Author + information
- Received August 23, 2001
- Revision received December 26, 2001
- Accepted January 10, 2002
- Published online April 3, 2002.
- Moti Haim, MD*,†,* (, )
- David Tanne, MD*,‡,§,
- Valentina Boyko, MSc*,‡,
- Tamar Reshef, MSc*,†,
- Uri Goldbourt, PhD*,‡,
- Jonathan Leor, MD*,‡,
- Yoseph A Mekori, MD*,† and
- Solomon Behar, MD*,‡
- ↵*Reprint requests and correspondence:
Dr. Moti Haim, Cardiology Department, Rabin Medical Center, Beilinson Campus, Jabotinsky Street, Petach-Tikva 49100, Israel
Objectives The goal of this study was to assess soluble intercellular adhesion molecule-1 (sICAM-1) level as a predictor of future acute coronary events in patients with chronic coronary heart disease (CHD).
Background Increased sICAM-1 concentration has been shown to be associated with the incidence of CHD in healthy persons. Its significance in patients with CHD has been scarcely investigated.
Methods We designed a prospective, nested case-control study. Sera were collected from patients with CHD enrolled in a secondary prevention trial that evaluated the efficacy of bezafibrate in reducing coronary events. We measured baseline sICAM-1 concentration in the sera of patients who developed subsequent cardiovascular events (cases: n = 136) during follow-up (mean: 6.2 years) and in age- and gender-matched controls (without events: n = 136).
Results Baseline serum concentrations of sICAM-1 were significantly higher in cases versus controls (375 vs. 350 ng/ml; p < 0.05). Each 100 ng/ml increase in sICAM-1 concentration was associated with 1.27 (95% confidence interval [CI]: 1.00 to 1.63) higher relative odds of coronary events. Soluble ICAM-1 concentration in the highest quartile (>394 ng/ml) was associated with significantly higher odds of coronary events (compared with the lowest quartile), even after multivariate adjustment (2.31, 95% CI: 1.02 to 5.50). After adding fibrinogen and total white blood cell count to the multivariate model, the relative odds were 2.12 (95% CI: 0.88 to 5.35) and 2.70 (95% CI: 1.10 to 7.05), respectively.
Conclusions Elevated sICAM-1 concentration in CHD patients is associated with increased risk of future coronary events independent of other traditional risk factors.
Inflammation, whereby adhesion molecules play an important role, has been implicated in the pathogenesis of atherosclerosis (1). Cellular adhesion molecules mediate the margination, adhesion and transendothelial migration of circulating mononuclear cells from the blood stream to the extravascular compartment and have an important role in the progression of atherosclerotic plaque (1,2). In addition, adhesion molecules and various other cytokines recruit and activate mononuclear cells to release matrix metalloproteinases, promoting plaque rupture and the initiation of acute coronary syndromes (1,3–5).
Various inflammatory markers have been proposed to assist in the prediction of subsequent coronary events among healthy persons (6–8). Soluble intercellular adhesion molecule-1 (sICAM-1) and other adhesion molecules have been found to be associated with subsequent incidence of coronary heart disease (CHD) among healthy men and women in some studies (7,9–11)but not in others (12).
In some cross-sectional studies among patients with CHD, sICAM-1 was a better predictor for the presence of atherosclerosis and CHD than other adhesion molecules (13). However, other studies did not find elevated levels of sICAM-1 in patients with stable or unstable CHD (14,15). One small study among CHD patients found higher baseline serum levels of several adhesion molecules in patients who subsequently had cardiovascular events than in patients who did not develop cardiovascular events during follow-up (16). However, that study was too small to account for confounding variables.
In the present study, we evaluated the association between sICAM-1 levels and risk of future acute coronary events over a long follow-up time in patients with chronic stable CHD.
The sample for the present study was selected from the Bezafibrate Infarction Prevention (BIP) study. This was a placebo-controlled, randomized, secondary prevention trial that evaluated the efficacy of the lipid-lowering drug bezafibrate in reducing death and non-fatal myocardial infarction (MI) in patients (n = 3,090) with CHD (17).
In the BIP study, CHD was defined as either a history of MI >6 months, but <5 years, before enrollment to the study or a history of angina pectoris confirmed by a positive coronary angiography, by nuclear scintigraphy or by a positive exercise test (17). Primary end points of the study were fatal MI, non-fatal MI or sudden death. These end points were reviewed and confirmed by an independent critical event committee whose members were blinded to the treatment assignment of study participants. Mean length of follow-up was 6.2 years (range: 4.7 to 7.6 years).
For the purpose of the present study, we planned a prospective, nested, case-control study. Case subjects (n = 134) were randomly selected patients who reached the primary end point and provided an adequate blood sample before randomization to the study. Each case was paired with a control subject. Matching criteria were age, gender and participating medical center. Controls (n = 134) were patients with CHD who remained free of any recurrent coronary events and stroke throughout the follow-up period and were alive at the end of study.
Blood samples for measurement of serum lipids, fibrinogen, blood chemistry and other laboratory tests were collected at randomization (baseline) and thereafter at regular intervals. All laboratory analyses were performed in a single central laboratory using standard automated procedures with commercial kits (17).
For the purpose of the present study, plasma samples that had been taken at baseline from each study participant and stored at −70°C were thawed and assayed for sICAM-1 using a commercial enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, Minnesota). A single laboratory technician, who was blinded to the case or control status of each sample, performed all the tests.
Statistical analyses were done using the SAS statistical software. To assess the significance of differences between means of continuous variables, among the matched pairs, we used the paired ttest procedure. The McNemar test for paired samples was used to assess differences between rates.
We used logistic regression analyses conditioned on the matching variables (age, gender and medical center) in order to estimate the increase in relative odds of achieving the primary end point, associated with each 100 ng/ml increase in sICAM-1 concentration and with each 60 mg/dl increase in fibrinogen concentration. Adjustment was made for various baseline characteristics.
Conditional logistic regression (conditioned on the same variables as above) was used to calculate the odds ratios (OR) (and 95% confidence intervals [CI]) according to quartiles of sICAM-1 serum level. Division to quartiles was done according to sICAM-1 concentrations in a random sample of BIP study participants who were free from any cardiovascular events by the end of the study (n = 310). Similarly, we calculated ORs in quartiles of fibrinogen serum level.
We adjusted for the use of study medication, serum lipid concentrations at baseline and, finally, for multiple risk factors for coronary events (diabetes mellitus, hypertension, body mass index, New York Heart Association [NYHA] functional class, history of MI, current smoking), fibrinogen and white blood cell count.
In addition, conditional regression analyses were done to calculate ORs of future coronary events in tertiles of sICAM-1 concentration according to low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) concentrations (divided to tertiles).
As expected from the study design and matching criteria, cases and controls were well matched in terms of age and gender. Cases and controls were fairly similar in terms of various coronary risk factors (Table 1). Cases were in poorer functional class as assessed by NYHA functional class, and they had a higher body mass index. Mean serum concentration of fibrinogen and total white blood cell count were significantly higher in the cases than in the control group, and serum triglycerides were somewhat higher in cases than in controls (Table 1).
sICAM-1 level, distribution and correlates
Serum concentration of sICAM-1 was normally distributed in the study sample, ranging from 19.4 ng/ml to 785.6 ng/ml, with a mean level of 362.3 ng/ml (SD: 104.6 ng/ml) and a median level of 345.5 ng/ml. Distribution of cases and controls according to sICAM-1 concentration is shown in Figure 1. The mean serum concentration of sICAM-1 was significantly higher in cases (375 ng/ml [range: 19 ng/ml to 786 ng/ml]) than in the control group (350 ng/ml [range: 107 ng/ml to 707 ng/ml]).
Soluble ICAM-1 concentration correlated with triglyceride concentration (r = 0.28, p < 0.01), fibrinogen level (r = 0.16, p = 0.01) and white blood cell count (r = 0.24, p < 0.01).
Occurrence of acute coronary events
Each elevation of 100 ng/ml (1 SD), in sICAM-1 concentration, was associated with a 1.27 (1.00 to 1.63) increased relative odds of subsequent acute MI or sudden death (Table 2). Adjustment for multiple confounders related to coronary risk, including fibrinogen, or for white blood cell count did not change the relative odds substantially (Table 2). For comparison, in a similar multivariate model, each 60 mg/dl (1 SD) increase in fibrinogen concentration was associated with a 1.19 (0.92 to 1.55) increased relative odds of subsequent coronary events.
To explore the possibility that there is a threshold concentration above which sICAM-1 levels are associated with acute coronary events, we evaluated the risk for future events according to sICAM-1 concentrations divided into quartiles (see Methods section).
The unadjusted analysis of the matched pairs revealed a significantly higher risk for subsequent MI or sudden death in the highest quartile of sICAM-1 level than in the first quartile (OR: 2.51, 95% CI: 1.21 to 5.45, p = 0.02). Adding the study medication to the model did not change the result, nor did adjustment for serum lipids (Table 3).
Finally, a model that included multiple confounders associated with coronary risk produced an OR of future MI or sudden death of 2.31 (95% CI: 1.02 to 5.50, p = 0.02) in the highest quartile of sICAM-1 level (Table 3).
After adjustment for the inflammatory mediator fibrinogen, the relative odds of coronary events was 1.31 in the second quartile of sICAM-1 concentration, 1.24 in the third and 2.12 in the fourth (Table 3). Adjusting for white blood cell count yielded relative odds of 1.43, 1.25 and 2.70 in the corresponding quartiles of sICAM-1 concentration (Table 3).
For comparison, we evaluated future coronary risk according to fibrinogen concentration divided to quartiles. The crude relative odds of future coronary events were: 0.6 (0.25 to 1.29) in the second quartile of fibrinogen concentration, 0.56 (0.27 to 1.13) in the third and 1.40 (0.73 to 2.75) in the fourth quartile (p for trend = 0.18). Adjustment for multiple confounders did not change these estimates.
Risk according to LDL-C and HDL-C level
The relative odds of future coronary events increased with increasing sICAM-1 concentrations in the lowest and middle tertiles of LDL-C concentrations but not in the highest tertile (Table 4). Risk increased with increasing sICAM-1 concentration regardless of HDL-C levels (Table 4).
Time to event analysis
A previous report in healthy men showed that the risk associated with elevated sICAM-1 concentration was noted after two years of follow-up (9). In this analysis, the relative odds of sudden death or MI associated with sICAM-1 levels above 394 ng/dl was 1.37 (0.39 to 5.06, p = 0.62) in the first two years of follow-up, and it increased to 3.5 (1.39 to 9.67, p = 0.01) in events occurring after two years.
In this study, each 100 ng/ml increase in sICAM-1 concentration was associated with a corresponding 27% increase in the risk of future coronary events. Coronary heart disease patients in the highest quartile of sICAM-1 concentration were at particularly high risk, with 2.5 higher relative odds, compared with patients in the lowest quartile. The risk remained elevated after adjustment for traditional coronary risk factors and after adjustment for fibrinogen and for total white blood cell count. Soluble ICAM-1 was a stronger predictor than fibrinogen, which was previously shown to be associated with cardiac mortality and stroke (18–22).
The predictive value of elevated sICAM-1 levels in patients with stable CHD was only rarely investigated prospectively. Soluble ICAM-1 levels were significantly higher in patients with stable angina pectoris who developed cardiac events than in patients who did not develop any events during follow-up (16). However, this was a small study with only seven patients in the events group, and it was therefore not feasible to adjust for any possible confounders (16). To the best of our knowledge, our study is the first one to evaluate, both prospectively and over a long follow-up period, the risk associated with elevated sICAM-1 levels in patients with symptomatic but stable CHD.
By contrast with our study, in a recent community study, sICAM-1 was not independently associated with future development of CHD (12). However, it should be noted that in Malik’s study only 25% of the study participants had CHD at baseline, whereas in our study all the patients had proven CHD. In addition, sera that were tested for sICAM-1 were kept at −20°C in Malik’s study, compared with −70°C in the present study (this could affect sICAM-1 concentrations differently over time). Malik et al. (12)evaluated the top tertile of sICAM-1 concentration (vs. lowest tertile), whereas in our study, we compared the upper quartile with the lowest quartile.
In the Atherosclerosis Risk in Communities (ARIC) study and in the Physician’s Health study, elevated baseline levels of sICAM-1 predicted future development of carotid artery atherosclerosis (11)and CHD (9,11)in healthy persons, even after adjustment for inflammatory mediators (11), including C-reactive protein, fibrinogen (9)and other adhesion molecules (11). In addition, sICAM-1 levels were significantly higher in healthy women who subsequently developed CHD than in women who did not develop CHD (7).
Serum levels of sICAM-1, vascular cell adhesion molecule-1 (VCAM-1) and E-selectin increase during acute coronary events (24–29)and remain elevated for six months, but not thereafter (26). The BIP study included only patients whose coronary event had occurred six months or more prior to that study. The sample for the present study was selected from the BIP study, therefore eliminating any possible confounding related to increased sICAM-1 concentration after an acute coronary event.
Soluble ICAM-1, a member of the immunoglobulin super family, plays an important role in the adhesion of mononuclear cells to endothelial cells, promoting progression of atherosclerotic plaques and transformation of the stable plaque into an unstable plaque through invasion of circulating monocytes into the atheroma (1,4,5). Thus, elevated sICAM-1 levels in patients with CHD may indicate not only a tendency toward enhanced progression of the atherosclerotic process but also an inherent propensity toward plaque instability and rupture, hence the increased risk of recurrent MI and sudden death over time in patients with elevated sICAM-1.
The results of the present study add to the insight gained previously from studies in healthy persons, which have shown the importance of cytokines and circulating adhesion molecules as predictors of future atherosclerosis and CHD (7–11). More recently, VCAM-1 was found to predict recurrent coronary events in patients with unstable coronary syndromes (29). In the present study, on-going inflammatory activity, marked by elevated sICAM-1 concentration, predicted future coronary events in patients with chronic stable CHD.
This study relied on a single measurement of sICAM-1 in sera that were kept frozen for 10 years. Others have used similar methodology and laboratory methods to investigate the role of adhesion molecules and cytokines in CHD (7–12).
Studies such as ours cannot provide data on the origin of circulating sICAM-1. Whether from the coronary circulation or from systemic inflammation, the evidence for the strong association between inflammatory activity and subsequent risk for coronary events is nevertheless substantial.
We did not measure C-reactive protein or other inflammatory mediators. And we cannot compare their predictive value to that of sICAM-1. However, this kind of comparison was beyond the scope of the present study, which aimed to evaluate the role of sICAM-1 as a predictor of future acute coronary events in patients with stable CHD.
Despite these limitations, our study sample provides an excellent opportunity to assess the relationship between sICAM-1 and long-term (6.2-year) risk for acute coronary events in CHD patients. Our patients were under close monitoring for recurrent events, which were confirmed by an independent critical events committee.
Patients with CHD with elevated sICAM-1 levels were at increased risk for future MI and sudden death.Appendix
A complete list of Bezafibrate Infarction Prevention (BIP) study investigators and participating medical centers has been previously published (17).
☆ Supported by a research grant from the Israeli Society for Prevention of Heart Attacks (ISPHA).
- Bezafibrate Infarction Prevention study
- coronary heart disease
- confidence interval
- high-density lipoprotein cholesterol
- low-density lipoprotein cholesterol
- myocardial infarction
- odds ratio
- soluble intercellular adhesion molecule-1
- vascular cell adhesion molecule-1
- Received August 23, 2001.
- Revision received December 26, 2001.
- Accepted January 10, 2002.
- American College of Cardiology Foundation
- Hillis G.S,
- Flapan A.D
- Pasterkamp G,
- Schoneveld A.H,
- Van der Wal A.C,
- et al.
- Libby P,
- Simon D.I
- Ridker P.M,
- Rifai N,
- Stampfer M.J,
- Hennekens C.H
- Ridker P.M,
- Buring J.E,
- Rigai N
- Hwang S.J,
- Ballantyne C.M,
- Sharrett R,
- et al.
- Porsch-Oezcueruemez M,
- Kunz D,
- Kloer H.U,
- Luley C
- Wallen N.H,
- Held C,
- Rehqvist N,
- Hjemdahl P
- The BIP Study Group
- Benderly M,
- Graff E,
- Reicher-Reiss H,
- Behar S,
- Brunner D,
- Goldbourt U
- the Bezafibrate Infarction Prevention (BIP) study group,
- Tanne D,
- Benderly M,
- Goldbourt U,
- et al.
- Danesh J,
- Whincup P,
- Walker M,
- et al.
- Yano K,
- Grove J.S,
- Chen R,
- Rodriguez B.L,
- Curb J.D,
- Tracy R.P
- Morisaki N,
- Saito I,
- Tamura K,
- et al.
- Balbay Y,
- Tikiz H,
- Baptiste R.J,
- Ayaz S,
- Sasmaz H,
- Korkmaz S
- Mulvihill N.T,
- Foley B,
- Murphy R,
- Crean P,
- Walsh M
- Mulvihill N,
- Foley B,
- Murphy R,
- Curtin R,
- Crean P,
- Walsh M