The Effect of Knee Rotation Angle on Patellofemoral Instability

Objectives: A lateralized tibial tubercle, as often measured by tibial tubercle–trochlear groove (TT-TG) distance, is a risk factor for patellofemoral instability (PI) and influences surgical management. TT-TG has been shown to be a function of both tibial tubercle position and rotation of the tibia relative to the femur, or knee rotation angle (KRA). However, the role of KRA in PI, as well as its utility in surgical decision making, remains a topic of investigation. The purpose of this study was to determine if a threshold value of KRA could be identified as an independent risk factor for PI. Methods: A retrospective review was conducted of patients with PI who underwent medial patellofemoral ligament reconstruction (MPFLr) at a single institution between 2018 and 2023. A comparison group was patients who underwent isolated partial meniscectomy (IPM) during that same period. Parameters of interest were TT-TG and KRA measured on MRI. Data analysis of patient factors and demographics, along with MRI measurements, was completed with SAS Viya (Cary, NC); significance was alpha <0.05. Results: 122 patients who underwent MPFLr (73 female, 49 male; mean age 21.1 [±9.8] years) and 96 patients who underwent IPM (47 female, 49 male; mean age 38 [±17.6] years) were identified. The average TT-TG for MPFLr and IPM was found to be significantly different at 15.3mm (±5.4) and 11.4mm (±3.5) respectively (p<0.001). The average KRA for MPFLr and IPM was also found to be significantly different at 5.9° (±4.1) and 4.7° (±2.8) respectively (p=0.018). Examining the relationship of TT-TG and KRA, via linear regression, revealed a positive correlation. For every 1mm increase in TT-TG, KRA increased by 0.20° (p=0.034). When examining MPFLr vs IPM via logistic regression, both KRA and TT-TG were significantly associated with MPFLr (p=0.034 and p<0.001 respectively). Additionally, female gender was also significantly associated with MPFLr (p=0.009). Age was significantly associated with MPFLr, for every one year decrease in patient age they were 0.1 times more likely to undergo MPFLr (p<0.001). Further analysis, via decision tree, demonstrated that a KRA>8.07° and TT-TG>15mm had a specificity of 76% and 85% respectively. Additionally, a binary analysis at these values was significantly correlated with MPFLr (p=0.001 KRA and p<0.001 TT-TG). Conclusions: The results of the study demonstrated that KRA is associated with PI even when controlling for TT-TG. Additionally, KRA>8.07 may be an independent risk factor for the development of instability.