OBJECTIVE: To examine effect of omentum transposition surgery (OT) in Alzheimer's disease (AD). METHODS: Within-subjects design, also known as repeated-measures design, was used. OT was performed on six biopsy-confirmed AD patients (three to the left and right hemispheres each). Follow-up was conducted over 16-50 months. Outcome measures included the sum of the sub-scores of the clinical dementia rating scale (CDRSS), dementia severity rating scale (DSRS), mini-mental status exam (MMSE) and neuropsychiatric inventory (NPI), all normalized to 0-1.0. Outcomes were compared to baseline values and to expected decline with and without cholinesterase inhibitors therapy (ChEI). RESULTS: Compared to baseline and to expected decline with ChEI, CDRSS scores were 22 and 39% less impaired at means of 14 and 25 months post-OT, and DSRS scores were 12 and 22% less impaired at means of 14 and 19 months post-OT (p<0.0001). Compared to baseline and expected course with and without ChEI, the MMSE scores of the left hemisphere OT patients were not significantly different for 11, 17 and 22 months respectively (p>0.49), while those of the right hemisphere OT patients more rapidly declined. The two patients with significant pre-operative behavioral problems markedly improved; NPI severity scores decreased by 23 (16%) and 78 (54%) points and were sustained for 22 and 42 months. DISCUSSION: OT yielded cognitive, functional or behavioral improvement for up to 3.5 years in these AD patients. Compared to randomized ChEI clinical trials, OT was 34 times more likely to produce clinically significant improvement. Basic research to identify the mechanisms underlying the therapeutic effect of omentum is warranted.