Background The analysis of synovial fluid is an important tool for diagnosing joint disease. When synovial fluid is removed, the white cell count (WCC) decreases with time, and an inflammatory liquid could become a false non-inflammatory specimen. Reagent strip testing of urine is a valid tool for the diagnosis of urinary tract infection, via the detection of leukocyte esterase activity. It has been used for the analysis of others body fluids. Synovial fluid test at the site of arthrocentesis using reagent strips could have potential benefits as a screening tool.

Objectives To evaluate the performance of leukocyte esterase reagent strips for diagnosis of inflammatory synovial fluid.

Methods Prospective single center study. We analyzed synovial fluids samples collected from patients in a tertiary university Hospital (November 2015- December 2016). Synovial fluid samples were tested within 1 hour after collection. We analyzed: The presence of leukocyte esterase using the leukocyte esterase reagent strips test (originally designed for urine test, URI-Clip Test, Menarini Diagnostics). It was recorded semi quantitatively: negative, 1+ (>25 WBC/uL), 2+ (>75 WBC/uL) or 3 +(>500 WBC/uL) by comparison with a standard color chart found on the container's label. The WCC, formula, glucose level. The WCC was measured by manual leukocyte counting, using saline as diluents. Cultures were also collected. We consider + if leucocyte esterase pad was more or equal than 1+ positive. The cut-off for the WCC (>2000cells/mm³) was used to differentiate between inflammatory and non-inflammatory specimens. We compared the WCC (reference standard diagnostic test) with the presence of leukocyte esterase using the leukocyte esterase reagent. Sensitivity (Se), specificity (Sp), PPV, NPV were determined. P-value smaller than 0.05 were considered significant.

Results During the study period, 125 joint fluid specimens were analyzed: 56 (44.8%) mechanical and 69 (55.2%) inflammatory. Of the mechanical fluids 33 (58.9%) were negative by leukocyte esterase reagent and of the inflammatory fluids 67 (97.1%) were positive. The Se and Sp of leukocyte esterase reagent was 97.1% and 58.9% respectively. The PPV was 74.4% and NPV was 94.3%. The 2 false-negative results (negative by leukocyte esterase reagent but more than 2000 WBC/mm³), showed a predominance of mononuclears (> =91%), the median WCC was 2 775/mm and median neutrophil percentage was 8.5%. For inflammatory fluids: semi-quantitative results (negative, 1+, 2+ and 3+) were significantly different regarding the main leukocyte, neutrophil and lymphocyte count (table).

Conclusions Our results demonstrate that leukocyte esterase reagent strips are a rapid, cheap, and sensitive tool to identify inflammatory synovial fluid. Leukocyte esterase reagent strips had an excellent Se but a poor Sp, it could be used as a screening tool in primary care practice. A positive result may indicate an inflammatory process, then the patient should be referred to a rheumatologist.

Disclosure of Interest None declared