Background and Objectives: Prior studies have shown that stroke patients treated with percutaneous left atrial appendage occlusion (LAAO) for non-valvular atrial fibrillation (NVAF) experience better outcomes than similar patients treated with warfarin. We investigated the impact of percutaneous left atrial appendage closure on post-stroke neurological outcomes in NVAF patients, compared with non-vitamin K antagonist oral anticoagulant (NOAC) therapy. Methods: Medical records for 1,427 patients in multiple registries and for 1,792 consecutive patients at 6 Korean hospitals were reviewed with respect to LAAO or NOAC treatment.Stroke severity in patients who experienced ischemic stroke or transient ischemic attack after either treatment was assessed with modified Rankin Scale (mRS) scoring at hospital discharge and at 3 and 12 months post-stroke. Results: mRS scores were significantly lower in LAAO patients at 3 (p<0.01) and 12 months (p<0.01) post-stroke, despite no significant differences in scores before the ischemic cerebrovascular event (p=0.22). The occurrences of disabling ischemic stroke in the LAAO and NOAC groups were 36.7% and 44.2% at discharge (p=0.47), 23.3% and 44.2% at 3 months post-stroke (p=0.04), and 13.3% and 43.0% at 12 months post-stroke (p=0.01), respectively.Recovery rates for disabling ischemic stroke at discharge to 12 months post-stroke were significantly higher for LAAO patients (50.0%) than for NOAC patients (5.6%) (p<0.01). Conclusions Percutaneous LAAO was associated with more favorable neurological outcomes after ischemic cerebrovascular event than NOAC treatment.

Background and Objectives: Prior studies have shown that stroke patients treated with percutaneous left atrial appendage occlusion (LAAO) for non-valvular atrial fibrillation (NVAF) experience better outcomes than similar patients treated with warfarin. We investigated the impact of percutaneous left atrial appendage closure on post-stroke neurological outcomes in NVAF patients, compared with non-vitamin K antagonist oral anticoagulant (NOAC) therapy. Methods: Medical records for 1,427 patients in multiple registries and for 1,792 consecutive patients at 6 Korean hospitals were reviewed with respect to LAAO or NOAC treatment.Stroke severity in patients who experienced ischemic stroke or transient ischemic attack after either treatment was assessed with modified Rankin Scale (mRS) scoring at hospital discharge and at 3 and 12 months post-stroke. Results: mRS scores were significantly lower in LAAO patients at 3 (p<0.01) and 12 months (p<0.01) post-stroke, despite no significant differences in scores before the ischemic cerebrovascular event (p=0.22). The occurrences of disabling ischemic stroke in the LAAO and NOAC groups were 36.7% and 44.2% at discharge (p=0.47), 23.3% and 44.2% at 3 months post-stroke (p=0.04), and 13.3% and 43.0% at 12 months post-stroke (p=0.01), respectively.Recovery rates for disabling ischemic stroke at discharge to 12 months post-stroke were significantly higher for LAAO patients (50.0%) than for NOAC patients (5.6%) (p<0.01). Conclusions Percutaneous LAAO was associated with more favorable neurological outcomes after ischemic cerebrovascular event than NOAC treatment.

BACKGROUND AND OBJECTIVES: There are no data comparing clinical outcomes of complex percutaneous coronary intervention (PCI) between biodegradable polymer-biolimus-eluting stents (BP-BES) and durable polymer-everolimus-eluting stents (DP-EES). We sought to evaluate the safety and efficacy of BP-BES compared with DP-EES in patients undergoing complex PCI. METHODS: Patients enrolled in the SMART-DESK registry were stratified into 2 categories based on the complexity of PCI. Complex PCI was defined as having at least one of the following features: unprotected left main lesion, ≥2 lesions treated, total stent length >40 mm, minimal stent diameter ≤2.5 mm, or bifurcation as target lesion. The primary outcome was target lesion failure (TLF), defined as a composite of cardiac death, target vessel-related myocardial infarction (TV-MI), or target lesion revascularization (TLR) at 2 years of follow-up. RESULTS: Of 1,999 patients, 1,145 (57.3%) underwent complex PCI: 521 patients were treated with BP-BES and 624 with DP-EES. In propensity-score matching analysis (481 pairs), the risks of TLF (3.8% vs. 5.2%, adjusted hazard ratio [HR], 0.578; 95% confidence interval [CI], 0.246–1.359; p=0.209), cardiac death (2.5% vs. 2.5%, adjusted HR, 0.787; 95% CI, 0.244–2.539; p=0.689), TV-MI (0.5% vs. 0.4%, adjusted HR, 1.128; 95% CI, 0.157–8.093; p=0.905), and TLR (1.1% vs. 2.9%, adjusted HR, 0.390; 95% CI, 0.139–1.095; p=0.074) did not differ between 2 stent groups after complex PCI. CONCLUSIONS: Clinical outcomes of BP-BES were comparable to those of DP-EES at 2 years after complex PCI. Our data suggest that use of BP-BES is acceptable, even for complex PCI.
Coronary Artery Disease ; Death ; Drug-Eluting Stents ; Follow-Up Studies ; Humans ; Myocardial Infarction ; Percutaneous Coronary Intervention ; Stents

BACKGROUND AND OBJECTIVES: This study aimed to confirm the effects of traditional holidays on the incidence and outcomes of out-of-hospital cardiac arrest (OHCA) in South Korea. METHODS: We studied 95,066 OHCAs of cardiac cause from a nationwide, prospective study from the Korea OHCA Registry from January 2012 to December 2016. We compared the incidence of OHCA, in-hospital mortality, and neurologic outcomes between traditional holidays, Seollal (Lunar New Year's Day) and Chuseok (Korean Thanksgiving Day), and other day types (weekday, weekend, and public holiday). RESULTS: OHCA occurred more frequently on traditional holidays than on the other days. The median OHCA incidence were 51.0 (interquartile range [IQR], 44.0–58.0), 53.0 (IQR, 46.0–60.5), 52.5 (IQR, 45.3–59.8), and 60.0 (IQR, 52.0–69.0) cases/day on weekday, weekend, public holiday, and traditional holiday, respectively (p<0.001). The OHCA occurred more often at home rather than in public place, lesser bystander cardiopulmonary resuscitation (CPR) was performed, and the rate of cessation of CPR within 20 minutes without recovery of spontaneous circulation was higher on traditional holiday. After multivariable adjustment, traditional holiday was associated with higher in-hospital mortality (adjusted hazard ratio [HR], 1.339; 95% confidence interval [CI], 1.058–1.704; p=0.016) but better neurologic outcomes (adjusted HR, 0.503; 95% CI, 0.281–0.894; p=0.020) than weekdays. CONCLUSIONS: The incidence of OHCAs was associated with day types in a year. It occurred more frequently on traditional holidays than on other day types. It was associated with higher in-hospital mortality and favorable neurologic outcomes than weekday.
Cardiopulmonary Resuscitation ; Epidemiology ; Heart Arrest ; Holidays ; Hospital Mortality ; Incidence ; Korea ; Mortality ; Out-of-Hospital Cardiac Arrest ; Prospective Studies

BACKGROUND AND OBJECTIVES: Screening and early diagnosis for heart failure (HF) are critical. However, conventional screening diagnostic methods have limitations, and electrocardiography (ECG)-based HF identification may be helpful. This study aimed to develop and validate a deep-learning algorithm for ECG-based HF identification (DEHF). METHODS: The study involved 2 hospitals and 55,163 ECGs of 22,765 patients who performed echocardiography within 4 weeks were study subjects. ECGs were divided into derivation and validation data. Demographic and ECG features were used as predictive variables. The primary endpoint was detection of HF with reduced ejection fraction (HFrEF; ejection fraction [EF]≤40%), and the secondary endpoint was HF with mid-range to reduced EF (≤50%). We developed the DEHF using derivation data and the algorithm representing the risk of HF between 0 and 1. We confirmed accuracy and compared logistic regression (LR) and random forest (RF) analyses using validation data. RESULTS: The area under the receiver operating characteristic curves (AUROCs) of DEHF for identification of HFrEF were 0.843 (95% confidence interval, 0.840–0.845) and 0.889 (0.887–0.891) for internal and external validation, respectively, and these results significantly outperformed those of LR (0.800 [0.797–0.803], 0.847 [0.844–0.850]) and RF (0.807 [0.804–0.810], 0.853 [0.850–0.855]) analyses. The AUROCs of deep learning for identification of the secondary endpoint was 0.821 (0.819–0.823) and 0.850 (0.848–0.852) for internal and external validation, respectively, and these results significantly outperformed those of LR and RF. CONCLUSIONS: The deep-learning algorithm accurately identified HF using ECG features and outperformed other machine-learning methods.
Artificial Intelligence ; Early Diagnosis ; Echocardiography ; Electrocardiography ; Forests ; Heart Failure ; Heart ; Humans ; Learning ; Logistic Models ; Machine Learning ; Mass Screening ; ROC Curve

BACKGROUND AND OBJECTIVES: There are no data comparing clinical outcomes of complex percutaneous coronary intervention (PCI) between biodegradable polymer-biolimus-eluting stents (BP-BES) and durable polymer-everolimus-eluting stents (DP-EES). We sought to evaluate the safety and efficacy of BP-BES compared with DP-EES in patients undergoing complex PCI. METHODS: Patients enrolled in the SMART-DESK registry were stratified into 2 categories based on the complexity of PCI. Complex PCI was defined as having at least one of the following features: unprotected left main lesion, ≥2 lesions treated, total stent length >40 mm, minimal stent diameter ≤2.5 mm, or bifurcation as target lesion. The primary outcome was target lesion failure (TLF), defined as a composite of cardiac death, target vessel-related myocardial infarction (TV-MI), or target lesion revascularization (TLR) at 2 years of follow-up. RESULTS: Of 1,999 patients, 1,145 (57.3%) underwent complex PCI: 521 patients were treated with BP-BES and 624 with DP-EES. In propensity-score matching analysis (481 pairs), the risks of TLF (3.8% vs. 5.2%, adjusted hazard ratio [HR], 0.578; 95% confidence interval [CI], 0.246–1.359; p=0.209), cardiac death (2.5% vs. 2.5%, adjusted HR, 0.787; 95% CI, 0.244–2.539; p=0.689), TV-MI (0.5% vs. 0.4%, adjusted HR, 1.128; 95% CI, 0.157–8.093; p=0.905), and TLR (1.1% vs. 2.9%, adjusted HR, 0.390; 95% CI, 0.139–1.095; p=0.074) did not differ between 2 stent groups after complex PCI. CONCLUSIONS Clinical outcomes of BP-BES were comparable to those of DP-EES at 2 years after complex PCI. Our data suggest that use of BP-BES is acceptable, even for complex PCI.

BACKGROUND AND OBJECTIVES: Screening and early diagnosis for heart failure (HF) are critical. However, conventional screening diagnostic methods have limitations, and electrocardiography (ECG)-based HF identification may be helpful. This study aimed to develop and validate a deep-learning algorithm for ECG-based HF identification (DEHF). METHODS: The study involved 2 hospitals and 55,163 ECGs of 22,765 patients who performed echocardiography within 4 weeks were study subjects. ECGs were divided into derivation and validation data. Demographic and ECG features were used as predictive variables. The primary endpoint was detection of HF with reduced ejection fraction (HFrEF; ejection fraction [EF]≤40%), and the secondary endpoint was HF with mid-range to reduced EF (≤50%). We developed the DEHF using derivation data and the algorithm representing the risk of HF between 0 and 1. We confirmed accuracy and compared logistic regression (LR) and random forest (RF) analyses using validation data. RESULTS: The area under the receiver operating characteristic curves (AUROCs) of DEHF for identification of HFrEF were 0.843 (95% confidence interval, 0.840–0.845) and 0.889 (0.887–0.891) for internal and external validation, respectively, and these results significantly outperformed those of LR (0.800 [0.797–0.803], 0.847 [0.844–0.850]) and RF (0.807 [0.804–0.810], 0.853 [0.850–0.855]) analyses. The AUROCs of deep learning for identification of the secondary endpoint was 0.821 (0.819–0.823) and 0.850 (0.848–0.852) for internal and external validation, respectively, and these results significantly outperformed those of LR and RF. CONCLUSIONS The deep-learning algorithm accurately identified HF using ECG features and outperformed other machine-learning methods.

BACKGROUND AND OBJECTIVES: This study aimed to confirm the effects of traditional holidays on the incidence and outcomes of out-of-hospital cardiac arrest (OHCA) in South Korea. METHODS: We studied 95,066 OHCAs of cardiac cause from a nationwide, prospective study from the Korea OHCA Registry from January 2012 to December 2016. We compared the incidence of OHCA, in-hospital mortality, and neurologic outcomes between traditional holidays, Seollal (Lunar New Year's Day) and Chuseok (Korean Thanksgiving Day), and other day types (weekday, weekend, and public holiday). RESULTS: OHCA occurred more frequently on traditional holidays than on the other days. The median OHCA incidence were 51.0 (interquartile range [IQR], 44.0–58.0), 53.0 (IQR, 46.0–60.5), 52.5 (IQR, 45.3–59.8), and 60.0 (IQR, 52.0–69.0) cases/day on weekday, weekend, public holiday, and traditional holiday, respectively (p<0.001). The OHCA occurred more often at home rather than in public place, lesser bystander cardiopulmonary resuscitation (CPR) was performed, and the rate of cessation of CPR within 20 minutes without recovery of spontaneous circulation was higher on traditional holiday. After multivariable adjustment, traditional holiday was associated with higher in-hospital mortality (adjusted hazard ratio [HR], 1.339; 95% confidence interval [CI], 1.058–1.704; p=0.016) but better neurologic outcomes (adjusted HR, 0.503; 95% CI, 0.281–0.894; p=0.020) than weekdays. CONCLUSIONS The incidence of OHCAs was associated with day types in a year. It occurred more frequently on traditional holidays than on other day types. It was associated with higher in-hospital mortality and favorable neurologic outcomes than weekday.

BACKGROUND AND OBJECTIVES: A risk prediction is needed even in the contemporary era of acute myocardial infarction (AMI). We sought to develop a risk scoring specific for patients with AMI being treated with guideline-adherent optimal therapies, including percutaneous coronary intervention and all 5 medications (aspirin, thienopyridine, β-blocker, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and statin). METHODS: From registries, 12,174 AMI patients were evaluated. The primary outcome was 1-year all-cause death or AMI. The Korea Working Group in Myocardial Infarction (KorMI) system was compared with the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX AMI), Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC), and Global Registry of Acute Coronary Events scores (GRACE) models. RESULTS: Ten predictors were identified: left ventricular dysfunction (hazard ratio [HR], 2.3), bare-metal stent (HR, 2.0), Killip class ≥II (HR, 1.9), renal insufficiency (HR, 1.8), previous stroke (HR, 1.6), regional wall-motion- score >20 on echocardiography (HR, 1.5), body mass index ≤24 kg/m2 (HR, 1.4), age ≥70 years (HR, 1.4), prior coronary heart disease (HR, 1.4), and diabetes (HR, 1.4). Compared with the previous models, the KorMI system had good discrimination (time-dependent C statistic, 0.759) and showed reasonable goodness-of-fit by Hosmer-Lemeshow test (p=0.84). Moreover, the continuous-net reclassification improvement varied from −27.3% to −19.1%, the integrated discrimination index varied from −2.1% to −0.9%, and the median improvement in risk score was from −1.0% to −0.4%. CONCLUSIONS: The KorMI system would be a useful tool for predicting outcomes in survivors treated with guideline-adherent optimal therapies after AMI.
Angioplasty ; Angiotensins ; Body Mass Index ; Coronary Disease ; Discrimination (Psychology) ; Drug Therapy ; Echocardiography ; Humans ; Korea ; Myocardial Infarction ; Percutaneous Coronary Intervention ; Registries ; Renal Insufficiency ; Stents ; Stroke ; Survivors ; Ventricular Dysfunction, Left

BACKGROUND AND OBJECTIVES: A risk prediction is needed even in the contemporary era of acute myocardial infarction (AMI). We sought to develop a risk scoring specific for patients with AMI being treated with guideline-adherent optimal therapies, including percutaneous coronary intervention and all 5 medications (aspirin, thienopyridine, β-blocker, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and statin). METHODS: From registries, 12,174 AMI patients were evaluated. The primary outcome was 1-year all-cause death or AMI. The Korea Working Group in Myocardial Infarction (KorMI) system was compared with the Assessment of Pexelizumab in Acute Myocardial Infarction (APEX AMI), Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications (CADILLAC), and Global Registry of Acute Coronary Events scores (GRACE) models. RESULTS: Ten predictors were identified: left ventricular dysfunction (hazard ratio [HR], 2.3), bare-metal stent (HR, 2.0), Killip class ≥II (HR, 1.9), renal insufficiency (HR, 1.8), previous stroke (HR, 1.6), regional wall-motion- score >20 on echocardiography (HR, 1.5), body mass index ≤24 kg/m2 (HR, 1.4), age ≥70 years (HR, 1.4), prior coronary heart disease (HR, 1.4), and diabetes (HR, 1.4). Compared with the previous models, the KorMI system had good discrimination (time-dependent C statistic, 0.759) and showed reasonable goodness-of-fit by Hosmer-Lemeshow test (p=0.84). Moreover, the continuous-net reclassification improvement varied from −27.3% to −19.1%, the integrated discrimination index varied from −2.1% to −0.9%, and the median improvement in risk score was from −1.0% to −0.4%. CONCLUSIONS The KorMI system would be a useful tool for predicting outcomes in survivors treated with guideline-adherent optimal therapies after AMI.