Use of Low-Dose Gabapentin for Aggressive Behavior in Vascular and Mixed Vascular/Alzheimer Dementia

Behavioral and psychological symptoms of dementia (BPSD) are common symptom manifestations in dementia. They may occur at any stage of the illness, although are more common in the latter stages. More than two-thirds of patients with dementia will suffer these symptoms during the course of the illness.¹ They can be very distressing for the patient and may crucially undermine the caregiver’s capacity to continue to provide care for the dementia sufferer in his or her usual setting. Aggressive behavior in dementia is common, often precipitates referral to specialist services, and is a major cause of institutionalization.² Although all efforts should be made to address these symptoms by trying to determine their underlying cause and using nonpharmacological approaches, recourse to pharmacological agents may be necessary.

Risperidone is the only drug licensed for BPSD in the United Kingdom and Ireland and is approved for short-term (up to 6 weeks) treatment of persistent aggression in patients with moderate-to-severe Alzheimer's dementia (AD) unresponsive to nonpharmacological approaches and when there is a risk of harm to self or others. However, although both typical and atypical antipsychotic medications have the best evidence-base, when used to treat BPSD,³ they have been associated with both an increased risk of cerebrovascular accidents (CVAs) and higher mortality in this context.⁴ Also, their use is associated with deterioration in cognitive functioning.⁵ A recent U.K. report has highlighted the need to significantly curtail the use of antipsychotic medication in patients with BPSD and emphasized the need to research alternative pharmacological agents to manage these symptoms.⁶ Although current guidelines recommend short time-frames for antipsychotic treatment of aggression in dementia, it is clear that these symptoms tend to be relatively persistent, with aggression showing an increased risk of severity over time.⁷ Therefore the recommendation that antipsychotics should be used for short periods of time may underestimate the tendency of these symptoms to persist over time. As Jeste⁴ has so aptly noted, the pharmacological management of BPSD is “a clinical conundrum …with no immediate or simple solutions.”

The “ideal drug for BPSD” should be effective against the target symptom(s); have a low incidence of side effects, including no detrimental effects on cognitive functioning, and few drug interactions.⁴ The difficulty for the clinician is the paucity of randomized clinical trials of psychotropic medication to provide guidance on the management of these symptoms, and the need to make choices on the basis of clinical experience and judgment. Anticonvulsants have been used to treat these symptoms, with the best evidence-base for carbamazepine. However, randomized, controlled trials (RCTs)^8–11 have produced equivocal results.¹² Also, common side effects of carbamazepine include dizziness, drowsiness, ataxia, and cognitive impairment. Less common, but serious, adverse effects include cardiac dysrhythmia, aplastic anemia, liver failure, Stevens-Johnson syndrome, and syndrome of inappropriate antidiuretic hormone (SIADH).¹³ Carbamazepine is associated with significant drug interactions, and this limits its usefulness.

Gabapentin is an anticonvulsant drug structurally related to the central nervous system inhibitory neurotransmitter gamma-aminobutyric acid (GABA). It potentiates GABAergic activity, although it does not bind to the same site as GABA.¹⁴ It is not hepatically metabolized, does not bind to serum proteins, has minimal drug–drug interactions, and has few adverse effects.¹⁴ The main adverse effect noted in the literature has been its tendency to cause sedation at higher doses.

In AD, an association has been found between behavioral problems and deficits of GABA in brain tissue,^15,16 and this may provide a theoretical framework for the possible effectiveness of drugs acting on these pathways in patients with BPSD.¹⁷ However, gabapentin has actions on other neurotransmitter systems, including serotonin and glutamate,¹⁶ and its mechanism of action in BPSD is unclear. There is a developing literature supporting the off-label use of gabapentin for the treatment of BPSD¹⁴ in addition to its reported effectiveness in managing inappropriate sexual behavior in patients with dementia.^18,19

Methods

Subjects

The study was a retrospective design of consecutive cases of vascular or mixed vascular/Alzheimer dementia referred to the Department and treated with gabapentin over a 6-month period. (In view of recent concerns of the cerebrovascular side effects of antipsychotic medication in treating BPSD, we decided to use gabapentin in those patients theoretically at greatest risk; that is, those with vascular dementia or mixed dementia.) Seven patients who had been referred to the Department of Old Age Psychiatry Dublin South East were included in the study. All met ICD-10²⁰ criteria for vascular dementia or mixed Alzheimer vascular dementia. Each patient had evidence of vascular risk factors and exhibited severe aggressive behavior creating serious concern for their caregivers.

All patients referred to the Old Age Psychiatry service are screened by the referring doctor for evidence of physical illnesses (e.g., infections, constipation, pain, medication side effects), and the view of the referring and treating doctor was that the aggressive behavior was unrelated to any underlying medical or physical health problem complicating their dementia; four were nursing home residents, and three were living at home with family caregivers. Their ages ranged from 63 to 80 (mean age: 74); there were 6 men, 1 woman. Patients engaged in a range of challenging behaviors, including physical aggression causing soft tissue injury, verbal rage episodes, sexually inappropriate behavior, and aggressive behavior resulting in noncompliance with assistance with ADLs.

The most common antipsychotic medication used to treat behavioral problems in these patients at the time of referral was quetiapine, with 6 of the 7 cases receiving daily doses ranging from 25 mg to 150 mg; 3 patients were on treatment with SSRIs previously for depression (2 on escitalopram, 1 on citalopram), and 1 patient was being treated with both quetiapine and molipaxin 75 mg daily (both for BPSD). No patient was on antiepileptic drug treatment at the time of referral.

All patients (see Table 1) were started on gabapentin at doses of 100 mg twice daily and increased to a total daily dose range from 200 mg daily to 600 mg daily. Antipsychotic medication was discontinued in four patients before treatment with gabapentin (as there was clearly no clinically significant response to it), and it was uneventfully discontinued subsequent to gabapentin treatment in the other three patients. Patients were on gabapentin treatment at the time of review from 3 to 5 months. All patients tolerated gabapentin well, and all patients had a significant response to it, both in terms of clinically significant responses. It was felt in each case that, if the aggressive behavior had continued, patients would have had to be transferred to inpatient care or to alternative specialist long-term care facilities.

Illustrative Cases

Case 1

A 74-year-old male nursing home resident with a history of vascular dementia was referred by his general practitioner with a several-year history of aggressive behavior, with a recent escalation in assaultive episodes. These episodes of physical aggression appeared to have no obvious precipitant and had resulted in one resident sustaining bruising and soft-tissue injuries. Staff expressed concern that they might not be able to manage the patient in his current setting. On assessment, he had moderate cognitive impairment, with very poor short-term memory and no recall of his behavioral outbursts. Routine investigations were normal, and physical examination revealed no obvious physical cause for his behavior. At the time of his referral, he was on treatment with memantine 20 mg daily, donepezil 10 mg daily, and quetiapine 25 mg daily. After review, he was begun on gabapentin 100 mg bid, which was increased to 200 mg bid after 1 week. Within 2 weeks of starting treatment with gabapentin, there were no further aggressive episodes or sexually-disinhibited behavior, and, as a result of this change, a clinical psychology referral request was deferred and subsequently cancelled. Treatment with quetiapine was gradually discontinued without recurrence of aggressive behavior. The improvement was sustained at review 3 months later. Nursing home staff were satisfied with the intervention and were happy to continue to manage the patient in the nursing home.

Case 2

A 76-year-old male nursing home resident with a background history of vascular dementia was referred to the Psychiatry of Old Age Service with behavioral difficulties. These included intermittent aggression, agitation, and sexually-inappropriate behavior toward nursing staff and other residents. He had longstanding oppositional and aggressive behavior toward staff assisting him with his personal hygiene, grooming, and dressing.

He had a history of a CVA in 2009, after which he had a carotid endarterectomy. On evaluation, he presented as a thin, elderly gentleman, with an anxious affect. He had severe cognitive impairment, with very poor short-term memory and receptive and expressive aphasia. He had no apparent evidence of delusional thoughts or perceptual abnormalities. His mood was euthymic. Routine investigations were normal, and physical examination revealed no obvious physical cause for his behavior. His medication at the time of assessment included quetiapine 50 mg bid, which he had been on for 8 months, without significant benefit. He had been on citalopram 20 mg for 1 year for treatment of depression. He had also been receiving donepezil 10 mg daily for 1 year. He was begun on gabapentin 100 mg bid, and, within 1 week, his agitation had lessened considerably, and there were no further episodes of aggression or intrusive behavior. Staff were able to attend satisfactorily to his personal hygiene. Gabapentin was well tolerated. The nursing home staff and patient’s family were pleased with the effect of the medication. Follow-up 4 months later revealed that the response to treatment with gabapentin had been maintained. Treatment with quetiapine medication was discontinued gradually and uneventfully.

Case 3

This patient was an 80-year-old man with a 4-year history of gradually deteriorating memory, referred by his general practitioner, who was concerned about his increasing irritability. He had a history of coronary artery disease, atrial fibrillation, and hypercholesterolemia, and was being treated with warfarin. At clinical review, he had patchy cognitive deficits, with poor short-term memory, nominal dysphasia, and visuospatial impairment. An MRI brain scan revealed deep white-matter ischemic changes. A diagnosis of mixed vascular/Alzheimer dementia was made. He had a history of verbal rage episodes, which were precipitated by relatively minor disagreements with his wife. These episodes resulted in his shouting angrily and making threatening gestures toward his wife. She felt intimidated by him, although he had never been physically aggressive toward her. Full medical review by the Medicine for Older People team had not revealed any obvious physical factors that may have contributed to his rage episodes. An initial trial with memantine had been followed by deterioration in these outbursts, and it was discontinued. He was begun on quetiapine, which was gradually titrated up to 75 mg bid, with only partial control of these episodes. At subsequent review, his wife reported that she could no longer care for him at home if his aggressive behavior could not be managed promptly. Gabapentin was begun at a dose of 100 mg bid and increased after 1 week to 200 mg tid. Within 2 weeks, his wife noted that the rage episodes had diminished significantly. He has remained well on this dose of gabapentin after 3 months, and treatment with quetiapine has been discontinued without further verbal aggression.

Discussion

These cases provide support for the hypothesis that gabapentin at low doses can be an effective treatment for aggressive behavior in the context of dementia with a significant vascular component. Treatment with gabapentin, in these cases, has allowed discontinuation of antipsychotic medication without adverse effect. Also, gabapentin has been effective in low doses (200 mg–600 mg daily) and is well tolerated. The low doses used in this sample contrast with other case series described, where doses up to 3,600 mg daily have been used.²¹ No patient in this group, as might have been expected in view of the low doses used, developed sedation while on treatment, and the response to treatment in each case has been prompt and sustained.

Although based on a small sample, and using open-label treatment, these cases support a growing literature that gabapentin may have a useful role in treating aggressive behavior in dementia. Criticism of this type of case series report is that there is an inherent risk of bias favoring those cases that have responded to gabapentin. This issue can only be resolved by rigorously-conducted randomized, placebo-controlled trials comparing gabapentin with other agents. It is clear in these reports that the reduction in behavioral problems is not a result of sedation, as the doses used were too low to cause this problem. Even though no standardized behavioral rating scales were used to measure aggression in these patients, it is clear from the clinical impression of the treating doctor and the assessment of their caregivers, both formal and informal, that the improvement in aggressive behavior was unequivocally clinically significant. However, the use of standardized rating scales such as the NPI²² would have been helpful in giving a more objective and measurable assessment of response to treatment and might be helpful in determining what particular behavioral problems might respond to gabapentin.

Concerns regarding the adverse effects of antipsychotic medications in the treatment of BPSD and the lack of licensed medication in this area have created an imperative to investigate alternative agents. Anticonvulsants have been studied in this context with equivocal results. Also, the majority of existing studies and case reports have used anticonvulsants in addition to antipsychotic medication.¹² The case series reported here suggests that gabapentin may be effective as a stand-alone treatment. Although a recent review by an expert committee in this area has recommended further trials of carbamazepine to treat these symptoms,³ there is a pressing need to carry out RCTs on gabapentin, with its better tolerability and safety profile.

We have highlighted here the effectiveness in this case series of gabapentin treatment for aggressive behavior and agitation occurring in vascular or mixed dementia where one might expect a higher risk of CVA using antipsychotic medication. Whether gabapentin is more effective in treating these symptoms in vascular dementia than other types of dementia would need to be tested in RCTs with patients with mixed samples of dementia sufferers.

Caution should be noted about the use of gabapentin in Lewy-body dementia, where dramatic worsening of neuropsychiatric symptoms has been reported after its use to treat behavioral symptoms.²³ A review of the literature yielded no reports of an increased risk of cerebrovascular events in patients with vascular dementia treated with gabapentin. Indeed, gabapentin is the only antiepileptic drug that has been specifically evaluated in stroke patients showing a high rate of long-term seizure freedom.²⁴

Although this report suggests that gabapentin is a useful treatment in the short-term treatment of BPSD and, in low doses, trials with a longer follow-up period are needed to establish whether it maintains its effectiveness over time and what its safety profile is in this situation. The clinical and economic significance of safely managing aggressive behavior in dementia sufferers cannot be overstated. It may enable caregivers to continue to care for their dependents in the home environment and prevent inpatient psychiatric hospitalization, with the disruption and difficulty that this may cause for patients and their families.

This report adds to an emerging literature that gabapentin may have a useful role in the pharmacological management of BPSD and that it may be an effective alternative to antipsychotic use in this situation. If RCTs support its effectiveness, particularly at lower doses, then we will be one step closer to Jeste’s⁴ ideal drug.