A novel method for assessing proximal tibiofibular joint on magnetic resonance images in patients with knee osteoarthritis

Abstract

Purpose: Knee osteoarthritis (OA) is well accepted as a whole organ disease that involves abnormalities of various structures. Proximal tibiofibular joint (ProxTibFibJ) is a synovial sliding joint that has been estimated to transmit one-sixth of the leg’s static load. Despite extensive studies on tibiofemoral and patellofemoral compartments, little attention has been paid to the ProxTibFibJ in OA research. One study has reported that proximal fibular osteotomy could significantly improve the clinical outcomes in patients with medial compartment OA. However, no study has delineated the measurement of ProxTibFibJ morphological parameters (ProxTibFibJ contacting area, load-bearing area, lateral stress-bolstering area and posterior stress-bolstering area) on magnetic resonance imaging (MRI) and investigated their correlations with knee OA structural abnormalities. This study, therefore, is to describe and validate a pragmatic method of measuring the ProxTibFibJ morphological parameters, and to investigate the longitudinal associations with joint abnormal structural changes in patients with knee OA. Methods: A total of 408 participants with knee OA were selected from the Vitamin D Effects on Osteoarthritis (VIDEO) Study, which was a multi-center, randomized and double-blind clinical trial. ProxTibFibJ morphological parameters were measured on coronal and sagittal MRI. The contacting area of ProxTibFibJ (S), its projection areas onto the horizontal (load-bearing area, Sτ), sagittal (lateral stress-bolstering area, Sφ) and coronal plane (posterior stress-bolstering area, Sυ) were assessed, respectively. MRI knee structural abnormalities, including cartilage defects, bone marrow lesions (BMLs) and cartilage volume, were evaluated at baseline and 2 years later. Ordinal logistic regression analyses were used to assess the cross-sectional associations between ProxTibFibJ morphological parameters and cartilage defects, BMLs, osteophytes, and JSN. Log binominal regression and linear regression models were used to assess the longitudinal associations between ProxTibFibJ morphological parameters and osteoarthritic structural changes. The reliabilities were examined by calculating the intra- and inter-observer correlation coefficients. Results: The average ProxTibFibJ fibular contacting area was 2.4 ± 0.7 cm2. The intra- and inter-observer correlation coefficients for all measures were excellent (all ≥0.90). In cross-sectional analyses, the ProxTibFibJ morphological parameters (S, Sτ, Sυ and Sφ) were significantly associated with radiographic medial JSN (OR 1.72 for S; 2.20 for Sτ; 1.65 for Sυ), radiographic medial osteophyte (OR 0.51 for Sφ) and MRI-assessed knee joint structural abnormalities including cartilage volume (β −0.07 for S; −0.09 for Sτ), cartilage defects (OR 1.63 for S; 1.95 for Sτ) and BMLs (OR 1.54 for S; 1.74 for Sτ) at medial tibiofemoral compartment. In longitudinal analyses, S (RR, 1.45) and Sτ (RR, 1.55) of ProxTibFibJ were significantly and positively associated with an increase in medial tibial cartilage defects over 2 years, after adjustment for age, gender, height, weight, ROA, tibial plateau bone area and intervention. Sτ (β, −0.07), Sυ (β, −0.07) and S (β, −0.06) of ProxTibFibJ were significantly and negatively associated with change in medial tibial cartilage volume, after adjusted for above covariates. Sτ (RR, 1.55) of ProxTibFibJ was positively associated with an increase in medial tibial BML, and Sφ (RR, 0.35) was negatively associated with an increase in medial femoral BMLs. No significant associations were found between ProxTibFibJ morphological parameters and osteoarthritic changes in lateral tibiofemoral compartment. Conclusions: This novel method to assess the morphological parameters of ProxTibFibJ using MRI is reproducible, and has clinical construct validity. The longitudinal associations with osteoarthritic changes suggest that higher load-bearing area of ProxTibFibJ is a potential risk factor for medial compartment OA. This may contribute to a theoretical basis of proximal fibular osteotomy in the treatment of medial tibiofemoral OA.