A long-term comparison of radiographic and clinical outcomes between 62% and 70% weight-bearing line targets in medial opening wedge high tibial osteotomy.

PURPOSE: The ideal intraoperative correction target for medial opening wedge high tibial osteotomy (MOWHTO) remains debated. This study compared 10-year outcomes between two correction targets, 70% and 62% weight-bearing line (WBL). METHODS: A total of 103 knees that underwent MOWHTO with a locking plate, performed by one surgeon and randomized to a 70% (n = 50) or 62% (n = 53) WBL target, were retrospectively reviewed. Hip-knee-ankle angle and WBL were measured at fixed time points; clinical scores were recorded at 10 years. Conversion to total knee arthroplasty (TKA) was the primary endpoint. Because Kaplan-Meier curves crossed, a landmark, time-varying analysis was used alongside standard tests. RESULTS: Six-month WBL averaged 71.4% (70% group) and 62.2% (62% group). Both groups lost similar correction by 10 years (-5.5 vs. -4.8 percentage points, p = 0.429), leaving 45% of the 62% group undercorrected versus 13% of the 70% group (p < 0.001). Undercorrected knees were converted to TKA more often than normally aligned knees (36.7% vs. 9.6%, p = 0.012). The proportional hazards assumption was violated for the group effect (p = 0.0025); after 10 years, conditional on event-free survival, the hazard ratio for 70% versus 62% was 0.54 (95% confidence interval 0.245-1.182, p = 0.123). Overall survivorship was similar (22.0% vs. 24.5% converted, p = 0.954), but this comparison was underpowered (4.8%) and should not be read as equivalence. Clinical scores did not differ significantly. CONCLUSIONS: Both targets lost similar correction over 10 years, but the 62% target was closer to the undercorrection range, linked to more TKA conversion. Hazards were not proportional over time: The 70% target carried an early, overcorrection-related risk, while the 62% target carried a later, undercorrection-related risk undetectable by an overall Cox comparison alone. A target close to, but not exceeding, 70% may lower the risk of both patterns, pending confirmation with an intermediate target and a larger, adequately powered sample. LEVEL OF EVIDENCE: Level II, retrospective long-term follow-up analysis of a prospectively randomized cohort.