Background: Acutely infected total knee arthroplasty (TKA) is commonly treated with debridement, antibiotics, and implant retention (DAIR). There are no direct comparative studies to determine whether debridement should be performed open or arthroscopically for infected TKA. The aim of this study is to compare the outcomes of open vs arthroscopic debridement of infected TKAs. Methods: All patients at a university teaching hospital with an infected TKA treated with DAIR between 2002 and 2017 were analyzed. The primary outcome was successful treatment defined using international consensus criteria. Secondary outcomes included antibiotic suppression, prosthesis retention, mortality, postoperative range of motion, and length of stay. Clinical, laboratory, surgical, and antibiotic treatment data were collected. Propensity score matching was performed to control for selection bias. Results: DAIR was used in 141 patients. The initial DAIR procedure was open for 96 patients and arthroscopic for 45 patients. The success rate was 29% greater for open DAIR (45% open vs 16% arthroscopic; P < .001). After propensity score matching, this benefit was estimated to be 36% (95% confidence interval, 22%-50%; P < .0001). When those on antibiotic suppression were also considered successfully treated, open DAIR was still superior by 34% (95% confidence interval, 18%-51%; P < .0001). Conclusion: For infected TKA, open DAIR is a more successful index procedure compared with arthroscopic DAIR. Open DAIR remained more successful even when antibiotic suppression is considered successful treatment.