Variability in Patellofemoral Alignment Measurements on MRI: Influence of Knee Position.

Abstract

The purpose of this study was to determine the variability of distances between four distinct anatomic landmarks used for the evaluation of patellofemoral alignment while implementing controlled changes in patient positioning during MRI. Limited MRI was performed of 12 knees in healthy volunteers (10 men, two women; mean age, 28 years) with the knees in four different positions. The four landmarks used were TT (the most anterior point of the osseous tibial tubercle), TG (the nadir of the cartilaginous trochlear groove), PT (the center of the patellar tendon insertion on the tibia), and PCL (the medial border of the posterior cruciate ligament at its insertion along the posterior tibia). TT-TG, PT-TG, TT-PCL, and PT-PCL distances were measured on the MR images. Each distance was measured with the knee at maximum patient comfort (rest), full extension, 15° of flexion, and 30° of flexion. Linear mixed models with random intercepts were implemented to determine variability between measurements and knee position. In general, measurements based on anatomic landmark and knee position varied greatly. The greatest variability in different knee positions was seen in mean TT-TG and PT-TG distances ± SD (TT-TG: rest, 18.1 ± 7.9 mm; full extension, 17.3 ± 5.3 mm; 15° of flexion, 11.4 ± 5.7 mm; 30° of flexion, 11.7 ± 6.0 mm; intraclass correlation coefficient [ICC] = 0.539; PT-TG: rest, 18.1 ± 6.3 mm; full extension, 17.9 ± 4.7 mm; 15° of flexion, 13.2 ± 5.2 mm; 30° of flexion, 11.8 ± 4.3 mm; ICC = 0.633). In contrast, knee position did not significantly affect distances for TT-PCL and PT-PCL (TT-PCL: rest, 23.5 ± 6.8 mm; full extension, 20.5 ± 5.5 mm; 15° of flexion, 22.8 ± 6.2 mm; 30° of flexion, 22.8 ± 6.7 mm; ICC = 0.484; PTPCL: rest, 23.4 ± 5.3 mm; full extension, 21.5 ± 4.5 mm; 15° of flexion, 22.3 ± 4.3 mm; 30° of flexion, 23.1 ± 4.8 mm; ICC = 0.509). On MRI, TT-PCL and PT-PCL measurements have significantly less variability compared with TT-TG and PT-TG measurements, regardless of knee position. Although further studies are warranted, the use of TT-PCL and PT-PCL measurements may offer more reliable assessment of tibial tubercle lateralization and patellofemoral alignment.