A palliative care perspective on deprescribing cholinesterase inhibitors in Alzheimer’s disease

The cornerstone of palliative care is to improve quality of life for individuals and their families facing life-threatening illness. Dementia is a progressive, life-threatening illness with no cure. Considering the globally aging population¹, dementia is more prevalent as an illness managed by palliative care with an anticipated exponential increase for palliative care service delivery across all care settings. The needs of those with advanced dementia are comparable to those with other advanced progressive and life-limiting illnesses such as metastatic cancer^2,3. Unfortunately, individuals with dementia continue to receive inadequate palliative care throughout their natural illness trajectory, despite mounting evidence that the palliative approach is favored by formal and informal caregivers³, decreasing the likelihood of having physical restraints, feeding tubes, transfers to hospital at the end of life and overall costs^4,5, and increasing outcomes of improved pain control⁶ and overall quality of life^7,8.

Alzheimer’s disease is the most common form of dementia¹ and is associated with decreased levels of acetylcholine secondary to loss of cholinergic neurons in the brain⁹. Acetylcholinesterase inhibitors (ChEIs) inhibit the cholinesterase enzyme from breaking down acetylcholine, enhancing synaptic levels of acetylcholine. Clinically, the severity of dementia is often assessed using a three-stage scale of mild, moderate and severe dementia¹⁰. In Canada, the USA, Europe and China, the three ChEIs donepezil, galantamine and rivastigmine are approved for the symptomatic treatment of mild to moderate Alzheimer’s disease while donepezil is also approved for patients with severe Alzheimer’s disease¹¹. Rivastigmine in the only ChEI available as a transdermal patch¹². Common side effects of ChEIs are gastrointestinal in nature and include nausea, vomiting, diarrhea and anorexia. Dizziness and headaches are also side effects and, for individuals with bradycardia or cardiac conduction disease, ChEIs are contraindicated due to an increased risk of syncope, falls and fractures¹¹. Considering that there is no evidence to suggest a difference in effectiveness among patients with different stages of disease, all three ChEIs are associated with mild improvements in cognitive function, behavior and activities of daily living. Some individuals and their family members living with Alzheimer’s disease may hope for a delay in disease progression, increased survival or a delay until institutionalization – unfortunately, the use of ChEIs along all stages of disease does not support these goals¹³. Within a palliative approach to care, given the small potential for benefit, a decision to treat means carefully weighing the potential harms in the context of the individual’s wishes, prognosis and goals of care.

Discontinuation of ChEIs, planned or unplanned, may occur at different points in the disease trajectory of Alzheimer’s disease, not just at the end of life when the oral route becomes compromised, or when an individual with advanced disease is actively dying. There are many entry points to discussion with the patient or substitute decision maker regarding deprescribing ChEIs. Opportunities to deprescribe include situations in which disease progression is greater on treatment with ChEIs than that prior to being treated. Being in the severe stage of Alzheimer’s disease, or in the event that the patient is in a situation where continued use of the ChEI poses unacceptable risks such as the end of life, poses another opportunity. More upstream than the actively dying phase, intolerable side effects or nonadherence to medications can trigger a deprescribing discussion¹⁴.

Despite progression of Alzheimer’s disease to its severe state, there continues to be hesitation to deprescribe ChEIs. There is a growing body of literature exploring the impact of discontinuing ChEIs, mostly focused on cognitive and behavioral outcomes^11,15,16. Some case series are emerging describing a constellation of withdrawal symptoms specific to ChEIs. The main symptoms of ChEI withdrawal symptoms are neuropsychiatric in nature and include hallucinations, delusions and altered level of consciousness^17–19. Further symptoms include agitation, insomnia, labile mood and difficulty concentrating^18,19. Symptoms associated with withdrawal of the ChEI medication begin to emerge between days 3 and 7 of discontinuation and symptoms can differ in terms of level of severity^18,19.

Importantly, reported cases show symptom improvement upon re-introduction of the original ChEI medication. In keeping with a palliative approach to care, it is prudent to anticipate withdrawal symptoms when there is an abrupt discontinuation of ChEIs and to explain possible withdrawal symptoms to family members and substitute decision makers. Further studies, including the validation of current ChEI discontinuation diagnostic tools¹⁸, are required in order to enhance our understanding of when to anticipate withdrawal symptoms. As a number of withdrawal symptoms from ChEIs are neuropsychiatric in nature, it is worth signaling the importance of its recognition rather than misdiagnosis as depression or delirium, as treatment is markedly different and can significantly impact a patient’s quality of life. It becomes clinically challenging at the end of life, where delirium is common and often multifactorial and difficult to reverse, given a limited prognosis and goals of care.

In order to avoid withdrawal symptoms when deprescribing ChEIs, it is suggested that ChEIs be tapered down over weeks rather than ceasing abruptly – this may not always be possible given an individual’s particular clinical circumstances¹⁴. There is no consensus as to a tapering process; however, the literature suggests reducing the ChEI by 25–50% every 1–2 weeks^17,20. Increased caution for patients with severe Alzheimer’s disease who have a baseline of hallucinations or delusions may be warranted as a recent study in institutionalized patients with such features suggested worse clinical outcomes despite a slow taper to discontinuation of the ChEIs¹⁷. There is also a lack of consensus and guidance on the management of ChEI withdrawal amongst individuals with Alzheimer’s disease receiving a palliative approach to care. Management may include non-pharmacologic approaches such as close observation for the presence and severity of symptoms post-discontinuation. When symptoms of withdrawal are considered to be severe or distressing to the patient, pharmacologic interventions may include re-starting the ChEI at the lowest dose possible. If re-starting the ChEI is being considered but the oral route is unavailable, the rivastigmine patch is an option as there is some evidence that, among patients with mild to moderate dementia, switching from donepezil to the rivastigmine transdermal patch, either immediately or after 7 days of discontinuation of donepezil, is generally safe and well tolerated²¹. When restarting the ChEI is not possible or appropriate, supportive care for withdrawal symptoms may also include pharmacologic considerations such as the use of anti-dopaminergic agents routinely used in the palliative care setting for delirium²².

Discontinuation of ChEIs is a complex clinical scenario and very relevant in palliative care. In any setting in which the palliative approach to care is being received for individuals with Alzheimer’s disease, treatment decisions are often embedded within goals of care and prognosis. In the case of ChEIs, as a known discontinuation syndrome is recognized, mitigating distress to both the patient and their families is of paramount importance. If the decision to discontinue ChEIs is being considered, or the loss of oral route is reasonably foreseeable, anticipation and early counseling of withdrawal symptoms, and a plan to manage those symptoms, is critical. As such, considerations to include the possibility of ChEI withdrawal into discussions with the patient or substitute decision-maker can help aid patients and their families in making transitions throughout the disease trajectory.

Transparency

Declaration of funding

No funding to disclose.

Declaration of financial/other relationships

G.-A.P. and P.L. have disclosed that they have no significant relationships with or financial interests in any commercial companies related to this study or article.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgements

None reported.