Background-Successful antitachycardia pacing (ATP) terminates ventricular tachycardia (VT) up to 250 bpm without the need for painful shocks in implantable cardioverter-defibrillator (ICD) patients. Fast VT (FVT) Ͼ200 bpm is often treated by shock because of safety concerns, however. This prospective, randomized, multicenter trial compares the safety and utility of empirical ATP with shocks for FVT in a broad ICD population. Methods and Results-We randomized 634 ICD patients to 2 arms-standardized empirical ATP (nϭ313) or shock (nϭ321)-for initial therapy of spontaneous FVT. ICDs were programmed to detect FVT when 18 of 24 intervals were 188 to 250 bpm and 0 of the last 8 intervals were Ͼ250 bpm. Initial FVT therapy was ATP (8 pulses, 88% of FVT cycle length) or shock at 10 J above the defibrillation threshold. Syncope and arrhythmic symptoms were collected through patient diaries and interviews. In 11Ϯ3 months of follow-up, 431 episodes of FVT occurred in 98 patients, representing 32% of ventricular tachyarrhythmias and 76% of those that would be detected as ventricular fibrillation and shocked with traditional ICD programming. ATP was effective in 229 of 284 episodes in the ATP arm (81%, 72% adjusted). Acceleration, episode duration, syncope, and sudden death were similar between arms. Quality of life, measured with the SF-36, improved in patients with FVT in both arms but more so in the ATP arm. Conclusions-Compared with shocks, empirical ATP for FVT is highly effective, is equally safe, and improves quality of life. ATP may be the preferred FVT therapy in most ICD patients.

Warfarin dosing remains challenging due to narrow therapeutic index and highly individual variability. Incorrect warfarin dosing is associated with devastating adverse events. Remarkable efforts have been made to develop the machine learning based warfarin dosing algorithms incorporating clinical factors and genetic variants such as polymorphisms in CYP2C9 and VKORC1. The most widely validated pharmacogenetic algorithm is the IWPC algorithm based on multivariate linear regression (MLR). However, with only a single algorithm, the prediction performance may reach an upper limit even with optimal parameters. Here, we present novel algorithms using stacked generalization frameworks to estimate the warfarin dose, within which different types of machine learning algorithms function together through a meta-machine learning model to maximize the prediction accuracy. Compared to the IWPC-derived MLR algorithm, Stack 1 and 2 based on stacked generalization frameworks performed significantly better overall. Subgroup analysis revealed that the mean of the percentage of patients whose predicted dose of warfarin within 20% of the actual stable therapeutic dose (mean percentage within 20%) for Stack 1 was improved by 12.7% (from 42.47% to 47.86%) in Asians and by 13.5% (from 22.08% to 25.05%) in the low-dose group compared to that for MLR, respectively. These data suggest that our algorithms would especially benefit patients requiring low warfarin maintenance dose, as subtle changes in warfarin dose could lead to adverse clinical events (thrombosis or bleeding) in patients with low dose. Our study offers novel pharmacogenetic algorithms for clinical trials and practice.