Neurologic signs predict periventricular white matter lesions on MRI

Objective: Periventricular white matter disease (PVWD) is associated with abnormalities on tests that involve complex cognitive processes, along with an increased risk of cerebrovascular events which are associated with significant morbidity in older patients. This study investigates whether the neurological examination can predict the presence of PVWD on magnetic resonance imaging (MRI). No prior studies have assessed whether the neurological examination can predict the presence of PVWD on MRI. Methods: A focused neurological examination was performed on a random selection of patients referred for a MRI of the brain. Staff neuroradiologists who were blinded to the results of the physical examination independently read the MRI scans. The MRI interpretations were divided into four categories based on the degree of PVWD: none, mild, moderate, severe. Results: Twenty-three subjects had some degree of PVWD, while 25 subjects had none. The total number of neurologic signs correlated significantly with the severity of PVWD even when adjusting for the effect of age (rho=0.67, p<0.001). Ninety-one percent of subjects with PVWD had three or more abnormal signs, while 76% of subjects without PVWD had fewer than three. Abnormalities with the three step motor sequencing and horizontal visual tracking tests were maximally predictive of PVWD. One or both of these tests were abnormal in 96% of subjects with PVWD, while 64% of subjects without PVWD had no problems with either test. Conclusion: Simple neurologic tests can predict the presence or absence of PVWD on MRI.