Change in Serum Markers Failed to Predict Persistent Infection after Two-stage Exchange ArthroplastyChange in Serum Markers Failed to Predict Persistent Infection after Two-stage Exchange Arthroplasty

Two-stage exchange arthroplasty is a viable choice for prosthetic joint infection (PJI). After removing the infected prosthesis and implanting an antibiotic-loaded spacer in first stage, the proper timing of reimplantation is crucial for successful treatment. So far, there is no gold reference to determine the eradication of PJI before reimplantation. The combination of serum indicators, synovial white cell count (WBC), culture results, intraoperative histology, and clinical symptoms is used extensively to guide the timing of reimplantation. However, the proper timing of reimplantation was ill defined. We wonder: (1) the utility of serum indicators, including erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), interleukin-6 (IL-6) and fibrinogen, when predicted failure of reimplantation; (2) correlation between primary culture results and serum markers’ change.


Introduction
Management of prosthetic joint infection (PJI) is a tough problem for our clinicians, with high prevalence after total joint arthroplasty (TJA) [1,2]. In North America and East Asia, two-stage exchange arthroplasty is wildly applied to chronic PJIs after TJA [3,4]. After removing the infected prosthesis and implanting an antibiotic-loaded spacer in first stage, the proper timing of reimplantation is crucial for successful treatment [5].So far, there is no gold reference to determine the eradication of PJI before reimplantation. The combination of serum indicators, synovial white cell count (WBC), culture results, intraoperative histology, and clinical symptoms is used extensively to guide the timing of reimplantation.
Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are two most acceptable serum indictors in predicting PJI [6][7][8][9][10][11]. But the relativity and utility of both indictors was suspected by several researches. In spite of their doubtful threshold cutoff levels, the normalization of both markers was reported failed to predict control of PJI [5][6][7]11].Several other serum biomarkers were studied by researchers. Interleukin-6 (IL-6) was proved to be useful in diagnose PJI [12][13][14], and Hoell et al reported the good utility of IL-6 in predicting failure of reimplantation [15]. Li et al reported fibrinogen had good performance in diagnosis of PJI [16]. However, more research is needed to determine the accuracy and reliability of these serum indicators in predicting the proper timing of reimplantation.
Instead of a numerical threshold, the 2018 International Consensus Meeting recommended that the down trend in serum markers was viable to determine the proper timing of reimplantation. But opposite result was found by Stambough et al [17]. Stambough et al [17] found the area under the receiver operator curves was 0.530 for percent or delta change in ESR and 0.482 in CRP when predicted persistent PJI, indicating both were poor markers. Considering the value of fibrinogen and IL-6 in diagnosing PJI, change in these serum markers may be viable to predict failure after reimplantation. Besides, instead of percent change, the utility of the value change in serum markers should be investigated.
As contradictory existed whether change in serum markers can guide the timing of reimplantation in two-stage exchange arthroplasty after total joint arthroplasty (TJA), we wonder: (1) the utility of serum indicators, including erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), interleukin-6 (IL-6) and fibrinogen, when predicted failure of reimplantation; (2) correlation between primary culture results and serum markers' change.

Patients
After the Institutional Review Board approval, we retrospectively reviewed all patients underwent two-stage reimplantation between 2014 and 2017 (n=226). All patients were confirmed to be chronic PJI, as acute hematogenous and perioperative infection were excluded [18]. We excluded 7 fungal PJI, 26 patients with potential inflammatory rheumatism disease (rheumatoid arthritis, ankylosing spondylitis, systemic lupus erythematosus, etc), and 30 PJI with hardware fixation, because of uncertain changes in serum indictors in these patients [11,19,20]. 15 patients were excluded for less than the minimum 2-year follow-up. Furthermore, 14 patients who referred to our hospital with antibiotic-loaded cement were excluded for lack of detailed data in serum indicators at the time of resection. 12 participants were excluded because they were satisfied with spacer and refused reimplantation. There are 122 patients left. In this research, MSIS criteria [21] was considered as gold reference to diagnose PJI before resection, and 13 PJIs were excluded without were identified before resection and reimplantation. The threshold values was over 30mm/hr for ESR and 10mg/L for CRP referred to MSIS criteria [21], and the upper limit was 12 pg/mL for and 1.25mg/mL for fibrinogen [16,22] when diagnose PJI. The

Treatment protocol
All patients underwent an institutional standard two-stage exchange arthroplasty, including removal of prosthesis, placement of antibiotic-loaded articulating cement spacer and thoroughly debridement at the time of first stage procedure.
After insertion of cement, all patients accepted 6-8 weeks inter venous antibiotic depends on culture sensitivity reports. For negative result, a broad-spectrum antibiotic therapy was applied. At least 2-week antibiotic holiday was stipulated before reimplantation. During two-stage revision, the antibiotic-loaded cement was removed. 4-6 L sterilized saline water was used to irrigate the joint after thoroughly debridement. Three to five samples were sent to frozen sections, aerobic and anaerobic cultures according to surgeon' suspicion.

Definition of persistent PJI and treatment success
Persistent PJI for this study was defined as :1. Re-infection after reimplantation, based on MSIS criteria, 2. Long-term antibiotic suppression after reimplantation, 3.
Death relating to PJI, 4. Matched MSIS criteria at the time of reimplantation, 5. A pathogen is isolated by culture from at least two separate tissue or fluid samples obtained from the affected prosthetic joint.
We determined treatment success using the Delphi consensus criteria [23,24], which matches the following: 1. a healed wound without fistula, drainage, pain, or infection recurrence caused by the same organism strain; 2. no subsequent surgical intervention for infection after reimplantation surgery; 3. no occurrence of PJIrelated mortality.

Statistical analysis
All of the statistical analyses were performed with the statistical software packages R (http://www.R-project.org, The R Foundation). Categorical data were summarized as absolute values and percentage. Continuous data were presented as median and interquartile range (IQR). The demographic and clinical characteristics between groups were compared with the use of the Student's t-test if they were normally distributed or the Mann-Whitney test if not normally distributed for continuous variables and the chi-square test or Fisher's exact test for categorical variables.
Receiver operating characteristic (ROC) curves were generated to determine the diagnostic value of each test for the assessment of persistent PJI. The area under the curve (AUC) was calculated. A p-value less than 0.05 was considered significant.

General information and patients' follow-up
Patient characteristics and follow-up results were showed in Table 1 The value of ESR, IL-6, CRP and fibrinogen were compared between no re-infected patients and failed patients. The detail of each serum markers was recorded in Table 3. At the time of diagnosing PJI, the re-infected group exhibited significant higher value in all four serum markers. As for reimplantation, higher value of CRP and fibrinogen was observed, while IL-6 and ESR were comparable, indicating that the higher value of CRP and fibrinogen may relate to re-infection. The accurate value of CRP and fibrinogen needed further research.
Does percent change or value change in serum markers guide the timing of reimplantation? Table 5 showed the percent change and value change from resection to reimplantation. No significant difference was found in neither value change of serum markers nor percent change of these. Furthermore, we depicted receiver  Table 5. The re-infected patient showed lower value change and higher percent change of ESR. And subgroup analysis between culture negative PJI and identified organism was exhibited in Table 6. No significant difference was found in neither value change nor percent change of all four serum markers.

Discussion
PJI is still a tough problem after total joint arthroplasty, and two-stage exchange arthroplasty was proved to be the viable treatment for prosthetic joint infection after total joint arthroplasty, with the success rate range from 65% to 100% [25].
The proper timing of reimplantation was crucial to the survivorship of implant.
Considering the unreliability of clinical symptom, lagging of pathology and scarcity of synovial fluid, serum biomarkers still played an important role in predicting the persistent infection after reimplantation. Though several authors failed to determine the threshold of ESR and CRP [6,7,10], the utility of fibrinogen and IL-6 in predicting failure after reimplantation was reported by some articles [13,26].
Our research investigated the value change and the percent change of four common serum markers (ESR, CRP, IL-6 and fibrinogen) between resection and reimplantation. The AUC of ROC was no more than 0.70 in all serum markers, and the combination of serum markers didn't improve the diagnose utility, indicated that neither value change nor percent change of these four serum markers were poor utility in determining reimplantation. Higher value of CRP was observed in both resection and reimplantation, and elevated fibrinogen in reimplantation seemed to be related to increased failure rate. Besides, resistant PJI or culture negative result was not associated with change in serum markers.
We found that preresection CRP and IL-6 was significant higher in failed group.
Besides, the value of fibrinogen and ESR had the tendency to be significance. This is no surprisingly given that the utility and threshold of the four serum markers had been thoroughly studied [16,22,27,28]. However, at the time of reimplantation, values of CRP and fibrinogen were significant higher in re-infection. Kusuma et al [6] reported 76 PJI after total knee arthroplasty, and the AUC of ROC was 0.62 in ESR and 0.39 in CRP. Xu et al [26] also investigated 109 hips who underwent two stage exchange arthroplasty. They found that the value of fibrinogen may be the promising marker in predicting persistent PJI, and the threshold of fibrinogen was 3.61 g/L. Qu et al [29] reported high specificity but low sensitivity when the threshold of IL-6 set at 8.12 pg/ml, and the AUC of ROC was 0.59. The value of serum markers may be useful in predicting re-infection after two-stage exchange arthroplasty, but further study was needed to determine the accurate threshold of them. Besides, other serum markers, like procalcitonin, TNF-ɑ and d-dimer, was reported to be viable by several articles [22,30,31].

Acknowledgments
The authors would like to thank all staff from the participating departments and clinics.

Funding sources
Not applicable.

Availability of data and materials
We do not wish to share our data, because some of the patient's data regarding individual privacy, and according to the policy of our hospital, the data could not be shared with others without permission. Written informed consent was obtained from all participants.

Consent for publication
Not applicable.    Flowchart of patients' characteristics