Tibial Slope on Conventional Lateral Radiographs in Anterior Cruciate Ligament–Injured and Intact Knees: Mean Value and Outliers

Abstract

Background: An increased tibial slope (TS) has been identified as a risk factor for anterior cruciate ligament (ACL) injury and graft failure after ACL reconstruction. However, different imaging modalities are used to determine the TS, resulting in divergent values. Consequently, no reference values and no consensus on thresholds can be reached, which in turn is mandatory for indicating correction osteotomies when facing outlier TS. Purpose: To determine the mean values of the TS and the incidence of their outliers in large cohorts of patients with ACL-injured and noninjured knees and to determine the feasibility of measuring TS on conventional lateral radiographs (CLRs). Study Design: Cross-sectional study; Level of evidence 3. Methods: TS of ACL-injured knees (n = 1000, group A) and ACL-intact knees (n = 1000, group B) was measured by 3 experienced examiners. Medial TS was measured on CLRs using the technique of Dejour and Bonnin. Patients with radiographs with poor image quality, osteoarthritis, previous osteotomies, or nondigital radiographs were excluded. The intra- and interrater reliability was calculated using the intraclass correlation coefficient. Results: The mean TS was significantly higher in group A than in group B (10.04°± 3° [range, 2°-22°] vs 9.02°± 2.9° [range, 1°-18°], respectively; P < .001). Significantly more participants in group A had TS larger than 12° (≥12°, 32.2% vs 19.8%, P < .001; ≥13°, 20.9% vs 11.1%, P < .001; ≥14°, 13.5% vs 5.7%, P < .001; ≥15°, 8% vs 2.7%, P < .001; ≥16°, 3.7% vs 1.4%, P = .0005), respectively. In contrast, significantly more participatns in group B had TS 8° or less (≤8°, 32.1% vs 42.7%, P < .001; ≤7°, 20% vs 30.9%, P < .001; ≤6°, 12.4% vs 19.8%, P < .001; ≤5°, 6.6% vs 12%, P = .0003; ≤4°, 2.8% vs 5.3%, P = .0045). The intraclass correlation coefficient revealed a good to excellent reliability throughout measurements. Conclusion: Median values for the TS were 9° for uninjured and 10° for ACL-injured knees on CLRs. Notwithstanding its statistical significance, this finding might be negligible in clinical practice. However, a significantly larger number of outliers were found in the ACL-injured group exceeding a TS of 12° and demonstrating an incremental proportion with increasing TS, serving as a potential threshold for correction osteotomy. Furthermore, CLRs in the largest cohort to date exhibited high reproducibility, proving the feasibility of CLRs as a routine measurement for TS.