Calprotectin emerges as top plasma biomarker separating acute joint implant infection from aseptic failure

Plasma samples from patients with infected hip or knee implants show a striking surge in calprotectin, a two-part immune protein made of the subunits S100A8 and S100A9, according to a proteomics study of a prospective Danish cohort. Using liquid chromatography coupled to mass spectrometry, researchers quantified 656 plasma proteins and found that the calprotectin heterodimer was the most significantly upregulated protein in acute periprosthetic joint infection (PJI) when compared with two non-infectious complications of joint replacement: aseptic failure and chronic pain after surgery. A targeted blood test for a calprotectin fragment produced by neutrophil elastase, called CPa9-HNE, confirmed the pattern in matched serum samples from the same patients.

The work, published in the journal Arthroplasty, draws on the Prosthetic-Related-Infection and Pain (PRIS) cohort, which enrolled patients in the North Denmark Region between 2011 and 2014. All participants had undergone total knee or hip arthroplasty and were referred because of unexplained pain or mechanical problems with their prosthesis. The study team analyzed plasma from patients classified as having acute infection (n = 18), aseptic failure (n = 41), or chronic pain (n = 18). Patients with chronic infection, culture-negative infection, or undetermined diagnoses were excluded from the main comparisons because their group sizes were too small, meaning the findings apply specifically to acute PJI.

Diagnosing PJI remains one of the more stubborn problems in revision surgery. The infection, driven largely by biofilm-forming bacteria that shield themselves in a protective matrix on implant surfaces, causes pain, functional decline, prolonged antimicrobial therapy, and increased morbidity and mortality. International consensus definitions from the International Consensus Meeting and the European Bone and Joint Infection Society integrate clinical, microbiological, and biochemical findings, yet no single gold-standard molecular test exists. Culture-negative infections and inconclusive cases are frequent, and blood-based markers currently in use, such as C-reactive protein, erythrocyte sedimentation rate, interleukin-6, and fibrinogen, lack disease specificity and can be confounded by systemic inflammation, low-grade infections, prior antibiotics, and biofilm-associated disease.

Distinguishing infection from aseptic failure and from chronic postsurgical pain adds another layer of difficulty because the symptoms overlap. Chronic pain, unlike infection or loosening, does not require revision surgery and is managed without further operations. Misdiagnosis cuts both ways: an unnecessary procedure exposes the patient to surgical risk and a higher likelihood of subsequent infection, while a missed infection can lead to persistent disease, repeated surgeries, and poor outcomes. Synovial fluid markers such as alpha-defensin and calprotectin perform well, but obtaining joint fluid requires an invasive aspiration that can be logistically demanding and may yield insufficient volume.

To map the blood-based protein landscape, the team used a bottom-up proteomic workflow with protein aggregation capture sample preparation, trypsin digestion on an automated liquid-handling system, and analysis on a timsTOF Pro2 mass spectrometer using data-independent acquisition with parallel accumulation-serial fragmentation. Protein identification and quantification were carried out in Spectronaut against the human reference proteome, and differential abundance was tested with Welch’s t-tests corrected for false discovery rate, with significance set at an adjusted p-value below 0.05. The raw analysis identified 675 proteins, of which 656 passed strict filtration for quantification.

The differential abundance analysis placed calprotectin at the top of the list in every infection comparison. In infection versus aseptic failure, both S100A8 and S100A9, the two subunits that form the calprotectin heterodimer, were the most significantly upregulated proteins, with S100A8 the most significant of all; C-reactive protein and the transitional endoplasmic reticulum ATPase VCP were also elevated. In infection versus chronic pain, 25 proteins were overabundant and 36 underabundant in the infection group, with acute-phase proteins including CRP, serum amyloid A1, and serum amyloid A2 joining calprotectin among the upregulated proteins. When the two non-infected groups were pooled into an exploratory “Sterile” group, the calprotectin complex remained the most significantly upregulated protein. Notably, the comparison between chronic pain and aseptic failure revealed no significant protein differences at all.

Functional enrichment analysis reinforced the infection signature. The upregulated proteins in infected patients were dominated by acute inflammatory response, defense response, and neutrophil-related pathways, while the downregulated fractions featured hydrolase activity and reverse cholesterol transport. A principal component analysis showed considerable heterogeneity across the cohort: the first two principal components, explaining 16.7 percent and 6.5 percent of the variance respectively, did not clearly separate the diagnostic groups, although later components began to distinguish chronic pain from infection visually. The authors note that this overlap underscores the need for targeted protein biomarkers rather than reliance on global proteomic patterns.

The confirmation step targeted neutrophil activity rather than protein abundance alone. CPa9-HNE is a competitive ELISA assay measuring a neo-epitope of S100A9 generated when neutrophil elastase cleaves calprotectin; both proteins are released when neutrophils activate. According to prior work cited by the authors, this fragment shows a 101-fold change between activated and inactive neutrophils, compared with only 0.6-fold for full-length calprotectin. In the 77 matched serum samples available from the LC-MS cohort, CPa9-HNE levels were significantly elevated in the infection group compared with both aseptic failure and chronic pain, with no difference between the two non-infected groups, mirroring the mass spectrometry results. A multivariable model adjusting for rheumatoid disease, inflammatory conditions, and biological immunotherapy showed minimal effect of these comorbidities, and sensitivity analyses excluding such patients or using unimputed data left the top findings essentially unchanged.

Calprotectin is already familiar to clinicians as a fecal marker for inflammatory bowel disease and, more recently, as a synovial fluid diagnostic for PJI, where multiple studies report high sensitivity and specificity and where it received a ‘Strong’ level of evidence recommendation at the International Consensus Meeting in 2025. One cited study found blood calprotectin sensitivity and specificity of 93.3 percent and 87.5 percent for PJI, compared with 80 percent and 87.5 percent for blood CRP. The new study extends this picture by showing that calprotectin’s rise is detectable at the systemic, blood plasma level in acute infection, and that its neutrophil-elastase-cleaved fragment carries a distinctive activity signal. Blood sampling offers practical advantages over joint aspiration, which is invasive and can be limited by insufficient fluid volume and postoperative changes in synovial fluid composition.

The authors are careful about the limits of the evidence. This was a hypothesis-generating, exploratory study without an independent external validation cohort, and the small groups, 18 patients each for infection and chronic pain, cannot capture the full heterogeneity of PJI. Clinical information on the microbial profile of infected patients was unavailable, preventing analysis by pathogen type. The authors also disclose commercial affiliations with Nordic Bioscience A/S, the developer and producer of CPa9-HNE, and state that the preliminary findings require independent third-party validation in future cohorts. Because the cohort contained too few chronic infection cases, the conclusions are restricted to acute PJI, and imputation of missing mass spectrometry values may compress variance in small cohorts even though sensitivity analyses suggested a minor effect. Within those boundaries, the study marks, to the authors’ knowledge, the first plasma-wide proteomic profile of PJI and the first use of CPa9-HNE in this context, laying groundwork for blood-derived calprotectin-based diagnostics that could, after clinical validation, help separate infected implants from painful but sterile ones and spare chronic pain patients unnecessary revision surgery.

Subject of Research: Blood-based proteomic biomarkers for diagnosing acute periprosthetic joint infection

Article Title: Calprotectin identified as a key differentially expressed plasma biomarker in periprosthetic joint infection against aseptic failure: a precision health proteomics study

Article References: Andree, S. L., Morsø, L. S., Skallerup, J., Larsen-Ledet, B., Søborg, M. V., Satriano, L., Bay-Jensen, A. C., Rasmussen, S., Thudium, C. F., & Stensballe, A. (2026). Calprotectin identified as a key differentially expressed plasma biomarker in periprosthetic joint infection against aseptic failure: a precision health proteomics study. Arthroplasty, 8(1), Article 67. https://doi.org/10.1186/s42836-026-00414-6

Image Credits: AI Generated

DOI: 10.1186/s42836-026-00414-6

Keywords: periprosthetic joint infection, calprotectin, S100A8, S100A9, CPa9-HNE, proteomics, mass spectrometry, biomarkers, joint arthroplasty, neutrophil activity, aseptic failure, chronic pain