How Close Am I? Shocks Decline When ICD Programmers Reminded of Evidence-Based Settings

May 14, 2013

DENVER, CO — A successful trainer likely knows: feedback on a performance should elicit improvement in performance. So it was with some success that regular reports on how clinicians were programming their patients' implantable cardioverter-defibrillators (ICD) made them more likely in the future to program them using evidence-based settings [1]. And that led to significant reductions in number of delivered shocks, both appropriate and inappropriate, in a prospective study covering >4000 patients with ICDs followed in a real-world practice setting.

The device-programming report system tested in the study called Shock-Less represented an old-school approach to an issue explored widely at this year's Heart Rhythm Society (HRS) 2013 Scientific Sessions --namely, how to get ICDs to deliver fewer shocks to rhythms that will not ultimately be fatal.

Importantly in the trial, the "therapy-programming reports" clinicians received were associated with at least 27% fewer of both main categories of unwanted shocks: "inappropriate" ones that are generally in response to supraventricular arrhythmias misinterpreted by the device as ventricular in nature, and "appropriate" shocks for ventricular arrhythmias, specifically those that may well have terminated on their own without being lethal.

Also, consistent with more traditional types of performance evaluation, the process itself seemed to have a positive effect. According to Dr Marc T Silver (WakeMed, Raleigh, NC), who presented Shock-Less at the HRS sessions, simply participating in the study seemed to spur clinicians to program devices more in line with the evidence-based targets. Their performance improved even before the treatment-programming reports were made available for the first time.

Dr Marc T Silver

And that seemed to catch some physicians by surprise. After his presentation, Silver said that, as someone who received the therapy-programming reports himself, "It is a character-building exercise. And if you can get hold of information like this on your own practices, I guarantee you will leave [feeling] a little smaller afterward, like I did."

At total of 4131 patients implanted with single- or dual-chamber Medtronic ICDs, with or without biventricular pacing, for either primary or secondary prevention were enrolled in the study at 118 predominantly North American sites from 2009 to 2012. The clinicians who programmed their ICDs' shock parameters received the therapy-programming reports nine to 12 months later, by which time 2693 patients, acting as a control group, had been entered. The 1438 subsequently entered patients, the intervention group, were similarly followed at 12 months and then every six months.

It is a character-building exercise. And if you can get hold of information like this on your own practices, I guarantee you will leave a little smaller afterward, like I did.

Throughout the study, clinicians programmed every device at their own discretion, and parameters they chose after the first therapy-programming report were compared with those from before the first report.

Access to the reports significantly improved adherence to the evidence-based programming criteria. The biggest improvements, Silver said, were in achieving a 30/40 programming target for number of intervals to detect (NID) ventricular fibrillation (VF) in primary-prevention devices, for which adherence went from 18.5% before the reports to 33.1% after the first reports. Also, the target of wavelet programming "on" was done in 64.3% of cases before and 83.3% after the start of reporting; and the evidence-based <300-ms limit for supraventricular tachycardia was chosen 64.2% and 79.1% of the times, respectively. All those improvements were significant (p<0.0001). Antitachycardia pacing was nearly always chosen regardless of the reports, >98% of the time before and after they started.

There was at least one shock episode in 10% of the patients managed before the reports and 8% of those in the intervention cohort. Every delivered shock was independently adjudicated as either appropriate or inappropriate.

The prevalence of shocks from any cause, the primary end point, dropped by 27% from before to after the start of reporting; interestingly, appropriate and inappropriate shocks each fell by precisely the same percentage. The hazard ratios edged slightly better in analyses adjusted for factors known to influence the likelihood of shock, Silver reported.

Adjusted* hazard ratio (95% CI) for shock episodes, after vs before the start of reporting, by type of shock, in the Shock-Less study

Type of shock episode HR (95% CI) p
All-cause shocks 0.71 (0.57–0.89) 0.002
Appropriate 0.72 (0.55–0.95) 0.021
Inappropriate 0.70 (0.49–0.98) 0.038

*Adjusted for whether it was the patient's first device implant; single- or dual-chamber device; primary or secondary prevention; age; history of smoking, AF, CAD, and coronary bypass surgery; NYHA class and whether ischemic or nonischemic; use of various cardiovascular drugs; and systolic and diastolic blood pressure.

Dr Andrea M Russo

After Silver's presentation, session comoderator Dr Andrea M Russo (Cooper University Hospital, Camden, NJ) commended the study's administrators for "performing your own performance-improvement initiative" and, essentially, disclosing the lag time between ICD-programming evidence-based recommendations and their implementation.

"Opportunity clearly exists for even more improvement," Silver said. "Achieving numbers in the 33% range for 30/40 [VF NID] I think is not what many of us might hope for one day." His own take on what the study implies was a bit ironic, given its promising outcome. Alternative approaches for increasing adherence to recommended programming criteria include getting ICD manufacturers to program the devices' nominal, out-of-the-box settings according to evidence-based guidelines. The effect would be to take the initiative for achieving appropriate programming targets in many cases out of physicians' hands.

"As someone who's become very interested in physician behavior," he said, "I regret to say that I believe that changing nominals may be the best way to move our profession forward. I say that with some degree of regret, but I think that's the truth."

I regret to say that I believe that changing nominals may be the best way to move our profession forward.

Speaking with heartwire , session comoderator Dr Gordon Tomaselli (Johns Hopkins University, Baltimore, MD) agreed that industry should be encouraged to routinely set ICD nominal programming in accordance with evidence-based guidelines. "They can make the nominal settings anything they want, so if they set them so that we have to actively set them away from the guidelines-based programming, that to me is a better situation than having [us] change them to be more like the guidelines."

The study was funded by Medtronic, from which Silver discloses receiving consulting fees or honoraria. Russo discloses receiving consulting fees or honoraria from Cameron Health, Boston Scientific, Medtronic, St Jude Medical, and Biotronik; research grants from Medtronic and Cameron Health; and fellowship support from Medtronic. Tomaselli said he had no disclosures.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.