| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Atherosclerosis and Lipoproteins |
From the Medical Department, Hanse-Hospital Stralsund, Teaching Hospital of the Ernst-Moritz-Arndt-University of Greifswald (R.M., M.S.), Institute of Clinical Chemistry and Laboratory Medicine (J.L.), Department of Neurology (U.S., C.K.), and Institute of Epidemiology and Social Medicine (U.J.), Ernst-Moritz-Arndt-University of Greifswald, Reference Method Laboratories, Institute for Standardization and Documentation in the Medical Laboratory (INSTAND), Duesseldorf (W.G.W.), Institute of Epidemiology and Social Medicine, University of Muenster (K.B.), Institute of Medical Microbiology and Hygiene, University of Lubeck (M.M., J.R.), and Hospital Itzehoe, Teaching Hospital of the University of Lubeck, Itzehoe (M.K.), Germany.
Correspondence to Rolf Mitusch, MD, Medical Department, Hanse-Hospital Stralsund, Teaching Hospital of the Ernst-Moritz-Arndt-University of Greifswald, Post Box 2341, D-18410 Stralsund, Germany. E-mail mitusch{at}uni-greifswald.de
| Abstract |
|---|
|
|
|---|
Methods and Results C pneumoniae DNA in PBMCs was determined by nested polymerase chain reaction and associated with IMT for 1032 healthy participants of a general population survey who were within the highest or lowest IMT distribution quartile. C pneumoniae DNA was more prevalent in those with increased IMT (13.4% versus 10.7%), but this was not significant in univariate and of borderline significance in multivariate analysis. Testing for potential effect modifications by known strong determinants of an increased IMT in group interaction analysis revealed an independent association between C pneumoniae DNA and IMT in normotensive subjects (odds ratio [OR], 2.06; 95% CI, 1.05 to 4.03; P=0.04) and in those <70 years old (OR, 1.84; 95% CI, 1.06 to 3.19; P=0.03).
Conclusions Asymptomatic atherosclerosis is associated with circulating C pneumoniae DNA independently of classical cardiovascular risk factors in normotensive subjects and those <70 years old.
C pneumoniae has been implicated in atherogenesis. We determined the association of chlamydial DNA in peripheral blood mononuclear cells with the carotid intima-media thickness from 1032 healthy subjects from a general population survey. A stratified group interaction analysis revealed an independent association in normotensive subjects and those <70 years old.
Key Words: Chlamydia pneumoniae infection atherosclerosis intima-media thickness
| Introduction |
|---|
|
|
|---|
Epidemiologically, it has been difficult to establish C pneumoniae infection as a distinct cardiovascular risk factor because a recognized marker for vascular infection is lacking. Whereas initial seroepidemiological studies reported an increased prevalence and level of C pneumoniae antibody in atherosclerosis, recent studies failed to show this association.3,4 C pneumoniae immunoassays experience several limitations. Antibody prevalence is high in the general population, individual levels vary considerably after seroconversion and reinfection, and test readings are subjective. It is unclear whether a certain antibody profile reflects past, persistent, or reactivated infection. In addition, the humoral immune response appears to be only weakly, if at all, correlated with the vascular presence of the intracellular pathogen.5 For these reasons, immunoassays measuring antibodies to C pneumoniae are considered unsuitable for diagnosis of vascular chlamydial infection.6
Taking these limitations into account, we decided to use direct detection of the pathogen in circulating peripheral blood mononuclear cells (PBMCs) by polymerase chain reaction (PCR) as a marker of systemic chlamydial infection in healthy subjects with known cardiovascular risk profiles from a large ongoing population survey. Presence or absence of circulating C pneumoniae DNA was then correlated with ultrasound measurements of the carotid artery intima-media thickness (IMT) as a surrogate marker of generalized subclinical atherosclerosis.
| Methods |
|---|
|
|
|---|
Anthropometric measurements and blood pressure were obtained using a standardized protocol. Prevalent hypertension was defined as systolic blood pressure
160 mm|Hg or diastolic blood pressure
95 mm|Hg or treatment in accordance with the MONICA-Augsburg protocol. Examinations included ultrasound measurements of the carotid arteries for subjects between 45 and 80 years old (2438 persons). A total of 532 women and 500 men with no history of myocardial infarction, stroke, or carotid artery stenosis were then included in our study whose common carotid IMT fell either within the highest (IMTq4) or the lowest (IMTq1) quartile of the sex-specific IMT distribution of all subjects examined.
Ultrasound Imaging
Trained certified technicians scanned the carotid arteries bilaterally for stenosis with B-mode ultrasound using a 5-MHz (range 3.5 to 7.0 MHz) linear array transducer and a high-resolution instrument (DIASONICS VST Gateway). The axial resolution of the probe at an insonation depth of 30 mm was 0.5 mm as determined with a tissue mimicking phantom (Gammex RMI 402GS). Stenosis was defined by a continuous wave Dopplerquantified occlusion of
50%. Plaque formation was defined as localized thickening of vessel wall structures that did not meet criteria for a stenosis. Additionally, scans from the distal straight portion (1 cm) of both common carotid arteries were recorded. The mean IMT was calculated by averaging 10 consecutive measurement points in 1-mm steps from the bulb by trained and certified readers. The maximum IMT was defined as the highest mean IMT of either the right or left side. Technicians and readers were not identical. Intrareader and inter-reader reliability was reported previously.9 Cutoff values for IMTq4 were >0.88 mm for females and >0.96 mm for males, and those for IMTq1 were <0.67 mm for females and <0.74 mm for males.
Laboratory Examinations
Serum and buffy coat samples were prepared from blood within 3 hours after collection. High-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and fibrinogen were determined by standard methods. High-sensitivity C-reactive protein (hsCRP) was determined by particle-enhanced immunonephelometry (hsCRP Kit; Dade Behring; measuring range 0.2 to 11 mg/L). Buffy coat samples for PCR were available from 997 subjects (96.6%). Genomic C pneumoniae DNA was amplified in a protocol described previously in detail for PBMCs and vascular tissue.10,11 Briefly, 1 to 10 µg DNA was purified from each sample and used in a nested PCR based on the species-specific HL-1/HR-1 and the nested IN-1/IN-2 primers, which yield a 128-bp product.12 For confirmation, nonradioactive DNA hybridization was performed using digoxigenin-labeled oligonucleotide HM-1 as the probe. PCR was controlled according to current guidelines.6
Statistics
The differences in cardiovascular risk factors between subjects within IMTq1 and IMTq4 were analyzed by univariate analysis, multivariate analysis by logistic regression, and by multivariate group interaction analysis. All values are given as means and 95% CI or as prevalences. Separate interaction terms were defined forC pneumoniae DNA positivity and presence of each of the following established risk factors: age (
70/>70 years), hypertension (yes/no), diabetes (yes/no), smoking (never/former or current), elevated LDL cholesterol (highest quartile among study participants, cutoff 4.55 mmol/L for females and 4.30 mmol/L for males), and elevated hsCRP (cutoff 3 mg/L). The resulting interaction terms had 4 categories: no risk factor/no C pneumoniae DNA, no risk factor/C pneumoniae DNA, risk factor/no C pneumoniae DNA, and risk factor/C pneumoniae DNA. This approach was chosen because it was anticipated that the known strong risk factors for increased IMT might conceal the Chlamydia-mediated effects. Univariate differences were tested by Wilcoxon and KruskalWallis or
2 statistics. Logistic regression analysis was applied to fit models of being in IMTq4 as a function of C pneumoniae DNA positivity, age, smoking behavior, diabetes mellitus, hypertension, HDL and LDL cholesterol, fibrinogen, and the interaction terms. Analyses were performed with PROC FREQ (CochranMantelHaenszel statistics), PROC GLM, and PROC LOGISTIC of the SAS 8.1 software system.
| Results |
|---|
|
|
|---|
|
|
Subjects who were positive for C pneumoniae DNA had a similar occurrence of plaque formation in the carotid arteries (67.5% versus 66.9%; NS) and a slightly higher prevalence of current smokers (24% versus 19%, NS) compared with those who were negative. Three percent of participants positive for C pneumoniae DNA reported a history of chronic bronchitis, which could possibly be associated with C pneumoniae infection.
Multivariate Analysis of Risk Factors by Logistic Regression
Table 2 shows the association of the above factors with increased sex-specific IMT in a multivariate analysis by logistic regression. An increase in age by 5 years doubled the risk (odds ratio [OR], 2.05) of being in IMTq4 after adjustment for important confounders. Diabetes resulted in a 3-fold and hypertension in a 2-fold increased OR for an increased IMT. An LDL cholesterol increase by 1 mmol/L was associated with a 40% higher risk. Current smokers had a 48% and individuals with circulating C pneumoniae DNA a 59% increased risk of being in IMTq4. However, the latter results were only of borderline significance.
|
Group Interaction Analysis of Circulating C pneumoniae DNA and Strong Modifiers of Risk
We also examined whether the relationship between C pneumoniae DNA and IMT was influenced by presence or absence of determinants for atherosclerosis. The factors age, hypertension, diabetes, smoking, elevated LDL cholesterol, and elevated hsCRP had been shown previously to be the major determinants of intima-media thickening in our population. Table 3 shows the results of these interaction analyses including the prevalence of C pneumoniae DNA in the 6 interaction categories.
|
In this stratified analysis, we found effect modifications by hypertension status (P=0.04), age >70 years (P=0.03), elevated hsCRP status (P=0.05), and diabetes (P<0.08). Accordingly, circulating C pneumoniae DNA was a risk factor for an increased IMT in normotensive subjects, in younger subjects (
70 years of age), in nondiabetic subjects, as well as in those with normal hsCRP levels. In addition, the data suggest that smokers as well as subjects with elevated LDL cholesterol with additional C pneumoniae DNA have a higher risk for intima-media thickening (NS).
Correlation of Circulating C pneumoniae DNA With Markers of Inflammation
In subjects positive for C pneumoniae DNA, markers of inflammation including hsCRP (3.47±8.34 mg/L versus 3.05±4.51 mg/L; P=0.14), fibrinogen (3.09±0.67 g/L versus 3.06±0.67 g/L; P=0.47), and ferritin (133.08±143.56 g/L versus 121.28±132,68 g/L; P=0.51) did not differ significantly compared with those who were negative.
| Discussion |
|---|
|
|
|---|
The influence of age on IMT was similarly strong in our general population survey. Age affects various cardiovascular risk factors in the elderly and is also known as a prominent confounder for patients >70 years of age, when IMT is used as the surrogate marker of atherosclerosis.20 An age-dependent weakening of the association between increased IMT and antichlamydial antibodies was also observed in 326 case-control pairs from the Atherosclerosis Risk in Communities (ARIC) study, which focused on asymptomatic carotid atherosclerosis.21
Most clinical studies on the association between infection and atherosclerosis have relied on the microimmunofluorescence test to detect the antichlamydial seroresponse. Because measuring the humoral immune response against a pathogen that chronically persists intracellularly cannot reflect disease status sufficiently, we applied a PCR to detect C pneumoniae DNA directly in circulating leukocytes.3,5,22 This diagnostic approach is based on the finding that circulating blood monocytes may act as a transport vehicle for C pneumoniae, spreading it systemically from a respiratory focus.2,23 PBMC infection results in a proatherogenic modification of the host gene expression profile, thus possibly linking monocyte infection and atherosclerosis. A recent microarray study showed C pneumoniae in human monocytic cells to induce transcriptional activation of a variety of genes with proinflammatory functions, including CD44 and endothelin-1.24 Also, previous studies have shown that monocyte infection resulted in shedding of proatherosclerotic mediators, such as TF-1 or CD 14.25,26,27
Use of direct detection of C pneumoniae has been applied in studies of patient populations and is reported here for the first time in a general population survey.3 The prevalences of C pneumoniae DNA in our study population are very similar to recently published studies that reported on C pneumoniae in 9% of healthy blood donors or 12% of patients undergoing coronary angiography.27,28 The latter study found no association between C pneumoniae DNA and angiographic coronary artery disease, whereas another study detected an association in men (9% versus 3%) but not among women.29 Other studies in patient populations found positive results in 26% to 59% of patients.10,30,31 One study could not detect C pneumoniae DNA in circulating leukocytes of patients undergoing coronary angiography.32 We are well aware that detection of circulating chlamydial DNA has not yet been validated as a method in a population survey. However, because this method directly reveals systemic presence of the pathogen or its components, we consider it to reflect chlamydial influence on atherosclerosis better than classical serological profiles. A positive result for circulating C pneumoniae DNA reflects current infection and may originate from recent asymptomatic respiratory infection. However, a recent longitudinal study shows the presence of circulating C pneumoniae DNA to become chronic in 50% of patients.33
We could neither detect an association between C pneumoniae DNA in PBMCs and markers of inflammation nor an additive effect of both on IMT. Systemic inflammation seems to be scarce in the presence of PBMC infection, and we could not confirm findings of a study showing an additive effect of inflammation and antibodies to C pneumoniae on incident coronary events.34
Increased carotid IMT, as determined by high-resolution ultrasound, is a well-established measure of early atherosclerosis and indicates an increased risk for future cardiovascular events.35 Because our study population had no definite clinical end points for cardiovascular disease (ie, myocardial infarction or stroke, and no ultrasound findings of high grade carotid stenosis), the present data support an association with early states of atherosclerotic disease. C pneumoniae DNA has been identified in early atherosclerotic lesions (Stary grade I and II) and in young persons, which supports the hypothesis of an interaction between atherosclerosis and the pathogen early in life.36,37 Identification of individuals with increased risks for cardiovascular diseases and early treatment of their modifiable risk factors is a prevention strategy to reduce the number of myocardial infarctions and strokes. Because an increased IMT can be caused by different factors, it rather reflects an accumulation of risk factors than a single cause, and thus goes beyond classical risk factor concepts. Several studies have investigated the relationship between C pneumoniae seropositivity and quantification of the carotid IMT21,38 and found significant associations even when different degrees of adjustments for cardiovascular risk factors and indicators of inflammation were applied. One study did not find any association.39
Our study is limited by its cross-sectional design, which does not allow establishment of a causal relationship between infection and atherosclerosis. However, the association between risk factor and end point is more reliable than in clinical studies because it is based on true disease prevalences. A potential weakness is the restriction of IMT measurement to the common carotid segment, which is justified by the better reproducibility of measurements from this site. A 5-MHz probe was used, which has a slightly lower axial resolution than probes with higher transducer frequencies but a better penetration, resulting in a lower number of dropouts attributable to poor image quality. IMT measurement was used exclusively to assign subjects to quartiles of IMT, and no conclusions have been drawn from measured IMT values. Our results are restricted to a white population and cannot be generalized to other ethnic groups.
We conclude that the proof of C pneumoniae DNA in circulating leukocytes carries an increased risk for thickening of the carotid artery IMT independent from classical cardiovascular risk factors in persons without hypertension and <70 years of age. The prospective clinical and therapeutic implication of the detection of circulating C pneumoniae DNA remains to be determined.
| Acknowledgments |
|---|
Received March 7, 2004; accepted October 28, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Bunk, I. Susnea, J. Rupp, J. T. Summersgill, M. Maass, W. Stegmann, A. Schrattenholz, A. Wendel, M. Przybylski, and C. Hermann Immunoproteomic Identification and Serological Responses to Novel Chlamydia pneumoniae Antigens That Are Associated with Persistent C. pneumoniae Infections J. Immunol., April 15, 2008; 180(8): 5490 - 5498. [Abstract] [Full Text] [PDF] |
||||
![]() |
S Alamowitch, J Labreuche, P-J Touboul, F Eb, P Amarenco, and for the GENIC Investigators Chlamydia pneumoniae seropositivity in aetiological subtypes of brain infarction and carotid atherosclerosis: a case control study J. Neurol. Neurosurg. Psychiatry, February 1, 2008; 79(2): 147 - 151. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Volanen, M. J. Jarvisalo, R. Vainionpaa, M. Arffman, K. Kallio, S. Angle, T. Ronnemaa, J. Viikari, J. Marniemi, O. T. Raitakari, et al. Increased Aortic Intima-Media Thickness in 11-Year-Old Healthy Children With Persistent Chlamydia pneumoniae Seropositivity Arterioscler. Thromb. Vasc. Biol., March 1, 2006; 26(3): 649 - 655. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2005 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |