Stockholm, Sweden - As indications for implantable cardioverter defibrillators (ICDs) broaden and patients receive them earlier in their disease course and live longer, the devices have become smaller and more versatile than ever. Those trends are on a kind of collision course, as increasing energy demands shorten device longevity, which means that more patients may eventually need multiple replacements during their lifetimes, perhaps with unforeseen consequences.

"The trend toward miniaturization is driving battery size down," Dr Robert G Hauser (Abbott Northwestern Heart Hospital, Minneapolis, MN) told heartwire. Meanwhile, the increase in device features, particularly cardiac resynchronization capability (CRT-D), is further shortening the average device's service life. "There's no doubt that, if the current trend continues, these devices will not last as long as we would like or as long as the average patient lives." The potential implications are higher total device costs and hospital charges, which would increase demands on reimbursement resources, according to Hauser.

Dr Bruce Wilkoff

As for the potential clinical cost—arising from more surgeries, lead failures, and pocket infections or hematoma—experts differ on its severity but agree that the risk rises with the number of device replacements. "Every time you change a device, there's risk," observed Dr Bruce Wilkoff (Cleveland Clinic Foundation, OH) for heartwire. "Once you open up the pocket, you increase the risk of lead failure, increase the risk of infection."

Although battery depletion is the most common reason for device replacement, upgrades and lead failures are also a cause. When old leads are exchanged for new ones, it's routine to implant a new pulse generator, according to Wilkoff. "You're not going to leave in an old device with new leads [when] the device is over 50%, maybe 80%, depleted." Also, he noted, many patients come to need a more versatile device; for example, their device may be switched from a single- or dual-chamber ICD to a CRT-D unit because they've developed symptoms. But "not everybody's going to need increased therapy, so you want to make change-outs for battery depletion as infrequent as possible.

"It would be nice if there were only one or two generators in a person's lifetime instead of three or four," Wilkoff said. "I think the devices should have a six- or seven-year longevity. That would be ideal."

Dr Robert G Hauser (Source: Minneapolis Heart Institute at Abbott Northwestern)

Although there are few, if any, reported data showing a rise in device-replacement rates, Hauser and colleagues recently presented numbers pointing to such a trend based on 1279 patients who received a total of 1429 ICDs of various designs at their center over the past 11 years [1]. In their report, presented at the recent European Society of Cardiology Congress (ESC) 2005, cumulative survival curves for patients with LVEF <40% compared with patients with LVEF >40% at device implantation were superimposed on "time-to-device-failure" plots, broken out by type of ICD. Device failure was defined as battery depletion, malfunction due to electronic or device-housing defect, or removal because of a recall or safety alert.

The graph suggests that "the cumulative probability that a single-chamber device will fail at just over five years is 50%," Hauser said. "The average patient with an ejection fraction <40% is going to live a bit over eight years. That tells you that most of these patients are going to require more than one ICD implant." Dual-chamber devices were serviceable for a slightly shorter period, increasing the likelihood of a second implant. And if the trend continues with the CRT devices, it could be that those patients are going to require two replacements, he said. "The left-ventricular lead requires more current to function, and that's driving early battery depletions around the country."

Dr Jeffrey L Anderson(Source: University of Utah)

"These trends are exactly what we're experiencing here," Dr Jeffrey L Anderson (University of Utah, Salt Lake City) told heartwire. "We're putting in devices that use more energy, that are more expensive, and that require more frequent replacement as people live longer. There's no question about it." The range in service life as device complexity varies can be wide, he says, noting that a conventional pacemaker he recently replaced in a patient had been in service nine to 10 years. "With a standard ICD, we can expect them to go five or six years. With CRT devices—I think I'm replacing the generators at about four years. . . . And if they have an ICD and they're using CRT, they're going to need replacement sooner."

Warren Watson (Source: Medtronic)

Although the balance of device and patient longevity may seldom be discussed with patients, industry routinely considers it, according to Warren Watson, vice president for Cardiac Rhythm Management, Implantable Product Development, at Medtronic. "It's a topic that gets discussed virtually every time we talk with our [physician] consultants on new product ideas."

Major design factors that influence ICD longevity include the battery's chemistry, its size relative to the capacitor, and the circuitry's energy efficiency, according to Watson, who joined Medtronic 29 years ago with an engineering background. "It's always a trade-off," he said.

But according to Watson, the shrinking size of ICDs since they were introduced has been accompanied by much greater longevity. The company's first ICD had a volume of 209 cc and lasted about two years, he said; today single- and dual-chamber units are about 35 cc with a six- to seven-year longevity, depending on how often they discharge. Even today's CRT-D devices, including the bulky connectors for at least three leads, are about 45 cc and may last three to four years, Watson said.

But given that the addition of CRT hastens battery drain, plus data from Hauser et al indicating that CRT-D implantations are poised to overtake the use of ICD-only devices, if they haven't already, the inverse relationship between device size vs longevity outlined by Watson seems likely to reverse.

Dr Stuart J Connolly

Dr Stuart J Connolly (McMaster University, Hamilton, ON) agrees that the addition of CRT capability to ICD technology accelerates battery drain. A CRT device "turns out to be fairly cheap when it's piggybacked on top of the defibrillator, but the cost of shorter longevity is often not factored in, people forget about that," he said to heartwire. The number of patients receiving the devices at his center, he said, has "more than doubled" over the past five years, but the replacement rate "has not changed dramatically over time."

Clinical implications . . . 

Anderson agrees that the pocket-infection risk is "at least as high" for replacements as it is for initial devices. Still, he said, the risk of major nonfatal complications such as severe infection or lead failure may be only 1% to 2%, and the risk of a fatal event "is something like a tenth of that." Anderson also said lead complications are more likely related to the initial procedure. "If the leads are in, they generally don't require replacement at nearly the rate that the pulse generators do," he said. "There's much less risk on the second go-round."

However, the risk may go up with the placement of additional leads, which are often needed for device upgrades. In a review of 104 patients who received single-chamber and 74 patients who received dual-chamber ICDs, device-related complication rates were 6% and 16%, respectively (p=0.01) [2]. Complications included lead dislodgment, ventricular-lead malfunction, and pocket infection or hematoma.

The limited number of reports focusing on the clinical impact of ICD system replacement are inconsistent. In a recent literature review [3], the complication rate associated with pacemaker replacement was found to be three times higher than that for initial insertions. Complications included infection, skin erosions, and lead-related problems. But a single-center review of 1170 patients who received ICDs showed a 1.2% rate of associated infection, with no difference between patients receiving initial devices and those receiving replacements [4]. When pocket hematoma developed in nearly 5% of 3000 pacemaker or ICD implantations in a recently reported series [5], the complication prolonged hospitalization in 2% and required another surgery in 1% of the total group.

. . . And impact on costs

"The cost of providing therapy for patients with ejection fractions <40% over the lifetime of these patients is going to increase, and it's going to increase substantially," Hauser said, adding that his pulse-generator replacement-cost analysis doesn't even consider the inevitable rise in hospital charges, clinician fees, and coverage of surgical complications. He referred to his recently published Viewpoint [6], in which he estimated that 70% of the more than 100 000 ICDs implanted in 2004 will be replaced because of battery depletion. The replacement pulse generators alone will cost an estimated $1.4 billion, according to the article, based on data from Hauser's institution and a multicenter registry he helped establish.

In their report at the ESC, Hauser and colleagues separately analyzed their institution's numbers from the past five years, a period when CRT emerged and blossomed as a mainstream HF therapy. Results showed a steady growth in ICD therapy paralleled by an explosion in the proportional use of more expensive and energy-hungry CRT-D devices, along with a steep and "disturbing" rise in overall device costs. "As indications expand, longer-lived ICDs will be needed if this therapy is going to be cost-beneficial," the group concluded.

Single-center total ICD-equipped implanted device use and total cost of devices by year, 2000-2005

Parameter	2000	2001	2002	2003	2004
Total ICDs (n)	148	144	201	282	326
ICD only (%)	98.6	96.5	81	68	56
ICD+CRT (%)	1.4	3.5	19	32	44
Total ICD cost (millions of dollars)	2.60	2.56	3.86	5.75	7.02
Total ICD-only cost (millions of dollars)	2.55	2.43	2.84	3.35	3.18
Total CRT+ICD cost (millions of dollars)	0.06	0.13	1.01	2.40	3.84

Anderson states: "I guess it's 'disturbing' if you're paying the bills. On the other hand, if you're the patient and there's extension of life, it may be good. . . . I think that, yes, there are things to consider. And yes, we ought to think about it a little more and put it into perspective." But the established efficacy of device therapy is so great, he said, that in the absence of a full analysis of risk and cost vs benefit suggesting otherwise, Hauser et al's prediction concerning growth in total device cost "isn't a showstopper."

Wilkoff seems to agree. Although he supports the availability of longer-lived pulse generators, he said, the cost and use trends from Hauser and colleagues "are just life." This kind of controversy about ICD therapy "wouldn't be happening if it didn't work," he said. "Patients are having symptoms that were not treatable before but are now treatable." With upgrades, he noted, "they feel better, live longer—and yes, it costs money to make people feel better and live longer."

Challenges to larger pulse generators

"We do not have the choice of implanting large-battery-capacity single-chamber or CRT ICDs because they are not available," Hauser wrote in his Viewpoint. "The ICD manufacturers have no incentive to provide long-lived pulse generators. Indeed, the opposite is true—frequent replacements increase sales and profits."

Wilkoff agrees that things would change if device manufacturers were given incentive to market longer-lasting ICDs. "People are not asking for longevity, they're asking for small," he said. "That's what sells." Patients are seldom clearly given a choice between smaller device size and longer device longevity, according to Wilkoff, because doctors realize that "they usually can't feel what's going to happen three or four years from now. They just see that they want it small and comfortable now."

Still, Wilkoff said, it should be up to clinicians to emphasize the value of ICD longevity for patients. "If doctors all said they needed seven-year devices, we'd have them."

He observed also that both doctors and hospitals profit from device upgrades and other replacements. But "there's nothing nefarious going on. There's a dynamic ambivalence about all this. . . . You don't know when somebody's going to develop heart failure, when you're going to need another lead. So the tendency is to try to do as much as you can, put in the best device with the most features, and make it the most flexible."

"We obviously have to convince physicians that a little larger ICD is going to provide more longevity and is more cost-effective and patients are going to appreciate it," Hauser said. As for the last point, he referred to a 2004 survey [7] that suggested the overwhelming majority of patients would trade small size for a longer-lived device—as long as they understood the implications of their choice: a smaller device that requires more frequent surgery or a larger one that lasts longer. Of 156 patients about to receive ICDs who were asked their preference, about 90% chose the larger device and 10% the smaller one (p<0.0001), independent of sex or body habitus.

How to improve ICD-battery longevity

Practice guidelines for pacemaker systems suggest a way to prolong battery life that may apply to devices that include CRT capability: tailor the programming to individual patients more consistently. "Optimal programming of output voltages, pulse widths, and AV delays can markedly decrease battery drain," note joint guidelines from major cardiology societies [8]. They refer to data suggesting that "expert programming of pacemaker generators can have a major impact on longevity, prolonging it by an average of 4.2 years compared with nominal settings."

Watson doubts that ICD service life overall will improve much with tailored programming. "The device is programmed when it's in the box, and most doctors will leave it that way," he said. "It's already optimally programmed for the vast majority of patients." However, he added, "if physicians choose not to program the antitachycardia pacing feature 'on,' they are missing an opportunity to affect the longevity of the device." Watson said Medtronic has data suggesting that antitachycardia pacing can reduce the number of high-energy shocks by 71%.

In his Viewpoint article [6], Hauser suggested that reimbursement incentives might shift preferences toward longer-lasting devices. "Several approaches could be adopted by payers and offered by manufacturers, including a program that would cap device costs for the life of an individual patient and reward ICD longevity," thereby reducing an individual's total number of devices and surgeries.

Pressuring manufacturers to market larger but longer-lived devices wouldn't work without changing physicians' preferences for smaller units, Wilkoff said. "I think education would be part of the solution, which is what Bob Hauser is doing."

Watson noted that CRT development is in its infancy, which provides the best opportunity for design changes to make an impact on ICD longevity. "CRT-D devices are really generation 1.5 at this point. We have focused on ensuring they have the current technology, but we have a long way to go to apply more leading-edge technology to reduce the size and improve longevity. If you look at the evolution of single-chamber or dual-chamber ICDs, you'd see the same sort of pathway of evolution."

Hauser, a former CEO of Cardiac Pacemakers, a company later purchased by Guidant, says that after leaving the company he divested himself of any equity in and other financial relationships with implantable cardiac rhythm device companies. Anderson says he has no equity interest in or financial relationships with medical product companies. Connolly reported that he has received research grants and consultancy fees from Guidant and St Jude Medical. Wilkoff said he is an advisor to and consultant for Medtronic, Guidant, and St Jude Medical and has participated in research sponsored by all three companies.