POS1186 STUDY OF SPONDYLODISCITIS WITHOUT BACTERIOLOGICAL DOCUMENTATION FROM A COHORT OF 142 PATIENTS WITH SUSPECTED INFECTIOUS SPONDYLODISCITIS ON IMAGING

Abstract

BackgroundThe incidence of infectious spondylodiscitis was estimated at 2.4/100,000 people in 2002. When faced with an image of spondylodiscitis on imaging, infectious spondylodiscitis is the most feared etiology. In recent years, several non-infectious spondylodiscitis etiologies have been described: Andersson lesion, crystal-induced discopathy, degenerative changes, etc (2). The identification of the germ by blood cultures or disc-vertebral puncture-biopsy allows the treatment to be best adapted antibiotic. Bacteriological investigation is inconclusive in about 30% (1). More and more undocumented spondylodiscitis are described.ObjectivesThe aim of this study is to describe a cohort of spondylodiscitis without bacteriological documentation and to compare it to spondylodiscitis with bacteriological documentation.MethodsUsing a “clinical data warehouse”, we searched for the key word “spondylodiscitis” in the documents associated with hospitalisations in the Rheumatology department of the Rouen University Hospital between 2010 and 2020. Four hundred and twenty-two records were analysed: 196 were excluded because they were not hospitalizedn in Rheumatology, among the 226 suspected spondylodiscitis, imaging allowed us to exclude 84 records for which there was no infectious spondylodiscitis and to retain 142 records. We collected demographic data, history, clinical symptoms, results of imaging, biological and bacteriological examinations. Statistical analysis was performed by Fisher’s exact test for qualitative data and by Mann-Whitney test for quantitative data.ResultsFour hundred and twenty-two patients were collected, 142 were analyzed. The average age was 65.5 +/- 14 years, 64.1% were male. Spinal pain was present in 96.5% of the cases without any difference between the two groups. One hundred and nine cases of spondylodiscitis were documented, of which 72 were confirmed by blood cultures and 36 by spinal disc biopsy. Thirty-three were not documented. Documented spondylodiscitis was more often febrile (41.3% vs 15.21% in undocumented cases, p = 0.006), had a greater biological inflammatory syndrome (mean CRP 152.4 +/- 112.6 mg/L versus 73 +/- 73.1 mg/L, p < 0.0001), and had been evolving for a shorter period of time than undocumented spondylodiscitis (54.1% vs 24.2% less than one month). Staphylococcus aureus was the most frequently retrieved bacteria 27.5%, followed by coagulase-negative staphylococci and streptococci (18.3% and 19.3% respectively). One hundred and twenty-one patients (85%) had an MRI at diagnosis; one hundred and two (71.8%) had a CT scan; and 81 patients (57%) had both examinations. The imaging analysis showed that there was no difference in soft tissue infiltration, erosions and abscesses. Probabilistic antibiotic therapy was proposed in 28/33 (84.5%) of cases. After collegial discussion, the alternative diagnoses retained were degenerative disc disease (4 cases), spinal gout (1 case), spondyloarthritis (1 case).ConclusionUndocumented spondylodiscitis is a recurrent problem in hospital practice. Alternative diagnoses are increasingly reported, their diagnosis is based on a collegial discussion.