![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Journal of Clinical Microbiology, February 2002, p. 718-720, Vol. 40, No. 2
0095-1137/01/$04.00+0 DOI: 10.1128/JCM.40.2.718-720.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Laboratoire de Microbiologie,1 Departement de Cardiologie et Chirurgie Cardio-Vasculaire, Centre Hospitalier Universitaire, Sfax, Tunisia,3 Laboratoire de Bactériologie, Centre National de Référence des Infections à Chlamydia, Université de Bordeaux 2, Bordeaux, France2
Received 30 July 2001/ Returned for modification 30 September 2001/ Accepted 9 November 2001
 |
ABSTRACT |
|---|
 Top Abstract Case reportReferences |
|---|
|
CASE REPORT |
|---|
|
TopAbstractCase reportReferences |
|---|
On physical examination, her blood pressure was 110/40 mm Hg, her pulse was 72 beats/min, and her axillary temperature was 39°C. Cardiac auscultation revealed systolic mitral and diastolic aortic murmurs. In addition, the liver and spleen showed hypertrophy. The rest of the physical examination was normal.
An electrocardiogram showed a diastolic hypertrophy of the left ventricle. A chest radiograph was normal. An echocardiographic examination confirmed the mitral and aortic insufficiency but did not show any vegetations. Transesophageal echocardiography revealed severe aortic and mitral valve regurgitation, a vegetation (7 by 6 mm) on the right of the aortic valve, and a vegetation (5 by 6 mm) on the mitral valve.
After this examination a diagnosis of endocarditis was strongly suspected. The white blood cell count was 3.6 x 109/liter, the hemoglobin concentration was 7 g/100 ml, and the erythrocyte sedimentation rate was 75 mm/h (normal rate, <10 mm/h). Three pairs of aerobic and anaerobic blood specimens for culture (Hemoline; bioMerieux, Marcy l'Etoile, France) drawn before administration of penicillin and gentamicin remained negative. Blood cultures were incubated at 37°C for a total of 15 days, examined daily, and subcultured on conventional media at 5, 10, and 15 days. Serum samples obtained on admission and 2 and 4 weeks later showed negative results when tested for antibodies to the following agents: Legionella pneumophila, Mycoplasma pneumoniae, and Coxiella burnetii. However, the concentration of immunoglobulin G (IgG) antibodies to C. pneumoniae, as determined by microimmunofluorescence (MIF) assay, was higher than 1:4,096 for all three serum samples. There were also cross-reactions with antigens of Chlamydia trachomatis and Chlamydia psittaci in the three serum samples. All three serum samples were positive for IgM antibodies to C. pneumoniae by the MIF test. A specific IgG antibody response for C. pneumoniae and declining C. pneumoniae IgM titers during the observation period were detected by a commercial enzyme-linked immunosorbent assay (ELISA) with C. pneumoniae (Sero CP; Savyon Diagnostics Ltd., Biomedical Diagnostics, Marne La Vallée,, France) as the antigen. No cross-reaction was found by the ELISA with C. trachomatis (Sero CT; Savyon Diagnostics) as the antigen.
On 9 May 1998, a diagnosis of culture-negative endocarditis presumed to be due to C. pneumoniae was made, and doxycycline therapy was prescribed.
In July 1998, the aortic and mitral valves were replaced. Histologically, there was marked destruction of the valves with calcification and fibrosis. A mixed inflammatory infiltrate was evident. Cultures of valve tissue were negative for aerobic and anaerobic bacteria. The samples obtained at valvectomy were placed in sucrose phosphate transport medium and were stored at -80°C until they were processed by molecular biology-based techniques, PCR, and in situ hybridization (ISH). One part of each dissected valve was extracted by the cetyltrimethylammonium bromide method for PCR amplification, and another part was fixed in 10% nonbuffered formalin, dehydrated, and embedded in paraffin for ISH. Primers that amplify a 463-bp fragment of the 16S rRNA gene of C. pneumoniae (8), a 300-bp fragment of the pmp genes of C. psittaci (12), a 129-bp fragment of the major outer membrane protein gene (5) of C. trachomatis, and a 466-bp fragment of the P1 adhesin gene of M. pneumoniae (3) were used in four different assays. The amplified products were detected by hybridization with specific internal biotinylated probes with a commercial kit (Gen-Etik; Dia Sorin, Saluggia, Italy). PCR of specimens of the aortic and mitral valve tissues was positive for C. pneumoniae but negative for C. psittaci, C. trachomatis, and M. pneumoniae. A digoxigenin-labeled 498-bp probe specific for the omp1 gene of C. pneumoniae was used for the ISH assays and was prepared by a nested PCR as described previously (1). APNOU and APNOL were the outer primers, and APN1 and APN2 were the inner primers (2). The ISH test with specimens of the the aortic and mitral valve tissues was positive for C. pneumoniae (Fig. 1).
|
Discussion. C. pneumoniae is a well-known pathogen in community-acquired respiratory infections and has been associated with cardiovascular diseases in seroepidemiological studies and by demonstration of the pathogen in atherosclerotic lesions. Its role in endocarditis remains little documented. Among the approximately 20 published reports of Chlamydia endocarditis, C. psittaci was considered the etiologic agent in 11 reports, C. trachomatis was considered the etiologic agent in 2 reports, and C. pneumoniae was considered the etiologic agent in the remaining ones (4, 6, 10, 13, 14). In most of the cases, diagnosis was established solely by serology. The demonstration of a cross-reaction between bacterial species has raised the possibility of misdiagnosis. Maurin et al. (14) demonstrated that the serological diagnosis of C. pneumoniae endocarditis was erroneous in at least two reports (6, 13) and that most of the patients were probably suffering from Bartonella-induced endocarditis rather than Chlamydia-induced endocarditis. In some cases, in addition to serology, the diagnosis was made on the basis of electron microscopy and/or immunofluorescence performed with cardiac valve tissue specimens. Immunohistochemistry techniques should not be considered reliable diagnostic tests because of their lack of specificity for some chlamydial antibodies. Moreover, culture of Chlamydia species from blood or cardiac valve vegetations either was not performed or was negative. In our patient, the diagnosis of C. pneumoniae endocarditis, based initially on serology, was confirmed by the demonstration of the presence of C. pneumoniae by PCR and ISH techniques with the cardiac valve tissue that had been removed.
Clinical features compatible with endocarditis, histological evidence of a recent episode of endocarditis, and serologic changes appropriate for a recent C. pneumoniae infection suggested that this patient had culture-negative endocarditis due to C. pneumoniae. However, C. pneumoniae serology remains problematic. Because the prevalence of antibodies to C. pneumoniae among the general population is high, a fourfold increase in IgG levels is diagnostic, whereas by the MIF technique only titers of 
1:512 are highly suspicious. Hyman et al. (11) reported antibody titers in almost 20% of culture-negative and PCR-negative, subjectively healthy adults. Conversely, C. pneumoniae antibodies are not detectable by the MIF assay in the majority of children with culture-documented C. pneumoniae pneumonia or asthma. The titers obtained by the MIF assay in our patient were very high and implicated Chlamydia as the cause of infection. Since the surfaces of the elementary bodies contain genus-reactive antigens that lead to some cross-reactions in the MIF test, it is recommended that the three Chlamydia species be simultaneously included as antigens in every MIF test. When cross-reactions are observed, the specific reaction will have an endpoint titer twofold or greater than the titers observed for the two other Chlamydia species. The complete absence of IgG antibodies to C. trachomatis by specific ELISA and the presence of IgG antibodies to C. pneumoniae eliminated infection with C. trachomatis and suggested that C. pneumoniae was the causal agent. Declining C. pneumoniae IgM titers during the observation period suggested primary infection. Our findings were in accordance with those of Marrie et al. (13). Their patient had also shown declining IgM titers, whereas the patients described by other investigators had no IgM response, suggesting reinfection (4, 15, 16).
In our patient, the detection of C. pneumoniae in the aortic and mitral valve tissues by PCR and ISH tests confirmed the association between C. pneumoniae and endocarditis, and the negative PCR results for C. trachomatis and C. psittaci excluded the possibility of infection with these two species. This is the first report in which the presence of C. pneumoniae has been demonstrated in cardiac valve tissue by hybridization methods. In a recent report, Schaad et al. (16) described a myocardial infarction associated with a positive serologic reaction for C. pneumoniae and a positive PCR for a specimen from the left aortic leaflet. Those investigators proposed that their patient had C. pneumoniae endocarditis with concurrent myocardial infection. The tropism of C. pneumoniae for the human vascular system, confirmed in vitro by its ability to infect and proliferate in human macrophages, endothelial cells, and aortic artery smooth muscle cells (7), confirms the potential role of C. pneumoniae in endocarditis.
Experience with the therapy of Chlamydia endocarditis is too limited to make any recommendations. C. pneumoniae is susceptible to tetracyclines and erythromycin (9) but is resistant to penicillin, ampicillin, and sulfisoxazole. In our patient, after a lengthy postoperative treatment (1 year) with doxycycline, the patient showed an improvement in health. The optimal duration for such treatment is unknown. In some cases of valve replacement, eradication of infection was obtained after a 1- to 4-month course of postoperative antibiotic treatment (6). In all cases of culture-negative endocarditis not responding to classic empiric antibiotics, Chlamydia infection should be suspected.
C. pneumoniae should be considered a cause of bacterial endocarditis in patients with negative blood cultures, and chlamydial serological studies should be complemented by molecular biology-based methods to demonstrate the presence of C. pneumoniae in the tissues, when possible.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractCase reportReferences |
|---|
ur. Vaiss. 10:1423-1425.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Antimicrob. Agents Chemother. | Clin. Microbiol. Rev. |
|---|---|
| Clin. Vaccine Immunol. | ALL ASM JOURNALS |
|---|
