Significance of serum MMP8 in predicting CVD end points

1
Significance of serum MMP-8 in predicting CVD end
points
  Anita M. Tuomainen1, Kristiina Nyyssönen2, Jari
A. Laukkanen2, Taina Tervahartiala1, Tomi-Pekka
Tuomainen2, Jukka T. Salonen3, Timo Sorsa1,
Pirkko J. Pussinen1   1 Institute of Dentistry,
University of Helsinki, and Department of Oral
and Maxillofacial Diseases, Helsinki University
Central Hospital, Finland 2 Research Institute of
Public Health, University of Kuopio, Finland 3
Oy Jurilab Ltd, Kuopio, Finland
Tuomainen AM et al. Arterioscler Thromb Vasc
Biol. 2007 Oct 11 Epub ahead of print
2
MMP-8 AND TIMP-1 IN CARDIOVASCULAR DISEASES

• Some tissue studies of MMP-8 protein and mRNA
  expression exist.
• A few serum or plasma analysis of MMP-8 and
  TIMP-1
• Decreased serum MMP-8 concentrations in heart
  failure and cerebral ischemia.
• Increased plasma MMP-8 correlates with presence
  and severity of coronary artery disease (CAD).
• Plasma MMP-8 associates with carotid plaque
  instability and morphology.
• Plasma TIMP-1 is increased in acute coronary
  syndrome.
• Serum TIMP-1 associates with the presence of
  carotid lesions.
• TIMP-1 expression is slightly upregulated in
  abdominal aortic aneurysm.
• No long-term follow-up studies on the role of
  MMP-8 and TIMP-1 in cardiovascular diseases (CVD).

3
AIM
Do serum MMP-8 and TIMP-1 or their ratio predict
CVD events?
4
STUDY DESIGN

• Kuopio Ischemic Heart Disease Risk Factor (KIHD)
  study
• baseline yrs 1991-93, follow-up yr 2002
• 1018 men, 46-64 years, from which 905 were free
  from CVD at baseline
• Blood pressure, BMI, intima media thickness (IMT)
  were measured.
• Data on sociodemographic background, health
  status, and smoking habits were acquired by
  self-administered questionnaires
• During the follow-up 153 men of the 905 free from
  CVD at baseline experienced an end point - fatal
  or non-fatal AMI, CHD, or CVD event and death
  from any cause.
• Data on morbidity, medication, and cause of death
  was obtained from National Death and National
  Hospital Discharge Data registers

5
CHARACTERISTICS OF MEN FREE FROM CVD AT BASELINE
Mean (SD) All (n905) IMTlt1 mm
(n753) IMT?1 mm (n152)
p-value Age (yrs) 55.7 (6.6)
54.9 (6.5) 59.4 (5.7)
lt0.001 BMI (kg/m2) 27.5 (3.6)
27.4 (3.6) 27.9 (3.3)
0.114 Blood pressure 135 (17)
134 (16) 142 (15)
lt0.001 Smoking (cig/day) 4.3 (8.4)
4.0 (7.9) 5.8 (10.4)
0.046 Alcohol (g/week) 80.0 (113.8)
80.2 (114.5) 79.4 (111.8)
0.940 Cholesterol (mmol/l) 5.5 (0.9)
5.5 (0.9) 5.6 (0.9)
0.122 TG (mmol/l) 1.57 (1.01) 1.54
(1.02) 1.74 (0.95)
0.029 HDL (mmol/l) 1.11 (0.29) 1.12
(0.29) 1.05 (0.27)
0.003 Fibrinogen (g/l) 3.13 (0.57)
3.10 (0.57) 3.26 (0.55)
0.002 Antibody levels to periodontal pathogens
(EU) 9.31 (4.34) 9.18 (4.21)
9.91 (4.91) 0.057 n
() p-value Dentures 479 (52.9)
389 (52.1) 90 (60.0)
0.076 Diabetic 51 (5.6)
38 (5.0) 13 (8.6)
0.087 Current smoker 245 (27.1)
200 (26.6) 45 (29.6)
0.441 High antibody level for periodontal
pathogens 133 (14.7) 103 (13.7)
30 (19.7) 0.054
For difference between men with IMTlt1mm and ?1
mm.
6
MMP-8 AND TIMP-1 CONCENTRATIONS IN SUBJECTS
ACCORDING TO END POINTS AND SMOKING HABITS
Median Selection parameter
(n) MMP-8 TIMP-1
MMP-8/TIMP-1 (µg/l) (ng/ml)
(mol/mol) IMT at baseline lt1mm (753) 8.3
190.1 14.3 1 mm (152) 9.1
191.8 15.5 p-value ns.
ns. 0.046 AMI in
follow-up no (819) 8.4 190.1
14.6 yes (86) 8.5 197.6
15.6 p-value ns. ns.
ns. CHD death no (872)
8.4 365.8 14.6 yes
(33) 9.3 188.1
15.4 p-value ns. ns.
ns. CVD death no (852) 8.4
190.9 14.7 yes (53)
9.3 186.8 14.6 p-value
ns. ns.
ns. All cause death no (802) 8.3
190.1 14.9 yes (103) 9.4
197.4 14.2 p-value ns.
ns. ns. Current
smoker no (662) 8.0 186.7
13.9 yes(243) 11.3 203.8
17.7 p-value lt0.001 0.005
0.001
 
ns. non-significant
7
MULTIVARIATE LINEAR REGRESSION MODEL FOR MMP-8
AND TIMP-1 CONCENTRATIONS IN MEN FREE FROM CVD AT
BASELINE
ß, p-value MMP-8
TIMP-1 MMP-8/TIMP-1
(µg/l) (ng/ml)
(mol/mol) Age (yrs) -0.135,
lt0.001 0.073, 0.041 Cholesterol
(mmol/l) -0.091, 0.011 Smoking
(cig/day) 0.051, 0.131 0.141, lt0.001 BMI
(kg/m2) -0.115, 0.001
-0.060, 0.068 Triglycerides (mmol/l)
0.107, 0.006 HDL (mmol/l)
-0.049, 0.148 0.128, 0.001 Fibrinogen
(g/l) 0.237, lt0.001
0.195, lt0.001 Dentures (no/yes) 0.053, 0.105
0.046, 0.163 Alcohol (g/week)
-0.049, 0.150
Independents added by the backward selection
method, excluded variables included systolic
blood pressure, diabetes, socioeconomic status,
mena IMT, and serum antibody levels to major
pariodontal pathogens. The 5 best predictors are
shown.
8
RELATIVE RISK (RR) FOR AN END POINT (/1 SD of
increasing serum MMP-8 concentrations or
MMP-8/TIMP-1 ratio) IN MEN FREE FROM CVD AT
BASELINE
RR (95 CI) / SD increase,
p-value End point With/without MMP-8
MMP-8/TIMP-1
(µg/l) (mol/mol) AMI
86/819 1.138 (1.009-1.284), 0.036
1.112 (0.957-1.293), 0.166 CHD death 33/872
1.188 (1.034-1.365), 0.015 1.206
(1.028-1.414), 0.022 CVD death 53/852
1.171 (1.026-1.338), 0.020 1.160
(0.986-1.364), 0.073 All cause death 103/802
1.136 (1.018-1.269), 0.023 1.127
(0.991-1.281), 0.068
Adjusted for age, BMI, smoking, diabetes,
systolic blood pressure, serum cholesterol
concentration, HDL, and plasma fibrinogen
concentration.
9
RELATIVE RISKS (RR) FOR AN END POINT IN MEN FREE
FROM CVD AT BASELINE, MEAN IMT ?1 mm
Adjusted for age, BMI, smoking, and plasma
fibrinogen concentration.
10
CUMULATIVE SURVIVAL RATE OF MEN FREE FROM CVD,
WITH IMT ? 1 mm
RR 3.03 (1.09-8.44) p0.034
MMP-8, highest quartile
Cumulative survival without CVD death
MMP-8, three lower quartiles
Follow-up (days)
Adjusted for age, BMI, smoking, and plasma
fibrinogen concentration.
11
CONCLUSIONS

• Serum MMP-8 is elevated in subclinical
  atherosclerosis.
• High serum MMP-8 and MMP-8/TIMP-1, but not TIMP-1
  alone, predict a poor CVD outcome in men,
  especially in those with subclinical
  atherosclerosis.
• MMP-8 serum analysis may be useful for assessing
  CVD risk.