Tracking Childhood Development with MR: The Next Steps

Assessment of normal brain maturation has become an established application of MR. Since its clinical inception, investigators have recognized MR’s ability to assess myelination at a specific age based on T1 and T2 signal characteristics. In this issue of the American Journal of Neuroradiology, Nakagawa et al (page 1129) demonstrate yet again MR’s ability to assess the normal sequence of myelination accurately and reproducibly. In a systematic fashion, these authors evaluated the appearance and progression of myelination within fiber bundles. Myelination was primarily evident in the brain stem and corticospinal tracts in subjects ranging in gestational age from 35 to 145 weeks. Their study is similar in scope to previous publications on myelination by Barkovich (1), Bird (2), Hittmair (3), and Deitrich (4). Their results are also similar, with only a few exceptions that varied from four weeks’ to several months’ difference. Their approach was unique in that they evaluated not only when myelination of a fiber bundle appears, but also when the same fiber might disappear because of progressive myelination of the surrounding nerve bundles. The authors attribute much of their work to improved resolution in MR, which has allowed further characterization of myelin as it progresses in the developing brain. These investigators remind us that conspicuity in MR can also be regarded as a dynamic process related to any changes over time. Yet this report, along with previous studies, brings into focus several important issues regarding the use of MR to assess normal development. As noted, there are excellent publications equating myelin appearance with gestational age, though neuroradiologists are generally unfamiliar with the phases of myelination shown on T1- and T2-weighted images. This important aspect of any MR interpretation of an infant is frequently overlooked or purposely avoided. There is little doubt that this information obtained by MR may prove invaluable in the diagnosis and treatment of progressive neurologic childhood disease. No neuroradiologist should be interpreting MR in infants of young children unless s/he first becomes generally familiar with the appearance of these developmental stages. Nakagawa et al’s investigation and the excellent works that have preceded it assert that the practicing neuroradiologist should have a basic knowledge of these developmental stages before s/he reaches for the dictaphone. Alternatively, a large void in the evaluation of brain development exists, and the application of MR to this end is also wanting. If we were to look at other monumental works in child development such as Greulich and Pyle’s atlas of normal female and male bone maturation (5), we would recognize just how much work lies ahead if we are to provide similar firm statistics about the developing brain. Unfortunately, despite several attempts, such a statistically based atlas of normal brain myelination correlated to spe