Summary

The study found that radiographic and MRI measurements of medial posterior tibial slope (PTS) are not comparable, particularly in patients with a PTS of ≥13˚, suggesting that radiographs may be more reliable for surgical decision-making in these cases.

Abstract

Introduction

Posterior tibial slope (PTS) measurements vary widely based on the imaging technique used, such as magnetic resonance imaging (MRI) or plain radiography, leading to conflicting results in similar patient cohorts. This suggests that factors beyond imaging modality, including variations in tibial morphology, may contribute to these discrepancies. Previous literature has identified a lack of correlation between different imaging modalities, but has not clearly defined the characteristics influencing this. This study aims to compare the differences between MRI-based medial PTS and radiographic medial PTS, subgrouping patients into those with increased versus normal PTS. The hypothesis is that radiographic and MRI PTS measurements are not comparable, particularly in patients with an increased PTS.

Methods

This study included 156 patients who underwent primary or revision ACLR and had proper lateral radiographs of the knee. A proper radiograph was defined as having less than 5 mm of offset between the posterior and distal overlap of both femoral condyles, with at least 10 cm of the tibial shaft visible. Medial PTS was measured using both radiographs and MRI. Radiographic PTS was measured as the angle between the medial tibial plateau and a line tangent to the anatomic axis of the tibia. PTS on MRI was defined as the angle between a line orthogonal to the medial plateau and the proximal anatomic axis on the central sagittal slice. Patients were categorized into a high-risk group for graft tear (≥13˚) and a normal PTS group (<13˚) based on literature-defined cut-offs. Associations between MRI and radiographic measurements were analyzed using Pearson correlation for each group. The reliability of PTS measurements was assessed by two blinded raters.

Results

The mean medial PTS was 11.9±3.4˚ in radiographic measurements and 3.4±3.5˚ in MRI measurements. In patients with a PTS of ≥13˚ on radiographic measurement, the difference in medial PTS between radiography and MRI was significantly larger than in patients with a PTS of <13˚ (10.2±3.8 vs 7.3±3, p<.001). A moderate and statistically significant correlation was found between radiographic and medial MRI PTS in patients with a PTS of <13˚ (r=0.39, p<.001). However, no correlation was observed between MRI and radiographic measurements in patients with a PTS of ≥13˚ (r=0.11, p=.35). Interobserver reliability was excellent (Cronbach’s α =0.902 and 0.851), as was intraobserver reliability for both radiographic and MRI measurements (Cronbach α=0.924 and 0.852), respectively.

Conclusion

Medial PTS did not correlate between radiographic and MRI measurements in patients with a PTS of ≥13˚ on radiographic measurement. This finding suggests the presence of a developed metaphyseal deformity or morphological variability in patients with a PTS of ≥13˚, potentially located below the region detectable in a short proximal segment MRI view. Therefore, based on this data, surgeons may consider using radiographs for surgical decision-making rather than relying on MRI-based PTS measurements. MRI tends to underestimate the magnitude of the PTS, especially in patients with increased PTS.