Author + information
- Benjamin Wyplosz, MD, PhD⁎ (, )
- Gilles Montalescot, MD, PhD,
- Agathe Subtil, PhD,
- Annick Dujeancourt,
- Marie-Laure Tanguy, MD,
- Rémi Choussat, MD,
- Loïc Capron, MD, PhD and
- Alice Dautry-Varsat, PhD
- ↵⁎Institut Pasteur, Unité de Biologie des Interactions, Cellulaires, CNRS, URA 2582, 25, rue du Dr Roux, 75015 Paris, France
To the Editor:The relevance of Chlamydia pneumoniae(Cpn) in atherosclerosis is debated (1). Reliable diagnostic tests for endovascular bacteria require biopsies during atherectomy or bypass surgery. They are performed at advanced stages of atherosclerosis and are not suitable for exploring if Chlamydiainfection is an additional risk factor. We developed a new diagnostic approach to detect endocoronary bacteria. We extracted DNA from balloons after dilatation, detected Cpnby polymerase chain reaction (PCR), and correlated amplification results with coronary artery disease (CAD) severity.
We prospectively studied 101 unselected patients. Symptomatic CAD was acute coronary syndrome or stable angina. Asymptomatic CAD was silent ischemia related to hemodynamically significant stenosis.
Washed balloon catheters were cut off and frozen in 180 μl ATL buffer (DNA Mini Kit tissue; Qiagen, Courtabouef, France). Proteinase K (20 μl) was added; the sample was vortexed and lysed (1 h, 56°C); 180 μl was used as sample, and 20 μl was added to 160 μl phosphate-buffered saline as positive control: 1 μl plasmid DNA solution was added, containing 100 copies/μl of a 487-bp fragment of the MOMP gene (Cpn0695), cloned using APN1 and APN2 primers (2) into the pGEM-T vector (Promega, Charbonnières, France). The procedure followed the manufacturer’s instructions.
A nested PCR protocol, taking care to avoid contaminations (3), was performed using a Perkin-Elmer (Courtabouef, France) thermal cycler. Primers were APN1 and APN2, and for the second round they were 5′CTACTGGAACAAAGTCTGCG3′ and 5′CAGATACGTGAGCAGCTCTC3′.
Cycle conditions were: 5 min at 94°C, 35 cycles of 30 s at 94°C, 60 s annealing at 55°C, 60 s at 72°C, and final extension 7 min at 72°C. In the second amplification, annealing was at 50°C. Amplicons appeared as a 390-bp product on agarose gels. Every balloon was studied in triplicate with positive and negative controls. The PCR allowed detection of ten copies of bacterial DNA per tube.
Potential associations between patient data and CAD severity were tested by univariate procedures using Student ttests for age and chi-square or Fisher exact tests for categorical variables.
Multivariate analysis relied upon stepwise logistic regression. Variables with p < 0.10 in univariate analysis were included in the model. The p values were two-sided, and values of ≤0.05 were considered significant. Analyses were performed with the SAS 8.2 software (SAS Institute, Cary, North Carolina).
We examined the presence of CpnDNA in balloons from 101 patients. Nine were excluded from analysis for technical reasons: presence of PCR inhibitors (n = 3) and accidental contamination (n = 6). Five were excluded for medical reasons: coexisting restenotic and atherosclerotic lesions (n = 2), occluded lesions that could not be crossed (n = 2), and rejection after cardiac transplantation (n = 1).
The clinical and angiographic data of the 87 remaining patients are summarized in Table 1.Except for clinical history, the 76 patients with atherosclerosis and the 11 patients with restenosis were similar for risk factors, clinical status, and number of diseased vessels.
We analyzed 204 balloons: 176 from atherosclerotic and 28 from restenotic patients. The mean number of balloons studied per patient was about 2.5 in both groups. Thirty (39.5% [95% confidence interval (CI): 28.4 to 51.4]) of the 76 patients treated for primary stenosis were PCR positive, compared with none (0% [95% CI: 0 to 28.5]) of the 11 patients with restenosis (p = 0.01).
Table 2shows the clinical and angiographic characteristics of the patients treated for native atherosclerosis. Main risk factors were similar in both groups. The PCR-positive patients reported a familial history of CAD more often (p = 0.05), suffered from more severe CAD, and had significantly more frequent personal history of coronary events (i.e., myocardial infarction ≥1 month, acute coronary syndrome, stable angina, or revascularization procedures) (63.3% vs. 28.3%; p = 0.003), symptomatic CAD (80% vs. 56.5%; p = 0.04), and multivessel disease at angiography (80% vs. 52.2%; p = 0.01) than PCR-negative patients.
We examined whether Cpnwas independently associated with symptomatic CAD and multivessel disease by studying the following in univariate analyses: age, gender, diabetes, hypertension, hypercholesterolemia, smoking, CAD familial history, medical history, and positive PCR. In univariate analyses, positive PCR and diabetes were associated with symptomatic CAD. Multivariate analyses showed that only detection of Cpnby PCR was positively associated with symptomatic CAD (odds ratio 4.04 [95% CI: 1.23 to 13.28]; p = 0.021). However, and not unexpectedly, diabetes was negatively correlated (odds ratio 0.17 [95% CI: 0.05 to 0.59]; p = 0.005) with the existence of coronary symptoms.
Smoking status and positive PCR were positively associated with multivessel disease in univariate analysis. In multivariate analysis, positive PCR was the only variable significantly associated with multivessel disease (odds ratio 3.67 [95% CI: 1.26 to 10.64]; p = 0.017).
In conclusion, we report on the first assessment of angioplasty balloons to detect bacteria in coronary artery lesions. Chlamydia pneumoniaeDNA was detected in 39.5% of patients with native atherosclerotic stenosis. An overview of 43 studies analyzing 2,679 atherosclerotic lesions removed at autopsy or during surgery or atherectomy reported a bacterial detection rate of 24% by PCR amplification and of 48% by immunohistochemistry (3). However, the true prevalence is unknown as detection rates varied between investigators.
To standardize the procedure, we used a plasmid as internal PCR quality control, which ensured a threshold of 10 bacteria/tube. Thus, negative results do not mean that the ChlamydiaDNA was absent but rather below this threshold. Negative controls were patients with restenotic lesions, mostly because of in-stent restenosis, resulting from the formation of a neointima different from native atherosclerosis. If contaminations had occurred, experimentally or from circulating bacteria, these balloons should have been as often positive as those of atherosclerotic patients.
Chlamydiaserology was not performed because numerous studies have shown major limitations (4,5).
We showed that Chlamydia pneumoniaeis associated with: 1) clinical CAD severity; 2) multivessel disease; and 3) native atherosclerosis. The association between Chlamydiadetection and coronary lesion severity, independently of the main risk factors, strengthens infection as an atherosclerosis risk factor. Our diagnostic tool could help target therapeutic trials on patients with endovascular bacteria at earlier atherosclerosis stages than in previous studies. It could also allow assessing a putative relationship between Cpnpresence and atherosclerosis progression.
The authors thank Ms. Béatrice Broquet and Ms. Michéle Swiatek for collecting balloons.
Please note: This work was funded by the Institut Pasteur. Dr. Wyplosz received the Marcel Simon Prize from the Société Nationale Française de Médecine Interne.
- American College of Cardiology Foundation
- Ong G.,
- Thomas B.,
- Mansfield A.,
- Davidson B.,
- Taylor-Robinson D.
- Dowell S.F.,
- Peeling R.W.,
- Boman J.,
- et al.
- Boman J.,
- Hammerschlag M.R.
- Maass M.,
- Gieffers J.,
- Krause E.,
- Engel P.,
- Bartels C.,
- Solbach W.