Verification of defibrillation efficacy by defibrillation threshold (DFT) testing during implantable cardioverter-defibrillator implantation is the current standard. Generally, defibrillation of ventricular fibrillation at 10 J below the maximum output of a device is felt to establish an adequate safety margin. Nonetheless, DFT testing adds to cost and carries some potential for morbidity, whereas its impact on outcomes in the modern era of defibrillator technology is unclear. We aimed to determine the frequency that DFT testing resulted in a change at device implant and to identify clinical and echocardiographic predictors of the need for DFT testing.

We reviewed all implantable cardioverter-defibrillators that were implanted at the London Health Sciences Centre (Ontario, Canada) from June 1999 to August 2003 and used multivariate analysis to determine variables associated with DFT test failures and elevated DFT values. When a defibrillation failure was not observed, a lowest energy to defibrillate (LED) was recorded.

Among 168 implants, DFT testing was successful with a minimum 10-J safety margin in 152 (90%), whereas the remaining 16 required changes at device implant. In a multivariate analysis, use of amiodarone was independently associated with DFT failure (odds ratio, 4.6; 95% confidence interval, 1.2-17.0). Significantly higher mean DFT/LED values were observed among patients on amiodarone (1.36 J; P = .0041). Those with nonischemic cardiomyopathy had a higher mean DFT/LED compared with those with ischemic cardiomyopathy (1.44J; P = .028).

Use of amiodarone is associated with a 4-fold increase in risk of DFT failure and subsequent need for changes at implant to achieve a safe threshold. Defibrillation threshold testing appears to be most useful for patients taking amiodarone.