Haematogenous prosthetic knee infections: Prospective cohort study of 58 patients.

BACKGROUND

Prosthetic joint infection (PJI) is a rare (incidence, 0.15% to 0.9%) but serious complication of knee arthroplasty. Haematogenous PJI of the knee (KhPJI) which accounts for 10% of cases, has been less studied than PJI due to other mechanisms. The primary objective of this study in patients with KhPJI of the knee was to determine the 2-year infection eradication failure rate after either exchange arthroplasty or arthrotomy/synovectomy/irrigation (ASI), combined with prolonged peri-operative antibiotic therapy, at a referral centre for complex osteo-articular infections.

HYPOTHESIS: ASI within 2 weeks after symptom onset and one-stage exchange arthroplasty produce similar 2-year success rates in patients with KhPJI of the knee.

MATERIAL AND

METHODS

A prospective observational cohort study was performed in patients managed for PJI of the knee between 2003 and 2015. The primary outcome measure was the occurrence of a septic event or of KhPJI -related death during a minimum follow-up of 2 years.

RESULTS

Of 265 patients with PJI after total knee arthroplasty, 58 (22.1%) had KhPJI with onset more than 3 months after the last arthroplasty procedure and were included in the study. Among them, one-third had immune deficiencies. The most common causative organisms were streptococci (n=25, 43%) and Staphylococcusaureus (n=20, 34%). The primary focus of infection was identified in only 64% of patients and was most often cutaneous (n=19, 33%) or dental (n=11, 19%). A septic event or KhPJI-related death occurred in 5/34 (15%) patients after one-stage exchange arthroplasty and 6/19 (32%) patients after ASI within 15 days after symptom onset (p=0.03). Patient characteristics, type of prosthesis, and causative organism were not significantly associated with failure to eradicate the infection.

CONCLUSION

ASI carried a high failure rate despite being performed within 15 days after symptom onset. One-stage exchange arthroplasty seems to be the best surgical option, particularly as the exact time of symptom onset may be difficult to determine. Identifying and eradicating the primary focus of infection is crucial.

LEVEL OF EVIDENCE: II, low-powered prospective cohort study.