Clinical MR Neuroimaging: Diffusion, Perfusion and Spectroscopy. Jonathan H. Gillard, Adam D. Waldman, and Peter B. Barker, eds. Cambridge: Cambridge University Press, 2004. 852 pages, $330, hardcover: ISBN 0-521-82457-5.

Neuroimaging technologies have demonstrated tremendous advances over the last decade. Radioligand-based procedures such as positron emission tomography (PET) are undergoing continuous refinement and development, and there is increasing clinical utility being demonstrated (eg, with combined PET and computed tomography [CT], ie, "PET-CT," offered in many centers). Radioligand-based procedures still have a place, and in fact are the only options for some types of imaging (eg, for imaging certain neurotransmitter systems or characterizing other biomarkers), but advances in magnetic resonance (MR)-based technologies have exponentially increased the feasibility of imaging certain brain structures and even some neurally mediated processes. Although most readers of this journal are likely to be familiar with standard anatomic magnetic resonance imaging (MRI) techniques and newer MR techniques such as functional MRI (or fMRI), there are several other MR-based approaches to neuroimaging that are actually well established and are used with increasing frequency (and in many centers routinely) for clinical evaluation. Advances in MRI, along with physical and fiscal limitations associated with radioligand-based approaches, have contributed to new ways of examining processes such as cerebral blood flow, neurochemistry, and membrane diffusion. In addition, novel approaches to characterizing neuroanatomy (eg, white matter tractography, which is of great interest and potential utility in brain injury) have also been achieved through MR. The text edited by Gillard et al1 extensively discusses 3 such classes of technologies that are derived from MR-based technologies and research.

Before discussing these classes of imaging procedures with specific populations, the editors present several chapters dealing with the general basic biophysics of MR-based imaging and the specific biophysical substrates of the 3 procedures to be covered in the text. The first of these procedures is magnetic resonance spectroscopy (MRS), a technique that permits characterization of the distribution and concentration of endogenous biocompounds in the living human brain. Although several primary neurochemicals and metabolites have relatively unique magnetic properties, the most commonly imaged neuronal compounds are those of N-acetyl aspartate, choline, and lactate, compounds that are known to be impacted by brain injury. MRS allows noninvasive access to biochemicals and biomarkers that might have required the use of a radioisotope-labeling agent, or might not even be captured by any other noninvasive approach.

The second technology discussed in the text is diffusion MRI. Diffusion MRI derives its utility from its sensitivity to motion at the molecular level, specifically the diffusion of water through neuronally associated membranes. This diffusion is differentially expressed in the brain depending on tissue type. The most common use of diffusion MRI in brain research is for characterizing fiber tracts. Using diffusion MRI, it is possible to display 3-dimensional images of bundles of axonal pathways. This will be an obvious benefit to studies of traumatic brain injury, although such investigations are preliminary at present.

The third technique discussed is that of MR perfusion. This is essentially an application of MR in measuring cerebral blood flow (which can also be examined using single photon emission computed tomography or PET), which does not require the use of radioisotopes and has spatial resolution superior to that with radioligand-based approaches.

These first 9 chapters are very detailed and clearly written, but might be at too "micro" a level for the interest or need of many readers. Given that each clinical chapter provides essentially a mini-review of the techniques and biophysical processes involved, it is probable that the average clinician could select any of the nonbiophysics chapters and readily appreciate the content without having read these first several chapters.

The remainder of this text consists of 37 chapters (divided among 7 major subject sections) on specific clinical syndromes, written by leading experts on clinical MR imaging, within each category of disorders. Sections include chapters on cerebrovascular disease, neoplasms, infection/inflammation/demyelination, seizure, neuropsychiatric and neurodegenerative disorders, trauma (to be discussed in greater detail later in this review), and pediatrics. The clinical chapters are generally very detailed and nicely summarize their respective literatures. The case reports within each chapter are also very useful in illustrating the clinical applicability of these techniques. These chapters should make for interesting reading even among readers who identify their work as being primarily among persons with brain injury, as individuals with traumatic brain injuries certainly may have concomitant or preexisting disorders from the above list, and as such this information may be of some utility in differential diagnosis. Of note, however, is that there is no necessarily explicit demonstration for ways in which these new technologies could potentially be used for differential diagnosis. Rather, most of these technologies are necessarily confined to well-defined populations. This is not a weakness per se, but rather reflects the status of the field. There is some to-be-expected unevenness across the clinical chapters, as some disorders have been extensively studied with each of these technologies while others have only a handful of published investigations. In addition, there are 80 contributing authors, which increases the variability of writing styles.

The section on head trauma would obviously be of greatest interest for the readership of this journal. This section consists of 3 chapters, one of which is a general overview and 2 others that are more content-based dealing with specific forms of imaging. The first chapter within the brain trauma section is by John Pickard and reviews acute pathophysiology of brain injury as well as early clinical management of brain injury. There is passing mention of the imaging technologies covered in this text that might be of use in examining brain injury, but primarily discussion is deferred to the other 2 chapters.

The next trauma chapter is on MRS after brain injury and has been written by William Brooks. This is a comprehensive review of the existing studies of MRS involving brain injury, an area in which MRS has seen a fair amount of publication. The review is comprehensive and covers studies published to date at the time of this review. There is some attention given to the potential clinical applications of MRS, but the authors correctly point out many of the methodological problems related to the use of MRS (particularly acutely in persons with severe traumatic brain injury [TBI]) limit its routine use. I did find a curious but significant statement in this chapter, which I assume to be related to a typographical or proofing error. On page 620, when discussing early indicators of brain injury severity, the text describes the "GCS score" as "a five-point scale ranging from normal (GCS = 5) to dead (GCS = 1)." Obviously, this does not refer to the Glasgow Coma Scale (GCS) but rather the Glasgow Outcome Scale (GOS), which is discussed correctly later as an outcome rating scale. Such an error could be confusing for a reader unfamiliar with traumatic brain injury.

The third chapter within the section on brain trauma is that by Peter Bradley and David Menon, which discusses diffusion- and perfusion-weighted MR imaging in TBI. There is a very good review of basic pathophysiology in TBI and its relevance to movement of both water molecules and blood through the brain. The chapter is also richly illustrated with examples of different neuropathological sequelae of brain trauma that demonstrate how perfusion- and diffusion-weighted images may assist in differential diagnosis of the physical effects of TBI. There is also a breakdown of the types of neuropathological findings within the general rubric of brain trauma, such as diffuse axonal injury versus focal contusions.

Within the pediatric section of the book, there is a chapter on MRS of hypoxic brain injury, which might be missed if an individual were looking for TBI-related chapters solely within the preceding brain trauma section. This chapter describes in great detail pathophysiological mechanisms and implications of decreased supply of oxygen to the brain. The chapter also very nicely describes and richly illustrates the use of MRS in neonatal and pediatric hypoxic events. Finally, there is also discussion and empirical presentation of studies that provide sensitivity and predictive value of MRS in hypoxic-ischemic encephalopathy. This is certainly very useful, and in fact would have been a nice addition to see in other trauma-related (and nontrauma) chapters. This is, of course, more likely a reflection of the greater developments of these technologies being studied in the population rather than any oversight by the authors in the other chapters.

Although fMRI is likely to be quite familiar to the readers of this journal, it is not addressed in the present text. The authors state in the foreword of the text that they did not address this particular technology because it is "well covered in other texts." While this statement is certainly true, some readers might mistakenly assume on the basis of its title that fMRI could be covered in this text. In this reviewer's opinion, a more compelling rationale for not including fMRI in a clinical neuroimaging text is simply that even though it is increasingly encountered in research and even in the popular press, it has not yet amassed a sufficient evidence base in support of its clinical utility across multiple populations when compared to diffusion, perfusion, and spectroscopy.

Another important set of issues that is not addressed explicitly in the text is that of imaging data reconstruction and subsequent statistical analyses once imaging data are mapped. Neuroimaging data are not really direct "images" at all but rather are visual depictions of enormous numbers of individual data points. As with any data set-whether derived from self-report checklists, paper-and-pencil tests, or cutting-edge technologies-there are numerous measurement, statistical, and data transformation issues that must be addressed before accepting the final product as reliable and valid, no matter how appealing or "self-evident" the output might appear. A chapter (or more) that specifically dealt with this topic could have addressed at least some of these concerns.

In spite of some shortcomings, I would strongly recommend the book for individuals in any area of neuroimaging research, given its broad scope of clinical populations, as well as its detailed coverage of biophysical principles underlying several MR-based technologies. I would also suggest it for radiologists, medical physicists, and other clinicians directly involved with the application of MR technologies. The text may be less useful for individuals whose practice is predominantly or exclusively in clinical evaluation and treatment of persons with brain injury.

-Joseph H. Ricker, PhD, ABPP (CN, RP)

Associate Professor Department of Physical Medicine and Rehabilitation and the Center for the Neural Basis of Cognition University of Pittsburgh, Pittsburgh, Pa

REFERENCE