Do DH, et al. - As in-hospital cardiac arrest (IHCA) due to pulseless electrical activity (PEA)/asystole is correlated with poor survival, researchers examined the predictive value electrocardiographic changes for risk of IHCA from PEA/asystole. In this retrospective case-control study, they used continuous electrocardiographic data from case and control patients. Three consecutive 3-hour blocks (block 3, 2, 1 in that order) were selected; in cases, block 1 immediately preceded cardiac arrest, whereas in controls, block 1 was chosen at random. On evaluating 91 cases (age 63.0±17.6, 58% male) and 1783 control patients (age 63.5±14.8, 67% male), they identified significant differences in electrocardiographic trends between case and control block 1, particularly in QRS duration, QTc, RR, and ST. A more common occurrence of new episodes of atrial fibrillation and bradyarrhythmias was observed before IHCA. The random forest was identified as the optimal model, that achieved an AUC of 0.829, 63.2% sensitivity, 94.6% specificity at differentiating case vs control block 1 on a validation set, and AUC 0.954, 91.2% sensitivity, 83.5% specificity at differentiating case block 1 vs case block 2. These findings suggest that relative to other time periods, the 3-hour window immediately preceding in-hospital cardiac arrest show significantly different trends in electrocardiographic parameters, and provide strong predictive information.