Background: Reports have demonstrated a circadian variation in the incidence of acute myocardial infarction, ventricular arrhythmias, and sudden cardiac death. We tested the hypothesis that a similar circadian variation exists for defibrillation energy requirements in humans.
Methods and results: We reviewed the time of defibrillation threshold (DFT) measurements in 134 patients with implantable cardioverter-defibrillators (ICDs) who underwent 345 DFT measurements. The DFT was determined in 130 patients at implantation, in 121 at a 2 months, and in 94 at 6 months. All patients had nonthoracotomy systems. The morning DFT (8 AM to 12 noon) was 15.1 +/- 1.2 J compared with 13.1 +/- 0.9 J in the midafternoon (12 noon to 4 PM) and 13.0 +/- 0.7 J in the late afternoon (4 to 8 PM), P < .02. In a separate group of 930 patients implanted with an ICD system with date and time stamps for each therapy, we reviewed 1238 episodes of ventricular tachyarrhythmias treated with shock therapy. To corroborate the hypothesis that energy requirements for arrhythmia termination vary during the course of the day, we plotted the failed first shock frequency for all episodes per hour. There was a significant peak in failed first shocks in the morning compared with other time intervals (P = .02).
Conclusions: There is a morning peak in DFT and a corresponding morning peak in failed first shock frequency. This morning peak resembles the peaks seen in other cardiac events, specifically sudden cardiac death. These findings have important implications for appropriate ICD function, particularly in patients with marginal DFTs.