Boston, MA - Clinical outcomes associated with dual-chamber and single-chamber pacing can be similar in patients with implantable cardioverter defibrillators (ICDs), according to a randomized study of dual-chamber devices with a programming feature that strictly limits ventricular activation in favor of the heart's native pacing capability [1,2].

Dr Brian Olshansky

The finding may alleviate concerns based on recent clinical trials that dual-chamber rate-adaptive pacing (DDDR) can worsen clinical outcomes compared with ventricle-only backup pacing (VVI) in patients with ICDs [3]. The new study, Inhibition of Unnecessary RV Pacing with AV Search Hysteresis in ICDs (INTRINSIC RV), conducted at 108 centers, supports prior suggestions that any adverse clinical effects derive from how much right-ventricular (RV) pacing the patient experiences, rather than dual-chamber pacing capability per se.

"The notion that dual-chamber programming poses an inherent safety risk is incorrect," Dr Brian Olshansky (University of Iowa, Iowa City) said when presenting the INTRINSIC RV study here at the Heart Rhythm Society 2006 Scientific Sessions. When ICDs in the trial were set to the dual-chamber mode and programmed to limit RV pacing using a programming feature called atrioventricular (AV) search hysteresis (AVSH), according to Olshansky, "outcomes with dual-chamber programming were as good as if not better than with single-chamber programming." The trial was powered to discern "noninferiority," but a post hoc analysis tantalized with a possible trend for DDDR superiority over VVI.

The findings are a "major call" for rethinking the role of DDDR programming in patients with ICDs but no consistent need for antibradycardia pacing, Olshansky told heartwire. Dual-chamber pacing largely fell out of favor in that population after the release of several studies, most notably the Dual Chamber and VVI Implantable Defibrillator (DAVID) trial published in 2002 [3].

As reported then by heartwire, the DAVID trial randomized 506 ICD recipients with mostly ischemic cardiomyopathy and an LVEF <40% to have their devices programmed to either DDDR with RV pacing initiated at a lower rate of 70 bpm or VVI backup pacing activated at 40 bpm. The DDDR mode was associated with a 61% increase in the relative risk of death or HF hospitalization (95% CI 1.06-2.44) over one year.

"Now we have some hints that there might at least be some subgroups that will do better with DDDR-AV search hysteresis pacing," although any markers for such patients have yet to be identified, said Olshansky when interviewed. As for the general applicability of the INTRINSIC findings, "We took a wide range of patients of all types," he said. "They came from academic centers, from private practices, from all over the country and in fact internationally."

According to Dr Bruce L Wilkoff (Cleveland Clinic Foundation, OH), his center still implants dual-chamber pacemakers but "a lot less often" than before the DAVID trial, for which he was lead author. "And until we can show some significant disadvantage to them, rather than a lack of a disadvantage, I think we'll continue to use a lot of single-chamber devices," he told heartwire. These days, the dilemma for a given patient is more likely to be "whether to go with a single-chamber or biventricular device."

Encouraging the INTRINSIC rate

Dr Bruce L Wilkoff

The INTRINSIC RV investigators implanted Guidant VITALITY AVT devices in 1530 patients with standard ICD indications and set them to the DDDR mode, with AVSH programmed to activate ventricular pacing at a lower threshold of 60 bpm. After seven days, they excluded patients who had shown an RV pacing prevalence of at least 20%. They then randomized the remaining 988 patients to continue with the same pacing criteria or have their devices reprogrammed to the backup VVI mode at a lower rate threshold of 40 bpm.

Over a mean of 10.4 months, Olshansky reported, 32 and 46 patients in the DDDR-AVSH and VVI groups, respectively, met the primary end point of all-cause death or HF hospitalization. The numbers were statistically the same. Mortality was also similar in the two groups.

Outcomes in INTRINSIC RV, DDDR-AVSH vs VVI pacing in ICD patients

End point	DDDR-AVSH, n=502	VVI, n=486	p*
All-cause mortality or HF hospitalization (%)	6.4	9.5	<0.001
All-cause mortality (%)	3.6	5.1	<0.001

The post hoc superiority analysis gave the composite-end-point difference between the two pacing modes a p value of 0.072, Olshansky reported, suggesting that appropriately powered trials might later find a clinical advantage for dual-chamber pacing.

For Wilkoff, the trial's "major design issues" include statistical power that limited the end-point comparison to noninferiority and the exclusion of more than a third of the original INTRINSIC RV cohort—those with a >20% RV-pacing prevalence in which AVSH was least likely to prevent ventricular pacing. In contrast, he noted, the DAVID trial took "all comers."

Wilkoff noted, however, that the hint of possible DDDR superiority might bode well for atrial-based pacing (how DDDR programming works when it's not pacing the right ventricle) compared with VVI, which is being tested in the ongoing DAVID-2 trial.

The INTRINSIC twist on DDDR pacing

Although INTRINSIC RV used Guidant devices, its findings aren't necessarily specific to that company's products, Olshansky observed for heartwire. The AVSH programming option is available from several manufacturers that use different AV hysteresis algorithms, especially in biventricular devices for patients with HF, he said. It allows the device to search automatically for native AV conduction and, within the limits of its programming, extends the AV-delay interval accordingly to allow natural RV pacing.

"Unless the patient has chronic atrial fibrillation, it makes the most sense to me to put in a dual-chamber device and program it to minimize ventricular pacing down to less than 20%," according to Olshansky. He favors a more consistent use of DDDR units over their selective use in ICD patients currently needing more than VVI. "It gives patients more options later," he said, by helping to avoid the risks and costs of device replacement should their pacing needs increase later. "In my opinion, most patients would actually do better getting a dual-chamber device implanted first."

According to Wilkoff, patients with VVI-only devices rarely deteriorate to the point of needing a DDDR-device upgrade. The risk of that happening has been only 4% in some trials, so any total cost savings from consistent use of DDDR units as a hedge against future pacing needs probably won't be great, he said.

However, with their additional atrial lead compared with the single lead in VVI-only devices, DDDR-capable ICDs do have advantages, agreed Wilkoff. The atrial lead can enhance the thoroughness of patient follow-up by allowing more sophisticated diagnostics, such as discrimination between atrial and ventricular arrhythmias, he said. And some investigators believe atrial pacing may reduce the risk of atrial fibrillation.

"The downside is that more things can fail. You have an extra lead, you get more venous occlusion, implantation time is longer, and infection rates could be higher," Wilkoff said. "And it certainly costs more to have a dual-chamber rather than a single-chamber device."

INTRINSIC's greater proportion of patients with NYHA class 3-4 HF could help account for differences between its results and those of DAVID, according to Olshansky. But "the more likely explanation," he said in his presentation, is that the mean RV pacing rate in INTRINSIC was 10% as compared with the "dramatically higher" 58% observed in the earlier study.

To support that idea and show some common ground between the two trials, Olshansky displayed a DAVID subanalysis in which an RV pacing rate exceeding 40% in the DDDR group was associated with more than triple the risk of a primary end point compared with rates <40% (p=0.03). The end point rate in the DDDR group with >40% RV pacing also trended greater than the rate for patients randomized to VVI pacing (p=0.07).

Those findings could suggest that an RV pacing rate <40% "may be acceptable," Olshansky said. But, "bearing in mind the dangers of overinterpreting nonrandomized post hoc analyses," he cautioned, an INTRINSIC subanalysis suggested that even lower is better: the risk of death or HF hospitalization dropped steadily as the prevalence of RV pacing fell from >50% to 10% to 19%.

Olshansky reports he is a consultant and on the speakers' bureau for Boston Scientific, the parent company of Guidant. Of his coauthors, Dr Scott Brown is a consultant for and Drs Maureen McGuire and Darin R Lerew are employees of Boston Scientific. Wilkoff reports having served as a consultant for and received research funding from Medtronic, Guidant, and St Jude Medical.