Experimental radiation injury: combined MR imaging and spectroscopy.

A model of radiation injury to the brain was developed in the cat. Definite radiation changes were demonstrated at magnetic resonance (MR) imaging in four of six cats. These changes consisted of high-intensity abnormalities on images obtained with a long repetition time (TR) and a long echo time (TE), which were initially noted 208-285 days after irradiation. These changes were associated with gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) enhancement on short TR and inversion-recovery (IR) pulse sequences. Gd-DTPA enhancement and the high intensity on the long TR/TE images were identified at the same time and became more prominent throughout the study. Chemical-shift imaging and phosphorus spectroscopy demonstrated no notable changes despite clear-cut MR evidence of abnormalities. Sodium imaging was positive in one case. Correlation of MR and pathologic findings revealed areas of radiation necrosis and wallerian degeneration that corresponded to areas of Gd-DTPA enhancement on short TR and IR images and to areas of high intensity on long TR/TE images. Peripheral to the areas of Gd-DTPA enhancement were nonenhanced zones of high-signal-intensity abnormality on long TR/TE images, which represented regions of demyelination without necrosis. Gd-DTPA-enhanced proton imaging was the most sensitive method for detecting radiation damage in this animal model.

Magnetic resonance imaging of cranial radiation lesions.

Fifty-six patients who previously received therapeutic cranial irradiation (CRT) were imaged by a 1.5 Magnetic Resonance (MR) System 0.1-11 years following CRT. Abnormal MR findings within the treatment volume unrelated to tumor, prior to surgery, or coexisting conditions were reviewed for an association with CRT. Twenty-four patients had MR abnormalities considered to be attributable to CRT. These were scored as mild (Grade I) in 6, moderate (Grade II) in 9, and severe (Grade III) in 9. Eight of these 24 patients with CRT findings on MR had CT abnormalities that correlated with the MR. Six lesions seen on computed tomography (CT) were Grade III abnormalities; all were judged as being visualized better by MR. Eight patients had significant neurologic dysfunction attributable to their CRT lesions, and 7 of these had Grade III lesions. Whereas the clinical significance of mild or moderate CRT effects seen on MR is uncertain, Grade III (severe) MR lesions correlate well with important clinical findings.

MR of brain radiation injury: experimental studies in cats.

Two of six cats receiving small-field, single-dose, brain irradiation of 35 Gy with 6 MeV photons developed brain abnormalities in the irradiated area on MR images at 6 and 8 months, respectively, after treatment. The lesions were of high intensity on T2-weighted images and did not enhance after IV administration of gadolinium-DTPA. An additional lesion in one of these cats displayed high signal on T2-weighted images and enhanced on T1-weighted images after IV gadolinium-DTPA. Pathologic correlation revealed that the nonenhancing T2-weighted lesions consisted of edema or demyelinated regions without inflammation while the gadolinium-enhanced lesion demonstrated necrosis with inflammatory infiltrate. Focal brain irradiation may produce noninflammatory demyelination and necrosis. These histologic entities may be potentially distinguished on MR with IV gadolinium-DTPA.

MRI of sickle cell cerebral infarction.

Eleven patients with sickle cell disease and neurological symptoms underwent MRI examination. Cerebral infarcts of two types were found, those in the vascular distribution of the middle cerebral artery and those in the deep white matter. In the patient whose hydration and whose oxygenation of erythrocytes has been treated, MRI offers diagnostic advantages over arteriography and CT.

High-field MRI of hemorrhagic cortical infarction.

High-field MRI is capable of differentiating acute, subacute, and chronic hemorrhagic cortical infarctions. In eight of nine patients, hemorrhage occurred in a vascular watershed zone. Acute hemorrhagic cortical infarction produces mild cortical low intensity on T2-weighted images outlined by subcortical edema (high intensity) and isointensity with normal cortex on T1-weighted images. Subacute hemorrhagic cortical infarction shows cortical high intensity first on T1-weighted images and later on T2-weighted images; it is also associated with subcortical edema. In the chronic stage, there is a marked persistent cortical low intensity on T2-weighted images. This is most prominent in the deeply infolded cortical gyri. The low intensity noted in acute and chronic hemorrhagic cortical infarction with T2 weighting appears to be related to two separate underlying histochemical states. The characteristic cortical low intensity observed on T2-weighted images in acute and chronic hemorrhagic cortical infarction is proportional to the square of the magnetic field strength.

Magnetic resonance imaging of hemorrhagic conditions.

There is a stereotyped progression in the appearance of hemorrhage or thrombosis in magnetic resonance imaging performed at 1.5 tesla (T). This progression begins with low intensity on T2 weighted images (WI), and progresses to high intensity on T1 and T2 WI associated eventually, in some cases, with peripheral low intensity on T2 WI. Depending upon the etiology and location of hemorrhage or thrombosis the latter hypointensity on T2 WI may or may not be present. Hemorrhages occurring in regions of the brain not having an intact blood-brain barrier do not usually display persistence of low intensity. Regardless of etiology, consistent patterns in the evolution of hemorrhage may be observed on MR.