Marcus E. Raichle. (interview)(Interview)

Marc Raichle is a Professor in the Departments of Radiology, Neurology, and Neurobiology at the Washington University School of Medicine. He is also a Senior Fellow in the McDonnell Center for Studies of Higher Brain Function at Washington University. He received a B.S. and M.D. from the University of Washington in Seattle and his training in neurology at the New York Hospital-Cornell University Medical College where he was introduced to research on brain circulation and metabolism by Fred Plum and Jerry Posner. He joined the faculty of Washington University in 1971 after serving 2 years in the United States Air Force at the School of Aerospace Medicine. His research has focused, broadly, on studies of the brain circulation and metabolism as a means of understanding human brain function in health and disease. He has been active in the development of cognitive neuroscience, serving since its inception as an advisor to the McDonnell-PEW Program in Cognitive Neuroscience. In his spare time he is an amateur oboe/English horn player, sailor, and recreational high-altitude physiologist.

JOCN: As one of the world's pioneers and authorities on brain imaging and, in particular, PET, how would you characterize its short history? Are the kinds of problems you think about now, the kinds you guessed you would be thinking about 10 years ago?

MER: The history of modern functional brain imaging, now exemplified by a combination of PET, fMRI, and ERPs, in my mind represents a remarkably successful merging of developments in imaging technology, neuroscience, and behavior. The pieces of the puzzle had developed quite separately until about 10 or so years ago when cognitive science joined neuroscience in using the newly developed PET techniques to measure changes in brain blood flow in relation to changes in normal human behavior. The power of this combined approach became apparent almost immediately although the learning curve remains steep.

I certainly did not envision my current scientific agenda when I began working with Fred Plum and Jerry Posner in the late 1960s on issues of brain metabolism and blood flow nor when I joined Mike Ter-Pogossian's group here at Washington University in 1971 to work with positron emitting radionuclides. I was intrigued by their unique properties for measuring regional brain metabolism and blood flow in humans. Little did I know at the time how regional and how unique. Luckily, I was in the right place at the right time as events unfolded rather quickly after my arrival in St. Louis with the introduction of X-ray CT in about 1972 and the invention of PET in our laboratory over the ensuing 2 years.

JOCN: PET was initially built to deal with medical issues, perhaps looking at cerebral stroke per se, or studying chemotherapeutic agents for brain tumor, or looking at neurotransmitters in psychiatric and degenerative disease. PET today seems mostly committed to the study of functional correlates of cognitive function. Is this true and, if so, why?

MER: Actually, at its inception PET had a very varied agenda in the minds of the people who created it. The physics and engineering people who developed the imaging devices themselves had what I would describe as a clinical nuclear medicine orientation. In my estimation they saw PET as the means by which clinical nuclear medicine could maintain a position in the clinical area along with X-ray CT, which was clearly getting all of the attention at the time. Nuclear medicine brain scans, which had been a staple of the practice of nuclear medicine, were quickly replaced by X-ray CT. Imaging had clearly captured everyone's imagination.

The people like myself who had been working with positron emitting radio isotopes such as oxygen-15, nitrogen-13, and carbon-11 (all short-lived isotopes) for the measurement of brain blood flow, metabolism, and the like, as well as the tissue autoradiographers like Lou Sokoloff and his many disciples, all saw PET as a means of safely doing autoradiography in humans. The agenda did include the notion of functional brain mapping but, as you suggest, it included a lot else as well. There was actually a rather frustrating tension between those who thought that it ought to be tried clinically in whatever way possible and those, like myself, that felt we should be exploring what the tool could tell us about the normal brain as well as selected disease regardless of the practical clinical application.

This latter attitude might sound a bit like its coming from someone with his head in the clouds (I know that is how it sounded to some of my nuclear medicine engineering/physics colleagues), but I remain convinced that it was the right course. The argument that I often heard was that when CT came along nobody took such a position. It was tried and it worked. Why shouldn't we do the same with PET? To me the answer was simple. CT merely did considerably better what clinicians had been doing for years with dangerous, unpleasant for the patient, and difficult to interpret tests such as angiography and pneumoencephalography (for those who do not know or do not remember, this involved injecting air into the …