Evaluation of T1ρ relaxation times in meniscal tears during static loading

Abstract

Purpose: Among its many functions, the meniscus serves to evenly distribute the load of the body and provide a smooth surface for knee joint articulation. The properties of the menisci are based on its biochemical composition which includes water, collagen and proteoglycan (PG). Previous studies have found meniscal tears to be associated with symptomatic osteoarthritis (OA) and articular cartilage loss. Thus the health of the meniscus may play an important role in the progression in OA of the knee joint. Magnetic resonance imaging techniques such as T1ρ relaxation time mapping allows for non-invasive quantification of the biochemical composition of joint tissues and has been used to detect degenerative changes. Meniscal tears are characterized by damage to collagen fibrils and the PG matrix and are most commonly seen in the posterior horn of the medial meniscus in osteoarthritic knees. Recent studies have reported elevated T1ρ values in menisci with tears compared to healthy controls and we hypothesize loading may provide additional information on meniscal health. The purpose of this investigation was to evaluate the biochemical relationship between healthy and diseased menisci as measured by T1ρ relaxation under static loading conditions. Methods: MR data for 38 subjects (age: 53.6±10.1 years, BMI: 25.7± 8.2 kg/m2, 50% female) were acquired on a 3T MR scanner with an 8-channel phased array knee coil using a custom-built loading apparatus. MR Images (sagittal T1ρ and high resolution 3D FSE CUBE) were acquired for subjects with and without an axial load equivalent to 50% body weight. Loaded scans were prospectively registered to unloaded scans using a rigid registration algorithm. Four meniscus compartments (lateral anterior horn (AHLAT), lateral posterior horn (PHLAT), medial anterior horn (AHMED), and medial posterior horn (PHMED)) were segmented on 3D FSE CUBE images and then superimposed upon registered T1ρ maps to extract relaxation time mean values. Morphological grading using modified Whole-Organ Magnetic Resonance Imaging Score (WORMS) was performed to determine the occurrence of meniscal tears. Subjects were then stratified into no meniscal tear (meniscal grades of 0-1) and meniscal tear (meniscal grade 2-4) groups. Statistical differences between groups in both unloaded and loaded conditions were performed using a linear regression model adjusting for age, BMI and gender in JMP software version 8 (SAS Institute, Cary NC). Results: The incidence of tears in a location other than PHMED (34.2%) was very low and therefore our analysis was focused on the PHMED. Under static loading T1ρ relaxation times were significantly higher (∆T1ρ +4.1ms, P<.05) in PHMED with a tear (18.6ms) compared to those without a tear (14.5ms) in PHMED. This same trend was seen between PHMED tear (18.4ms) and no tear (15.3ms) groups in the unloaded condition (∆T1ρ +3.1ms, P<.05). There was no significant difference between loaded and unloaded meniscal T1ρ in tear or no tear groups. Conclusions: Meniscal tears had significant effects on T1ρ values in both unloaded and loaded conditions. Static loading did not have significant effects on T1ρ values in either meniscal tear or no tear groups compared to unloaded conditions. Unlike hyaline articular cartilage which has been shown to have changes in T1ρ under loaded conditions, the meniscus is mainly composed of Type 1 collagen which does not easily deform under static loading and may explain why no significant T1ρ differences were observed under varied loading conditions. Future analysis will incorporate T2 relaxation measures and stratification of subjects based on severities of osteoarthritis to probe how overall joint health may affect meniscus response to acute loading.