Abstract
Background/Objectives: Periprosthetic joint infection (PJI) remains one of the most devastating complications of arthroplasty, with early diagnosis crucial for successful management. The serum neutrophil-lymphocyte ratio (NLR) and monocyte-lymphocyte ratio (MLR) have been proposed as simple, inexpensive inflammatory biomarkers, but their diagnostic performance in PJI remains unclear. This meta-analysis aimed to compare the diagnostic accuracy of serum NLR and MLR in detecting PJI. Materials and Methods: A systematic literature search was conducted in PubMed, Web of Science, and Scopus up to April 2025. Twenty-nine eligible studies (n = 14,040 patients; 3418 with PJI, 10,622 without PJI) reporting diagnostic metrics for serum NLR or MLR were included. Extracted data comprised mean biomarker values, cut-off thresholds, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Non-parametric statistical tests and subgroup analyses were applied to examine performance across infection types and PJI definitions. Results: Both serum NLR and MLR were significantly elevated in PJI patients compared with aseptic cases (p < 0.001 and p = 0.003, respectively). Pooled diagnostic accuracy was moderate: mean AUC 0.719 for NLR and 0.700 for MLR. For NLR, mean sensitivity was 69.9% and specificity 69.8%, with an average cut-off of 2.88. For MLR, mean sensitivity was 68.2% and specificity 70.4%, with an average cut-off of 0.34. Subgroup analyses indicated superior diagnostic performance of NLR in acute infections and variability depending on the PJI definition employed (p = 0.037). Strong correlations were observed between standardized mean differences in biomarker levels and corresponding diagnostic accuracy, particularly for NLR (ρ = 0.802, p = 0.002). Conclusions: Serum NLR demonstrates slightly superior diagnostic accuracy over serum MLR in identifying PJI, especially in acute settings. Both markers are inexpensive and widely accessible but show only moderate discriminative capacity, supporting their role as adjunctive rather than standalone diagnostic tools. Further large-scale prospective studies with harmonized methodologies are needed to refine biomarker thresholds and integrate them into multimodal diagnostic algorithms.