OBJECTIVE: To define the distal femur rotation pattern in a Brazilian population, correlating such pattern with the one suggested by the arthroplasty instruments, and analyzing the variability of each anatomic parameter. MATERIALS AND METHODS: A series of 101 magnetic resonance imaging studies were evaluated in the period between April and June 2012. The epidemiological data collection was performed with the aid of the institution's computed imaging system, and the sample included 52 male and 49 female patients. The measurements were made in the axial plane, with subsequent correlation and triangulation with the other plans. The posterior condylar line was used as a reference for angle measurements. Subsequently, the anatomical and surgical transepicondylar axes and the anteroposterior trochlear line were specified. The angles between the reference line and the studied lines were calculated with the aid of the institution's software. RESULTS: The mean angle between the anatomical transepicondylar axis and the posterior condylar line was found to be 6.89º, ranging from 0.25º to 12º. For the surgical transepicondylar axis, the mean value was 2.89º, ranging from -2.23º (internal rotation) to 7.86º, and for the axis perpendicular to the anteroposterior trochlear line, the mean value was 4.77º, ranging from -2.09º to 12.2º. CONCLUSION: The anatomical transepicondylar angle showed mean values corresponding to the measurement observed in the Caucasian population. The utilized instruments are appropriate, but no anatomical parameter proved to be steady enough to be used in isolation.

Keywords: Knee; Prosthesis; Image; Magnetic resonance imaging; Alignment.

OBJECTIVE: To compare intraoperative measurements of the knee with those obtained by magnetic resonance imaging, in order to validate the latter method for use in anthropometric studies. MATERIALS AND METHODS: We studied 20 knees in 20 patients with osteoarthritis, all of whom underwent total arthroplasty between August and December of 2013. We took six measurements in the distal femur and two in the proximal tibia. Using the information system of the institution, we made the measurements on magnetic resonance imaging scans that had been obtained in the axial plane. Intraoperative measurements were obtained using a caliper, after the initial cuts made during the arthroplasty. The anatomical parameters determined by magnetic resonance imaging were the same as those determined by intraoperative measurement. The intraclass correlation coefficient was used in order to assess the level of agreement in anthropometric measurements of the knee performed by magnetic resonance imaging and by intraoperative measurement. RESULTS: Statistical analysis revealed a highly significant correlation between the knee anthropometric parameters of the knee determined by intraoperative measurement and those determined by magnetic resonance imaging. CONCLUSION: The dimensions of osteoarthritic knees measured by magnetic resonance imaging were similar to those measured intraoperatively. Therefore, magnetic resonance imaging can be considered a reliable method for use in large-scale anthropometric studies that will allow the available implants to be adapted and improved.

Keywords: Knee; Diagnostic imaging; Magnetic resonance imaging; Prostheses; Osteoarthritis; Anthropometry.