Epidemiology of heart failure

1
Epidemiology of heart failure

• Darrel Francis
• Clinical Senior Lecturer in Cardiology
• Imperial College School of Medicine

2

• Difficulties of case definition
• Aetiology
• Prevalence and Incidence
• Temporal trends
• Global burden

3
Case definition
4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
The greatest challenge in the epidemiology of
heart failure difficulty defining cases

• Symptoms are protean
• Signs are commonly found in subjects without
  heart failure
• Tachycardia, crepitations, leg oedema(!)
• Tests what is normal?
• why is this a challenge?

10
Why are normal ranges so important here?

• After all, we can study blood pressure without
  predeciding normal range

BP is easier because there is only one way to
measure it (or at least experts have developed
conventions)
11
There is more than one test for heart failure!

• No tests, just symptoms signs

12
Criterion
Point value Category I
history Rest dyspnea 4 Orthopnea 4 Paroxysmal
nocturnal dyspnea 3 Dyspnea while walking on
level area 2 Dyspnea while climbing 1 Category
II physical examination Heart rate abnormality
(1 point if 91 to 110 beats per 1 or 2
minute 2 points if more than 110 beats per
minute) Jugular venous elevation (2 points if
greater than 2 or 3 6 cm H2O 3 points
if greater than 6 cm H2O plus hepatomegaly or
edema) Lung crackles (1 point if basilar 2
points if more than 1 or 2 basilar) Wheezing

3 Third heart sound
3 Category III chest
radiography Alveolar pulmonary edema
4 Interstitial pulmonary
edema 3 Bilateral
pleural effusion
3 Cardiothoracic ratio greater than 0.50
3 Upper zone flow redistribution
2 8 to 12 points
definite HF 5 to 7 points possible HF lt4 points
unlikely HF
Boston Criteria for diagnosing Heart Failure
from Marantz et al Circulation 198877607-12.
13
(No Transcript)
14
No tests gives very poor validity

• Less than half of those identified by clinical
  judgement alone are confirmed by subsequent tests
• Even more unreliable for women than for men

15
ESC gives guidelines for definition of HF
Eur Heart J 2005 26 11151140 .
16
There is more than one test for heart failure!

• No tests, just symptoms signs
• Brain Natriuretic Peptide (blood test)
• Radionucleide ventriculography (MUGA) or contrast
  ventriculography
• 2d Echocardiography
• Magnetic Resonance Imaging
• Tissue Doppler Imaging

17
(No Transcript)
18
Impact of difficult case definition?

• Difficult to safely compare absolute rates beyond
  study
• may be due to ascertainment method, not a true
  difference in the population
• Reasonably safe to look for relationships (e.g.
  with age) within any one study

19
Aetiology
20
Aetiology of heart failure in a UK
population Coronary Artery Disease
52 Idiopathic 13 Valve Disease
10 Cardiomyopathy 10 Hypertension 4 Alcohol
4 Atrial Fibrillation 3 (Wood, 2002)
21
A more honest breakdown?
Cowie, Hillingdon heart failure study, Eur Heart
J 1999 20 421428
22
Aetiology does change

• In the 1950s, Hypertension was the commonest
  aetiology
• Garrison GE, McDonough JR, Hames CG, Stulb SC.
  Prevalence of chronic congestive heart failure in
  the population of Evans County, Georgia. Am J
  Epidemiol 196683338-344.
• Since then, primary prevention (antihypertensive
  therapy) has dramatically reduced this proportion
• Kannel WB, Ho K, Thorn T. Changing
  epidemiological features of cardiac failure. Br
  Heart J 199472S3-S9

23
Prevalence
24
(No Transcript)
25
Case study Heart of England Screening study
Invited random sample (n1617) of all men and
women aged over 45 years registered at GP
practices in the West Midlands. All patients
who agreed to participate were assessed in their
own general practice by clinical history
(including prescribed drugs), determination of
New York Heart Association functional
class, clinical examination, resting 12 lead
electrocardiography, and echocardiography
including Doppler studies. Defined heart
failure according to ESC criteria appropriate
symptoms (NYHA II or worse) plus objective
evidence of cardiac dysfunction. EFlt40
"definitely impaired 40-50 "borderline"
(40-50) Did not attempt to diagnose diastolic
dysfunction.
Davis et al, BMJ 20023251156-60.
26
Heart of England Screening study Over 2 of
patients (3 of men and 1.7 of women) screened
had definite heart failure. Probable heart
failure was seen in around a further 1 of
patients. From these prevalence rates they
estimate about 369,000 men aged gt45 in the UK
with definite heart failure, and 300,000 women,
giving a total of around 669,000. If probable
cases of heart failure are included, there are an
estimated 497,500 men and 404,000 women, a total
of 901,500 people aged 45 and over who have heart
failure in the UK today. Prevalence of heart
failure increases steeply with age, so that while
around 1 of men and women aged under 65 have
heart failure, this increases to about 7 of
those aged 75-84 years and 15 of those aged 85
and above.
27
Potentially complex contributory factors
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
Does the 2 prevalence cover all types of
patient?
32
Prevalence of Congestive Heart Failure by Age and
Sex NHANES 1999-2002
Source CDC/NCHS and NHLBI.
33
(No Transcript)
34
Prevalence of heart failure by deprivation, 1998,
England and Wales
Ellis C et al (2001) Health Statistics Quarterly
11 17-24
www.heartstats.org
35
(No Transcript)
36
Temporal trends
37
Prevalence is rising
38
Congestive Heart Failure Episodes by Sex United
States 1970-2002
Source CDC/NCHS.
39
Why is prevalence rising?
Levy, NEJM 2002 3471397-1402
40
Survival
41
Levy, NEJM 2002 3471397-1402
42
Levy, NEJM 2002 3471397-1402
43
Levy, NEJM 2002 3471397-1402
44
Levy, NEJM 2002 3471397-1402
45
(No Transcript)
46
Survival after diagnosis of cancer or heart
disease

100
Men
80
Survival
60
Bladder Ca
40
Prostate Ca
Bowel Ca
20
Lung Ca
0
60
48
36
24
12
0
Months from diagnosis
Stewart S, EJHF 2001 3315-322
47

Survival after diagnosis of cancer or heart
disease in women Where does breast cancer lie?
100
80
60
Survival
MI
40
Ovarian Ca
Bowel Ca
Heart Failure
20
Lung Ca
0
60
48
36
24
12
0
Months from diagnosis
48
Epidemiology ofHeart failure

• Numbers heavily depend on methods
• clinical assessment is unreliable
• Marked increase in prevalence with age
• Risk factors are similar to those of coronary
  artery disease
• Mortality worse than most cancers
• Increasing survival with modern therapy leads to
  increasing prevalence

49
Today a problem of the developed world, but