ABSTRACT
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.
Subject(s)
Artificial Intelligence , Atrial Fibrillation/diagnosis , Aged , Aged, 80 and over , Atrial Fibrillation/therapy , Clinical Decision Rules , Cohort Studies , Female , Humans , Machine Learning , Male , Models, Statistical , Pacemaker, Artificial , Prospective Studies , ROC Curve , Registries , Republic of KoreaABSTRACT
BACKGROUND: The efficacy of implantable cardioverter defibrillators (ICDs) for primary prevention is controversial in patients with nonischemic heart failure (HF). We evaluated the mortality and predictors of mortality in patients with prophylactic ICD implantation for ischemic and nonischemic HF. METHODS: From 2008 to 2017, 1097 patients (667, nonischemic HF and 430, ischemic HF) who underwent prophylactic ICD implantation, were identified from the Korean National Health Insurance Service database. We used propensity score overlap weighting to correct the differences between two groups. RESULTS: Those with ischemic HF were older (67.0 ± 10.1 vs 61.8 ± 14.2 years), more often male (71.4% vs 63.7%), and had more comorbidities than patients with nonischemic HF. During a median follow-up of 37.3 months (interquartile range [IQR], 14.2-53.8 months), all-cause mortality was higher in unweighted patients with ischemic HF than in those with nonischemic HF (10.9 vs 6.4 per 100 person-years; hazard ratio [HR], 1.74; 95% confidence interval [CI], 1.38-2.20; P < .001). However, after weighting, the annual all-cause mortality rate was similar in both groups (9.5 vs 8.8 per 100 person-years), with no significant difference in the risk of all-cause mortality (HR, 1.08; 95% CI, 0.68-1.71; P = .755). Older age and chronic kidney disease were independent predictors of all-cause mortality in both groups. There was no significant difference in cardiac and noncardiac mortality between the weighted nonischemic and ischemic HF groups. CONCLUSIONS: The all-cause, cardiac, and noncardiac mortality rates were similar between patients with nonischemic and ischemic HF who underwent prophylactic ICD implantation.
ABSTRACT
BACKGROUND AND OBJECTIVES: There are limited published data on the incidence and cost associated with cardiac implantable electrical device (CIED) infection for Asian patients. We analyzed the infection burden associated with the implantation of CIEDs in Korea. METHODS: In the Health Insurance Review & Assessment Service (HIRA) database during the period from January 1, 2014 to December 31, 2016, we identified 16,908 patients with CIED implantation. CIED infection was defined as either: 1) Infection-related diagnosis code by the Korean Standard Classification of Diseases after any CIED procedure; or 2) CIED removal along with systemic infection. RESULTS: The proportions of first implantation and replacement were 77.6% and 22.4%, respectively. During the follow-up period of 17.1±10.6 months, a total of 462 patients had CIED infection with incidence of 1.95 per 100 person-years with higher infection rate in replacement than first implantation (3.97 vs. 1.4 per 100 person-years, p<0.001). The average cost per person was US$ 16,584 (pacemaker, $13,736; implantable cardioverter defibrillator, $28,402; cardiac resynchronization therapy, $29,674). The risk factors of CIED infection were generator replacement (adjusted hazard ratio [aHR], 3.14; 95% confidence interval [CI], 2.60-3.78), diabetes mellitus (aHR, 1.94; 95% CI, 1.58-2.38), and congestive heart failure (aHR, 1.86; 95% CI, 1.51-2.28). CONCLUSIONS: The rate of CIED infection in Korea was 1.95 per 100 person-years with average cost of US$ 16,584. The most important risk factor was generator replacement. This result suggests that generator replacement should be performed cautiously to avoid CIED infection.
ABSTRACT
BACKGROUND AND OBJECTIVES: Implantation of cardiac implantable electronic devices (CIED), including permanent pacemakers (PM), implantable cardioverter-defibrillators (ICD), and cardiac resynchronization therapy (CRT) devices, has increased significantly over the past several years. However, limited data exists regarding temporal trends of CIED implantations in Asian population. This study aimed to investigate temporal trends of CIED treatment in Korea. METHODS: Using the National Health Insurance Service database of the entire Korean adult population, temporal trends of CIED procedures between 2009 and 2016 were evaluated. Additionally, temporal changes in the prevalence of patients' comorbidities were evaluated. RESULTS: A total of 35,421 CIED procedures (new implantations: 27,771, replacements: 7,650) were performed during the study period. The mean age of new CIED recipients and the prevalence of comorbidities, including hypertension, diabetes mellitus, heart failure, stroke, and atrial fibrillation, increased substantially with time. Compared to 2009, the number of new implantations of PM, ICD, and CRT devices increased by 2.0 (1,977 to 3,910), 3.6 (230 to 822), and 4.9 (44 to 217) times in 2016, respectively. The annual new implantation rate of CIED also increased accordingly (5.1 to 9.3 for PM, 0.6 to 1.9 for ICD, and 0.1 to 0.5 for CRT devices, per 100,000 persons). CONCLUSIONS: The number of CIED implantation increased substantially from 2009 to 2016 in Korea. Also, the patients with CIED have been changed to be older and have more comorbidities. Therefore, the burden of health care cost in patients with CIED would be expected to increase in the future.
