Biomedical Engineering Reference
In-Depth Information
Table 6.3 Characteristics of randomized studies assessing ICD efficacy as part of secondary prevention [39]
Reduction in mortality
in the ICD group (%)
Study
Years
Patients (n)
Main inclusion criteria
31
AVID
1993-1997
1,013
Resuscitation for VF, VT with syncope, LVEF
£
40 %
30
CIDS
1990-1998
659
Resuscitation for VF, VT with syncope,
VT with a cycle length £ 400 ms, LVEF £ 35 %
CASH
1987-1998
288
Resuscitation for VF
39
important to have an approximately equal number of sub-
jects in the treatment and control groups at each center.
compared with amiodarone was found in patients with an
LVEF > 34 %. By contrast, patients with an LVEF of 20-34 %
had a significantly reduced mortality with the use of ICD in
years 1 and 2. A similar trend was shown in the group with
an LVEF < 20 %, but the results were not statistically
significant because of the small number of patients.
Prospective/retrospective study
A retrospective study investigates the effect of a phenom-
enon that was used in treatment in the past. Thus, it is
rather an analysis and interpretation of data obtained pre-
viously. In a prospective study, the protocol, methods, and
study subject characteristics are designed first; only then
are subject enrollment and the actual study initiated.
Canadian Implantable De fi brillator Study (CIDS)
The CIDS study was conducted in 659 patient-survivors of
ventricular fibrillation who had documented VT with syncope,
sustained VT with presyncope or stenocardia, and an
LVEF < 35 %. They were randomized into groups with phar-
macological treatment or ICD implantation. The study findings
confirmed a 30 % reduction in mortality in the ICD group. A
subgroup analysis identified patients who benefited most from
ICD treatment (a 50 % relative risk reduction in all-cause
mortality): age > 70, LVEF < 35 %, NYHA classes III and IV.
These patients exhibited a 1-year mortality rate of 30 % when
treated with amiodarone, but only 14 % with ICD treatment.
Single-blind/double-blind study
When performing controlled studies, the result can be
biased by patient or investigator expectations. Thus, the
use of placebo makes sense only when the patient is
unaware whether he or she is taking an active substance or
placebo; such studies are referred to as single-blinded. If
the investigator is also unaware whether the patient is
treated with placebo or the active substance, the study is
referred to as double-blinded. Studies in which even the
team processing the data is unaware of which group is
receiving which treatment are referred to as triple-blinded.
Single-center/multicenter study
When a large number of patients needs to be recruited,
studies are conducted in multiple centers or worldwide.
Results are not biased by regional differences in the treat-
ment of enrolled subjects.
Cardiac Arrest Study Hamburg (CASH)
The CASH study was performed in 288 patients after cardiac
arrest secondary to VT who were randomized into groups
with pharmacological treatment or ICD implantation. At the
2-year follow-up, mortality from sudden death was 2 % in
the group treated with ICD implantation (99 patients) com-
pared with 11 % in the group of 189 patients treated with
amiodarone or metoprolol. All-cause mortality was 12.1 %
in the ICD group versus 19.7 % in the group treated with
amiodarone or metoprolol, which means a relative reduction
in all-cause mortality of 39 %.
6.2.1.1 Secondary Prevention Studies
Clinical studies analyzing secondary prevention of SCD
(Table 6.3 ) show a consistently higher efficacy of ICD treat-
ment compared with antiarrhythmic drugs [ 39 ] .
Antiarrhythmics Versus Implantable De fi brillator (AVID)
The AVID study was conducted in 1,013 patients with near-
fatal ventricular fibrillation or sustained hemodynamically
unstable VT and an LVEF < 35 %. They were randomized
into groups with pharmacological treatment with amiodarone
or ICD implantation. In the pharmacological treatment
group, 97.4 % of patients received amiodarone and 2.6 %
received sotalol. The study was stopped early when a
significant reduction in mortality - 38 % at 1-year follow-up
and 31 % at 3 years - was shown in patients with an ICD
compared with the group of patients taking an antiarrhyth-
mic drug. A subgroup analysis revealed that patients with an
LVEF < 34 % benefited most from ICD. No benefit from ICD
6.2.1.2 Primary Prevention Studies
Multicenter Automatic De fi brillator Implantation
Trial (MADIT)
The MADIT was conducted from 1990 to 1996 and included
196 patients with ischemic cardiomyopathy, an LVEF < 35 %,
unsustained VT, and sustained VT resistant to procainamide
infusion during electrophysiologic testing. They were ran-
domized into groups with pharmacological treatment with
antiarrhythmic drugs or ICD implantation. The study was
stopped prematurely. In the ICD group, only 16 % of patients
died versus 39 % in the pharmacological treatment group (a
54 % relative risk reduction in all-cause mortality); of these
 
Search WWH ::




Custom Search