Neuropathological Correlates of Cumulative Benzodiazepine and Anticholinergic Drug Use

Abstract

BACKGROUND: Benzodiazepines and anticholinergic drugs have been implicated in causing cognitive decline and potentially increasing dementia risk. However, evidence for an association with neuropathology is limited. OBJECTIVE: To estimate the correlation between neuropathology at death and prior use of benzodiazepines and anticholinergic drugs. METHODS: We categorized 298 brain donors from the population-based Medical Research Council Cognitive Function and Ageing Study, according to their history of benzodiazepine (including Z-drugs) or anticholinergic medication (drugs scoring 3 on the Anticholinergic Cognitive Burden scale) use. We used logistic regression to compare dichotomized neuropathological features for those with and without history of benzodiazepine and anticholinergic drug use before dementia, adjusted for confounders. RESULTS: Forty-nine (16%) and 51 (17%) participants reported benzodiazepine and anticholinergic drug use. Alzheimer's disease neuropathologic change was similar whether or not exposed to either drug, for example 46% and 57% had intermediate/high levels among those with and without anticholinergic drug use. Although not significant after multiple testing adjustments, we estimated an odds ratio (OR) of 0.40 (95% confidence interval [95% CI] 0.18-0.87) for anticholinergic use and cortical atrophy. For benzodiazepine use, we estimated ORs of 4.63 (1.11-19.24) and 3.30 (1.02-10.68) for neuronal loss in the nucleus basalis and substantial nigra. There was evidence of neuronal loss in the nucleus basalis with anticholinergic drug use, but the association reduced when adjusted for confounders. CONCLUSIONS: We found no evidence that benzodiazepine or anticholinergic drug use is associated with typical pathological features of Alzheimer's disease; however, we cannot rule out effects owing to small numbers.