Donepezil as a safe alternative treatment after maculo‐papular eruption related to rivastigmine in Lewy body disease: a case report and pharmacovigilance data

Rivastigmine, a reversible and noncompetitive acetylcholinesterase inhibitor (AChI), provides significant symptomatic benefit for neuropsychiatric symptoms in Lewy body disease (LBD). It is available as oral capsules and transdermal patches. Few adverse effects have been reported, gastrointestinal being the most frequent during initiation. Discontinuation due to adverse effects is rare and may be associated with faster cognitive decline. Here we report the case of an extensive erythematous maculopapular eruption following rivastigmine introduction in a patient diagnosed with LBD, who provided informed consent for this case report. This 79-year-old Caucasian man had a medical history of hypertension, diabetes and dyslipidaemia. His medication included ramipril, ezetimibe, simvastatin, metformin and acetylsalicylic acid, with no recent changes. He had no history of drug-related allergic reaction or dermatologic condition. He was assessed for major neurocognitive disorder with recurrent visual hallucinations and was diagnosed with LBD. Rivastigmine transdermal patches were introduced at 4.6 mg per 24 h (Fig. 1). The treatment was well tolerated initially with significant decrease of visual hallucinations, so dosage was increased 3 months later up to 9.5 mg per 24 h. A month later, the patient developed a pruritic erythematous maculopapular eruption at the patch application site. Suspecting contact dermatitis, his geriatrician switched to oral rivastigmine at 1.5 mg twice daily. Two weeks later, the patient visited the hospital emergency room because of the extension of the eruption over 70% of the body surface, including limbs, back and abdomen. No other symptoms were reported such as gastrointestinal symptoms, fever, mucosal tissue injury or Nikolsky sign. An antihistaminic treatment (cetirizine) was introduced. However, as the skin lesions did not regress after a week (Fig. 2), rivastigmine was finally discontinued and significant improvement in the skin condition was observed within 48 h. Given the efficacy of rivastigmine in improving LBD symptoms, his geriatrician decided to introduce donepezil, another AChI, at the dose of 5 mg per day with close monitoring, 1 month after the resolution of cutaneous lesions. The dose was increased to 10 mg after 4 weeks and over the next 6 months, without any adverse effect. Skin tests with rivastigmine 1.5 mg, including prick tests and 30% patch tests in petrolatum, yielded negative on day 2 and day 5. Most common cutaneous adverse reaction due to transdermal patches is contact dermatitis, usually confined to the application site but generalization is possible. Commonly, contact dermatitis is often related to an adhesive component of the patch rather than the rivastigmine molecule itself. However, in our case, the eruption spread after switching to oral capsules, indicating that the adverse effect was likely due to rivastigmine itself. These observations align with evidence from the French pharmacovigilance database, including more than 400 declarations of rivastigmine-related adverse effects. Almost all cutaneous adverse effects associated with transdermal patches were reported as contact dermatitis at the application site, with significant improvement after switching to oral capsules. Using the Naranjo scale for assessing causality, we found a high probability that the adverse effect was related to rivastigmine (score = 6; probability range: 5–8). Although allergological tests were negative, intradermal tests were not performed and this marked the first application of the rivastigmine reagent. After comprehensive multidisciplinary assessment including allergologists, dermatologists, geriatricians, pharmacovigilance experts and pharmacists, the allergic hypothesis was the most likely. A similar case of generalised skin eruption after switching from patches to capsules had been previously reported, where the patient presented maculopapular rash limited to the application site at the dose of 4.6 mg. Our patient showed no dermatological sign at this dosage. Few cases of cutaneous reaction to rivastigmine have been reported (1-5) (See Table 1). This is the first case showing no cross-reaction with another AChI. In a previous report, a patient introduced to oral galantamine experienced similar skin symptoms. Our patient has been treated for 6 months with donepezil without any recurrence of skin lesions or other adverse effects. Local reaction to patch When switching to capsules: highly pruritic maculopapular eruption (upper trunk, chest, arms and back that progressively expanded on her abdomen and buttocks) Although rivastigmine, galantamine and donepezil belong to the same therapeutic class, they have distinct structural, pharmacological and mechanism-of-action characteristics. Rivastigmine inhibits both acetylcholinesterase and butyrylcholinesterase, galantamine acts as a positive allosteric modulator of nicotinic acetylcholine receptors, while donepezil is a specific AchI. This difference could explain the absence of cross-allergy with donepezil. Some limitations include negative allergological tests and the absence of biological test or biopsy during clinical presentation. Although this case is unique and requires external validity, it suggests that a patient diagnosed with LBD should not be deprived of a symptomatically useful therapeutic class, despite a severe cutaneous reaction. AChI remain a keystone in treating neuropsychiatric symptoms in LBD. For our patient, switching to donepezil after a generalised eruption did not lead to recurrence of the skin lesions, suggesting that donepezil may be an alternative treatment when experiencing cutaneous adverse reactions to rivastigmine. M. C., S. P., H. C., G. B., C. D., V. B. and C. D. declare no conflict of interest. A. S. declares conflict of interest with Novartis, Bioproject, Leo Pharma and Sanofi. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.