Significant increase of pathogen detection rate by dry arthroscopic biopsies at suspected low-grade infection following total knee arthroplasty: a prospective observational study.

Abstract

The differentiation between stiff-knee and low-grade periprosthetic joint infection (PJI) is the current diagnostic challenge in total knee (TKA) revision arthroplasty. The aim of this study was to investigate the additional value of dry biopsies, compared to wet biopsies, in patients presenting with knee stiffness following primary TKA. Single center, prospective observational study. Consecutive patients with joint stiffness of unknown origin following primary TKA were enrolled. Patient assessment followed the diagnostic standard algorithm. During diagnostic arthroscopy, synovial fluid (synovial WBC, PMN%) and five dry biopsies (dry) were collected. Then fluid was infused and another five microbiology (wet) and five histological biopsies gathered, all from identical locations. The primary outcome parameter was the difference between the pathogens in wet and dry biopsies. 71 patients (61% females, 67 ± 10years) were eligible. Preoperative blood serology mean CRP (0.7 ± 1.5mg/dl; p = 0.852), WBC (6.6 ± 1.7 G/l; p = 0.056), and synovial fluid mean WBC (1639 ± 2111; p = 0.602), PMN% (38 ± 28; p = 0.738) did not differ between patients with negative, positive wet or dry biopsies. The histology was in 11% positive (p = 0.058). In 32% at least one pathogen was detected, 48% from wet, 44% from dry biopsies. An inhomogeneous distribution was found. Cutibacterium acnes (100%) was solely found in wet, Micrococcus luteus (75%), Staphylococcus capitis (67%), and Micrococcus lylae (100%) were predominantly found in dry biopsies. Additional dry biopsies increased the pathogen detection rate by 49%. The addition of dry biopsies to the current standard diagnostic algorithm for PJI increased the pathogen detection rate by 49%.