A Game of Shadows 3


by Edison McDaniels, MD

Beginning in the mid 1970’s, when the first CT scans became available, the ease and ability with which neurosurgeons could localize anatomic lesions within the skull skyrocketed. The initial CT era, which spanned about 15 years, saw the development of higher resolution imaging as well as faster machines. An entire head CT in 1990 took about 10-12 minutes or even longer. Today, about 20 seconds. This means very few patients have a serious enough injury to bypass CT and go straight to surgery. If they can be stabilized at all, a stop in CT will generally increase their chance of survival by allowing detailed imaging of not just the brain, but the chest, abdomen, pelvis, and other injuries as well. In fact, such imaging in trauma not only allows the recognition of injuries requiring surgery, it also allows the recognition of injuries suitable for nonoperative observation—injuries that would have been operated in the past but, under close observation and supervision, are not operated today.

That 15 years also saw the development of contrast agents of various sorts. These contrast solutions, you might think of them as dyes, make tumors which would otherwise be invisible on CT or MRI visible. Many tumors have the same density as the surrounding brain and are only rendered truly visible by this contrast solution. A large tumor will give itself away by the distortion it causes to the surrounding structures whether or not it is itself visible, but such distortion may not be apparent post-op. In such a case, regrowth of the tumor can be checked for with contrast solutions. This has had a tremendous effect on modern day neuroimaging and neurosurgery.

E MCD 004

MRI of a large brain tumor sitting between the frontal lobes. It is only visible because the patient has been given a contrast agent, otherwise it would appear quite similar to normal brain—though distortions in that normal brain would still announce its presence to the informed eye. This tumor is almost certainly benign (not cancer).

By the way, the difference between CT and plain x-ray is that CT does have a 3D aspect to it, which allows very fine discernment of tissue density. What looks like a solid structure on plain x-ray, the heart for instance, actually can be seen as a multichambered muscular organ on CT. In a practical sense, this means a radiologist can tell the difference between normal gas inside the small intestine, and abnormal gas within the wall of the small intestine, as might be present if the small intestine has been damaged by tumor, infection, or trauma. This means the surgeon can operate early rather than waiting for the patient to develop signs of peritonitis—life threatening infection—first.

In fact, the ability to discern the fine details of anatomy is so good on CT and MRI, that some medical schools are now foregoing real human cadavers and teaching anatomy virtually, using programs composed of thousands of pictures (CT and/or MRI slices) of the human body.

MRI has come into its own since about 1995. MRI is extraordinarily sensitive in terms of anatomic detail. One can see very fine vessels, perhaps just 2-3 mm in diameter, on MRI. Very small tumors, too small even to be operated upon, are also visible. In fact, sometimes we find things which aren’t tumors at all but incidental curiosities. These are things we would only have found at autopsy in the past. As a neurosurgeon, I see several patients a month with such findings. We generally don’t operate on these incidental findings.


CT of a large, life-threatening hemorrhage (the white stuff). It’s on the right side of the brain. This patient is almost certainly comatose. Without an emergent craniotomy, he or she will not survive. With surgery, his or her long term outcome might be surprisingly good if operated in time. In neurosurgery, Time is Brain. Time is life. This is one of the first lessons a neurosurgeon in training learns.

Sometimes the incidental findings seen on MRI are important. A small lesion shows up in the frontal lobe, perhaps too small to cause symptoms just yet. Is it a tumor? Or just a curiosity? Sometimes we can’t know. Such cases have to be individualized. Occasionally surgery is recommended to biopsy the lesion or remove it entirely. More often a wait and see approach is taken, wherein the imaging is repeated in a few months looking for growth or some other change.

Up until recently, most all imaging has been anatomic. Today that is changing. More and more, functional MRI is coming to the forefront. fMRI actually observes and measures physiology—neuronal activity and connectivity. This is useful to avoid damage to elegant areas of the brain during tumor surgery, or to identify the focus of seizure activity. Or, in the case of post-surgical recovery or after a stroke, to assess neural plasticity. Much of this is still experimental and only available at tertiary centers, but stay tuned. It has been called Neurosurgery 2.0.

By the way, functional imaging of the human brain has another promising use. It is now possible to see nonorganic disease—psychiatric disorders—in action. This technology is in its infancy, but imagine being able to visualize how a human thinks in an objective fashion. Schizophrenia may show activity is one particular area, mania in another. One can imagine scanning a patient to check on the efficacy of a particular treatment or medication. Taking it still further, might it be possible to ferret out a murderer in this fashion, or maybe even a future murderer?

In other words, be careful what you think—big brother just might be watching…

Edison McDaniels, MD, is a board certified neurosurgeon practicing in the American South. Follow him on twitter @surgeonwriter and read his fiction on Amazon in both paperback and kindle. 

Send to Kindle