Topical Tranexamic Acid Use in Knee Periprosthetic Joint Infection Is Safe and Effective

Abstract

Tranexamic acid (TXA) has been shown to decrease hemoglobin loss and reduce the need for transfusions in primary hip and knee arthroplasty. Our study evaluated the safety and efficacy of topical TXA in revision TKA for periprosthetic joint infection (PJI). We performed a retrospective review of patients who underwent removal of hardware with antibiotic spacer placement (stage 1) and/or revision TKA (stage 2) for PJI at our institution between September 2007 and July 2013. During that time, 49 patients underwent stage-1 procedures (20 knees with TXA, 29 without TXA) and 47 patients underwent stage-2 revisions (28 with TXA, 19 without TXA). We evaluated hemoglobin loss, need for transfusion, reinfection rate, length of stay (LOS), complications, and mortality with and without the use of TXA in these patients. All data sets were analyzed with a two-sample t-test. Average follow-up was 3.15 years (range, 1-7 years). TXA use led to a significantly lower percentage drop in the postoperative lowest hemoglobin compared with the preoperative hemoglobin in stage-1 surgeries (19.8 vs. 30.05%, p = 0.0004) and stage-2 revisions (24.5 vs 32.01%, p = 0.01). In both groups, TXA use was associated with a significant reduction in transfusion rates (stage-1, 25 vs 51.7%, p = 0.04; stage-2, 25 vs. 52.6%, p = 0.05). There was a nonstatistical decreased LOS in both groups in which TXA was used (stage 1, 5.15 vs. 6.72 days, p = 0.055; stage 2, 5.21 vs. 6.84 days, p = 0.09). There was no difference in the reinfection rate (4 vs. 4, p = 0.56) or mortality rate between groups (0 vs. 2 non-TXA group). A single upper extremity deep vein thrombosis occurred in a stage-1 patient who received TXA, and no pulmonary embolism occurred. We show that topical TXA is safe and effective for use in both stages of revision TKA for PJI. Previous studies have shown TXA to aggravate a staphylococcal infection in mice; however, topical TXA doesn't appear to negatively effect on the treatment of PJI in our patients and did not increase the reinfection, complication, or mortality rate.