Activation syndrome caused by paroxetine in patients with dementia with Lewy bodies


Dr Takemi Kimura MD PhD, Division of Clinical Research, Kikuchi National Hospital, 208 Fukuhara, Koshi 861-1116, Japan. Email:


Activation syndrome is a severe adverse effect caused by antidepressants. It is more frequently induced by paroxetine than by other selective serotonin reuptake inhibitors. However, there have been no descriptions of demented patients with activation syndrome derived from paroxetine. Herein, we report on three cases of dementia with Lewy bodies who suffered from activation syndrome induced by paroxetine. These symptoms rapidly disappeared following discontinuation of paroxetine. The present cases lead us to speculate that paroxetine is injurious to patients with dementia with Lewy bodies.


Paroxetine is a selective serotonin reuptake inhibitor (SSRI) that has been used widely in the treatment of depression. Recently, paroxetine has been also prescribed to patients with panic disorder, social anxiety disorder, and obsessive–compulsive disorder. The adverse effects of paroxetine include gastrointestinal disturbance, drowsiness, sexual dysfunction, hyponatremia, activation syndrome (AS), and discontinuation syndrome. Of these adverse effects, AS is the most serious and significant because it is intimately associated with the risk of suicidal behavior in patients with depression.1

Kosaka et al. suggested a new type of dementia, dementia with Lewy bodies (DLB), by presenting 12 cases.2 Consensus guidelines for the clinical and pathological diagnosis of DLB were reported by the Consortium on DLB international workshop.3 Fluctuation in cognitive function, persistent well-formed visual hallucinations, and spontaneous motor features of parkinsonism are core features with diagnostic significance in discriminating DLB from Alzheimer's disease and other dementias. The DLB consortium has since revised the guidelines, incorporating new information and giving greater diagnostic weight to severe neuroleptic sensitivity, disorders in REM sleep, and reduced striatal dopamine transporter activity on functional neuroimaging as features suggestive of a diagnosis of DLB.4 In association with extreme sensitivity to neuroleptics, we suppose that patients also exhibit increased sensitivity to other drugs. It is thought that AS is a severe adverse effect and a hypersensitive symptom induced by paroxetine.5 Herein, we report on three patients with DLB who suffered from AS caused by paroxetine and discuss pharmacotherapy in DLB.


Case 1

This patient was an 87-year-old woman who was previously high functioning. Seven years prior to presentation, she sometimes experienced pain in her mouth. Although her dentist examined her repeatedly, no significant changes were observed in her mouth; the patient was not satisfied with this outcome. Two years ago the patient felt intermittent pain in the mouth and started to become less and less active. When she was experienced chest pain and difficulty breathing 8 months ago, she was transported to an emergency hospital, but no respirato-circulatory abnormalities were found. Because the respirato-circulatory attack had caused the patient considerable distress and continued to do so, she visited a clinic and was prescribed paroxetine (10 mg/day). Three days after beginning treatment, she got up at 4 o'clock in the morning and repeatedly telephoned her daughters. Because she was highly talkative, disinhibitory, irritable, and aggressive, she was admitted to our hospital.

A neuropsychiatric examination revealed a manic-like state, including talkativeness, hyperanakinesia, disinhibition, agitation, irritability, and insomnia. There were no unusual findings on laboratory tests. Cranial magnetic resonance imaging (MRI) indicated mild atrophy of the cerebral cortices. The patient's score on the Mini-Mental State Examination (MMSE) was 22.

Agitation and irritability disappeared 3 days after discontinuation of paroxetine, whereas talkativeness, hyperanakinesia, disinhibition, and insomnia improved after 4 days. However, the patient could only walk taking very small steps and exhibited rigidity, disorders in REM sleep, orthostatic hypotension, syncope, and constipation, and occasionally presented with cenesthopathy and a delusion that her daughter had had an affair with unknown men and had died. The appearance of these psychiatric symptoms was influenced by fluctuations in the patient's cognitive functions. The psychiatric symptoms were treated with aripiprazol (10 mg/day).

Case 2

This patient was an 85-year-old woman who had been well previously. Seventeen years ago, she presented appetite loss, insomnia, headache, general fatigue, decreased volition, depressive mood, and lying down all day long. She was admitted to a psychiatric hospital. Antidepressant treatment for 2 months reduced her symptoms, but after that she experienced similar symptoms on 14 separate occasions and was treated in hospital each time. Since 1 year ago, the patient had described visual hallucinations of children in addition to having a depressive mood. Although she was at home, she kept repeating that she was going to go home. When she came to our hospital, she told us that unknown men came to her home every evening and she also asked where her sister (who had already died) was. A neuropsychiatric examination revealed visual hallucinations, fluctuations in cognitive function, delusions (nurturing syndrome), depressive mood, decreased volition, dry mouth, dysarthria, rigidity, difficulty walking, and REM sleep disorders. Laboratory tests were normal. Cranial MRI showed mild atrophy of the cerebral cortices and an old small infarction in the right putamen. The patient's MMSE score was 14.

Two days after switching from maprotiline (50 mg/day) to paroxetine (10 mg/day) to decrease the symptoms of dry mouth, dysarthria, and depressive state, the patient started suffering from insomnia and restlessness. She cut her left forearm using a kitchen knife approximately 20 times. When she was admitted to our hospital, she said that she was feeling irritable, did not know what to do, and cut her forearm in confusion.

Two days after discontinuation of paroxetine, the patient's irritability, akathisia, and insomnia improved. Visual hallucinations, fluctuations in cognitive function, and insufficient volition were observed 10 days after discontinuation of paroxetine. The remaining symptoms were relieved by donepezil (1.5 mg/day) treatment for 1 week.

Case 3

This patient was an 89-year-old woman who had been well previously. Five years ago, she suffered from memory impairment and severe constipation and sometimes became anxious and/or entered a panic state. Two years ago, she presented with auditory hallucinations, delusions of persecution, and wandering at night, and occasionally failed to recognize her family. The patient's intermittent agitation at night distressed and exhausted her caregivers. Therefore, the patient visited a psychiatric hospital and was given paroxetine (20 mg/day). Several days after starting treatment, the patient was extremely talkative, restless, and irritable. Consequently, she was admitted to our hospital.

A neuropsychiatric examination indicated a manic-like state, including insomnia, talkativeness, irritability, and akathisia. There were no irregular findings on laboratory tests. Cranial computed tomography revealed mild atrophy of the cerebral cortices. The patient's MMSE score was 11.

Two days after discontinuation of paroxetine, irritability and akathisia disappeared. Another 2 days after that, insomnia and talkativeness improved. However, the patient still showed fluctuations in cognitive function, rigidity, and autonomic dysfunction, including orthostatic hypotension, syncope, and constipation. Fluctuations in cognitive function were reduced by donepezil (1 mg/day) treatment for 2 weeks.


Case 1 is considered to be mildly demented, because her MMSE score was 22. This patient presented with fluctuations in cognitive function and parkinsonism (rigidity and difficulty walking), which are the core symptoms of the consensus guidelines for the clinical diagnosis of DLB.3 Thus, this patient was diagnosed as probable DLB, with the diagnosis supported by the occurrence of REM sleep disorders and orthostatic hypotension, which are often found in patients with DLB.

With an MMSE score of 14, Case 2 is likely moderately demented. This patient showed all the core symptoms of DLB, including visual hallucinations, fluctuations in cognitive function, and parkinsonism (rigidity and difficulty walking). According to the guidelines,3 this patient was diagnosed as probable DLB. The presence of REM sleep disorders supported the diagnosis.

Case 3 was considered to be moderately demented based on her MMSE score of 11. She presented with fluctuations in cognitive function and parkinsonism (rigidity), which are the core symptoms reported in the guidelines.3 This patient was diagnosed as probable DLB and the autonomic dysfunction observed corresponds with the diagnosis.

Activation syndrome is a severe adverse effect caused by antidepressants and it occurs more frequently with paroxetine than with any other SSRI. The discussion paper from the US Food and Drug Administration lists 10 symptoms of AS, specifically anxiety, agitation, panic attacks, insomnia, irritability, hostility, impulsivity, akathisia, hypomania and mania.6 Several days after starting paroxetine, all three patients in the present series exhibited a manic-like state or akathisia with suicidal behavior. These symptoms disappeared 2–4 days after discontinuation of the drug and so we suppose that the paroxetine was responsible for the AS in the present cases. Different symptoms of AS were observed in the three cases. Case 2 exhibited irritability and akathisia, but not manic-like symptoms, whereas Case 3 did not exhibit the hyperanakinesia and disinhibition observed in Case 1, indicating that the symptoms shown by Case 1 are more similar to the manic state than those of the other two patients. Although the precise reason for the differences is not known, it is likely that the severity of DLB, vascular changes in the brain, and the dose of paroxetine administered may have contributed to the different symptoms seen. The differences between the manic-like state of AS in DLB and the manic state induced by antidepressants are: (i) obvious irritability and/or akathisia without exaltation is seen in the manic-like state of AS; and (ii) the manic-like state of AS improves several days after discontinuation of the antidepressant, but this does not occur in cases of a manic state induced by antidepressants. In all three patients in the present series, irritability and akathisia were resolved soon after discontinuation of paroxetine.

It is assumed that AS appears in approximately 4.3% of patients taking paroxetine,7 whereas AS in adults is considered rare.8 To our knowledge, there are no reports of AS caused by paroxetine in patients with dementia. Previously, we described that paroxetine was useful in the treatment of patients with frontotemporal dementia to relieve their behavioral symptoms and without causing AS.9 Based on these findings, the present report of three cases of AS following paroxetine treatment is extremely interesting and suggests that AS with paroxetine treatment may be found in patients with DLB, but not other dementias.

The reason for the high prevalence of AS in patients with DLB is not known. The SSRIs selectively increase 5-hydroxytryptamine (5-HT) receptor binding in the brain and it is possible that excessive stimulation of 5-HT receptors is linked to AS. In addition, 5-HT neurons regulate dopaminergic and norepinephrinergic systems in the brain. Therefore, rapid changes in 5-HT may induce abnormal activity in other monoaminergic neurons, leading to AS-like symptoms. Because patients with DLB show severe adverse effects to neuroleptics, there may be an underlying fragility of the monoaminergic system in the brain of patients with DLB. Given this increased fragility, paroxetine most likely causes AS in these patients.

Tricyclic and tetracyclic antidepressants aggravate the cognitive function of patients with DLB as a result of their anticholinergic action. Thus, some psychiatrists use SSRIs or serotonin–norepinephrine reuptake inhibitors (SNRIs) to reduce the depressive state observed in DLB patients. Because there is no evidence supporting the efficacy of these drugs, the usefulness of these antidepressants is highly contentious. As described in the present report, paroxetine induced AS in patients with DLB. Considering these pharmacological findings, administration of SSRIs or SNRIs to DLB patients to treat their depressive state is unlikely to be beneficial and is instead likely to be injurious to these patients as a result of the severe adverse effects of the drugs. As shown by Case 2, there are some patients with DLB who converted from depression. In these patients, antidepressants are effective for the treatment of the depression prior to the appearance of DLB, whereas the drugs may become ineffective and give rise to adverse effects once patients have DLB. Therefore, it is likely that antidepressants are not useful for the treatment of DLB patients. In our clinical experience, donepezil (5 mg/day) treatment occasionally makes DLB patients irritable and defiant, whereas a smaller dose of donepezil relieves their depressive state, visual hallucinations, and fluctuations in cognitive function, as shown by Cases 2 and 3. Based on these results, a small dose of donepezil may be useful for some DLB patients.

In the clinical field of the aged, it is not easy to differentiate dementia from depression. Furthermore, it is difficult to distinguish DLB from depression, because a depressive state is frequently found in DLB patients more than in individuals with Alzheimer's disease. The present case report suggests that there is a risk associated with the use paroxetine in DLB patients, which emphasizes the importance of an accurate diagnosis of DLB. To provide convincing evidence of the risk, more cases need to be evaluated.