Hereditary thrombophilia

1
Hereditary thrombophiliacardiovascular disease
(CVD)
Dr. A. Lubetsky Natl. Hemophilia Center Inst. of
Thrombosis Hemostasis Sheba Medical Center
2
Burden of CVD

• 61 million americans have CVD
• 6 million hospitalizations/year
• 950,000 deaths/year
• 40 of all death causes
• Cost 2003 209 billion for direct health costs
• 142 billion for productivity loss

3
Establishment
The Roman Cohort
?
?
?
?
?
?
?
?
?
?
?
?
?
?
4
(No Transcript)
5
The Framingham cohorts - timeline
1971
1948
2004
Omni study
Original cohort 5,209 ??
Offspring cohort 5,124 ??
3rd generation cohort?? currently
recruiting Aim- 3,500
Examine life style and other risk factors for CVD
Examine genetic risk factors for CVD
6
The Framingham cohorts major findings
1948
1967
1961
1960
1976
1978
1988
IHD
STRESS
good cholesterol
STROKE
1970
1994
7
A role of blood coagulation ?
8
Veins and arteries are not similar
Arteries Veins
high low Vessel pressure
high low Flow velocity
low high Capacity
Vessel wall injury is main factor. Interplay of endothelial cell injury, platelet activation and inflammation. Relative circulatory stasis. Activation of coagulation. Natural anticoagulants play a significant role. Mechanism
Less important Important factor Thrombophilia
9
Risk factor management
Anti platelet Rx
Thrombolysis PCI
10
CVD time consuming..
TIME
Are young people more sensitive to the effects
of hypercoagulability ???
Many years
Few minutes
11
First evidence..

• The Northwick Park Heart Study (NPHS)
• 1511 white males aged 40-64
• 109 first major CHD event during FU
• Mean Fibrinogen level 290mg
• For each 1 s.d. (59mg) rise in level 84
  increase in risk for heart events

Meade TW et al., Lancet, 1986
12
Why is Fibrinogen an attractive candidate ?

• A prominent part of the atheromatous plaque
• Major constituent of the occlusive thrombus
• Influenced by smoking
• Marked genetic contribution to blood levels (48
  - 61)

13
Why is it not ?

• An acute phase reactant
• Influenced by inflammation ? (confounder effect)
• Fibrin deficient mice capable of forming
  atheromatous plaques
• Many studied failed to associate ß chain
  polymorphisms to risk of CVD.

14
Role of Fibrinogen ???

• Fibrinogen has no role (disease marker)
• Effect is very small compared to other risk
  factors (need large N)
• Effect apparent only in presence of other risk
  factors

15
Other candidates..

• FVII
• FXIII
• EPCR thrombomodulin
• tPA
• PAI 1
• Platelet membrane glycoproteins (GP IIb/IIIa, GP
  Ib-IX-V, GP Ia-IIa)

16
Classic thrombophilia polymorphisms(FVL-PT-MTHFR)

• Known since 1993
• Very common
• Increase risk of VTE
• Interact with other risk factors

17
Is classical thrombophilia associated with CVD ?

• A meta-analysis
• 56 studies
• 54,547 patients
• Endpoints proven MI, stroke or PVD
• Tested FVL, PT, MTHFR

RJ Kim et al. Am Heart J, 2003
18
Factor V Leiden
OR (CI95) N Studies
1.10 (0.88-1.36) 19,360 20 MI
1.27 (0.86-1.87) 15,239 15 Stroke
0.91 (0.38-2.16) 385 1 PVD
1.21 (0.99-1.49) 25,053 33 All
19
Prothrombin G20210A
OR (CI95) N Studies
1.28 (0.94-1.73) 12,727 14 MI
1.30 (0.91-1.87) 6,675 10 Stroke
0.26 (0.03-2.04) 385 1 PVD
1.32 (1.03-1.69) 16,945 23 All
20
MTHFR
OR (CI95) N Studies
1.05 (0.86-1.27) 7,780 14 MI
1.46 (1.19-1.79) 4,319 11 Stroke
- - - PVD
1.20 (1.02-1.41) 12,104 25 All
21
All polymorphisms
OR (CI95) N Studies
1.21 (0.99-1.49) 25,053 33 FVL
1.32 (1.03-1.69) 16,945 23 PT
1.20 (1.02-1.41) 12,104 25 MTHFR
22
Patient selection age lt 55
OR (CI95) N Studies
1.37 (0.96-1.97) 5,696 14 FVL
1.66 (1.13-2.46) 5,472 23 PT
1.41 (1.13-1.76) 4,390 12 MTHFR
23
Similar results in MI patients
OR (CI95) N Studies
1.26 (0.94-1.67) 3,395 6 FVL
0.89 (0.59-1.35) 4,478 3 PT
Age gt 55
OR (CI95) N Studies
1.34 (0.94-1.91) 2,542 6 FVL
1.86 (0.99-3.51) 1,159 3 PT
Age lt 55
SM Boekholdt et al. Circulation, 2001
24
The Copenhagen City Heart Study
1998
1991
1981
1976
9,253 individuals tested for FVL
IHD 834 (365 AP, 469 ac.MI) Ischemic stroke
410
K. Juul, Blood, 2002
25
CCHS 3 sub studies
K. Juul, Blood, 2002
26
Frequency of FVL in CCHS case-ctrls
Stroke MI IHD Ctrls
231 493 448 7,907 N (all)
16 (6.9) 47 (9.5) 34 (7.6) 612 (7.7) FVL -
1 (0.4) 1 (0.2) 2 (0.4) 17 (0.2) FVL
0.92 (0.56-1.52) 1.25 (0.92-1.78) 1.01 (0.71-1.44) OR (CI95)
27
Frequency of FVL in CCHS cohort
Stroke MI IHD Ctrls
410 469 365 7,907 N
22 (5.4) 30 (6.4) 26 (7.1) 612 (7.7) FVL -
1 (0.2) 1 (0.2) 2 (0.5) 17 (0.2) FVL
0.70 (0.47-1.06) 0.83 (0.59-1.18) 0.96 (0.67-1.39) OR (CI95)
28
The Copenhagen City Heart Study
29
Meta analyses (6 studies CCHS data)
IS children IS MI lt 50 y MI
4.79 1.05 1.61 1.2
3.26-7.03 0.73-1.50 1.09-2.38 1.03-1.39
30
Conclusions

• No clinically significant association of CVD with
  thrombophilia
• Interaction with other risk factors
• More significant in young (lt 55) patients
• No interaction with gender, smoking, diabetes,
  BMI, cholesterol, LDL cholesterol, Lp(a),
  fibrinogen levels.

31
Limitations

• Heterogeneity (study populations, selection of
  controls, outcome definitions, study size, degree
  of matching).
• Possible publication bias.
• Possible interaction with other CVD risk factors
  (e.g homocysteine blood level).
• Studies include only living patients.

32
Is there an interaction ?
33
(No Transcript)
34
OCA and the risk for MI
BC Tanis, NEJM, 2001
Population based case-ctrl study 248?aged 18
49 with MI compared to 925 random controls
(random digit dialing)
BC Tanis, BJH, 2003
Population based case-ctrl study 217?aged 18
49 with MI compared to 763 random controls
(random digit dialing)
35
Prothrombotic defects confer no risk for MI in
young women
FVL mutation
PT mutation
all ctrls MI
55 42 13 Y
925 721 204 N
980 763 217 all
all ctrls MI
23 18 5 Y
957 745 212 N
980 763 217 all
OR 1.1 (0.6-2.1)
OR 1.0 (0.4-2.7)
BC Tanis, BJH, 2003
36
Risk for MI not affected by OCA
all ctrls MI FVL/PT
49 36 13 Y
562 446 116 N
611 482 129 all
all ctrls MI FVL/PT
25 20 5 Y
338 258 80 N
363 278 85 all
OR 1.4 (0.7-2.7)
OR 1.9 (0.6-5.5)
OCA
OCA-
BC Tanis, NEJM, 2001
37
Positive interaction with other risk factors
Prothrombotic defect Prothrombotic defect
YES NO
OR (CI95) OR (CI95)
12.5 (5.7-27) 6.5 (4.3-9.7) Smoking
11.1 (1.1-114) 4.2 (2.7-6.6) Hypertension
1.7 (0.3-10.4) 3.7 (1.9-7.3) High cholesterol
4.0 (1.6-9.8) Diabetes
3.7 (1.3-10.2) 2.6 (1.7-3.8) Obesity
BC Tanis, BJH, 2003
38
More of the same HRT, prothrombotic defects and
the risk for non fatal MI in post menopausal
women
Population based case-ctrl study 232?post
menopausal (30 79) with MI compared to 723
ctrls (stratified according to BP status due to
different funding sources)
BM Psaty, JAMA, 2001
39
HRTs do not affect MI risk
HTN
HTN-
all ctrls MI HRT
163 124 39 Y
297 212 85 N
460 336 124 all
all ctrls MI HRT
186 145 41 Y
269 242 67 N
455 347 108 all
OR 0.78 (0.5-1.2)
OR 1.02 (0.7-1.6)
BM Psaty, JAMA, 2001
40
FVL does not affect MI risk
HTN
HTN-
all ctrls MI FVL
23 19 4 Y
467 363 104 N
490 382 108 all
all ctrls MI FVL
25 20 5 Y
435 316 119 N
460 336 124 all
OR 0.67 (0.3-1.5)
OR 0.66 (0.2-1.8)
BM Psaty, JAMA, 2001
41
PT mutation and HTN affect MI risk
HTN-
HTN
all ctrls MI PT
16 11 5 Y
444 325 119 N
460 336 124 all
all ctrls MI PT
15 7 8 Y
478 378 100 N
493 385 108 all
OR 1.24 (0.4-3.7)
OR 4.32 (1.57-12.1)
BM Psaty, JAMA, 2001
42
which disappears after stratification for HRT use
HRT-
HRT
all ctrls MI
7 5 2 Y
301 236 65 N
308 241 67 all
all ctrls MI
8 2 6 Y
177 142 35 N
185 144 41 all
OR 1.45 (0.28-7.66)
OR 10.9 (2.15-55.2)
OR 4.32 (1.57-12.1)
43
HRT, prothrombotic defects and the risk for non
fatal MI in post menopausal women
Interaction between PT mutation and HRT
use increase MI risk only in hypertensive women
!!!
44
(No Transcript)
45
The Copenhagen City Heart Study
Stroke MI IHD Ctrls
410 469 365 7,907 N
22 (5.4) 30 (6.4) 26 (7.1) 612 (7.7) FVL -
1 (0.2) 1 (0.2) 2 (0.5) 17 (0.2) FVL
FVL homozygosity does not increase risk for CVD
46
Combined carrier status of PT and FXIIIA Leu34
alleles as risk factors for MI evidence of a
gene-gene interaction

• MI a multifactorial disease
• Interaction of several genes may affect risk
• FXIIIA leu34 proposed as a protective allele
• Other studies yield contradicting results

C Butt, Blood, 2003
47
A unique experimantal sorrounding..
Newfoundland

• A small population (530,000)
• Descendents of 1718th century Irish settlers
• Little inward migration
• A homogenous genetic background

48
Study protocol
49
Increased risk in PT carriers only
all ctrls MI PT
21 5 16 Y
979 495 484 N
1000 500 500 all
all ctrls MI FVL
46 23 23 Y
954 477 477 N
1000 500 500 all
OR 1.0 (0.6-1.8)
OR 3.3 (2.6-4.0)
C Butt, Blood, 2003
50
which changes after stratification for FXIIIA
leu34
all ctrls MI FXIII34L
13 1 12 Y
979 495 484 N
992 496 496 all
all ctrls MI FXIII34L-
8 4 4 Y
979 495 484 N
987 499 488 all
OR 12.3 (1.6-94.8)
OR 1.02 (0.25-4.1)
51
Newfounland, Canada

• FVL was not a risk factor
• PT and FXIII 34Leu interact to increase the risk
• Risk was age and gender dependent

52
Conclusions

• There is no firm evidence to support a strong
  association between hereditary thrombophilia and
  CVD
• Interaction with other genetic or environmental
  risk factors may increase CVD risk
• Acquired thrombophilic markers may as well
  interact with hereditary risk factors

53
Practical points

• There is no sense in screening for thrombophilia
  in CVD patients
• Excluding APL Abs and elevated homocysteine blood
  level such screening will not affect possible
  treatment