Scroll through and look for fireworks

It’s a strange feeling. You scan what you think is a normal person, and have taken all precautions to make sure that there are no indications of medical complications. But once you look at the scans, there is something wrong with the brain you are looking at. As in this case, I scanned a person some while ago as part of my study of the brain in healthy ageing.

The finding is not as obvious as one should think. Why? Unless the lesions are so vivid that anybody can see it, training is required. Furthermore, research projects are not aimed at detecting or diagnosing pathology.

As you might see, this is no regular structural scan. Well, it is not a fancy new technique either, but it is not structural in the sense that you normally see it in articles. This scan is a so-called FLAIR sequence, a technique used mostly to look for lesions inside the brain in areas close to the fluid surrounding it. The sequence is normally not used in standard research protocols, not even ageing research or clinical research. But in this case, I have applied this sequence. To our own and the subject’s fortune, this might have saved the subject from multiple brain lesions in the near future.

In fact, the subject’s neuropsychology was above average on all cognitive domains. And the normal structural (MPRAGE and T2) scans looked “about” normal, given the subject’s age (>65). At this age, you can often see quite a lot of changes in the brain that would otherwise — i.e. in young brains — be termed abnormal. However, the big difference was on the FLAIR sequence.

One of our radiologists told me to “scroll through the brain quickly and look for flashes”, just as a first approcimation to detecting brain pathology. So I’ve done that ever since. Just that simple trick has actually been helpful, this case being the prime example. Above, you can see how the lesions pop out as white sparks in the brain.

For my subject, it means that we have detected a stenosis in both arteries supporting the brain. If untreated, they would eventually have blocked the bloodstream to the brain and caused widespread neuronal damage, maybe even be life threatening. And this finding could only have been made by looking at the FLAIR. Of course, the professional radiologist or neurologist could have spotted this on the other regular scans. But they would only have seen it as part of a

So questions aside on this particular subject — what does this signify for the neuroethics of brain imaging in general?

It makes one very pertinent point: research scans are not made to detect brain pathology. But why should this be so? I, for one, would certainly not claim that all my subjects were representatives of healthy brains unless properly checked. So how then about the increasing stream of studies that (claim to) study healthy individuals? Maybe the discussion about incidental findings has focused too much on the benefit and risks for subjects and researchers, and maybe too little on the research aspects? How sure can we be that our subjects are indeed as healthy as we claim? For non-clinical projects run by clinically untrained (and uninterested?) PIs, what is the chance that healthy is indeed healthy?

I, for one feel that little bit more certain. Once this subject is excluded from my analysis, I feel that my claim about studying healthy ageing is more valid. As a bonus, the subject has had a benefit from participating in my project. You lose some, but as in this case you really gain something else, too.

-Thomas