PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of jmmcrLink to Publisher's site
 
JMM Case Rep. 2016 October; 3(5): e005067.
Published online 2016 October 31. doi:  10.1099/jmmcr.0.005067
PMCID: PMC5343147

Sternal wound infection caused by Gordonia bronchialis: identification by MALDI-TOF MS

Abstract

Introduction:

Gordonia spp. infections are uncommon. However, a few clinical cases have been reported in the literature, particularly those involving immunocompromised hosts. Advanced microbiology diagnosis techniques, such as matrix-assisted laser desorption ionization-time of flight MS (MALDI-TOF MS), have been recently introduced in clinical microbiology laboratories in order to improve microbial identification, resulting in better patient management.

Case presentation:

Here, we present a new clinical case of persistent wound infection caused by Gordonia bronchialis in a 64-year-old woman after a mitral valve replacement, using two MALDI-TOF-based systems for identifying this micro-organism.

Conclusion:

Both MALDI-TOF systems were able to identify Gordonia spp.; thus, providing a useful tool that overcomes the current limitations of phenotypic identification associated with this micro-organism. Although the technique validation deserves additional verification, our study provides guidance about MALDI-TOF as a fast and easy method for Gordonia spp. identification.

Keywords: Gordonia, MALDI-TOF, sternal wound infection

Introduction

Gordonia species are aerobic actinomycetes that rarely infect humans, incidences that do occur are most notably in the setting of intravascular catheter-related infections (Johnson et al., 2011) and usually involve immunocompromised patients. There are approximately 35 validly named species in the genus (Conville & Witebsky, 2015), and the most frequently isolated species from human infections are Gordonia aichiensis, G. araii, G. bronchialis, G. otitidis, G. polyisoprenivorans, G. rubripertincta, G. sputi and G. terrae (Guerrero Gómez et al., 2014). G. bronchialis was first identified in soil samples, as well as in sputum samples obtained from patients with pulmonary disease (Tsukamura, 1971).

We present a case of persistent sternal wound infection by G. bronchialis in an immunocompetent woman. Our main purpose is to show our experience and published ones in applying matrix-assisted laser desorption ionization-time of flight MS (MALDI-TOF MS) technology to identify this micro-organism.

Case report

A 64-year-old woman was admitted to hospital because of dehiscence of a sternal wound, after a mitral valve replacement that was performed 2 months earlier due to severe insufficiency. She presented a clinical history of rheumatic mitral stenosis, which was treated with closed mitral valvulotomy 35 years previously, resulting in a mitral insufficiency. Twenty-three years previously she had suffered a bacterial endocarditis due to viridans group streptococci that led to cerebral embolism.

On examination, a white material was found to be exuded from the sternal wound when pressed over the wound margins. A computed tomography scan of the chest showed a dehiscence of the surgical wound, with swelling of soft tissue above the sternum and osteitis of the sternal bone. Apart from a C-reactive protein level of 2.6 mg dl−1 and an albumin level of 3.1 g dl−1, laboratory studies were unremarkable.

Empirical treatment with clindamycin (300 mg/6h i.v.) and ceftazidime (2 g/8h i.v.) was started. The treatment was changed to imipenem (500 mg/6h i.v.) and ciprofloxacin (750 mg/12h p.o.) after a preliminary microbiology laboratory report of growth of an actinomycete with presumed susceptibility to several antimicrobials. Surgical debridement of the wound was performed. This treatment was maintained for 3 weeks, but successive wound cultures continued showing the presence of the actinomycete organism. Because the symptoms did not improve, sternal cerclage was removed and antibiotic therapy was shifted to teicoplanin (400 mg/24h i.v.) plus ciprofloxacin (750 mg/12h p.o.) and rifampin (600 mg/24h p.o.) for 2 weeks, followed by ciprofloxacin plus rifampin for an additional6 weeks, resulting in wound healing.

Culture of wound samples on chocolate and blood agar plates for 72 h at 37 °C in aerobic conditions yielded creamy-white, dry, wrinkled and non-haemolytic colonies. After these 3 days, a colour change was observed in the colonies from white to yellowish. Colony appearance showed synnemata and no aerial hyphae (see Fig. 1). Gram staining yielded Gram-positive short coryneform rods without branching. Modified Ziehl–Neelsen staining confirmed slight acid-fastness. Both conventional Ziehl–Neelsen and auramine stains were negative. The micro-organism was non-spore-forming, and catalase and urease positive. Casein, hypoxanthine, tyrosine and gelatine were not decomposed. Arylsulfatase production was negative within 3 days. Nitrate was not reduced to nitrite and indole was not produced. With the API NH strip (bioMérieux) acid was produced from glucose, fructose and sucrose. 16S rRNA gene sequence analysis using the blast algorithm showed 99.9 % similarity to G. bronchialis strain DSM 43247 (GenBank accession no. NR074529.1).

Fig. 1.
Evolution of colony appearance.

An antimicrobial-susceptibility assay was performed using Etest strips (bioMérieux) on Mueller–Hinton agar with 5 % defibrinated horse blood and 20 mg β-NAD l−1 (MH-F; Oxoid). Readings were taken after 48 h of incubation, and susceptibility categories were defined according to Clinical and Laboratory Standards Institute (CLSI) guidelines for mycobacteria, nocardiae and other actinomycetes (CLSI, 2011). The isolate was resistant to clindamycin (MIC=8 mg l−1), and susceptible to amoxicillin/clavulanic (0.016 mg l−1), ceftriaxone (0.5 mg l−1), imipenem (0.008 mg l−1), ciprofloxacin (0.06 mg l−1), amikacin (0.06 mg l−1), tobramycin (0.12 mg l−1), clarithromycin (2 mg l−1), minocycline (0.25 mg l−1), linezolid (1 mg l−1) and co-trimoxazole (0.03 mg l−1). Although no susceptibility breakpoints have been established for vancomycin and teicoplanin by the CLSI, MIC values were low (0.25 and 1 mg l−1, respectively).

The isolate was analysed by two MALDI-TOF MS-based systems, a Bruker Biotyper (Bruker Daltonics) and a Vitek MS (bioMérieux). Identification of G. bronchialis (99.9 % identity) was obtained with the Vitek MS (saramis 3.0 software) following the procedure recommended by the manufacturer. Briefly, target slides were inoculated into the spots by picking a freshly grown overnight colony and overlaid with 1 µl matrix solution, α-cyano-4-hydroxycinnamic acid. The same result was attained with the Bruker Biotyper (version 3.1 software), using a complete protocol of protein extraction with formic acid and acetonitrile, following the Bruker Biotyper instructions, but the score value (1.72) was lower than the one defined in the manufacturer’s criteria (≥2.00) for acceptance of identification at the species level.

Discussion

Gordonia is a Gram-positive rod with mycolic acids in its structure, which confer partially acid-fast staining. It has been reported to cause a variety of infections after coronary bypass surgery, such as sternal wound infection, bacteraemia, osteomyelitis, pleural infection and recurrent breast abscess (Richet et al., 1991; Sng et al., 2004; Werno et al., 2005; Johnson et al., 2011; Siddiqui et al., 2012; Guerrero Gómez et al., 2014).

Other cases of sternal wound infection by G. bronchialis have been reported previously and are summarized in Table 1. All cases had a history of previous cardiac surgery, including two outbreaks probably related to intraoperative transmission from a nurse.

Table 1.
Summary of reported cases of sternal wound infection due to G. bronchialis

Phenotypic identification of Gordonia spp. is not conclusive, and biochemical profiles can lead to incorrect identification of isolates as non-tuberculosis mycobacteria or other actinomycetes, especially with the genus Rhodococcus, since both are coryneform, aerial-hyphae negative and weak modified acid-fast. Molecular methods, such as 16S rRNA gene sequencing, have significantly improved organism identification, but the results are not generated in a short time and these methods are not available in all laboratories. Recently, MS-based systems emerged as a reliable, fast and cost-effective tool for the identification of bacteria and fungi in routine diagnosis.

Few studies have evaluated the accuracy of MALDI-TOF MS for the identification of Gordonia. Vasquez et al. (2013) reported the first clinical diagnostic identification of G. bronchialis by applying a Bruker Biotyper (score values of 1.515 and 1.892 using different methods of sample preparation). Hsueh et al. (2014) evaluated the performance of the Bruker Biotyper with seven strains of Gordonia species. The system identified two out of three G. bronchialis and a G. sputi isolate, but the identification scores obtained were below 2. Moreover, three Gordonia amicalis were identified as G. rubripertincta, with score values lower than 1.7, because G. amicalis is not included in the Bruker Biotyper 3.1 software. Lam et al. (2015) reported the reliability of the Bruker Biotyper in identifying two G. sputi strains (scores 2.039 and 2.026) and one G. bronchialis (score 1.743). Both research groups used a complete protocol of extraction with formic acid and acetonitrile following the Bruker Biotyper instructions, as in our case.

Other authors, such as Barberis et al. (2014), applied a simplified method of extraction for the Bruker Biotyper system, as previously described (a colony was inoculated on the MALDI-TOF plate and sequentially overlaid with 0.5 µl formic acid and 1 µl matrix; Theel et al., 2012), to a collection of Gram-positive rods of clinical origin, including three Gordonia strains (none were G. bronchialis). Only one G. terrae strain was identified with a score ≥2.0, but following published recommendations (Alatoom et al., 2012; Bizzini et al., 2011), an additional identification of G. terrae was accepted at the species level. These recommendations apply lower cut-off scores for identification of Gram-positive rods (≥1.5 for the genus level and ≥1.7 for the species level). We repeatedly tried to identify our isolate by this simplified extraction method, but no result was obtained in any case. Titécat et al. (2014) failed to identify G. bronchialis with the Bruker Biotyper system with an easier sample preparation method (score lower than 1.5). They applied a small amount of bacterial sample on the target plate, which was overlaid only with 1 µl matrix without adding formic acid. Retesting the isolate with the recommended protein extraction method yielded an identification of Arthrobacter castelii with a score of 1.967.

A PubMed search using Gordonia and Vitek MS as keywords yielded no published studies evaluating the performance of this system for Gordonia identification. However, our clinical isolate was identified by Vitek MS as G. bronchialis with a high level of identity, applying an easier method of sample preparation compared to the one recommended by the manufacturer.

In summary, although more information is required about the reliability of MALDI-TOF MS for the identification of Gordonia species, and other actinomycetes in general, it would be expected that more cases of infections by these micro-organisms could be diagnosed with these new identification approaches. In this aim, a high degree of suspicion for the presence of these micro-organisms in the sample, after direct staining, is important in order to prolong the incubation of cultures.

Notes

Abbreviations:

CLSI
Clinical and Laboratory Standards Institute
MALDI-TOF
matrix-assisted laser desorption ionization-time of flight

References

  • Alatoom A. A., Cazanave C. J., Cunningham S. A., Ihde S. M., Patel R.(2012). Identification of non-diphtheriae Corynebacterium by use of matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 50160–163.10.1128/JCM.05889-11 [PMC free article] [PubMed] [Cross Ref]
  • Barberis C., Almuzara M., Join-Lambert O., Ramírez M. S., Famiglietti A., Vay C.(2014). Comparison of the Bruker MALDI-TOF mass spectrometry system and conventional phenotypic methods for identification of Gram-positive rods. PLos One 9e10630310.1371/journal.pone.0106303 [PMC free article] [PubMed] [Cross Ref]
  • Bizzini A., Jaton K., Romo D., Bille J., Prod'hom G., Greub G.(2011). Matrix-assisted laser desorption ionization-time of flight mass spectrometry as an alternative to 16S rRNA gene sequencing for identification of difficult-to-identify bacterial strains. J Clin Microbiol 49 693–696.10.1128/JCM.01463-10 [PMC free article] [PubMed] [Cross Ref]
  • Chang J. H., Ji M., Hong H. L., Choi S. H., Kim Y. S., Chung C. H., Sung H., Kim M. N.(2014). Sternal osteomyelitis caused by Gordonia bronchialis after open-heart surgery. Infect Chemother 46110–114.10.3947/ic.2014.46.2.110 [PMC free article] [PubMed] [Cross Ref]
  • CLSI (2011). Susceptibility Testing of Mycobateria, Nocardiae, and Other Aerobic Actinomycetes; Approved Standard, 2nd edn, M24-A2. Wayne, PA: Clinical and Laboratory Standards Institute.
  • Conville P. S., Witebsky F. G.(2015). Nocardia, Rhodococcus, Gordonia, Actinomadura, Streptomyces and other aerobic actinomycetes. Manual of Clinical Microbiology, 11th edn, 504–535. Edited by Jorgensen J. H., Pfaller M. A., Carroll K. C., Funke G., Landry M. L., Richter S. S., Warnock D. W., editors. Washington, DC: American Society for Microbiology.
  • Guerrero Gómez, C., Casañ C., Antequera P., Candel C., Blázquez R.(2014). Catheter-related bloodstream infection caused by Gordonia terrae in a bone-marrow transplant patient: case report and review of the literature. JMM Case Rep 1.
  • Hsueh P. R., Lee T. F., Du S. H., Teng S. H., Liao C. H., Sheng W. H., Teng L. J.(2014). Bruker biotyper matrix-assisted laser desorption ionization-time of flight mass spectrometry system for identification of Nocardia, Rhodococcus, Kocuria, Gordonia, Tsukamurella, and Listeria species. J Clin Microbiol 522371–2379.10.1128/JCM.00456-14 [PMC free article] [PubMed] [Cross Ref]
  • Johnson J. A., Onderdonk A. B., Cosimi L. A., Yawetz S., Lasker B. A., Bolcen S. J., Brown J. M., Marty F. M.(2011). Gordonia bronchialis bacteremia and pleural infection: case report and review of the literature. J Clin Microbiol 491662–1666.10.1128/JCM.02121-10 [PMC free article] [PubMed] [Cross Ref]
  • Lam J. Y., Wu A. K., Leung W. S., Cheung I., Tsang C. C., Chen J. H., Chan J. F., Tse C. W., Lee R. A., et al. (2015). Gordonia species as emerging causes of continuous-ambulatory-peritoneal-dialysis-related peritonitis identified by 16S rRNA and secA1 gene sequencing and matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). J Clin Microbiol 53671–676.10.1128/JCM.02971-14 [PMC free article] [PubMed] [Cross Ref]
  • Nguyen D. B., Gupta N., Abou-Daoud A., Klekamp B. G., Rhone C., Winston T., Hedberg T., Scuteri A., Evans C., et al. (2014). A polymicrobial outbreak of surgical site infections following cardiac surgery at a community hospital in Florida, 2011-2012. Am J Infect Control 42432–435.10.1016/j.ajic.2013.11.021 [PMC free article] [PubMed] [Cross Ref]
  • Richet H. M., Craven P. C., Brown J. M., Lasker B. A., Cox C. D., McNeil M. M., Tice A. D., Jarvis W. R., Tablan O. C.(1991). A cluster of Rhodococcus (Gordona) bronchialis sternal-wound infections after coronary-artery bypass surgery. N Engl J Med 324104–109.10.1056/NEJM199101103240206 [PubMed] [Cross Ref]
  • Siddiqui N., Toumeh A., Georgescu C.(2012). Tibial osteomyelitis caused by Gordonia bronchialis in an immunocompetent patient. J Clin Microbiol 503119–3121.10.1128/JCM.00563-12 [PMC free article] [PubMed] [Cross Ref]
  • Sng L. H., Koh T. H., Toney S. R., Floyd M., Butler W. R., Tan B. H.(2004). Bacteremia caused by Gordonia bronchialis in a patient with sequestrated lung. J Clin Microbiol 422870–2871.10.1128/JCM.42.6.2870-2871.2004 [PMC free article] [PubMed] [Cross Ref]
  • Theel E. S., Schmitt B. H., Hall L., Cunningham S. A., Walchak R. C., Patel R., Wengenack N. L.(2012). Formic acid-based direct, on-plate testing of yeast and Corynebacterium species by Bruker Biotyper matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 503093–3095.10.1128/JCM.01045-12 [PMC free article] [PubMed] [Cross Ref]
  • Titécat M., Courcol R. J., Loı¨ez C., Wallet F.(2014). Difficulty with Gordonia bronchialis identification by Microflex mass spectrometer in a pacemaker‐induced endocarditis. JMM Case Rep 1.
  • Tsukamura M.(1971). 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 6815–26.10.1099/00221287-68-1-15 [PubMed] [Cross Ref]
  • Vasquez M. A., Marne C., Villuendas M. C., Arazo P.(2013). Osteomielitis esternal subaguda por Gordonia bronchialis tras cirugía cardiaca. Enferm Infecc Microbiol Clin 31559–560.10.1016/j.eimc.2013.02.012 [PubMed] [Cross Ref]
  • Werno A. M., Anderson T. P., Chambers S. T., Laird H. M., Murdoch D. R.(2005). Recurrent breast abscess caused by Gordonia bronchialis in an immunocompetent patient. J Clin Microbiol 433009–3010.10.1128/JCM.43.6.3009-3010.2005 [PMC free article] [PubMed] [Cross Ref]
  • Wright S. N., Gerry J. S., Busowski M. T., Klochko A. Y., McNulty S. G., Brown S. A., Sieger B. E., Ken Michaels P., Wallace M. R.(2012). Gordonia bronchialis sternal wound infection in 3 patients following open heart surgery: intraoperative transmission from a healthcare worker. Infect Control Hosp Epidemiol 331238–1241.10.1086/668441 [PubMed] [Cross Ref]

Articles from JMM Case Reports are provided here courtesy of Microbiology Society