issue of JMRI (1). The authors report on the use of proton MR spectroscopy to evaluate the water-to-lipid ratio as a quantitative means of characterizing vertebral marrow in Glycogen Storage Disease (GSD) Ib patients, with and without treatment.
The introduction is a well-done and comprehensive explanation of GSD type Ib and the effect of treatment on this disease. Nonetheless, the body of the article has, we believe, several weaknesses.
The authors present characteristics and results of quantitative proton MR spectroscopy in the vertebral body L1 in healthy volunteers and in GSD Ib patients. They also include figures with spectra showing increased water signal and an almost complete loss of lipid signal in the bone marrow of the L1 vertebral body in patients with GSD Ib. These findings are well known and have been published previously in different journals; for example, in an article in the August 2001 issue of AJR by the same authors (2). It is surprising, however, that the authors do not correlate their findings with MR imaging findings. In the August 2001 article the authors conclude that in patients with GSD Ib, an increased myelopoetic activity of bone marrow caused by functionally impaired leukocytes can be shown by MR imaging (2). It is puzzling, to say the least, that they failed to reiterate that important point in the current article.
Also surprising, is that in the first paragraph of the discussion they state that low signal intensity on T1weighted images is not necessarily a sign of high marrow cellularity, since osteomyelofibrosis and osteomyelosclerosis result in low signal in the presence of low cellularity (1). The first part of this statement is quite true. On the other hand, attempting to diagnose myelofibrosis with only one T1-weighted sequence is quite wrong. In our review of imaging findings in patients with myelofibrosis, we stress that to diagnose myelofibrosis, MR imaging must demonstrate a very low signal and a mainly homogeneous signal in both T1-and T2weighted images (3). The T2-weighted sequence is very important since diminution of marrow signal intensity on T1-dependent pulse sequences is also shared by other pathologies such as tumors, osteomyelitis, and metastases of various origins. Therefore, shorter T1 is less specific for bone pathologies. Moreover, T2weighted images are considered more specific in the case of myelofibrosis, since tumors and acute inflammation present longer T2, with increasing signal intensity, whereas fibrosis diminishes signal intensity (3).
The authors conclude that in patients with GSD Ib, the proliferation of myelopoieitic cells causes a nearly complete displacement of lipid cells in the central bone marrow and that no further increase of water content can be demonstrated in GSD Ib patients undergoing treatment. As this is a conclusion, what guidelines do the authors propose for diagnosing or treating patients with GSD Ib?
It seems that publishing this article after, rather than concurrently with, the August article could be misleading, as in that paper they suggested MR imaging as a tool for monitoring treatment in patients with GSD Ib.