Temporal lobe abnormalities in dementia and depression: a study using high resolution single photon emission tomography and magnetic resonance imaging.

OBJECTIVES: Perfusion SPECT and MRI were used to test the hypothesis that late onset depression is associated with brain abnormalities. METHODS: Forty depressed patients (DSM-III-R major depressive episode, not demented at two year follow up) were recruited who were either drug free, or on a stable dose of antidepressants for at least three weeks, as well as 22 demented patients (DSM-IIIR and NINCDS/ADRDA criteria for probable Alzheimer's disease). Patients were imaged at rest with a high resolution single slice 12 detector head scanner (SME-Neuro 900) and the cerebral perfusion marker 99mTc-exametazime (HM-PAO). Temporal lobe templates were fitted with brains pitched by 20 degrees-30 degrees. A subgroup of 41 patients (22 depressed) were also scanned using a Siemens Magnetron 1.0 Tesla magnetic resonance imager, using a FLAIR imaging sequence for the assessment of white matter hyperintensities, and a Turbo FLASH sequence for the measurement of medial temporal lobe width. RESULTS: Demented patients showed reduced perfusion, particularly in the left temporoparietal cortex. In these regions of interest, patients with late onset depression tended to have perfusion values intermediate between patients with early onset depression and demented patients. Differences in changes in white matter between demented and early and late onset depressive patients did not reach conventional levels of significance. Temporal lobe width differed between demented and depressed patients, but not between early and late onset depressed patients. Perfusion and temporal lobe width were not associated, but reductions of perfusion were associated with periventricular white matter changes. Mini mental state examination scores were associated with temporal perfusion in demented patients and with changes in deep white matter in depressed patients. Finally, severity of depressive symptoms was associated with decreased perfusion in frontotemporal and basal ganglia regions of interest. CONCLUSION: A cumulative effect of duration of illness on regional cerebral perfusion could not be confirmed. Late onset depression may show more abnormalities of deep white matter and of left temporoparietal perfusion than early onset depression, but the underlying pathology remains to be established.