Congestive Heart Failure Collaborative, October 14, 2004
Preventing Readmissions
Kenneth A. LaBresh, MD, FAHA, FACC
V.P., Medical Affairs and Quality Improvement, MassPRO
Clinical Associate Professor, Brown University
ACC/AHA Guidelines for Evaluation and Management of Chronic Heart Failure 2001
HF can be prevented
HF has established risk factors
HF is a progressive condition with asymptomatic and symptomatic stages
HF morbidity and mortality can be reduced by stage specific treatments
Hunt, et al. J Am Coll Cardiol. 2001; 38:2101-2113.
ACC/AHA Proposed Stages of HF
STAGE A High risk for developing HF
STAGE B Asymptomatic LV dysfunction
STAGE C Past or current symptoms of HF
STAGE D End-stage HF
Hunt, et al. J Am Coll Cardiol. 2001; 38:2101-2113.
Neurohormonal Activation in Heart Failure
Hypertrophy, apoptosis, ischemia,
arrhythmias, remodeling, fibrosis
Angiotensin II
Norepinephrine
Morbidity and Mortality
? CNS sympathetic outflow
? Cardiac sympathetic activity
? Renal sympathetic activity
Sodium retention
Myocyte hypertrophy
Myocyte injury
Increased arrhythmias
Disease progression
?1
b1
b1
b2
?1
? Vascular sympathetic activity
Vasoconstriction
?1
Activation
of RAS
Adrenergic Pathway in Heart Failure Progression
Beta-blocker Therapy in Heart Failure
Potential Beneficial Effects
Protection from
Catecholamine
Toxicity
Renin Angiotensin System
Reversal of Remodeling
Up-regulation
of b-adrenergic
Receptors
Ancillary
Factors
Major Placebo Controlled Trials of ?-Blockade in Heart Failure
34% ?
Cumulative Mortality (%)
Days
20
15
5
0
10
P=.0062 (adjusted)
Metoprolol CR/XL
(n=1990)
Placebo (n=2001)
US Carvedilol Trials1
Probability of Event-free Survival?
Carvedilol (n=696)
Placebo (n=398)
Days
P<.001
0.0
0
100
200
300
400
65% ?
1.0
0.8
0.7
0.9
MERIT-HF2
Survival (% of Patients)
100
90
80
60
70
0
600
0
400
300
200
100
Days
Carvedilol (n=1156)
Placebo (n=1133)
500
600
0
400
300
200
100
500
35% ?
P=.00013
COPERNICUS4
Days
0.0
200
400
800
1.0
0.8
0.6
P<.0001
34% ?
Bisoprolol (n=1327)
Placebo (n=1320)
CIBIS-II3
0
600
Survival
1Packer M et al. N Engl J Med. 1996;334:1349–1355. 2MERIT-HF Study Group. Lancet. 1999;253:2001–2007. 3CIBIS-II Investigators. Lancet. 1999;353:9–13. 4Packer M et al. N Engl J Med. 2001;344:1651–1658.
COPERNICUS: All-Cause Mortality
% Survival
Packer M et al. N Engl J Med. 2001;344:1651–1658.
900
600
300
0
P=.0012
P=.0002
P<.0001
For any
reason
For cardiovascular
reason
For heart
failure
Placebo
Carvedilol
29%
33%
600
400
200
0
450
300
150
8
20%
COPERNICUS: Number of Hospitalizations
Packer M et al. Circulation. 2002;106:2194–2199.
0
5
10
15
20
25
30
All Patients
(n=2289)
Higher-Risk Patients
(n=624)
Number of Events
Krum H et al. JAMA. 2003;289:712–718.
0
60
180
All Patients
(n=2289)
Higher-Risk Patients
(n=624)
Number of Events
8 Weeks
8 Weeks
Deaths
Death or Hospitalization for Any Reason
Placebo
Carvedilol
COPERNICUS: Early Clinical Outcomes
120
25
19
15
3
153
134
63
44
Does Subspecialty Care Affect CHF Outcome
SUPPORT
1298 Hospitalized Patients
743 (57%) Cardiolgist (C)
555 (43%) Primary Care (P)
C P
Age 63 71
Male 71% 52%
EF known 69% 47%
EF < 20% 52% 39%
SUPPORT Results
Cardiologists vs. Primary Care:
RHC 2.9 times more likely
Coronary angio 3.9 times more likely
Hospital costs 43% higher
ACE-I 64% IN BOTH groups
Short term mortality similar
4.6 year follow up mortality 20% lower for cardiologists (rr 0.80 (0.66 - 0.96))
Aurebach ad, AIM 2000;132:191-200
Impact of Subspeciality Care
Upstate New York 10 hospitals
Three patient groups
I Noncardiologist n = 977
II Cardiologist Attending n = 419
III Cardiology Consult n = 1058
Group I more NH, more comorbidity
more females, less B-Blocker use
Upstate New York Results
Results (cont.)
Philbin E, CHEST 116:2, 346 - 354
Hospital Based CHF Clinic
Retrospective analysis before (n = 407) and after (n = 357) implementation of a CHF program in 1994
Elements of the Program
Multidisiplinary team
Inpatient and outpatient treatment protocols
Patient and family education
Follow up telephone calls
Outpatient infusion center
Outcomes
HF Management Meta Analysis
McAlister FA, et al., JACC 2004;44:810-819
29 studies involving 5,039 patients
Conclusions
HF management programs decrease HF hospitalizations in a wide variety of formats
Multidisciplinary approaches, particularly HF clinics have been demonstrated to reduce mortality as well
15 0f 18 studies that evaluated costs demonstrate cost savings as well
SPAN-CHF
Specialized Primary and Networked Care in Heart Failure
Kimmelsteil et al., Circ 2004;110:1450-1455
Characteristics of the System
1. Focus on patients who have a resent HF hospitalization
2. Administered by nurses as case managers with strong expertise in HF with the support of HF physicians acting as consultants to the nurses.
Telephone monitoring and limited home visits
Communication between the nurse and the PCP
5. Provided a three-month “active intervention” followed by surveillance out to 1 year
Hypothesis
A uniform disease management program will:
Reduce the cost of medical care
Increase the time to hospitalization or death
Decrease the frequency of admission for HF.
Increase the total number of days alive out of hospital
Improve indices of health-related quality of life
Characteristics of the Study Population
Inclusion
Patients d/c with a primary diagnosis of heart failure
Etiology of HF may include:
IHD, Dilated CMP, corrected valve disease or regurgitant valve disease deemed uncorrectable