In the decision as to whether to replace an implantable cardioverter
defibrillator (ICD) that has become subject to a safety advisory,
the guiding equation is more likely to be "risk vs risk" than risk
vs benefit. The potential hazards of leaving a device in depend on
the safety issue and can be apparent, the risks inherent to device
replacement may be less well defined, and knowing what can promote
them would presumably inform the decision.

In a retrospective experience from a dozen major device-implantation
centers across Canada [1], published in the December 2008 issue of
Heart Rhythm, advisory-ICD-replacement complication rates were in
line with prior studies of the issue and complication risk factors
included some that make perfect sense but at least one that was
unexpected.

One message from the case-control study, according to principal
coauthor Dr Andrew D Krahn (London Health Sciences Centre, ON), is
that ICD-generator replacements are not as easy as often
perceived. "From a surgical perspective, this is a really, really
simple operation. It may feel like nothing can go wrong when it
takes you 20 minutes and by the time you're done it feels like
you've barely started," he told heartwire.

"Whereas when you have to address lead issues, it¡¦s a more involved
procedure, and the mind connects that with more to go wrong."

But, he said, the risks of generator replacement appear to be on the
same order of magnitude.

Krahn said the rates he usually cites for lead extraction are about
2% for major complications and 0.5% for death. In the current
analysis of 451 patients with advisory-ICD replacement followed for
a year, 6.0% developed major complications; the rate included a
0.44% mortality. Major complications were defined as death, nonfatal
MI, cardiogenic shock, or reoperation due to hematomas, infections,
malfunctions, or significant pain.

Pocket infections requiring device extraction accounted for more
than a third of the major complications and were involved in both of
the deaths.

The major-complication rate compares with the 5.8% rate for advisory-
ICD replacements over a mean of 2.7 months at 17 Canadian centers
reported by many of the same coauthors in a 2006 issue of the
Journal of the American Medical Association [2].

In the current analysis, the rate of minor complications was 3.1%;
they consisted entirely of medically managed pain or incisional
infection and postoperative exacerbation of a medical condition.

As noted in the report, with first author Dr Paul A Gould
(University of Western Ontario, London), "Data from our current
study suggest that ICD-generator advisories are potentially as
complex to intervene on as advisories with ICD leads."

In an accompanying editorial, Dr Mark H Schoenfeld (Yale University,
New Haven, CT) makes the point with even greater urgency [3]. "What
is clear from this study, and all too often underappreciated, is
that there is no such thing as a 'minor' procedure, and that
replacing a device under advisory cannot be considered 'just' a
generator replacement."

In a case-control analysis that looked for independent risk
predictors, the 41 patients who suffered complications were matched
by age and sex with 82 patients in whom advisory ICDs were safely
replaced.

From among a long list of potential risk factors, two emerged as
independently significant in multivariate analysis. The risk of
complications more than doubled for every procedure previously
performed on the same pocket (odds ratio 2.53 [95% CI 1.14¡V5.62];
p=0.022).

That makes sense, according to Krahn. "When patients have a
pacemaker or ICD put in for the first time, infection is quite
unusual. When they come in to have a device or lead replaced, or
anything that involves going back into a scarred area, that's when
infections are more likely to happen." Their data, he said, suggest
that multiple procedures up the risk of complications in general.

Also, surprisingly, according to Krahn, the safest advisory-device
replacements apparently were performed by two operators, generally a
fellow working with a more experienced physician.

"This is an operation that relatively junior people do without a
great deal of supervision because it's so simple, usually," Krahn
said, "and yet having two people present is associated with a little
lower risk." The difference in the analysis could have been by
chance, he said, "but the other possibility is that there's
something about two people keeping an eye on each other that makes
them that much more vigilant or fastidious."

According to the editorial, "Whether enhanced vigilance explains
the 'protective' effect of replacement procedures performed by two
operators . . . is unclear. . . . What we do know is that there is
an inverse correlation between patient outcomes and the volume of
ICD-implantation procedures performed by a physician, and that has
not been reflected in the present study."


Gould PA, Gula LJ, Champagne J, et al. Outcome of advisory
implantable cardioverter-defibrillator replacement: 1-year follow-
up. Heart Rhythm 2008; DOI:10.1016/j.hrthm.2008.09.020. Available
at: http://www.heartrhythmjournal.com.
Gould PA, Krahn AD, for the Canadian Heart Rhythm Society Working
Group on Device Advisories. Complications associated with
implantable cardioverter-defibrillator replacement in response to
device advisories. JAMA 2006; 295:1907-1911. Abstract
Schoenfeld MH. The "natural" history of implantable defibrillators
under advisory. Heart Rhythm 2008; DOI:10.1016/j.hrthm.2008.10.010.
Available at: http://www.heartrhythmjournal.com.