Prevalence of unexpected intraoperative cultures (UPIC) in revision total knee arthroplasty (TKA) and risk of periprosthetic joint infection (PJI)

Prevalence of unexpected positive intraoperative cultures (UPIC) in revision total knee arthroplasty (TKA) is reported to be 8.32% with largely varying findings in individual studies (5.92–62.07%) due to a heterogeneity in preoperative diagnostic work-up and diagnostic criteria for periprosthetic joint infection (PJI). When diagnosed as unsuspected PJI, re-revision rate is reported to be as high as 18.45% compared to 2.94% in patients with UPIC classified as contamination (p = 0.0237). We performed a monocentric retrospective analysis of revision TKA surgery to determine prevalence of UPIC, risk of developing a subsequent PJI and re-revision rate after 24 months compared to those without UPIC.

All presumed aseptic total knee revision procedures from 2008 until 2017 were retrospectively reviewed. Single stage femorotibial revisions of TKA with at least 3 intraoperative cultures and a minimum follow-up of 24 months were included, whereas partial revisions or revisions from unicompartimental knee prosthesis to TKA were excluded. Demographic data were collected as well as initial reason for revision, preoperative aspiration results, micro-organism identified and postoperative antibiotics’ administration. Primary outcome was survival rate after 24 months with differentiation based on UPIC interpretation: sterile, contamination, unsuspected PJI.

Of 256 included cases, 75 (29.3%) were found to have at least 1 UPIC during revision surgery, 72% of whom a single UPIC was found. Thirty-four cases (13.28%) were interpreted as unsuspected PJI and therefore treated with antibiotics. In total, 13 of the 256 patients (5.08%) had to be re-revised during follow-up with 7 PJI (2.73%), none of them developing reinfection defined as PJI with the same microorganism. A statistically significant trend towards earlier failure in the unsuspected PJI group compared to the contaminant group was not observed. However, a higher incidence of re-revisions attributed to septic reasons was noted in the former group.

Presence of UPIC in presumed aseptic revision TKA does not correlate with a lower survival rate, but a higher incidence of subsequent PJI was observed in cases interpreted as unsuspected PJI.