Background: This study aimed to assess the most appropriate thresholds for neutrophils to lymphocytes ratio (NLR), platelets to lymphocytes ratio, monocytes to lymphocytes ratio (MLR), and platelets to mean platelet volume ratio in patients who had a suspected prosthetic knee infection. Furthermore, we evaluated the diagnostic accuracy of our proposed thresholds by men and women.
Methods: A total of 261 consecutive patients affected by painful total knee arthroplasty were included. International Consensus Meeting diagnostic criteria were used to identify periprosthetic infections. Sensitivity, specificity, positive, and negative predictive values were calculated for each cutoff value obtained. The area under the receiver operating characteristic curve was evaluated.
Results: NLR reported the best diagnostic accuracy. MLR and NLR reported higher area under the curves in men and women. We obtained an MLR value ≥0.30 (optimal cutoff value for men) and ≥0.17 (optimal cutoff value for women). In men, the sensitivity and the specificity were 0.71 and 0.77, respectively, versus 0.82 and 0.29, in women. For NLR, we obtained a value ≥2.52 (best cutoff value for men) and ≥2.46 (best cutoff value for women). These cutoffs reported a sensitivity and specificity of 0.71 and 0.88 versus 0.54 and 0.76 in men and women, respectively.
Conclusion: These biomarkers present a low diagnostic accuracy in periprosthetic joint infection detection. Men whose MLR and NLR were above cutoff values had a 77 and 88% probability of presenting a septic prosthetic failure. NLR of at least 2.46 was reasonably sensitive for women who have a periprosthetic knee infection.
Level of evidence: Diagnostic study, Level II.
Keywords: diagnostic accuracy; monocytes to lymphocytes ratio; neutrophils to lymphocytes ratio; periprosthetic joint infection; platelets to lymphocytes ratio; platelets to medium platelet volume ratio.
Copyright © 2022 Elsevier Inc. All rights reserved.