We began our series on “Advances in Neuroimaging” with first our blog on Increased Field Strength – and then yesterday’s blog on Dilated Perivascular spaces. Today, we will discuss the need for tailored protocols in properly investigating Mild Brain Injury and the existence of Post Concussion Syndrome, aka, Subtle Brain Injury.
The difference between a radiologist’s “call” of a normal or abnormal scan may also be dependent on the protocols that a given center is using. In a speech in 2002 I called for what I described as the forensically guided scan. Another more politically correct way to call it is a “tailored protocol or protocol for a specific condition.” When using a tailored protocol, you have a much greater probability of finding an abnormality related to the condition that you are specifically looking for. There can be no argument that a flexible approach to neuroimaging isn’t appropriate. If you find something, it is worth the effort. (Of course corporations and insurance companies will always whine about any extra money of effort to identify any pathology.)
If the radiologist is not directed to seek subtle pathology and applies the same techniques and level of inquiry that would be used in an acute setting where the search is for acute hemorrhage, the chances of a scan finding subtle yet significant pathology go down exponentially.
One of the areas where there may be a major advantage in a tailored protocol, is in the area of hemosidrin staining. Hemosidrin is the stain on brain tissue, left behind by blood that has now been reabsorbed back into the blood system. As an analogy, think of getting a glob of ketchup on a white shirt. When the ketchup first lands, it is clearly visible, has three dimensional mass and continues to spread. You quickly wipe it off, stop the spreading, but there is a bright red spot where the ketchup had been. You wash the shirt, the ketchup is all gone, but a stain remains. A bleed does the same thing to brain tissue. It at first has dimensional mass and will show up on a CT scan. Later, when stil fresh, it will likely show up on a conventional MRI. But as all of the significant mass of the blood has been reabsorbed, all that will be left is the stain. Neuropathologists have known this for generations, because they can see this stain on autopsy. Recent advances in MRI protocols, have created ways in which the magnet and the computer that inteprets the data, can identify this “hemosidrin” staining.
Again, an excerpt from a deposition of a leading neuroradiologist provides significant illumination:
1 Q. With respect to neurotrauma, what
2 typically would have a higher concentration of
3 iron in it?
4 A. Well, injuries — iron is basically
5 a by-product of blood. And when we look at
6 iron in the body, we usually see it when
7 it’s not in red blood cells or normal
8 structures. We usually think of it as
9 hemosiderin, which is sort of an iron stain,
10 basically.
11 So hemosiderin in trauma is what
12 you would look for after the acute phase.
13 If you have someone who is injured
14 immediately and is actively bleeding, or has
15 a lot of bleeding going on in the brain,
16 that’s a very different picture than the iron
17 that I’m talking about. The iron that I’m
18 talking about is leftover or by-product.
19 So after the fact, you had some
20 kind of bleeding in the brain, for whatever
21 reason, but a trauma you would be talking
22 about sometimes shearing injuries can bleed.
23 You can have contusions that will bleed.
24 Hematomas in the brain.
25 But when the blood is basically
1 gone, what’s left behind is a stain, a
2 hemosiderin stain. And that hemosiderin shows
3 up very black on MRI scans. Susceptibility
4 weighed sequences are designed to bring that
5 out.
Other areas where tailored protocols may come into play is increasing the proximity of the MRI slices thru the brain from the standard 2 mm slices to one mm. In essence, this improvement allows us to see pathology that might exist between the layers of the 2mm slices. Another potential advancement which is not getting much attention yet, is to increase the pixel size of the scan to 1024 by 768, (similar in size to the standard resolution of most laptops) from what is typically something more equivalent to 360 pixels by 240 (more the size of a typical Youtube video.) This type of resolution is now common when scanning for tumors. Why not brain injury? Partially priorities, but also because not enough thought is given as to where to aim this higher resolution. My suggestion is that they should be aimed: at the frontal lobes, particularly the underside of the frontal lobes, the lower brain structures and at the brain stem, areas that are difficult to image conventionally because the structures are small and the skull close by.
With all tailored protocols, there is always a cost benefit analysis. They cost more, require more attention from the neuroradiologist, and in some cases, can involve some loss of image sharpness. They may also extend the amount of time that the patient must stay in the scanner, which can be unpleasant and exceedingly boring. Faster scanning times are eliminating some of that disadvantage, but mostly these scans aren’t done, because no one demands it.
If you are having a scan done on someone with Post Concussion Symptoms, insist on at least the hemosidrin investigation, and hopefully the 1 mm slices. Someday, I believe that the 1024 x 768 resolution will be the norm, at least in the areas most likely susceptible to mild brain injury pathology.
Tomorrow:
Diffusion Tensor Imaging
No comments yet.