Patellar maltracking with dynamic MRI: clinical feasibility and diagnostic utility

To evaluate (1) the clinical feasibility of adding a dynamic MRI sequence during isometric quadriceps contraction to a routine knee MRI protocol and (2) the association between contraction-induced changes and clinically diagnosed patellar maltracking, including whether this information adds discriminative value beyond static imaging.

This retrospective case-control study included all patellar maltracking patients (cases) and healthy volunteers (controls) undergoing knee MRI supplemented with dynamic series at a single tertiary referral center between 2021 and 2025. Exclusion criteria included prior knee surgery, severe imaging artifacts, or incomplete exams. Image quality, adverse events, and acquisition data were collected. Two musculoskeletal radiologists independently performed patellofemoral measurements on the static and dynamic series. Statistical analysis included t-tests, age/sex-adjusted regression, ICC(2,1) for reliability, and AUC/DeLong analysis for incremental diagnostic value.

One hundred thirteen patients and 20 controls underwent knee MRI with dynamic series. Image quality was rated "good" or "excellent" in 128/133 (96.2%) subjects. No protocol-related adverse events were recorded. In both cohorts, patellar height, lateral translation, and tilt were greater on dynamic than static series (p < 0.05). There was a larger increase in TT-TG distance between static and dynamic images in patients with patellar maltracking compared to controls (4.35 mm vs 1.58 mm, p = 0.003). Inter-reader ICC(2,1) estimates ranged from 0.36 (95% CI, 0.20-0.50) to 0.93 (95% CI, 0.91-0.95).

The supplementary dynamic MRI sequence performed during quadriceps contraction is clinically feasible. The contraction-induced increase in TT-TG distance was greater in maltracking patients than controls, though its incremental value beyond static imaging was not statistically confirmed in this sample.