Relationship Between Pisa Syndrome and Cholinesterase Inhibitor Use for Elderly Adults with Alzheimer's Disease

To the Editor: Drug-induced persistent dystonia called Pisa syndrome (PS), or pleurothotonus, was first described in 1972.1 Antiemetics, atypical neuroleptics, and antidepressants have been reported to cause PS. Cholinesterase inhibitors (CHEIs) were added later. PS is defined as an acquired, persistent truncal dystonia that appears to be potentially reversible. It is characterized by involuntary lateral flexion of the body and head associated with a backward axial rotation so that the person seems to be leaning like the leaning tower of Pisa.1, 2 This is a description without statistical analysis of individual case narratives of individuals with PS taking CHEIs reported in 15 articles in 13 years (including 10 cases in the north of France). The published observations are similar to the French cases. Individuals are not aware of their dystonia. There is no definite lateralization in the direction of the dystonic posturing. To which side the individual tilts is not often mentioned in articles. The most common adverse effects of CHEIs are related to cholinergic function and include nausea, vomiting, anorexia, and insomnia. The first cases were described in 2000.3 Twenty-seven cases of individuals with PS with probable Alzheimer's disease (AD) with CHEI were described between 2000 and 2014.3-6 The connection between CHEI treatment and PS is likely because, in most cases, the dystonia stopped with CHEI discontinuation, suggesting that CHEIs alone can induce reversible PS in individuals with AD. Predisposing factors that were considered, in addition to AD and taking CHEIs, were old age, being female, presence of an organic brain disorder, and previous treatment with classic neuroleptics.1, 2, 7 There are few data on the prevalence and incidence of PS induced by CHEIs because of the small number of cases and because randomized trials do not study long-term adverse reactions to drugs.7 Some authors estimate the incidence at 2 per 10,000 individuals per year taking CHEIs, which is fewer than with antipsychotics. In a Canadian study, a prevalence rate over a 5-year period of 8.3% was reported, whereas a German study found a lower prevalence rate of 0.037%.1, 2, 7 The three available CHEIs (donepezil, rivastigmine, galantamine) have been implicated independent of the dose or pharmaceutical form used (oral or transdermal patch).4-6, 8 There are two types of dystonia. Some older adults develop clinical features of acute dystonia, whereas others develop delayed dystonia. The duration of CHEI use before the onset of dystonia is variable: from a few days to several years. The underlying pathophysiology of drug-induced PS is not well known. A dopaminergic–cholinergic imbalance, induced by CHEI use and resolved with CHEI withdrawal, is thought to be the main causal factor of PS.1-3, 9 A recently published U.S. pharmacovigilance study of 26 individuals with AD with PS taking CHEIs also supported a potential dopaminergic-cholinergic imbalance as an underlying mechanism for PS.8 Other neurotransmitters, such as serotonin or noradrenalin, may also cause PS.1 PS may therefore be a syndrome with multiple etiologies. It is unclear whether CHEI-induced dystonia increases the risk of falling or whether individuals with PS lean but do not fall, just like the Tower of Pisa. Individuals can maintain activities such as sitting and walking.4, 5 The treatment of PS is not well established. The persistent truncal dystonia appears to be reversible with CHEI discontinuation (Table 1). Administration of the minimally effective dose or discontinuation of CHEI is recommended.4, 5 Other pharmacological strategies that have brought about an improvement in PS include amantadine and injections of botulism toxin.5, 9 In most cases, dystonia is noted again after a gradual reintroduction of CHEIs (a different one, the same one, or a different pharmaceutical form of the same one: oral or transdermal). Memantine is therefore proposed as an alternative to treat AD for an optimal long-term care strategy.10 Dystonia resolved in 7 days Rivastigmine initiated 10 weeks later Dystonia noted again Dystonia resolved with CHEI discontinuation 1 week later Dystonia resolved in 1 month Rivastigmine initiated 2 months later Dystonia noted again after 10 days Dystonia resolved with CHEI discontinuation 3 weeks later Dystonia resolved 1 week later Donepezil reinitiated 10 weeks later Dystonia noted again CHEIs discontinued Dystonia resolved 1 week later Risperidone discontinued Dystonia resolved in 6 days No improvement Botulinum toxin injections Dystonia resolved Dystonia resolved Donepezil initiated later Dystonia noted again Dystonia resolved in 16 days Dystonia resolved Rivastigmine initiated later Dystonia noted again 2 months later Dystonia resolved with CHEI discontinuation 1 week later Dystonia resolved 2 weeks later Rivastigmine initiated later Treated previously with rivastigmine Also given atypical antipsychotics Dystonia resolved Rivastigmine initiated later CHEI discontinued because of bradycardia Treated previously with donepezil Also given atypical antipsychotics Treated with tranquillizers Given atypical antipsychotics Also given sertraline and valproic acid Memantine initiated later Dystonia resolved Rivastigmine initiated later Dystonia resolved Rivastigmine initiated later 1 relapse 5 months later Dystonia resolved No recurrence This report suggests a connection between PS and CHEIs. The widespread use of CHEIs may cause an increasing number of PS cases. Physicians should alert individuals and their families about the possibility of PS. To improve care quality, systematic efforts to enhance clinician awareness and reporting all adverse events with CHEIs use have become essential. Effectiveness, safety, and pharmacovigilance of CHEIs will consequently be improved. The authors thank Evelyne Guillautou for her assistance. Conflict of Interest: None. Author Contributions: The authors contributed equally to this work. Sponsor's Role: No sponsor.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.