Reliability of Total Nucleated Cell Counts in the Setting of Hip Arthroplasty

Abstract

Abstract Background Total nucleated cell (TNC) count and differential are used to classify joint effusions as inflammatory or noninflammatory. Further diagnostic evaluation and management is contingent on this classification. TNC count can be measured by automated analyzers or by manual assessment using a hemocytometer. Studies have raised concerns regarding the accuracy of TNC counts measured by automated instruments, particularly in the setting of joint arthroplasty. The objective of this study was to determine whether metallosis, a complication of total hip arthroplasty in which metal debris accumulates in periprosthetic tissues and synovial fluid, is associated with inaccurate TNC counts in synovial fluid. Methods We compared the accuracy of cell counts measured by the Sysmex XN-1000 and Beckman Coulter Iris iQ200 with the gold standard of manual assessment using a hemocytometer in synovial fluid from patients with suspected metallosis and in fluid obtained from controls from patients with native joints and a history of arthroplasty for other indications. Results TNC counts produced by automated analyzers were associated with increased levels of discordance (relative to manual counts) in patients with metallosis. Metallosis was not associated with increased levels of discordance for RBC counts or WBC differentials. The Sysmex XN flagged all but 1 metallosis sample for manual verification of the results. Conclusions Automated methods are generally reliable for analysis of synovial fluid. TNC counts can be inaccurate in the context of metallosis following total hip arthroplasty. Laboratories should correlate automated cell counts with a microscopic assessment of the specimen, as recommended by instrument manufacturers.