Background TKA and unicompartmental knee arthroplasty (UKA) are both utilized to treat unicompartmental knee arthrosis. While some surgeons assume UKA provides better function than TKA, this assumption is based on greater final outcome scores rather than on change in scores and many patients with UKA have higher preoperative scores. Questions/purposes We therefore asked whether TKA would demonstrate (1) better change in clinical outcome scores from preoperative to postoperative states and (2) better survivorship than UKA. Methods We evaluated 4087 patients with 5606 TKAs and 179 patients with 279 UKAs performed between 1978 and 2009. Patients with TKA were older and heavier than patients with UKA (mean age, 68 versus 66 years; mean BMI, 32 versus 29). We compared preoperative, latest postoperative, and change in Knee Society Clinical Rating System (KSCRS), SF-12, and WOMAC scores. Minimum followup was 2 years (UKA: mean, 7 years; range, 2.0-23 years; TKA: mean, 6.5 years; range, 2.0-33 years). Preoperative outcome measure scores (WOMAC, SF-12, KSCRS) were higher in the UKA group. Results Patients with UKA had higher postoperative KSCRS and SF-12 mental scores. Changes in score for all WOMAC domains were similar between groups. Total KSCRS changes in score were similar between groups, although patients with TKA had higher knee scores (49 versus 43) but lower function scores than UKA (21 versus 26). Cumulative revision rate was higher for UKA than for TKA (13% versus 7%). Kaplan-Meier survivorship at 5 and 10 years was 95% and 90%, respectively, for UKA and 98% and 95%, respectively, for TKA. Conclusions While patients with UKA had higher preand postoperative scores than patients with TKA, the changes in scores were similar in both groups and survival appeared higher in patients with TKA.