European Journal of Clinical Microbiology & Infectious Diseases

, Volume 26, Issue 9, pp 667–670

Vertebral osteomyelitis caused by Actinobaculum schaalii: a difficult-to-diagnose and potentially invasive uropathogen

Authors

  • P. Haller
    • Division of Infectious Diseases and Hospital EpidemiologyUniversity Hospital Basel
  • T. Bruderer
    • Microbiology LaboratoryUniversity Hospital Basel
  • S. Schaeren
    • Department of Orthopedic SurgeryUniversity Hospital Basel
  • G. Laifer
    • Division of Infectious Diseases and Hospital EpidemiologyUniversity Hospital Basel
  • R. Frei
    • Microbiology LaboratoryUniversity Hospital Basel
  • M. Battegay
    • Division of Infectious Diseases and Hospital EpidemiologyUniversity Hospital Basel
  • U. Flückiger
    • Division of Infectious Diseases and Hospital EpidemiologyUniversity Hospital Basel
    • Department of Internal MedicineKantonsspital Olten
Case Report

DOI: 10.1007/s10096-007-0345-x

Cite this article as:
Haller, P., Bruderer, T., Schaeren, S. et al. Eur J Clin Microbiol Infect Dis (2007) 26: 667. doi:10.1007/s10096-007-0345-x

Abstract

We report the first case of vertebral osteomyelitis caused by Actinobaculum schaalii and review all cases of A. schaalii identified at our institution between 2002 and 2005. A. schaalii causes urinary tract infections – especially in elderly people – occasionally with septic complications.

Introduction

Actinobaculum schaalii was first described in 1997 [1]. It is a Gram-positive, straight to slightly curved, nonmotile, facultatively anaerobic rod rarely reported as a cause of urinary tract infections (UTI). It belongs to the genus Actinobaculum, which is closely related to Actinomyces and Arcanobacterium. Before the implementation of 16S rDNA sequencing, this bacterium was misclassified as Actinomyces [1]. The genus Actinobaculum also contains Actinobaculum suis, which is a known pathogen in veterinary medicine that causes cystitis and metritis with abortions in pregnant sows [2, 3]. Two additional Actinobaculum species have recently been proposed: Actinobaculum massiliae and A. urinale. The former was isolated from the urine of a patient with UTI [4, 5] and from the pus of a patient with a superficial skin infection [6], while A. urinale was isolated from the urine of two patients with UTI [7, 8].

Isolating and identifying A. schaalii using standard laboratory techniques is difficult because of its slow growth under aerobic conditions and its resemblance to normal bacterial flora on skin and mucosa. The inability of most commercial biochemical identification systems to recognize A. schaalii represents an additional difficulty in terms of the correct identification. Therefore, for patients with a high suspicion of UTI but negative routine urinary culture, the proposed approach is to carry out a urine Gram stain followed by prolonged incubation of the urine in 5% CO2, if significant leucocyturia and gram-positive rods on microscopy indicate the presence of a UTI [9]. The increasing use of 16S rDNA sequencing has also enabled A. schaalii to be identified and reported more frequently [9, 10].

Since January 2002 at the University Hospital Basel, a 700-bed primary and tertiary care center, all relevant bacterial isolates that can not be identified unambiguously by standard biochemical methods are identified by 16S rDNA sequencing (Microseq 500 Bacterial identification Kit; Applied Biosystems, Foster City, Calif.). All bacteria growing in specimens supposed to be sterile and monocultures from unsterile sites are considered to be relevant.

Between January 2002 and December 2005 we identified five patients with A. schaalii isolates, including one patient with vertebral osteomyelitis (spondylodiscitis) due to A. schaalii. To our knowledge, this is the first report of a patient with A. schaalii vertebral osteomyelitis.

Case report

This 71-year-old male patient was admitted to our hospital with immobilizing low-back pain that had persisted for 10 days. Pain was more intense with flexion of the hip and coughing and it radiated to his left leg. Apart from a discrete hypesthesia at the lateral rim of his left foot (S1) he showed no neurological symptoms. Four to six weeks prior to admission the patient had experienced chills, nocturnal sweating and intermittent dysuria. Noteworthy in his personal history were a past hepatitis B-infection and excision of the upper left lobe of the lung and ribs I–IV for tuberculosis. He was heterozygous for factor V Leiden and had experienced thromboembolic disease 3 years earlier. He suffered from paroxysmal ectopic atrial tachycardia and arterial hypertension with hypertensive cardiopathy. His regular medication consisted of phenprocoumon, bisoprolol and ramipril. At admission the patient was not able to walk and had bilateral lumbar tenderness. Breath sounds over the left lung were diminished. He presented with tachycardia of 130 beats per minute, a blood pressure of 117/97 mmHg and a temperature of 37.5°C. The C-reactive protein (CRP) was 55 mg/L and the leukocyte count was 8.09 × 109/L. A computed tomography (CT) of the lumbar spine showed no signs of the clinically suspected vertebral osteomyelitis. Because of a urine culture with >100,000 CFU/ml of mixed flora, a UTI was diagnosed, and the patient was treated with ciprofloxacin for 4 days followed by 2 days of sulfamethoxazol/ trimethoprim. The initially elevated infection parameters (maximum leukocytes: 12.65 × 109/L; maximum CRP: 112 mg/L) decreased gradually. Due to the persistence of the immobilizing low-back pain, a magnetic resonance imaging scan of the lumbar spine was performed 2 weeks after admission; this revealed vertebral osteomyelitis at L3/L4 with slight destruction of the vertebral endplates. The CT-guided biopsy of the disc L3/L4 yielded moderate amounts of A. schaalii and Corynebacterium striatum. The cytology showed granulocytic inflammation. A. schaalii was also isolated from blood cultures drawn on the same day (in one of four bottles) and was therefore considered to be the causative pathogen. The empiric treatment with vancomycin was changed to amoxicillin/ clavulanic acid 3 × 2.2 g/day intravenous (iv) after 2 days, when the susceptibility testing with the Etest method (AB Biodisk, Solna, Sweden) demonstrated a minimum inhibitory concentration (MIC) of 0.38 mg/L. A. schaalii was also susceptible to ceftriaxone (MIC = 0.125 mg/L), clindamycin (MIC = 0.023 mg/L), rifampicin (MIC < 0.002 mg/L), doxycyclin (MIC = 0.125 mg/L) and vancomycin (MIC = 0.25 mg/L). During the 8 weeks of intravenous treatment the low-back pain improved and the patient was mobilized with a soft brace. Ten months later the patient showed a nearly complete recovery with only mild stress-dependent low-back pain. Radiologically the vertebral osteomyelitis was healed and the infection parameters were completely normal.

In addition to this patient, A. schaalii was identified in four more patients during the observation period at our institution. All four were elderly people who presented with cystitis; A. schaalii was cultured from their urine (Table 1: cases 2–5).
Table 1

Clinical characteristics of all reported patients with Actinobaculum schaalii infection, including five new cases

Case no.

Age (years)

Sex

Clinical presentation

Specimens with A. schaalii

Concomitant flora

Underlying genitourinary tract pathologies

Treatment

Reference

1

71

M

Vertebral osteomyelitis

Blood disc biopsy

Corynebacterium striatum

 

Amoxicillin/clavulanic acid

a

2

89

F

Cystitis

Urine

  

Amoxicillin/clavulanic acid

a

3

92

M

Cystitis

Urine

 

Overflow incontinence after prostatectomy

Ciprofloxacin

a

4

92

F

Cystitis

Urine

Escherichia coli

 

Norfloxacin

a

5

83

M

Cystitis

Urine

 

Urinary catheter, overflow incontinence

Ciprofloxacin

a

6

64

M

Chronic pyelonephritis

Blood

Unknown

Unknown

Unknown

[1]

7

5

M

Pyelonephritis

Urine

 

Pyeloureteral junction obstruction

Amoxicllin/ clavulanic acid, vancomycin, pyeloplasty

[14]

8

68

M

Sepsis

Blood

Actinobaculum urinale

Chronic renal failure

Metronidazole, piperacillin/tazobactam

[7]

9

70

M

Urosepsis

Blood urine

 

Prostatic hyperplasia

Cefuroxime and gentamicin

[10]

10

63

M

Urosepsis

Blood

Coagulase-negative staphylococci

Carcinoma vesicae

Meropenem, gentamicin, nephrostomy catheter

[10]

11

63

M

Pyelonephritis

Pus from kidney cyst

 

Prostatic hyperplasia, congenital cystic kidney malformation

Ampicillin, gentamicin-ciprofloxacin, surgical drainage

[10]

12

9mo

F

Cauda equina syndrome

Intradural abscess

Nonhemolytic streptococci

 

Penicillin, metronidazole, surgical drainage

[10]

13

70

F

Urosepsis

Blood urine

  

Cefuroxime and gentamicin

[10]

14

77

F

Cystitis

Urine

  

Nitrofurantoin followed by ampicillin

[10]

15

84

M

Cystitis

Urine

  

Cefuroxime

[10]

16

65

F

Cystitis

Urine

Aerococcus urinae

Pyelonephrosis

Amoxicillin followed by amdinocillin

[10]

17

84

F

Cystitis

Urine

Staphylococcus aureus

Urinary catheter

Sulfamethizole

[10]

18

65

M

Urosepsis

Blood

 

Paraplegia, kidney stones

Cefuroxime, ciprofloxacin, metronidazole nephrectomia

[10]

19

78

M

Urosepsis

Blood urine

Aerococcus urinae

Urinary catheter

Ceftriaxone, amoxicillin, ciprofloxacin, co-trimoxazole

[9]

20

72

M

Recurrent cystitis

Urine

Citrobacter koseri

Urinary incontinence, intermittent catheterization

Unknown

[9]

21

39

M

Sepsis

Urine

Mixed anaerobes

Urinary incontinence, intermittent catheterization

Ciprofloxacin and metronidazole

[9]

aNew cases reported at our institution

The 0.52-kb partial 16S rDNA sequences of our isolates were deposited in GenBank under the following accession numbers: EF151124 (case 1), EF151125 (case 2), EF151126 (case 3), EF151127 (case 4), and EF151128 (case 5).

Discussion

Our five new cases and the 16 cases previously reported (Table 1) show that older people are probably more susceptible to A. schaalii: Of the 21 patients, 18 were between 63 and 92 years of age. Interestingly, 13 patients had an underlying genitourinary tract pathology, which may be an additional predisposing condition for infection with this pathogen. A. schaalii seems to cause predominantly UTI: 17 patients had a diagnosis of UTI (nine had cystitis, three pyelonephritis and five urosepsis). In 15 patients, A. schaalii was isolated in urine or pus from the kidney. However, in eight patients, A. schaalii was isolated from the blood, indicating that this pathogen may become invasive.

UTI with bacteremia is a frequent cause of vertebral osteomyelitis [11, 12]. In our patient, the history of chills and intermittent dysuria some weeks before the acute onset of back pain support the hypothesis that UTI with A. schaalii bacteremia caused the vertebral osteomyelitis as a septic complication. The biopsy of the disc L3/L4 yielded not only A. schaalii but also Corynebacterium striatum. However, as only A. schaalii was found in the blood cultures, we assumed it to be the causative agent. Indeed, Corynebacterium species colonize the skin and mucous membranes and are usually considered to be contaminants when recovered in the laboratory. Moreover, to date, C. striatum has been reported to cause vertebral osteomyelitis in only one patient [13]. The blood cultures of this patient tested positive for C. striatum and he was subsequently treated with ciprofloxacin. Two weeks later he presented with lumbar back pain and a CT-scan showed the destruction of L3/L4. However, at that moment the blood cultures remained negative, and no disc biopsy was performed to confirm C. striatum as the causative pathogen.

In conclusion, A. schaalii may cause UTI and septic complications such as vertebral osteomyelitis, particularly in elderly patients with underlying genitourinary tract pathology. The incidence of A. schaalii infections is probably underestimated, since it may be missed by routine culture procedures. This is confirmed by the fact that we found five patients with A. schaalii infections during the 48 months following implementation of 16S rDNA sequencing. In patients with a high suspicion of UTI but negative routine urinary culture, A. schaalii should be considered as a possible causative agent.

Acknowledgements

We are indebted to Dr. P.D.J. Sturm, Department of Medical Microbiology, Radboud University Nijmegen Medical Center, for providing additional detailed information about cases numbers 20 and 21.

Copyright information

© Springer-Verlag 2007