Irrigation and Debridement with Chronic Antibiotic Suppression Is as Effective as 2-Stage Exchange in Revision Total Knee Arthroplasty with Extensive Instrumentation

BACKGROUND:Extraction of implants because of periprosthetic infection (PJI) following complex revision total knee arthroplasty (rTKA) with extensive instrumentation is a daunting undertaking for surgeon and patient alike. The purpose of the present study was to evaluate whether infections following complex rTKA are better treated with 2-stage exchange or irrigation and debridement (I&D) with modular component exchange and antibiotic suppression in terms of infection control, reoperation, and function. METHODS:We reviewed rTKAs that had been performed for the treatment of PJI from 2005 to 2016. Extensive instrumentation was defined as the presence ≥1 of the followingmetaphyseal cones/sleeves, distal femoral replacement, periprosthetic fracture instrumentation, or fully cemented stems measuring >75 mm. Cases were categorized according to the initial treatment (I&D with antibiotic suppression or initiation of 2-stage exchange). RESULTS:Eighty-seven patients with PJI and extensive instrumentation were identified56 patients who were managed with I&D with suppression and 31 who were managed with the initiation of 2-stage exchange. The rate of success (defined as no reoperation for infection) was similar for the 2 groups (62.5% the I&D group compared with 67.7% for the 2-stage group; p = 0.62). The rate of mortality was also similar (39.3% for the I&D group compared with 38.7% for the 2-stage group; p = 0.96). Of the 31 patients in the 2-stage group, 18 (58.1%) underwent reimplantation with a revision replacement. Of those 18 patients, 13 were still infection-free at the time of the most recent follow-up; however, when the analysis was expanded to all 31 patients in the 2-stage group, only 13 (41.9%) both had a successful reimplantation and did not require additional surgery for infection. Nine (29.0%) of the 31 patients in the 2-stage group never underwent the second stage, and 4 (12.9%) of the 31 required arthrodesis at the second stage. In contrast, 35 (62.5%) of the 56 patients in the I&D group were successfully managed, without additional surgery for the treatment of infection. At the time of the latest follow-up (mean, 3.2 years; range, 2 to 13 years), more patients in the I&D group were ambulatory (76.8% in the I&D group compared with 54.8% in the 2-stage group; p = 0.05) and maintained a functional bending knee joint (85.7% in the I&D group compared with 45.2% in the 2-stage group; p < 0.001). CONCLUSIONS:In the treatment of periprosthetic infection of rT