Opioid treatment for agitation in patients with advanced dementia.

BACKGROUND Some patients with advanced dementia cannot convey the experience of pain verbally and may react to pain with aggressive and agitated behaviors. We hypothesized that unrecognized pain could contribute to agitation and that low dose opioid therapy might reduce agitation by reducing pain. We therefore attempted to determine the effect of opioids on agitation in demented patients. METHODS We administered placebo for 4 weeks and a long-acting opioid for another 4 weeks to nursing home patients with advanced dementia and severe agitation despite treatment with psychotropic drugs. Patients and study nurses did not know if the medication administered was placebo or opioid. We measured the Cohen-Mansfield Agitation Inventory (CMAI) score at baseline and every two weeks. RESULTS Among 47 patients who entered the study, 25 completed the two phases. The median age for the 25 patients was 85.5 years. Analyses of the data of these 25 patients and of the patients <85 years-old showed no significant differences in agitation level between the placebo and opioid phases. However, among the 13 patients who completed the study and were > or =85 years old, the agitation level at the end of the opioid phase was significantly lower than at the end of the placebo phase (mean change in CMAI score: -6.4; 95% confidence interval (CI): -10.96, -1.8). The decrease in agitation in the patients > or =85 years old persisted after adjusting for sedation. The results remained unchanged when we expanded the analyses to include four > or =85 patients who dropped out of the study after the second week of the opioid phase. CONCLUSION Low dose, long-acting opioids can lessen agitation that is difficult to control in very old (> or =85) patients with advanced dementia.