Ask the Expert: Cognitive Decline

The worldwide prevalence of Alzheimer's disease (AD) is now estimated to be 30 million, with four times as many cases expected by 2050. Because of the high comorbidity of AD and psychiatric disorders, psychiatrists will increasingly treat patients with AD. Successfully caring for patients with AD (and their family members and caregivers) requires a multimodal approach, one aspect of which is addressing cognitive decline. Other important aspects (beyond the scope of this review) include:

Two categories of medications have been found to be modestly effective in the treatment of cognitive symptoms: cholinesterase inhibitors (ChE-I’s) and NMDA antagonists. The pathology of AD includes degeneration of cholinergic neurons in brain networks involved in memory and attention. ChE-I’s block the metabolism of acetylcholine and increase its availability in synapses, thereby improving cognition. Typically, ChE-I’s slow or halt cognitive decline for 6 to 12 months, after which expected cognitive decline begins again. There is some evidence that ChE-I’s may decrease caregiver burden and may delay nursing home placement. Interestingly, discontinuing ChE-I’s can result in marked and abrupt declines in cognition and functioning in some patients.

The relevant APA Practice Guideline recommends that ChE-I's be offered to patients with mild-to-moderate AD. All three ChE-I’s have been FDA approved for mild-to-moderate AD, whereas only donepezil has been FDA approved for moderate-to-severe AD (roughly a Folstein MMSE score of 14 or less). These medications are likely of limited utility in the advanced stages of AD, especially in institutionalized patients. ChE-I’s may be effective for dementia due to Lewy bodies and for vascular dementia. In mild cognitive impairment (MCI), donepezil has been shown to slow conversion to AD for up to 2 years, but no effect was found at 3 years; galantamine may increase mortality in patients with mild cognitive impairment.

Side effects arise because of increased cholinergic activity: nausea, vomiting, diarrhea, bradycardia (possibly complicated by hypotension and syncope), insomnia, and nightmares (cholinergic systems are implicated in REM sleep). Gastrointestinal side effects can cause weight loss, which can be problematic in frail elders. ChE-I’s must be titrated slowly to reduce the chances of side effects. If a ChE-I is accidentally stopped, it should be retitrated as described below. (Retching and Mallory-Weiss tears have been reported when Che-I’s have been restarted at high doses.) The three ChE-I’s have similar side effect profiles, but are titrated differently:

Patients on ChE-I’s should not be prescribed other cholinergic agents. For patients undergoing electroconvulsive therapy, it should be noted that ChE-I’s will prolong the neuromuscular block caused by succinylcholine. Beta-blockers and centrally acting calcium channel blockers may increase the risk of bradycardia in patients on ChE-I’s.

These agents appear to be equally effective. If one is ineffective, a second can be started after a suitable washout period (to prevent excess cholinergic stimulation). If the patient does not tolerate this class, a trial of memantine may be considered.

Memantine is the lone member of the class of NMDA antagonists. Overexcitation of glutamatergic neurons, mediated by NMDA receptors, may be part of the pathophysiology of AD. Thus, blockade of the NMDA receptor can slow cognitive decline. As with ChE-I’s, the benefits of memantine are modest. Memantine is FDA approved for moderate-to-severe AD, although there is evidence to support its use in mild-to-moderate disease. Titration of memantine begins at 5 mg/day for 1 week, then 5 mg twice a day for 1 week, then 10 mg in the morning and 5 mg in the evening for 1 week, then 10 mg twice a day. The maximum dose in renal impairment is 10 mg/day. Memantine is generally well tolerated, with the most common side effects being nausea, dizziness, diarrhea, and agitation. There is mixed evidence regarding the combination of ChE-I and memantine, with the most recent study showing no benefit of adding memantine to a ChE-I.

Unfortunately, the history of AD includes multiple agents that have shown to be ineffective or even deleterious. These include vitamin E, vitamin B12, folate, omega-3 fatty acids, conjugated equine estrogen, gingko biloba, steroids, and COX-2 inhibitors. Management of vascular risk factors (e.g., with statins for dyslipidemia) may be helpful. Cognitive stimulation exercises (e.g., puzzles) are alluring, but there is little evidence of benefit.

One final note: older adults are commonly prescribed medications that worsen cognition, especially anticholinergics (which also antagonize the effects of ChE-I’s). Numerous psychotropic medications have anticholinergic properties, including tricyclic antidepressants, paroxetine, various atypical antipsychotics, benztropine, trixehyphenidyl, dipenhydramine, and hydroxyzine. Many agents for urinary incontinence are anticholinergic. Benzodiazepines (which are disproportionately prescribed to elders) and opioids also affect cognition. Every effort must be made to discontinue such medications if possible.