Dynamic measurement of patello-femoral joint alignment using weight-bearing magnetic resonance imaging (WB-MRI).

Objective: Aim of our work was to compare standard and weight-bearing WB-MRI to define their contribution in unmasking patello-femoral (PF) maltracking and to define what measurement of patellar alignment is the most reliable.

Methods: We prospectively collected 95 non consecutive patients, clinically divided into 2 groups: group A (the control group), including 20 patients (negative for patellar maltracking), and group B including 75 patients (positive for patellar maltracking). The patients underwent a dedicated 0.25 T MRI, in supine and WB position, with knee flexion of 12-15°. The following measurements were performed: Insall-Salvati index (IS), lateral patellar displacement (LPD), lateral patello-femoral angle (LPA) and lateral patellar tilt (LPT). Quantitative and qualitative statistical analyses were performed to compare the results obtained before and after WB-MRI. Measurements were subsequently performed on both groups.

Results: Group A patients showed no statistically significant variations at all measurements both on standard and WB-MRI. On the basis of measurements made on standard MRI, group B patients were divided into group B1 (23 patients) (negative or positive at 1 measurement) and group B2 (52 patients) (positive at 2 or more measurements). After WB-MRI, group B1 patients were divided into group B1a (6 patients), in case they remained positive at 0/1 measurement, and group B1b (17 patients), in case they became positive at 2 or more measurements. All group B2 patients confirmed to be positive at 2 or more measurements at WB-MRI. Quantitative statistical analysis showed that LPT and LPA were the most reproducible and clinically useful measurements. Qualitative statistical analysis performed on standard and WB-MRI demonstrated that LPT was the best predictive measurement.

Conclusions: This study demonstrates both the high diagnostic value of WB-MRI in unmasking PF-maltracking and the best predictive value of LPT measurement.