Preventing Inappropriate ICD Shocks

Abstract & Commentary

By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville.

Source: Moss AJ, et al. Reduction in inappropriate therapy and mortality through ICD programming. N Engl J Med 2012;367:2275-2283.

This study, the Multicenter Automatic Defibrillator Implantation Trial – Reduce Inappropriate Therapy (MADIT-RIT), was designed to test the hypothesis that programming implantable cardioverter-defibrillator (ICD) therapies with higher detection rates or delayed detection before initiation of therapy would decrease inappropriate shocks in ICD patients. The study enrolled 1500 patients at 98 ICD implanting centers. Patients could have any form of ischemic heart disease, had to be in sinus rhythm at the time of enrollment, and met approved criteria for implantation of an ICD or CRT-D device for primary prevention of sudden death. Patients were randomized to three types of programming. The conventional therapy group received a device programmed with two detection zones: 1) a ventricular tachycardia zone for heart rate between 170-199 beats per minute with a 2.5-second delay and atrial discriminators, and 2) a second zone beginning at 200 beats per minute for faster VT or VF with a 1-second delay before delivery of therapy, either antitachycardia pacing (ATP) or shock. Group two patients received a device programmed with a monitor only zone between 170-199 beats per minute with a therapy prescribed at 200 beats per minute after a 2.5-second delay. Patients in the delayed therapy group received a device programmed to three detection zones: one at 170-199 beats per minute with rhythm detectors on and a 60-second delay before initiation of therapy, a second VT detection zone beginning at 200 beats per minute with rhythm detection on and a 12-second delay before therapy, and a third zone at 250 beats per minute with a 2.5-second delay before initiation of therapy. The devices used in this trial were commercially available from Boston Scientific. The decision for dual-chamber ICD vs a CRT-D device was at the discretion of the investigator. The primary endpoint was the first occurrence of inappropriate therapy, including both ATP or shock. All therapy electrograms were reviewed by a core laboratory and classified as appropriate or inappropriate by a three-member committee. The secondary endpoints were death from any cause and first episode of syncope.

The clinical characteristics of the 1500 patients enrolled were similar to those for all patients receiving primary prevention ICDs in the United States, except for a younger mean age. Patients with both ischemic and non-ischemic heart disease were included in equal proportions. Approximately equal numbers of patients received dual-chamber ICDs and CRT-D devices. As compared with the conventional therapy group, the high rate and delayed therapy groups were less likely to receive both appropriate therapy and inappropriate therapy during a follow-up duration of 1.4 years. In the conventional therapy group, 22% received appropriate therapy and 20% received inappropriate therapy. In the high-rate therapy group, 9% of patients received appropriate therapy and 4% received inappropriate therapy. In the delayed therapy group, 6% of patients received appropriate therapy and 5% received inappropriate therapy. This reduction in therapy was due to much less utilization of antitachycardia pacing. When the total occurrences of therapy were examined, application of antitachycardia pacing in the high-rate therapy and delayed therapy group was reduced by more than 75% compared to the conventional therapy group. Life-table estimates of the time to first occurrence of inappropriate therapy showed that the conventional therapy group had a 29% probability of inappropriate therapy at 2.5 years while the inappropriate therapy rate at that time point was only 6% in both the high-rate and delayed therapy groups. Mortality was also reduced in both the high-rate therapy and delayed therapy groups. There were 34 deaths in the conventional therapy group, compared to 16 deaths in the high-rate group and 21 deaths in the delayed therapy group. The incidence of syncope was similar in the three groups.

The authors conclude that ICD programming to less aggressive detection settings reduces inappropriate and unnecessary therapies and may have beneficial effects on mortality.


It’s not a real surprise that less aggressive detection zones will reduce therapy delivery in primary prevention ICD recipients, but MADIT-RIT provides the most definitive data supporting this conclusion. What is an important finding is that there appears to be additional benefits in terms of overall mortality with almost no downside. We’re not told if there were any episodes of slow sustained VT without syncope that fell below the programmed detection criteria, but this should be an unusual occurrence in a primary prevention population. The incidence of syncope was the same in all three groups, pointing out that not all syncope in low ejection fraction cardiac patients is due to VT.

Most ICDs in the United States are placed for primary prevention, but the default settings are more appropriate for patients with a history of sustained monomorphic VT. Although wavelet morphology and atrial discrimination algorithms may be helpful, MADIT-RIT shows that they are often inadequate tools for preventing clinically unnecessary shocks. Manufacturers should consider changing the default settings while still allowing electrophysiologists to modify the programming when needed.