Skip to main content
Journal of Clinical Microbiology logoLink to Journal of Clinical Microbiology
. 2007 May 2;45(7):2324–2326. doi: 10.1128/JCM.00400-07

Fatal Bioprosthetic Aortic Valve Endocarditis Due to Cardiobacterium valvarum

Walter Geißdörfer 1,*, René Tandler 2, Christian Schlundt 3, Michael Weyand 2, Werner G Daniel 3, Christoph Schoerner 1
PMCID: PMC1933016  PMID: 17475754

Abstract

Cardiobacterium valvarum was isolated from the blood of a 71-year-old man with fatal aortic valve endocarditis. The API NH system was used for phenotypic characterization of the C. valvarum strain. This is the first case of infective endocarditis caused by C. valvarum in Germany and the first case worldwide affecting a prosthetic valve and lacking an obvious dental focus.

CASE REPORT

A 71-year-old patient was admitted to our cardiac department due to complicated severe endocarditis. His medical history revealed surgery 2 years previously with insertion of bypasses and replacement of a stenotic aortic valve with a bioprosthesis. No infective endocarditis was diagnosed at that time. Two months prior to admission, he had suffered backache resulting from a disc prolapse, which was conservatively treated with repeated local injections of nonsteroidal anti-inflammatory drugs. In the last weeks before admission, he developed a general weakness accompanied by fever, chills, edema and petechiae in the legs, and psychomotoric slowing. Upon admission to a nearby hospital (day 1), arterial hypotension was treated with noradrenalin. Transesophageal echocardiography revealed endocarditis with vegetations on the bioprosthetic aortic valve. Blood cultures were taken immediately and showed growth of fastidious gram-negative rods after 4 days of incubation. In response to this finding, empirical antibiotic therapy was adjusted to ceftriaxone, rifampin, and amikacin. When an atrioventricular block, grade III, appeared on day 6, the patient was transferred in stable cardiorespiratory condition to the intensive care unit of our department of cardiology. Laboratory tests showed anemia (hemoglobin, 9.6 g/dl), increased concentration of C-reactive protein (107 mg/liter), leukocytosis (16,900 leukocytes/μl), and a reduced platelet count (74,000 cells/μl). Blood cultures taken at that time were incubated for 3 weeks but remained negative. For a more detailed assessment, transesophageal echocardiography was repeated; it showed a large floating vegetation on a degenerated, insufficient aortic valve and a hypodense region extending to the aortic bulbus suspicious of a large perivalvular abscess cavity. In the cerebellum, two new ischemic areas, indicative of septic embolism, were identified by cranial computerized tomography and magnetic resonance tomography. No indications of further embolisms were found in the spleen, liver, kidneys, or eyes upon sonographic and ophthalmoscopic examinations. Magnetic resonance tomography of the lumbar spine was suspicious of erosive osteochondrosis (L3/4) and spondylodiscitis, presumably resulting from nonsteroidal anti-inflammatory drug injections.

On day 9, the patient was transferred to our center for cardiac surgery for a high-risk prosthetic valve replacement. Intraoperatively, the aortic valve prosthesis showed massive inflammatory destruction and a large abscess near the anulus, including nearly the complete circumference of the aortic bulbus. Although it was impossible to remove all of the inflamed tissue, a mechanic aortic valve prosthesis (St. Jude Medical Regent, 19 mm) was implanted. Assisted by counterpulsation therapy with an intra-aortic balloon catheter, the patient postoperatively showed a fairly stable hemodynamic condition, but he died 1 day after surgery from severe septic shock as a consequence of the severe cardiac deterioration.

In a nearby laboratory, the blood culture isolate (VA1941/06) was tentatively identified as a probable HACEK organism on the basis of gram staining (gram-negative rods) and testing for oxidase (positive), catalase (negative), and indole (positive). Because conventional identification of these organisms can be challenging, and because of the patient's severe clinical condition, the strain was sent to our department of clinical microbiology for fast characterization. 16S rRNA gene sequence analysis revealed 99.8% identity to the sequence of Cardiobacterium valvarum (GenBank accession number AF506987). Routine PCR testing of the explanted aortic valve prosthesis by broad-range (“eubacterial”) PCR (8) revealed the identical 16S rRNA gene sequence, confirming C. valvarum as the etiological agent of endocarditis.

Cultures from the explanted valve specimen were incubated for 13 days but remained negative, indicating the effectiveness of the antibiotic treatment. Strain VA1941/06 grew slightly better on chocolate agar at 37°C with 5% CO2 than on sheep blood agar and formed round, opaque, weakly alpha-hemolytic, glistening colonies of 0.5 mm after 3 days (Fig. 1). Bacterial cells appeared as straight gram-negative rods, located singly or in pairs and of variable length depending on the medium (Fig. 1). The API 20NE (bioMérieux) system was positive for tryptophanase only, and the resulting identification code 2000004 was unsuitable for identification, as previously described by Han et al. (3, 4). With the API NH system, reactions for glucose, fructose, maltose, sucrose, indole, lipase (weak), and alkaline phosphatase were positive, whereas penicillinase, ornithine decarboxylase, urease, beta-galactosidase, proline arylamidase, and gamma-glutamyl transferase were negative. The respective API NH algorithm resulted in identification code 7134, corresponding to Haemophilus parainfluenzae.

FIG. 1.

FIG. 1.

(A) Colony morphology of C. valvarum on chocolate agar after 4 days of growth at 37°C with 5% CO2. (B to D) Gram stains from blood culture (B), agar with sheep blood (C), and chocolate agar (D).

In an agar diffusion test and determination of the MIC with Etest (Biodisk, Solna, Sweden) on nonstandardized chocolate tryptone agar, the C. valvarum strain showed high susceptibility to penicillin (MIC, 0.008 mg/liter), amoxicillin (MIC, 0.016 mg/liter), piperacillin, cefazolin, cefotiam, ceftriaxone (MIC, 0.016 mg/liter), ceftazidime, imipenem (MIC, 0.012 mg/liter), meropenem, gentamicin, tobramycin, amikacin, ciprofloxacin, moxifloxacin (MIC, 0.006 mg/liter), fosfomycin, erythromycin, tetracycline, chloramphenicol, vancomycin, and linezolid.

The genus Cardiobacterium, with its sole species, C. hominis, was established in 1964 for a group of fastidious gram-negative bacteria of the HACEK group (9). C. hominis is found in most (∼70%) individuals as part of the normal flora of the nose and throat. It appears to be the rarest agent of infective endocarditis within the HACEK group, with a total of 61 reported cases (1, 2, 6). The species C. valvarum, as a new member of the HACEK group, was first described in 2004 in Houston, TX, as the agent of endocarditis (4). Two additional cases have been reported, from Washington, DC (5), and Marseilles, France (1). A fourth case, from Chicago, IL, has been published without clinical data (7). Ours is therefore the fifth case of endocarditis caused by C. valvarum worldwide, the second in Europe, and the first in Germany. From our experience of 175 cases of infective endocarditis diagnosed from explanted heart valves, this case is the only one caused by a HACEK organism.

Strain VA1941/06 is available from the Culture Collection of the University of Göteborg, Göteborg, Sweden (CCUG 53031). Its phenotypic and biochemical features resemble those of the reported strains (1, 3-5), including their sensitivity to a broad range of antibiotics. In contrast, however, it grew better on chocolate agar than on sheep blood agar. Indole production, as a key reaction for identification of gram-negative bacteria, was reported to be positive for two (1, 4) and negative in five (3, 5) isolates of C. valvarum. Our strain tested indole positive. This confirms that indole production is variable within C. valvarum and is therefore not appropriate for differentiating C. valvarum from indole-positive C. hominis. Because API 20NE is not suitable for identification of C. valvarum due to its fastidious growth, we used the API NH identification system, which is commonly used for identification of Haemophilus spp. and Neisseria spp. of clinical importance. The outcome of the reactions was well defined but resulted in a misidentification as Haemophilus parainfluenzae, which does not grow on sheep blood agar. Therefore, API NH may be used for determination of key reactions in the case of fastidious gram-negative rods such as C. valvarum, but for final identification, molecular tools, e.g., 16S rRNA gene sequence analysis, are indispensable.

As with the other cases described in the literature, the onset of disease in our case was insidious, with moderate or no fever, and infection resulted in extensive tissue destruction, necessitating valvular replacement. Unlike all other published reports, in which the patients had a congenital bicuspid aortic valve, our patient had a prosthetic aortic valve as a risk factor, and he could not be restored to health as a consequence of the massive destruction of heart tissue. A fatal outcome is also unusual in cases of endocarditis caused by C. hominis, where the prognosis for both native and prosthetic valve endocarditis is usually favorable (6). The oral cavity is the probable reservoir of C. valvarum, because in all three of the published cases the patients experienced oral abscesses or dental treatment without antibiotic prophylaxis prior to endocarditis. In addition, other C. valvarum strains have all been isolated from subgingival pockets and tooth plaque (3). In contrast, our patient had no significant dental treatment prior to the onset of disease, and no further focus was known. However, the condition of his teeth and gums was poor, and they may have been a reservoir for C. valvarum.

In conclusion, this is the first case of endocarditis due to C. valvarum affecting a prosthetic valve, and it shows that physicians should be aware of C. valvarum as a rare cause of endocarditis. In addition, our case supports the value of applying molecular methods in the detection and identification of unusual agents of infective endocarditis.

Nucleotide sequence accession number.

The 16S rRNA gene sequence of strain VA1941/06 has been deposited in the GenBank database under accession number DQ645464.

Acknowledgments

We are grateful to Laura Artz, of the Laboratory Spranger and Partner (Ingolstadt, Germany), for providing strain VA1941/06 and for discussion of the results.

Footnotes

Published ahead of print on 2 May 2007.

REFERENCES

  • 1.Bothelo, E., F. Gouriet, P. E. Fournier, V. Roux, G. Habib, F. Thuny, D. Metras, D. Raoult, and J. P. Casalta. 2006. Endocarditis caused by Cardiobacterium valvarum. J. Clin. Microbiol. 44:657-658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Brouqui, P., and D. Raoult. 2001. Endocarditis due to rare and fastidious bacteria. Clin. Microbiol. Rev. 14:177-207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Han, X. Y., and E. Falsen. 2005. Characterization of oral strains of Cardiobacterium valvarum and emended description of the organism. J. Clin. Microbiol. 43:2370-2374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Han, X. Y., M. C. Meltzer, J. T. Woods, and V. Fainstein. 2004. Endocarditis with ruptured cerebral aneurysm caused by Cardiobacterium valvarum sp. nov. J. Clin. Microbiol. 42:1590-1595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Hoover, S. E., S. H. Fischer, R. Shaffer, B. M. Steinberg, and D. R. Lucey. 2005. Endocarditis due to a novel Cardiobacterium species. Ann. Intern. Med. 142:229-230. [DOI] [PubMed] [Google Scholar]
  • 6.Malani, A. N., D. M. Aronoff, S. F. Bradley, and C. A. Kauffman. 2006. Cardiobacterium hominis endocarditis: two cases and a review of the literature. Eur. J. Clin. Microbiol. Infect. Dis. 25:587-595. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Petti, C. A., C. R. Polage, and P. Schreckenberger. 2005. The role of 16S rRNA gene sequencing in identification of microorganisms misidentified by conventional methods. J. Clin. Microbiol. 43:6123-6125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Relman, D. A., T. M. Schmidt, R. P. MacDermott, and S. Falkow. 1992. Identification of the uncultured bacillus of Whipple's disease. N. Engl. J. Med. 327:293-301. [DOI] [PubMed] [Google Scholar]
  • 9.Slotnick, I. J., and M. Dougherty. 1964. Further characterization of an unclassified group of bacteria causing endocarditis in man: Cardiobacterium hominis gen. et sp. Antonie Leeuwenhoek 30:261-272. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Clinical Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES