Abstract
Aims
We have reviewed the published classifications of long-bone osteomyelitis. This review demonstrated the limitations and poor recognition of existing classifications. We have designed a new system which includes four easily identifiable variables which are Bone involvement, Antimicrobial availability, Soft tissue coverage and Host status. This is called the B.A.C.H. classification system. In this study, we aim to retrospectively validate this classification in a cohort of osteomyelitis cases.
Methods
We identified 100 patients who had received surgery for osteomyelitis between 2013–2015 in a single specialist centre. Each patient was classified retrospectively by two assessors who were not involved in the initial patient care. Osteomyelitis was confirmed in each patient by a validated composite protocol.
Results
All patients in this series could be classified using each of the B.A.C.H. variables. Seventy-four patients were categorised as B1, 13 as B2 and 13 as B3. Thirty-four patients revealed no growth of microorganisms (Ax). Fifty-four were A1, 11 A2 and one patient was classified as A3. For rare organisms (e.g. Corynebacterium spp.), classification required specialised infectious disease knowledge. Twenty-four patients needed soft tissue procedures (C2) and 76 had their wound closed primarily (C1). Twenty patients did not need optimisation prior to surgery and were deemed as H1. The remaining 80 patients needed optimisation prior to surgery and were deemed as H2.
Conclusions
All patients were classifiable when using the B.A.C.H. system. This system offers a simple method of stratifying long-bone osteomyelitis and may give an indication of severity and the need for specialist intervention. However, there were difficulties in classification of rare causative organisms. This validation has been performed in a single specialist centre for osteomyelitis and requires both internal prospective and external validation to evaluate its reproducibility.
