Ethnic influences on stroke risk

1
Ethnic influences on stroke risk

• Francesco P Cappuccio MD MSc FRCP FFPH

2
Selected leading causes of death worldwide in 1990
3M (70) in developing countries
Number of deaths (million)
Lancet 19973491269-76
3
Prevalence of severe disability in men
(per 1,000)
Age groups
Lancet 19973491347-52
4
Mortality due to leading global risk factors
Ezzati M et al. Lancet 20023601347-60
5
Stroke is preventable!

• Time trends
• rapid change in stroke mortality is likely to
  have resulted from change in incidence rates, so
  factors determining onset of disease must have
  changed

6
Trends in age-adjusted mortality from stroke in
the US
7
Stroke is preventable!

• Time trends
• rapid change in stroke mortality is likely to
  have resulted from change in incidence rates, so
  factors determining onset of disease must have
  changed
• Geographic variations
• large international differences in stroke
  mortality
• they are not fixed, e.g. in Japanese migrants

8
Prevalence of hypertension in populations of
African origin (ICSHIB)
Prevalence of hypertension in populations of
European vs African origin
Age and sex-adjusted prevalence ()
Age and sex-adjusted prevalence ()
R Cooper et al Am J Pub Health 1997 87 160-8
R Cooper et al. BMC Medicine 2005 3 2
9
Mortality from Stroke amongst Japanese migrants
Deaths per 100,000 population
Am J Epidemiol 1990131579-88
10
Stroke is preventable!

• Time trends
• rapid change in stroke mortality is likely to
  have resulted from change in incidence rates, so
  factors determining onset of disease must have
  changed
• Geographic variations
• large international differences in stroke
  mortality
• they are not fixed, e.g. in Japanese migrants
• Causes of stroke
• many can be avoided
• effect reversible in a few years, e.g. RCTs

11
Results from a meta-analysis of 10 trials of
anti-hypertensive drug therapy
Total of 9278 active treatment and 9264 control
patients
12
Risk Factors for Stroke

• Inherent biological traits
• age, sex, ethnic background
• Physiological characteristics
• blood pressure, fibrinogen, BMI, homocysteine,
  etc.
• Pathological factors
• atrial fibrillation, diabetes, sickle cell
  disease
• Behaviour
• smoking, diet, alcohol, OC
• Social characteristics
• social class
• Environmental Features
• temperature, season, etc.
• Genetics
• candidate genes (AGT, Na-channel, G-protein,
  adducin, ...)

13
Blood Pressure, Stroke and CHD
14
Variations by ethnic groups in the UK

• Burden of vascular disease
• Detection, management and control of hypertension
• Application of national guidelines
• Assessment of risk
• Non-drug therapy
• Pharmacological treatment

15
IHD and CVD Mortality in England Wales (1983)
in people aged 20-69 yrs
Balarajan R. BMJ 1991302560-4
16
Cause of death from vascular disease in US
blacks and whites(1991 NYC Medical Examiners
Office)
MEN WOMEN (n417)
(n170) ______________________________
Age at death (years) 51.7 vs 51.2 54.7 vs 61.5
Atherosclerotic 0.4 (0.2 - 0.5) 0.4 (0.2 -
0.8) Hypertensive 2.2 (1.4 - 3.4) 3.1
(1.5 - 6.5) Age-adjusted OR (95
CI) plt0.01, plt0.001
Hypertension 1998311070-6
17
Incidence of first ever stroke in London (1995-6)
12.2 yrs younger!
Stewart JA et al. Br Med J 1999318967-71
18
Types of stroke
19
Incidence of first ever stroke subtype in London
(1995-6)
age-adjusted
Stewart JA et al. Br Med J 1999318967-71
20
Prevalence of hypertension by age and ethnic
group in South London
BP gt160 and/or gt95 mmHg or on therapy
Heart 199778555-63
21
Detection, Management and Control of Hypertension
in S. London (1994-6)
22
Stroke Mortality and Quality of Hypertension
Control
Du et al. BMJ 1997
23
Risk Assessment and Treatment Choices

• Prevention and management of CVD based on overall
  absolute risk of disease, rather than individual
  risk factor management
• Recent guidelines now adopt this paradigm shift
  (JBS-2, BHS IV, NSF for CHD, NICE)
• Risk estimates based on 10-year prospective
  experience of Framingham cohort (sub-urban
  American white middle-class men and women)
• The burden of cardiovascular disease is not
  distributed equally among society.
• Ethnic groups have disproportionately high
  riskbut

24
Relationship between CHD and CVD risks by ethnic
origin
1.40 (1.35 - 1.45)
1.30 (1.27 - 1.34)
1.48 (1.40 - 1.55)
Cappuccio FP et al. Br Med J 20023251271-4
25
British evidence

• None prospective
• CHD and CVD risk by Framingham score not
  consistently related in ethnic groups Cappuccio
  FP e al. BMJ 20023251271-4
• Discrepancies between predicted risks by ethnic
  group and SMR by country of birth Quirke TP et
  al. Heart 200389785-6
• Inconsistent predictions between Framingham,
  FINRISK and SCORE in ethnic groups Bhopal R et
  al. J Pub Health 20052793-100
• Age-adjustment to reduce inconsistency of
  Framingham risk across ethnic groups Aarabi M et
  al. Eur J Cardiovasc Prev Rehab 20051246-51

26
ETHRISK A modified Framingham CHD and CVD risk
calculator for British black and minority ethnic
groups
www.epi.bris.ac.uk/CVDethrisk/
Heart 2006921595-1602
27
In people of African origin
A slave trader licking a slaves face to assess
his fitness for the voyage across the Atlantic

• high blood pressure more common
• low plasma renin activity
• sensitive to changes in sodium intake
• sodium retention and volume expansion

28
BHS - NICE Guidelines
29
ALLHAT blood pressure control blacks vs
non-blacks
below 140/90 mmHg
Wright JT et al. JAMA 20052931595-1608
30
ALLHAT outcomes in blacks vs non-blacks
Wright JT et al. JAMA 20052931595-1608
31
Stroke ethnic minority groups

• Compared to whites, Africans Caribbeans
• have a lower incidence of CHD
• have a higher risk of stroke, renal failure
  and LVH
• BP more sensitive to salt restriction
  benefit more from dietary advice alone
• low plasma renin and angiotensin ? reduced
  response to ACE-i ARBs (or ?-blockers) as
  monotherapy however, see ALLHAT
• Improved efficacy to ACE-inhibitors/ ARBs in
  combination with diuretics or CCBs
• Compared to whites, South Asians
• have a greater incidence of CHD
• have also a higher risk of stroke and renal
  failure
• some sub-groups have high blood pressure and
  some have very high smoking rates
• metabolic abnormalities more common