Magnetic resonance imaging in the central nervous system: comparison with computed tomography.

It is obvious that MRI has assumed a prominent role in the evaluation of disease in the brain and spine. However, CT has not been totally replaced in the evaluation of the CNS, and probably still remains the imaging modality of choice for most applications outside the CNS. Given a choice of CT or MRI in the evaluation of the CNS, certain recommendations can be made on the basis of the specific indication. It should be emphasized that these recommendations are quite dependent on the current level of technology and are thus subject to change. While not always based on rigorous comparisons, they represent a distillation of our own clinical experience (over 6,000 patients) and that of others. There is obviously a significant overlap between CT and MRI in their clinical applications in the CNS. For this reason, the recommendations in Table 1 are listed in descending order of preference, i.e., strong preference at the top, weaker preference at the bottom. In the brain MRI is indicated for the exclusion of early disease on the basis of its greater sensitivity. It is particularly useful in the posterior and middle fossae, where CT is limited by beam-hardening artifact from bone. MRI is preferred for the evaluation of MS and other diseases in the periventricular region, e.g., deep white matter infarcts and interstitial edema caused by hydrocephalus. Small extraaxial fluid collections are better seen by MRI than CT, particularly in the middle and posterior cranial fossae and at the vertex. Evaluation of the craniovertebral junction is enhanced by the ability to image directly in the sagittal plane. While MRI has clearly preempted many applications, CT is still the examination of choice in several clinical settings. Patients on respirators and those with cardiac pacemakers or intracranial aneurysm clips are currently excluded from MRI. Bone detail is better evaluated by CT; thus trauma victims with fractures and patients with suspected osseous abnormalities of the temporal bones should be studied by CT. Acute trauma patients are better studied by CT because they frequently cannot be sedated and therefore may not be able to lie motionless for the 5 to 20 min required for acquisition. In addition, trauma patients and those with acute strokes where hemorrhage is suspected should be studied by CT, where the blood can be specifically identified. The higher spatial resolution and shorter scan time of CT currently make it the examination of choice in the orbit and in the evaluation of suspected pituitary microadenomas.(ABSTRACT TRUNCATED AT 400 WORDS)