Objectives: To investigate whether medial temporal lobe atrophy predicted outcome in patients with minor cognitive impairment and whether assessment of the medial temporal lobe could increase the predictive accuracy of age and delayed recall for outcome. Quantitative and qualitative methods of assessing the medial temporal lobe were also compared. Methods: Patients with minor cognitive impairment older than 50 years (n=31) were selected from a memory clinic and were followed up for on average 1.9 years. The medial temporal lobe was assessed in three different ways: volumetry of the hippocampus, volumetry of the parahippocampal gyrus, and qualitative rating of medial temporal lobe atrophy (MTA). Outcome measures were Alzheimer type dementia or cognitive decline at follow up. Delayed recall was tested with a verbal learning test. Results: Ten patients had experienced cognitive decline at follow up, of whom seven had probable Alzheimer type dementia. All medial temporal lobe measurements were associated with cognitive decline at follow up (p trend analysis between 0.001 (hippocampus) and 0.05 (parahippocampal gyrus)). Only the hippocampal volume and MTA score were associated with Alzheimer type dementia at follow up (p trend analysis respectively 0.003 and 0.01). All medial temporal lobe measurements increased the predictive accuracy of age and the delayed recall score for cognitive decline (p increase in predictive accuracy varied between <0.001 (hippocampus) and 0.02 (parahippocampal gyrus and MTA score)) and the hippocampal volume and the MTA score increased the predictive accuracy of age and the delayed recall score for Alzheimer type dementia (p= 0.02). Conclusions: The ability to detect patients at high risk for Alzheimer type dementia among those with minor cognitive impairment increases when data on age and memory function are combined with measures of medial temporal lobe atrophy. Volumetry of the hippocampus is preferred, but qualitative rating of medial temporal lobe atrophy is a good alternative.