Patellofemoral and tibiofemoral alignment in a fully weight‐bearing upright MR: Implementation and repeatability

Abstract

3 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;47:841-847.

Alignment differs between patellofemoral osteoarthritis cases and matched controls: An upright 3D MRI study

“…We acquired MR images in two scanners to benefit from the advantages of each. Our method used a single high field strength image for each participant to provide the high resolution needed for assigning bony coordinate systems, and a low field strength vertically open‐bore scanner to rapidly capture alignment in various weightbearing positions …”

“…Post‐image processing methods for calculating 3D alignment have been previously reported in detail . Briefly, we manually segmented bony outlines of the femur, patella, and tibia of the single high resolution (3T) image to create participant‐specific anatomical surface models .…”

“…Using the rapid, lower resolution images obtained in the 0.5T vertically open bore scanner, we manually segmented bony surface outlines for the femur, patella, and tibia. We then registered (shape‐matched) the high‐resolution anatomic surface models (with bone coordinate system attached) to each of the lower‐resolution bony outlines using an iterative closest points algorithm . Next, we calculated alignment of the patella relative to the femur (flexion, medial tilt, and proximal, lateral, and anterior translation) using the assigned joint coordinate system (see Figure for specific alignment labeling conventions).…”

“…Our method used a single high field strength image for each participant to provide the high resolution needed for assigning bony coordinate systems, and a low field strength vertically open-bore scanner to rapidly capture alignment in various weightbearing positions. 25,35,36 We acquired rapid images in a vertically oriented openbore 0.5T MRI scanner (ParaMed MROpen Genoa, Italy) with a commercial knee surface coil (see Table 1). Participants were scanned in upright (two-legged) at four knee flexion angles (0˚, 15˚, 30˚, 45˚).…”

“…For standing images, participants stood two-legged in the scanner (Figure 1). 25 A 12-inch goniometer was used to estimate knee flexion angle, and consistent methods were used to maintain position during image acquisition, including the placement of pressure bars intended to aid participants in achieving stillness during image acquisition (participants were instructed not put weight through the bars). 25 We acquired high-resolution sagittal images in a traditional closed-bore 3T MRI scanner (Philips Achieva Best, NL) with a dual SENSE Flex-M coil configuration (see Table 1).…”