Taiki Kambe, Kaoru Mitsuhashi
Fluctuating cognition is a core feature of dementia with Lewy bodies (DLB). The DLB Consortium also recognizes syncope and other transient episodes of unresponsiveness (TEU) as supportive clinical features [1]. The boundaries and management of TEU remain poorly defined. A recent report highlighted unexplained coma or unresponsiveness in Lewy body disorders [2], and a Japanese case report described syncope attributed to autonomic dysfunction in Lewy body disease that resolved after low-dose donepezil [3]. However, longitudinal accounts of TEU in medication-sensitive older adults with probable DLB remain limited, including dose-time relationships during very-low-dose donepezil titration and after withdrawal of sedating medications [1, 4-7]. We describe such a case. An 85-year-old Japanese woman developed cognitive decline, anxiety, and dependence in daily activities, with concurrent bradykinesia. Approximately 4 months before donepezil initiation, her Mini-Mental State Examination score was 18/30. Brain magnetic resonance imaging (MRI) showed no acute lesion or focal atrophy, electroencephalography showed no epileptiform discharges, and perfusion scintigraphy demonstrated left medial occipital hypoperfusion. Ramelteon 8 mg/day had been prescribed for nocturnal anxiety and insomnia, and low-dose quetiapine was later added for the same indication. Sleep improved, but marked daytime somnolence occurred even at quetiapine 12.5 mg/day, prompting discontinuation. Ramelteon was reduced to 4 mg/day, but nocturnal anxiety recurred and quetiapine was reintroduced at 6.25 mg/day with caution. Tramadol also caused somnolence and was replaced with nonopioid analgesics, suggesting sensitivity to sedating medications. During 1–2 months before donepezil initiation, more profound episodes emerged at home and at adult day care. They were characterized by abrupt, sleep-like unresponsiveness with closed eyes and no response to voice or touch, lasting several minutes to approximately 1 h. When seated, she sometimes remained sitting, but at other times she slumped and required support; other episodes occurred in bed. Breathing remained regular, with no diaphoresis or pallor. No convulsions, tongue biting, forced eye deviation, persistent focal neurologic deficit, or postevent confusion were observed, and recovery was gradual. Formal orthostatic testing before donepezil initiation did not demonstrate orthostatic hypotension. Two prolonged episodes, each lasting approximately 40 min, prompted emergency evaluation. In both, the Glasgow Coma Scale score was 6 (E1V1M4), but respiration and circulation remained stable. Blood tests, computed tomography, brain MRI, electrocardiography, cerebrospinal fluid analysis, toxicology screening, and electroencephalography showed no cardiopulmonary, structural neurologic, infectious, toxicologic, epileptic, or metabolic cause. At our initial assessment, 1 month before donepezil initiation, she had cognitive fluctuation and daytime somnolence, with brief periods of reduced responsiveness, described as “spaced out” that improved with verbal prompting. Parkinsonism was present (Unified Parkinson's Disease Rating Scale part III, 36). Probable REM sleep behavior disorder was reported. She met consensus criteria for probable DLB, and occipital hypoperfusion on perfusion scintigraphy and the presence of TEU further supported this clinical impression [1]. Dopamine transporter imaging and myocardial metaiodobenzylguanidine scintigraphy were not performed. Donepezil was initiated primarily to address cognitive decline and fluctuating cognition rather than TEU itself. Given frailty and medication sensitivity, treatment was started at 0.75 mg/day by quartering a 3-mg tablet. Apart from transient appetite loss for 2 days, no adverse effects were observed. Two weeks later, the dose was increased to 1.5 mg/day. Thereafter, TEU frequency decreased. TEU recurred intermittently over the next 17 months. Discontinuation of quetiapine at 5 months and ramelteon at 13 months was not associated with an obvious change in TEU frequency or pattern. Because TEU frequency increased slightly at 17 months, donepezil was increased to 2.5 mg/day. One further TEU, lasting 20 min, occurred 4 days after dose escalation. No additional episodes were documented thereafter during 18 months of follow-up. Gait and appetite did not worsen after dose escalation. See Figure 1. A Japanese report described syncope attributed to autonomic dysfunction in Lewy body disease that improved after low-dose donepezil [3]. Our case adds a complementary perspective on TEU in probable DLB with cognitive fluctuation, longitudinal dose-time data, and the course after withdrawal of sedating medications. TEU decreased after titration to 1.5 mg/day and, after escalation to 2.5 mg/day, did not recur apart from one episode 4 days later. Although quetiapine required cautious use in this medication-sensitive patient with probable DLB, lack of improvement after discontinuation of quetiapine and ramelteon made medication-induced oversedation less likely. Important limitations remain: despite unrevealing emergency evaluation and negative orthostatic testing, physiologic monitoring during later episodes was limited; dopamine transporter imaging and myocardial metaiodobenzylguanidine scintigraphy were not performed; and quarter-tablet dosing may have introduced variability. This case suggests that cholinergic optimization, including very-low-dose initiation and stepwise donepezil titration, may be a pragmatic management strategy for TEU in a medication-sensitive older adult with probable DLB. Taiki Kambe conceived the report and drafted the manuscript. Kaoru Mitsuhashi contributed to data collection and critical revision of the manuscript. Both authors read and approved the final manuscript. The authors have nothing to report. The authors have nothing to report. Written informed consent for publication was obtained. The authors declare no conflicts of interest. The data supporting this case report are not publicly available because they contain potentially identifiable personal clinical information but may be available from the corresponding author on reasonable request.