Cara Cipriano1, Aparna Maiti1, Gregory Hale1, William Jiranek1. 1. Department of Orthopaedic Surgery, Virginia Commonwealth University, 1112 East Clay Street, McGuire Hall Annex, Richmond, VA 23298. E-mail address for C. Cipriano: caracipriano@gmail.com. E-mail address for A. Maiti: amaiti2@vcu.edu.
Abstract
BACKGROUND: Toll-like receptors (TLRs) 1 and 6 are consistent molecular indicators of the host inflammatory response against bacterial infection. Our aims were to determine whether TLR elevation could be detected in infected periprosthetic tissues and to assess the utility of these biomarkers as tests for detecting a periprosthetic joint infection. METHODS: Fifty-nine patients undergoing revision total joint arthroplasty (twenty-seven hips and thirty-two knees) were prospectively evaluated for periprosthetic joint infection according to currently recommended diagnostic criteria. Nine patients were excluded because of insufficient work-up, leaving fifty available for study. Of these, twenty-one were categorized as infected and twenty-nine as noninfected. Periprosthetic tissues were collected intraoperatively, and total RNA was extracted by standard techniques. Expression of TLR messenger RNAs was assessed by first-strand complementary DNA synthesis from 1 μg of total RNA followed by real-time PCR (polymerase chain reaction). Results were normalized relative to the housekeeping gene GAPDH (glyceraldehyde 3-phosphate dehydrogenase). Expression of TLRs 1, 6, and 10 in the infected and noninfected groups was compared with use of the Student t test. The receiver operating characteristic curve, area under the curve (AUC), sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-) were calculated to determine the accuracy of each TLR for predicting periprosthetic joint infection at its optimal diagnostic threshold. RESULTS: Mean TLR1 mRNA expression was significantly elevated in infected compared with noninfected samples (0.600 compared with 0.005, p = 0.0003); the same was true of TLR6 (0.208 compared with 0.0165, p = 0.0059) but not of TLR10 (0.00019 compared with 0.00014, p = 0.6238). The AUC was 0.995 for TLR1, 0.883 for TLR6, and 0.546 for TLR10. The optimal threshold for diagnosing periprosthetic joint infection was 0.0924 for TLR1 (sensitivity = 95.2%, specificity = 100%, LR+ = 13.80, LR- = 0.91) and 0.0215 for TLR6 (sensitivity = 85.7%, specificity = 82.8%, LR+ = 4.98, LR- = 0.83). CONCLUSIONS: In our pilot study, TLR1 expression in periprosthetic tissues most accurately predicted periprosthetic joint infection. This measure of the host response may be particularly helpful in detecting culture-negative infections and avoiding false positives resulting from contamination. LEVEL OF EVIDENCE: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.
BACKGROUND: Toll-like receptors (TLRs) 1 and 6 are consistent molecular indicators of the host inflammatory response against bacterial infection. Our aims were to determine whether TLR elevation could be detected in infected periprosthetic tissues and to assess the utility of these biomarkers as tests for detecting a periprosthetic joint infection. METHODS: Fifty-nine patients undergoing revision total joint arthroplasty (twenty-seven hips and thirty-two knees) were prospectively evaluated for periprosthetic joint infection according to currently recommended diagnostic criteria. Nine patients were excluded because of insufficient work-up, leaving fifty available for study. Of these, twenty-one were categorized as infected and twenty-nine as noninfected. Periprosthetic tissues were collected intraoperatively, and total RNA was extracted by standard techniques. Expression of TLR messenger RNAs was assessed by first-strand complementary DNA synthesis from 1 μg of total RNA followed by real-time PCR (polymerase chain reaction). Results were normalized relative to the housekeeping gene GAPDH (glyceraldehyde 3-phosphate dehydrogenase). Expression of TLRs 1, 6, and 10 in the infected and noninfected groups was compared with use of the Student t test. The receiver operating characteristic curve, area under the curve (AUC), sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR-) were calculated to determine the accuracy of each TLR for predicting periprosthetic joint infection at its optimal diagnostic threshold. RESULTS: Mean TLR1 mRNA expression was significantly elevated in infected compared with noninfected samples (0.600 compared with 0.005, p = 0.0003); the same was true of TLR6 (0.208 compared with 0.0165, p = 0.0059) but not of TLR10 (0.00019 compared with 0.00014, p = 0.6238). The AUC was 0.995 for TLR1, 0.883 for TLR6, and 0.546 for TLR10. The optimal threshold for diagnosing periprosthetic joint infection was 0.0924 for TLR1 (sensitivity = 95.2%, specificity = 100%, LR+ = 13.80, LR- = 0.91) and 0.0215 for TLR6 (sensitivity = 85.7%, specificity = 82.8%, LR+ = 4.98, LR- = 0.83). CONCLUSIONS: In our pilot study, TLR1 expression in periprosthetic tissues most accurately predicted periprosthetic joint infection. This measure of the host response may be particularly helpful in detecting culture-negative infections and avoiding false positives resulting from contamination. LEVEL OF EVIDENCE: Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.
Authors: Jacqueline C M Sitt; James F Griffith; Fernand M Lai; Mamie Hui; K H Chiu; Ryan K L Lee; Alex W H Ng; Jason Leung Journal: Eur Radiol Date: 2016-08-23 Impact factor: 5.315
Authors: Luca Maria Sconfienza; Alberto Signore; Victor Cassar-Pullicino; Maria Adriana Cataldo; Olivier Gheysens; Olivier Borens; Andrej Trampuz; Klaus Wörtler; Nicola Petrosillo; Heinz Winkler; Filip M H M Vanhoenacker; Paul C Jutte; Andor W J M Glaudemans Journal: Eur Radiol Date: 2019-06-27 Impact factor: 5.315
Authors: Monique A E Schoeman; Angela E Oostlander; Karien Ede Rooij; Edward R Valstar; Rob G H H Nelissen Journal: J Orthop Res Date: 2017-04-13 Impact factor: 3.494
Authors: Alberto Signore; Luca Maria Sconfienza; Olivier Borens; Andor W J M Glaudemans; Victor Cassar-Pullicino; Andrej Trampuz; Heinz Winkler; Olivier Gheysens; Filip M H M Vanhoenacker; Nicola Petrosillo; Paul C Jutte Journal: Eur J Nucl Med Mol Imaging Date: 2019-01-26 Impact factor: 9.236