Criteria for Differentiation of Non-Bacterial and Haematogenous Osteomyelitis: A Case-Control Study With Prospective Verification of the Outcomes

Abstract

Background. Patients with haematogenous and non-bacterial osteomyelitis have similar clinical symptoms (pain in the nidus area, soft tissue swelling, fever) and laboratory signs (increased erythrocyte sedimentation rate, leukocyte count, C-reactive protein concentration). The criteria for distinguishing these two states are not determined.Objective. Our aim was to determine diagnostic criteria to differentiate haematogenous and non-bacterial osteomyelitis.Methods. The study included data of patients under the age of 18 years with non-bacterial or haematogenous osteomyelitis hospitalised to two clinical centres from 2009 to 2016. The diagnosis was established and re-verified according to archival data (medical history) and after two years of observation (at least once a year). Clinical, anamnestic and laboratory data (haemoglobin, leukocytes, leukocyte formula, platelets, ESR and C-reactive protein, CRP) as well as the results of radiation diagnostics (X-ray, CT scan, MRI or osteosyntigraphy) obtained at the disease onset were taken into account as potential diagnostic criteria.Results. Out of 145 patients with non-bacterial or haematogenous osteomyelitis, the diagnosis was re-verified in 138, of them non-bacterial osteomyelitis — in 91, haematogenous osteomyelitis — in 47. The following criteria had the highest diagnostic value for establishing cases of non-bacterial osteomyelitis: detection of bone destruction foci surrounded by osteosclerosis area [sensitivity (Se) 1.0; specificity (Sp) 0.79]; absence of fever (Se 0.66; Sp 0.92); the number of bone destruction foci > 1 (Se 0.73; Sp 1.0); CRP 55 mg/L (Se 0.94; Sp 0.73); negative results of bacteriological examination of the material from the bone destruction focus (Se 1.0; Sp 0.67).Conclusion. Diagnostic criteria for differentiation of non-bacterial and haematogenous osteomyelitis have been described. Further research on the efficacy of using these criteria to reduce the risk of diagnostic errors, decrease the diagnostic pause, reduce the risk of non-bacterial osteomyelitis complications is needed.