LAUSR

Study Design: Prospective observational cohort study. Objective: To determine whether biofilms exist on spinal instrumentation recovered during revision surgery in which microbial cultures were negative. Background: Biofilm bacteria are extremely difficult to detect by conventional culture methods used in the standard hospital setting. Chronic infections in which bacteria form biofilms have been demonstrated to slow healing and prevent bony fusion. These slime encased microbial communities serve to isolate the bacteria from the body's immune responses, while simultaneously providing metabolic resistance to antimicrobial therapy. Methods: Traditional debridement wound cultures were taken from each specimen and sent for microbiological analyses. Bacterial DNA testing was performed using polymerase chain reaction (PCR) electrospray ionization-mass spectrometry (ESI-MS). Based on the PCR/ESI-MS results, specific crossed immune electrophoresis was used to detect the bacterial species within biofilms observed on the removed instrumentation. In addition, fluorescent in situ hybridization (FISH) probes corresponding to the bacterial species identified by PCR/ESI-MS were used with confocal microscopy to visualize and confirm the infecting bacteria. Results: Fifteen patients presented for surgical revision of thoracolumbar spinal implantation: four for clinical suspicion of infection, six for adjacent segment disease (ASD), one with ASD and pseudoarthrosis (PA), three with PA, and one for pain. Infections were confirmed with PCR/ESI-MS for all four patients who presented with clinical infection, and for five of the patients for whom infection was not clinically suspected. Of the presumed non-infected implants, 50% demonstrated the presence of infectious biofilms. Half of the revisions due to pseudoarthrosis were shown to harbour biofilms. The revisions that were performed for pain demonstrated robust biofilms but did not grow bacteria on traditional culture media. Conclusions: Culture is inadequate as a diagnostic modality to detect indolent/subclinical biofilm infections of spinal instrumentation. The PCR/ESI-MS results for bacterial detection were confirmed using species-specific microscopic techniques for both bacterial nucleic acids and antigens. Biofilms may contribute to pseudoarthrosis and back pain in postoperative wounds otherwise considered sterile.