Neurosurgical Tests

Neurosurgeons have a relatively limited repertoire of testing-really only seven imaging and functional tests are ordered on a routine basis: x-rays; CAT scan; myelography; MRI; angiography; EEG; and finally EMG/nerve conduction studies.


This requires little explanation-everyone is familiar with x-rays, so how about a little physics? X-rays are a form of electromagnetic radiation-just like visible light, microwaves and TV/radio waves. Without getting to far afield, like all forms of electromagnetic radiation, x-rays have a dual nature of both particle and wave. Waves of course have a length-from the top of one crest to the top of the next-and a frequency-how fast are the crests coming at you (wavelength and frequency are inversely related). X-rays have a much shorter wavelength than visiblelight and a much higher frequency than visible light, but they are both still forms of electromagnetic radiation. X-rays are not radioactive, and they can not make you radioactive or "glow in the dark"- unless you're made of phosphorous, in which case you already glow in the dark and shouldn't be getting x-rays.

X-Rays are high energy particles/waves which can do you harm (most commonly thought by damaging your DNA leading to cell death or some forms of cancer).X-Rays are a form of ionizing radiation; they have enough energy to detach electrons from some atoms, leading to free-radical formation, and damage to molecules such as DNA.

X-rays show bone reliably; they will also show air and some other dense objects, but they do not show the brain, spinal cord, nerves, discs (although they will show the disc space), ligaments, muscle (although shadows can be seen), or any of the other "soft" structures. X-rays in neurosurgery are about bone and alignment of bones; any other interpretation is "iffy" at best (needs to be confirmed by some other imaging modality).Here are some normal images.

Computerized Tomography-CAT Scan

A CAT scanner is a big donut-shaped machine that your body slides through. Generally there's plenty of room, so just about everyone can tolerate them. It delivers a fine beam of x-rays in a circular path that goes through the body. A detector,located 180 degrees opposite the beam, detects how much x-ray absorption occurred in that thin slice of your body. A computer then stacks all the slices together, and now we have a three-D view of the body. Even though it uses x-rays, a CT scan will show "soft-tissue" structures-not as well as an MRI, but considerable better than plain X-rays. Intravenous contrast can be given to make the picture even better.

CT scanners are everywhere now-most people have one in their garage, and because of their availability, speed, ability to detect most acute/severe pathologies, they've largely replaced x-rays in the ER. Newer scanners have the ability to show the cerebral blood vessels nearly as well as an angiogram, and CT angiography may one day replace conventional angiography (more on angiography latter). The major downside is the need for x-rays and the damage this form of electromagnetic radiation can cause. Still, the dose is quite small. The other risk for CT scanning is the dye that is used- most people report a flushing sensation, and occasionally it is centered in the crotch, giving people the impression that they've wet themselves. More important are the allergic reactions and the very small risk of kidney damage when contrast is used. Lab tests should be performed before a contrasted CT scan of any kind.

Here are some CT pictures, both the slice images and the 3-D reconstructions.

Here's a CT angiogram video demonstrating the brain arteries, and the ability to rotate the image (an excellent pre-surgical tool):

How about some spine CT scans:


Myelography is used to image the spine; outside of x-rays it is the oldest spinal imaging techniques. Using both x-rays and dye the spinal cord, cauda equina, and the spinal nerves are actually visualized. The accuracy is greatly enhanced with a post-myleographic CT scan.

The procedure is invasive-a radiologist (a few neurosurgeons still do their own myelography) will place a needle between the bones of the back (rarely the neck) under x-ray guidance and inject dye into the spinal sac. That dye now outlines everything within the spinal sac-cord, nerves, anything. X-rays are taken of the spine with the patient laying down-usually on their belly,then tilted, and often while standing (to see what happens to the bones and nerves when a patient bears weight), even when flexing forward or arcing back. The CT scan then follows.

Myelograms are dynamic studies- CT and MRIs are static studies-you lay flat on a table-usually spine patients are more comfortable laying down so the pathology we see with these two studies will often understate the degree of the problem. This is why for imaging of the spine CT/Myelography is the gold-standard. Yes, that's right, myelography combined with CT imaging is better, much better, much, much better than the fancy, expensive, loud MRIs. Now, that is not to say that everyone with a spine issue needs myelography-most small to intermediate problems can be detected very accurately with an MRI-however, if the issue is beyond "common", get a myelo/CT.

I "borrowed" this youtube video from a radiology group-I hope they don't mind. It shows the procedure well-if you don't like needles, or people being stuck with needles, skip it, although you'll miss the music.

Magnetic Resonance Imaging-MRI

Personally, I'm not as enamored with MRIs as a lot of people, but I have to admit the images that they generate are very cool. Twenty-five years ago we could only dream of images that were as clear as photographs; details presented so dramatically that we had to rethink some of our cherished anatomic and physiologic tenets. In time, however, like most new and shiny things, we began to realize that MRIs are not perfect; in fact, they are, at times quite misleading. To be perfectly fair, the images aren't misleading, our interpretations of those images are misleading- there are many situations in which we don't know what we are seeing. This has lead to the first rule of MRI: MRIs lie. It doesn't happen often, but with enough regularity that our interpretations of those exquisitely detailed images can't be trusted (so maybe the first rule of MRI is that our interpretations of MRIs lie-but that's to long for a first rule).

An MRI machine is a big noisy "torpedo-tube". Because even a small amount of motion will severely degrade the images, it is not uncommon for people to require sedation for the 30 to 60 minute studies. It's also not uncommon for people to refuse going into the tube (or bore) due to claustrophobia. Here's a picture of smiling patient just before being slid in as far as his ankles-wonder what he looked like an hour later.

Like CT scans, MRI has it's own form of intravenous contrast, and like CT dye, MRI dye can cause problems. It's very rare , but if you have severe kidney failure, exposure to MRI dye can lead to a condition called Nephrogenic Systemic Fibrosis. Less than 300 people have developed NSF since it's initial description in 1997, and 90% of those patients were on dialysis. Take home message: before a contrasted MRI you will need to have your kidney function checked.

All that being said, you need to know that MRIs are amazingly safe. There are no x-rays, no ionizing radiation, just a very powerful magnetic field that aligns all the hydrogen atoms ( and a few other atoms) in a particular direction. The image is created when a radiofrequency pulse (ie a radio wave) disturbs this alignment briefly. After the pulse, the atoms of the body "relax" back into the alignment that the very powerful magnetic field dictates. As the atoms relax they give off a signal that the MR detectors pick up, and a computer does the rest. Imagine a thousand clocks that are all pointing straight up at the number 12 (all your hydrogen atoms inside the magnetic field); along comes a mischievous little boy (the radiofrequency pulse) who resets all the clocks simultaneously-he's a very talented mischievous little boy- to 12:05, and then runs away giggling.The clocks sense the change and quickly snap back to 12 o'clock while yelling at the fleeing miscreant. The MRI detectors listen for the amount of yelling, which tells the computer how many clocks (or hydrogen atoms) are in this particular area.

Unlike the CT scan, we have a number of variables here: how long the little boy holds the clocks in their new position; how much he turns the hands of the clock: 12:05, 12:08; 11:44, etc.(each of these different times will generate a different "yell", ie signals from the clocks, ie atoms); we can even adjust which clocks or atoms are involved. So MRIs give us a number of different images from the same area of the body-we call them sequences. Some sequences are good for seeing "normal" anatomy, others for seeing pathology (usually by detecting increased water content), others for seeing signs of stroke (acute and chronic), or blood, etc., etc.. The flexibility of MRIs is excellent-so excellent it can be confusing-recall MRI rule #1. MRIs can also be used to evaluate the blood vessels-MRA, for Magnetic Resonance Angiogram. MRAs are inferior to both real angiograms and CT Angiograms (CTA).

Now for some images-I'll keep this to "normal" studies for now. Notice that the sequences that demonstrate normal anatomy look different between the brain and spine-the brain is a T1 sequence and the spine sequences are T2.

MRI has some very special applications that are becoming more "main-stream". One of the most clinically relevant, ie the one most people will benefit from, is in the diagnosis of acute stroke, or even TIAs (transient ischemic attacks). A CT scan can take more than 24 hours to show changes of small stokes, and remains normal in TIAs. With increasing ability to treat acute strokes, diagnosing them accurately becomes critical, and, as the results of these treatments is heavily predicated on timely administration,"we've gotta know now" becomes the typical refrain from the ER. The images to the left are from three different patients who are suffering from acute cerebral ischemia (heart-attack of the brain, or a brain-attack). Prompt treatment (ideally before four hours) will not only recover neurologic function (sight, movement, etc) but may very well save a life.

Another fascinating application of MRI is functional MRI- or simply fMRI. This study detects changes that are due to neural activity; it's predicated on the fact that with activity the blood flow to the area increases, and we are able to detect those increases. The test is very simple-just a standard MRI, and for a time the patient is asked to perform simple tasks-tap a finger, speak, answer a question, etc.. The computer can then detect the brain location that subserves those tasks. The picture to the right is a composite fMRI demonstrating multiple tasks each represented by a different color.

MRI scanning has some contraindications. You should never have an MRI if you have a pacemaker, a brain aneurysm clip, or pieces of metal in your eye. Relative contraindications include implanted insulin pumps, free or open wires that have been implanted in the body, cochlear or other types of ear implants, some forms of heart valves, or vascular clips in other parts of the body. Movements and heating of these implants have occurred, with fatal results. Pregnancy used to be a contraindication to MRI, but is now the procedure of choice to image the fetus-outside of office-based ultrasound.


This is also called an angiogram, or a blood vessel test. It is an invasive study; a radiologist will "catheterize" an artery-usually the femoral artery (in the groin)-occasionally an arm artery is used. Then, using x-ray guidance, a small catheter-skinny IV- is guided up to the base of the brain, or heart, or kidneys, or whatever body part you want to see, and then dye is injected. The dye flows through the arteries, is collected by the veins and returns to the heart. X-ray films are taken very rapidly during the artery phase and then the venous phase. It is the best way of evaluating blood-vessels, but it's got some risks-all toll between one in a hundred to one in two-hundred patients will have problems. Most problems are minor-bleeding at the puncture site (which is why patients will stay down 4-6 hours with a weight over the site); some problems are more severe-small stokes from the catheter hitting areas of plaques inside the arteries and knocking them off so that they float (or embolize) downstream. Dye reactions are also seen, and the dye can harm your kidneys (who's job it is to remove the dye from the body). Like all things medical, the benefits have to be weighted against the risks. When accurate and detailed vascular (blood vessel) anatomy has to be known, an angiogram is the best method of getting that information.

Here's a long video of an interventional neurologist explaining and demonstrating cerebral angiography. It's about 6 minutes long, and if you're about to have an angiogram it's worth watching. Warning: there's needles and blood.

Before leaving angiograms I'd like to expand on the interventional aspects. Through an angiogram we can occlude an aneurysm, shut down blood flow to a blood vessel abnormality, stent a narrowed artery, etc.. It's a rapidly expanding field, and a new field- like neurosurgery, not all interventionists are created equal, so do your home-work!

Electroencephalography EEG

Brainwaves-good name for a book (but not as good as Hybrid)- that's what EEGs detect. An array is placed on your head, preferably while you're sleepy, and all the electrical activity of the brain is detected and represented by a graph/tracing. It's non-invasive, painless, but a little long -up to an hour. It's the best way of detecting seizures (many seizures are subclinical-ie those around the patient couldn't detect it- the patient is "just acting funny"). It will also help diagnosis some less common entities in the outpatient setting, and in the ICU it's good for helping to determine the etiology of a coma/decreased level of consciousness. We also use EEG monitoring during surgery, and can use it for long-term monitoring (an array is placed surgically under the skull and you walk around in the hospital for up to a week with wires dangling from your head)-this is quite useful in children and occasionally adults who have a difficult to detect seizure focus.

Electromyography and Nerve Conduction Studies-EMG-NCS

These are outpatient studies, although we will use them during certain types of surgery. Most of the time we are looking for a source of limb pain, numbness, and/or weakness. Nerves conduct electricity-much like household wires; like wires, nerves require some insulation to conduct well. A whole host of things will damage the wires, or their insulation and impair the nerves ability to conduct the signal from the CNS to the target site. Sometimes the offending agent will be in the spinal column; other times it maybe within the nerve itself (usually the tiny blood vessels that feed the nerve get damaged), but both situations can present in similar fashion making diagnosis, and subsequent treatment, difficult. EMG/NCS can help sort this out.

First some definitions:

Neuropathy: a disease process that affects the nerves of the peripheral nervous system. This definition is extremely broad and in clinical practice, a neuropathy excludes process' that are associated with the spinal roots or the cauda equina. The most common causes of neuropathies are diabetes, nutritional deficiencies/alcoholism, hereditary, and idiopathic (our fancy word that means "we don't know"). Neuropathies classically present with symmetric (the same from side-to-side) numbness and pain (almost always a burning type of pain- occasionally a very disagreeable pins-and-needles pain). Neuropathies usually involve more than one nerve, but there are patterns in which only one peripheral nerve is involved. Neuropathies will often attack the insulation (the Schwann cells or just their myelin sheaths), the axon, or both.

Compressive neuropathy: any process that causes physical compression of a peripheral nerve-once again we exclude nerve roots and the cauda equina. Think carpal tunnel-compression of the median nerve at the wrist, or an ulnar neuropathy-compression of the ulnar nerve (the "funny-bone") at the elbow. There are other less common compressive neuropathies as well.

Radiculopathy: now we get to the nerve roots and cauda equina. Anything that causes compression/distortion/traction will cause a very specific set of symptoms- a deep boring achy pain that is "in the bone", along with a level specific distribution of numbness and weakness (ie L5 radiculopathy will cause numbness in the top of the foot and the big toe with weakness in the hamstrings and ability to lift the foot and toes). While nothing in medicine is written in stone-radiculopathies are close-all your symptoms need to match to a specific level.

EMG/NCS help to differentiate and quantify (how bad is it) the problem. There are two parts-the EMG part and the nerve conduction part.

The EMG requires very small needles to be inserted into the muscles, and then recordings are taken. A normal muscle will be fairly quiet, while a muscle who's nerve is irritated will react very differently to the needle; it's rest activity, and the response to muscle contraction will be altered. Matching the muscles involved to known nerve innervation patterns (ie this nerve innervates or runs these specific muscles) will tell us what nerve or spinal root is involved. A signal can also be sent into the nerve to assess how fast the spinal roots are responding-if they are slow, or the return "echo" is very attenuated, there is probably nerve root damage. An EMG can also detect inherent diseases of the muscles-myopathies.

Nerve conduction studies are easier. Small patches are placed on the skin in various places on the limb and tiny currents are run between them. The nerves will pick up this current and conduct it down the line. How fast it conducts, and how much of the signal is lost will tell us the condition of the nerves. A compressed nerve-ie the median nerve at the wrist, or carpal tunnel, will not conduct quickly and only a fraction of the signal will get through the compressed portion of the nerve. Nerves that have damage to their myelin (insulation) or their axon (wires) will also conduct poorly, but over longer distances than a nerve that's just being compressed at a single discrete point.

EMG/NCS aren't fun, but there also nothing to lose sleep over.

Alright, that's enough of testing, let's get into some pathology. The first section is rather small,congenital pathology of the brain.