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 25, Number 8—August 2019
Research Letter

Erwinia billingiae as Unusual Cause of Septic Arthritis, France, 2017

Isabelle Bonnet, Baptiste Bozzi, Eric Fourniols, Stéphane Mitrovic, Olivia Soulier-Escrihuela, Florence Brossier1, Wladimir Sougakoff, Jérôme Robert, Stéphane Jauréguiberry, Alexandra AubryComments to Author , on behalf of the Pitié-Salpêtrière Infection Ostéo-articulaire group
Author affiliations: Assistance Publique–Hôpitaux de Paris, Hôpitaux Universitaires Pitié Salpêtrière–Charles Foix, Paris, France (I. Bonnet, B. Bozzi, E. Fourniols, S. Mitrovic, O. Soulier-Escrihuela, F. Brossier, W. Sougakoff, J. Robert, S. Jauréguiberry, A. Aubry); Sorbonne Université, Cimi-Paris, U1135, Paris (I. Bonnet, F. Brossier, W. Sougakoff, J. Robert, A. Aubry)

Cite This Article


In 2017 in France, we treated a patient with knee septic arthritis caused by Erwinia billingiae after trauma involving a palm tree. This rare pathogen could only be identified through 16S rRNA gene sequencing. For bacterial infections after injuries with plants, 16S rRNA gene sequencing might be required for species identification.

The prevalence of acute septic arthritis in Western Europe is ≈4–10 cases/100,000 inhabitants (1). We report a case of posttraumatic knee septic arthritis in an immunocompetent patient in France that was caused by Erwinia billingiae, a gram-negative environmental bacterium of the family Enterobacteriaceae. We also review the characteristics of Erwinia species and infections.

On April 9, 2017, a 65-year-old man with an unremarkable medical history was admitted to an emergency unit in Nice, southern France, for painful right knee swelling that occurred a few hours after a Phoenix palm tree needle pierced the area. The foreign body was partly removed, and the wound was sutured. The patient was discharged without any knee pain and given a prescription for amoxicillin/clavulanic acid (1 g 3×/d for 6 d).

On April 22, the patient was admitted to the emergency unit of our hospital in Paris because of sudden right knee pain and fever. Synovial fluid collected by knee puncture the day of his admission to the orthopedic unit (April 23) contained 118 × 109 leukocytes/L, consisting of 64% polymorphonuclear cells, 33% lymphocytes, and 3% other leukocytes; no microorganism could be identified after Gram staining and cultures. A second knee puncture was performed 3 days after admission, and gram-negative rods grew within 2 days solely within the anaerobic blood culture vial (BacT/ALERT SN; bioMérieux, Subcultures of the blood culture vial were positive after 24 hours of incubation at 37°C on blood agar (Trypticase Soy agar + 5% horse blood and Mueller Hinton 2 agar + 5% sheep blood; bioMérieux) and Drigalski agar (BD, under aerobic conditions and chocolate agar (BD) under microaerobic conditions.

Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (Bruker Daltonik, was performed on colonies and failed to correctly identify the species. Therefore, we performed species identification by 16S rRNA amplification and sequencing with primers RNA-S (16S, 5′-AGAGTTTGATCCTGGYTCAG-3′) and RNA-AS (16AS, 5′-CTTTACGCCCARTAAWTCCG-3′) at a hybridization temperature of 52°C. We amplified a 521-bp sequence that matched the E. billingiae genome of 2 isolates with 99.4% similarity (GenBank accession nos. JQ929658 and JN175337). Other closely related species displayed lower similarities: Pantoea rwandensis (99.0%), Erwinia persicina (98.9%), Pantoea coffeiphila (98.7%), Erwinia tasmaniensis (98.5%), and Erwinia aphidicola (98.3%). Following guidelines of the Antibiogram Committee of the French Society for Microbiology (, we tested the E. billingiae isolate with the antimicrobial drugs recommended for Enterobacteriaceae; the isolate was susceptible to all these drugs, including ampicillin.

Because of the lack of clinical improvement, the joint was washed on day 6 after admission. After this intervention, an empiric antimicrobial drug treatment was started with amoxicillin/clavulanic acid (2 g 3×/d intravenously). Once results of drug susceptibility testing became available (i.e., 10 days after admission), his treatment was switched to cefotaxime (2 g 3×/d intravenously) and ciprofloxacin (500 mg 2×/d orally for 8 d), followed by ciprofloxacin (500 mg 2×/d alone for 38 additional days). Total duration of treatment was 45 days. The clinical evolution of this patient was favorable; he fully recovered and had no relapses up to 1 year after treatment completion.

In the past, some members of the Erwinia genus were reassigned to the genera Enterobacter or Pantoea. Erwinia spp. are ubiquitous in the environment, especially in water ecosystems and soils. Plant-associated Erwinia species comprise epiphytic nonpathogenic (i.e., E. billingiae and E. tasmaniensis) and pathogenic (i.e., E. amylovora and E. pyrifoliae) species. The MALDI-TOF mass spectrometry system failed to identify the bacterium, even though E. billingiae is contained in the database for either method used (direct deposit or on-plate formic acid treatment). Future expansion of the database with more spectra will likely improve the performance of the MALDI-TOF mass spectrometry system for E. billingiae identification. Indeed, the database contains fewer spectra of E. billingiae (n = 4) than those of frequently encountered species in medical microbiological laboratories, such as Escherichia coli (n = 14) and Staphylococcus aureus (n = 10).

To further investigate Erwinia infections in humans, we reviewed reports available in PubMed published during 1967–2017 written in English by using the keywords “Erwinia” and “infection” (Table). Among the 17 cases reported, the sites of infection were diverse, and most (53%, 9/17) cases occurred after a direct inoculation during an injury with a plant (Table). We found no reports of osteoarticular infections with Erwinia; the only other E. billingiae case reported was a dermohypodermitis (Table). In that case, as in the case we report here, an injury with a plant was reported.

This case report illustrates the importance of the methods used for bacterial identification to correctly diagnose such infections. Biochemical methods (28) and MALDI-TOF mass spectrometry (as done in our investigation) could result in misidentification. This report highlights the usefulness of analyzing MALDI-TOF mass spectrometry scores before assigning a species identity and sequencing the 16S RNA gene for bacteria not identifiable by conventional methods.

Dr. Bonnet is a clinical microbiologist in the Bacteriology Laboratory, Pitié Salpêtrière–Charles Foix University Hospital, in Paris, France. She is also part of research team 2 (Bacteriology), Centre d’Immunologie et des Maladies Infectieuses, Cimi-Paris, INSERM, U1135, Sorbonne Université, Paris, France. Her research interests relate to microbiology, especially antimicrobial drug resistance, mycobacteria, and infectious disease.



  1. Mathews  CJ, Weston  VC, Jones  A, Field  M, Coakley  G. Bacterial septic arthritis in adults. Lancet. 2010;375:84655. DOIPubMedGoogle Scholar
  2. Slotnick  IJ, Tulman  L. A human infection caused by an Erwinia species. Am J Med. 1967;43:14750. DOIPubMedGoogle Scholar
  3. Gilardi  GL, Bottone  E, Birnbaum  M. Unusual fermentative, gram-negative bacilli isolated from clinical specimens. I. Characterization of Erwinia strains of the “lathyri-herbicola group”. Appl Microbiol. 1970;20:1515.PubMedGoogle Scholar
  4. von Graevenitz  A. Erwinia infection from environmental sources. JAMA. 1971;216:1485. DOIPubMedGoogle Scholar
  5. Wechsler  A, Bottone  E, Lasser  R, Korenman  G. Brain abscess caused by an Erwinia species. Report of a case and review of the literature. Am J Med. 1971;51:6804. DOIPubMedGoogle Scholar
  6. Mason  GI, Bottone  EJ, Podos  SM. Traumatic endophthalmitis caused by an Erwinia species. Am J Ophthalmol. 1976;82:70913. DOIPubMedGoogle Scholar
  7. Umenai  T, Saitoh  Y, Takano  S, Shoji  E, Tanaka  K, Ishida  N. Significance of Erwinia in the vagina as causative agents of urinary tract infections. Tohoku J Exp Med. 1979;129:1034. DOIPubMedGoogle Scholar
  8. Williams  AJK, Scott  RJD, Lightfoot  NF. Erwinia herbicola as a cause of bacterial endocarditis. J Infect. 1986;12:713. DOIPubMedGoogle Scholar
  9. Shin  SY, Lee  MY, Song  J-H, Ko  KS. New Erwinia-like organism causing cervical lymphadenitis. J Clin Microbiol. 2008;46:31568. DOIPubMedGoogle Scholar
  10. Prod’homme  M, Micol  LA, Weitsch  S, Gassend  JL, Martinet  O, Bellini  C. Cutaneous infection and bactaeremia caused by Erwinia billingiae: a case report. New Microbes New Infect. 2017;19:1346. DOIPubMedGoogle Scholar




Cite This Article

DOI: 10.3201/eid2508.181073

Original Publication Date: June 25, 2019


2Group members are listed at the end of this article.

Table of Contents – Volume 25, Number 8—August 2019

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:

Alexandra Aubry, AP-HP, Hôpitaux Universitaires Pitié Salpêtrière–Charles Foix, Bactériologie-Hygiène, 47-83 Boulevard de l’Hôpital, Paris 75013, France

Send To

10000 character(s) remaining.


Page created: July 16, 2019
Page updated: July 16, 2019
Page reviewed: July 16, 2019
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.