Involuntary Weight Loss After Switching Acetylcholinesterase Inhibitors

Weight loss is common among patients with dementia and may decrease their quality of life and increase their risk of mortality; thus, prompt attention to weight loss is warranted. A variety of factors have been associated with weight loss in these patients, including dementia treatments such as acetylcholinesterase inhibitors (AChEIs) and memantine. Weight loss has been reported to occur in up to 20% of patients receiving an AChEI, but it is not known whether particular agents in this drug class are more likely to cause weight loss than others. It is also unclear whether patients who lose weight while taking one agent in this drug class can be safely treated with another agent in this drug class. The authors report a case that suggests that weight loss associated with the use of one AChEI does not necessarily predict weight loss from another AChEI. Their report also suggests that when clinicians encounter weight loss in a patient receiving an AChEI, switching to a different AChEI may be a reasonable strategy if the patient’s weight loss appears to be related to this treatment. 

Key words: Acetylcholinesterase inhibitors, galantamine, donepezil, weight loss. 
____________________________________________________________________________________________________________________________________________

The acetylcholinesterase inhibitors (AChEIs) donepezil, galantamine, and rivastigmine are commonly used to treat symptoms of Alzheimer’s disease and other dementias. In addition to increasing cholinergic activity in the brain, these agents increase peripheral cholinergic activity, the latter of which may cause adverse gastrointestinal effects, including nausea, vomiting, and diarrhea.^1,2 Although there are some conflicting data, weight loss (with or without gastrointestinal symptoms) is a significant risk of AChEI use and has been reported to occur in up to 20% of patients.^1,3 It is not known whether any particular AChEI is more likely to cause weight loss, nor whether patients who have lost weight on one such agent can be safely treated with another. We report the case of a patient with vascular dementia who had tolerated donepezil well for many years but lost almost 50 lb when switched to galantamine. Once his donepezil regimen was restarted, he promptly regained most of his weight. This case demonstrates that suspected AChEI-related weight loss in a patient with dementia may not necessarily contraindicate trying another AChEI.

Case Presentation 

An 86-year-old man was admitted to the nursing home because of falls and weight loss. He had an approximate 10-year history of progressive dementia, most consistent with vascular dementia. Approximately 7 years before  his nursing home admission, he was prescribed donepezil 10 mg every night at bedtime, and his daughter, with whom he previously resided, reported clear cognitive improvement. He tolerated donepezil well without any gastrointestinal or other adverse effects, his appetite remained good, and his weight remained stable, ranging between 210 lb and 220 lb (body mass index [BMI], 31-33 kg/m²).

The patient’s medical history was significant for diabetes mellitus with mild neuropathy (hemoglobin A_1c generally in the 7% to 8% range); hypertension; hyperlipidemia; sickle cell trait; a remote history of schizophrenia now in remission; and adenocarcinoma of the prostate, for which he underwent a prostatectomy in 1995 with no evidence of recurrence. There was no history of alcohol use. In addition to donepezil, his medications included risperidone 0.5 mg every night at bedtime, citalopram 20 mg every day before noon, amlodipine 10 mg daily, lisinopril 5 mg
daily, hydrochlorothiazide 25 mg daily, aspirin 325 mg daily, rosuvastatin 20 mg daily, gabapentin 200 mg every morning, goserelin 10.8-mg implant every 3 months, insulin glargine 57 U daily, and oral cyanocobalamin 1000 µg daily.

Seventeen months before being admitted to the nursing home, as a result of changes to the hospital formulary, donepezil was discontinued and galantamine SA 24 mg every night at bedtime was initiated. Over the subsequent 17 months, he steadily lost a total of 48 lb (nadir, 170 lb; BMI, 25 kg/m²). Both the patient and his daughter described his appetite as poor since the change to galantamine, and he reported uncharacteristically skipping meals; however, he never developed any nausea, vomiting, diarrhea, or other specific gastrointestinal symptoms. Except for some minor changes to his insulin dosage and discontinuation of rosuvastatin, there were no other significant medication, health, or psychosocial changes. About 1 month before the patient’s admission to the nursing home, his outpatient provider empirically changed his citalopram to mirtazapine 15 mg every night at bedtime in an attempt to improve his appetite, but this proved to be ineffective.

Shortly after being admitted to the nursing home, galantamine was discontinued and donepezil 10 mg every night at bedtime was restarted, given his previous positive response to this treatment. No other salient medication changes were made, and he was maintained on his prescribed daily 2000-calorie, carbohydrate-controlled diet, plus snacks as desired. He did not receive nutritional supplements. Within 2 weeks of being switched back to donepezil, he reported improvement in his appetite and he regained 30 lb within the first 3 months in the facility (203 lb; BMI, 30.4 kg/m²). 

(Continued on next page)

Discussion 

Although our patient’s rapid weight gain after being switched back to donepezil may have been attributed to his concurrent admission to the nursing home, there was no evidence that there were any contributory problems at his daughter’s home. Even if there had been, that could not explain the almost immediate onset of weight loss that occurred when galantamine was substituted for
donepezil. This double switch in medications makes a compelling case for his weight loss being causally related to the galantamine; his other medications were unlikely to have any significant anorectic or orexigenic effects and were not changed.

It is disconcerting that our patient’s weight loss was not addressed sooner, but this is not particularly unusual based on our experience. His weight loss may have gone unaddressed for so long because it was overshadowed by his chronically suboptimal glycemic control and because he was never underweight. Nevertheless, involuntary weight loss in persons with dementia should always be addressed promptly, as it is associated with functional impairment, decreased quality of life, and increased mortality.^3-6 

For patients receiving an AChEI, the overall incidence of weight loss is reported to be between 17% and 20%.^1-3,7 To some extent this weight loss may be dose-dependent. Gallini and colleagues⁸ reported the case of a patient with dementia and Parkinson’s disease who lost weight when treated with galantamine and regained weight when the dosage was reduced. Also, as occurred with our patient, weight loss may result without any associated nausea, vomiting, or diarrhea. Stewart and Gorelik³ reported eight cases of significant weight loss in patients with dementia who were treated with AChEIs, five of who had no nausea or other gastrointestinal adverse effects. In contrast, a retrospective study published by Gillette-Guyonette and colleagues⁹ found a similar incidence of weight loss between patients with Alzheimer’s disease treated with AChEIs and those who were untreated; however, these patients were treated in an academic geriatric clinic, which likely resulted in greater attention being paid to weight loss. Also, two prospective but nonrandomized studies by Guerin and colleagues^6,10 found an inverse relationship between weight loss and the use of AChEIs, suggesting a possible protective effect. The mechanism of weight loss associated with AChEIs is not entirely clear, but it may be related to subclinical nausea³ or to increased cholinergic activity in the nucleus accumbens, which has been associated with taste aversion¹¹ and satiety.¹²

We could only locate one other report in the medical literature examining the incidence of weight loss between the various AChEIs.¹ In contrast to our case report, this multifacility, retrospective analysis reviewed Minimum Data Set information for nursing homes throughout the United States and found a lower incidence of weight loss among residents treated with galantamine than with donepezil or rivastigmine. It is not clear why our patient lost weight with galantamine but not donepezil. Although both agents inhibit acetylcholinesterase, only galantamine also acts as a nicotinic receptor modulator, potentiating nicotinic cholinergic neurotransmission.¹³ It is unclear whether this activity provides additional cognitive benefits, but nicotinic agonists have been shown to decrease food intake in mice and in humans.¹⁴ 

Conclusion 

Weight loss in patients with dementia is generally an ominous sign, predicting poor quality of life and an increased risk of mortality, and the weight is rarely regained. A variety of factors can cause involuntary weight loss in these patients, including use of AChEIs; thus, patients on these agents must be closely monitored for accelerated weight loss. Clinicians should also keep in mind that nausea or other gastrointestinal adverse effects may not necessarily accompany weight loss in these patients. 

It is still not known whether AChEI-related weight loss is more often a class effect or whether it is more commonly associated with one specific agent in this drug class. Regardless, our case report demonstrates that switching agents may be a viable strategy for some patients who lose weight while on an AChEI. In addition, before prescribing or switching any AChEI, clinicians should conduct a diligent risk-benefit analysis, as the potential benefits of these agents are modest, even subtle, and the risk of significant weight loss can be devastating. If the liabilities of a drug are determined to exceed its benefits, or if there has been no demonstrable benefit, the drug must be discontinued, as would occur under any geriatric prescribing situation.