Previously, we reported a relationship between silent cerebral infarction (SCI), as detected by magnetic resonance imaging (MRI), and late onset major depression. In the present study, we clarify the clinical features of the depressive phase of patients with major depression and SCI, and their response to antidepressant pharmacotherapy. Using clinical charts, we retrospectively examined patients with depression, who were first admitted for antidepressant pharmacotherapy. All patients were classified according to the MRI findings and the age on admission (older or younger than 50 years) into either the young SCI(–) group (n = 23), the elderly SCI(–) group (n = 27) or the elderly SCI(+) group (n = 20).The characteristics of the clinical features were evaluated at the time of admission, after 2 weeks of treatment and at the time of discharge using the Hamilton rating scale for depression (HAMD). These data were compared between each patient group. No differences in the clinical features, as evaluated by HAMD, were observed between the three groups at the time of admission. However, the mean length of treatment was significantly longer and the treatment response, as evaluated by the total HAMD score, was significantly worse in the elderly SCI(+) group than in the other two groups, when examined after 2 weeks of treatment and at the time of discharge. The elderly SCI(+) group demonstrated higher scores in feelings of guilt, suicide, retardation and hypochondriasis than the young SCI(–) group and the elderly SCI(–) group after two weeks of treatment, and higher scores in early insomnia, late insomnia, somatic anxiety and hypochondriasis at the time of discharge. Our findings suggest that while the presence of SCI does not affect the clinical features observed at the time of admission, it does affect the treatment response to antidepressant pharmacotherapy.