Objective: Postoperative infection is a common but costly complication. The neutrophil-lymphocyte ratio is a promising marker for the identification of postsurgical infectious events. We aimed to perform this meta-analysis to assessed the accuracy of the neutrophil-lymphocyte ratio for the prediction of postsurgical infection.

Methods: We searched PubMed, Embase, Web of Science, and Cochrane Library without language restriction from their inceptions to April 2022, and checked reference lists of included studies. Studies were included if they assessed predictive accuracy of neutrophil-lymphocyte ratio for postsurgical infection. We estimated its predictive value and explored the source of heterogeneity. The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool was used to assess methodological quality and the Deeks' test to evaluate publication bias. The bivariate model and hierarchical summary receiver operating characteristic (HSROC) curve were used for meta-analysis and generated a summary receiver operating characteristic space (ROC) curve.

Results: Our search returned 379 reports, of which 12 fulfilled the inclusion criteria, accounting for 4375 cases. The bivariate analysis yielded a pooled sensitivity of 0.77 (95%C.I.: 0.65-0.85) and specificity of 0.78 (95%C.I.: 0.67-0.86). Pooled positive LR and negative LR were 3.48 (95%C.I.: 2.26-5.36) and 0.30 (95%C.I.: 0.20-0.46), respectively. A negative LR of 0.30 reduces the post-test probability to 2% for a negative test result. The area under of receiver operating characteristic curve was 0.84 (95%C.I.: 0.80-0.87). Subgroups comparisons revealed difference by study design, surgical site, presentence of implant, time of sampling, type of infection event and prevalence of infection. The Deeks' test showed no publication bias. The sensitivity analysis showed no study affected the robustness of combined results.

Conclusions: Low-certainty evidence suggests that the neutrophil-lymphocyte ratio is a helpful marker for predicting postoperative infectious complication. The negative predictive value of the neutrophil-lymphocyte ratio enables for reliable exclusion of postoperative infection.Trial registrationPROSPERO registration number CRD42022321197. Registered on 27 April 2022.

Keywords: Infection; Meta-analysis; Neutrophil-lymphocyte ratio; Postoperative complication; Prediction.

© 2023 The Authors. Published by Elsevier Ltd.