Periprosthetic joint infection (PJI) is a devastating complication that can occur after total joint arthroplasty (TJA). It often requires long-term antibiotic therapy and multiple revision surgeries, and even with optimal treatment, it can significantly decrease a patient’s quality of life. Additionally, the risk of mortality is high for those affected by PJI (Wildeman et al., 2021). In addition to the significant impact on individual patients, the socioeconomic burden of PJI is substantial and expected to increase in the future (Peel et al., 2013) The one-year and five-year risk of PJI after total hip arthroplasty (THA) is 0.7% and 1.1%, respectively, while for total knee arthroplasty (TKA), the corresponding figures are 0.7% and 1.4% (Kurtz et al., 2018). The overall one-year and five-year survival rates after PJI diagnosis were 88.7% and 67.2% for THA and 91.7% and 71.7% for TKA, respectively (Kurtz et al., 2018). In Germany, the total number of total joint arthroplasty (TJA) procedures is projected to increase by 45% for TKA and 23% for THA between now and 2040 (Rupp et al., 2020). Similarly, in the US, the incidence rate of primary TKAs is projected to increase by 43% from 2020 to 2050 (Klug et al., 2021). Among these revision surgeries, the biggest share is performed due to PJI (25%), followed by mechanical loosening (19%) as the second most common reason for TKA revisions (Bozic et al., 2015).

The treatment of infected and aseptic failures after TJA differs significantly and can impose a significant burden on patients (Andersson et al., 2010). Therefore, it is essential to avoid treating a non-infected joint as an infected one, and vice versa, as this can lead to increased morbidity, unnecessary costs, and avoidable surgical interventions (Moojen et al., 2014). Accurate diagnostics are thus of paramount importance in clinical practice to ensure appropriate treatment and avoid these adverse outcomes (Kurtz et al., 2022). In some cases, diagnosing PJI is straightforward, as clear clinical findings such as the presence of a sinus tract or pus around the implanted prosthesis are considered confirming diagnostic criteria (Parvizi et al., 2018; McNally et al., 2021). However, in many cases, these confirming criteria are not present, making PJI diagnostics challenging (Balato et al., 2020). Diagnosis typically relies on laboratory tests such as serology or synovial fluid analysis, microbiological analysis of tissue specimens or synovial fluid, as well as histological and radiographic findings. In recent years, efforts have been made to improve diagnostic accuracy. In 2011, the Musculoskeletal Infection Society (MSIS) published PJI criteria that classify “major” criteria, including the presence of a communicating sinus tract and two positive periprosthetic cultures, and “minor” criteria, such as elevated ESR/CRP, elevated synovial leukocyte count, elevated synovial polymorphonuclear (PMN)%, purulent material, isolated organism in one culture, and intraoperative frozen sections with histology (Parvizi et al., 2011). In 2013, the Infectious Diseases Society of America (IDSA) provided its own PJI diagnostic criteria with the aim of standardizing diagnostics (Osmon et al., 2013). Unlike the MSIS criteria, the IDSA criteria do not include elevated inflammatory markers but consider other factors such as the growth of a virulent organism from a single culture or the presence of acute inflammation from histopathology of the periprosthetic tissue. In 2013, the International Consensus Meeting (ICM) introduced a new minor criterion - the leukocyte esterase in synovial fluid measured by a urine strip test. Later, Parvizi et al. updated the ICM concept by introducing a scoring system based on the different sensitivity and specificity of the markers in 2018 (Parvizi et al., 2018). This updated system included promising new markers such as alpha-defensin in synovial fluid and D-dimer in serum. In 2021, the European Bone and Joint Infection Society (EBJIS) criteria were introduced, classifying cases as “unlikely,” “likely,” and “confirmed” All these criteria rely on various clinical, laboratory, microbiological, and histological analyses, as well as intraoperative findings, to establish a diagnosis (see Table 1 ).

Despite the improvements made in recent decades that have made correct diagnosis of PJI more likely through the introduction of different PJI criteria, the identification of a novel biomarker that is highly specific and sensitive for PJI could enable easier and more accurate diagnosis of this devastating disease, ultimately improving patient outcomes. Therefore, the objective of this systematic review is to identify and evaluate novel biomarkers for preoperative PJI diagnostics.

Over the last five years, 19 studies have reported on new biomarkers for PJI, with 15 of these studies specifically focused on parameters in the synovial fluid. Most of the studies analyzed proteins (nine studies), followed by molecules (three studies), exosomes (two studies), DNA (two studies), interleukins (one study), and lysosomes (one study). Calprotectin is a promising and frequently analyzed marker ( Table 1 ) (Warren et al., 2021; Grassi et al., 2022; Warren et al., 2022). In scenarios with a threshold of ≥50 mg/mL, the calprotectin point-of-care performance showed a high sensitivity of 98.1% and specificity of 95.7% ( Figures 2 , 3 ). LCN2 is another hopeful marker that nearly perfectly discriminates between controls and confirmed infections in a small cohort of patients (72 patients/22 PJI) ( Figures 2 , 3 ) (Vergara et al., 2019). D-lactate, which has been analyzed in a large cohort of patients, is also noteworthy, revealing 94% sensitivity and 78% specificity ( Figures 2 , 3 ) (Yermak et al., 2019; Karbysheva et al., 2020). The sensitivity and specificity of these markers are comparable to those of established markers. A review by Sigmund et al. presented the performance of established and novel serum inflammatory biomarkers. The sensitivity and specificity of established markers such as erythrocyte sedimentation rate (ESR) or white blood cell count (WBC) demonstrated similar sensitivity and specificity in comparison to new markers. C-reactive protein (CRP) with a cut-off of 3-32mg/L showed a sensitivity of 62-100% and specificity of 64-96%, while procalcitonin demonstrated a maximum sensitivity of 90% and specificity of 100% (Sigmund et al., 2021).

More than two-thirds of the studies analyzed biomarkers from synovial fluid, but it’s important to note that diagnostic hip aspirations are unsuccessful in up to one-third of patients (Christensen et al., 2022). Five of the 19 studies analyzed only preoperative aspirates. Furthermore, there is a discordance of approximately 20% between preoperative aspirate culture and intraoperative synovial fluid culture, which makes relying solely on synovial fluid in the preoperative setting for diagnosing PJI challenging (Li et al., 2021). A meta-analysis of 14 studies that pooled preoperative aspiration culture data revealed an average sensitivity of 67.6% (range 28% to 100%) and an average specificity of 98.4% (range 96% to 100%) (Rodriguez-Merchan, 2018).

Inflammation triggers a series of signaling cascades, and different markers investigated in PJI are linked to these cascades, either up or down. For instance, calprotectin is secreted by neutrophils (Stríz and Trebichavský, 2004) which play a vital role in PJI diagnostics as PMN%. Likewise, the measurement of calprotectin in synovial fluid is significantly associated with PMN% and is an important marker for diagnosing PJI (Burri et al., 2013; Lisowska-Myjak et al., 2016; Honar et al., 2022). Similarly, D-lactate concentration is linked to microbial load. The concentration of D-lactate seems to depend on the number of bacteria, as higher levels of D-lactate were observed in culture-positive PJI compared to culture-negative PJI (Yermak et al., 2019). Given the interdependence of markers in the inflammation signaling cascades, it is not surprising that relying on a single new marker alone may not revolutionize PJI diagnostics.

Alpha defensin, which is a diagnostic marker included in the ICM 2018 criteria ( Table 1 ), was initially hailed as a game-changing diagnostic marker. However, as it became known that alpha defensin is released by neutrophilic granulocytes and acts as part of the non-specific immune system, it was not surprising that the hoped-for diagnostic breakthrough was followed by disappointment. In the literature, the sensitivity of alpha-defensin for the diagnosis of PJI has been reported to range from 67% to 100%, and the specificity from 89% to 99% (Kasparek et al., 2016; Wyatt et al., 2016; Balato et al., 2020). Renz et al. calculated a sensitivity of 84% using the MSIS criteria, 67% using the IDSA criteria, and 54% using the PRO-IMPLANT/EBJIS criteria (Renz et al., 2018). Such variation according to the various criteria presents a challenge in clinical application. Therefore, the routine use of alpha-defensin testing is not recommended in the literature and should only be performed as an additional diagnostic test. Costs also have to be taken into account when using alpha-defensin for diagnostics. ELISA for alpha-defensin is much more expensive than the leukocyte esterase test strip (£0.11 [US$0.17] per test), costing around £500 [US$760] per test (Wyatt et al., 2016).

Besides cost, availability is a major concern for implementing new biomarkers in clinical practice. Established markers like CRP and synovial WBC are commonly used in medical centers, whereas newer markers like alpha defensin are rarely used for routine diagnosis of PJI. Alpha defensin is only collected in 19.4% of cases, while microbiological (97.7%), leukocyte count (74.8%), and PMN% (65.8%) are the most frequently measured parameters in diagnostic setting (Ahmad et al., 2016). Furthermore, storage of specimens poses a challenge as certain markers, such as cytokines and s-Pecam1, require specific temperatures during transportation, which can complicate logistics. As a result, introducing these markers in clinical practice can be difficult.

Modern genomic sequencing diagnostics may offer a solution to the challenges associated with biomarkers and conventional microbial diagnostics. While culture-based detection methods remain the gold standard, they are plagued by several limitations, including low sensitivity. Microbiological culture only detects the pathogen in 44-80% of cases (Malhotra and Morgan, 2004; Williams et al., 2004). One major factor that significantly affects the probability of detecting a pathogen through culture-based methods is the duration of the culture (Saleh et al., 2003; Schäfer et al., 2008). Additionally, contamination and resulting false positive findings can also be problematic (Yee et al., 2013). To overcome these limitations, culture-independent, molecular biology-based methods can be employed as an alternative diagnostic tool. In particular, plasmatic detection of circulating free DNA through Next Generation Sequencing (NGS) has shown promise as a diagnostic method for patients with bloodstream infections. Metagenomic NGS (mNGS) offers the ability to identify multiple organisms in a single sample (Gu et al., 2019). Early studies have suggested that NGS-based diagnostics are more effective than conventional culture-based methods for detecting bloodstream infections (Grumaz et al., 2016; Decker et al., 2017; Grumaz et al., 2019). In the case of PJI, Echeverria et al. identified the pathogen in 35 cases, including four cases that were deemed culture-negative (57%) (Echeverria et al., 2021). Having a pathogenic marker such as circulating free DNA could be beneficial as it specifically identifies present bacteria compared to nonspecific markers. Thus, NGS could be utilized to identify the pathogen in cases where culture-based methods are ineffective.

Several limitations of this systematic review must be acknowledged. First, the study compared three different PJI criteria, which are the most commonly used ones. The MSIS and ICM criteria were used in six studies. Sigmund et al. conducted a retrospective study of 206 PJI patients, of which 101 (49%) were diagnosed with PJI using the EBJIS definition, 99 (48%) with the IDSA definition, and 86 (42%) with the ICM definition. A total of 84 cases (41%) had an infection based on all three criteria. The novel EBJIS definition appears to be more sensitive for PJI diagnosis compared to the IDSA and ICM definitions. All infections classified by the IDSA or ICM criteria were identified by the EBJIS definition, indicating that the EBJIS definition is superior to the IDSA and ICM criteria for PJI diagnostics. However, only two studies in this systematic review used the EBJIS definition, which was introduced recently in 2021 (Sigmund et al., 2022). However, the present systematic review is limited by several factors. Secondly, only two studies in our review employed the recently introduced EBJIS definition, which limits the generalizability of our findings. Thirdly, the use of different cut-off values for biomarkers across studies makes comparison challenging. Finally, the limited availability of information on the time elapsed since the arthroplasty procedure may affect the accuracy of some biomarkers, as their diagnostic performance may vary in the early postoperative period (Yi et al., 2014).

Based on the current analysis, no novel biomarker investigated in the past five years for diagnosing PJI has been proven to outperform the already established diagnostic parameters. Further studies may demonstrate the usefulness of additional markers, such as calprotectin, in the established PJI diagnostic criteria.

SM, NW, and MR contributed to conception and design of the study. SM and NW organized the database. SM and NW performed the statistical analysis. SM wrote the first draft of the manuscript. NW and MR wrote sections of the manuscript. All authors contributed to the article and approved the submitted version.