with left bundle branch block (LBBB), 136 with wide QRS complexes without LBBB, and 1512 not having these findings (control). We compared and analysed these three groups for improvement of LV function at follow-up echocardiography. In patients who met CRT indications (patients with LBBB or wide QRS complexes without LBBB), logistic regression was performed to identify risk factors for no improvement of LV. No improvement of LV was defined as LVEF ≤ 35% at follow-up echocardiography or the composite adverse outcomes: death, heart transplantation, extracorporeal membrane oxygenation, or use of a ventricular assist device before follow-up echocardiography. A classification tree was established using the binary recursive partitioning method to predict the outcome of patients who met CRT indications. In a median follow-up of 11 months, LVEF improvement was observed in 24.3%, 15.4%, and 40.5% of patients with LBBB, wide QRS complexes without LBBB, and control, respectively. Patients meeting CRT indications had higher 3 month mortality rates than the control (24.6% vs. 17.7%, P = 0.002). Multivariable logistic regression analysis revealed that large LV end-systolic dimension [odds ratio (OR) 1.10, 95% confidence interval (CI) 1.05-1.15, P < 0.001], low LVEF (OR 0.92, 95% CI 0.87-0.98, P = 0.006), diabetes requiring insulin (OR 6.49, 95% CI 2.53-19.33, P < 0.001), and suboptimal medical therapy (OR 6.85, 95% CI 3.21-15.87, P < 0.001) were significant factors predictive of no improvement. A decision tree analysis was consistent with these results. CONCLUSIONS: Patients with CRT indications had higher mortality during their follow-up compared with control. LV function improvement was rare in this population, especially when they had some risk factors. These results suggest that the uniform waiting period before CRT implantation could be reconsidered and individualized.

C-reactive protein and statins in heart failure with reduced and preserved ejection fraction.

Background: High C-reactive protein (CRP) levels are associated with poor outcomes of heart failure (HF), and statins are known to reduce CRP levels. We investigated the prognostic value of CRP and statin in patients with HF with reduced and preserved ejection fraction (EF). Methods: Altogether, 3,831 patients from the Korean Acute Heart Failure registry were included and stratified according to the tertiles of CRP levels (T1: CRP < 0.30 mg/dL, T2: 0.30-1.14 mg/dL, and T3: CRP > 1.14 mg/dL). HF with reduced EF (HFrEF), HF with mildly reduced EF (HFmrEF), and HF with preserved EF (HFpEF) were defined as left ventricular ejection fraction (LVEF) ≤ 40%, 41-49%, ≥50%, respectively. The primary endpoints were all-cause, in-hospital, and post-discharge mortality. Results: No significant correlation was observed between CRP levels and LVEF (r = 0.02, P = 0.131). The prevalence of risk factors increased gradually from T1 to T3 in both the types of HF. Overall, 139 (3.6%) and 1,269 (34.4%) patients died during the index admission and follow-up (median: 995 days), respectively. After adjustment, each increase in the CRP tertiles was independently associated with in-hospital mortality (HFrEF: OR 1.58 and 95% CI 1.09-2.30, HFmrEF: OR 1.51 and 95% CI 0.72-3.52, and HFpEF: OR 2.98, 95% CI 1.46-6.73) and post-discharge mortality (HFrEF: HR 1.20, 95% CI 1.08-1.33, HFmrEF: HR 1.38 and 95% CI 1.12-1.70, and HFpEF: HR 1.37, 95% CI 1.02-1.85). In only patients with LVEF > 40% with highest CRP tertile, statin-users showed better survival trend than those without statins. Conclusion: CRP is an excellent prognostic marker for HFrEF, HFmrEF, and HFpEF, implying that the neurohumoral and inflammatory pathways might be independent pathways. Statins may be beneficial in HF patients with increased CRP levels. Clinical trial registration: [https://clinicaltrials.gov/], identifier [NCT013 89843].

Integrated Analysis of Two Probucol Trials for the Secondary Prevention of Atherosclerotic Cardiovascular Events: PROSPECTIVE and IMPACT.

AIMS: In this study, we integrated two randomized control trials, PROSPECTIVE and IMPACT, to address the effect of probucol on cerebrocardiovascular events and carotid intima-media thickness (IMT) in Japanese, Korean, and Chinese patients with coronary artery disease (CAD). METHODS: A total of 1,025 patients from the PROSPECTIVE and IMPACT studies were enrolled. The time to the first major adverse cerebrocardiovascular event, in addition to carotid IMT and lipid levels, was compared between the control and probucol groups. RESULTS: In the integrated analysis, the adjusted hazard ratio (HR) and 95% confidence interval (CI) were 0.67 and 0.44-1.03, respectively, indicating a tendency to show the effect of probucol on cerebrocardiovascular events in secondary prevention. We also found no significant differences between the control and probucol groups in the mean IMT of the carotid arteries and its changes. However, we found a significant decrease in cerebrocardiovascular events in patients with reduced levels of HDL cholesterol (HDL-C) (≥ 6.25 mg/dL) compared with those with levels ＜6.25 mg/dL (p=0.024), without any increase in adverse events such as severe ventricular arrhythmias. CONCLUSION: We demonstrated a marginal effect of probucol on cerebrocardiovascular events in Asian patients with CAD, with reasonable safety profiles. A larger study may be needed to support the effect of probucol for cardiovascular prevention.

Impact of successful restoration of sinus rhythm in patients with atrial fibrillation and acute heart failure: Results from the Korean Acute Heart Failure registry.

BACKGROUND: Restoring and maintaining sinus rhythm (SR) in patients with atrial fibrillation (AF) failed to show superior outcomes over rate control strategies in prior randomized trials. However, there is sparse data on their outcomes in patients with acute heart failure (AHF). METHODS: From December 2010 to February 2014, 5,625 patients with AHF from 10 tertiary hospitals were enrolled in the Korean Acute Heart Failure registry, including 1,961 patients whose initial electrocardiogram showed AF. Clinical outcomes of patients who restored SR by pharmacological or electrical cardioversion (SR conversion group, n = 212) were compared to those of patients who showed a persistent AF rhythm (AF persistent group, n = 1,662). RESULTS: All-cause mortality both in-hospital and during the follow-up (median 2.5 years) were significantly lower in the SR conversion group than in the AF persistent group after adjustment for risk factors (adjusted hazard ratio [HR]; 95% confidence interval [CI] = 0.26 [0.08-0.88], p = 0.031 and 0.59 [0.43-0.82], p = 0.002, for mortality in-hospital and during follow-up, respectively). After 1:3 propensity score matching (SR conversion group = 167, AF persistent group = 501), successful restoration of SR was associated with lower all-cause mortality (HR [95% CI] = 0.68 [0.49-0.93], p = 0.015), heart failure rehospitalization (HR [95% CI] = 0.66 [0.45-0.97], p = 0.032), and composite of death and heart failure rehospitalization (HR [95% CI] = 0.66 [0.51-0.86], p = 0.002). CONCLUSIONS: Patients with AHF and AF had significantly lower mortality in-hospital and during follow-up if rhythm treatment for AF was successful, underscoring the importance of restoring SR in patients with AHF.

Detection and classification of arrhythmia using an explainable deep learning model.

BACKGROUND: Early detection and intervention is the cornerstone for appropriate treatment of arrhythmia and prevention of complications and mortality. Although diverse deep learning models have been developed to detect arrhythmia, they have been criticized due to their unexplainable nature. In this study, we developed an explainable deep learning model (XDM) to classify arrhythmia, and validated its performance using diverse external validation data. METHODS: In this retrospective study, the Sejong dataset comprising 86,802 electrocardiograms (ECGs) was used to develop and internally variate the XDM. The XDM based on a neural network-backed ensemble tree was developed with six feature modules that are able to explain the reasons for its decisions. The model was externally validated using data from 36,961 ECGs from four non-restricted datasets. RESULTS: During internal and external validation of the XDM, the average area under the receiver operating characteristic curves (AUCs) using a 12­lead ECG for arrhythmia classification were 0.976 and 0.966, respectively. The XDM outperformed a previous simple multi-classification deep learning model that used the same method. During internal and external validation, the AUCs of explainability were 0.925-0.991. CONCLUSION: Our XDM successfully classified arrhythmia using diverse formats of ECGs and could effectively describe the reason for the decisions. Therefore, an explainable deep learning methodology could improve accuracy compared to conventional deep learning methods, and that the transparency of XDM can be enhanced for its application in clinical practice.

Predicting survival in heart failure: a risk score based on machine-learning and change point algorithm.

OBJECTIVE: Machine learning (ML) algorithm can improve risk prediction because ML can select features and segment continuous variables effectively unbiased. We generated a risk score model for mortality with ML algorithms in East-Asian patients with heart failure (HF). METHODS: From the Korean Acute Heart Failure (KorAHF) registry, we used the data of 3683 patients with 27 continuous and 44 categorical variables. Grouped Lasso algorithm was used for the feature selection, and a novel continuous variable segmentation algorithm which is based on change-point analysis was developed for effectively segmenting the ranges of the continuous variables. Then, a risk score was assigned to each feature reflecting nonlinear relationship between features and survival times, and an integer score of maximum 100 was calculated for each patient. RESULTS: During 3-year follow-up time, 32.8% patients died. Using grouped Lasso, we identified 15 highly significant independent clinical features. The calculated risk score of each patient ranged between 1 and 71 points with a median of 36 (interquartile range: 27-45). The 3-year survival differed according to the quintiles of the risk score, being 80% and 17% in the 1st and 5th quintile, respectively. In addition, ML risk score had higher AUCs than MAGGIC-HF score to predict 1-year mortality (0.751 vs. 0.711, P < 0.001). CONCLUSIONS: In East-Asian patients with HF, a novel risk score model based on ML and the new continuous variable segmentation algorithm performs better for mortality prediction than conventional prediction models. CLINICAL TRIAL REGISTRATION: Unique identifier: INCT01389843 https://clinicaltrials.gov/ct2/show/NCT01389843 .

Heart failure and atrial fibrillation: tachycardia-mediated acute decompensation.

AIMS: Tachycardia is a reversible event that may cause hemodynamic decompensation but may not necessarily cause direct damages to the myocardium. To evaluate the clinical outcomes of patients with heart failure (HF) and atrial fibrillation (AF), whose acute decompensation was tachycardia mediated. METHODS AND RESULTS: The Korean Acute Heart Failure registry was a prospective registry that consecutively enrolled 5625 patients with acute HF. Patients were classified into three groups according to the rhythm and aggravating factor: (i) 3664 (65.1%) patients with sinus rhythm (SR), (ii) 1033 (18.4%) patients with AF whose decompensation was tachycardia-mediated, AF-TM (+), and (iii) N = 928 (16.5%) patients with AF whose decompensation was not tachycardia-mediated, AF-TM (-). The primary outcomes were in-hospital and post-discharge 1 year all-cause mortality. At admission, the mean heart rate was 90.8 ± 23.4, 86.8 ± 26.8, and 106.3 ± 29.7 beats per minute for the SR, AF-TM (-), and AF-TM (+) groups, respectively. The AF-TM (+) group had more favourable characteristics such as de novo onset HF, less diabetes, ischaemic heart disease, and higher blood pressure than the AF-TM (-) group. In-hospital mortality rates were 5.1%, 6.5%, and 1.7% for SR, AF-TM (-), and AF-TM (+) groups, respectively. In logistic regression analysis, the AF-TM (+) group had lower in-hospital mortality after adjusting the significant covariates (odds ratio, 0.49; 95% confidence interval, 0.26-0.93). The mortality rate did not differ between SR and AF-TM (-) groups. During 1 year follow-up, 990 (18.5%) patients died. In univariate and multivariate Cox proportional regression analyses, there was no difference in 1-year all-cause mortality between the three groups. CONCLUSIONS: In patients with HF and AF, patients whose acute decompensation is tachycardia-mediated have better in-hospital, but similar post-discharge outcomes compared with those with SR or those with AF whose decompensation is not tachycardia-mediated. CLINICAL TRIAL REGISTRATION: ClinicalTrial.gov NCT01389843.

Beneficial Effect of Left Ventricular Remodeling after Early Change of Sacubitril/Valsartan in Patients with Nonischemic Dilated Cardiomyopathy.

Background and Objectives: Evidence for effectiveness of early change from angiotensin II receptor blockers (ARBs) or angiotensin-converting enzyme inhibitors (ACEIs) to sacubitril/valsartan is lacking. We aimed to investigate whether early changes to sacubitril/valsartan could improve outcomes in patients with nonischemic dilated cardiomyopathy (DCM) in real-world practice. Materials and Methods: A total of 296 patients with nonischemic DCM who were treated with ARB or ACEI continuously (group A, n = 150) or had their medication switched to sacubitril/valsartan (group S, n = 146) were included. The sacubitril/valsartan group was divided into early change (within 60 days, group S/E, n = 59) and late change (group S/L, n = 87) groups. Changes in echocardiographic parameters from the time of initial diagnosis to the last follow-up were analyzed. Results: Patients in group S showed greater left ventricular (LV) end-diastolic dimension (EDD) (group A vs. S, 61.7 ± 7.4 vs. 66.5 ± 8.0, p < 0.001) and lower LV ejection fraction (LVEF) (28.9 ± 8.2% vs. 23.9 ± 7.5%, p < 0.001) than those in group A at initial diagnosis. During a median follow-up of 76 months, patients in group S/E, ∆ LVEF (%) and ∆ LVESD (mm) were significantly improved compared with those in patients in group A (group A vs. S/E, ∆ LVEF, p = 0.036; ∆ LVESD, p = 0.023) or S/L (group S/E vs. S/L, ∆ LVEF, p = 0.05; ∆ LVESD, p = 0.005). Among patients whose medications were switched to sacubitril/valsartan, those with an earlier change showed a significant correlation with greater LVEF improvement (r = -0.367, p < 0.001) and LV reverse remodeling (r = 0.277, p < 0.001). Conclusions: in patients with nonischemic DCM, an early switch to sacubitril/valsartan was associated with greater improvement in LV function. Patients might benefit in terms of LV function by early switching to sacubitril/valsartan.