Quantitative assessment of meniscus was performed using a 3 T clinical system (Siemens Skyra) in combination with a 15 channel knee coil (for signal transmission/reception) with $AdT_{1\rho}$ and $AdT_{2\rho}$ mapping technique¹ in 17 asymptomatic volunteers and 17 subjects with early radiographic OA (KL = 1,2). The cohorts were matched for sex and age. Sagittal images of the knee were acquired using a preparation block consisting of a train of 0, 4, 8, 12 and 16 adiabatic fast passages (AFP) hyperbolic secant pulses (pulse duration = 6 ms) of the HSn family, here HS4; followed by a gradient recalled echo (FLASH) readout (Table 1). For $AdT_{2\rho}$, the AFP pulses were placed between two adiabatic half passage pulses (AHP). The RF peak amplitude for both $AdT_{1\rho}$ and $AdT_{2\rho}$ was $\gamma B_{1max}$ = 800 Hz. Adiabatic longitudinal and transverse relaxation time constants in rotating frame ( $AdT_{1\rho}$ and $AdT_{2\rho}$ ) were calculated by mono-exponential fitting of the signal intensity decay on a pixel-by-pixel basis in four different regions of interest (ROIs): anterior horn medial (AHMED), posterior horn medial (PHMED), anterior horn lateral (AHLAT), and posterior horn lateral (PHLAT), (Fig.1 and Fig.2). MRI of all the subjects were assessed by an experienced radiologist (A.G.) and scored using MOAKS². Proton density (PD) turbo spin echo (TSE) coronal, PD 3D-TSE (SPACE) fat-suppressed (FS) sagittal, and T1 TSE coronal (Table 1) were used for MOAKS assessment.

The prevalence of meniscal degeneration in all subjects was assessed for each compartment of the meniscus in two steps. First, $AdT_{1\rho}$ and $AdT_{2\rho}$ relaxation times were directly used to assess meniscal degeneration in volunteers and patients. Second, they were used to assess meniscal lesion based on meniscus MOAKS scores as ‘no lesion’ (meniscus MOAKS = 0) and ‘lesion’ (meniscus MOAKS > 0) independent of their symptoms and KL. The segmentation was carried out using in-house developed MATLAB application (Mathworks, Natick, MA), and the statistical analysis was conducted using SPSS software (IBM SPSS Statistics, New York, USA).

Stratifying the data by asymptomatic subjects and patients yielded statistically different $AdT_{1\rho}$ and $AdT_{2\rho}$ relaxation times for PHMED. Both $AdT_{1\rho}$ and $AdT_{2\rho}$ in PHMED of patients were longer than their respective asymptomatic compartments (Fig.3) (p = 0.007 for $AdT_{1\rho}$, p = 0.005 for $AdT_{2\rho}$). Furthermore, $AdT_{1\rho}$ in PHLAT of patients was statistically significantly longer (p = 0.006) than the respective asymptomatic compartment. There was a trend for longer $AdT_{2\rho}$ in AHLAT of patients as compared to their respective asymptomatic compartments (p = 0.058).

Classifying the subjects based on meniscus MOAKS scores produced statistically significant difference in $AdT_{1\rho}$ and $AdT_{2\rho}$. Similarly, both $AdT_{1\rho}$ and $AdT_{2\rho}$ of PHMED with lesion were longer than their relevant compartments with no lesion (Fig.4), (p < 0.001 for $AdT_{1\rho}$, p = 0.011 for $AdT_{2\rho}$). Moreover, $AdT_{1\rho}$ value of AHLAT with lesion was longer than its relevant compartment with no lesion (p = 0.023).

In summary, a significant difference in $AdT_{1\rho}$ and $AdT_{2\rho}$ was found for PHMED considering both of the assessment steps.

Our findings indicate that $AdT_{1\rho}$ and $AdT_{2\rho}$ measurements could not only distinguish meniscal degeneration in subjects with clinical symptoms but could also distinguish the lesion in subjects clinically assessed by radiologists. The length of $AdT_{1\rho}$ and $AdT_{2\rho}$ was directly related to the severity of meniscal degeneration.

The significant difference in $AdT_{1\rho}$ and $AdT_{2\rho}$ for PHMED was consistent in all our assessments. The longer values of $AdT_{1\rho}$ and $AdT_{2\rho}$ for patients and subjects with lesion indicate the severity of meniscal degeneration in PHMED, which is consistent with findings of others^3,4. Fithian et al. found that the posterior two-third of the medial meniscus the collagen fibers were orientated obliquely with respect to each other, while in the rest of meniscal tissue the fibers were highly orientated in the circumferential direction. During load transmission the posterior two-thirds of the medial meniscus may interrupt the circumferential fiber orientation causing over circumferential tension and might explain the higher prevalence of lesion in PHMED⁵.