1. New solutions in heart failure:
Drugs and Devices
Martin R Cowie
Professor of Cardiology
Imperial College London (Royal Brompton Hospital)
m.cowie@imperial.ac.uk

2. NICE 2010
Low Ejection Fraction
HF with normal
Ejection Fraction

3. Ever increasing evidence base....
EMPHASIS
SHIFT
RAFT

4. Ever increasing evidence base....
EMPHASIS
SHIFT
RAFT

5. Neurohormonal activation

6. Neurohormonal antagonism
Post-MI HF Mild Moderate Severe
LV dysfunction CHF CHF CHF
AIRE/SAVE/TRACE SOLVD Treatment CONSENSUS
ACE-I (ramipril/captopril/trando) (enalapril) (enalapril)
Beta CAPRICORN US Carvedilol/MERIT/CIBIS COPERNICUS
Blocker (carvedilol) (carvedilol/metoprolol/biso) (carvedilol)
RALES
Aldosterone EPHESUS EMPHASIS-HF
(spironolactone)
Blocker (eplerenone) (eplerenone)
Moderate-Severe HF
NYHA II mild CHF NYHA III/IV
OPTIMAAL ELITE (Losartan)
ARB (Losartan) ValHEFT/CHARM
VALIANT (Valsartan/Candesartan)
(Valsartan)
ACE-I: Angiotensin Converting Enzyme-Inhibitor
ARB: Angiotensin II Receptor Blocker

7. Aldosterone antagonist therapy for heart failure
due to LVSD
Probability
of Survival
• The Randomised Aldactone Evaluation
1.00 Study (RALES)
0.95
0.90
• 1663 patients with NYHA III or IV heart
RRR=0.30 (0.18-0.40) failure and ejection fraction ≤35% who
0.85
RRR = 30% were already treated with ACE inhibitor,
0.80 P<0.001 diuretic digoxin
0.75
Spironolactone
0.70 • Spironolactone 25mg od vs placebo,
0.65 with patients followed for an average of
0.60
2 years
Placebo
0.55
P < 0.001 • 30% reduction in the risk of death
0.50 (p<0.001) and 35% reduction in risk of
0.45 hospitalisation (p<0.001) among
0.00 patients randomised to spironolactone
0 3 6 9 12 15 18 21 24 27 30 33 36
Months
Pitt et al, N Engl J Med, 1999

8. Aldosterone antagonist therapy for
heart failure after MI
EPHESUS trial
• 3313 patients were randomised to eplerenone
Cumulative
Incidence (%) 25 mg/day and 3319 to placebo (in addition to
‘standard’ medical therapy).
40
35 p=0.008 • Mean follow-up of 16-months. Among those
30 RRR=0.15 taking eplerenone there was:
25 Placebo – 15% relative risk reduction in all-cause death
20 (p=0.008)
15 Eplerenone – 13% relative risk reduction in cardiovascular
10 death or hospitalisation (p=0.002)
5 – 21% relative risk reduction in sudden cardiac
0 death ( p=0.03)
0 3 6 9 121518212427303336
Months since Randomization • Compared with spironolactone, eplerenone was
No. at Risk
Placebo 3313 3064
3319 3125
2983 2830 2418 1801 1213 709 323 99 2 0 0 less likely to cause gynaecomastia or breast
Eplerenone 3044 2896 2463 1857 1260 728 336 110 0 0 0
tenderness, but K+ monitoring was still
essential.
Pitt et al, N Engl J Med, 2003

9. NEJM 2011; 364: 11-21

10. NEJM 2011; 364: 11-21

12. Concomitant medication
NEJM 2011; 364: 11-21

13. 37% RRR
NEJM 2011; 364: 11-21

14. 24% RRR

15. 23% RRR 42% RRR

16. EMPHASIS-HF Study
SAFETY ADVERSE EVENTS
Patients with an adverse event (AE)*
Outcome Eplerenone Placebo (N=1373) P Value
(N=1360)
All 979 (72) 1007 (73.6) 0.37
Hyperkalemia – n (%) 109 (8) 50 (3.7) <0.001
Hypokalemia – n (%) 16 (1.2) 30 (2.2) 0.05
Renal failure – n (%) 38 (2.8) 41 (3.0) 0.82
Hypotension – n (%) 46 (3.4) 37 (2.7) 0.32
Gynecomastia and other 10 (0.7) 14 (1.0) 0.54
breast disorders – n (%)
*Investigator reported adverse events

17. EMPHASIS-HF Study
SAFETY: DRUG DISCONTINUATIONS DUE TO AE
Patients with an adverse event* leading to
drug withdrawal — no. (%)
Outcome Eplerenone Placebo P Value
(N=1360) (N=1373)
All 188 (13.8) 222 (16.2) 0.09
Hyperkalemia – n (%) 15 (1.1) 12 (0.9) 0.57
Hypokalemia – n (%) 0 3 (0.2) 0.25
Renal failure – n (%) 4 (0.3) 6 (0.4) 0.75
Hypotension – n (%) 0 3 (0.2) 0.25
Gynecomastia and other 2 (0.1) 2 (0.1) 1.00
breast disorders – n (%)
*Investigator reported adverse events

18. EMPHASIS-HF Study
SAFETY: PRESPECIFIED ADJUDICATED EVENTS
Eplerenone Placebo Hazard Ratio P
Outcome
(N=1364) (N=1373) (95% CI) Value
Hospitalization for
worsening renal 9 (0.7) 8 (0.6) 0.97 (0.37, 2.58) 0.95
failure*
Hospitalization for
4 (0.3) 3 (0.2) 1.15 (0.25, 5.31) 0.85
hyperkalemia*
EMPHASIS HF study results presentation. Presented at AHA congress 2010.
http://click.heartemail.org/?qs=c809010216325f9c50c94e221d4e3fd62e92e966356857c348c68a0675e1e1a3. Accessed November 21, 2010

19. Important addition to therapy....
For mild HF with low EF
NNT
• To prevent one patient
experiencing the primary
endpoint, per year of
follow up, is 19
• To postpone one death, per
year of follow up, is 51
NB Eplerenone not yet licensed for treatment of EMPHASIS population

20. Ever increasing evidence base....
EMPHASIS
SHIFT
RAFT

21. Heart rate strongly associated with mortality
The CIBIS-2 study (n=2539)
One-year mortality (%)
18
16 Bisoprolol
14 Placebo
12
10
8
6
4
2
0
Baseline HR Baseline HR Baseline HR
72 bpm 72-84 bpm >84 bpm
Lechat P, et al. Circulation. 2001;103:1428-1433.

22. Reduction of heart rate and outcomes
in CHF trials
Changes in mortality (%)
60
PROFILE
40 XAMOTEROL
20 PROMISE
VHeFT
(prazosin)
0
CIBIS VHeFT
-20 BHAT SOLVD (HDZ/ISDN)
ANZ
-40 NOR
TIMOLOL GESICA CONSENSUS
-60 MOCHA
US
-80 CARVEDILOL
-100
-18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10
Changes in heart rate (bpm)
Kjekshus J, et al. Eur Heart J. 1999;1(suppl.H):H64-H69.

23. Ivabradine: ‘pure’ heart rate reduction
open
closed closed
RR
Pure
0 mV heart rate
reduction
-40 mV
-70 mV
Ivabradine
If inhibition reduces the diastolic depolarization slope, and thereby
lowers heart rate
Thollon C, et al. Brit J Pharmacol. 1994;112:37-42.

24. Systolic Heart failure treatment with
the If inhibitor ivabradine Trial

25. Primary objective
To evaluate whether the If inhibitor ivabradine
improves cardiovascular outcomes in patients with:
1. Moderate to severe chronic heart failure
2. Left ventricular ejection fraction  35%
3. Heart rate  70 bpm in sinus rhythm
4. Best recommended therapy
Ivabradine 5mg bd or placebo, titrated to
7.5mg/5mg/2.5mg according to tolerability

26. Study end points
Primary composite end point
 Cardiovascular death
 Hospitalization for worsening heart failure
Other end points
 All-cause / CV / HF death
 All-cause / CV / hospitalization for heart failure
 Composite of CV death, hospitalization for HF or nonfatal MI
 NYHA class / Patient & Physician Global Assessment
Median study duration 22.9 months, maximum 41.7 months
Swedberg K, et al. Eur J Heart Fail. 2010;12:75-81.

27. Baseline characteristics
Ivabradine Placebo
3241 3264
Mean age (y) 60.7 60.1
Male (%) 76 77
Ischemic etiology (%) 68 67
NYHA II (%) 49 49
NYHA III/IV (%) 51 51
Previous MI (%) 56 56
Diabetes (%) 30 31
Hypertension (%) 67 66
Swedberg K, et al. Lancet. 2010;376:875-885.

28. Baseline characteristics
Ivabradine Placebo
3241 3264
Mean heart rate (bpm) 80 80
Mean LVEF (%) 29 29
Mean SBP (mm Hg) 122 121
Mean DBP (mm Hg) 76 76
eGFR (mL/min/1.73 m2) 75 75
Swedberg K, et al. Lancet. 2010;376:875-885.

29. Chronic heart failure
background treatment
Patients (%)
100
91 91 Ivabradine
89 90
84 83 Placebo
80
61 59
60
40
22 22
20
3 4
0
Beta-blockers ACEIs and/or ARBs Diuretics Aldosterone Digitalis ICD/CRT
antagonists
Swedberg K, et al. Lancet. 2010;376:875-885.

30. Background beta-blocker
treatment
Patients (%)
100 Ivabradine
89 89
Placebo
80
60 56 56
40
26 26
20
0
Beta-blockers at At least 50% target daily Target daily dose
randomization dose
Swedberg K, et al. Lancet. 2010;376:875-885.

31. Heart rate is a predictor of CV death and/or
hospitalizations for HF
Patients with primary composite end point in the placebo group (%)
50 ≥87 bpm
P<0.001
40
80 to <87 bpm
75 to <80 bpm
30
72 to <75 bpm
70 to <72 bpm
20
10
0 Months
0 6 12 18 24 30
Risk increases by 3% per 1 bpm increase, and by 16% per 5 bpm increase
Böhm M, et al. Lancet. 2010;376:886-894.

32. Mean heart rate reduction
Heart rate (bpm)
90
80
80
75 Placebo
75
70 Ivabradine
67
64
60
50
Months
0 2 weeks 1 4 8 12 16 20 24 28 32
Swedberg K, et al. Lancet. 2010;376:875-885.

33. Primary composite end point
(CV death or hospital admission for worsening HF)
Cumulative frequency (%)
40
HR = 0.82 (0.75–0.90)
P < 0.0001 Placebo
30 18% RRR
Ivabradine
20
10
0
Months
0 6 12 18 24 30
Swedberg K, et al. Lancet. 2010;376:875-885.

34. Hospitalization
for worsening heart failure
Cumulative frequency (%)
30
HR = 0.74 (0.66–0.83)
P < 0.0001 Placebo
26% RRR
20
Ivabradine
10
0
0 6 12 18 24 30 Months
Swedberg K, et al. Lancet. 2010;376:875-885.

35. Cardiovascular death
Cumulative frequency (%)
30
HR = 0.91 (0.80–1.03)
P = 0.128
9% RRR (P=0.12) Placebo
20
Ivabradine
10
0
0 6 12 18 24 30 Months
Swedberg K, et al. Lancet. 2010;376:875-885.

36. Death from heart failure
Cumulative frequency (%)
10
HR = 0.74 (0.58–0.94)
P = 0.014
26% RRR Placebo
5
Ivabradine
0
0 6 12 18 24 30 Months
Swedberg K, et al. Lancet. 2010;376:875-885.

37. Effect of ivabradine in
prespecified subgroups
Test for interaction
Age
<65 years
≥65 years
Sex
Male
Female
Beta-blockers
No
Yes
Etiology of heart failure
Nonischemic
Ischemic
NYHA class
NYHA class II
NYHA class III or IV
Diabetes
No
Yes
Hypertension
No
Yes
Baseline heart rate
<77 bpm P = 0.029
≥77 bpm
0.5 1.0 1.5
Hazard ratio
Favors ivabradine Favors placebo
Swedberg K, et al. Lancet. 2010;376:875-885.

38. Incidence of selected
adverse events
Patients with an event
n= 6492
Ivabradine Placebo
P value
n=3232, n (%) n=3260, n (%)
All serious adverse events 1450 (45%) 1553 (48%) 0.025
All adverse events 2439 (75%) 2423 (74%) 0.303
Symptomatic bradycardia 150 (5%) 32 (1%) <0.0001
Asymptomatic bradycardia 184 (6%) 48 (1%) <0.0001
Atrial fibrillation 306 (9%) 251 (8%) 0.012
Phosphenes 89 (3%) 17 (1%) <0.0001
Blurred vision 17 (1%) 7 (<1%) 0.042
Swedberg K, et al. Lancet. 2010;376:875-885.

39. Treatment discontinuation
Patients with an adverse event,
leading to withdrawal
Ivabradine Placebo
P value
n=3232, n (%) n=3260, n (%)
All adverse events 467 (14%) 416 (13%) 0.051
Symptomatic bradycardia 20 (1%) 5 (<1%) 0.002
Asymptomatic bradycardia 28 (1%) 5 (<1%) <0.0001
Atrial fibrillation 135 (4%) 113 (3%) 0.137
Phosphenes 7 (<1%) 3 (<1%) 0.224
Blurred vision 1 (<1%) 1 (<1%) 1.000
Swedberg K, et al. Lancet. 2010;376:875-885.

40. Important addition to therapy....
For those in sinus rhythm with HR > 70bpm and low EF
NNT
• To prevent one patient
experiencing the primary
endpoint, per year of follow
up, is 26
• To postpone one
hospitalisation for HF, per year
of follow up, is 27
NB Ivabradine not yet licensed for treatment of SHIFT population

41. Ever increasing evidence base....
EMPHASIS
SHIFT
RAFT

42. New device implant rate 1999-2009
ICD
CRT
Ten year average growth rate 15.1%

43. High energy device implant rates across Europe
Source: Eucomed 2009

44. NEJM 2010; 363: 2385-95

45. RAFT design

48. 25% RRR

49. 25% RRR

50. Subgroups with more benefit?
QRS ≥ 150 msec
LVEF < 20%
LBBB

51. Conclusions

52. So – how should this affect practice for those
with low EF HF?
• Life-saving therapy should include:
– ACEI (or ARB),
– Β-blocker, and
– aldosterone antagonist (eplerenone) or good reason for not!
• Once β-blockade maximised, if in sinus rhythm and HR >
70bpm, ivabradine should be added
• For patients with severe LV systolic dysfunction, mild-
moderate HF, LBBB and optimal drug therapy, increasingly
likely that CRT-D (rather than CRT-P or ICD alone) will be
recommended
• NICE will issue new guidance in 2012/13 for device therapy.
• Good monitoring always required!!
These statements are a personal opinion - NOT official recommendations!