T2 Mapping: Methodology

T2 relaxation time in in vivo imaging is generally performed using 1.5 T or 3 T MRI scanners. Various MRI sequences can be used for acquisition including spin echo, multislice multiecho (MSME), fast spin echo, and three-dimensional spoiled gradient recalled. A comparison of these sequences for T2 measurement has been previously performed, demonstrating various differences in quantified T2 values between sequences. Thus, the direct comparison of T2 values between sequences is not recommended.

In addition to the type of MRI sequence used for quantification, an imaging parameter called “echo time” must also be optimized for T2 quantification. In general, the greater the number of echo times, the more accurate the quantification. Many studies have used four echo times, while others including the Osteoarthritis Initiative (OAI) study have used seven (echo times = 10, 20, 30, 40, 50, 60, and 70 ms). While it would be ideal to use the greatest number of echo times with the least time separation between them, these values are optimized based on limitations of the MRI hardware and as well as scan time—the greater the number of echoes, the longer the scan time.

The scan time is an important factor to consider when designing an MRI sequence for T2 measurement. The scan time is ideally as short as possible, in order to ensure feasibility of T2 relaxation time measurements in clinical practice. Compared to single-echo spin echo and single-slice multiecho spin echo sequences, the application of an MSME spin echo sequence considerably reduces the acquisition time. However, potential sources of error are introduced by using an MSME spin echo sequence. Watanabe et al 5 reported that the average T2 value measured with multislice acquisition was shorter than that measured with single-slice acquisition. However, they found only a relatively small decrease in T2 and observed no obvious inter-slice variation in T2 values when multislice acquisition was used. They concluded that multislice acquisitions for T2 measurements are clinically applicable. Thus, large clinical trials such as the OAI are using MSME acquisition.

Cartilage T2 maps are created using the following process: Typically, T2-weighted multiecho, spin echo images with varying echo times and identical repetition times are acquired. Second, T2 maps are computed assuming exponential signal decay. The T2 relaxation time value for each pixel in an image is calculated by fitting the measured signal intensity S at each echo times to a monoexponential decay function:

where S ₀ is the signal intensity at zero echo times. The last step of image processing is the creation of T2 maps containing the calculated T2 values of each pixel. A T2 map can be visualized by creating a color-coded representation of the T2 values where high values signify cartilage degeneration (Fig. 29.1). Such maps can be used to localize areas of cartilage degeneration, which may be helpful in surgical planning.

Because many factors—such as previous injury, knee alignment, and altered mechanical loading patterns—may contribute to OA, the disease is heterogeneous and may develop in different areas of the knee. Thus, cartilage T2 is often assessed on a knee level as well as in a localized manner in order to capture the heterogeneous nature of OA degeneration. Generally, T2 is quantified as the mean T2 value in a region of interest (ROI). Research studies often report mean T2 values in various regions of the knee cartilage including the patella, medial femur, lateral femur, medial tibia, lateral tibia, and trochlea. Further subdivision of ROIs into weight-bearing and nonweight-bearing regions has been performed in order to assess the effects of mechanical loading on the pathogenesis of OA. Thus, subdividing the analysis into subregions facilitates the assessment of different areas of the joint that may behave uniquely due to varied mechanical load patterns and weight-bearing properties.

In addition to this subregional analysis, recent studies have developed novel image-processing tools to assess the cartilage layers. Cartilage is composed of three primary layers: The “superficial layer” is closest to the cartilage surface and contains collagen fibers that are oriented parallel to the surface. The “transitional layer” is the largest zone that is adjacent to the “superficial layer” and has collagen fibers that are randomly oriented. The “deep layer” contains collagen fibers that are oriented perpendicular to the joint surface. MRI laminar analysis is a method that partitions the regions of interest into cartilage layers; studies have demonstrated elevated T2 values in the superficial layer compared to the deep layer. Thus, in addition to mean subregional values, laminar analysis may provide additional insight on the pathogenesis of cartilage degeneration.

Another technique that captures localized changes to cartilage is gray-level co-occurrence matrix (GLCM) texture analysis, a method developed by Haralick et al. 6 This technique has been used to assess the spatial distribution of cartilage T2. The GLCM determines the frequency that neighboring gray-level values occur in an image. Various GLCM texture parameters including contrast, variance, and entropy can be calculated in each region. Each texture parameter provides unique information on the spatial distribution of T2 values in the cartilage. Preliminary studies have shown that subjects with OA have a more heterogeneous distribution of T2 values than controls 7, 8, 9 (as reflected by higher values of GLCM contrast, entropy, and variance), demonstrating that the mean and heterogeneity of cartilage T2 pixels may be indicative of early cartilage matrix degeneration. Fig. 29.1 illustrates two representative T2 maps from a control and a subject at risk for OA, respectively. While both subjects do not have cartilage abnormalities, the subject from the incidence cohort has greater mean T2, GLCM contrast, GLCM variance, and GLCM entropy of cartilage T2.

T2 Mapping: In Vivo Research Studies

T2 mapping has been used to assess a multitude of research topics including pathogenesis of OA in various joints including the knee and hip, the efficacy of cartilage repair surgical procedures, and relationship with physical activity. This section will focus on research studies relating to knee cartilage.

Studies have utilized cartilage T2 to study OA in order to understand the role of cartilage biochemistry in the initiation and progression of disease. Cross-sectional studies have demonstrated elevated cartilage T2 values in subjects with OA as compared to controls, demonstrating alterations in cartilage biochemistry with disease. Another interesting characteristic of OA is the relationship between cartilage biochemistry and cartilage morphology. Studies have shown an inverse relationship between cartilage T2 and cartilage thickness, and that elevated T2 at baseline is associated with cartilage loss at 12 months. Other studies have assessed the relationship between cartilage degeneration and degeneration in other tissues of the knee including the bone marrow and the meniscus, and have demonstrated elevation in cartilage T2 in relation to bone marrow edema pattern and meniscus degeneration. These studies highlight the complexity of OA and that degenerative changes in one tissue may cause a cascade of degenerative changes in other neighboring regions of the knee.

Assessing the longitudinal changes in joint biochemistry, morphology, and symptoms has been a key focus in recent studies on OA. The OAI is a national and multicenter, ~5,000-patient natural history and prevalence database of OA images. This study aims to evaluate the pathogenesis of OA and to classify biomarkers that can predict the development and progression of the disease. The OAI is a cross-sectional and longitudinal dataset that includes both MRI and radiographic images of subjects scanned annually over 8 years. MRI for the assessment of cartilage morphology and cartilage T2 are available, and the longitudinal follow-up facilitates assessment of the progression of the disease. Recent studies have utilized the OAI database to assess whether cartilage T2 relaxation time can predict longitudinal changes in joint morphology—having an elevated T2 relaxation time at baseline was predictive of developing joint degeneration 3 years later. These results suggest that the initial changes to the cartilage matrix may be indicative of future joint degeneration and demonstrate the utility of the OAI dataset in the study of biomarkers for OA progression.

While a multitude of research studies have used T2 to assess cartilage degeneration during the pathogenesis of OA, recently T2 has been used to assess the outcomes of cartilage repair. Previous studies have measured T2 relaxation time following chondrocyte transplantation and microfracture surgical repair techniques, demonstrating changes in T2 following repair. These studies demonstrate that cartilage T2 measurements may supplement the standard MRI clinical protocol for the noninvasive assessment of treatment efficacy.

Overall, these studies demonstrate that T2 mapping may be a viable biomarker for early degenerative cartilage disease, and may be useful for the early detection of OA as well as for therapeutic assessment.