Magnetic resonance imaging of short T2 relaxation components in the musculoskeletal system

Abstract

One of the unusual features of magnetic resonance (MR) imaging of the musculoskeletal system is the fact that many of the tissues of most interest such as tendons, ligaments, menisci, periosteum, and cortical bone have short T2s and produce no signal when imaged with conventional pulse sequences. They have been described as MR “invisible.” The lack of signal from the invisible tissues provides a useful dark background against which abnormalities that have an increase in signal can be recognized, but it also creates problems because the absence of signal means that it is not possible to characterize these tissues by measuring their mobile proton densities, T1s or T2s, actively manipulate their image contrast, differentiate adjacent tissues with short T2s, visualize short T2 internal structures within them, or observe effects due to contrast enhancement. These problems have meant that from a technical point of view, study of the MRinvisible tissues has lagged far behind that of visible tissues. The T2 which represents the cut-off between visible and invisible tissues was about 10 ms using spin echo sequences on older MR systems [1] but with conventional gradient echo sequences and modern clinical machines, the cut-off has become shorter and is probably 1–2 ms for images with clinically useful spatial resolution. Of particular interest have been ultrashort echo time (UTE) pulse sequences with TEs as short as 8 μs. These were originally described in the 1990s [2, 3] but improved machine performance has led to images using these sequences with signal detectable from previously MRinvisible tissues at spatial resolutions comparable to those obtainable with conventional sequences in clinically acceptable scan times.