Synovial fluid calprotectin in diagnosing periprosthetic joint infection: A meta-analysis

Purpose Periprosthetic joint infection (PJI) is one of the most debilitating complications following joint replacement surgery. Synovial biomarkers, such as Calprotectin, have become valuable in the diagnosis of PJI. This meta-analysis aimed to investigate the role of synovial Calprotectin as a diagnostic test in PJI. Methods This meta-analysis was conducted with adherence to PRISMA guidelines. PubMed, Cochrane, Web of Science, and Google Scholar were searched until February 2022. Inclusion criteria were as follows: all studies in which the patients with joint replacements were evaluated for PJI; synovial Calprotectin was the biomarker of choice to diagnose PJI; standardized guidelines were used as the gold standard for the diagnosis; and a comparison between the guidelines and Calprotectin results was made. Diagnostic parameters such as sensitivity, specificity, diagnostic odds ratio (DOR), positive predictive value, negative predictive value, and area under the curve (AUC) were calculated for the included studies to evaluate synovial Calprotectin for PJI diagnosis. Results The total number of the included patients was 618 from eight studies. The pooled sensitivity, specificity, and diagnostic odds ratio of Calprotectin test were 92% (95%CI: 84%-98%), 93% (95%CI: 84%-99%), and 187.61 (95%CI: 20.21–1741.18), respectively. The results showed that the negative and positive likelihood ratios of the Calprotectin test were 0.07 (95%CI: 0.02–0.22) and 9.91 (95%CI: 4.11–23.93), respectively. The SROC showed that the area under the curve for Calprotectin test was 0.935. Conclusion Synovial Calprotectin is a valuable biomarker as it provides a reliable and rapid diagnosis of PJI. It has the potential to be used in clinical practice due to its high sensitivity and specificity that are comparable to the other utilized biomarkers. Another advantage is its low cost relative to other biomarkers.


Introduction
Periprosthetic joint infection (PJI) is defined as an infection of prosthesis and the surrounding soft tissues and is considered one of the most debilitating complications following joint replacementsurgery. PJI contributes to around 14% of all knee and hip revision arthroplasties [1], leading to an enormous healthcare and economic burden that adds up to $1.62 billion in the USA alone [2]. However, the accurate and timely diagnosis of PJI remains quite challenging with wide variation based on the standard adopted guidelines.
Efforts to standardize the diagnosis of PJI yielded various guidelines of pre-operative and intra-operative criteria by the Musculoskeletal Infection Society (MSIS) and International Consensus Meetings (ICMs) [3][4][5]. In 2018, a new validated 1 3 and updated version of the MSIS criteria was defined by Parvizi et al. with a higher sensitivity of 97.7% compared to the original MSIS (79.3%) and ICM definition (86.9%), with a similar specificity of 99.5% [4]. However, the inclusion of microbial cultures in these criteria remains a setback due to their poor reliability (sensitivity and specificity), particularly with low-grade micro-organism infections [6].
While serologic markers such as CRP, D-dimer, ESR have been widely used in the diagnosis of PJI, they are highly influenced by various systemic and confounding factors [7,8]. The emergence of new diagnostic modalities has made synovial biomarkers of particular interest, including synovial WBC, leukocyte esterase, Alpha-Defensin, and Calprotectin, which have shown promising potential as diagnostic tools in PJI.
Calprotectin, also known as cystic fibrosis antigen, is a protein complex mainly secreted by neutrophils as part of the inflammatory response and plays a role in leukocyte migration and stimulation [9]. Different testing methods have also been explored to detect synovial Calprotectin, including enzyme-linked immunosorbent assay (ELISA) and lateral flow testing, showing promising results [10,11]. Several studies have reported the efficacy of synovial Calprotectin in the diagnosis of PJI; however, further understanding of the underlying pathophysiology and diagnostic accuracy is warranted. Therefore, high-quality evidence is needed to highlight the reliability of synovial Calprotectin as a diagnostic tool in PJI.
This meta-analysis aimed to investigate the role of synovial Calprotectin as a diagnostic test in PJI and measure its reliability and validity in terms of sensitivity, specificity, diagnostic odds ratio (DOR), positive predictive value, negative predictive value, and area under the curve (AUC).

Materials and methods
This systematic review and meta-analysis were conducted with strict adherence to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [12]. The focus was studies that compared Calprotectin, as a biomarker to diagnose PJI, with gold standard criteria such as the MSIS and ICM-2018.

Information sources and search strategy
Electronic databases of PubMed, Cochrane, Web of Science, and Google Scholar were searched from inception till February 2022. The following keywords were used: "Periprosthetic joint infection" OR "Prosthesis-related infections" AND "Synovial" AND "Calprotectin." Two independent reviewers screened the titles and abstracts, and the full-text review was done for the eligible studies as per the belowmentioned criteria.

Eligibility criteria
All articles were included if the following criteria were met: Patients with joint replacements being evaluated for PJI. Synovial fluid aspiration was done for PJI diagnosis. Standardized diagnostic criteria, such as MSIS and ICM-2018, were used to diagnose PJI. Calprotectin was used as a biomarker to diagnose PJI. A comparison between Calprotectin and the diagnostic criteria was done. We only included accessible articles that were published in English.

Exclusion criteria
Studies that did not use standardized criteria were excluded. We also excluded studies that did not use Calprotectin among the biomarkers for PJI diagnoses. Patients who had a first-stage revision before being investigated for PJI with Calprotecin were excluded.

Data collection process and data items
We collected the following data items: Author's name, study year, country of origin, age, sex, number of participants, diagnostic criteria, detection method, Calprotectin cutoff point, Calprotectin sensitivity, Calprotectin specificity, Calprotectin positive predictive value, Calprotectin negative predictive value, Area Under the Curve, Calprotectin concentration in septic and aseptic joints.

Risk of bias in individual studies
Two of the authors evaluated the methodological quality of the included studies using the QUADAS-2 tool, which is composed of four key domains; patient selection, index test, reference standard, and flow and timing [13]. Signaling questions were applied to evaluate the risk of bias and clinical applicability. The risk of bias is judged as "low," "high," or "unclear" (when insufficient data are reported to permit a judgment). Any disagreement between the two authors was resolved by a discussion with a senior author.

Statistical analysis
For all the studies, we constructed a 2 × 2 contingency table, then the sensitivity, specificity, diagnostic odds ratio, positive and negative predictive values were 1 3 calculated for each study. Moreover, we pooled the prevalence of the disease in the included studies using a random effect model with double arcsine transformation to calculate the diagnostic parameters that need prevalence to be calculated (PPV and NPV). When more than one threshold was used by any of the included studies, the threshold with the largest Yourdon index was used in the analysis. The mentioned diagnostic parameters were pooled using a random effect model. In addition, the summarized receiver operating characteristic (SROC) curve was constructed using these diagnostic parameters. The heterogeneity of the included studies was investigated using the Cochrane Q and I2 statistic. All the mentioned analyses except the SROC were conducted using Meta XL, version 5.3 (EpiGear International, Queensland, Australia). The SROC was generated using MetaDTA: Diagnostic Test Accuracy Meta-Analysis v2.01 [14].

Study Selection
The search yielded 160 articles, 22 of which were duplicates that were removed manually and electronically. After screening using title/abstract, 125 were excluded. The remaining 13 articles were screened using a full-text form, and five of them were excluded. Finally, eight articles were included in this study. The detailed selection process is described in Fig. 1.

Characteristics of the included studies
A total number of 618 patients from eight studies were included in this meta-analysis. The percentage of septic patients in the included studies according to the ICM or MSIS criteria was 39.5% (244/618) while 60.5% of them Full-text articles assessed for eligibility (n = 13) Full-text articles excluded, due to not matching eligibility criteria (n = 5) Studies included in qualitative synthesis (n = 8) Studies included in quantitative synthesis (meta-analysis) (n=8)  [17] and Grassi et al. [18] used the value of more than or equal 50 mg/l as their cutoff point for positive tests. On the other hand, the cutoff point for positive tests varied between Trotter et al. [10], Zhan et al. [11], and Grzelecki et al. [19], which were > = 14 mg/l, > 173 ug/ml, and > = 1.5 mg/l, respectively. In addition, Warren et al. [20] used two cutoff points for positive Calprotectin tests which were more than or equal 50 mg/l and more than or equal 14 mg/l. The more than or equal 50 mg/l showed higher Youden index value; hence, it was the one that was used in the analysis. Grassi et al. reported all parameters for both ELISA and POC test. However, the POC test they used was a protoyle, and therefore, the parameter calculated for ELISA test was included in the analysis. The characteristics of the included studies are described in Table 1. Figure 2 illustrates the quality assessment of the included studies using QUADAS-2 tool criteria.

Sensitivity
The Calprotectin test sensitivity model included six studies.

Positive and negative likelihood ratio
The positive likelihood ratio model included five studies while the negative likelihood ratio model included six studies. The pooled positive likelihood ratio was 9.91 (

Diagnostic Odds Ratio
The diagnostic odds ratio model for the Calprotectin test included five studies. The pooled diagnostic odds ratio was 187.61 ( Fig. 7

Discussion
This meta-analysis reported an excellent pooled diagnostic value of Calprotectin in the diagnosis of PJI in comparison with gold standard methods. The pooled sensitivity    (Grassi). Higher sensitivity and specificity were reported using ELISA with a higher PLR and a lower NLR. Grzelecki et al. used Immunoturbidimetric Calprotectin immunoassay with a threshold of 1.5 mg/L in diagnosing hip and knee PJI. They reported 95% sensitivity and specificity. This demonstrated that the method used can affect the diagnostic accuracy and lateral flow immunoassay might be inferior to other methods. Moreover, with the available literature, a threshold for diagnosis cannot be determined and more studies are needed. Low cost, availability, and previous utilization for other pathologies are considered advantages for the use of Calprotectin in diagnosing PJI. In comparison with other available biomarkers, Calprotectin showed promising and comparable results. A pooled sensitivity and specificity of Alpha-Defensin of 95% and 96%, respectively, were reported in two recent meta-analyses [21,22]. The meta-analysis by Wyatt et al. [23] reported pooled diagnostic sensitivity and specificity of leukocyte esterase for PJI were 0.81and 0.97, respectively. Furthermore, IL-6 showed a pooled sensitivity of 83% and a pooled specificity of 91% in the meta-analysis by Yoon et al. [24]. Future comparative controlled studies are needed to draw a solid conclusion on the value of Calprotectin in comparison with other available biomarkers in the diagnosis of PJI.
A recently published meta-analysis by Xing et al. investigated the role of Calprotectin in diagnosing PJI. However, this article did not account for the fact that the majority of  the included studies used different cutoff points. Moreover, they did not specify which cutoff point was used when they conducted the analysis on the studies that reported different cutoff points. On the other hand, our study accounted for this limitation by using the Youden Index. This index guided our decision on which cutoff value to use in our analysis. Using Youden to guide us on which cutoff point to use in the analysis explains the variation in the results between our article and Xing's meta-analysis. In addition, this meta-analysis included eight studies with a total number of 618 patients, a 15% larger sample size when compared the Xing's metaanalysis. The larger sample size results in lower standard of error and hence lower confidence intervals across all the analyses. The new study made huge effect on the conference intervals, which is reflected on the reliability of our results.

Limitations
Several limitations should be acknowledged in this study. First, the low number of the included studies hindered our ability to perform sensitivity analysis for different Calprotectin cutoff points or testing techniques. Second, since there is no standard technique or cutoff point for testing synovial fluid Calprotectin, different studies used different techniques and different values, which can impact the diagnostic accuracy. Accordingly, future large-scale prospective randomized trials are required to address these problems. The study by Wouthuyzen-Bakker [17] included some of the patients who were recruited in their previous study in 2017 [16], which might have created some crosspoints in our analysis. However, not all the patients were included and the sequel study included a significant number of patients. Another limitation is the fact that most of the included studies did not mention any information about blinding, sampling point time, and adjustment for confounding variables, which increases the risk of both confounding and selection biases. Finally, our analysis revealed high heterogeneity among the included studies, which can be explained by different cutoff points and Calprotectin testing techniques used by the included studies.

Conclusion
Based on this meta-analysis, Synovial Calprotectin is a reliable and valid biomarker for PJI. It has the potential to be used in clinical practice due to its high sensitivity and specificity that are comparable to the other utilized biomarkers. Another advantage is its low cost relative to other tests. The role of Calprotectin in PJI diagnosis still needs to be elucidated in randomized trials.
Funding Open Access funding provided by the Qatar National Library.
Data availability Not applicable.
Code availability Not applicable.

Declarations
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.

Consent to participate Not applicable.
Consent for publication Not applicable.

Conflicts of Interest
The authors have no financial or proprietary interests in any material discussed in this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.