Transdermal estrogen patches for aggressive behavior in male patients with dementia: a randomized, controlled trial

To investigate the efficacy and tolerability of transdermal estrogen patches for the adjunctive treatment of aggressive behavior in male patients with advanced dementia.

The study was designed as an 8-week, randomized, controlled trial in acute aged psychiatry inpatient units and specialized nursing homes in Melbourne, Australia, between 1998 and 1999. The participants were 27 men with established dementia, identified as displaying aggressive behavior not responding to treatment for at least 2 weeks prior to referral. The instruments used to measure aggressive behavior were the Rating Scale for Aggressive Behavior in the Elderly (RAGE), the Cornell Scale for Depression in Dementia (CSDD) and the Mini-mental State Examination (MMSE). Physical examination was performed and biochemistry and serum hormone concentrations were measured at baseline and at 8 weeks. Concomitant psychotropic medication use was recorded and analyzed.

There was no significant difference in aggressive behavior at 8 weeks, but significant "rebound" in aggressive behavior (change in scores between week 8 and week 10, p < 0.009) and benzodiazepine use in the estrogen group (p < 0.03), following removal of the patches. Only behavioral items of the CSDD improved in the experimental group (p = 0.031). The use of patches was associated with a significant rise in serum estrogen (p < 0.001) but not with a significant decrease in serum testosterone (p = 0.077). There were no adverse effects associated with their use.

The use of transdermal estrogen yielding up to 100 microg per day was not associated with discernable improvement in aggressive behavior in men with advanced dementia. Small subject numbers, multiple variables and a lack of statistical power impair interpretation of these results. However, the tolerability and apparent rebound effect on removal of patches indicate the need for larger studies in this area.