Assessment of vastus medialis muscle quality in knee osteoarthritis using ultrashort echo time magnetization transfer MR imaging

Abstract

Objectives: Knee osteoarthritis (KOA) is a leading cause of disability, driven by cartilage degeneration, joint inflammation, and biomechanical stress. Beyond these changes, quadriceps muscle (QM) degradation, particularly in the vastus medialis (VM), has emerged as a key contributor to symptomatic KOA. Clinical MRI primarily offers qualitative assessments of muscle alterations, with limited ability to evaluate compositional changes, especially in collagen content. Novel ultrashort echo time magnetization transfer (UTE-MT) is a quantitative technique that can be used to evaluate muscle in KOA.

Materials and methods: This cross-sectional study included 54 participants (mean age: 53 ± 17 years), comprising healthy volunteers and patients with knee osteoarthritis (KOA). Knee MRI was performed to map the magnetization transfer ratio (MTR) of the VM. Intramuscular fat infiltration was classified using the Goutallier score (GS). The performance of UTE-MTR was compared with that of the GS assessed on T1-weighted imaging (T1WI) and with UTE-T1 and UTE-T2* measurements. Correlation analysis was applied to assess the relationship between VM MRI biomarkers and KOA severity, classified according to the Kellgren-Lawrence (KL) grading. An ordinal regression model was designed to estimate the severity of KOA based on UTE-derived measurements.

Results: UTE-MTR of the VM demonstrated a strong negative correlation with KL groups (r = - 0.82), outperforming T1WI (r = 0.55) for GS classification, as well as UTE-T1 (r = - 0.60) and UTE-T2* (r = - 0.50) techniques. The ordinal regression model identified UTE-MTR as the only statistically significant indicator of OA severity (adjusted odds ratio: 0.13).

Conclusions: UTE-MTR is a promising quantitative imaging biomarker for assessing VM muscle degradation in KOA.

Keywords: Collagen; Knee osteoarthritis; Magnetization transfer; Ultrashort echo time.

Fig.1.

Representative clinical sagittal T1WI-FSE and quantitative UTE-MTR, UTE-T1, and UTE-T2* maps from four subjects with varying KL grades. Each row represents a different imaging technique while each column shows maps from a subject with a defined KL grade. The columns from left to right: (1) a 20-year-old female with a KL grade of 0 (A-0:T1WI-FSE; B-0: UTE-MTR; : C-0:UTE-T1; D-0: UTE-T2*); (2) a 58-year-old male with a KL grade of 1 (A-1:T1WI-FSE; B-1: UTE-MTR; : C-1:UTE-T1; D-1: UTE-T2*); (3) a 72-year-old female with a KL grade of 2 (A-2: T1WI-FSE; B-2: UTE-MTR; : C-2:UTE-T1; D-2: UTE-T2*); and (4) a 78-year-old female with a KL grade of 3 (A-3: T1WI-FSE; B-3: UTE-MTR; C-3: UTE-T1; D-3: UTE-T2*).

Fig.2.

Boxplots and pair-comparisons of quantitative UTE measurements of the VM muscle across different OA groups: (A) UTE-MTR vs. KL group; (B) UTE-T1 vs. KL group; (C) UTE-T2* vs. KL group.

Fig.3.

Jittered scatterplots of Spearman correlations of semiquantitative (GS) and quantitative (UTE) MRI measurements of the VM muscle across different OA groups: (A) GS based on the T1WI-FSE vs. KL group; (B) UTE-MTR vs. KL group; (C) UTE-T1 vs. KL group; (D) UTE-T2* vs. KL group.