Abstracts

Cerebellar atrophy is a recognized complication of chronic epilepsy, including temporal lobe epilepsy (TLE). It has been shown that the cerbellum contributes to a variety of higher cognitive abilities, including specific types of memory. The neural circuitry underlying classical conditioning, a form of procedural memory, has been shown to be dependent on the cerebellum. The purpose of the current study was to determine if classical eyeblink conditioning, a cerebellum-dependent learning task, was affected by cerebellar atrophy in chronic epilepsy. 77 TLE subjects (26 males, 51 females) and 59 healthy controls (26 males, 33 females) underwent classical eyeblink conditioning. consisting of pairing a conditioned stimulus (CS, headphone delivered 75 dB 1-kHz tone), with an unconditioned stimulus (UCS, a 5 psi air puff to left eye), that elicited the unconditioned response (UCR, an eyeblink measured by infrared photobeam). Quantitative volumetric processing of MRI scans with manual tracing of the cerebellum was performed for a consecutive series of 49 TLE and 57 control subjects. Corrected cerebellar volumes (ICV) were significantly (3.8%) smaller in the TLE group compared to the controls (p=.028). Epilepsy patients exhibited significantly fewer conditioned responses on the classical conditioning task (59%) compared to controls (67.5%, p=.045). Among controls, the partial correlation (ICV as covariate) of cerebellar volume with classical eyeblink conditioning was significant (r=0.49, p=.005). There was no association of classical conditioning performance with volumes of segmented (gray or white matter) total brain or lobar (temporal, frontal, parietal, occipital) tissue or total CSF volume (all p[apos]s ns). Among epilepsy subjects, the relationship between classical conditioning and cerebellar volume was disrupted with a non-significant correlation (r=-.10, p=.52), and there were no significant associations between classical conditioning and segemented volumes of segmented whole brain or lobar tissue volumes or CSF volume (all p[apos]s ns). Compared to healthy controls, TLE patients exhibited significantly smaller cerebellar volumes and poorer classical eyeblink conditioning, a form of procedural learning. Among the controls there was a specific association between classical conditioning performance and cerebellar volume. In contrast, this unique structure-function relationship was disrupted and non-significant in patients with TLE. Further research into the cognitive consequences of volumetric abnormalities distant to the site of seizure onset may help to elucidate the consequences of these volumetric abnormalities on circuitry-specific cognitive abilities, leading to a better understanding of the etiology of the overall cognitive burden associated with TLE. (Supported by NIH NS 2-RO1 37738 and MO1 RR03186)