The literature is mixed on the long-term fate of knees that undergo manipulation under anesthesia (MUA). The purpose of this study is to evaluate the long-term outcomes and survivorship of patients who required a MUA after primary total knee arthroplasty (TKA) compared with a cohort of patients who did not undergo a MUA. Between 2003 and 2007, 2,193 patients (2,783 knees) underwent primary TKA with 2-year minimum follow-up; 182 knees (6.5%) had a MUA. Patients who had a manipulation were younger (p < 0.001) and had worse preoperative range of motion (ROM) (p < 0.001). Postoperative ROM, Knee Society clinical (KSC), functional, and pain (KSP) scores, revisions, and survivorship were compared between MUA and no MUA. Mean follow-up was 9.7 years. MUA patients had lower postoperative ROM (p < 0.001), change in ROM (p < 0.001), KSC (p < 0.001), KSP (p < 0.001), and change in KSP scores (0.013). Revisions occurred in 18 knees (9.9%) in the MUA group compared with 77 knees (3%) without a MUA (p < 0.001). Most common reason for revision after MUA was continued stiffness (50%). Relative risk for revision after one MUA was 2.01 (95% confidence interval [CI], 1.1–3.8, p < 0.001) and after three or more MUAs were 27.02 (95% CI, 16.5–44.1, p < 0.001). Ten-year survival after MUA was 89.4% (95% CI, 87.1–91.7%) compared with 97.2% (95% CI, 96.9–97.5%) without a MUA (p < 0.001). Patients who undergo a MUA after primary TKA may have a knee at risk with higher revision rates, worse long-term clinical scores, ROM and survivorship.