ICD Mortality Trials

	5-24-99 - It has been less than 20 years
since the first ICD was implanted in 1981. It is now
taken for granted that the ICD has a benefit in 
prolonging life. Quality of life studies have not been
done in detail but what studies are available do NOT
suggest a major impact of the ICD on quality of life.
Patients who have an ICD may have fewer hospitalizations.

Trial Results

	The MADIT trial's results support use of the ICD
as a preventive therapy. MADIT chose patients with 
coronary artery disease and low ejection fraction who also
had nonsustained VT. These patients had baseline EPS and
if they had inducible VT that was not suppressed by 
procainamide, they were given either an ICD or 
conventional therapy - which in many patients included
amiodarone.
	The major result of MADIT was that there was close
to a 50% reduction in mortality. MADIT has been accepted
by many doctors as very strong evidence in favor of using
prophylactic ICD therapy.

	The Multicenter UnSustained Tachycardia Trial 
(MUSTT) further supports the ICD being useful in patients
who have not yet had a VT episode, but who are at high risk
for sudden death. Patients had an EF less than 40% and at
least one episode of unsustained ventricular tachycardia.
They had a baseline EPS. If they had inducible VT, they got
either guided anti-arrhythmic therapy or no anti-arrhythmic
therapy.
	All patients received best other medical therapy.
The patients who got EP-guided therapy almost always got an
anti-arrhythmic drug and had follow-up EPS. If VT was no
longer inducible, this was considered success of the anti-
arrhythmic drug and patients were then treated with that
drug.
	In reality however, it was difficult to find a drug
that suppressed VT and half the patients in the guided 
therapy group of MADIT got an ICD. So, as the study went on,
more and more patients got ICDs.
	The original idea was that most patients would end
up on drugs but by the end of the study most patients were
ending up with an ICD. The main result of MADIT was a
reduction of arrhythmic death and cardiac arrest. Most of 
the benefit of the guided therapy approach seemed to come
from ICD use. However, ICD use in that group was not 
randomized, so it is hard to draw conclusions about whether
the ICD is as beneficial as it seems in this trial.
	Still, it is widely accepted that in spite of the
MADIT problems, MUSTT provides support for primary 
prevention of sudden death with an ICD.

To Implant or Not to Implant

	It seemed that several of the panel members were
eager to implant ICDs. One doctor would implant an ICD in
just about any patient he sees who has a low EF and who has
unsustained ventricular tachycardia. He doesn't think doing
an EP study is necessary or even advisable.
	Another doctor reminded us of one negative trial.
The CABG-Patch trial studied patients who had bypass 
surgery and who also had low EF and a positive signal-
averaged ECG. Patients were given an ICD or were not, and 
were followed for up to several years. This study was 
stopped early when it appeared there would be no way to 
show any difference between treatments.
	Indeed, in the CABG-Patch study there was almost no
difference in outcome, whether or not a patient received an
ICD. This study does show that ICD therapy is not
beneficial in all patients with low EF. The results suggest
that we need to identify which patients benefit from ICD
therapy.

Examples

	One scenario was the patient with dilated
cardiomyopathy who has unsustained VT. There is no evidence
about whether such a patient should receive an ICD, but he
would probably get one if treated by an American doctor.
	Another questionable group is patients with Class 4
CHF who are awaiting transplant. American doctors may feel
they are under pressure from the transplant services in 
their hospitals to implant an ICD in these patients, even
though there is no evidence from trials for this.

Guidelines

	There are now guidelines for use of device therapy.
These include grades of evidence and levels of 
recommendation. Currently the strongest recommendations
are based on multiple clinical trials. Weaker 
recommendations are based on the results of a single 
clinical trial. Weakest of all recommendations are based
on the opinion of experts. The panel agreed that 
guidelines have limited use because most patients do not
fit neatly into the categories.

Conclusion

	There is increasing evidence that the ICD is 
beneficial and some doctors are not willing to wait for
evidence from clinical trials to treat their patients with
an ICD. It was noted that the Sudden Cardiac Death Heart
Failure Trial (SCD-HFT) is addressing a fundamental issue,
namely, should all patients with heart failure and low 
ejection fraction get an ICD?