BACKGROUND: Due to the potential reduction of morbidity and mortality, unicompartmental knee arthroplasty (UKA) may represent an interesting solution for older patients with unicompartmental arthritis. It was our hypothesis that UKA can represent an alternative to total knee arthroplasty (TKA) for patients older than 75. We, thus, aimed to compare in those patients (1) functional results, (2) rates of forgotten joint, and (3) survivorships of UKA vs TKA. METHODS: In this retrospective matched-pair study, 101 patients who underwent UKA in our institution were included and then matched one-to-one with TKA group based on age, gender, body mass index, preoperative Knee Society Score (KSS). Inclusion criteria were age between 75 and 90 years on the day of surgery, knee arthroplasty performed for primary osteoarthritis or osteonecrosis of the knee. All patients were evaluated clinically (using KSS, Knee Injury Osteoarthritis Outcome Score [KOOS], and Forgotten Joint Score) at 1, 2, and every 5 years, thereafter. Survivorships of UKA and TKA implants were also compared. RESULTS: At last follow-up, patients from UKA group had better KSS than in TKA group, (respectively, KSS function 82.8 ± 12.2 vs 79.2 ± 13.1 [P = .0448] and KSS knee 88.2 ± 8.9 vs 82.3 ± 12.5 [P = .0005]). Knee Injury Osteoarthritis Outcome Scores were also higher in UKA group (all P \textless .001) as well as the rate of forgotten knees (42% vs 25% P = .01). Sixteen-year survivorships free from revision for any reason were similar in the 2 groups (91.8% vs 94.6% P = .66). CONCLUSION: The results of our study showed that UKA provide higher function and better forgotten joint scores with similar survivorship, compared to TKA, for patients older than 75.