By Cara N. Pellegrini, MD
Assistant Professor of Medicine,
San Francisco VA Medical Center
Dr. Pellegrini reports no financial relationships relevant to this field of study.
SOURCE: Healey JS, et al. Cardioverter defibrillator implantation without induction of ventricular fibrillation: A single-blind, non-inferiority, randomized, controlled trial (SIMPLE). Lancet 2015;385:785-791.
Traditionally, the placement of an implantable cardioverter-defibrillator (ICD) concludes with induction of ventricular fibrillation (VF) to confirm that the new device is able to appropriately sense the arrhythmia and terminate it while there is still the opportunity to reposition the ventricular lead or add an additional lead if necessary. This defibrillation testing has never been shown to independently decrease mortality or improve outcomes. A growing number of implanting physicians are moving away from performing defibrillation testing routinely due to greater confidence in the reliability of modern-day, high-energy devices and increasing concern about the potential deleterious effects of even brief VF and/or the high-current density of the defibrillation itself.
The multinational SIMPLE study, led by Drs. Healey and Connolly at McMaster University in Ontario, Canada, examined whether abstaining from defibrillation testing at initial device implant was noninferior to testing, with regard to success of first appropriate shock and occurrence of arrhythmic death. They enrolled 2500 patients into this randomized, controlled trial, with broadly defined inclusion criteria. Over a mean follow-up of 3.1 years, 90 patients who did not undergo defibrillation testing had experienced a failed appropriate shock or arrhythmic death, as compared to 104 patients in the testing group, a nonsignificant difference (hazard ratio [HR], 0.86; confidence interval [CI], 0.65–1.14, P non-inferiority < 0.001). Safety endpoints and overall mortality were similar between groups, with no deaths in this study directly attributable to defibrillation testing. Notably, about 1% of all patients having defibrillation testing needed chest compressions or an unplanned intubation, compared to 0.1% in the no-testing group (P = 0.007). The authors concluded that there was no clear benefit to routine defibrillation testing with regard to shock efficacy or arrhythmic death prevention, and that complications from testing appeared to be rare.
This well-conducted study has two main messages. First, not testing should probably be the default. They made a special effort to minimize exclusion criteria and make their study very generalizable. A bit more than one-third of their patients were receiving an ICD for secondary prevention, 30% had NYHA class III heart failure symptoms, and almost 30% received a resynchronization device (biventricular defibrillator). These patients are among those who are most likely to have higher defibrillation thresholds, making testing theoretically more important, and the lack of difference between groups despite their meaningful inclusion is notable. As those getting a resynchronization device have the distinction of receiving disease-modifying therapy with the potential for improvement of their ventricular function, and also being at higher risk for heart failure exacerbation with defibrillation testing in subgroup analysis, this group should be specifically spared from defibrillation testing where possible. Some benefits of not testing are not even accounted for in this study, in which presumably both groups underwent similar anesthesia protocols to achieve patient blinding. By not performing defibrillation testing, the risks and costs of deeper levels of sedation can be eliminated.
The second important lesson is that, should testing be judged appropriate for a certain patient, it can probably be performed with minimal additional risk. Patients receiving right-sided devices — where the vector of defibrillation may be less favorable — putting them at higher risk for a failed shock, were explicitly excluded from this study (although they did constitute about 1% of each arm). Patients with subcutaneous ICDs were also excluded. Data are sparse for those with hypertrophic cardiomyopathy, channelopathies, or sarcoidosis; congenital heart disease patients also may be expected to have higher defibrillation thresholds and were not studied. Finally, those with a recalled lead or starting a new antiarrhythmic medication that could increase defibrillation threshold may merit testing. Indeed, individualized risk-benefit calculus is appropriate here.
Not addressed by this study is the role of other potentially safer methods of assessing the integrity of the ICD system, such as upper limit of vulnerability testing. This technique utilizes the relationship between the energy required for VF induction and that required for VF termination to determine the presence of an adequate safety margin, with VF being induced in only the minority of procedures. Where device testing is believed to be necessary, this practice may further lower risk. For the majority of patients, though, we now have data that are at least just as good to keep it SIMPLE.