Managing abnormal eating behaviours in frontotemporal lobar degeneration patients with topiramate


Dr Shunichiro Shinagawa, MD, PhD, Department of Psychiatry, Jikei University School of Medicine, 3-25-8 Nishi-shinbashi, Minato-ku, Tokyo 105-8461, Japan. Email:


Abnormal eating behaviours are specific to frontotemporal lobar degeneration and increase caregiver burden. Topiramate, an anticonvulsant, suppresses cravings for alcohol and other substances and is a potential treatment for binge eating. However, there are few reports on topiramate efficacy for abnormal eating behaviours in frontotemporal lobar degeneration patients. We present three Japanese frontotemporal lobar degeneration patients with abnormal eating behaviours. Topiramate was effective, especially for compulsive eating, in cases with distinct lobar atrophy, but not for all abnormal eating behaviours.


Frontotemporal lobar degeneration (FTLD) features prominent behavioural disturbances such as loss of insight, disinhibition, apathy, mood changes, stereotypic behaviours and abnormal eating behaviours at presenile or senile ages.1 High prevalence of altered food preferences and eating habits in FTLD patients, especially with frontotemporal dementia (FTD) and semantic dementia,2–5 increase caregiver burden. Research has shown that selective serotonin reuptake inhibitors improve behavioural symptoms, especially stereotyped behaviours.6 However, definitive psychopharmacologic strategies for abnormal eating behaviours are lacking. Herein, we suggest topiramate, an anticonvulsant, as a potential treatment for this problem.


To maintain patient privacy, personal information was altered. Written informed consent was obtained after explaining treatment benefits and possible side effects. The Neuropsychiatric Inventory (NPI) was administered to evaluate all behavioural problems.7 To evaluate abnormal eating behaviour quantitatively, a caregiver-based eating behaviour questionnaire was administered.4 It consisted of 36 items investigating five domains: swallowing, appetite, food preferences (including sweets and food fads), eating habits (including stereotypic eating behaviours and diminished etiquette), and other oral behaviours (e.g. food cramming, indiscriminate eating). If caregivers agreed that a particular behaviour occurred, they were asked to rate the frequency and severity, to derive a frequency × severity score. The questionnaire is available from the authors or the Journal website. Patients' weights changes were also measured.

Case 1

A 78-year-old woman visited us with a 2-year history of irritability, worsening conversation responses and going out without purpose. She eventually repeated other's words in daily conversation and abandoned housework. Neuropsychiatric examination showed disinhibition, euphoria, compulsive eating behaviour and apathy. She had history of hypertension and no family history of dementia. Laboratory test results were unremarkable. Cranial magnetic resonance imaging showed bilateral dorsolateral prefrontal cortical dominant atrophy. FTLD, frontal variant-FTD, was diagnosed. Dietary change was highly problematic. She ate the same meals compulsively, with extremely rapid oral overfilling. Eliminating foods from her home (i.e. moving or hiding unnecessary foods) was attempted, but with little success. Fluvoxamine (maximum 150 mg/day) for 10 weeks did not improve her eating behaviours. Then, topiramate 50 mg/d was commenced for 2 weeks, and the dosage was subsequently increased to 100 mg/d. Oral overfilling and compulsive rapid eating disappeared. Stereotypic eating of the same meals decreased. Other stereotypic behaviours such as stereotypic speaking and stereotypic walking also decreased to some extent. No adverse effects occurred. The patient's pretreatment NPI and eating behaviour questionnaire scores were 20 and 82, respectively, and after 6 weeks, they decreased to 16 and 28. The patient's body weight was 5 kg greater than her weight when her condition began, but she lost 3 kg during the 6 weeks of treatment. After treatment, there were no further weight fluctuations. Among eating behaviour questionnaire subscales, eating habits and other oral behaviours showed moderate relief.

Case 2

A 47-year-old man showed inappropriate speech at work and became increasingly argumentative 2 years before visiting us. His work errors increased, and he was eventually forced to retire. Neuropsychiatric examination showed fluent aphasia, anomia, impaired comprehension of word meanings, agitation, euphoria and stereotypic behaviour. He did not have any past history of mental of physical disease, and he had no family history of dementia. Laboratory test results were unremarkable. Cranial magnetic resonance imaging showed cortical atrophy, dominant in the left temporal pole. FTLD, temporal variant-FTD (semantic dementia), was diagnosed. Six months later, sexually deviant speech toward his family and sweet food preferences gradually manifested, producing a major caregiver burden for his family. The patient was given an atypical antipsychotic agent, quetiapine (maximum 50 mg/day), for 4 weeks, which produced no improvement and caused him to complain of drowsiness. Quetiapine was stopped, and topiramate 25 mg/day was commenced. After two weeks, the dosage was then titrated up to 100 mg/day, and the patient did not complain drowsiness. Sweet food preferences persisted, but the familial burden decreased. Sexually deviant speech toward his family continued, but it was not as persistent. The patient's pretreatment NPI and eating behaviour questionnaire scores were 35 and 60, respectively, and after 6 weeks of treatment, they were 25 and 22. There was no significant change in body weight during the observation period. Among eating behaviour questionnaire subscales, appetite and eating habits showed moderate relief.

Case 3

A 57-year-old man visited us for an accurate diagnosis and medical advice. He had a 2-year history of apathy and difficulty going to work. He had been diagnosed with depression. Despite his taking antidepressants, the patient's apathy did not improve, so he visited us. Neuropsychiatric examination showed loss of interest in personal affairs and responsibilities, social withdrawal, apathy and euphoria. He did not have a depressed mood, previous manic episodes or a family history of mood disorder. He had a history of hyperlipidaemia. Laboratory test results were unremarkable. A few months later, he began to eat other people's food in public refrigerators and supermarket items without permission. He was euphoric, and mood stabilizers (e.g. valproic acid up to 600 mg/day for 6 weeks) were not effective. Cranial magnetic resonance imaging showed no remarkable localized atrophy, but frontal dominant diffuse atrophy was observed. FTLD without lobar atrophy was strongly suspected. Topiramate 50 mg/day was commenced and then titrated up to 150 mg/day after two weeks. However, there was little improvement in his abnormal eating behaviours, and no improvement in other behavioural problems such as apathy and euphoria. Thus, he was hospitalized afterward. The patient's pretreatment NPI and eating behaviour questionnaire scores were 32 and 66, respectively, and after 6 weeks of treatment, they were 29 and 58. There was no significant change in body weight during the observation period. (A summary of the three FTLD patients with abnormal eating behaviours is shown in Table 1.)

Table 1. Three patients with abnormal eating behaviours
 Case 1Case 2Case 3
  1. FTD, frontotemporal dementia; FTLD, frontotemporal lobar degeneration; MMSE, Mini-mental State Examination; NPI, Neuropsychiatric Inventory.

DiagnosisFrontal variant-FTDTemporal variant-FTDFTLD without lobar atrophy
Previous medicationFluvoxamine 150 mg/dQuetiapine 50 mg/dValproic acid 600 mg/d
Maximum dose of topiramate100 mg/d100 mg/d150 mg/d
MMSE score161020
NPI total (pretreatment)203532
NPI total (after 6 weeks of treatment)162529
Eating behaviour questionnaire (pretreatment)826066
Eating behaviour questionnaire (after 6 weeks)282258
Change in body weight (during 6-week period)−3 kgNoneNone
Adverse effectsNoneNoneNone


This report highlights possible resolution of severe abnormal behaviours in FTLD patients using topiramate, a relatively new anticonvulsant that has been increasingly used as a mood stabilizer in bipolar disorder. Appetite suppression is a well-known side effect of topiramate, which has been suggested as a treatment for obesity, particularly when complicated by binge eating.8–10 It has also shown promise in suppressing cravings in alcoholism,11 drug abuse,12 and pathological gambling.13

Though the mechanism of its effects on abnormal behaviours in FTLD remains unclear, one possibility is that topiramate suppresses appetite or enhances satiety through neuropeptide Y or glutamate systems. Neuropeptide Y is one of the most potent anorexigenic peptides within the hypothalamus.14 Also, glutamate antagonism of the nucleus tractus solitarius decreases food intake.15 Topiramate reduces glutamate neurotransmission by antagonizing α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPA)/kainate glutamate receptors.16 A second possible mechanism is that topiramate might directly induce weight loss. Generally, it has been hypothesized that carbonic anhydrase inhibitors induce weight loss by inhibiting carbonic anhydrase-mediated de novo lipogenesis.17 Although the patient in case 1 lost 3 kg, there were no significant changes in body weight in the other two cases. Therefore, we believe that the effect of topiramate on abnormal eating behaviour may not be due to just lost appetite. A third possible mechanism is that topiramate affects eating behaviour by altering the rewarding properties of food. Previous researches suggest that topiramate is effective in patients with alcohol and cocaine dependence,11,12 possibly by reducing the rewarding properties of these drugs through its GABAergic and antiglutamatergic properties combining to modulate cortico-mesolimbic dopamine function. A fourth possible mechanism of topiramate for abnormal eating behaviour is its general effect on impulsivity rather than a specific effect on eating behaviour. Topiramate is associated with reductions of other impulsive behaviours, such as alcohol and cocaine use,11,12 binge eating and purge behaviours in bulimia nervosa,18 and aggression in borderline personality disorder.19 These disorders may share a similar biological background of impulsivity, and topiramate might benefit these conditions by an anti-impulsivity effect that spans diagnostic categories. In our series, topiramate was effective especially for stereotypic and compulsive eating behaviours in cases with distinct lobar atrophy, but not for all abnormal eating behaviours or all other behavioural problems. The reason for topiramate being ineffective in case 3 is unclear, but it may relate to the patient's initial apathy and lack of typical lobar atrophy. Furthermore, the abnormal eating behaviour in case 3 was associated with euphoria in contrast to the other two cases, which showed stereotypic and compulsive eating behaviour. Therefore, we assume that the mechanism that affects abnormal behaviours in FTLD is the third or fourth hypothesis.

This effect was noted in previous FTLD case reports.20,21 However, these reports did not quantitatively evaluate eating behaviour as in this report. Therefore, we believe our results to be important. However, our report is based on only three cases; our sample size is too small to draw conclusions, just speculations. A future prospective, placebo-controlled study with more patients is needed to clarify topiramate's effects. We hope this new pharmacological approach will provide an option for FTLD treatment and relieve caregiver burden.