If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Wenming Zhang, Department of Orthopaedic Surgery, The First Affiliated Hospital of Fujian Medical University, Chazhong Road No. 20, Fuzhou 350005, China, Tel: +8613950391800; Fax: 0591-87983333.
Synovial fluid neutrophil gelatinase-associated lipocalin (SF-NGAL) level in prosthetic joint infection group was significantly higher than aseptic failure group.
•
The SF-NGAL level does not affect by previous antibiotics use.
•
This study shows the potential clinical application of SF-NGAL for prosthetic joint infection diagnosis.
Abstract
Objectives
Neutrophil gelatinase-associated lipocalin (NGAL) in synovial fluid (SF) may have potential diagnostic value for prosthetic joint infection (PJI). This study aimed to study (i) the diagnostic performance of SF-NGAL levels for diagnosing PJI and (ii) the impact of previous antibiotic use on the SF-NGAL levels.
Methods
Consecutive patients who needed hip or knee revision surgery were included prospectively from January 2017 to September 2018. The patients were divided into the PJI group and the aseptic failure group. An enzyme-linked immunosorbent assay was used to determine the SF-NGAL level.
Results
A total of 50 of 78 included patients were diagnosed with PJI. The median SF-NGAL level was 3633 ng/ml (interquartile range [IQR], 1332-10,737) in the PJI group and 26.8 ng/ml (IQR, 12.4-52.5) in the aseptic failure group (P <0.0001). When the SF-NGAL threshold was 263 ng/ml, the area under the curve was 0.98, the sensitivity was 92.9%, and the specificity was 98%. The median level was 5779 ng/ml (IQR, 1425-13,072) in the antibiotic group and 2590 ng/ml (IQR, 932-8970) in the nonantibiotic group (P = 0.1).
Conclusion
SF-NGAL level can be used as a diagnostic indicator of PJI. The use of antibiotics before sampling does not affect the SF-NGAL level.
Prosthetic joint infection (PJI) is a serious complication that is difficult to diagnose after joint arthroplasty and seriously affects the physical and mental health of patients (
). The number of joint arthroplasty surgeries is still increasing rapidly, and the total number of patients with PJI is increasing accordingly. According to existing diagnostic criteria and protocols (
Hanssen AD. Seronegative infections in hip and knee arthroplasty: periprosthetic infections with normal erythrocyte sedimentation rate and C-reactive protein level.
). In particular, the previous use of antimicrobial therapies will significantly decrease the sensitivities of either serum or the synovial fluid (SF) biomarkers (
). Inaccurate diagnosis may lead to incorrect treatment options and poor prognosis.
Current studies suggest that PJI is a relatively limited infectious disease, and the SF biomarkers more directly reflect the nature and extent of inflammation (
). Multiple studies have suggested that the sensitivity and specificity of the SF biomarkers of PJI are superior to those of traditional serum markers (
Hanssen AD. Seronegative infections in hip and knee arthroplasty: periprosthetic infections with normal erythrocyte sedimentation rate and C-reactive protein level.
). Neutrophil gelatinase-associated lipocalin (NGAL) is a member of the antibacterial peptide and lipocalin family. It is one of the proteins secreted by neutrophils and participates in various inflammatory responses.
NGAL has been used for the early identification of bacterial and viral infections in blood and cerebrospinal fluid (
found that the level of NGAL in the SF of patients with PJI significantly increased, and the sensitivity and specificity of the SF-NGAL level for the diagnosis of PJI were 100%.
used chemiluminescence microparticle immunoassay to detect NGAL content in the SF of 44 patients who underwent joint arthroplasty and found that the SF-NGAL level could be used to accurately distinguish PJI from aseptic patients, with a sensitivity and specificity of 86.3% and 77.2%, respectively. NGAL is a potential SF biomarker for the diagnosis of PJI. However, only a few existing studies about this issue are available, and the diagnostic ability of SF-NGAL level has not been fully verified. Moreover, these studies did not explore whether the previous antibiotic treatment affected the level of the SF-NGAL.
We conducted a prospective study using enzyme-linked immunosorbent assay (ELISA) to detect the level of the SF-NGAL in patients undergoing joint revision surgery. The purpose was to (i) study the diagnostic value of SF-NGAL levels for PJI and to compare it with other existing laboratory indicators and (ii) to study the impact of antibiotic use before sampling on the level of the SF-NGAL.
Methods
Study population
This was a prospective study in a single osteoarticular infection referral center, in which consecutive patients who needed revision surgery after hip and knee arthroplasty due to PJI or aseptic prosthesis loosening from January 2017 to September 2018 were included. The study center has 35 beds and the number of treated patients with PJI is about 55 per year. The exclusion criteria were (i) patients without complete clinical and laboratory data for the diagnosis of PJI, (ii) the NGAL test could not be performed because of insufficient volume of the SF (<1 ml), (iii) the SF was visibly mixed with a large amount of blood, and (iv) patients had a recurrent infection. The study was approved by the institutional ethics review committee (2017-093), and all patients included in the study signed written informed consent forms. The International Consensus Meeting criteria (2013) were used to diagnose PJI (
). Patients who met the PJI diagnostic criteria were included in the PJI group, and those who did not meet the criteria were included in the aseptic failure (AF) group. In addition, we divided patients in the PJI group into two subgroups based on the criteria described by Tasukayama as follows: acute infection and chronic infection (
). The patients with PJI were divided into the antibiotic (ABX) group and NO-ABX group according to whether antibiotics were applied within 2 weeks before sampling.
Sample collection
In our center, joint aspiration was considered as a regular diagnostic workup for early diagnosis of PJI when we meet suspected patients with PJI or patients who need revision surgery after hip or knee replacement. Preoperatively, the SF was taken using joint aspiration from the patients we included. The knee joint was directly punctured through the suprapatellar bursa. The hip joint was punctured under ultrasound guidance. If the preoperative puncture failed and PJI was not excluded, the SF was drawn before arthrotomy. We collected 0.5 ml of the SF for the white blood cell count (SF-WBC) and neutrophil classification. A total of 1 ml of the SF was frozen at -80°C for NGAL determination. The remaining SF was used for microbiological culture.
Microbiological culture
A total of 0.1 ml of the SF was directly inoculated into blood plates for bacterial and fungal culture. The remaining SF was injected into Bactec Plus/F aerobic or Bactec Peds Plus/F blood culture bottles and anaerobic blood culture bottles (Becton-Dickinson, Germany). The plates and bottles were incubated for 14 days in a Bactec 9050 automatic incubator (Becton-Dickinson, Germany). The periprosthetic tissue was cut into pieces, added to the broth for grinding, and then cultured for aerobic and anaerobic bacteria on a blood plate for 14 days. The Vitek II system (bioMérieux, USA) was used for microbial identification and antibiotic susceptibility testing.
Analysis of NGAL using ELISA
The frozen SF samples were removed from the -80°C freezer and thawed at room temperature. The samples were centrifuged and diluted appropriately before assaying. NGAL Duo Set ELISA Kits (R & D Company, Minneapolis, Minnesota) were used to determine NGAL levels according to the manufacturer's instructions. The level of NGAL in the SF was measured in duplicate. The SF-NGAL levels did not affect the management of each patient.
Statistical analysis
Statistical analysis was performed using Prism 6.0 (GraphPad, Inc., La Jolla, CA, USA) and SPSS 22.0 statistical software. The chi-square test was used for comparing the categorical variables, and the t-test was used for the continuous data. The nonparametric rank-sum test was used to compare the measured variables of each group (after testing for homogeneity of variance). The data of the SF-NGAL levels, serum C-reactive protein (CRP) levels, erythrocyte sedimentation rate (ESR) levels, the SF-WBC count, and percentage of the SF polymorphonuclear cells (SF-PMN) were plotted in the receiver operating characteristic (ROC) curves to analyze the optimal threshold, the area under the curve (AUC), sensitivity and specificity. P <0.05 was considered statistically significant and the confidence interval (CI) was established at 95%. Spearman correlation analysis was used to analyze related trend variables in SF-NGAL levels, serum CRP levels, ESR levels, and SF-WBC.
Results
General data
A total of 83 patients who underwent hip or knee revision surgery were initially included. Three patients with incomplete clinical data and two patients with insufficient SF volumes were excluded. Finally, 78 patients were enrolled. Among them, 50 patients were included in the PJI group (Appendix Table 1 ) and 28 patients were included in the AF group. There were no significant differences in age (P = 0.137, t = 1.50), sex (P = 0.7, χ2 = 0.191), or involved joint (P = 0.6, χ2 = 0.223) between the two groups (Table 1). The levels of CRP and ESR and the SF-PMN count between the PJI group and the AF group were significantly different (P = 0.0001), as was the SF-WBC count (P = 0.013, t = 2.53).
Statistical data was calculated from chi-squared test. Abbreviations: ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; SF-PMN = percentage of synovial fluid cells that are polymorphonuclear cells; SF-WBC = synovial fluid white blood cell count; N/A= not applicable.
Statistical data was calculated from chi-squared test. Abbreviations: ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; SF-PMN = percentage of synovial fluid cells that are polymorphonuclear cells; SF-WBC = synovial fluid white blood cell count; N/A= not applicable.
Heterogeneity of variance, data expressed as IQR. Statistical data was calculated from t-test.
Note: aStatistical significance.
b Heterogeneity of variance, data expressed as IQR. Statistical data was calculated from t-test.
c Statistical data was calculated from chi-squared test.Abbreviations: ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; SF-PMN = percentage of synovial fluid cells that are polymorphonuclear cells; SF-WBC = synovial fluid white blood cell count; N/A= not applicable.
A total of 40 cases were positively cultured. The most common pathogen was Staphylococcus aureus in 10 (25%) patients, followed by coagulase-negative staphylococci in seven patients, and gram-negative bacilli in seven patients (17.5%), and other pathogens in 16 patients (Appendix Table 1). Of these, two patients were diagnosed with polymicrobial infections.
NGAL levels in the SF in the PJI and AF groups
The median SF-NGAL level in the PJI group was 3633.6 ng/ml (IQR, 1332.4-10,737.2), which was significantly higher than the 26.8 ng/ml (IQR, 12.42-52.53) in the AF group (P <0.0001) (Fig. 1).
Figure 1The levels of SF-NGAL in the PJI and AF groups. The horizontal lines denote median group values with IQRs. The median of SF-NGAL in the PJI group was significantly higher than that in the AF group. Abbreviations: AF, aseptic failure; PJI, prosthetic joint infection; SF-NGAL, synovial fluid neutrophil gelatinase-associated lipocalin.
When the diagnostic threshold was set to 263.5 ng/ml, the SF-NGAL level had the best sensitivity and specificity. The sensitivity of the SF-NGAL level for PJI was 92.9%, and the specificity was 98%. The AUC of the SF-NGAL levels was 0.98, which was similar to that of the SF-WBC (0.99). The AUC of the SF-PMN, CRP levels, or ESR levels were 0.96, 0.89, and 0.90, respectively (Fig. 2). When the thresholds of CRP levels, ESR levels, SF-WBC, SF-WBC, and SF-PMN were set to 8.7 mg/l, 35.5 mm/h, 3005/μl, and 57.6%, respectively, this panel of indicators had the best diagnostic value with the best sensitivity and specificity (Table 2).
Figure 2The ROC curves for diagnostic performance comparison among SF-NGAL and SF-WBC, SF-PMN, CRP and ESR. Abbreviations: AUC, area under the curve; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; ROC, receiver operating characteristic; SF-NGAL, synovial fluid neutrophil gelatinase-associated lipocalin; SF-PMN, percentage of synovial fluid polymorphonuclear cells; SF-WBC, synovial fluid white blood cell count.
Comparison of the laboratory indicators between the ABX and NO-ABX groups
In the PJI group, 25 (50%) patients had a history of antibiotic use within 2 weeks before sampling (Table 3). The most commonly used antibiotics were vancomycin, third-generation cephalosporins, and quinolones in ABX group. The median level of the SF-NGAL was 5779.6 ng/ml (IQR, 1425.7-13,072) and 2590.7 ng/ml (IQR, 932.9-8970.8) in the ABX group and in the NO-ABX group, respectively (P = 0.1). The SF-WBC and CRP levels were lower in the ABX group than in the NO-ABX groups (P values of 0.047 and 0.04, respectively).
Table 3Comparison of laboratory indicators for the ABX and NO-ABX group.
In this study, the SF-NGAL levels were positively correlated with these inflammation-related indicators (Table 4). The SF-NGAL level was moderately correlated with the SF-WBC (rho = 0.64, P = 0.0001), weakly correlated with the SF-PMN (rho = 0.152, P = 0.293), and moderately correlated with serum indicators CRP levels (rho = 0.303, P = 0.032) and ESR (rho = 0.222, P = 0.121). In addition, CRP levels was strongly correlated with ESR (rho = 0.549, P = 0.0001), and the SF-WBC was weakly and nonsignificantly correlated with the SF-PMN (rho = 0.195, P = 0.176).
Table 4Correlation analysis of laboratory indicators.
Subgroups analysis of potential factors in SF-NGAL
We concluded several potential factors which might influence the value of SF-NGAL (Appendix Tables 1, 2). In the PJI group, 30 (60%) patients had a history of nonsteroidal anti-inflammatory drug use within 4 weeks before sampling and near half of the patients had at least one concomitant disease, such as obesity (n = 18), diabetes (n = 14), and hypertension (n = 28). However, compared with the value of SF-NGAL in subgroups, there was no significant difference in whether a patient had a smoking habit or had a concomitant disease. We also calculated the similar results in the AF group (Table 5).
Table 5Subgroups analysis of potential factors in SF-NGAL
Statistical data was calculate from t-test. Abbreviations: NSAIDs, Nonsteroidal anti-inflammatory drugs; SF-NGAL, synovial fluid neutrophil gelatinase-associated lipocalin.
Smoking history
Yes
6113±7402
0.283
Yes
101±249
0.627
No
8580±7035
No
191±480
Prior NSAIDs
Yes
6681±8105
0.931
Yes
36±39
0.114
No
6881±6882
No
231±462
Obesity
Yes
5684±6600
0.181
Yes
96±271
0.572
No
8794±8261
No
175±403
Hypertension
Yes
5342±6272
0.201
Yes
43±47
0.118
No
7952±7963
No
238±482
Diabetes
Yes
7159±7817
0.544
Yes
107±320
0.679
No
5890±5993
No
163±347
a Heterogeneity of variance, data expressed as mean value ± SD.
b Statistical data was calculate from t-test.Abbreviations: NSAIDs, Nonsteroidal anti-inflammatory drugs; SF-NGAL, synovial fluid neutrophil gelatinase-associated lipocalin.
). SF biomarkers are proteins or peptides that can be tested conveniently and stably with convenient methods in clinical practice. This has significance for the rapid and accurate diagnosis of PJI (
). To that end, we evaluated the diagnostic value of this marker for PJI and provided a novel test for the rapid diagnosis of PJI methods. Our results reveal that NGAL is a potential SF biomarker for diagnosing PJI that can accurately distinguish PJI from AF.
This prospective study showed that the median level of the SF-NGAL in the PJI group, determined using ELISA, was 135-fold greater than that in the AF group, and there was a very narrow overlap between the two groups. When the threshold of the SF-NGAL was set to 263.5 ng/ml, the SF-NGAL level had good diagnostic accuracy. The AUC of the ROC curve was 0.98, with a sensitivity and specificity of 92% and 98%, respectively. This is similar to the result of
, who studied 16 SF biomarkers in 95 patients (29 with PJI and 66 with AF). Of these biomarkers, NGAL was the one of the five best markers, with 100% sensitivity and specificity.
measured NGAL in SF using chemiluminescence microparticle immunoassay and showed that the diagnostic sensitivity and specificity were 86.3% and 77.2%, respectively, which were slightly inferior to those in the current study. The median and threshold values of SF-NGAL in this current study are different from those of the studies by Deirmengian et al. (2200 ng/ml) and Vergara et al. (1536.5 ng/ml). The possible reasons may be different detection reagents or methods and heterogeneity of patients selected.
The current study also demonstrated that the sensitivity of SF-NGAL is higher than that of serum CRP levels or ESR levels. Moreover, the specificity of SF-NGAL is also high. This is because PJI is mainly confined to the involved joints, and the local immune response may act more strongly than the systemic immune response. Moreover, the local immune response is not affected by the inflammatory status at other body sites, such as pneumonia and odontitis. Therefore, SF biomarkers should potentially have higher sensitivity and specificity than those of serum biomarkers.
Most general physicians in primary hospitals often treat patients with suspected PJI with antibiotics, without correct and prompt aspiration. This will inhibit the pathogenic activity in the joints, affecting the microbial culture results and weakening the host's inflammatory response to these pathogens, reducing serum inflammation marker levels and even the SF-WBC count (
), and seriously interfering with the accuracy of PJI diagnosis. Our prospective real-world study addressed this problem by including all patients regardless of whether antibiotics had been previously administered. The results reveal that the level of the SF-NGAL is not affected by the previous use of antibiotics. Conversely, the levels of CRP and the SF-WBC in patients in the ABX group were significantly lower than those in the NO-ABX group. This novel potential diagnostic indicator may not require stopping antibiotics before sampling or surgery. Thus, it will shorten the waiting period and be more suitable for clinical practice.
NGAL is mainly present in the peroxidase granules of neutrophils, the level of which might highly depend on the number of neutrophils (
). Our study indeed showed that the SF-NGAL level was moderately positively correlated with the SF-WBC and the SF-PMN levels. However, the sensitivity of the SF-NGAL level was slightly higher than that of the SF-WBC or the SF-PMN. NGAL is an active antipathogenic factor released by neutrophils during inflammation, making its local level higher than that in no-infectious tissue. The commonly used automatic cytological counting requires at least 1-3 ml of the SF to process, whereas the ELISA test of NGAL only requires 0.5 ml. In some patients, there may not be enough SF (
Our study had some limitations. First, we should note the heterogeneity of the population studied and the relatively small sample size. Second, there is still no real gold standard for diagnosing PJI, and the definition we used may give rise to misdiagnosis. Finally, the influence of inflammatory joint disease on the diagnostic efficacy of the SF-NGAL level was not investigated because too few of these patients were included to conduct a powerful analysis.
In total, NGAL in the SF has both high sensitivity and specificity for the PJI diagnostics. The previous use of antibiotics does not affect the level of the SF-NGAL. The SF-NGAL can be used as a potential laboratory indicator for diagnosing PJI.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Funding statement
This work was supported by Fujian Education and Scientific Research Projects for Young Teachers (grant number JAT170241) National Science Foundation for Young Scientists of China (grant number 81702168) and Natural Science Foundation of Fujian Province (grant number 2018I0006, 2018Y4003, 2019I0011). Funds were used for laboratory testing fee and research coordinator salary.
Ethics approval and consent to participate
This study was approved by the institutional review board issued on January 2, 2017, with process-number 2017-093.
Author contributions
Z.H., Z.Z. and M.L.: Designing the study, performing the test, collecting and analyzing the data, and writing the manuscript. W.L.: Writing and editing the manuscript. Z.Z., and X.F.: Analyzing the data and editing the manuscript. Z.Z.: Performing the microbiological culture and editing the manuscript. W.Z.: Designing and supervising the study and editing the manuscript. The authors have seen and agreed to the submitted version of the paper. We state that if accepted, the paper will not be published elsewhere in the same form, in English or in any other language, without written consent of the copyright holder.
Hanssen AD. Seronegative infections in hip and knee arthroplasty: periprosthetic infections with normal erythrocyte sedimentation rate and C-reactive protein level.
00474-X&id=ga1.jpg){kind=link}
00474-X&id=gr1.jpg){kind=link}
00474-X&id=gr2.jpg){kind=link}