Diagnostic Value of Neutrophil-Lymphocyte Ratio in Predicting Post-Operative Infection after Orthopedic Surgery: A Systematic Review and Meta-Analysis

Zhan Peng,Yukun Jia,Jin Li,Guangye Wang
DOI: https://doi.org/10.1089/sur.2024.002
Abstract:Objective: This study aims to evaluate the predictive value of neutrophil-lymphocyte ratio (NLR) in determining infection after orthopedic surgery. Methods: A comprehensive search was conducted in PubMed, EBASE, CNKI, and Wanfang databases to identify relevant studies. The quality of the included studies was assessed using QUADAS-2. Data extraction was performed to calculate sensitivity, specificity, and other indicators. Bivariate mixed-effects meta-analysis was conducted using Stata software. The sources of heterogeneity were evaluated, and a summary receiver operating characteristic curve was generated. Results: A total of 16 literatures comprising 18 studies involving 3737 patients were included in this analysis. NLR demonstrated moderate sensitivity (0.77) and specificity (0.69) in diagnosing orthopedic post-operative infection, with an area under the curve of 0.80 and diagnostic odds ratio of 7.76. Significant heterogeneity was observed among the studies, primarily due to variations in surgical type, infection type, blood test timing, and NLR cutoff value. Fagan nomogram indicated that NLR could increase the positive posterior probability to 72% and decrease the negative posterior probability to 25%. The pooled effect of the likelihood ratio dot plot for diagnosis fell in the lower right quadrant. Deek funnel plot suggested no publication bias in this study. Conclusion: NLR holds certain value in diagnosing infection after orthopedic surgery and can provide additional information to assess the risk of infection. However, its predictive performance is influenced by various factors, and it cannot be used as a sole criterion for confirming the diagnosis. Prospective studies should be conducted in the future to optimize the diagnostic threshold and explore its combination with other indicators.
What problem does this paper attempt to address?