A Series Of Studies On The Diagnosis Of Periprosthetic Joint Infection Based On Preoperative Joint Aspiration

BackgroundJoint aspiration is a key step for detecting periprosthetic joint infections(PJIs)before surgery.Although guidelines have been constantly updated to improve the process and algorithm for detection,there are still unresolved issues that need to be further investigated.For example,in clinical practice,centrifugation is used to remove color interference from erythrocytes in blood-synovial fluid samples before leukocyte esterase(LE)strip testing.However,the impact of centrifugation requires further study.In addition,the use of a saline solution lavage and reaspiration to obtain samples for patients with insufficient synovial fluid(dry tap)remains controversial.There has been no robust evidence to support or oppose the use of this technique.In addition,several factors remain unknown,including the proteomic changes in the synovial fluid of PJI patients compared to that of non-PJI patients and whether better biomarkers exist among the numerous proteins in the synovial fluid.The transformation and optimization of excellent biomarkers into convenient and clinically available diagnostic methods is worth exploring.Thus,to improve and enrich the preoperative diagnosis system for PJI based on joint aspiration,this series of studies aims to investigate the issues mentioned above.MethodsFrom April 2015 to December 2018,the clinical data of patients who underwent preoperative diagnostic aspiration performed by the author were recorded to establish a PJI cohort for this clinical research study.After being used in the clinic,the remaining joint fluid was frozen in the-80 degree freezer as a sample bank for basic research.The diagnosis of PJI was based on the modified Musculoskeletal Infection Society(MSIS)criteria.The clinical research section comprised the following time periods.①From April 2016 to October 2017,133 aspirations were included in this section.In 110 cases,LE strip tests were performed before and after centrifugation.The strips in the other 23 cases could not be read without centrifugation due to color disturbance caused by blood contamination.The results were recorded according to the following different color grades on a color chart:grade 3(++),grade 2(+),and grade 1(others).The results before and after centrifugation were compared,and the corresponding sensitivity,specificity,positive predictive value(PPV),negative predictive value(NPV)and corresponding 95%confidence interval(CI)were calculated.②From April 2015 to August 2018,286 aspirations were included.If more than 1.0 ml synovial fluid was obtained,then we directly used the synovial fluid for culture.If less than 1.0 ml synovial fluid was aspirated,10 ml of saline was injected and reaspirated for culture.The sensitivity,specificity,PPV,NPV and corresponding 95%CI were calculated.③For basic research studies,fifty synovial fluid samples obtained from May 2015 to March 2017 were selected as the discovery phase,and 117 joint fluid samples were prospectively collected from March 2017 to July 2018 as the verification phase.The protein spectrum of the synovial fluid was detected by mass spectrometry,and the data were analyzed.In addition,from August 2018 to January 2019,40 joint fluid samples were prospectively collected to verify the results by using selected reaction monitoring mass spectrum(SRM-MS)analysis.④A lateral flow device to rapidly test a-defensin was developed by using color emulsion-based immunochromatography.The test results from the lateral flow device were compared with those from enzyme-linked immunosorbent assays(ELISAs),and the consistency between the two methods was compared.ResultsA total of 572 joint aspiration procedures were performed by the author,including 376 procedures for patients who underwent joint arthroplasty.After excluding 10 failed aspirations and 25 cases of unclear diagnoses,there were 175 PJI and 166 non-PJI cases.In addition,there were 242 joint fluid samples frozen in the sample bank,including 106 samples from PJI cases,125 samples from non-PJI cases and 11 samples from unclear cases.The clinical research studies were performed as follows:①After centrifugation,almost every sample was lighter in color than before centrifugation.Although most samples showed minor changes and remained in the same grade,18.6%(8/43)and 17.9%(12/67)of the cases were downgraded in the PJI and non-PJI groups,respectively.Before centrifugation,when grade 3(++)was used as the positive threshold,the sensitivity and specificity were 97.7%(86.2%-99.9%)and 100%(94.3%-100%),respectively.After centrifugation,when grades 2 and 3(+and++,respectively)were used as the positive threshold,the sensitivity and specificity were 92.5%(80.9%-97.6%)and 100%(94.3%-100%),respectively.②Saline lavage and reaspiration were used in 82 cases,while direct aspiration was performed in 204 cases.The overall saline lavage ratio was 28.6%(82/286).For procedures involving the knee and hip,the saline lavage ratios were 15.0%(21/140)and 41.8%(61/146),respectively.The sensitivity and specificity of all aspirations were 0.795(0.7200.857)and 0.957(0.909-0.984),respectively.For "dry tap"cases,the sensitivity and specificity were 0.851(0.717-0.938)and 0.857(0.697-0.952),respectively.③Among the different proteins evaluated in PJI and non-PJI cases,myeloid cell nuclear differentiation antigen(MNDA)and proteinase 3(PRTN3)were excellent indicators to diagnosis PJI in basic research studies.Both MNDA and PRTN3 can be stably detected by mass spectrometry in the discovery phase and the validation phase.This result could be further verified by using SRM-MS analysis.When MNDA and PRNT3 were used in combination to diagnose PJI,the area under the receiver operating characteristic curve(AUC)was 0.975.④The test for a-defensin with the lateral flow device could be completed in approximately 20 minutes,and the sensitivity,specificity,PPV and NPV of the device were 0.857,0.933,0.857 and 0.933,respectively.The ELISA for a-defensin had an AUC of 0.969.When considering 15.44 μg/ml as the positive threshold,the sensitivity,specificity,PPV and NPV were 1.000,0.900,0.824 and 1.000,respectively.The two methods were in good agreement with each other(kappa=0.752).ConclusionIn conclusion,these studies found the following:①The influence of centrifugation should be considered when interpreting the LE strip test results.For samples tested without centrifugation,we recommended using grade 3(++)as the positive threshold,while for samples tested using centrifugation,the threshold should be reduced to both grade 3 and grade 2(++and＋).②Using a saline lavage and reaspiration to obtain samples from patients with insufficient synovial fluid before surgery may be a good option.This technique should be reconsidered by surgeons since it may have value for the diagnosis of PJI.③Both MNDA and PRTN3 were outstanding synovial biomarkers in diagnosing PJI and have promising applications in the future.④Although the conditions need to be further improved,the a-defensin lateral flow assay based on immunochromatography could be completed in approximately 20 minutes,and the results were similar to those of ELISA,which shows its potential to be used in clinical practice in the future.Moreover,the later flow assay also lays a foundation for the translation of MNDA and PRTN3 detection.