Abstract
Background and hypothesis Gram staining is a fundamental part of the early diagnosis of septic arthritis and helps direct initial antimicrobial therapy. Though research shows uniform high specificity, its sensitivity is low with a comparatively wide spread of values. We hypothesise that a significant component of the variation in Gram stain sensitivity may be attributable to suboptimal laboratory technique and limited experience. Method This is a retrospective observational study. Data from a total of 269 native and prosthetic joint aspirates were collected over two 12-month periods: October 2017 to October 2018 (cohort 1) and January 2023 to January 2024 (cohort 2). Between these periods, the Gram staining technique remained the same; however, the laboratory achieved accreditation from the United Kingdom Accreditation Service (UKAS). This accreditation was achieved through improved staff training in Gram staining technique, and a policy of second opinions for challenging samples was introduced. Data were analysed using MS Excel (Microsoft Corporation, Redmond, Washington, United States) and StatsKingdom software. Results Following UKAS accreditation, Gram stain sensitivity improved from 25.6% to 44.7%. A one-tailed proportions test demonstrated this to be statistically significant at p = 0.035. Regarding secondary outcomes, S. aureus was the most prevalent organism. It was cultured in isolation or as a polymicrobial finding in 51% of aspirates in cohort 1 and in 34% of samples in cohort 2. Crystal arthropathy and septic arthritis were not mutually exclusive. In cohort 1, 12% crystal arthropathy cases also grew microorganisms. In cohort 2, 15% of crystal arthropathies also had a positive culture. Conclusion In keeping with previous work, this study also shows Gram staining has relatively low sensitivity in the diagnosis of septic arthritis. However, retrospective analysis shows that significant improvements can be made through training and supportive laboratory practices and may go some way to explain the variation in Gram stain sensitivity in the wider literature.