Comparison of Posterior Tibial Slope Measurement Methods Based on 3-Dimensional Reconstruction

The posterior tibial slope (PTS) influences knee sagittal balance and is linked to anterior cruciate ligament (ACL) injury and meniscal pathology. Various imaging modalities assess PTS, but differences in measurement techniques and reliability remain a concern.

The purpose of this article was to evaluate the accuracy of calculating the lateral plateau slope using lateral radiograph (LR) and magnetic resonance imaging (MRI)-derived asymmetry. It was hypothesized that measurement error would be high and variable across imaging modalities and that 3-dimensional (3D) reconstruction would serve as a consistent and accurate reference for PTS assessment.

Cohort study (Diagnosis): Level of evidence, 2.

Between March 2020 and October 2023, ACL reconstruction patients were screened. PTS was measured on knee LRs, tibial long-bone LRs (LLRs), preoperative MRI, and 3D-reconstructed images from postoperative computed tomography scans. Two orthopaedic surgeons performed 2 measurements each. Reliability was assessed using intraclass correlation coefficients (ICCs), and agreement was analyzed using Bland-Altman plots, considering ±2° limits of agreement (LOA) as acceptable.

Among 132 screened patients, 32 met the inclusion criteria. Intrarater reliability was high for LR and LLR (ICC, 0.86-0.93), while interrater reliability was moderate for LR (ICC, 0.72) and excellent for LLR (ICC, 0.90). MRI showed moderate-to-good reliability (ICC, 0.65-0.79) but high variability (43.75%-65.60% within LOA). Three-dimensional reconstructions demonstrated excellent reliability (ICC, 0.93-0.97) with low variability. MRI-derived asymmetry (calculated as the difference between the PTS of the lateral tibial plateau (LTP) and the medial tibial plateau) showed good agreement with 3D imaging (ICC, 0.81; 73.44% within LOA). LLR had the highest agreement with 3D imaging for medial PTS (ICC, 0.88; 78.12% within LOA). No single method accurately estimated lateral PTS, but combining LLR-based medial PTS with MRI-derived asymmetry improved agreement with 3D LTP (ICC, 0.80; r = 0.826; P < .001).

Medial PTS measurement using anterosuperior and posterosuperior reference points in lateral views (LR and LLR) proved to be highly accurate and consistent. In MRI, only the lateromedial PTS asymmetry demonstrated reliability and consistency. To estimate lateral PTS, combining the LLR-based medial PTS with the MRI-derived lateromedial PTS difference yielded intra- and interrater reliability comparable with existing methods but improved accuracy compared with MRI-based measurements alone.