T2 MAPPING OF THE ARTICULAR CARTILAGE AS A BIOMARKER FOR OA IN A POPULATION BASED COHORT: THE ROTTERDAM STUDY

Abstract

While quantitative T2 mapping has been well established as a non-invasive tool for assessing (early) OA progression, there have been a limited number of studies investigating this in large population-based cohorts, other than the OAI. The objectives of the current study were to evaluate 1) the relationship between T2 relaxation times and patient characteristics age and BMI, and 2) the relationship between T2 relaxation times and MRI-based OA status, in a population-based study. We included 1327 knees from 665 females (mean age 59.8 years;SD:3.7 and mean BMI 26.9 kg/m2; SD:4.7) from a subcohort of the population based Rotterdam Study. All knees underwent 1.5 Tesla MRI at two different time-points (mean time between scans 5.1 years; SD: 0.4) comprising sagittal dual-echo FSE proton density–weighted, spoiled gradient echo with fat suppression, and fast imaging employing steady-state acquisition (FIESTA) sequences. The second time-point included an additional sagittal T2-weighted FSE sequence with fat suppression (4 echo times (15, 30, 45, 60 ms), repetition time 4100ms, field of view 160 x 160mm; slice thickness 3.2mm; matrix 256 x 256) for the purpose of T2 mapping. T2 relaxation times were calculated in six femoral and tibial regions of interest of full-thickness tibiofemoral cartilage: LF, LT, pLF, MF, MT, and pMF. Differences between groups were calculated using Student's t-tests for continuous data. Associations between T2 relaxation times and MRI Osteoarthritis Knee Score (MOAKS)-based tibiofemoral OA were analyzed using logistic regression with age and BMI adjustments. We defined early OA as OA developed between time-points, and late OA as OA established at the first time-point, prior to T2 mapping sequencing. Statistical analyses were performed using R (version 3.6.3). One hundred thirty-six knees had MRI defined OA at baseline, and 235 knees had OA at follow-up, meaning 99 knees developed OA the follow-up. T2 relaxation times were positively correlated with BMI, which was most pronounced in the lateral compartment (R 2 range: 0.38-0.44, all p-values <0.001). Very weak associations were found between T2 relaxation times and age for both compartments (R 2 range: 0.01-0.08). Knees with OA at follow-up had higher T2 values compared to knees without OA in the LF weight-bearing cartilage (35.7 vs. 35.3 ms, p=0.04), LT weight-bearing cartilage (35.8 vs. 34.8 ms, p<0.001), pLF cartilage (37 vs. 35.6 ms, p<0.001), MT weight-bearing cartilage (35.1 vs. 34.7, p=0.02), and pMF cartilage (37.9 vs. 36.9 ms, p<0.001). At the first time-point the LT (OR:1.11; 95%CI:1.04-1.18), pLF (OR:1.32; 95%CI:1.21-1.45), MF (OR:1.12; 95%CI:1.03-1.22), MT (OR:1.12; 95%CI:1.02-1.22), and pMF (OR:1.18; 95%CI:1.08-1.29) cartilage were significantly associated with OA after adjustments. At follow-up, LT (OR:1.08; 95%CI: 1.02-1.14), pLF (OR:1.27; 95%CI: 1.18-1.37) and pMF (OR:1.17; 95%CI:1.09-1.26) were associated with OA. In knees that developed OA during follow-up, T2 values of the pLF cartilage (OR:1.16; 95%CI:1.04-1.29) were associated with OA development. In this population-based cohort, early OA was associated with T2 values of the pLF, whereas advanced OA was associated with LT, pLF, MF, MT, and pMF cartilage segments. T2 values had very weak correlations with age, whereas weak to moderate correlations were observed between T2 relaxation values and BMI. none none. The authors are grateful to the participants and staff of the Rotterdam Study CORRESPONDENCE ADDRESS: e.oei@erasmusmc.nl