Skip to main content
Journal of Clinical Microbiology logoLink to Journal of Clinical Microbiology
. 2013 Jun;51(6):1970–1972. doi: 10.1128/JCM.03377-12

Septic Arthritis Caused by Noncapsulated Haemophilus influenzae

Sandra Le Quellec a, Olivier Gaillot d, Franck Chotel c, Anne-Marie Freydière a, Frédéric Laurent e,f, François Vandenesch a,f,g, Anne Doléans-Jordheim a,b,✉
PMCID: PMC3716039  PMID: 23515545

Abstract

Since the introduction of type b Haemophilus influenzae vaccination, noncapsulated H. influenzae has become responsible for most cases of invasive H. influenzae diseases. In our two cases of septic arthritis, we isolated strains with β-lactamase-positive amoxicillin-clavulanate resistance (BLPACR). Thus, the increasing prevalence of BLPACR should be taken into account when empirical therapy is chosen for septic arthritis.

CASE REPORTS

Patient 1.

A 1-year-old girl was admitted to the Pediatric Hospital in Lyon, France, in November 2011 with fever and inflammatory signs in her left elbow. She was born prematurely at 31 weeks of amenorrhea and had received 3 injections of diphtheria–tetanus–whole-cell pertussis–Haemophilus influenzae type b (DTPw/Hib) vaccine during her first year of life. One week before admission, she presented with a bilateral conjunctivitis associated with a progressive cutaneous rash. The diagnosis of a viral infection was entertained, and she was treated with acetaminophen (international nonproprietary name, paracetamol). A day before admission, she had a swollen, erythematous left elbow with a limited range of motion. Oral amoxicillin treatment was initiated.

Upon admission, the patient was febrile (38.5°C) and asthenic. The left elbow was flexed, swollen, warm, slightly erythematous, and painful upon mobilization. The white blood cell count was 22.00 × 109/liter, with 70% neutrophils (15.42 × 109/liter), 20% lymphocytes, and 10% monocytes, and the C-reactive protein (CRP) concentration was 235.3 mg/liter. An X ray of her left elbow was normal, and an ultrasound scan showed a moderate intra-articular effusion and a superior radial metaphysis subperiostal abscess. Aspirate from the articulation was purulent. Arthrotomy was performed promptly to drain the elbow joint, and empirical antibiotic treatment with intravenous cefamandole (150 mg/kg of body weight/day) was initiated.

Haemophilus influenzae was identified from the elbow aspirate and from blood cultures by using matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) technology (Vitek mass spectrometry [MS]; bioMérieux, France) on strains collected on chocolate agar plates (bioMérieux, France). The H. influenzae antiserum type b was negative, and the nitrocefin test was positive. According to the antibiogram realized with an ATB Haemo EU (08) panel (bioMérieux, France), the strain was resistant to ampicillin but sensitive to amoxicillin-clavulanate, cefotaxime, moxifloxacin, tetracycline, rifampin, co-trimoxazole, and chloramphenicol. Upon receipt of the antibiogram, the antibiotic treatment was immediately modified to intravenous ceftriaxone (50 mg/kg/day) for 14 days. The patient's temperature settled within a week, and the CRP concentration slowly decreased. The patient was discharged with an oral antibiotic treatment with amoxicillin-clavulanate (80 mg/kg/day) for 5 more weeks.

The National Reference Center of Haemophilus influenzae (Lille, France) identified a noncapsulated biotype III Haemophilus influenzae strain. The noncapsulated characteristics were confirmed by PCR (negative for the bexA gene). Antimicrobial susceptibility testing was carried out using the agar diffusion method. The isolated strain had a penicillinase and a mutation of the penicillin-binding protein 3 (PBP3) gene, which conferred resistance to ampicillin and amoxicillin-clavulanate (MIC = 2 μg/ml). The nonenzymatic resistance to β-lactam due to a mutation of the PBP3 gene was based on the increase in MIC and confirmed by PCR amplification of the ftsI gene, encoding PBP3. According to the new antibiogram, the antibiotic treatment was modified to oral co-trimoxazole (30 mg kg−1 day−1 sulfamethoxazole and 6 mg kg−1 day−1 trimethoprim) for 2 weeks. At the end of the treatment, i.e., 2 months after admission, the patient's elbow had improved significantly, with a range of motion back to normal. The leukocyte count had decreased to 11.5 × 109/liter, and the CRP concentration was lower than 0.3 mg/liter. A control ultrasound scan reported a decrease of the intra-articular effusion.

Patient 2.

A 66-year-old woman with fever and right-hip pain sought medical attention in the emergency department of the Edouard Herriot Hospital (Lyon, France) in November 2011. The patient had a long medical history, with rheumatoid polyarthritis diagnosed at the age of 17 years, and she had been treated with corticosteroids. Her past medical history was also remarkable for corticoid-induced diabetes, dyslipidemia, hypertension, and bilateral total hip and knee replacements. The rheumatoid polyarthritis was associated with rheumatoid vasculitis, which resulted in restrictive respiratory syndrome and ulcers of the lower extremities. The ulcers were known to be infected by Staphylococcus aureus and had been treated with local wound care. Over the course of her treatment, multiple antibiotic-induced iatrogenic events occurred, including cytolysis, Quincke's edema, febrile neutropenia, and tendinitis.

Upon admission, the patient presented with fever, right-hip pain, and symptoms of depression. The pelvic X ray was unremarkable. The leukocyte count was 11.6 × 109/liter (no differential count of leukocytes was available), and the CRP concentration was 257 mg/liter. The aspirate from the hip was purulent, containing Gram-negative coccobacilli. The causative bacterium and the antibiogram were identified as previously described and showed an H. influenzae strain resistant to ampicillin because of a penicillinase (nitrocefin test positive) and sensitive to amoxicillin-clavulanate, cefotaxime, moxifloxacin, tetracycline, rifampin, co-trimoxazole, and chloramphenicol. Blood cultures were sterile. An empirical antibiotic treatment taking into account the patient's allergy history was initiated and included intravenous imipenem (500 mg 4 times/day) associated with injections of gentamicin (15 mg/kg/day). A right-hip arthrotomy was performed 3 days later. The joint was drained and the right-hip prosthesis replaced at the same time. Blood, bones, and prosthesis material tissue cultures were all negative, but H. influenzae was isolated from the hip fluid. Antibiotic treatment was thereby completed by associating intravenous ofloxacin (200 mg 2 times/day) to the intravenous imipenem injections. Clinical and biological inflammatory markers improved within 2 weeks. The patient was discharged with her regular medication, which was associated with an antibiotic treatment comprising subcutaneous ertapenem (1 g 2 times/day) and oral ofloxacin (200 mg 2 times/day) for 3 months. A mortified tooth was considered to be the source of the H. influenzae infection and was extracted.

The H. influenzae strain was identified by the National Reference Center of H. influenzae as a biotype II noncapsulated H. influenzae strain. It was sensitive to ceftriaxone (MIC = 0.015 μg/ml), moxifloxacin, tetracycline, and co-trimoxazole and resistant to ampicillin (because of a penicillinase) and to amoxicillin-clavulanate (MIC = 2 μg/ml) (because of a mutation of the PBP3 gene similar to that in the strain from patient 1).

Non-type b and, more specifically, noncapsulated H. influenzae strains are uncommon causes of invasive diseases in children and adults. However, since the introduction of H. influenzae type b vaccine, the relative incidence of such infections has increased (1–4). Noncapsulated H. influenzae has mostly been reported in association with septicemia (1–3, 5), meningitis (1, 3, 4), and pneumonia (1, 2). However, rare cases of septic arthritis have also been also described (6–9).

The incidence of H. influenzae septic arthritis in children has decreased significantly since the introduction of vaccine-based immunization. In a retrospective study in the United Kingdom of 16 cases of H. influenzae septic arthritis in children, 14 cases occurred prior to immunization and 2 occurred after immunization (5). Another study reported no new cases of H. influenzae type b septic arthritis after the introduction of H. influenzae type b vaccine, whereas H. influenzae type b was responsible for 41% cases of septic arthritis prior to the immunization programs (10). Only 4 cases of septic arthritis due to non-type b H. influenzae have been reported, 2 in children with H. influenzae type b vaccination (on schedule for the children's ages) and 2 in adults whose H. influenzae strains were sensitive to antibiotics (Table 1). In the present report, we offered two new cases of septic arthritis, one of which occurred in an immunized child, and both H. influenzae strains presented the same two resistance mechanisms to antibiotics.

Table 1.

Characteristics of septic arthritis due to non-type b H. influenzae cases reported in the literature

Patient Age (yr) Sex Comorbidity(ies) H. influenzae serotype Resistance phenotype Reference
1 1 Female Premature birth, underweight Noncapsulated BLPACR Our study
2 66 Male Rheumatoid polyarthritis, long-term corticotherapy Noncapsulated BLPACR Our study
A 3 Female Type a β-Lactamase producer 6
B 0.4 Male Type f Wild type 7
C 46 Female Rheumatoid polyarthritis Noncapsulated Wild type 8
D 40 Male Chronic alcoholism, micronodular cirrhosis Noncapsulated Wild type 9

Empirical treatment for septic arthritis is directed toward the most common causes, i.e., Staphylococcus aureus, non-aureus streptococci, and Kingella kingae among children. Such regimens are frequently ineffective against penicillinase-producing H. influenzae, as shown in the two patients reported in this article. Beta-lactamase production was determined with a nitrocefin disk by the chromogenic cephalosporin method. The modification of penicillin-binding protein type 3 (PBP3) is an emergent resistance mechanism presented by H. influenzae. PBP3 is one of the eight PBPs in H. influenzae, and it is encoded by the ftsI gene (11). Several mutations of the ftsI gene are responsible for amino acid substitutions within the highly conserved motifs in the transpeptidase domain (S379-S-N, K513-T-G) that alter the transpeptidase activity of PBP3 (11–13). These mutations are therefore responsible for amoxicillin-clavulanate resistance (11–13). H. influenzae strains with both β-lactamase production and alteration in PBP3 are reported as β-lactamase positive and amoxicillin-clavulanic acid resistant (BLPACR) (11). While the prevalence of BLPACR strains remains low, it is probably underestimated because of the lack of a consensus-defining breakpoint and consistent technical performances (11). The incidence of BLPACR strains is highly varied among countries (Japan, 1.3% to 11%; France, 14%; United States, 0.15%) (11, 13). Dabernat and Delmas described the BLPACR prevalence among all isolates of noncapsulated H. influenzae in children of 5 years of age or less, and the prevalence was equal to 6.4% within the 2001-2008 period and to 2.4% among invasive isolates (14). Finally, as reported in 2011, among 141 H. influenzae strains with mutations in the transpeptidase domain of the ftsI gene, 47 (58.8%) were BLPACR (15). The phenotypic description of our strains as BLPACR was not found using the ATB Haemo (08) (bioMérieux, France) method, and our current H. influenzae antibiogram method should therefore be reevaluated in order to detect the modification of the PBP3 resistance mechanism.

The cases described in this report showed that a noncapsulated H. influenzae strain was capable of emerging as a pathogen causing septic arthritis in a child vaccinated with H. influenzae type b conjugate vaccine and in an immunosuppressed adult. Taking into account the resistance mechanisms occurring in the noncapsulated H. influenzae strain, empirical antibiotic therapy for septic arthritis should include coverage of the resistant strain described herein to avoid further articular and systemic deterioration.

ACKNOWLEDGMENTS

We acknowledge Pascal Rhéaume and Lars Petter Jordheim for assistance with the preparation of the manuscript.

Footnotes

Published ahead of print 20 March 2013

REFERENCES

  • 1. Bajanca P, Caniça M. 2004. Emergence of nonencapsulated and encapsulated non-b-type invasive Haemophilus influenzae isolates in Portugal (1989–2001). J. Clin. Microbiol. 42:807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Brown VM, Madden S, Kelly L, Jamieson FB, Tsang RS, Ulanova M. 2009. Invasive Haemophilus influenzae disease caused by non-type b strains in northwestern Ontario, Canada, 2002–2008. Clin. Infect. Dis. 49:1240–1243 [DOI] [PubMed] [Google Scholar]
  • 3. Dworkin MS, Park L, Borchardt SM. 2007. The changing epidemiology of invasive Haemophilus influenzae disease, especially in persons ≥65 years old. Clin. Infect. Dis. 44:810–816 [DOI] [PubMed] [Google Scholar]
  • 4. Tsang RS, Sill ML, Skinner SJ, Law DK, Zhou J, Wylie J. 2007. Characterization of invasive Haemophilus influenzae disease in Manitoba, Canada, 2000–2006: invasive disease due to non-type B. Clin. Infect. Dis. 44:1611–1614 [DOI] [PubMed] [Google Scholar]
  • 5. Shoaib A, Rethnam U, Bansal R, Clay N. 2007. The effects of mass immunization on Haemophilus influenzae type B-related orthopaedic disease. J. Pediatr. Orthop. B 16:236–238 [DOI] [PubMed] [Google Scholar]
  • 6. de Almeida AE, Schroeder LF, Caldeira S, da Silva M, Batista PR, Gallo MP, de Filippis I. 2008. Septic arthritis due to Haemophilus influenzae serotype a in the post-vaccination era in Brazil. J. Med. Microbiol. 57:1311–1312 [DOI] [PubMed] [Google Scholar]
  • 7. Glatman-Freedman A, Litman N. 1996. Septic arthritis caused by an unusual type of Haemophilus influenzae. J. Infect. 32:143–145 [DOI] [PubMed] [Google Scholar]
  • 8. Melhus A, Svernell O. 1998. Polyarticular septic arthritis caused by non-encapsulated Haemophilus influenzae biotype I in a rheumatic adult. Scand. J. Infect. Dis. 30:630–631 [DOI] [PubMed] [Google Scholar]
  • 9. Lester A, Beck Pedersen P. 1991. Serious systemic infection caused by non-encapsulated Haemophilus influenzae biotype III in an adult. Scand. J. Infect. Dis. 23:111–113 [DOI] [PubMed] [Google Scholar]
  • 10. Howard AW, Viskontas D, Sabbagh C. 1999. Reduction in osteomyelitis and septic arthritis related to Haemophilus influenzae type B vaccination. J. Pediatr. Orthop. 19:705–709 [PubMed] [Google Scholar]
  • 11. Tristram S, Jacobs MR, Appelbaum PC. 2007. Antimicrobial resistance in Haemophilus influenzae. Clin. Microbiol. Rev. 20:368–389 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. San Millan A, Giufré M, Escudero JA, Hidalgo L, Gutierrez B, Cerquetti M, Gonzales-Zorn B. 2011. Contribution of ROB-1 and PBP mutations to the resistance phenotype of a β-lactamase-positive amoxicillin/clavulanic acid resistant Haemophilus influenzae carrying plasmid pB1000 in Italy. J. Antimicrob. Chemother. 66:96–99 [DOI] [PubMed] [Google Scholar]
  • 13. Matic V, Bozdogan B, Jacobs MR, Ubukata K, Appelbaum PC. 2003. Contribution of β-lactamase and PBP amino acid substitutions to amoxicillin/clavulanate resistance in β-lactamase-positive, amoxicillin/clavulanate-resistant Haemophilus influenzae. J. Antimicrob. Chemother. 52:1018–1021 [DOI] [PubMed] [Google Scholar]
  • 14. Dabernat H, Delmas C. 2012. Epidemiology and evolution of antibiotic resistance of Haemophilus influenzae in children 5 years of age or less in France, 2001-2008: a retrospective database analysis. Eur. J. Clin. Microbiol. Infect. Dis. 31:2745–2753 [DOI] [PubMed] [Google Scholar]
  • 15. Barbosa AR, Giufré M, Cerquetti M, Bajanca-Lavado M-P. 2011. Polymophism in ftsI gene and β-lactam susceptibility in Portuguese Haemophilus influenzae strains: clonal dissemination of β-lactamase-positive isolates with decreased susceptibility to amoxicillin/clavulanic acid. J. Antimicrob. Chemother. 66:788–796 [DOI] [PMC free article] [PubMed] [Google Scholar]

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

RESOURCES