| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Department of Cardiology and
2
Central Laboratories, Athens Euroclinic, Athens 11521, Greece.
3
Department of Cardiology, St. Thomas’ Hospital, London
SE1 7EH, United Kingdom.
4
Clinical and Molecular Epidemiology Unit, Department of
Hygiene and Epidemiology, University of Ioannina, School of Medicine,
Ioannina 45110, Greece.
5
Department of Medicine, Tufts University, School of
Medicine, Boston, MA 02111.
aAddress correspondence to this author at: Central Laboratories, Athens Euroclinic, 9 Athanassiadou St., Athens 11521, Greece. Fax 30-1-6416555; e-mail haliassos{at}moleculardiagnostics.gr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: In a cross-sectional study, we examined 103 consecutive patients undergoing cardiac catheterization for suspected CAD. We assessed the association of preprocedural CRP concentrations with clinical presentation (unstable angina) and angiographic features of coronary lesions.
Results: Twenty patients had unstable angina. Independent predictors of unstable angina included increased CRP [odds ratio (OR), 2.93 per 10-fold increase in CRP; 95% confidence interval (CI), 1.28–6.69; P = 0.01] and the presence of macroscopic thrombus (OR, 7.08; 95% CI, 1.33–37.8; P = 0.02). Thirty-two culprit lesions had macroscopic thrombus or eccentric/irregular discrete morphology without total occlusion. Increased CRP was the strongest predictor of such features (OR, 2.04 per 10-fold increase in CRP; 95% CI, 1.03–4.04; P = 0.04), and the effect was independent of the presence of unstable angina.
Conclusions: Among patients with suspected CAD undergoing coronary angiography, increased CRP is strongly associated with unstable angina and with specific high-risk features of the culprit coronary lesions.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The protocol was approved by the appropriate ethics committees, and patients provided informed consent. Coronary angiography was performed according to standard techniques through the right femoral artery. A low osmolarity ionic contrast medium was used during the procedure. Coronary lesions were classified according to American College of Cardiology/American Heart Association (ACC/AHA) nomenclature (10). A lesion was classified as type A if it was discrete (length <10 mm), with smooth contour, concentric with little or no calcification, less than totally occlusive, nonangulated (<45°), not ostial in location, and without major branch involvement or thrombus present. A diffuse (length >20 mm) lesion as well as a degenerated saphenous vein graft or a lesion that was totally occluded >3 months or extremely angulated (>90°) was classified as type C. All other lesions were classified as type B.
In addition to a history and physical with focus on cardiovascular risk
factors, patients had blood drawn for a lipid profile (including LDL,
HDL, triglycerides, and cholesterol), CRP, and C. pneumoniae
and CMV serologies. For patients with unstable angina symptoms, the
blood was drawn in evacuated glass tubes with gel and clot activator
(Vacutainer SST; Becton-Dickinson Inc.) <7 days from the onset of
symptoms and 
3 h before the catheterization. Serum CRP was measured
on an automated biochemistry analyzer with a particle-enhanced
immunoturbidimetric method using latex particles coated with monoclonal
anti-CRP antibodies and turbidimetry reading of the precipitate at 552
nm (CRPLX on Cobas Integra 700; F. Hoffmann-La Roche Diagnostics).
Typical values in healthy subjects are <8 mg/L (11). The
stated detection limit of the method is 0.25 mg/L, and the day-to-day
imprecision (CV), as evaluated on 10 consecutive days using duplicate
measurements of CRP T control N (Hoffmann-La Roche; target value, 4.4
mg/L) was 2.8%. C. pneumoniae and CMV serologies were
determined by ELISA: the EIA method (Labsystems) for C.
pneumoniae, and the CMV AxSYM method (Abbott Diagnostics) for CMV.
Values in healthy individuals are 
40 kilounits/L for
Chlamydia IgG, 12 kilounits/L for Chlamydia IgM
and IgA, and 15 kilounits/L for CMV IgG and IgM, according to the
manufacturers. CRP values were logarithmically transformed to achieve
normality. Cholesterol and triglycerides were measured enzymatically,
and HDL-and LDL-cholesterol were determined using direct methods on the
same Cobas Integra 700 automated analyzer, along with the CRP
concentrations.
For the troponin I (TnI) measurement, we used the Cardiac Troponin I
assay on the Stratus CS fluorometric analyzer (Dade Behring Inc.). The
detection limit of this assay is 0.03 µg/L, and the cutoff for
prognostication according to the manufacturer can be as low as 0.1
µg/L. CRP was 
8 mg/L in 37 of 101 patients in whom it was measured
(37%). This cutoff point of 8 mg/L represents the 97.5th percentile of
510 outpatients from our hospital (260 males, 250 females; age range,
28–65 years) without detected health problems, who presented for a
yearly check-up during the period in which we evaluated the new CRP
assay.
We evaluated the associations of various clinical, laboratory, and angiographic predictors with an unstable angina presentation using univariate and multivariate logistic regressions with maximization of likelihood. Multivariate models used backward elimination with likelihood ratio criteria for the selection of variables. Similarly, we evaluated whether clinical and laboratory predictors were associated with the presence of specific high-risk "acute" angiographic characteristics of the culprit lesion.
Not all features that qualify a lesion as type B or C in the ACC/AHA classification system are associated with an acute inflammatory process. Therefore, for the main analysis, we defined a priori that high-risk/acute lesions would include those with thrombus and those that were discrete, eccentric, and/or irregular and not totally occluded. This selection was decided a priori, because (a) a thrombus usually characterizes an acute process, and (b) long and concentric lesions and those that are totally occluded are more likely to be more longstanding, even if superimposed acute events cannot be ruled out. Similar to length, unfavorable location and proximal tortuosity are characteristics of type B or C lesions, because of increased difficulty in percutaneous transluminal coronary angioplasty, but there is no reason to believe that they would be predictors of acute inflammation. Lengthy lesions take longer to develop, but it is unknown whether they may also provide a higher risk for acute events. Therefore, in a sensitivity analysis, we considered both long and short eccentric/irregular lesions without total occlusion as being high-risk/acute lesions.
Analyses were conducted in SPSS 9.0 (SPSS Inc.). P values are two-tailed.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). None of the patients had TnI >0.6 µg/L on admission as
well as on catheterization. No subjects had CMV IgM antibodies.
Fifty-six percent had IgG antibodies to Chlamydia, whereas
IgA antibodies were less frequent and IgM antibodies were rare.
|
Twelve patients had no significant coronary lesions, 38 had type A, and
46 had type B lesions (Table 2
). A total of 32 lesions either had thrombus or were discrete
eccentric/irregular lesions without total occlusion. Fifteen patients
underwent surgical revascularization, 34 patients had plain balloon
angioplasty, and 18 patients underwent stent implantation.
|
Of all candidate laboratory and clinical predictors, unstable angina
was associated primarily with increased CRP (Table 3
). Unstable angina was present in only 7 of 64 patients with CRP
<8 mg/L, but was present in 12 of 37 patients with higher CRP values.
None of the nine patients without CMV IgG had unstable angina, but the
finding was far from being statistically significant (P
= 0.20). There was no discernible association with any of the
Chlamydia serologies when titers were examined as discrete
(positive/negative) or continuous variables. In multivariate modeling,
CRP was still the only variable selected. When angiographic
characteristics of the culprit lesion were also considered, the two
variables independently associated with a clinical presentation of
unstable angina were increased CRP [odds ratio (OR), 2.93 per 10-fold
increase; 95% confidence interval (CI), 1.28–6.69; P
= 0.011] and macroscopic thrombus on angiography (OR, 7.08; 95% CI,
1.33–37.8; P = 0.022). The fit regression model was:
logit(unstable angina) = -1.46 + 1.96(macroscopic thrombus) +
1.08log10(CRP).
|
The presence of a high-risk lesion with acute characteristics on
angiography was most strongly associated with a clinical presentation
of unstable angina and with increased CRP (Table 3
). Such lesions were
present in only one-fourth (16 of 64) of patients with CRP <8 mg/L,
whereas they were present in 16 of 37 patients with higher CRP values.
There was no discernible association with any of the
Chlamydia or CMV serologies. In multivariate modeling, the
odds of a high-risk/acute lesion increased 2.44-fold (95% CI,
0.84–7.12; P = 0.056) in the presence of unstable
angina and 1.76-fold (95% CI, 0.86–3.58; P = 0.072)
per each 10-fold increase in the CRP concentration. The fit regression
model was: logit(high-risk/acute lesion) = -0.80 +
0.56log10(CRP) + 0.89(unstable angina). There was
no strong evidence for a statistical interaction between CRP and
unstable angina in increasing the risk of high-risk lesions with acute
characteristics.
Among patients with unstable angina, when both CRP and TnI were considered, the magnitude of the association between CRP and high-risk lesions with acute characteristics was not altered (OR, 2.15 per 10-fold increase), whereas increased TnI (0.1 µg/L or higher) did not seem to be independently related with such lesions (OR, 0.3; P = 0.50). However, this analysis of independent effects should be interpreted very cautiously given the sparsity of the data, and an independent TnI effect could easily have been missed. Among patients with unstable angina and high-risk lesions, all patients with TnI >0.1 µg/L also had increased CRP (1.5 mg/L or higher). On the other hand, five patients with unstable angina and high-risk lesions had increased CRP of 1.5 mg/L or higher and did not have documented TnI >0.1 µg/L.
The same strength of associations was observed when we considered as
high-risk/acute lesions all those that either had thrombus or were
eccentric/irregular and nonoccluded regardless of their length. The
odds increased 1.97-fold for each 10-fold increase in CRP and 2.95-fold
in the presence of unstable angina. Patients with totally occluded
lesions had the lowest average CRP values (Fig. 1
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our study is the first to link CRP concentrations with lesion morphology. Other investigators (12) suggested that although TnI correlates with angiographic lesions likely to precede major cardiac events, CRP shows no such relationship. High-risk lesions in that study were defined largely according to the ACC/AHA classification and included type C lesions and total occlusions. Although such lesions are definitely associated with a decreased chance of successful percutaneous transluminal coronary angioplasty, they do not necessarily reflect an acute process. In fact, totally occluded lesions mean completed events.
In the present study, we isolated angiographic features that are more likely to be characteristic of an acute process from those that, although unfavorable for clinical outcome and treatment options, are unlikely to reflect whether the pathophysiology is acute rather than chronic. Using this distinction, we documented that CRP is associated with the presence of features such as the presence of thrombus or eccentric location and irregularity. Patients with totally occluded lesions tended to have low CRP, consistent with the hypothesis that CRP reflects acute inflammatory pathophysiology, not established events. It is conceivable that patients with totally occluded lesions have no active inflammatory component, and therefore, no significant increase in CRP values is expected. Our study did not have adequate power to examine the independence of increased TnI and CRP in predicting a high-risk/acute lesion morphology specifically in patients with unstable angina, but the available data suggested that the strength of the association of CRP was also maintained in this group of patients once TnI was taken into account.
Infectious agents may be responsible for triggering the coronary inflammatory process (7)(8). In our study, CMV and C. pneumoniae serologies did not provide any insight into potentially causal relationships. In particular, the very high CMV seroprevalence (91%) in our study population led to a low power to test hypotheses for associations with this marker. Other investigators have observed a relationship among CAD, CRP concentrations, and CMV seropositivity, but the CMV seroprevalence in their cohort was only 64% (13). Perhaps most importantly, serologic markers may not be fully adequate for characterizing the local coronary disease process for CMV or Chlamydia. Alternative approaches are thus warranted, such as evaluation of DNA concentrations (14) or molecular evaluation of the coronary plaques.
Angiography offers only visual information and has limitations. For example, we could assess only the presence of macroscopic thrombi, whereas microscopic thrombi are not uncommon. Furthermore, some macroscopic features defy categorization as being more suggestive of acute rather than chronic processes. Plaque length is one such example, but we have shown that the results are similar regardless of how this variable is handled in the analysis. Most importantly, vulnerability to rupture is related to microscopic characteristics of coronary plaques, such as increased macrophages, reduced smooth muscle cells, a large soft lipid core, and a thin plaque cap (15). Such features cannot be appreciated by angiography. Thus, some non-critical-appearing lesions may be at high risk for rupture. Such classification errors would mean that the true relationship between lesion morphology and CRP is probably even stronger than what we observed using angiography alone.
In conclusion, inflammation may be implicated in the transformation of coronary plaques to unstable stenoses with specific high-risk features. The identification of markers indicating the propensity of a plaque to undergo disruptive transformation is of clinical importance, and CRP measured on automated analyzers may be a simple and useful marker in this regard.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  F. Otsuka, S. Sugiyama, S. Kojima, H. Maruyoshi, T. Funahashi, K. Matsui, T. Sakamoto, M. Yoshimura, K. Kimura, S. Umemura, et al. Plasma Adiponectin Levels Are Associated With Coronary Lesion Complexity in Men With Coronary Artery Disease J. Am. Coll. Cardiol., September 19, 2006; 48(6): 1155 - 1162. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Avanzas, R Arroyo-Espliguero, J Cosin-Sales, G Aldama, C Pizzi, J Quiles, and J C Kaski Markers of inflammation and multiple complex stenoses (pancoronary plaque vulnerability) in patients with non-ST segment elevation acute coronary syndromes Heart, August 1, 2004; 90(8): 847 - 852. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Vlachoyiannopoulos, P. G. Kanellopoulos, J. P. A. Ioannidis, M. G. Tektonidou, I. Mastorakou, and H. M. Moutsopoulos Atherosclerosis in premenopausal women with antiphospholipid syndrome and systemic lupus erythematosus: a controlled study Rheumatology, May 1, 2003; 42(5): 645 - 651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. W. Chan, D. L. Bhatt, D. P. Chew, J. Reginelli, J. P. Schneider, E. J. Topol, and S. G. Ellis Relation of Inflammation and Benefit of Statins After Percutaneous Coronary Interventions Circulation, April 8, 2003; 107(13): 1750 - 1756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Bhatt and E. J. Topol Need to Test the Arterial Inflammation Hypothesis Circulation, July 2, 2002; 106(1): 136 - 140. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
