We examined the association between the Executive Interview (EXIT25), a bed
side measure of executive control, and regional magnetic resonance imaging
(MR[) pathology among 52 consecutive geriatric patients presenting to a uni
versity dementia assessment clinic, Left frontal (p <.002), left medial (p
<.03), right frontal (p <.02), and right medial (p <.02) cortical lesions s
ignificantly worsened EXIT25 scores, even after adjusting for age, global c
ognitive impairment (our the Mini-Mental State Examination), and the severi
ty of cortical dementia on the Qualitative Evaluation of Dementia [QED]. Th
e EXIT25's associations with right hemisphere lesions did not persist after
adjusting for left frontal lesions. Left posterior lesions did not signifi
cantly affect the EXIT25. Similarly, left frontal circuit pathology worsene
d EXIT25 scores (p <.05). Pathology in left anterior subcortical structures
showed a trend (p =.052). EXIT25 scores were not affected by right subcort
ical pathology, nor by pathology in either hippocampus. We conclude that th
e EXIT25 is specifically affected by frontal system MRI lesions, particular
ly on the left. This conclusion is consistent with earlier functional neuro
imaging studies associating EXIT25 performance with left mesiofrontal perfu
sion.