MRI  [ back to Neuroimaging ]
MRI or magnetic resonance imaging is a neuroimaging technique used to depict abnormalities of soft tissues in finer detail, and greater visual clarity, than CAT scan without ionizing radiation. Developed for clinical use in the late 1970s, it became a common imaging modality for brain injury during the 1980s and 1990. During this period the technology behind it has been greatly improved, producing many "generations" of new scanning machines. With today’s MRI, a patient requiring a brain scan is placed on his back with his head inside a large donut shaped scanning device. The device uses super-cooled electric coils made of titanium alloys to generate a powerful, vertical magnetic field which is held at a constant strength measured in tesla units. This is known as the "primary" magnetic field, because during the scanning process, secondary and weaker magnetic fields called "gradients" will be generated through radio frequency pulses aimed perpendicular the main magnetic field. The strength of the field is measured in units called Tesla. Machines in common clinical use generate a magnetic field of between 1.5 and 2 Tesla. The fastest MRI machines available for clinical use have a magnetic field of 3 Tesla, which is 30,000 times more powerful than the earth's. As of late 2004 some Bay Area hospitals in California were just getting their first 3 Tesla MRI scanners. It will be a long while, before rural hospitals get one too.

The patient’s brain responds to the magnetic field because it is composed of 90% water, which in turn is made up of molecules of hydrogen and oxygen which carry a single proton (called a "water proton") in each atom of hydrogen. The earth’s natural magnetic field is much weaker than the one generated by an MRI scanner. Normally the water protons in a person’s brain spin randomly in every different direction as they collide with and bounce off each other. When the machine is switched on, and a powerful magnetic force is applied to the patient’s brain, the water protons react by spinning much faster, precessing (changing their spin angle to an open cone shape) and flipping or realigning the direction of their spin along or against the plane of the vertical magnetic field (some pointing north and others south). When the primary magnetic field is suddenly switched off, the rapidly spinning protons will not revert immediately to the spin energy, spin angle and directional orientation they occupied just before they were zapped with extra energy from the primary magnetic field.

Rather, at various rates within different brain locations, they will gradually release their excess spin energy and achieve progressively lower states of new thermal equilibrium. This gradual "spin relaxation time" is accompanied by emission of detectable radiofrequency "signals." The stronger the magnetic field generated by the MRI machine magnets, the stronger the "signal," the more quicker the signal is returned and the easier it is for the radiologist to "read" the signal. When 63% of the water protons have returned to their original state of lower energy configuration, the moment called T1, the MRI scanner measures the signals from all the water protons and locates the signals from different points of origin within the patient’s brain in the context of a grid pattern. The grid is arranged in rows of tiny symmetrical blocks of space called "voxels" measuring 1x1 millimeter.

The strength of the radio-frequency emissions or signals coming from different points within the patient’s brain at T1 vary in strength according to a variety of factors including tissue location, type of tissue, tissue density (simply the number of water proteins occupying each voxel), and the physiologic state of tissue health. MRI scans tend to be done at T1 (time one) and T2 (time two) of proton spin relaxation. T2 is derived from quick application of the secondary or "gradient" magnetic field forces perpendicular the to the primary vertical field. This causes the water protons to "nutate", i.e. to increase the width of the open-cone pattern of their T1 spiral, and to gradually spread their spin axis like a gyroscope until it flattens out to a horizontal position of 180 degrees. When the gradient field is suddenly switched off, the nutated water protons will gradually relax their spins. T2 is reached when 63% have returned to their former spin configurations.

The radiofrequency emission data from the spinning water protons in the patient’s brain are gathered at T1 and T2, digitized and used by the computer to reconstruct a 3 dimensional image of the location and appearance of the soft tissues and fluids within his brain. T1 weighted MRI images represent fluids within the brain (such as blood or cerebro-spinal fluid) as being dark. T2 weighted images show the same fluids to be light in color. In both kinds of MRI images, air and bone are depicted as dark and fat as light colored. The T2 image is less crisp and sharp than the T1, but better at depicting subtle abnormalities of soft tissue integrity such as a ruptured disc. During the 1980s MRI moved neuro-imaging to a higher level than CAT scan, because it could detect smaller, more subtle lesions in the brain from head trauma patients, even in a small proportion of closed head injury patients who were rendered unconscious only very briefly and who presented with Glasgow Coma Scores in the 13-15 range. The very concept of "mild traumatic brain injury" was formed in part because of the lessons learned from MRI, that traumatic lesions may exist but remain invisible until the sensitivity of neuroimaging technology is increased to the point where they can be seen.

MRI has been used to build a data base of the "normal" uninjured brain, and to track volume changes in the normal brains of men and women as they age. This is important, because we need an accurate standard of comparison for someone with a suspected TBI and a normal brain matched to the age and gender of the patient. A recent study showed that normal adult males lose 1.5% of the volume of their hippocampus between the third and fifth decades of life, whereas women tend to lose none during that particular age span. J. Neuroscience 1/1/01 21(1):194-200.

MRI can visualize small white matter lesions, but "small" is a relative term. MRI still cannot detect lesions to individual brain cells or diffuse cellular brain damage. Right now, the best technique for detecting small white matter lesions uses an element called gadolinium which is dripped into the bloodstream. Gadolinium is ferromagnetic, meaning it is strongly attracted to magnets. During the acute phase of brain injury, gadolinium will leak out of the microvessels of the brain's blood supply into the spaces where brain tissue died and it will show up as extremely bright or hyperintense spots on the scan. Gadolinium is less useful in the post-acute stage, when damaged blood vessels have been sealed up. There is also a technique called gradient spin echo, which helps visualize hemosiderian. This is an iron bearing deposit left behind in the brain tissue by acute damage to micro-vessels of the blood's brain supply. Where an MRI with gadolinium and a gradient spin echo MRI are both positive, you can be quite certain that head trauma caused diffuse brain damage with rupture of small blood vessels.

Because even the most powerful MRI scanners of today’s technology cannot penetrate to the cellular level of the brain, the majority of patients diagnosed with mild TBI will continue to have negative MRIs. Many of these patients were clearly concussed (manifested by dazing, confusion and a very brief period of post-traumatic amnesia following closed head trauma), but did not lose consciousness. As today’s MRI technology stands, a negative MRI neither rules in or nor rules out a mild TBI and leaves the question open for other diagnostic techniques such as clinical history, neuropsychological testing, PET scans and SPECT scans. Unfortunately one still sees defense "experts" (neurologists, psychiatrists and even neuropsychologists) testify that a negative MRI is incompatible with traumatic brain injury and rules it out. In a trial, it should be pointed out that today’s MRI machines have limited sensitivity and can only show damage to a selected block of brain tissue of 1.5 millimeters. Thus millions of potentially damaged brain cells within such a block of tissue are hidden from view. A closed head trauma may cause many separate islands of cellular damage smaller than what can be detected by MRI. Such swiss cheese damage cannot be detected when the holes are too small. Unfortunately even a tiny hole can ruin the capacity of a cup to hold water. Although the brain is not a cup, small holes in the brain will certainly decrease the speed, efficiency and reach of its mental operations, as has been proven over and over in published studies in peer reviewed journals. In the future, MRI may be much more useful than it is now in visualizing the effects of a TBI. Right now MRI spectroscopy is a better technique for indirect detection of diffuse shear damage on the cellular level, but is still a research tool.