Skip directly to site content Skip directly to page options Skip directly to A-Z link Skip directly to A-Z link Skip directly to A-Z link
Volume 6, Number 4—August 2000

Bacteremia and Endocarditis Caused by a Gordonia Species in a Patient with a Central Venous Catheter

Article Metrics
citations of this article
EID Journal Metrics on Scopus
Olivier Lesens*Comments to Author , Yves Hansmann*, Philippe Riegel†, Rémy Heller†, Mohamed Benaissa-Djellouli*, Martin Martinot*, Hélène Petit‡, and Daniel Christmann*
Author affiliations: *Service des Maladies Infectieuses et Tropicales, Clinique Médicale A, Hôpitaux Universitaires, Strasbourg, France; †Laboratoire de Bactériologie, Faculté de Médecine, ‡Service de Chirurgie Cardio-Vasculaire, Hôpitaux Universitaires, Strasbourg, France

Cite This Article


We report the first case of endocarditis caused by a Gordonia species genetically related to G. sputi but exhibiting some atypical biochemical features in a 31-year-old woman with a central venous catheter. This unusual pathogen may be a new cause of opportunistic infections in patients with severe underlying diseases.

Gordonia spp. are gram-positive coryneform bacteria, recently identified as a cause of systemic infections in three patients (1,2), including two associated with indwelling implantable subcutaneous central venous catheters (2). We report Gordonia spp.'s capacity to infect a patient's undamaged cardiac valve through an implantable subcutaneous central catheter.

Case Report

The patient was a 31-year-old woman, who had undergone splenectomy, at age 9 years for severe double heterozygous hemoglobinopathy (beta-thalassemia and hemoglobin E disease). Multiple blood transfusions at that time were complicated by hepatitis C with cirrhosis and secondary hemochromatosis, treated at home with twice weekly deferoxamine by subcutaneous central venous catheter. Hemochromatosis was complicated by diabetes, adrenal insufficiency, and peripheral neuropathy. In September 1997, the patient became ill with Staphylococcus aureus bacteremia associated with localized renal and cutaneous abscesses; transesophageal echocardiography showed neither valvular vegetation nor a valvular defect suggestive of endocarditis. The bacteremia was successfully treated with intravenous fosfomycin, cefotaxime and netilmicin, followed by ciprofloxacin and oxacillin for 8 weeks. The subcutaneous central venous port was also changed after 6 weeks of treatment.

The patient remained afebrile until December 1998, when fever and chills developed. Physical examination 1 week after onset of symptoms revealed a temperature of 39°C and a new mitral systolic murmur. The site of the subcutaneous central venous port showed no signs of infection. Results of clinical laboratory tests showed a leukocyte count of 31.4x109/L, an erythrocyte sedimentation rate of 67 mm/h, and a C-reactive protein plasma level < 5 mg/L. Transthoracic echocardiography revealed a mitral valvular vegetation (10x5 mm). Two blood cultures (one obtained from the central venous catheter) were performed at admission. One more peripheral blood sample was drawn for culture 12 hours later. All three cultures were positive for a gram-positive coryneform bacterium showing no extensive branching. This organism produced dry, raised, salmon-to-orange colonies without aerial hyphae. As the organism was weakly acid fast, according to the Kinyoun acid-fast stain modified for aerobic actinomycetes, we presumptively identified this organism as Rhodococcus sp. sensu lato. According to disk diffusion, the organism was susceptible to penicillin G, amoxicillin, cefotaxime, ceftriaxone, imipenem, gentamicin, netilmicin, ciprofloxacin, vancomycin, and erythromycin but was resistant to fosfomycin, ceftazidime, trimethoprim-sulfamethoxazole, and streptogramin. By E-test, the MICs of penicillin G, amoxicillin, and ceftriaxone were 0.047, 0.064 and 0.25 µg/mL, respectively. The patient was successfully treated with intravenous amoxicillin (3 g, 4/day) and intravenous netilmicin (150 mg, 2/day) for 1 week, then with intravenous amoxicillin alone for 3 weeks. This treatment was followed by home treatment with 2 g perfusion of ceftriaxone for 2 weeks. The central venous catheter was left in situ. At 1 year follow-up, the patient was not infected with Gordonia sp., and echocardiographic findings were consistent with mitral valve insufficiency without oscillating intracardiac mass on valve.

To accurately identify the organism, we analyzed the p-bromophenacyl esters of mycolic acids of the isolate by using high-performance liquid chromatography, obtaining a pattern consistent with that of Gordonia sp. The number of peaks and retention times were similar to those exhibited by the G. sputi type strain ATCC 29627. Biochemical test results were positive for the hydrolysis of urease and esculin but negative for the hydrolysis of xanthine, adenine, tyrosine, and hypoxanthine. When inoculated with aerobic, low-peptone carbohydrate slants, the strain produced acid from trehalose but not from L-rhamnose and D-mannitol. These biochemical characteristics fit those of Gordonia species, especially G. bronchialis (3). To determine partial 16S rRNA gene sequence, the two eubacterial universal primers P8-27 (5'-AGA GTT TGA TCC TGG CTC AG-3') and P1392-1372 (5'-AAG GCC CGG GAA CGT ATT CAC-3') were used for the amplification, then a direct sequencing method with an internal primer P535-514 (5'-GTA TTA CCG CGG CTG CTG GGC AC-3') 5' labeled with fluorescein isothiocyanate was performed (4). The sequence obtained coincides with the 450 5' base pairs of the 16S rRNA gene and matches totally that of G. sputi present in the GenBank-EMBL database. This sequence did not match other bacterial sequences, including those of other Gordonia species. The sequence of the isolate differed from the sequence of the G. aichiensis type strain by two nucleotides. DNA-DNA similarity experiments, according to the stringent nuclease S1 method, showed 55% DNA relatedness with G. sputi type strain and less than 28% with the type strain of other species including G. aichiensis. (The genetic definition of a species is more than 70% DNA similarity.) Thus, this isolate does not fit in any of the recognized Gordonia species, although it is taxonomically close to G. sputi.


The recent differentiation of Gordonia spp. as a distinct genus is the outcome of a taxonomic history complicated by several reclassifications (Table 1). Twelve organisms now belong to the genus Gordonia, including three species discovered in 1999 (5-9). To review the spectrum of clinical diseases in humans caused by Gordonia spp., we performed a PubMed search for 1966 to 1999, using all the designations included in Table 1. Only G. bronchialis (10), G. rubropertincta (11), G. sputi (1,12), and G. terrae (2,13,14) have been shown to be pathogenic in humans (Table 2). They are derived from soil and may also be isolated in the sputa from patients with chest disorders (15). In an outbreak of sternal-wound infections (10), G. bronchialis was isolated from the hand, scalp, and vagina of a nurse as well as from her dog.

Systemic Gordonia spp. infections have been described in three patients. Buchman et al. (2) reported two cases of Gordonia spp. bloodstream infection associated with a Hickman catheter in two immunocompetent patients receiving long-term parenteral nutrition. Both strains were susceptible to vancomycin and gentamicin. In one case, the Gordonia sp. isolated from blood cultures and from a broth culture of the catheter tip was not clearly identified but was close to G. rubropertincta. The patient received intravenous vancomycin for 5 days and intravenous gentamicin for 19 days; the catheter was removed after 2 days. In the second case, the microorganism isolated from blood cultures was identified as G. terrae, and the patient was treated with intravenous vancomycin for 6 weeks with the catheter left in situ.

The only case of bacteremia caused by G. sputi was described in a 34-year-old immunocompromised patient (1) with metastatic melanoma treated with intravenous interleukin-2. The bacterium was thought to have reached the bloodstream through extensive desquamative skin rashes caused by the interleukin-2 treatment or by contamination of the catheter, although cultures from these specimens were negative. Because G. sputi was susceptible to ß-lactams, vancomycin, aminoglycosides, doxycycline, and rifampicin, the patient was first treated with amoxicillin-clavulanate (1,000 and 125 mg, respectively, every 3 hours). After 1 week of treatment, a second set of blood cultures yielded the same organism, and the treatment was changed to a combination of amikacin and piperacillin, which was successful (1). As with other Gordonia species, G. sputi may be isolated in the sputum of patients with pulmonary disease (15). Mediastinitis caused by G. sputi after coronary artery bypass grafting was recently described in an immunocompetent patient (12). The patient was treated with cefmetazole sodium (2 g per day for 3 weeks) and piperacillin sodium (2 g per day for 2 weeks) after surgical soft tissue debridement.

In our case, Gordonia sp. systemic infection associated with an implantable subcutaneous central venous catheter was complicated by endocarditis. The diagnosis was assessed as definitive on the basis of Duke criteria (one major and three minor, including echographic evidence of an oscillating intracardiac mass with a new regurgitant murmur, two persistently positive blood cultures yielding Gordonia sp., fever >38°C, and intravenous deferoxamin use). Our patient had neither neutropenia nor immunosuppressive medications, but underlying diseases may have impaired the immune system and facilitated infection. The bacteremia may be caused by manipulations of the implantable subcutaneous central venous catheter during routine home use. However, even if no other source for the bacteremia had been evident, no inflammation was observed at the port site, and semiquantitative culture was not available. An environmental investigation was not performed.

We conclude that Gordonia spp. may cause opportunistic infections, in particular bacteremia and endocarditis, in patients with severe underlying diseases and indwelling central catheters.

Dr. Lesens is a specialist in infectious diseases in the Department of Infectious Diseases, Strasbourg Hospital, France. He is pursuing a PhD in epidemiology with scientific interests centered on the epidemiology of Staphylococcus aureus bacteremia.



The authors thank JP Euzéby (author of the folder "list of bacterial names with standing in nomenclature," for providing helpful comments about Gordonia' s classification.



  1. Riegel  P, Ruimy  R, de Briel  D, Eichler  F, Bergerat  JP, Christen  R, Bacteremia due to Gordona sputi in an immunocompromised patient. J Clin Microbiol. 1996;34:20457.PubMedGoogle Scholar
  2. Buchman  AL, McNeil  MM, Brown  JM, Lasker  BA, Ament  ME. Central venous catheter sepsis caused by unusual Gordona (Rhodococcus) species: identification with a digoxigenin-labeled rDNA probe. Clin Infect Dis. 1992;15:6947.PubMedGoogle Scholar
  3. McNeil  MM, Brown  JM. The medically important aerobic actinomycetes: epidemiology and microbiology. Clin Microbiol Rev. 1994;7:357417.PubMedGoogle Scholar
  4. Heller  R, Jaulhac  B, Charles  P, de Briel  D, Vincent  V, Bohner  C, Identification of Mycobacterium shimoidei in a tuberculosis-like cavity by 16S ribosomal DNA direct sequencing. Eur J Clin Microbiol Infect Dis. 1996;15:1727. DOIPubMedGoogle Scholar
  5. Stackebrandt  E, Rainey  FA, Ward-Rainey  NL. Proposal for a new hierarchic classification system, Actinobacteria classis nov. Int J Syst Bacteriol. 1997;47:47991. DOIGoogle Scholar
  6. Takeuchi  M, Hatano  K. Gordonia rhizosphera sp. nov. isolated from the mangrove rhizosphere. Int J Syst Bacteriol. 1998;48:90712. DOIPubMedGoogle Scholar
  7. Kummer  C, Schumann  P, Stackebrandt  E. Gordonia alkanivorans sp. nov., isolated from tar-contaminated soil. Int J Syst Bacteriol. 1999;49:151322. DOIPubMedGoogle Scholar
  8. Kim  SB, Brown  R, Olfield  C, Gilbert  SC, Goodfellow  M. Gordonia desulfuricans sp. nov. a benzothiophenedesulphurizing actinomycete. Int J Syst Bacteriol. 1999;49:184551. DOIPubMedGoogle Scholar
  9. Linos  A, Steinbuchel  A, Sproer  C, Kroppenstedt  RM. Gordonia polyisoprenivorans sp. nov., a rubber-degrading actinomycete isolated from an automobile tyre. Int J Syst Bacteriol. 1999;49:178591. DOIPubMedGoogle Scholar
  10. Richet  HM, Craven  PC, Brown  JM, Lasker  BA, Cox  CD, McNeil  MM, A cluster of Rhodococcus (Gordona) bronchialis sternal-wound infections after coronary-artery bypass surgery. N Engl J Med. 1991;324:1049.PubMedGoogle Scholar
  11. Hart  DHL, Peel  MM, Andrew  JH, Burdon  JGW. Lung infection caused by Rhodococcus. Aust N Z J Med. 1988;18:7901.PubMedGoogle Scholar
  12. Kuwabara  M, Onitsuka  T, Nakamura  K, Shimada  M, Ohtaki  S, Mikami  Y. Mediastinitis due to Gordonia sputi after CABG. J Cardiovasc Surg (Torino). 1999;40:6757.PubMedGoogle Scholar
  13. Drancourt  M, McNeil  MM, Brown  JM, Lasker  BA, Maurin  M, Choux  M, Brain abscess due to Gordona terrae in an immunocompromised child: case report and review of infections caused by G. terrae. Clin Infect Dis. 1994;19:25862.PubMedGoogle Scholar
  14. Drancourt  M, Pelletier  J, Cherif  AA, Raoult  D. Gordona terrae central nervous system infection in an immunocompetent patient. J Clin Microbiol. 1997;35:37982.PubMedGoogle Scholar
  15. Tsukamura  M. Proposal of a new genus, Gordona, for slightly acid-fast organisms occurring in sputa of patients with pulmonary disease and in soil. J Gen Microbiol. 1973;25:66581.
  16. Martin  T, Hogan  DJ, Murphy  F, Natyshak  I, Ewan  EP. Rhodococcus infection of the skin with lymphadenitis in a nonimmunocompromised girl. J Am Acad Dermatol. 1991;24:32832. DOIPubMedGoogle Scholar




Cite This Article

DOI: 10.3201/eid0604.000410

Table of Contents – Volume 6, Number 4—August 2000

EID Search Options
presentation_01 Advanced Article Search – Search articles by author and/or keyword.
presentation_01 Articles by Country Search – Search articles by the topic country.
presentation_01 Article Type Search – Search articles by article type and issue.



Please use the form below to submit correspondence to the authors or contact them at the following address:

Olivier Lesens, Service des Maladies Infectieuses et Tropicales, Clinique Médicale A, Fédération des Services de Médecine Interne, Hôpital Civil, 1 place de l'Hôpital - B.P. 426- 67091 Strasbourg Cedex, France; fax: 03 88 11 64 64

Send To

10000 character(s) remaining.


Page created: December 16, 2010
Page updated: December 16, 2010
Page reviewed: December 16, 2010
The conclusions, findings, and opinions expressed by authors contributing to this journal do not necessarily reflect the official position of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions. Use of trade names is for identification only and does not imply endorsement by any of the groups named above.