A systematic review of amnestic and non-amnestic mild cognitive impairment induced by anticholinergic, antihistamine, GABAergic and opioid drugs.

Background: Mild cognitive deficits are experienced by 18% of community-dwelling older adults, many of whom do not progress to dementia. The effect of commonly used medication on subtle impairments in cognitive function may be under-recognized.

Objective: The aim of the review was to examine the evidence attributing amnestic or non-amnestic cognitive impairment to the use of medication with anticholinergic, antihistamine, GABAergic or opioid effects.

Methods: MEDLINE and EMBASE were searched for randomized, double-blind, placebo-controlled trials of adults without underlying central nervous system disorders who underwent detailed neuropsychological testing prior to and after oral administration of drugs affecting cholinergic, histaminergic, GABAergic or opioid receptor pathways. Seventy-eight studies were identified, reporting 162 trials testing medication from the four targeted drug classes. Two investigators independently appraised study quality and extracted relevant data on the occurrence of amnestic, non-amnestic or combined cognitive deficits induced by each drug class. Only trials using validated neuropsychological tests were included. Quality of the evidence for each drug class was assessed based on consistency of results across trials and the presence of a dose-response gradient.

Results: In studies of short-, intermediate- and long-acting benzodiazepine drugs (n = 68 trials), these drugs consistently induced both amnestic and non-amnestic cognitive impairments, with evidence of a dose-response relationship. H(1)-antihistamine agents (n = 12) and tricyclic antidepressants (n = 15) induced non-amnestic deficits in attention and information processing. Non-benzodiazepine derivatives (n = 29) also produced combined deficits, but less consistently than benzodiazepine drugs. The evidence was inconclusive for the type of cognitive impairment induced by different bladder relaxant antimuscarinics (n = 9) as well as for narcotic agents (n = 5) and antipsychotics (n = 5). Among healthy volunteers >60 years of age, low doses of commonly used medications such as lorazepam 0.5 mg, oxybutynin immediate release 5 mg and oxycodone 10 mg produced combined deficits.

Conclusion: Non-amnestic mild cognitive deficits are consistently induced by first-generation antihistamines and tricyclic antidepressants, while benzodiazepines provoke combined amnestic and non-amnestic impairments. Risk-benefit considerations should be discussed with patients in order to enable an informed choice about drug discontinuation or substitution to potentially reverse cognitive adverse effects.