Skip to main content
Journal of Clinical Orthopaedics and Trauma logoLink to Journal of Clinical Orthopaedics and Trauma
. 2013 Nov 11;4(4):194–198. doi: 10.1016/j.jcot.2013.10.001

Escherichia coli – Marauding masquerading microbe

Amy Lee a,, Patrick Coleman b
PMCID: PMC3880952  PMID: 26403882

Abstract

Background

Escherichia coli is a rare cause of monoarticular septic arthritis, but is an even rarer cause of polyarticular septic arthritis.

Case description

We report an unusual case of polyarticular septic arthritis with an atypical presentation caused by E. coli, the source of which was a left pyelonephritis. Our patient developed E coli sepsis resulting in polyarticular septic arthritis (PASA) in the absence of typical risk factors except for pre-existing osteoarthritis. The joints involved were the hip, ankle, sternoclavicular and L5/S1 joints. Of interest, ankle pain was not reported or evident until correlated with nuclear medicine scans. Furthermore, sternoclavicular joint involvement presented as left shoulder pain, resulting in an initial misdiagnosis of left shoulder septic arthritis. The patient was treated with surgical washout and antibiotic therapy. He was subsequently discharged from rehabilitation having returned to his baseline level of mobility. Future consideration will be given to total hip arthroplasty.

Literature review

There are no reported cases of E. coli PASA involving more than three joints in the absence of any recognized risk factors for septic arthritis.

Purpose and clinical relevance

Asymptomatic involvement of joints can occur in polyarticular septic arthritis and should be considered in all cases of monoarticular septic arthritis (MASA). We believe that clinical suspicion is the key to early and comprehensive diagnosis of polyarticular septic arthritis particularly when presenting in an atypical fashion with an atypical pathogen. Strong consideration should be given to performing nuclear imaging in cases of monoarticular septic arthritis where polyarticular involvement cannot be definitively ruled out.

Keywords: Escherichia coli sepsis, Polyarticular septic arthritis, Septic arthritis, Nuclear medicine imaging, Bone-Gallium scan

1. Introduction

Escherichia coli is a rare cause of monoarticular septic arthritis, with most presentations caused by Staphylococcus aureus. In polyarticular septic arthritis, E. coli is an even rarer cause with poor prognosis, especially in the elderly where diagnosis and treatment is often delayed due to pre-morbid medical conditions and the nature of the presentation.

We identified two hundred and thirteen reported cases of PASA, of which twenty-five were from the same center aggregated over a thirteen year period.1 Fifty of these cases were caused by gram-negative bacilli however in only five of these was E. coli isolated.2,3,8 Recognized risk factors for septic arthritis were identified in all five cases and all cases involved three or less joints. Blood cultures were not always positive, and surgical intervention was not always undertaken. The majority of presentations were diagnosed on the clinical purulent appearance of a joint aspirate. This is the first case of E. coli sepsis resulting in PASA involving more than three joints to be reported. It is remarkable due to the absence of any recognized risk factors for septic arthritis. Also, the joints involved (hip, ankle, sternoclavicular joint, L5/S1 joint) are not commonly implicated. Of interest, ankle pain was not reported or evident until correlated with nuclear medicine scans. Furthermore, sternoclavicular joint involvement presented as left shoulder pain, resulting in an initial misdiagnosis of left shoulder septic arthritis.

Nuclear medicine scans are crucial in the diagnosis of foci of infections in these instances. Our case demonstrates that clinical suspicion, confirmed with nuclear imaging, is critical to early diagnosis and successful treatment of atypical presentations of PASA to minimize the potential associated morbidity and mortality.

2. Case report

A 75-year-old man with a history of severe bilateral arthritis of the hips, knees, ankles and lower back, self-referred to the Emergency Department because of increasing pain in his left hip and left shoulder that impeded his mobility. He reported a 72 h history of non-specific malaise, anorexia and diarrhea, which he attributed to a restaurant meal three days prior. He had suffered no trauma or falls to account for his pain. He reported self-limiting intermittent dysuria 2–3 weeks preceding his presentation. No other significant recent or past clinical history was elicited. His regular medications consisted of fish oil, glucosamine and acetaminophen as required. Premorbidly, the patient lived alone and, though he mobilized with two walking sticks, he was fully functionally independent and drove his car daily.

Physical examination revealed a confused man with a flushed facies. His pulse, blood pressure, respiratory rate and temperature were 107 beats per minute, 107/57 mmHg, 20 respirations per minute and 39.2 °C respectively. He had decreased range of movement in both hips but his left hip was clinically irritable. The remainder of the examination was unremarkable. X-rays of the pelvis and hips showed severe bilateral osteoarthritis with no evidence of joint effusion. Initial blood work revealed a C-reactive protein of 300 mg/L; a white cell count of 15.2 × 108/L; a platelet count of 79 × 109/L; and a serum creatinine of 135 μmol/L. Parenteral gentamicin and piperacillin/tazobactam were administered after a mid-stream urine (MSU) and blood for culture were obtained.

In light of ongoing left hip pain, worsening left shoulder pain and persistent fevers, the patient was brought to the operating room one day after admission. An arthroscopic washout of the patient's left shoulder was performed. Minimal fluid was aspirated, with a negative Gram stain and no organisms grown. Open washout of the left hip revealed turbid low viscosity fluid from which E. coli was cultured. This was sensitive to ampicillin, amoxicillin/clavulanic acid, gentamicin and ceftriaxone. The MSU subsequently grew E. coli with identical antibiotic sensitivities.

A CT scan of abdomen and sequential bone-gallium scan (Figs. 1 and 2) were performed. The patient was treated with parenteral piperacillin/tazobactam for 12 days followed by parenteral ceftriaxone for 22 days and subsequently, oral amoxicillin/clavulanic acid for six weeks. He was discharged home on day 47 of his admission, having spent 27 days on the rehabilitation ward. Orthopaedic follow-up for consideration of total hip arthroplasty was arranged.

Fig. 1.

Fig. 1

A whole body bone scan of the patient highlighting increased activity and uptake at the neck, left sternoclavicular joint, shoulders, lumbar spine, hips and ankles.

Fig. 2.

Fig. 2

A fusion CT-Gallium whole body scan of the patient showing increased areas of gallium uptake at the a) neck, b) left sternoclavicular joint, c) left hip, d) first sacral vertebral body, e) fifth lumbar vertebral body and e) left kidney.

3. Discussion

Septic arthritis, whether MASA or PASA, is an uncommon condition with potentially devastating sequalae. S. aureus is by far the most common organism implicated in cases of non-gonococcal septic arthritis.1,9 Though E. coli septic arthritis is rare, it is the most common organism isolated in cases of Gram-negative septic arthritis. There is little published literature clarifying whether this relates to MASA, PASA or combined incidences.4 Extra-intestinal sites of E. coli sepsis are varied but tend to cause more serious disease, even in non-compromised or non-immunosuppressed patients.3 As one would expect, the presence of diabetes, leukemia, cirrhosis, cancer, intravenous substance abuse, renal disease or immunosuppression increases the risk of septic arthritis.9 Aside from premorbid medical conditions such as degenerative joint disease, rheumatoid arthritis and corticosteroid therapy, age greater than 60 years and recent bacteremia are additional risk factors for developing septic arthritis. Poor prognostic signs include positive blood cultures and symptoms lasting more than twenty days.4 Of interest, and for reasons, which are poorly understood, the sternoclavicular joint appears to be peculiarly susceptible to Gram-negative anaerobic bacilli.10

MASA has a particularly poor prognosis in the elderly due to pre-existing co-morbidities, delayed presentations and decreased natural immunity.1 PASA has a poor prognosis regardless of age due to increased bacterial load and the fact it is usually associated with multiple co-morbidities. This accounts for a death rate of approximately 30%.1,2 PASA is an uncommon entity, with only a few single-digit case series in the published literature. The exception is the series of 25 cases reported by Dubost et al accrued over a 12-year period.1 The average age at presentation was 62 years and the mean number of affected joints was four. In PASA, the most commonly affected joints in order of incidence were the knees, elbows, and shoulders whereas MASA usually affected the knees, hips and shoulders. PASA commonly presents with concurrent rheumatic disease, usually rheumatoid arthritis. This combination carries a significantly higher mortality rate in the context of typically chronic advanced joint destruction coupled with corticosteroid therapy.1 Dubost outlined an increasing incidence of PASA by decade since before 1960, a decreasing trend of S. aureus as the causative organism, an association with pre-existing rheumatoid arthritis and a steady increase in cases caused by Gram-negative bacilli observed over the past fifty years.1 In the past two decades, there have been four reported cases of E. coli PASA but, unlike our patient, all four cases had positive risk factors for septic arthritis.2,3,8 Despite these trends over time (Table 1), cases requiring surgical treatment and mortality rates have remained constant.1

Table 1.

Characteristics of PASA by decade of publication.

Characteristic Before 1960 1961–1970 1971–1980 1981–1990 1991 – Present
Number 20 42 58 64 33
Age 47.9 58.6 53.9 50.5 60.6
Number of joints involved 4.25 3.33 3.02 3.97 3.81
Organism
 S. aureus 14 20 21 14 20
 E. coli 5
 Other gram negative bacilli 2 8 19 21 1
 Others 1 8 14 27 4
Underlying disease
 Rheumatoid arthritis 14 21 27 18 13
 Diabetes mellitus 0 5 5 7 2
 Steroid therapy 4 13 17 14 14
 Systemic lupus erythematosus 0 2 6 8 1
Surgical intervention 8/20 16/35 23/17 27/51 15/26
Imaging used 10 documented up to 1976 7
Positive blood cultures 9/14 19/27 28/36 43/52 19/26
Fatal cases 7/20 14/42 15/54 18/63 8/29

Early diagnosis remains paramount in achieving optimal outcomes for presentations of PASA, especially in the elderly. In many instances, blood cultures and even joint aspirate cultures remain negative despite clinical evidence of septic arthritis. In all cases, antibiotic therapy was commenced, regardless of whether a causative organism was found.1–10 The timing and decision to surgically intervene appears variable1–10 with no adherence to any clearly defined guidelines. Although clinical appearance of joint fluid appears to be the most widely accepted standard for diagnosis of septic arthritis,1–10 attempting this is difficult in PASA with multiple joints involved that may not be painful. Nuclear medicine scans are crucial in the diagnosis of foci of infections in these instances. Our case demonstrates that clinical suspicion, confirmed with nuclear imaging, is critical to early diagnosis and successful treatment of atypical presentations of PASA to minimize the potential associated morbidity and mortality.

Conflicts of interest

No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

Ethical approval

Each author certifies that his institution approved the reporting of this case and that all investigations were conducted in conformity with ethical principles of research.

Acknowledgments

We thank Dr Nicholas Coatsworth, infectious diseases registrar at Royal North Shore Hospital, Sydney, Australia, for reviewing the article, and for treatment input and advice.

References

  • 1.Dubost J.J., Fis I., Denis P. Polyarticular septic arthritis. Medicine (Baltimore) 1993;72:296–310. doi: 10.1097/00005792-199309000-00002. [DOI] [PubMed] [Google Scholar]
  • 2.Fernández V., Anniccherico F.J., Alonso J.L., Gutiérrez J., Lasa B. Urinary infection by E. coli complicated with multiple abscesses. An Sist Sanit Navar. 2007;30:293–296. doi: 10.23938/ASSN.0212. [DOI] [PubMed] [Google Scholar]
  • 3.Johnson J.R., Gajewski A., Lesse A.J., Russo T.A. Extraintestinal pathogenic Escherichia coli as a cause of invasive nonurinary infections. J Clin Microbiol. 2003;41:5798–5802. doi: 10.1128/JCM.41.12.5798-5802.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Kelly P.J. Bacterial arthritis in the adult. Orthop Clin North Am. 1975;6:973–981. [PubMed] [Google Scholar]
  • 5.Mcculloch M., Brooks H., Kalantarinia K. Isolated polyarticular septic arthritis: an atypical presentation of meningococcal infection. Am J Med Sci. 2008;335:323–326. doi: 10.1097/MAJ.0b013e318142bb0d. [DOI] [PubMed] [Google Scholar]
  • 6.Muhayidina A.D., Said M.S. Septic arthritis in rheumatology: review of five cases. J Clin Med Res. 2009;1:173–177. doi: 10.4021/jocmr2009.08.1254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Newman E.D., Davis D.E., Harrington T.M. Septic arthritis due to gram negative bacilli: older patients with good outcome. J Rheum. 1988;15:659–662. [PubMed] [Google Scholar]
  • 8.Rodriguez C.L., Ferran N.A., Lyons K., Evans R.O. Bilateral glenohumeral septic arthritis secondary to retroperitoneal abscess. Bull NYU Hosp Jt Dis. 2009;67:378–380. [PubMed] [Google Scholar]
  • 9.Shirtliff M.E., Mader J.T. Acute septic arthritis. Clin Microbiol Rev. 2002;15:527–544. doi: 10.1128/CMR.15.4.527-544.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Ziment I., Davis A., Finegold S.M. Joint infection by anaerobic bacteria: a case report and review of literature. Arthritis Rheum. 1969;12:627–635. doi: 10.1002/art.1780120610. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Clinical Orthopaedics and Trauma are provided here courtesy of Elsevier

RESOURCES