Researchers found that the natural antimicrobial peptide LL‑37 (and HBD‑3) spikes a lot in joint tissue when there’s a low‑grade infection, making it a very accurate marker for diagnosing infected joint replacements.

Safe diagnosis of periprosthetic joint infection (PJI) is of utmost importance for successful exchange arthroplasty. However, current diagnostic tools show insufficient accuracy in the clinically common and challenging chronic low-grade infections. To close this diagnostic gap, reliable (bio)markers display the most promising candidates. Antimicrobial peptides (AMPs) are part of the innate immune response towards microbial growth. Recently we could show significant intraarticular levels of human cathelicidin LL-37 and β-defensin-3 (HBD-3) with high diagnostic accuracy in PJI synovial fluid. Consequently, these promising biomarkers were evaluated in PJI synovial membrane and synoviocytes, which may significantly facilitate histological diagnosis of PJI to improve outcome of septic joint replacement. In this prospective single-center controlled clinical study (diagnostic level II), consecutive patients with total hip (THR) and knee (TKR) replacements were included undergoing primary arthroplasty (n = 8), surgical revision due to aseptic loosening (n = 9) and septic arthroplasty with coagulase-negative staphylococci (n = 8) according to the criteria of the Musculoskeletal Infection Society (MSIS). Semiquantitative immunohistochemical (IHC) analysis of LL-37, HBD-3 and HBD-2 in synovial membrane and isolated synoviocytes based on Total Allred Score (TS) and Immunoreactive Remmele and Stegner score (IRS) was performed. For statistical analysis, SPSS 26.0/R3.6.3 (p < 0.05) was used. The AMPs LL-37 and HBD-3 were significantly elevated (up to 20×) in synovial membranes from PJI compared to aseptic loosening or primary arthroplasty. The area under the curve (AUC) in a receiver operating characteristic curve analysis was equal to 1.0 for both scores revealing excellent diagnostic accuracy. Isolated synoviocytes as cellular AMP source showed comparable results with a significant LL-37/HBD-3-increase up to 3 × in PJI. In contrast, local HBD-2 levels were negligible (p > 0.23) upon PJI with a lower diagnostic accuracy (AUC = 0.65) in analogy to our previous findings with synovial fluid. Our results implicate AMPs as promising and specific biomarkers for the histological diagnosis of PJI.