INTRODUCTION
Bartonella quintana is a Gram-negative bacillus belonging to the α2 subgroup of Proteobacteria. The first recognized clinical manifestation of this louse-borne pathogen was trench fever,1 described first in Europe during World War I and then re-emerging during World War II, resulting in large-scale epidemics.2 The infection is ubiquitous and has been described in Japan, Peru, Mexico, and Burundi.1 Later, B. quintana was recognized as an etiologic agent of cutaneous bacillary angiomatosis, bacteremia, infective endocarditis (IE), and chronic lymphadenopathy in patients infected with HIV.3,4 In the past decade, B. quintana has gained importance as an agent of blood culture negative endocarditis (BCNE) in immunocompetent patients.5,6 Since then, many cases of endocarditis due to B. quintana and other species of the genus Bartonella have been reported.7,8 They are thought to account for 1% to 3% of IE in Europe and Canada. Fournier and others found that IE due to B. quintana endocarditis is associated with alcoholism, homelessness, and infection by body louse, which is the principal vector.8 Direct methods including culture and PCR from cardiac valves, when available, are the best tools. However, the culture of these bacteria is difficult, so diagnosis of IE due to B. quintana is mainly by serology. Previously, we have shown that microimmunofluorescence (MIF) has a positive predictive value of 95.5% at a cut off titer of 1:800 for IgG antibodies against B. quintana.9 The other advantage of MIF is that it can be performed for Bartonella species and the other causative agents of BCNE such as Coxiella burnetii, in parallel.10 However, serological cross-reactions between Bartonella spp. and Chlamydia spp. and Coxiella burnetii exist.5,11 Western immunoblotting with cross-absorption is more specific.12 Indeed, many cases of supposed Chlamydia endocarditis have been demonstrated to be in fact Bartonella endocarditis.13,14
From Tunisia in North Africa, there are no reports of B. quintana endocarditis, but a case of chlamydial IE has been reported.15 The aim of our study was to retrospectively diagnose the etiologic agents of endocarditis in patients suffering from BCNE in Sfax, Tunisia.
PATIENTS AND METHODS
Patients.
Sera from patients suspected to have endocarditis were collected at the CHU Habib Bourguiba hospital of Sfax from January 1997 to December 2003. IE was diagnosed using the modified Duke criteria.16
Serology.
Screening: sera were screened for Chlamydia pneumoniae, C. psittaci, and C. trachomatis in an MIF assay using antigens prepared in house from embryonic eggs. IgG titers of 1:64 or above were considered positive. Sera were also tested using C. burnetii phase II antigens (BioMérieux, Marcy L’Etoile, France). The cutoff for IgG was1:200. Sera positive either for Chlamydia species or C. burnetii were stored at −80°C until their use in the “Unité des Rickettsies” in Marseille, France.
Serology for Bartonella spp. and C. burnetii was performed using a MIF assay as previously described.9 The cutoff value to diagnose endocarditis was 1/800 for both IgG against phase I C. burnetii and Bartonella spp.10
Western immunoblotting was performed on samples positive for Bartonella as previously reported.17 Western blot analysis was performed both before and after cross-absorption with B. quintana and B. henselae antigens.
For one patient, diagnosis of chlamydial endocarditis was made previously by PCR on valve tissue.15 For this patient, a Western blot with C. pneumoniae antigens provided by MicroBix (Biosystems Inc., Ontario, Canada) before and after cross-absorption with either C. pneumoniae or B. quintana was performed.
Molecular biology.
DNA extraction was performed from sera with a QIAamp blood kit (Qiagen, Hilden, Germany). We performed two light cycler nested PCRs (LCN) in one step according to our previous report18 using external and internal primers amplifying rnpb gene19 and fur gene.20 Sequences of primers are shown in Table 1. A negative control of serum was added every three samples. The positive control was Bartonella elizabethae.
Purified PCR products were sequenced using an ABI 3100 automated sequencer (Perkin-Elmer, Coigniers, France). Nucleotide sequences were edited with the Autoassembler package (version 1.4; Perkin Elmer, Coigniers, France). Multiple alignments with Bartonella spp. sequences available from GenBank were carried out using the CLUSTAL W software, version 1.81.
RESULTS
At screening, 40 sera were positive against Chlamydia spp. antigens whereas one serum was positive for C. burnetii. Using specific MIF assay in the French national reference center, the patient with positive serology for C. burnetii was shown to have phase I IgG titers at 1:1600, thus confirming the diagnosis of Q fever endocarditis.21 Among the 40 patients with positive results for Chlamydia spp., 13 were positive with Bartonella spp. with IgG titers > 1:800 but were negative for C. burnetii antibodies (Table 2). Western blot was able to confirm diagnosis of B. quintana in 12 cases (92%) and B. henselae in one case. Of note, none of the remaining 27 Chlamydia positive and Bartonella negative sera were sampled from patients with IE.
PCR results are shown in table 2. LCN PCR amplifying rnpb gene was positive in 58.3% whilst LCN PCR targeting the fur gene in 92.3 Among the 13 patients with positive Bartonella serology, the diagnosis of definite endocarditis according to Duke criteria was made in 11 cases. During the period of the study, 112 definite diagnoses of IE were made in the cardiology department at CHU of Sfax, 51 (45.5%) of which were classified as BCNE. Therefore, these 11 cases of B. quintana endocarditis accounted for at least 9.8% of all definite IE. For the remaining two patients (nos. 10 and 13), lack of cardiac echography did not allow us to confirm the diagnosis. These two cases were considered possible diagnosis of IE. Interestingly, for patient no. 4 a diagnosis of endocarditis due to C. pneumoniae was previously made via PCR of 16S rRNA gene on cardiac valve tissue and by in situ hybridization.15 For this patient, LCN PCR targeting rnpb gene on serum was negative but LCN PCR targeting fur gene was positive. The sequence amplified was identical to B. quintana with a percentage homology of 98%. Furthermore, the western blot with cross absorption for this patient showed an infection with B. quintana. Epidemiologic data was available for 12 patients. The mean age of patients with Bartonella IE (possible or definite) was 39.58 ± 12.55 years with a range of 20 to 56 years. Sex ratio M/F was 3.33 with 10 (76.9%) men and 3(23.1%) women. Eleven patients (92%) were of rural origin with poor living conditions. Eight patients had contact with animals (66.6%). Seven patients (58%) had a history of valvulopathy whilst the patient with B. henselae had a prosthetic valve. One patient was immunocompromised; he had drug induced lupus and was undergoing treatment with corticosteroids. Seven (58%) patients presented with fever. One patient first presented with renal insufficiency and had extra-capillary glomerulonephritis. Five cases had mitral valve involvement, two had aortic valve involvement, four had mid-nenent a both valves and for the last case there is no data. All patients but three had received β-lactam and aminoglycoside therapy. The patient who did not receive antibiotics had is-chemic cerebral vascular accident and died before commencing treatment. Seven (58%) patients underwent valvular replacement and recovered. Unfortunately, valves removed from these patients were not stored. One patient did not undergo surgery and recovered.
DISCUSSION
In this study, using a variety of techniques, we have confirmed Bartonella IE in patients in Tunisia. These methods have been previously reported to be both specific and sensitive.9,10,17,18 Using MIF, all 13 patients had titers ≥1:800. Such high titers have a positive predictive value of 95.5% for endocarditis.9,10 Western blot with cross-absorption was used to discriminate between species.12 Moreover, LCN PCR amplifying rnpb gene was positive in 7 cases (58.3%), this percentage was similar to our previous report in which sensitivity of LCN PCR with ribC gene was 58.1%.18 The LCN PCR on sera targeting fur gene increased the sensitivity of the results; it was positive in 12 cases (92.3%). This is somewhat surprising because the two LCN PCR target genes have only one copy in the Bartonella genome.20,21 Our study reemphasizes the fact that the existence of cross-reactions between Bartonella and Chlamydia can lead to wrong diagnosis. Patients with B. quintana endocarditis were shown to display cross-reacting antibodies to C. pneumoniae, C. psittaci,5 and C. burnetii.11 It has been previously demonstrated that almost all cases of Chlamydia endocarditis were in fact Bartonella endocarditis.13 The major Chlamydia-cross-reacting protein antigen was about 60 to 70 kDa, which may correspond to a heat shock protein. Although Bartonella endocarditis has been well characterized by culture and/or PCR from valves, diagnosis of Chlamydia endocarditis was often based solely on serology. There is only one report of the presence of Chlamydia pneumoniae in cardiac valve by PCR and in situ hybridization.15 Interestingly the case of this patient corresponds to the patient no. 4 in our series. However, serology, western blot and PCR against Bartonella were not performed for the patient in this report.15 Using Western blot, we have shown that this patient had in fact Bartonella endocarditis. The presence of C. pneumoniae in his cardiac valve could be explained by the fact that this bacterium has a tropism for the human vascular system. It has especially been associated with cardiovascular diseases and detected in atherosclerotic plaques.22
Twelve (92%) of our cases were due to B. quintana, B. henselae was found in only one case. In the literature B. quintana accounts for three quarter of Bartonella endocarditis.8 B. quintana endocarditis has been associated with alcoholism, homelessness and body lice.8 In our study, no patient was homeless or an alcoholic and none had evidence of body lice. However, 11 (92%) of the infected were from rural areas and 8 (66.6%) had contact with animals. Although the only known reservoir of B. quintana is human23 the bacteria has recently been demonstrated in fleas.24 These may therefore play a role in the transmission of the bacteria to humans from a non human reservoir. A study performed in rodents, Psammomys obesus, in Tunisia showed that 49% were infected with Bartonella species with a recrudescence in August and September.25
In our series, patients treated with aminoglycosides and β-lactam antibiotics were cured. However, valve replacement was necessary in the majority of cases due to extensive valvular damage as previously reported.26 We have previously recommended that effective antibiotic therapy for Bartonella endocarditis should include an aminoglycoside for at least 14 days to achieve a bactericidal effect.26,27 Patient no. 7, who did not receive antibiotics, died.
In our study, BCNE accounted for 45% of IE. The incidence of BCNE may vary considerably from 1 to 79%.28 It is reported to be higher in developing countries (53 to 79% in India, 47.5% in south Africa and 48% in Pakistan).28,29 The main cause of BCNE is previous antibiotic therapy. Many etiological agents have been incriminated such as HACEK group bacteria, Bartonella, C. burnetii, and Tropheryma whipplei.30 Several studies of Bartonella endocarditis have previously been published with percentages ranging from 0 to 4.5% of all IE.7,31–33 In our series, we found a higher prevalence of Bartonella endocarditis (at least 9.8%), which is probably due to differences in living conditions or higher temperature, but in any way to homelessness. Data about body louse infection in our patients was lacking (Figure 1).
In conclusion, B. quintana appears to be a frequent agent of BCNE in Tunisia. Therefore specific serology should be performed systematically whenever diagnosis of IE is suspected. If left untreated, Bartonella endocarditis can be fatal. The treatment recommended for these infections should include aminoglycosides for at least 2 weeks in association with amoxicillin for 6 weeks.26,27 However, more prospective studies are needed to draw conclusions with regard to epidemiologic characteristics and clinical aspects of this infection.
Sequence of primers used in LCN PCR
| Sequences | Length of amplified fragment | |
|---|---|---|
| Primers of rnpb gene | ||
| External primers forward | 5′-GRTCCGGGGAGGAAAGTC-3′ | |
| External primers reverse | 5′-GGAAAAACGRTGCCGGGTA-3′ | 256 pb |
| Internal primers forward | 5′-AACCTACCCRAATGACTG-3′ | |
| Internal primers reverse | 5′-CCATYCATCTRGGACGGAT-3′ | 190 pb |
| Primers of fur gene | ||
| External primers forward | 5′-GCAAGAAYTACGCAYTGCAGG-3′ | |
| External primers reverse | 5′-AMATCAATGAGATGATCATGGTG-3′ | 263 pb |
| Internal primers forward | 5′-GTCAAMGACGTATTATTCTTGA-3′ | |
| Internal primers reverse | 5′-CCACCACTAAARSTATGGCG-3′ | 175 pb |
Results of serology, Western blot, PCR, antibiotic therapy, and outcome of patients with B. quintana endocarditis from cardiology service of CHU of Sfax, Tunisia
| Serology | LCN PCR | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Case/sex/age | Chlamydia | BQ | BH | WB | rnpb gene | fur gene | Final diagnosis | Antibiotic therapy | Operated | Outcome |
| Antibiotics: amik, amicacin; amphoB, amphotericin B; Ampi, ampicillin; doxy, doxycyclin; genta, gentamicin; oflo, ofloxacin; Peni G, penicillin G; vanco, vancomycin; sxt, cotrimoxazole. BH, B. henselae; BQ, B. quintana; Cb, Coxiella burnetii. LS, lost of sight; WB, Western blot. | ||||||||||
| * For patients 2 and 13, no data were available. | ||||||||||
| † Diagnosis of endocarditis due to C. pneumoniae was made by PCR performed on the valve. | ||||||||||
| 1/M/44 | + | 6400 | 6400 | BQ | BQ | BQ | BQ | Ampi + gena/ciflox | Yes | Recovery |
| 2/F* | + | 1600 | 3200 | BQ | BQ | BQ | BQ | |||
| 3/M/46 | + | 1600 | 1600 | BQ | Negative | BQ | BQ | Peni G + genta/cefo + ampi oflo | Yes | Recovery |
| 4/F/56 | + | 6400 | 3200 | BQ | Negative | BQ | BQ† | Peni G + genta/doxy + sxt | Yes | Recovery |
| 5/M/31 | + | 3200 | 3200 | BQ | BQ | BQ | BQ | No antibiotics | No | Death |
| 6/M/44 | + | 6400 | 6400 | BQ | BQ | BQ | BQ | Ampi + genta + amphoB | Yes | Recovery |
| 7/M/48 | + | 3200 | 3200 | BQ | Negative | BQ | BQ | Peni G + genta/cefo + oflo | Yes | Recovery |
| 8/F/26 | + | 3200 | 1600 | BQ | Negative | BQ | BQ | Cefo + genta | Yes | Recovery |
| 9/M/44 | + | 3200 | 3200 | BQ | Negative | BQ | BQ | Ampi + genta | No | LS |
| 10/M/56 | + | 800 | 400 | BQ | BQ | BQ | BQ | No antibiotics | No | LS |
| 11/M/40 | + | 6400 | 3200 | BQ | BQ | BQ | BQ | Ampi + genta | No | Recovery |
| 12/M/20 | + | 3200 | 3200 | BQ | BQ | BQ | BQ | Ampi + amik/oflo + vanco + amik | Yes | Recovery |
| 13/M/20* | + | 1600 | 1600 | BH | Negative | Negative | BH | |||
Proportion of Bartonella endocarditis among IE in Europe and North Africa.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 72, 5; 10.4269/ajtmh.2005.72.503
Proportion of Bartonella endocarditis among IE in Europe and North Africa.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 72, 5; 10.4269/ajtmh.2005.72.503
Proportion of Bartonella endocarditis among IE in Europe and North Africa.
Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 72, 5; 10.4269/ajtmh.2005.72.503
Authors’ addresses: Abir Znazen, Unitè des Rickettsies, Centre National de la Recherche Scientifique, Unitè Mixte de Recherche 6020, Facultè de Mèdecine, Universitè de la Mèditerranèe. 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France et Laboratoire de Microbiologie, Centro Hospitalo-Universitaire Habib Bourguiba Sfax, Route El Ain Km 05, 3028 Sfax, Tunisia. Jean-Marc Rolain and Didier Raoult, Unitè des Rickettsies, Centre National de la Recherche Scientifique, Unitè Mixte de Recherche 6020, Facultè de Mèdecine, Universitè de la Mèditerranèe, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France, Telephone: 33-4-91-32-43-75, Fax: 33-4-38-77-72, E-mail: Didier.Raoult@medecine.univ-mrs.fr. Nader Hammami and Adnane Hammami, Laboratoire de Microbiologie, Centro Hospitalo-Universitaire Habib Bourguiba Sfax, Route El Ain Km 05, 3028 Sfax, Tunisia. Samir Kammoun, Service de Cardiologie, Centro Hospitalo-Universitaire Hedi Chaker Sfax, Route El Ain Km 05, 3028 Sfax, Tunisia.
Acknowledgments: The authors thank Hachicha Sajiaa for his assistance to perform this work and E. Mathai for reviewing the manuscript. Among potential conflicts of interest, D. Raoult has deposed a patent on serological diagnosis of infective endocarditis, mainly based on the article: Rolain JM, Lecam C, Raoult D. Simplified serological diagnosis of endocarditis due to Coxiella burnetii and Bartonella.
Clin Diagn Lab Immunol 2003;10(6):1147–1148.
REFERENCES
- 1↑
Raoult D, Roux V, 1999. The body louse as a vector of reemerging human diseases. Clin Infect Dis 29 :888–911.
- 3↑
Spach DH, Callis KP, Paauw DS, Houze YB, Schoenknecht FD, Welch DF, Rosen H, Brenner DJ, 1993. Endocarditis caused by Rochalimaea quintana in a patient infected with human immunodeficiency virus. J Clin Microbiol 31 :692–694.
- 4↑
Koehler JE, Quinn FD, Berger TG, Leboit PE, Tappero JW, 1992. Isolation of Rochalimaea species from cutaneous and osseous lesions of bacillary angiomatosis. N Engl J Med 327 :1625–1631.
- 5↑
Drancourt M, Mainardi JL, Brouqui P, Vandenesch F, Carta A, Lehnert F, Etienne J, Goldstein F, Acar J, Raoult D, 1995. Bartonella (Rochalimaea) quintana endocarditis in three homeless men. N Engl J Med 332 :419–423.
- 6↑
Spach DH, Kanter AS, Daniels NA, Nowowiejski DJ, Larson AM, Schmidt RA, Swaminathan B, Brenner DJ, 1995. Bartonella (Rochalimaea) species as a cause of apparent “culture-negative” endocarditis. Clin Infect Dis 20 :1044–1047.
- 7↑
Raoult D, Fournier PE, Drancourt M, Marrie TJ, Etienne J, Cosserat J, Cacoub P, Poinsignon Y, Leclercq P, Sefton AM, 1996. Diagnosis of 22 new cases of Bartonella endocarditis. Ann Intern Med 125 :646–652.
- 8↑
Fournier PE, Lelievre H, Eykyn SJ, Mainardi JL, Marrie TJ, Bruneel F, Roure C, Nash J, Clave D, James E, Benoit-Lemercier C, Deforges L, Tissot-Dupont H, Raoult D, 2001. Epidemiologic and clinical characteristics of Bartonella quintana and Bartonella henselae endocarditis: a study of 48 patients. Medicine (Baltimore) 80 :245–251.
- 9↑
Fournier PE, Mainardi JL, Raoult D, 2002. Value of Microimmunofluorescence for Diagnosis and Follow-up of Bartonella Endocarditis. Clin Diagn Lab Immunol 9 :795–801.
- 10↑
Rolain JM, Lecam C, Raoult D, 2003. Simplified serological diagnosis of endocarditis due to Coxiella burnetii and Bartonella. Clin Diag Lab Immunol 10 :1147–1148.
- 11↑
La Scola B, Raoult D, 1996. Serological cross reactions between Bartonella quintana, Bartonella henselae, and Coxiella burnetii. J Clin Microbiol 34 :2270–2274.
- 12↑
Houpikian P, Raoult D, 2003. Diagnostic methods. Current best practices and guidelines for identification of difficult-to-culture pathogens in infective endocarditis. Cardiol Clin 21 :207–217.
- 13↑
Maurin M, Eb F, Etienne J, Raoult D, 1997. Serological cross-reactions between Bartonella and Chlamydia species: Implications for diagnosis. J Clin Microbiol 35 :2283–2287.
- 14↑
Mann P, Nye F, Williams G, Walker A, Amadi A, 2003. From trench fever to endocarditis. Postgrad Med J 79 :655–656.
- 15↑
Gdoura R, Pereyre S, Frikha I, Hammami N, Clerc M, Sahnoun Y, Bebear C, Daoud M, de Barbeyrac B, Hammami A, 2002. Culture-negative endocarditis due to Chlamydia pneumoniae. J Clin Microbiol 40 :718–720.
- 16↑
Li JS, Sexton DJ, Mick N, Nettles R, Fowler VGJ, Ryan T, Bashore T, Corey GR, 2000. Proposed modifications to the duke criteria for the diagnosis of infective endocarditis. Clin Infect Dis 30 :633–638.
- 17↑
Houpikian P, Raoult D, 2003. Western immunoblotting for Bartonella endocarditis. Clin Diagn Lab Immunol 10 :95–102.
- 18↑
Zeaiter Z, Fournier PE, Greub G, Raoult D, 2003. Diagnosis of Bartonella endocarditis by a real-time nested PCR assay using serum. J Clin Microbiol 41 :919–925.
- 19↑
Pitulle C, Strehse C, Brown JW, Breitschwerdt EB, 2002. Investigation of the phylogenetic relationships within the genus Bartonella based on comparative sequence analysis of the rnpB gene, 16S rDNA and 23S rDNA. Int J Syst Evol Microbiol 52 :2075–2080.
- 20↑
Park SY, Kelminson KL, Lee AK, Zhang P, Warner RE, Rehkopf DH, Calderwood SB, Koehler JE, 2001. Identification, characterization, and functional analysis of a gene encoding the ferric uptake regulation protein in Bartonella species. J Bacteriol 183 :5751–5755.
- 21↑
Tissot-Dupont H, Thirion X, Raoult D, 1994. Q fever serology: cutoff determination for microimmunofluorescence. Clin Diag Lab Immunol 1 :189–196.
- 22↑
Campbell LA, Kuo CC, Grayston JT, 1998. Chlamydia pneumoniae and cardiovascular disease. Emerg Infect Dis 4 :571–579.
- 23↑
Jacomo V, Kelly PJ, Raoult D, 2002. Natural History of Bartonella Infections (an Exception to Koch’s Postulate). Clin Diagn Lab Immunol 9 :8–18.
- 24↑
Rolain JM, Franc M, Davoust B, Raoult D, 2003. Molecular detection of Bartonella quintana, B. koehlerae, B. henselae, B. clarridgeiae, Rickettsia felis and Wolbachia pipientis in cat fleas, France. Emerg Infect Dis 9 :338–342.
- 25↑
Fichet-Calvet E, Jomaa I, Ben Ismail R, Ashford RW, 2000. Patterns of infection of haemoparasites in the fat sand rat, Psammomys obesus, in Tunisia, and effect on the host. Ann Trop Med Parasitol 94 :55–68.
- 26↑
Raoult D, Fournier PE, Vandenesch F, Mainardi JL, Eykyn SJ, Nash J, James E, Benoit-Lemercier C, Marrie TJ, 2003. Outcome and treatment of Bartonella endocarditis. Arch Intern Med 163 :226–230.
- 27↑
Rolain JM, Brouqui P, Koehler JE, Maguina C, Dolan MJ, Raoult D, 2004. Recommendations for treatment of human infections caused by Bartonella species. Antimicrob Agents Chemother 48 :1921–1933.
- 28↑
Millar B, Moore J, Mallon P, Xu J, Crowe M, Mcclurg R, Raoult D, Earle J, Hone R, Murphy P, 2001. Molecular diagnosis of infective endocarditis–a new Duke’s criterion. Scand J Infect Dis 33 :673–680.
- 29↑
Tariq M, Alam M, Munir G, Khan MA, Smego RA Jr, 2004. Infective endocarditis: a five-year experience at a tertiary care hospital in Pakistan. Int J Infect Dis 8 :163–170.
- 30↑
Brouqui P, Raoult D, 2001. Endocarditis due to rare and fastidious bacteria. Clin Microbiol Rev 14 :177–207.
- 31↑
Simon-Vermot I, Altwegg M, Zimmerli W, Flückiger U, 1999. Duke criteria-negative endocarditis caused by Bartonella quintana. Infection 27 :283–285.
- 32
Werner M, Fournier PE, Andersson R, Hogevik H, Raoult D, 2003. Bartonella and Coxiella antibodies in 334 prospectively studied episodes of infective endocarditis in Sweden. Scand J Infect Dis 35 :724–727.
- 33↑
Lamas CC, Eykyn SJ, 2003. Blood culture negative endocarditis: analysis of 63 cases presenting over 25 years. Heart 89 :258–262.