Posterior Tibial Slope Measurements of the Medial and Lateral Plateaus Vary Widely Between Magnetic Resonance Imaging and Computed Tomography

Abstract

This study aims to compare posterior tibial slope (PTS) measurements of the medial tibial plateau (MTP) and lateral tibial plateau (LTP) on magnetic resonance imaging (MRI) versus computed tomography (CT) to determine the agreement of measurement between imaging modalities. Patients aged 15-65 years with concurrent MRI and CT imaging were initially included. Knees with significant arthrosis (Kellgren-Lawrence grade >2), proximal tibia fracture, or artifact obscuring visualization were excluded. Two independent raters measured PTS of the MTP and LTP on paired MRI and CT. Interrater and intrarater reliability were assessed using the intraclass correlation coefficient (ICC). Intermethod agreement was assessed using ICC and Bland-Altman analyses. An acceptable Bland-Altman limit of agreement (LOA) was set at ±2°, requiring 95% of measurement differences between imaging modalities to fall between ±2° for an acceptable level of agreement. 46 knees in 45 patients met final inclusion criteria. Interrater reliability was good for MRI (ICC 0.78-0.83) and moderate-to-good for CT (ICC 0.64-0.80) studies. Intrarater reliability was moderate-to-excellent (ICC 0.64-0.94). Intermethod agreement between MRI and CT was poor at the MTP (ICC 0.34-0.42) and moderate at the LTP (ICC 0.59-0.70). Bland-Altman analysis demonstrated high variability of PTS measurements between MRI and CT: 0.16° (95% LOA -6.10-6.41°) for MTP for Rater 1; 0.22° (95% LOA -5.01-5.45°) for LTP for Rater 1; -0.95° (95% LOA -7.22-5.33°) for MTP for Rater 2; -0.99° (95% LOA -6.48-4.85°) for LTP for Rater 2, with only 47.83 to 60.87% of measurement differences falling within the predetermined acceptable LOA of ±2°. Although the interrater and intrarater reliability was moderate-to-excellent, the degree of agreement between PTS measurements on MRI and CT was highly variable at both medial and lateral plateaus. Although some variability may have been due to the study's limitations, PTS measurements at individual plateaus may not be interchangeable between MRI and CT. Level III, retrospective cohort study.