Background/Aims: The reimbursement policy for cryptogenic stroke (CS) was expanded in November 2018 from recurrent strokes to the first stroke episode. No reports have demonstrated whether this policy change has affected trends in implantable loop recorder (ILR) utilization. Methods: We identified patients who received an ILR implant using the Korea Health Insurance Review and Assessment Service database between July 2016 and October 2021. Patients meeting all the following criteria were considered to have CS indication: 1) prior stroke history, 2) no previous history of atrial fibrillation or flutter (AF/AFL), and 3) no maintenance of oral anticoagulant for ≥4 weeks within a year before ILR implant. AF/AFL diagnosed within 3 years after ILR implant or before ILR removal was considered ILR-driven. Results: Among 3,056 patients, 1,001 (32.8%) had CS indications. The total ILR implant number gradually increased for both CS and non-CS indications and the number of CS indication significantly increased after implementing the expanded reimbursement policy. The detection rate for AF/AFL was 26.3% in CS patients over 3 years, which was significantly higher in patients implanted with an ILR within 2 months after stroke than those implanted later. Conclusions The expanded coverage policy for CS had a significant impact on the number of ILR implantation for CS indication. The diagnostic yield of ILR for AF/AFL detection seems better when ILR is implanted within 2 months than later. Further investigation is needed to demonstrate other clinical benefits and the optimal ILR implantation timing.

Background: Worldwide, sepsis is the leading cause of death in hospitals. If mortality rates in patients with sepsis can be predicted early, medical resources can be allocated efficiently. We constructed machine learning (ML) models to predict the mortality of patients with sepsis in a hospital emergency department. Methods: This study prospectively collected nationwide data from an ongoing multicenter cohort of patients with sepsis identified in the emergency department. Patients were enrolled from 19 hospitals between September 2019 and December 2020. For acquired data from 3,657 survivors and 1,455 deaths, six ML models (logistic regression, support vector machine, random forest, extreme gradient boosting [XGBoost], light gradient boosting machine, and categorical boosting [CatBoost]) were constructed using fivefold cross-validation to predict mortality. Through these models, 44 clinical variables measured on the day of admission were compared with six sequential organ failure assessment (SOFA) components (PaO 2 /FIO 2 [PF], platelets (PLT), bilirubin, cardiovascular, Glasgow Coma Scale score, and creatinine).The confidence interval (CI) was obtained by performing 10,000 repeated measurements via random sampling of the test dataset. All results were explained and interpreted using Shapley’s additive explanations (SHAP). Results: Of the 5,112 participants, CatBoost exhibited the highest area under the curve (AUC) of 0.800 (95% CI, 0.756–0.840) using clinical variables. Using the SOFA components for the same patient, XGBoost exhibited the highest AUC of 0.678 (95% CI, 0.626–0.730). As interpreted by SHAP, albumin, lactate, blood urea nitrogen, and international normalization ratio were determined to significantly affect the results. Additionally, PF and PLTs in the SOFA component significantly influenced the prediction results. Conclusion Newly established ML-based models achieved good prediction of mortality in patients with sepsis. Using several clinical variables acquired at the baseline can provide more accurate results for early predictions than using SOFA components. Additionally, the impact of each variable was identified.