Long-term clinical outcomes of image-guided percutaneous coronary intervention in acute myocardial infarction from the Korea Acute Myocardial Infarction Registry.

Imaging modalities for percutaneous coronary intervention (PCI), such as intravascular ultrasound (IVUS) or optical coherence tomography (OCT), have increased in the current PCI era. However, their clinical benefits in acute myocardial infarction (AMI) have not been fully elucidated. This study investigated the long-term outcomes of image-guided PCI in patients with AMI using data from the Korean Acute Myocardial Infarction Registry. A total of 9,271 patients with AMI, who underwent PCI with second-generation drug-eluting stents between November 2011 and December 2015, were retrospectively examined, and target lesion failure (TLF) at 3 years (defined as the composite of cardiac death, target vessel myocardial infarction, and ischemia-driven target lesion revascularization) was evaluated. From the registry, 2,134 patients (23.0%) underwent image-guided PCI (IVUS-guided: n = 1,919 [20.6%]; OCT-guided: n = 215 patients [2.3%]). Based on propensity score matching, image-guided PCI was associated with a significant reduction in TLF (hazard ratio: 0.76; 95% confidence interval: 0.59-0.98, p = 0.035). In addition, the TLF incidence in the OCT-guided PCI group was comparable to that in the IVUS-guided PCI group (5.3% vs 4.7%, p = 0.903). Image-guided PCI, including IVUS and OCT, is associated with favorable clinical outcomes in patients with AMI at 3 years post-intervention. Additionally, OCT-guided PCI is not inferior to IVUS-guided PCI in patients with AMI.

Optimal Dose of Edoxaban for Very Elderly Atrial Fibrillation Patients at High Risk of Bleeding: The LEDIOS Registry.

BACKGROUND AND OBJECTIVES: Optimal anticoagulation in very elderly patients is challenging due to the high risk of anticoagulant-induced bleeding. The aim of this study was to assess outcomes of on-label reduced-dose edoxaban (30 mg) in very elderly patients who had additional risk factors for bleeding. METHODS: This was a multi-center, prospective, non-interventional observational study to evaluate the efficacy and safety of on-label reduced-dose edoxaban in atrial fibrillation (AF) patients 80 years of age or older and who had more than 1 risk factor for bleeding. RESULTS: A total of 2448 patients (mean age 75.0±8.3 years, 801 [32.7%] males) was included in the present study, and 586 (23.9%) were 80 years of age or older with additional risk factors for bleeding. Major bleeding events occurred frequently among very elderly AF patients who had additional bleeding risk factors compared to other patients (unadjusted hazard ratio [HR], 2.16; 95% confidence interval [CI], 1.16-4.02); however, there were no significant differences in stroke incidence (HR, 1.86; 95% CI, 0.98-3.55). There were no significant differences for either factor after adjusting for age and sex (adjusted HR, 1.65; 95% CI, 0.75-3.62 for major bleeding; adjusted HR, 1.13; 95% CI, 0.51-2.50 for stroke). CONCLUSIONS: In very elderly AF patients with comorbidities associated with greater risk of bleeding, the incidence of major bleeding events was significantly increased. In addition, risk of stroke showed tendency to increase in same population. Effective anticoagulation therapy might be important in these high-risk population, and close observation of bleeding events might also be required. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT03554837.

Multi-scale V-net architecture with deep feature CRF layers for brain extraction.

BACKGROUND: Brain extraction is a computational necessity for researchers using brain imaging data. However, the complex structure of the interfaces between the brain, meninges and human skull have not allowed a highly robust solution to emerge. While previous methods have used machine learning with structural and geometric priors in mind, with the development of Deep Learning (DL), there has been an increase in Neural Network based methods. Most proposed DL models focus on improving the training data despite the clear gap between groups in the amount and quality of accessible training data between. METHODS: We propose an architecture we call Efficient V-net with Additional Conditional Random Field Layers (EVAC+). EVAC+ has 3 major characteristics: (1) a smart augmentation strategy that improves training efficiency, (2) a unique way of using a Conditional Random Fields Recurrent Layer that improves accuracy and (3) an additional loss function that fine-tunes the segmentation output. We compare our model to state-of-the-art non-DL and DL methods. RESULTS: Results show that even with limited training resources, EVAC+ outperforms in most cases, achieving a high and stable Dice Coefficient and Jaccard Index along with a desirable lower Surface (Hausdorff) Distance. More importantly, our approach accurately segmented clinical and pediatric data, despite the fact that the training dataset only contains healthy adults. CONCLUSIONS: Ultimately, our model provides a reliable way of accurately reducing segmentation errors in complex multi-tissue interfacing areas of the brain. We expect our method, which is publicly available and open-source, to be beneficial to a wide range of researchers.

Computational processing of brain images can enable better understanding and diagnosis of diseases that affect the brain. Brain Extraction is a computational method that can be used to remove areas of the head that are not the brain from images of the head. We compared various different computational methods that are available and used them to develop a better method. The method we describe in the paper is more accurate at imaging the brain of both healthy individuals and those known to have diseases that affect the brain than the other methods we evaluated. Our method might enable better understanding and diagnosis of diseases that affect the brain in the future.

Impact of Positron Emission Tomography Viability Imaging: Guided Revascularizations on Clinical Outcomes in Patients With Myocardial Scar on Single-Photon Emission Computed Tomography Scans.

BACKGROUND: Positron emission tomography (PET) viability scan is used to determine whether patients with a myocardial scar on single-photon emission computed tomography (SPECT) may need revascularization. However, the clinical utility of revascularization decision-making guided by PET viability imaging has not been proven yet. The purpose of this study was to investigate the impact of PET to determine revascularization on clinical outcomes. METHODS: Between September 2012 and May 2021, 53 patients (37 males; mean age = 64 ± 11 years) with a myocardial scar on MIBI SPECT who underwent PET viability test were analyzed in this study. The primary outcome was a temporal change in echocardiographic findings. The secondary outcome was all-cause mortality. RESULTS: Viable myocardium was presented by PET imaging in 29 (54.7%) patients. Revascularization was performed in 26 (49.1%) patients, including 18 (34.0%) with percutaneous coronary intervention (PCI) and 8 (15.1%) with coronary artery bypass grafting. There were significant improvements in echocardiographic findings in the revascularization group and the viable myocardium group. All-cause mortality was significantly lower in the revascularization group than in the medical therapy-alone group (19.2% vs. 44.4%, log-rank P = 0.002) irrespective of viable (21.4% vs. 46.7%, log-rank P = 0.025) or non-viable myocardium (16.7% vs. 41.7%, log-rank P = 0.046). All-cause mortality was significantly lower in the PCI group than in the medical therapy-alone group (11.1% vs. 44.4%, log-rank P < 0.001). CONCLUSION: Revascularization improved left ventricular systolic function and survival of patients with a myocardial scar on SPECT scans, irrespective of myocardial viability on PET scans.

Effect of cholesterol variability on the incidence of cataract, dementia, and osteoporosis: A study using a common data model.

The effects of cholesterol variability on cataracts, dementia, and osteoporosis remain controversial. Using a common data model, we investigated the effects of variations in cholesterol levels on the development of cataracts, dementia, and osteoporosis. Patients who received statin therapy between 2011 and 2020 and those with 3 or more tests for total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) levels were included. The patients were divided into those with a coefficient of variation (CV) of TC higher than the mean (high-CV group) and those with a lower CV of TC (low-CV group). Moreover, 1:1 propensity score matching was conducted based on demographic variables. Cataract, dementia, or osteoporosis was defined as having a diagnostic, drug, or surgical code based on the cohort definition. Of the 12,882 patients, cataracts, dementia, and osteoporosis were developed in 525 (4.1%), 198 (1.5%), and 438 (3.4%) patients, respectively. The stratified Cox proportional hazards model showed that the incidences of cataracts and osteoporosis were 1.38 and 1.45 times greater in the high-CV group than in the low-CV group, respectively. Our study revealed that TC variability is associated with developing cataracts and osteoporosis.

Long-term impact of angiotensin receptor-neprilysin inhibitor based on short-term treatment response in heart failure.

AIMS: The long-term effect of angiotensin receptor-neprilysin inhibitor (ARNI) remains uncertain in patients who have experienced improvements in left ventricular (LV) systolic function or significant LV reverse remodelling following a certain period of treatment. It is also unclear how ARNI performs in patients who have not shown these improvements. This study aimed to assess the impact of prolonged ARNI use compared with angiotensin-converting enzyme inhibitors (ACEIs)/angiotensin receptor blockers (ARBs) in patients with and without significant treatment response after 1 year of heart failure (HF) treatment. METHODS AND RESULTS: The present study enrolled patients with HF with reduced ejection fraction (HFrEF) who were treated with either ARNI or ACEIs/ARBs within 1 year of undergoing index echocardiography. After 1 year of treatment, patients were reclassified into the following groups: (i) patients with HF with improved ejection fraction and persistent HFrEF and (ii) patients with and without LV reverse remodelling based on the follow-up echocardiography. The effect of ARNI versus that of ACEIs/ARBs in each group was assessed from the time of categorizing into new groups using the composite event of all-cause mortality and HF hospitalization. A total of 671 patients with HFrEF (age, 66.4 ± 14.1 years; males, 66.8%) were included, and 133 (19.8%) composite events of death and rehospitalization for HF were observed during the follow-up (median follow-up, 44 [interquartile range, 34-51] months). ARNI had a significantly lower event rate than ACEIs/ARBs in patients with HF with improved ejection fraction (7.0% vs. 30.4%, P = 0.020) and those with persistent HFrEF (17.6% vs. 49.7%, P < 0.001). Irrespective of whether patients exhibited LV reverse remodelling (15.8% vs. 31.1%, P = 0.001) or not (15.0% vs. 54.9%, P < 0.001), ARNIs were associated with a significantly lower event rate than ACEIs/ARBs. CONCLUSIONS: Regardless of significant treatment response measured by either LVEF or LV reverse remodelling after 1 year of treatment, the extended utilization of ARNI demonstrated a more favourable prognosis than that of ACEIs/ARBs in patients with HFrEF.

Diagnostic Value of a Wearable Continuous Electrocardiogram Monitoring Device (AT-Patch) for New-Onset Atrial Fibrillation in High-Risk Patients: Prospective Cohort Study.

BACKGROUND: While conventional electrocardiogram monitoring devices are useful for detecting atrial fibrillation, they have considerable drawbacks, including a short monitoring duration and invasive device implantation. The use of patch-type devices circumvents these drawbacks and has shown comparable diagnostic capability for the early detection of atrial fibrillation. OBJECTIVE: We aimed to determine whether a patch-type device (AT-Patch) applied to patients with a high risk of new-onset atrial fibrillation defined by the congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke, vascular disease, age 65-74 years, sex scale (CHA2DS2-VASc) score had increased detection rates. METHODS: In this nonrandomized multicenter prospective cohort study, we enrolled 320 adults aged ≥19 years who had never experienced atrial fibrillation and whose CHA2DS2-VASc score was ≥2. The AT-Patch was attached to each individual for 11 days, and the data were analyzed for arrhythmic events by 2 independent cardiologists. RESULTS: Atrial fibrillation was detected by the AT-Patch in 3.4% (11/320) of patients, as diagnosed by both cardiologists. Interestingly, when participants with or without atrial fibrillation were compared, a previous history of heart failure was significantly more common in the atrial fibrillation group (n=4/11, 36.4% vs n=16/309, 5.2%, respectively; P=.003). When a CHA2DS2-VASc score ≥4 was combined with previous heart failure, the detection rate was significantly increased to 24.4%. Comparison of the recorded electrocardiogram data revealed that supraventricular and ventricular ectopic rhythms were significantly more frequent in the new-onset atrial fibrillation group compared with nonatrial fibrillation group (3.4% vs 0.4%; P=.001 and 5.2% vs 1.2%; P<.001), respectively. CONCLUSIONS: This study detected a moderate number of new-onset atrial fibrillations in high-risk patients using the AT-Patch device. Further studies will aim to investigate the value of early detection of atrial fibrillation, particularly in patients with heart failure as a means of reducing adverse clinical outcomes of atrial fibrillation. TRIAL REGISTRATION: ClinicalTrials.gov NCT04857268; https://classic.clinicaltrials.gov/ct2/show/NCT04857268.

Statin therapy reduces dementia risk in atrial fibrillation patients receiving oral anticoagulants.

AIMS: Atrial fibrillation (AF) is linked to an increased risk of dementia, even in stroke-free patients. The impact of statin therapy on dementia risk is unclear in AF patients receiving oral anticoagulant (OAC) (vitamin K antagonist and direct-acting OAC). We sought to investigate the impact of statin therapy on dementia risk in AF patients receiving OAC. METHODS AND RESULTS: Using the Korean National Health Insurance Service database, 91 018 non-valvular AF (NVAF) patients from January 2013 to December 2017 were included in the analysis. Of the total, 17 700 patients (19.4%) were in the statin therapy group, and 73 318 patients (80.6%) were in the non-statin therapy group. The primary endpoint was the occurrence of dementia. The median duration of follow-up was 2.1 years. Statin therapy was associated with a significantly lower dementia risk than non-statin therapy for CHA2DS2-VASc scores ≥2 (hazard ratio = 0.77, 95% confidence interval 0.64-0.90, P = 0.026) in NVAF patients receiving OAC. The statin therapy group had a significantly lower dementia risk in a dose-dependent relationship compared with the non-statin therapy group (P for trend <0.001). CONCLUSION: In NVAF patients who received OAC, statin therapy lowered the dementia risk compared with no statin therapy. Furthermore, statin therapy is associated with a dose-dependent reduction in dementia risk.

Sex differences between serum uric acid levels and cardiovascular outcomes in patients with coronary artery disease after stent implantation.

Background: The relationship between elevated serum uric acid (SUA) levels and cardiovascular outcomes after stent implantation remains uncertain. This study sought to evaluate the impact of SUA on 12-month cardiovascular outcomes after stent implantation. Methods: We performed a retrospective study of patients who successfully underwent stent implantation and enrolled 3,222 patients with coronary artery disease (CAD) from a single center. SUA levels were measured before stent implantation. The patients were divided into six groups (<4, 4-4.9, 5-5.9, 6-6.9, 7-7.9 and ≥ 8 mg/dL) at SUA intervals of 1.0 mg/dL. The incidence of cardiovascular outcomes in the six groups was monitored for 1 year after stent implantation and the hazard ratios were estimated. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) for cardiovascular outcomes were estimated using a Cox proportional hazard regression analysis. The primary endpoint was all-cause death. The secondary endpoint was a composite of all-cause death, myocardial infarction, target vessel revascularization, stent thrombosis and stroke. The follow-up duration was 12 months. Results: Over the 12-month follow-up period, there were 101 all-cause deaths and 218 MACCE. After adjustment for several parameters, the group with SUA levels of more than or equal to 8 mg/dL had significantly higher hazard ratios in the incidence of all-cause death or MACCE. The group with <4.0 mg/dL had significantly higher hazard ratios in all-cause death only in male patients. In contrast, there were no significant differences observed for cardiovascular outcomes in female patients. Conclusions: Our study identified a U-shaped association between SUA levels and cardiovascular outcomes during 12-month follow-up for males, but not for females. Further studies are warranted to clarify the sex differences between SUA levels and clinical outcomes.

Predictors of stroke or systemic embolism in patients with non-valvular atrial fibrillation with CHA2 DS2 -VASc score of 0.

BACKGROUND: Anticoagulant therapy has been important for stroke prevention in patients with atrial fibrillation (AF). However, it was not recommended due to its relatively higher risk of bleeding than its lower risk of stroke in patients with a CHA2 DS2 -VASc score of 0. HYPOTHESIS: This study aimed to evaluate the predictors of stroke in AF patients with very low risk of stroke. METHODS: Between 1990 and 2020, 542 patients with non-valvular AF (NVAF) with a CHA2 DS2 -VASc score of 0 followed up for at least 6 months were enrolled. Patients with only being woman as a risk factor were included as a CHA2 DS2 -VASc score of 0 in this study. The primary outcome was stroke or systemic embolism. RESULTS: The primary outcome rate was 0.78%/year. In Cox hazard model, age of ≥50 years at diagnosis (hazard ratio [HR] 6.710, 95% confidence interval [CI] 1.811-24.860, p = .004), LVEDD of ≥46 mm (HR 4.513, 95% CI 1.038-19.626, p = .045), and non-paroxysmal AF (HR 5.575, 95% CI 1.621-19.175, p = .006) were identified as independent predictors of stroke or systemic embolism. Patients with all three independent predictors had a higher risk of stroke or systemic embolism (4.21%/year), whereas those without did not have a stroke or systemic embolism. CONCLUSION: The annual stroke or systemic embolism rate in NVAF patients with CHA2 DS2 -VASc score of 0 was 0.78%/year, and age at AF diagnosis, LVEDD, and non-paroxysmal AF were independent predictors of stroke or systemic embolism in patients considered to have a very low risk of stroke.

COVID-19 Vaccination-Induced Ventricular Fibrillation in an Afebrile Patient With Brugada Syndrome.

A 43-year-old man presented with cardiac arrest 2 days after the second coronavirus disease 2019 (COVID-19) vaccination with an mRNA vaccine. Electrocardiograms showed ventricular fibrillation and type 1 Brugada pattern ST segment elevation. The patient reported having no symptoms, including febrile sensation. There were no known underlying cardiac diseases to explain such electrocardiographic abnormalities. ST segment elevation completely disappeared in two weeks. Although there were no genetic mutations or personal or family history typical of Brugada syndrome, flecainide administration induced type 1 Brugada pattern ST segment elevation. This case suggests that COVID-19 vaccination may induce cardiac ion channel dysfunction and cause life threatening ventricular arrhythmias in specific patients with Brugada syndrome.

Comparison of Factors Associated With Direct Versus Transferred-in Admission to Government-Designated Regional Centers Between Acute Ischemic Stroke and Myocardial Infarction in Korea.

BACKGROUND: There has been no comparison of the determinants of admission route between acute ischemic stroke (AIS) and acute myocardial infarction (AMI). We examined whether factors associated with direct versus transferred-in admission to regional cardiocerebrovascular centers (RCVCs) differed between AIS and AMI. METHODS: Using a nationwide RCVC registry, we identified consecutive patients presenting with AMI and AIS between July 2016 and December 2018. We explored factors associated with direct admission to RCVCs in patients with AIS and AMI and examined whether those associations differed between AIS and AMI, including interaction terms between each factor and disease type in multivariable models. To explore the influence of emergency medical service (EMS) paramedics on hospital selection, stratified analyses according to use of EMS were also performed. RESULTS: Among the 17,897 and 8,927 AIS and AMI patients, 66.6% and 48.2% were directly admitted to RCVCs, respectively. Multivariable analysis showed that previous coronary heart disease, prehospital awareness, higher education level, and EMS use increased the odds of direct admission to RCVCs, but the odds ratio (OR) was different between AIS and AMI (for the first 3 factors, AMI > AIS; for EMS use, AMI < AIS). EMS use was the single most important factor for both AIS and AMI (OR, 4.72 vs. 3.90). Hypertension and hyperlipidemia increased, while living alone decreased the odds of direct admission only in AMI; additionally, age (65-74 years), previous stroke, and presentation during non-working hours increased the odds only in AIS. EMS use weakened the associations between direct admission and most factors in both AIS and AMI. CONCLUSIONS: Various patient factors were differentially associated with direct admission to RCVCs between AIS and AMI. Public education for symptom awareness and use of EMS is essential in optimizing the transportation and hospitalization of patients with AMI and AIS.

The Busan Regional CardioCerebroVascular Center Project's Experience Over a Decade in the Treatment of ST-segment Elevation Myocardial Infarction.

OBJECTIVES: The Regional CardioCerebroVascular Center (RCCVC) project was initiated to improve clinical outcomes for patients with acute myocardial infarction or stroke in non-capital areas of Korea. The purpose of this study was to evaluate the outcomes and issues identified by the Busan RCCVC project in the treatment of ST-segment elevation myocardial infarction (STEMI). METHODS: Among the patients who were registered in the Korean Registry of Acute Myocardial Infarction for the RCCVC project between 2007 and 2019, those who underwent percutaneous coronary intervention (PCI) for STEMI at the Busan RCCVC were selected, and their medical data were compared with a historical cohort. RESULTS: In total, 1161 patients were selected for the analysis. Ten years after the implementation of the Busan RCCVC project, the median door-to-balloon time was reduced from 86 (interquartile range [IQR], 64-116) to 54 (IQR, 44-61) minutes, and the median symptom-to-balloon time was reduced from 256 (IQR, 180-407) to 189 (IQR, 118-305) minutes (p<0.001). Inversely, the false-positive PCI team activation rate increased from 0.6% to 21.4% (p<0.001). However, the 1-year cardiovascular death and major adverse cardiac event rates did not change. Even after 10 years, approximately 75% of the patients had a symptom-to-balloon time over 120 minutes, and approximately 50% of the patients underwent inter-hospital transfer for primary PCI. CONCLUSIONS: A decade after the implementation of the Busan RCCVC project, although time parameters for early reperfusion therapy for STEMI improved, at the cost of an increased false-positive PCI team activation rate, survival outcomes were unchanged.

Thoracoscopic Implantation of Epicardial Left Ventricular Lead for Cardiac Resynchronization Therapy.

(1) Background: Limited data exist on the safety and efficacy of epicardial left ventricular (LV) lead placement using video-assisted thoracoscopic surgery (VATS) for cardiac resynchronization therapy (CRT). (2) Methods: Acute and post-discharge outcomes of CRT were compared between patients with epicardial LV leads (Epicardial-LV group, n = 13) and those with endocardial LV leads (Endocardial-LV group, n = 243). (3) Results: Epicardial LV leads were implanted via VATS alone (n = 8) or along with mini-thoracotomy (n = 5), for failed endocardial implantation (n = 11) or recurrent lead dislodgement (n = 2). All epicardial procedures under general anesthesia with one-lung ventilation were successfully completed in 1.0 ± 0.4 h without phrenic nerve stimulation. LV pacing thresholds in the epicardial-LV (1.5 ± 1.0 V) and endocardial-LV (1.3 ± 0.8 V) were comparable (p = 0.651). All patients were discharged alive post-VATS 8.8 ± 3.9 days. During the follow-up (34.3 ± 28.6 months), all patients with epicardial LV leads stayed alive except for one cardiac death post-CRT 14 months and one heart transplantation post-CRT 30 months. All epicardial LV leads maintained stable performance without dislodgement/significant changes in pacing threshold/impedance. LV lead dislodgement occurred only in endocardial-LV (7/243, 2.9%). Efficacy in both groups was comparable in terms of QRS narrowing, increase in LV ejection fraction, and survival free of cardiac death, or heart-failure-related hospitalization. (4) Conclusions: Epicardial LV lead placement using VATS can be a safe and effective alternative to endocardial implantation, with comparable acute and post-discharge outcomes achieved by both approaches.

Prevalence and Characteristics of Atrial Tachycardia From Noncoronary Aortic Cusp During Atrial Fibrillation Catheter Ablation.

BACKGROUND AND OBJECTIVES: Atrial tachycardias (ATs) from noncoronary aortic cusp (NCC) uncovered after radiofrequency ablation for atrial fibrillation (AF) are rarely reported. This study was conducted to investigate the prevalence and clinical characteristics of NCC ATs detected during AF ablation and compare their characteristics with de novo NCC ATs without AF. METHODS: Consecutive patients who underwent radiofrequency catheter ablation for AF were reviewed from the multicenter AF ablation registry of 11 tertiary hospitals. The clinical and electrophysiological characteristics of NCC AT newly detected during AF ablation were compared with its comparators (de novo NCC AT ablation cases without AF). RESULTS: Among 10,178 AF cases, including 1,301 redo ablation cases, 8 (0.08%) NCC AT cases were discovered after pulmonary vein isolation (PVI; 0.07% in first ablation and 0.15% in redo ablation cases). All ATs were reproducibly inducible spontaneously or with programmed atrial stimulation without isoproterenol infusion. The P-wave morphological features of tachycardia were variable depending on the case, and most cases exhibited 1:1 atrioventricular conduction. AF recurrence rate after PVI and NCC AT successful ablation was 12.5% (1 of 8). Tachycardia cycle length was shorter than that of 17 de novo ATs from NCC (303 versus 378, p=0.012). No AV block occurred during and after successful AT ablation. CONCLUSIONS: Uncommon NCC ATs (0.08% in AF ablation cases) uncovered after PVI, showing different characteristics compared to de-novo NCC ATs, should be suspected irrespective of P-wave morphologies when AT shows broad propagation from the anterior interatrial septum.

Artificial intelligence predicts clinically relevant atrial high-rate episodes in patients with cardiac implantable electronic devices.

To assess the utility of machine learning (ML) algorithms in predicting clinically relevant atrial high-rate episodes (AHREs), which can be recorded by a pacemaker. We aimed to develop ML-based models to predict clinically relevant AHREs based on the clinical parameters of patients with implanted pacemakers in comparison to logistic regression (LR). We included 721 patients without known atrial fibrillation or atrial flutter from a prospective multicenter (11 tertiary hospitals) registry comprising all geographical regions of Korea from September 2017 to July 2020. Predictive models of clinically relevant AHREs were developed using the random forest (RF) algorithm, support vector machine (SVM) algorithm, and extreme gradient boosting (XGB) algorithm. Model prediction training was conducted by seven hospitals, and model performance was evaluated using data from four hospitals. During a median follow-up of 18 months, clinically relevant AHREs were noted in 104 patients (14.4%). The three ML-based models improved the discrimination of the AHREs (area under the receiver operating characteristic curve: RF: 0.742, SVM: 0.675, and XGB: 0.745 vs. LR: 0.669). The XGB model had a greater resolution in the Brier score (RF: 0.008, SVM: 0.008, and XGB: 0.021 vs. LR: 0.013) than the other models. The use of the ML-based models in patient classification was associated with improved prediction of clinically relevant AHREs after pacemaker implantation.

Correction: A randomized, open-label, parallel, multi-center Phase IV study to compare the efficacy and safety of atorvastatin 10 and 20 mg in high-risk Asian patients with hypercholesterolemia.

[This corrects the article DOI: 10.1371/journal.pone.0245481.].