Article Text

THU0439 Bacterial Identification and Bad Prognostic Factors During Infectious Spondylodiscitis
  1. P. Guyader1,
  2. A. Bisson-Vaivre1,
  3. M. Verdet1,
  4. A.-C. Bouyeure2,
  5. S. Micu3,
  6. D. Alcaix3,
  7. O. Vittecoq1,
  8. X. Le Loët1,
  9. C. Zarnitsky3,
  10. V. Goëb4
  1. 1Rheumatology
  2. 2Biostatistic, University Hospital Rouen, Rouen
  3. 3Rheumatology, Le Havre Hospital, Le Havre
  4. 4Rheumatology, Amiens University Hospital, University Picardy Jules Verne, Amiens, France


Objectives Define the most relevant investigations to define the organism(s) involved in infectious spondylodiscitis and put in evidence the risk factors of poor outcome (death, relapse, recurrence) or after-effects at middle-term and long-term.

Methods Retrospective bicentric study on 306 patients affected by an infectious spondylodiscitis and followed-up between 2000 and 2009. Study on medical records and systematic telephonic call of patients and/or general practitioner. An uni and multivariate statistical analysis was performed.

Results The probability to isolate an organism was significantly lower when there was no fever (p<0,01; OR= 2,19 [1,23-3,94] or no inflammatory syndrome (p<0,01; OR=3,35 [1,27-8,81] as well as a recent spine surgery (p<0,001; OR=0,29 [0,16-0,55]), antibiotic exposure (p<0,01; OR=0,43 [0,23-0,78]) or malignancy past history (p<0,01; OR=0,42 [0,22-0,81]. A percutaneous discal and vertebral biopsy (PDVB) (p<0,0001; OR=0,29 [0,16-0,55]) even with PCR analysis (p<0,0001; OR=0,17 [0,08-0,36]) was not associated with a better probability to identify the germ. The diagnostic delay was significantly longer in the group of patients without isolated organism (p<0,01). An initial neurologic trouble (p<0,0001), an elevated age (p<0,0170), liver cirrhosis (p<0,01; OR=6,4 [1,74-19,05]) or epidural abscess (p<0,001; OR=3,07 [1,42-6,61]) were significant and independent factors of poor outcome. Back (p<0,01; OR=2,36 [1,28-4,34]) and radicular pains (p<0,01; OR=2,73 [1,46-5,10]), a recent spine surgery (p<0,01; OR=2,85 [1,50-5,37]) were associated to back pain sequelae at middle-term. They were significantly associated to spine stiffness (p<0,001; OR=2,85 [1,50-5,37]), persistant neurologic trouble (p=0,0365; OR= 2,62 [1,03-6,65]) or disc injury (p<0,001; OR=2,85 [1,50-5,37]). Long-term pain was significantly more frequent when the spondylodiscitis began after a back surgery (p<0,02; OR=3 [1,09-8,23]).

Conclusions The probability to isolate an organism was lower when clinical and biological presentations were less severe. However, the antibiotic therapy allowed similar vital and functional prognostic even if the the organism was identified or not. Our results also showed that a second PDBV or even a chirurgical biopsy were not justified.

References Spilf. Primary infectious spondylodiscites, and following intradiscal procedure, without prosthesis. Recommendations. Med Mal Infect2007;37:269-77.

Zimmerli W. Clinical practice. Vertebral osteomyelitis. N England J Med2010;362:1022-9

O’Daly BJ. Long-term functional outcome in Pyogenic spinal infection. Spine (Phila Pa 1976) 2008;33:246-53

Disclosure of Interest None Declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.