An integrated microfluidic system for rapid isolation and detection of live bacteria in periprosthetic joint infections

Abstract

Periprosthetic joint infection (PJI) is difficult to treat and the incidence is between 1% and 2% in primary arthroplasties. Implant-associated infections usually arise via either primary infections from bacterial invasion at the time of implant surgery or secondary infections from hematogenous sources. The two-stage re-implantation protocol that consists of extensive debridement at the first stage followed by delayed re-implantation is currently the standard process for chronic PJI with a success rate between 82% to 95%. Furthermore, re-implantation arthroplasty should be only performed after ensuring the complete eradication of bacterial infection to avoid devastating complications. However, it is still a challenge in clinical practice to accurately determine the eradication of infections before or during implantation. Conventional diagnostic methods such as measurements of serum C-reactive protein or interleukin-6 levels, culture of joint aspirates, and microscopic examination of tissue biopsy are either non-specific or relatively time-consuming. For critical decision-making before or during the re-implantation surgery, a quick method with high sensitivity and specificity is therefore of great need. Previous studies reported bacterial ribosomal ribonucleic acids (rRNAs) as a target for the diagnosis of infections since rRNAs are highly conserved among bacterial species and abundant in amount. By using universal primers, the presence of bacterial rRNA could be amplified by using reverse-transcription polymerase chain reaction (RT-PCR). Currently the RT-PCR method for detection of bacterial rRNA is highly sensitive with a limit of detection (LOD) as low as a pictogram level. However, RT-PCR signals could only indirectly distinguish live from dead bacteria based on the degradation of rRNA in the tissue. Furthermore, the whole detection procedure of 16s rRNA RT-PCR is labor-intensive. Therefore, an integrated microfluidic system was presented in this work, which could distinguish the existence of live bacteria within 1 hour with a LOD of 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sup> colony formation unit (CFU). In this study, the fabrication of the microfluidic chip was improved so that the consistency of the transported liquid volume was increased. Moreover, by using an ethidium monoazide (EMA) assay, the cumbersome pre-treatment process of rRNA in live bacteria can be alleviated. This is the first time that a microfluidic platform was reported to detect live bacteria successfully in PJI samples.