We studied the midbrain SERT availability in patients with major depression and assessed the relation of SERT occupancy by citalopram to the treatment response. 21 non-medicated patients with major depression and 13 healthy controls were examined by [(123)I]-ADAM SPECT. The midbrain SERT availability (SERT V(3)'') was calculated using individual MRI scans. In 13/21 patients SPECT was repeated 7 days after oral medication with citalopram (10 mg/day). We found no significant difference in the mean midbrain SERT availability between the studied patients with major depression and healthy controls (0.86 +/- 0.27 vs. 0.71 +/- 0.44, p = 0.069). The mean SERT occupancy accounted to 61%. The degree of SERT blockade by citalopram did not correlate with the reduction in HAMD total score. Treatment with low-dosed citalopram caused individually variable occupancy of the midbrain-SERT and a rapid clinical improvement in 54% of the investigated patients.