Pain distribution and functional outcomes after lateral unicompartmental versus total knee arthroplasty for isolated lateral compartment osteoarthritis with anterior knee pain

Zhenbao Lu¹Xiaodan Lin²Jianfu Zhu³Xu Wang⁴Xiaohong Fan⁵Jiliang Chen⁴Qiujin Xia³Chengshou Lin^3*Qingshan Xu^3*Qijin Wang^3*

• ¹Department of Orthopedics, Mindong Hospital, Fujian Medical University, Ningde, China
• ²Department of Neurology, Mindong Hospital of Ningde City, Ningde, China
• ³Department of Orthopedics, Mindong Hospital of Ningde City, Ningde, China
• ⁴Department of Orthopedics,, Mindong Hospital of Ningde City, Ningde, China
• ⁵Mindong Hospital of Ningde City, Ningde, China

Background: Isolated lateral compartment osteoarthritis (LCOA) with anterior knee pain (AKP) represents a lateral tibiofemoral–patellofemoral phenotype. Evidence comparing lateral unicompartmental knee arthroplasty (LUKA) and total knee arthroplasty (TKA) in this subgroup is limited for compartment-specific pain. This study compared pain distribution and function when both procedures were performed within a lateral parapatellar pathway. Methods: This retrospective cohort included 115 patients with isolated LCOA and AKP who underwent LUKA (n = 52) or TKA (n = 63) and completed 24 months of follow-up. All cases used a lateral parapatellar approach with patellar denervation and indication-based lateral retinacular release. The primary outcome was the Knee Society Score (KSS) at 24 months. Secondary outcomes were WOMAC, exploratory visual analogue scale (VAS) scores for lateral tibiofemoral (LC-VAS) and patellofemoral pain (PFC-VAS) at 3 and 24 months, perioperative metrics, complications, and reoperation-free survival. Group comparisons used unadjusted tests. For 24-month KSS and WOMAC, exploratory multivariable linear regression was complemented by propensity score weighting (stabilised inverse probability of treatment weighting [IPTW] and overlap weighting, with balance assessed using standardised mean differences [SMD]). Results: Baseline characteristics and preoperative scores were comparable. LUKA was associated with shorter operative time, less blood loss, and shorter hospital stay. At 3 months, KSS was higher and WOMAC lower after LUKA, with lower LC-VAS and similar PFC-VAS. At 24 months, no statistically significant between-group differences were detected in KSS, WOMAC, LC-VAS, and PFC-VAS. Regression estimates were small (KSS adjusted mean difference 0.51, 95% confidence interval (CI) −0.58 to 1.60, and WOMAC −0.62, 95% CI −1.85 to 0.62). Propensity score–weighted estimates were consistent, and 95% CIs remained within published minimal clinically important difference (MCID) thresholds for KSS (5 points) and WOMAC total (10 points). No revisions occurred after LUKA. One infection treated with debridement, antibiotics, and implant retention occurred after TKA. Conclusion: Within this standardised pathway, LUKA enabled faster early recovery and lower lateral pain with less surgical burden, while no statistically significant between-group differences were detected in 24-month pain distribution or function. The data primarily describe early recovery trajectories and pain distribution in this phenotype and warrant multicentre confirmation.