Prophylactic Implantation of a Defibrillator in Patients with Myocardial Infarction and Reduced Ejec

1
Prophylactic Implantation of a Defibrillator in
Patients with Myocardial Infarction and Reduced
Ejection FractionA Review of MADIT II
Brian McGovern, M. D. Massachusetts General
Hospital with Dr. Stephen Servoss, M.D.
2
Sudden Cardiac Death (SCD)

• 340,000 out-of-hospital or in an ED cardiac
  arrests per year in U.S.
• 50 in absence of acute MI
• 95 out-of-hospital mortality
• Given high mortality, goal is prevention

Bigger, NEJM 2002346931-2
3
Patients at risk for sudden cardiac death
Huikiri et al, NEJM 2001
4
Earlier Clinical Trials and Conclusions

• Secondary Prevention of SCD
• The Antiarrhythmics versus Implantable
  Defibrillators (AVID)
• Primary Prevention of SCD CAD and low EF
• The Multicenter Automatic Defibrillator
  Implantation Trial (MADIT)
• The Multicenter Unsustained Tachycardia Trial
  (MUSTT)
• CABG Patch Trial (No mortality benefit in ICD
  group at 3 years follow-up)

continued
5
Secondary prevention of SCD AVID
Earlier Clinical Trials and Conclusions
AVID
Huikiri et al, NEJM 2001
continued
6
Secondary prevention of SCD AVID(continued)
Earlier Clinical Trials and Conclusions

• 1016 patients randomized to ICD v. AAD therapy
• VF VT with syncope or sustained VT with EF lt
  0.40 and symptoms (near-syncope, CHF or angina)
• AAD mainly amiodarone (82 at 24 months)
• Early discontinuation at 3 years
• ICD associated with survival benefit at 3 years
• mortality ICD 24.6, AAD 35.9
• RRR mortality 31, ARR 11.3
• Low use of beta-blocker in AAD group
• 39.4 in ICD group, 10.1 in AAD

AVID Investigators, NEJM 1997
7
Primary prevention of SCD CAD and low EF
Earlier Clinical Trials and Conclusions
AVID
MADIT, MUSTT
Huikiri et al, NEJM 2001
continued
8
Primary prevention of SCD CAD and low
EF(continued)
Earlier Clinical Trials and Conclusions

• 3-4 million patients in the US have had prior MI
  with LVEF lt 0.30
• 340,000 new cases each year
• Trials of ICD therapy largely encouraging
• MADIT and MUSTT mortality benefit with ICD in
  high-risk subgroup
• CABG Patch no difference in survival with
  defibrillator implantation

Bigger, NEJM 2002346931-2
9
Earlier Clinical Trials and Conclusions
Primary prevention MADIT

• 196 patients randomized to ICD v. medical therapy
• prior MI, LVEF lt 0.35, assymptomatic NSVT
• inducible, nonsuppressible VT on EPS
• Early termination at 5 years
• ICD associated with survival benefit
• Mortality at 5 years ICD 16, conventional 39
• ARR 23, Hazard ratio ICD for mortality 0.46
• Low use of beta-blocker in AAD group
• 27 in ICD group, 5 in AAD

continued
Moss et al, NEJM 1996
10
Earlier Clinical Trials and Conclusions
MADIT Kaplan-Meier curve
Benefit was evident within the first few months
after implantation.
Moss et al, NEJM 1996
11
Earlier Clinical Trials and Conclusions
Primary prevention MUSTT

• 704 patients randomized to EP versus non-EP
  guided therapy
• CAD, LVEF lt 0.40, assymptomatic NSVT
• inducible VT on EPS
• if nonsuppressible, eligible for ICD
• ICD associated with reduced mortality with
  respect to both EPG and non-EPG medical therapy

Buxton et al, NEJM 1999
continued
12
Earlier Clinical Trials and Conclusions
MUSTT all cause mortality
Buxton et al, NEJM 1999
13
Earlier Clinical Trials and Conclusions
CABG Patch Trial

• 900 patients
• LVEF lt 0.36
• Abnormal signal averaged ECG
• No mortality benefit in ICD group at 3 years
  follow-up

Bigger, Whang et al, NEJM 1997
continued
14
Earlier Clinical Trials and Conclusions
CABG Patch no difference in mortality
Bigger, Whang et al, NEJM 1997
15
MADIT II Trial further assessment of ICD therapy
in patients with CAD and reduced EF

• MADIT and MUSTT mortality benefit with ICD in
  high-risk subgroup
• CABG Patch no mortality benefit in the setting
  of total revascularization
• Additional benefit to be gained?
• Screening for assymptomatic NSVT
• Possible benefit in those with negative EP

Bigger, NEJM 2002346931-2
16
MADIT II Trial Goal

• Evaluation of potential survival benefit of
• prophylactic ICD in patients with prior MI and
  LVEF
• lt 0.30 (with or without NSVT, in the absence of
• electrophysiological testing and not recently
• revascularized)

Moss et al, NEJM 2002
17
MADIT II Trial Overview

• Well designed study with excellent follow-up
• Low rate of crossover
• High frequency of adjunctive medications in both
  groups
• Primary endpoint all cause mortality

continued
18
MADIT II Trial Overview(Continued)

• Industry support
• Timing of mortality benefit
• Subpopulation that benefits
• difference in benefit depending on definition of
  MI?
• will the EP study identify those to benefit most?

continued
19
MADIT II Trial Overview(Continued)

• CHF admissions
• myocardial stunning from increased shocks
• RV pacing
• longer life expectancy
• Cost effectiveness

20
MADIT II Trial Study Design

• 100 support by Guidant (Indianapolis, Ind.)
• Enrollment began 7/11/97
• 71 US centers, 5 European centers
• Goal 1,200 patients
• 95 power for 38 reduction in 2 year mortality
  with postulated mortality rate 19 in
  conventional arm
• 32 randomization to the ICD and conventional
  arms, respectively
• Inclusion criteria
• prior MI gt 30 days prior to implantation
• AND LVEF lt 0.30
• (Frequent PVCs on Holter--eliminated given high
  prevalence)

Moss et al, NEJM 2002
21
MADIT II Trial Definition of Prior MI

• Abnormal Q wave on ECG
• Elevated cardiac markers on admission for
  suspected MI
• Fixed defect on thallium scanning
• Localized akinesis on ventriculography with
  obstructive CAD on coronary angiography

Moss et al, NEJM 2002
22
MADIT II Trial LVEF Assessment

• Angiography
• Radionuclide scanning
• Echocardiography

Moss et al, NEJM 2002
23
MADIT II Trial Exclusion Criteria

• Alternate indication for ICD
• NYHA IV CHF
• Coronary revascularization within prior 3 months
• Advanced cerebrovascular disease
• Women of childbearing age not using
  contraceptives
• Advanced, life-threatening noncardiac disease

Moss et al, NEJM 2002
24
MADIT II Trial Endpoints

• Primary endpoint all cause mortality
• Secondary endpoints
• cardiac mortality
• arrhythmic cardiac mortality

Moss et al, NEJM 2002
25
MADIT II Trial Statistical Analysis

• Weekly monitoring undertaken using triangular
  sequential design
• Preset criteria to terminate study if ICD either
  superior or inferior to conventional therapy
• P values were two-tailed

Moss et al, NEJM 2002
26
MADIT II Trial Sequential Monitoring
continued
Moss et al, NEJM 2002
27
MADIT II Trial Sequential Monitoring (continued)

• Measure of the cumulative differences in survival
  between the two treatments.
• Positive value indicates that defibrillator
  treatment was superior to conventional medical
  therapy.
• Variance is closely related to the number of
  deaths.
• Three boundaries are shown
• Inefficacy of the implantation of a defibrillator
  as compared with conventional medical therapy
• No difference between the groups
• The superiority of treatment with a defibrillator

28
Moss et al, NEJM 2002
29
MADIT II Trial Demographic Data
30
MADIT II Trial Cardiac Findings at Enrollment ()
Moss et al, NEJM 2002
31
MADIT II Trial NYHA Functional Class ()
Moss et al, NEJM 2002
32
MADIT II Trial Medications at Last Contact ()
Moss et al, NEJM 2002
33
MADIT II Trial Crossover
Moss et al, NEJM 2002
34
MADIT II Trial Compliance

• 8,749 follow-up visits
• 94 attendance in conventional group
• 97 attendance in ICD group
• 3 patients lost to follow up

Moss et al, NEJM 2002
35
MADIT II Trial Mortality at an Average of 20
Months
19.8
14.2
Hazard ratio 0.69 95 CI 0.51-0/93
Moss et al, NEJM 2002
36
MADIT II Trial Mortality by Cause
Moss et al, NEJM 2002
37
MADIT II Trial Kaplan-Meier Estimates
KaplanMeier Estimates of the Probability of
Survival in the Group Assigned to Receive an ICD
and the Group Assigned to Receive Conventional
Medical Therapy. The difference in survival
between the two groups was significant (nominal
P0.007, by the log-rank test).
Moss et al, NEJM 2002
38
MADIT II Trial Timing of Survival Benefit
RRR
12
28
28
9 months
Moss et al, NEJM 2002
39
MADIT II Trial Early Subgroup Analysis
Hazard Ratios and 95 Percent Confidence Intervals
for Death from Any Cause in the Defibrillator
Group as Compared with the Group Assigned to
Receive Conventional Medical Therapy, According
to Selected Clinical Characteristics
Moss et al, NEJM 2002
continued
40
MADIT II Trial Early Subgroup Analysis
(Continued)

• The hazard ratios in the various subgroups were
  similar, with no statistically significant
  interactions.
• The dotted vertical line represents the results
  for the entire study (nominal hazard ratio, 0.66,
  without adjustment for the stopping rule).
• The horizontal lines indicate nominal 95 percent
  confidence intervals.
• LVEF denotes left ventricular ejection fraction,
  and NYHA .

Moss et al, NEJM 2002
41
MADIT II Trial Adverse Effectsin ICD Group

• No deaths associated with ICD implantation
• 13 lead problems (1.8)
• 5 nonfatal infections (0.7)

Moss et al, NEJM 2002
42
MADIT II Trial Incidence of New or Worsened CHF
p 0.09
19.9
14.9
of pts adm for CHF
Moss et al, NEJM 2002
43
MADIT II Trial Timing of Survival Benefit
MADIT
MADIT II
Moss et al, NEJM 1996
MUSTT
Moss et al, NEJM 2002
Buxton et al, NEJM 1999
continued
44
MADIT II Trial Timing of Survival Benefit
(continued)
In contrast with the earlier MADIT Trial, in
which the survival rate improved within the first
few months after the implantation of the device,
survival benefit in MADIT II began approximately
9 months after ICD implantation. The difference
may reflect
MADIT II

• Somewhat lower mortality rate
• in the conventional therapy
• group
• Absence of risk stratification
• for arrhythmia as an entry
• criterion
• Lower cutoff value for the
• ejection fraction
• More vigorous medical
• treatment

Moss et al, NEJM 2002
45

• The financial implications are profoundyoure
  talking about hundreds of millions, if not
  billions of dollars.
• Dr. Douglas Zipes, ACC President Heartwire,
  3/19/02

46
MADIT II Trial Cost of ICD Therapy

• Hardware
• 65,000 ICDs implanted in US in 2000
• 18,000-25,000 / system
• total cost 1.2-1.6 billion / year
• Rehospitalization costs
• ablation, ICD explantation, generator
  replacement, lead revision, CHF
• 1,008 patients in AVID registry

Larsen et al, Circulation 2002
47
Cost Effectiveness Ratio
Total cost ICD - total cost conventional therapy
average total survival benefit
additional cost of ICD therapy year of life
saved

Larsen et al, Circulation 2002
48
AVID Cost Effectiveness

• Total average cost over 3-year follow-up
• 85,522 ICD
• 71,421 medical therapy
• Average survival benefit 0.21 years
• C/E ratio 66,677 per year of life saved

Larsen et al, Circulation 2002
49
Cost effectiveness of ICD versus drug therapy
Larsen et al, Circulation 2002
50
MADIT Trial Cost Effectiveness Ratio

• Total average costs over 4 year period
• ICD 97,560
• non-ICD 75,980
• Average survival ICD 3.66 y, non-ICD 2.80 y
• C/E ratio 27,000

Mushlin et al, Circulation 1998
51
MADIT II Conclusions

• In patients with reduced EF and history of MI,
  ICD implantation confers a mortality benefit.
• There is a concerning trend toward an increase in
  CHF admissions in the ICD group.
• Cost considerations are substantial, but will
  require an assessment of full survival benefit.
• Subgroup analysis, in particular assessment of
  EPS in risk stratification, will help transition
  these results into clinical practice.

52
Looking Ahead

• Retrospective look at EPS for risk stratification
• Further assessment of CHF risk
• Results of sudden cardiac death in heart failure
  trial (SCD-HeFT)
• Future cost-effectiveness analysis