Infected or Not Infected? Next-Generation DNA Sequencing Is Not the Answer...Yet

Commentary Infections of the musculoskeletal system continue to be a vexing challenge for orthopaedic surgeons. Postoperative infections, particularly those in cases in which implants are used for fixation or reconstruction, are common complications in orthopaedic surgery. Metallic implants serve as a nidus for the formation of biofilm, on which bacteria can survive, colonize, and hide from the immune system1. The presence of a biofilm prevents systemic antibiotics from effectively eradicating pathogenic bacteria, leading to high rates of reoperation in which the wounds undergo irrigation, debridement, and implant removal. The infection rates in orthopaedic trauma and arthroplasty cases have been reported to range from 1% to 30%2,3. The diagnosis of a surgical site infection, particularly indolent infection, can be challenging4. The clinical signs can be subtle, often without fever or obvious purulence. Systemic markers are not always reliable, and tissue and fluid bacterial cultures can lack sensitivity5–7. Often, the diagnosis of surgical site infection is made by the treating surgeon on the basis of radiographic findings of implant loosening or new radiolucencies at the bone-implant interface. However, these findings also can be found in cases of aseptic loosening. Therefore, there is a critical need for sensitive and specific assays to identify and characterize pathogenic bacteria in orthopaedic surgical wounds. Assays that can detect bacterial DNA through highly sensitive molecular biology techniques have the potential to improve diagnostic accuracy. Next-generation DNA sequencing (NGS) is under investigation as a tool to diagnose infection by identifying the presence of bacterial DNA (specifically, the gene for 16S ribosomal RNA) and characterizing the bacterial species through DNA sequencing8. The potential advantages of NGS include the ability to identify the presence of bacteria with use of a very small sample. In addition, such techniques do not require the varied culture environments needed to identify specific bacterial species. Point-of-care NGS analysis of musculoskeletal tissue and implant biofilm, if accurate, would be a major advance in orthopaedic care. Natoli et al. report on a prospective study in which patients undergoing “clean” orthopaedic trauma procedures (i.e., procedures for the treatment of closed fractures, hardware removal, or nonunion management) had intraoperative tissue sent for both bacterial culture and NGS. The authors found very little correlation between bacterial cultures and NGS results. Only 3 of 9 positive culture cases were also positive on NGS, with common species in only 2 of the 3. The vast majority of bacteria identified by NGS were either nonpathogenic bacteria or bacteria that are infrequently associated with clinical infection. The authors also found no correlation between the diagnostic assay results and the outcome of clinical infection. Two of 9 patients who had positive cultures developed a clinical infection, only 1 of 27 patients with a positive NGS finding developed an infection, and the NGS speciation in the latter case did not match the culture results of the clinical infection. It should be noted that the study was not designed to follow patients for clinical infection, and therefore some infections may have been missed. It is clear from the data by Natoli et al., as well as data from the shoulder arthroplasty literature9,10 and lower extremity arthroplasty literature11, that NGS should not be considered as a substitute for, or even a complement to, bacterial cultures in clean trauma cases. NGS is extremely sensitive and can identify bacteria that are generally innocuous and clinically unimportant or that present at sufficiently low levels so as to not result in clinical infection. These “false positive” results could lead to overtreatment and increased morbidity. Nevertheless, advances that can address the limitations of conventional bacterial culturing methods are desperately needed. Further work on developing molecular biology techniques that can detect and characterize specific pathogenic bacteria with greater sensitivity than cultures would allow for a more accurate approach12. In addition, NGS can be studied in cases of clinic infection to determine the agreement between culture results showing pathogenic bacteria and specific NGS findings for these pathogens. In the meantime, the study by Natoli et al. takes an important step toward improving our understanding of the clinical utility and the limitations of NGS in the diagnosis of surgical site infections, specifically in orthopaedic trauma surgery.