A recent article associated discrete regions of abnormal T2 prolongation in the splenium of the corpus callosum with aging and radiation therapy (1). We recently encountered a patient with autosomal dominant, early-onset Alzheimer disease (AD) due to presenilin-1 (PS-1) mutation with an identical callosal appearance.
The patient’s cognitive decline began at 39 years of age. Within a year, her job performance deteriorated, forcing employers to assign her progressively less challenging duties before termination. Her parents reported no family history of neurologic or psychiatric disorders. Initial evaluation in 2001 found moderate global cognitive impairment on the Mini–Mental State Examination (MMSE 16/30). Routine studies, including brain MR imaging and electroencephalography, were reported as normal or nondiagnostic. Subsequent evaluation at 45 years of age found the patient completely dependent on her parents for basic self-care activities. Physical examination revealed irregular, generalized myoclonic movements and unsteady gait. Cognitive testing showed severe impairment (MMSE 3/30). Blood tests detected a point mutation (M146 L) in the PS-1 gene that has been previously reported in association with early-onset AD.
Brain MR imaging (1.5T) at age 45 revealed diffuse cerebral and cerebellar cortical atrophy. Axial fluid-attenuated inversion recovery (FLAIR) images revealed a discrete thin band of hyperintense signal in the anterior subependymal region of the splenium (Fig 1A). Corresponding views on T2-weighted images showed that CSF signal intensity obscured assessment of the anterior splenium (Fig 1B). Axial FLAIR images also showed two thin, paired linear foci of increased signal intensity in the vicinity of the medial lemniscus in the pons, as well as a narrow (<5 mm) rim of hyperintensity around the margins of the superior aspect of the lateral ventricles. More superior FLAIR sections failed to demonstrate definite reduced signal intensity in the motor cortex, a finding anecdotally associated with AD at 3T. Other white matter structures were normal on all additional sequences. No sagittal FLAIR images were obtained. There were no areas of abnormal restriction on diffusion-weighted images or abnormal enhancement after intravenous administration of gadolinium-diethylenetriamine pentaacetic acid contrast medium.
Pekala et al (1) reported that focal signal intensity abnormalities in the splenium were common in otherwise normal older patients and in those who have undergone brain radiation therapy. In both groups, splenium hyperintensity correlated with leukoaraiosis elsewhere. The authors noted that FLAIR sequences allowed detection of juxtaventricular lesions that were obscured by adjacent high signal intensity from the ventricles on T2-weighted images. Observations from the present case support their hypothesis: volume-averaged ventricular fluid caused high signal intensity on conventional T2-weighted images that eclipsed the distribution of hyperintensity in the anterior splenium. The authors concluded that splenium lesions do not necessarily indicate diseases such as glioma or multiple sclerosis and admonished radiologists to be aware of this common incidental finding.
On the other hand, our patient, a young woman without history of radiation therapy, showed an identical callosal appearance. Although available radiographic descriptions indicate that some patients with early-onset AD due to PS-1 mutations show distinctive white matter abnormalities, abnormal callosal signal intensity has not been reported (2, 3). In retrospect, the images depicted in Aoki et al (2) may merely represent prominent perivascular spaces in the posterior white matter. Unfortunately, they only displayed T1- and T2-weighted images, not FLAIR or proton density–weighted studies. Accordingly, the paucity of abnormal white matter signal intensity closely resembles that in our patient. These previous reports attributed leukoaraiosis to ischemia secondary to vascular amyloid deposition. The lack of other typical deep white matter signal intensity changes, coupled with the uncommon incidence of ischemic injury in the posterior corpus callosum, make this mechanism less tenable for the case described here. Another explanation is that FLAIR hyperintensity in the anterior subependymal splenium merely represents the same benign process responsible for causing the thin rim of hyperintensity around the lateral ventricles that is commonly observed with MR imaging, especially in older patients, as Pekala et al (1) demonstrated in Figure 1. Alternatively, gliosis in this region may plausibly result from focally severe axonal attrition from parieto-occipital or posterior cingulate zones linked via the splenium. This conjecture suggests that, beside cases representing “incidental” findings, abnormal FLAIR signal intensity might be detectable in other degenerative conditions with a similar distribution of neocortical injury (e.g., prion disease, posterior cortical atrophy) (4).
Suppression of ventricular fluid signal by using the FLAIR sequence unmasked the abnormality hidden by the conventional T2-weighted sequence. A, Axial FLAIR image, revealing a discrete thin band of signal intensity hyperintensity in the anterior subependymal region of the splenium. B, T2-weighted image, showing that CSF signal intensity obscured assessment of the anterior splenium (arrow).
- Copyright © American Society of Neuroradiology
