| Objective Periprosthetic joint infection(PJI)is the most serious complication after artificial joint replacement.Timely diagnosis of PJI and accurate identification of pathogens are the key steps to successful treatment of PJI.At present,the pathogen detection methods in clinical work are in lack of standardization,so it is urgent to compare the diagnostic efficiency of different detection methods.Methods This study retrospectively analyzed artificial joint replacement cases of hip and knee performed in our hospital between December 2018 and January 2020.Demographic parameters were collected on patients who met the inclusion criteria.Patients who met the inclusion criteria were tested for requisite plasma and articular fluid markers and undertaken intraoperative frozen pathology to benchmark the diagnostic criteria of PJI.The 2018 PJI diagnostic criteria of the American Society of Musculoskeletal Infection(MSIS)was used in this study.Patients who met the inclusion criteria were collected intraoperative synovial fluid,periprosthetic tissue and prosthetic components.Then,the periprosthetic tissue were homogenized,and the prosthetic components were vortexed and sonicated.The specimens were inoculated to plate culture on aerobic and anaerobic conditions.Statistical analysis software SPSS 20.0 was used for statistical analysis.The t-test or chi-square test was used to compare the demographic parameters of patients in the PJI group and the aseptic loosening group(non-PJI group).The sensitivity,specificity,positive predictive value,negative predictive value and accuracy of each method were calculated.The detection efficiency of pathogenic microorganisms by different culture methods,the stratified analysis of whether antibiotics were used or not,the detection efficiency of pathogenic microorganisms by different culture methods,and the influence of antibiotics on the detection efficiency of pathogenic microorganisms by different culture methods were compared by Pearson chi-square test,among which,pair comparison was performed by Bonferroni method.Receiver operating characteristic curve(ROC)was used to evaluate the diagnostic efficiency.Kappa test was used for consistency test.Paired-chi-square test was used to compare the positive detection rates between different methods and gold standard.P < 0.05 was considered statistically significant.Finally,the microorganism spectrums cultured under different conditions were compared.Results After reviewing relevant data,87 cases were diagnosed as PJI and 144 cases were diagnosed as non-PJI.There was no difference in age and sex ratio between PJI group and non-PJI group(P=0.387;P = 0.473).The proportion of knee revision in the PJI group was higher than that in the non-PJI group(44.8% vs 26.4%,P=0.004).The first-stage revision rate in the non-PJI group was higher than that in the PJI group(97.2% vs20.7%,P<0.001).Preoperative antibiotic use rate in PJI group was higher than that in nonPJI group(54.0% vs 35.4%,P=0.006).Pathogenic microorganism was detected in 33 cases of synovial fluid group,64 cases of tissue homogenate group and 83 cases of sonication fluid group,respectively.In terms of sensitivity,the sensitivity and 95%confidence interval of the synovial fluid group(32.18%(22.6%-43.1%))were significantly lower than that of the tissue homogenate group(60.92%(49.9%-71.2%)),and were also significantly lower than that of the sonication fluid group(77.01%(66.8%-85.4%))(Pearson chi-square test Bonferroni method: a-b;a-b),while there was no significant difference in sensitivity between the tissue homogenate group and the sonication fluid group(Bonferroni method: b-b).In terms of specificity,there was no significant difference between the tissue homogenate group(92.36%(86.7%-96.1%))and the synovial fluid group(96.53%(92.1%-98.9%)),and there was no significant difference between the tissue homogenate group and the sonication fluid group(88.89%(82.6%-93.5%))(Bonferroni method: ab-a;ab-b),and the specificity of the synovial fluid group was significantly higher than that of the sonication fluid group(Bonferroni method: ab).There was no significant difference in the positive predictive value among the three groups(P=0.299).There was no significant difference in negative predictive value between the tissue homogenate group(79.6%(72.7%-85.5%))and the synovial fluid group(70.2%(63.3%-76.5%)),and there was no significant difference in negative predictive value between the tissue homogenate group and the sonication fluid group(86.5%(79.9%-91.5%))(Bonferroni method: ab-a;ab-b),furthermore,the negative predictive value of sonication fluid group was significantly higher than that of synovial fluid group(Bonferroni method: b-a).In terms of accuracy,there was no significant difference between the tissue homogenate group(80.52%)and the synovial fluid group(72.29%),and there was no significant difference between the tissue homogenate group and the sonication fluid group(84.42%)(Bonferroni method: ab-a;ab-b),furthermore,the accuracy of sonication fluid group was significantly higher than that of the synovial fluid group(Bonferroni method: b-a).In the longitudinal comparison among groups,the sensitivity relationship of the three groups remained as before,regardless of the use of antibiotics or not(Bonferroni method: a-b-b,P=0.000;P=0.001);There was no statistical difference in specificity among three groups with the use of antibiotics or not(P=0.458;P=0.051);There were no significant difference in positive predictive values with the use of antibiotics or not(P=0.828;P=0.330);With the use of antibiotics,the negative predictive values relationship of the three groups remained as before(Bonferroni method: a-abb,P=0.026).Without the use of antibiotics,there was no significant difference in the negative predictive values of the three groups(P=0.050).With the use of antibiotics,the accuracy relationship of the three groups remained as before(Bonferroni method: a-ab-b,P=0.007).Without the use of antibiotics,there was no significant difference in the negative predictive values of the three groups(P=0.229).In horizontal comparison within the group,the specificity,negative predictive value,and accuracy of the synovial fluid group were negatively affected by the use of antibiotics(P=0.034;P=0.001;P= 0.000);The use of antibiotics had negative effects on specificity,negative predictive value and accuracy of tissue homogenate group(P=0.042;P=0.001;P=0.001);The use of antibiotics had no effect on each term of the diagnostic efficiency of sonication fluid group(P>0.05).Using the Receiver Operating Characteristic Curve(ROC Curve)and Area Under Curve(AUC),it was found that whatever under the overall condition or with the use of antibiotics or not,the diagnostic efficiency of synovial fluid was significantly weaker than tissue homogenate and sonication fluid(P < 0.05),and there was no significant difference in the diagnostic efficiency between tissue homogenate and sonication fluid(P>0.05).Under the overall condition or with the use of antibiotics or not,based on the comparison with the gold standard,Kappa consistency test and paired chi-square test found that thefollowing relationship existed in diagnostic consistency and positive detection rate: the combination of three methods ≥ the combination of two methods≥ sonication fluid≥ tissue homogenate ≥ synovial fluid.Compared with the relatively monotonous cultured microbial spectrum of the synovial fluid group,the microbial spectrums of the homogenate group and the sonication fluid group was richer and more various,including a few rare pathogenic microorganisms and mixed microorganism species.Conclusion The culture of inflammatory periprosthetic-tissue homogenate and the sonication fluid of periprosthetic joint components could effectively improve the diagnostic efficiency of PJI based on pathogenic microorganism detection,and the preoperative use of antibiotics has no effect on the diagnostic efficiency of the latter. |
PDF Full Text Request