Pre-revision detection of infection in failed total joint replacements (TJR) is essential to allow appropriate management planning. Unfortunately, low-grade infection is often difficult to detect. The use of molecular biology may offer increased sensitivity in this setting. We have analysed the use of the Polymerase Chain Reaction (PCR) to diagnose infection in pre-operative aspirates in a group of patients undergoing revision arthroplasty. We prospectively tested 50 aspirates in 50 patients with failed TJR (34 hips and 16 knees). Antibiotics were omitted for 2 weeks prior to aspiration. The aspirate was sent for microbiological culture in aerobic and anaerobic conditions. An aliquot was retained for PCR analysis which involved DNA extraction then amplification of an 882 base pair segment of the Universal 16S RNA gene. In 33 patients who subsequently underwent revision arthroplasty multiple specimens were taken from around the joint for microbiological and histological examination and the presence or absence of pus was noted. The patient was deemed to be infected if one of these criteria was found: 2 or more intra-operative cultures positive for the same organism; an acute inflammatory response on histology; pus in the joint at revision ¹.

PCR was positive in 29 cases. Aspiration microbiology was positive in 13 cases. Of the 33 cases revised, 15 patients were deemed to be infected using the previously established criteria, described above. Compared to preoperative aspiration microbiology PCR had a sensitivity of 92% and specificity of 54%. Compared to the published criteria for infection, PCR was 93% sensitive and 61% specific. If rheumatoid cases are excluded the specificity improves to 71%.

It was concluded that PCR has the ability to amplify very small amounts of target DNA. The apparently high false positive rate compared to aspiration microbiology may indicate that PCR is picking up DNA from contaminating or non-viable organisms (treated or phagocytosed), giving poor specificity. However, microbiology is known to have poor sensitivity on pre-operative aspiration samples, and some of the microbiology results may be false negative. Compared to the criteria for infection after revision our results for PCR are more encouraging, especially for non-rheumatoid patients. These patients are part of an ongoing study to identify the most reliable criteria for pre-operative diagnosis of infection in total joint replacement.