To examine the relationship between Chlamydia pneumoniae (C. pneumoniae) seropositivity and carotid intima-media thickness (IMT) in three selected groups: 1) hypertensives; 2) white coat hypertensives; and 3) normotensives. Epstein-Barr antibodies were also measured.
The 340 participants underwent 24-h ambulatory blood pressure (BP) monitoring, clinic BP measurements, ultrasound carotid measurements, and serologic analysis (microimmunofluorescence and ELISA).
Significant differences in IMT were found between the three groups, regarding both mean internal carotid artery (MICA) and mean common carotid artery (MCCA) (one-way ANOVA). In all three groups, no association was found between the carotid IMT and the C. pneumoniae or Epstein-Barr seropositivity.
When the effect of BP is isolated, there is no association between C. pneumoniae antibody levels and carotid IMT.
Chlamydia pneumoniae (C. pneumoniae) is an obligate intracellular human pathogen responsible for a significant portion of atypical pneumonia and associated with a variety of chronic sequels, the most significant of which is atherosclerosis.1
Many experimental and serologic studies have demonstrated significant associations between the C. pneumoniae microorganism and coronary artery atherosclerosis or myocardial infarctions.2,3 However, a number of studies have failed to verify such an association.4 Similarly, the association between C. pneumoniae infection and carotid intima-media thickness (IMT) in the general population remains controversial.5,6
This inconsistency between the findings of previous reports regarding the carotid atherosclerosis may be the result of performing different serologic methods (ELISA, microimmunofluorescence test), or it could be explained by the clinical procedures used to define hypertension. Hypertension is an important risk factor that contributes to the genesis and establishment of the vascular disease, hence its presence in a sample should be carefully traced by the investigators to avoid significant methodologic errors.
The present study attempted to isolate the effect of blood pressure (BP) onto the carotid IMT by contrasting three sample populations—hypertensives, white coat hypertensives, and healthy normotensives—using BP as the grouping factor. In this way, the group-individual association between C. pneumoniae and carotid IMT could be compared and analyzed among the three groups and derive conclusions based on their exclusive relationship, ignoring the BP factor. To correctly identify sustained, white coat hypertensives and normotensives, both clinic and 24-h ambulatory blood pressure measurements were performed.
Epstein-Barr antibodies were also measured to clarify whether a possible association between carotid IMT measurements and C. pneumoniae antibodies exists, as well as other intracellular microorganisms, similarly common in Hellenic community.
Hypertensives and white coat hypertensives were selected among patients consecutively referred to the Outpatient Medical Clinic of Alexandra Hospital. The control group consisted by healthy normotensive subjects going to the hospital for routine medical check-up. The selection of the participants was based on the same criteria that were used in our previous studies, regarding the definition of hypertensives white coat hypertensives and normotensives.7,8
Five hypertensives, 2 white coat hypertensives, and 1 control subject refused to participate. Three hypertensives and one normotensive were excluded because they did not meet the serologic criteria (IgG ≥512 or IgA ≥512). The final sample consisted of 90 hypertensives, 60 white coat hypertensives, and 45 normotensives. All subjects gave their informed consent to participate in the study, which had been approved by the local scientific committee.
Clinic BP measurements
Blood pressure was measured three times during each examination, during three separate examinations on 3 different days, using a mercury sphygmomanometer.9
Ambulatory BP monitoring
Ambulatory blood pressure monitoring was performed using oscillometric Spacelabs 90209 equipment (Spacelabs, Redmond, WA).
The presence of C. pneumoniae-specific IgA and IgG antibodies in serum were determined by the microimmunofluorescence test. Elementary bodies of C. pneumoniae TW 183 strain, were used as a representative strain type (BIOS Gmbh Postfach 1640 D-82166 Grafelfing Munchen, Germany). The IgA titers ≥40 and IgG ≥80 in serum were considered as an increased antibody response.
The ELISA method (Gull Laboratories, Inc.) was used to detect antibodies of the Epstein-Barr virus.
Ultrasound carotid measurements
The scans were performed with a high-resolution ultrasound Doppler system (Acuson 128 XP, Mountain View, CA) using a 7-MHz linear transducer.
Definition of white coat hypertension
White coat hypertension was defined as the combination of increased clinic BP (>140 mm Hg systolic or 90 mm Hg diastolic BP) with an average daytime ambulatory BP of <135/85 mm Hg.10
Data are presented as mean ± SD. Analysis was performed using the SPSS 8.00 statistical package (SPSS, Chicago, IL). Independent sample t test was used to examine whether there is a significant difference in mean internal carotid artery (MICA) or mean common carotid artery (MCCA) IMT measurements between C. pneumoniae seropositive and seronegative subjects. Using the C. pneumoniae IgA and IgG titers as a continuous variable, the partial correlation was also used to indicate the existence of a possible relation between these parameters. Partial correlation was also used to clarify the connection between carotid IMT and antibodies against Epstein-Barr. One-way ANOVA was used to examine the difference in IMT between all three groups. Pearson's χ2 correlation was used to verify the association between C. pneumoniae antibodies and smoking habit and gender. Multiple linear and logistic stepwise regression analyses were performed, as appropriate, to control for conventional cardiovascular risk factors. A P value < .05 was considered significant.
In the three groups, the C. pneumoniae seropositive and seronegative groups, with regard to both IgG and IgA, did not significantly differ in socioeconomic status, age, blood cholesterol, triglycerides, glucose, HDL, LDL, C-reactive protein, and body surface area.
No statistically significant association between C. pneumoniae IgG, IgA seropositivity, and IMT measurements of MICA or MCCA was found regarding all three groups. Similarly, no correlation was found between Epstein-Barr antibodies (viral capsid antibody IgG and Epstein-Barr nuclear antibody) and carotid IMT. Epstein-Barr IgM and Epstein-Barr virus early antibodies were not detected in any of the participants examined.
The one-way ANOVA indicated a significant difference in MCCA IMT between controls and hypertensives (P < .000) and between controls and white coat hypertensives (P < .000). Regarding MICA, ANOVA revealed a significant difference between controls and white coat hypertensives (P < .005) and between controls and hypertensives (P < .009).
The IgG and IgA antichlamydial titers were higher in hypertensives (crosstabs, P < .002 and P < .000, respectively) and in white coat patients (crosstabs, P < .007 and P < .000, respectively) when both groups were compared to controls.
In the group of hypertensives a significant association was found between C. pneumoniae IgG seropositivity and the male sex (crosstabs, P < .03) as well as the smoking habit (crosstabs, P < .02), whereas IgG seropositive hypertensives had also significantly higher clinic systolic and diastolic BP values compared to IgG seronegatives (t test, P < .04 and P < .01. respectively). After controlling for age, the associations remained significant. Similarly, in the white coat group a positive association was observed between IgA seropositivity and smoking habit, as well as IgA seropositivity and gender (P < .02 and P < .014, respectively).
In hypertensives and in controls, a positive relationship was observed between MICA, MCCA, and smoking habit (in hypertensives P < .006 and P < .000, respectively, and in controls P < .04 and P < .003, respectively) which persisted after controlling for cardiovascular risk factors.
In all three groups a positive association was found between IMT of MICA, MCCA, and diabetes mellitus (in hypertensives P < .02 and P < .04, respectively, in controls P < .003 and P < .001, respectively, and in white coat hypertensives P < .04 and P < .03, respectively), hyperlipidemia (in hypertensives P < .001 and P < .003, respectively, in controls P < .002 and P < .001, respectively, and in white coat hypertensives P < .02 and P < .001, respectively), age (in hypertensives P < .02 and P < .01, respectively, in controls P < .002 and P < .001, respectively, and in white coat hypertensives P < .004 and P < .002, respectively), and obesity (in hypertensives P < .01 and P < .03, respectively, in controls P < .03 and P < .02, respectively, and in white coat hypertensives P < .02 and P < .01, respectively), and each association persisted after controlling for the other cardiovascular risk factors.
The findings of this study support that either past or chronic C. pneumoniae infection is not associated with carotid IMT. No statistically significant difference in IMT was demonstrated between the C. pneumoniae IgA/IgG seropositive and seronegative groups in all three groups. Seropositive and seronegative groups for both IgG and IgA antibodies in each group did not significantly differ regarding age, cholesterol, triglycerides, glucose, HDL, LDL, and body surface area. We found an association between C. pneumoniae seropositivity and smoking and male sex; this is in agreement with other reports.11 Although C. pneumoniae seropositive participants were older than seronegatives, no statistically significant association was found between age and C. pneumoniae seropositivity in the three groups.
It should be mentioned that C. pneumoniae IgG seropositive hypertensives had significantly higher clinic systolic and diastolic BP values compared to IgG seronegatives. Differences in clinic BP levels were also observed in white coat hypertensives but not at a statistically significant level, probably due to the sample size. Furthermore, antichlamydial titers were found higher in hypertensives and in white coat patients when both groups were compared to controls. These findings are in agreement with our previous reports,7,8 indicating a positive association between serologic evidence of C. pneumoniae infection and essential hypertension.
One source of skepticism could be the absence of significant association between carotid IMT and C. pneumoniae in the group of hypertensives, as both of these factors have been found, in previous studies,8 to be significantly associated with BP. One possible explanation for this paradox could be the existence of an unknown (nonlinear or noncontinuous) functional relationship between C. pneumoniae antibody titers and carotid IMT that cannot be captured by the performed tests (t test, correlation).
The existence of the well-known association between carotid IMT and sex, age, diabetes mellitus, hypercholesterolemia, and smoking was also observed.
The microimmunofluoresence test was used to determine C. pneumoniae seropositivity. The microimmunofluoresence test is generally accepted and the most widely used technique. However, it has been mentioned that the detection of C. pneumoniae antibodies does not ensure the existence of an infection,12 whereas other studies do not establish an association between the presence of the C. pneumoniae microorganism and the corresponding serologic evidence. In consideration of these limitations, it should be noted that chlamydial serology may not be used as a powerful and reliable marker for the presence of chlamydial infection.
No association existed between Epstein-Barr antibodies and carotid IMT. Antibodies against this virus were measured to elucidate whether our data indicate a specific association between carotid thickness and C. pneumoniae antibodies. The choice of Epstein-Barr was based on certain similarities to the C. pneumoniae microorganism. It is also an intracellular agent that is similarly widespread in the Hellenic community. In addition, the C. pneumoniae organism is an obligate intracellular organism that diverts the host cell's energy production mechanisms and lyses the host cell at the end of the replication cycle.13
Several studies5 are suggesting a positive relationship between C. pneumoniae infection and carotid thickness. Melnick et al14 in 1993, in a prospective cohort study, was the first to indicate an association between C. pneumoniae antibody titers and asymptotic carotid atherosclerosis. However, other reports found no link between them.6 It is well known that hypertension, together with other risk factors, is associated with the genesis of atherosclerosis, thus most of the previous studies have made an effort to eliminate this interference by using specific statistical methods. However, It should be emphasized that until the present the definition of hypertension has been based only on clinic BP measurements at the office by sphygmomanometer and on hypertension history. Nevertheless, several studies indicate that the definition of hypertension may differ if 24-h ambulatory BP monitoring is used, instead of BP readings. It is well known that BP may vary when measured at different times during the day or over short periods.15
Consequently, it is possible that a significant number of hypertensives might have been classified as normotensives in many studies and this might be the explanation for the opposing findings.
Assuming that the definition of hypertension—to the way BP was measured—might be responsible for the discrepancy of these results, the present study has used 24-h ambulatory BP monitoring, together with clinic sphygmomanometric measurements, for the classification of hypertensives, white coat hypertensives, and normotensives.
In our study, participants had no advanced atherosclerotic lesions; thus, our results are consistent with the conclusions of previous reports6 on the lack of association between early carotid atherosclerosis and C. pneumoniae infection.
In conclusion, our findings indicate that no relationship exists between C. pneumoniae and Epstein-Barr seropositivity on the one hand and carotid atherosclerotic lesions on the other. Nevertheless, due to the possible existence of a noncontinuous function, it could be speculated that the association between C. pneumoniae infection and carotid IMT would be stronger in population with more advanced carotid lesions, but this hypothesis needs to be further examined by combined serologic and experimental trials.
We extend our appreciation to D.M. Stamatakis for his valuable contribution in regards to the statistical analysis and procedures used in this article.