Abstract: BACKGROUND: In patients with severe medial compartment osteoarthritis of the knee combined with moderate lateral compartment osteoarthritis, it is controversial whether medial monondylar replacement alone can achieve satisfactory clinical benefit.
OBJECTIVE: To compare and analyze the difference in clinical outcomes of unicompartmental knee arthroplasty and total knee arthroplasty in patients with heterogeneous degenerative changes of the knee (medial compartment Kellgren-Lawrence grade III-IV combined with lateral compartment Kellgren-Lawrence grade II). 
METHODS: Knee arthroplasty patients with severe medial ventricular osteoarthritis combined with mild and moderate lateral ventricular osteoarthritis shown by preoperative knee X-ray were selected and divided into unicompartmental knee arthroplasty group and total knee arthroplasty group according to different operation methods, and 50 patients were included in each group according to 1:1 pairing. Western Ontario and McMaster University Osteoarthritis Index (WOMAC), American Knee Association Score, joint amnesia score, postoperative gait parameters, and complications were collected at 3, 6 months, 1 and 2 years after surgery, and the relevant data were statistically analyzed to compare the differences in postoperative efficacy between the two groups. 
RESULTS AND CONCLUSION: (1) The American Knee Association Score and WOMAC scores of the unicompartmental knee arthroplasty group were better than those of the total knee arthroplasty group at 3, 6 months and 1 year after surgery (P < 0.05). Two years after surgery, American Knee Association Score and WOMAC score were not significantly different between two groups (P > 0.05). (2) There was no significant difference in joint amnesia score between the two groups 3 months after operation (P > 0.05). The joint amnesia score was higher in the unicompartmental knee arthroplasty group than that in the total knee arthroplasty group at 6 months, 1 year and 2 years after surgery (P < 0.05). (3) At 1 year after surgery, step speed and step length were better in the unicompartmental knee arthroplasty group than those in the total knee arthroplasty group (P < 0.05). No statistically significant differences in step frequency were detected between the two groups (P > 0.05). (4) No complications occurred in both groups within 2 years after operation. (5) Follow-up showed that unicompartmental knee arthroplasty was clinically equivalent to total knee arthroplasty in patients with severe medial compartment osteoarthritis (Kellgren-Lawrence III-IV) combined with moderate lateral compartment degeneration (Kellgren-Lawrence II) 2 years after surgery. Unicompartmental knee arthroplasty had a significant biomechanical advantage within 1 year after surgery. No significant progression of lateral interventricular osteoarthritis was observed during follow-up (Kellgren-Lawrence grade increase), but further long-term observation is needed. Before making decisions, doctors should establish a multi-dimensional evaluation framework (including patient age, exercise load expectation, unicompartmental knee arthroplasty operation amount of the surgeon, etc.), and fully inform patients of the risk of lateral compartment progression, and decide whether to perform unicompartmental knee arthroplasty after comprehensive assessment. 

Key words: knee osteoarthritis, lateral intercompartment, unicompartmental knee arthroplasty, total knee arthroplasty, knee joint function, gait, indication, retrospective study