Refining Diagnostic Profiles of Quantitative MRI Parameters in Cartilage Assessment by Computational Model Techniques

Objectives: Quantitative MRI (qMRI) techniques such as T1ρ, T2, T2*, and T1 mapping are promising for the noninvasive assessment of cartilage, yet their exact structural and/or compositional correlates remain to be defined.

Methods: Using a clinical 3-T system (Philips, Achieva), spatially resolved T1, T1ρ, T2, and T2* maps of intact cartilage samples (n = 8) were generated, and mean parameter values were calculated in zonal and regional regions of interest (ROIs). Samples underwent histologic evaluation (Mankin classification) to ensure their integrity. For cross-referencing, a discretized numerical model capturing the distinct compositional and structural tissue properties of cartilage as a function of sample depth, that is, fluid fraction (FF), proteoglycan (PG), and collagen (CO) content and collagen fiber orientation (CFO), was implemented based on data in the literature. Pixel-wise and ROI-specific, qMRI parameters and modeled tissue parameters were correlated using Spearman’s correlation coefficient ρs.

Results: Significant correlations were found between computationally modeled parameters and T1, T2, and T2*, in particular in the central region (T1: ρs ≥ 0.7 [FF, CFO], ρs ≤ −0.8 [CO, PG]; T2 and T2*: ρs ≥ 0.67 [FF, CFO], ρs ≤ −0.71 [CO, PG]). For T1ρ, correlations were considerably weaker and fewer.

Conclusion: In a basic research context, qMRI parameters are further characterized in terms of their correlates. Although no parameter is specific toward any particular cartilage constituent, T1, T2, and T2* are more reflective of tissue compositional features than T1ρ.