Graphical Representation of Cell Count in Native Joint Aspirates

Objectives: The aim of the study was to describe the graphical synovial cell distribution as LMNE matrix patterns in native knee aspirates and to explore their association with septic arthritis, crystal arthropathy, rheumatoid arthritis and haemarthrosis. Methods: In total, 117 patients with knee joint effusions of various origins underwent aspiration of the knee joint and the aspirate analysed using the Yumizen H500 automated cell counter. Microscopy of the aspirate was performed at the same time in cases with a cluster in the so-called noise field or previously unknown cell distribution. Results: The graphical type differentiation of automated synovial cell count analyses of native joints allowed 9 different types to be recognised: Type I as particle type, type II as infection type, type III as combination type of infection and particles, type IV as indifferent type, type V as haematoma type and type VI as combination type of infection and haematoma, type VII as particle type with haematoma, type VIII in rheumatoid arthritis and type IX as mixed type combining infection and rheumatoid arthritis. Differentiation of LMNE types into infection types (types II, III, VI and IX) and non-infection types (types I, IV, V, VII and VII) resulted in a sensitivity of 100%, a specificity of 93.9%, an accuracy of 96.6%, a positive predictive value (PPV) of 92.7%, a negative predictive value (NPV) of 100%, a positive likelihood ratio (LR+) of 16.5 and a negative likelihood ratio (LR−) of 0. Conclusions: Graphical LMNE matrix analysis of native knee aspirates identified reproducible patterns associated with septic arthritis, crystal arthropathy, rheumatoid arthritis, and haemarthrosis, which may be helpful in distinguishing these clinical entities, suggesting that graphical cell count interpretation may provide useful adjunctive information to the standard diagnostic work-up. The study is a preliminary descriptive study and requires validation using a larger number of patients.