Dr. Marcotte is director of the psychiatric outpatient clinic Marcotte and Associates in Fayetteville, NC.

Acknowledgments:The author reports no financial, academic, or other support of this work.  



Are there safe treatments for elderly patients with dementia and aggression? This article describes the use of gabapentin, a nonmetabolized antiepileptic drug, for control of aggression in the elderly patient with dementia. The drug’s relative safety and ease of use are demonstrated to assist in the control of aggressive behavior. The objective of the study was to determine the effectiveness of gabapentin in the acute management of behavioral disturbance in patients who had failed to respond to previous medications and had failed their nursing home placement.


With the aging of our population, psychiatrists will increasingly be called upon to provide services to nursing homes, general hospitals, and families of loved ones with dementia and aggression. Although traditional and atypical antipsychotics are commonly employed in the management of aggression, antiepileptic drugs are being used with growing frequency.

Dementia is a major health concern today. It is expected that in the next 50 years, the worldwide population of people >80 years of age will increase by 6-fold, to 370 million.1 With the aging of our population and enhanced life expectancy, the large number of baby boomers will shortly reach the ages at which dementia tends to occur, presenting a greater challenge to medical resources and the economic welfare of families.2,3 Those of us who have worked with families that maintain an elderly demented patient in their home are witness to the emotional stress and financial burden that caretaking involves. Factors that affect the family’s ability to care for an elderly relative in the home include incontinence, aggressive behaviors toward others, behaviors resulting in self-injury (eg, falling), or wandering from the premises.

Many nursing homes are equipped with devices that monitor a patient’s whereabouts (eg, anklet or door alarms). However, while nursing home staff may be familiar with patients who wander or are incontinent, they may not be equipped to handle aggressive behaviors that threaten staff members or other residents of the nursing home. Aggressive behavior, considered an immediate crisis within the patient’s home or the nursing home, frequently leads to psychiatric hospitalization. To maintain the possibility of having the patient return to the nursing home, families are often taxed with additional costs, such as paying for the vacant nursing home bed during the patient’s hospitalization in a psychiatric unit.

Patients with dementia hospitalized for other medical procedures in a general hospital have longer lengths of stay.4 Lyketsos and colleagues4 studied 823 patients in a general hospital and found that the average length of stay for patients with dementia exceeded that for patients without dementia by 4 days. There were higher costs of hospitalization and greater complications. Unfortunately, that study did not differentiate delirium from dementia. A substantial portion of those patients who exhibited demented behavior may have qualified for the diagnosis of delirium.5 Patients with a diagnosis of dementia who were admitted to a general hospital were found not to have higher rates of mortality in the hospital. Another study by Lyketsos and colleagues6 noted that of patients with dementia, 27% had apathy, 24% had depressive disorders, and 24% had aggression and agitation. Although apathy and depression were noted to have significant effects on the individual and earlier nursing home referral, a worse prognosis accompanied those patients who had aggression and agitation. Such symptoms also increased the cost of caregiver burden.7

Not only does aggressive and behavioral disturbance such as agitation lead to early nursing home placement, it can lead to expulsion from the nursing home. Behaviors that include aggression toward others result in more costly expenditure, greater morbidity, higher mortality, and increased financial burden.4,6 In addition, the symptoms of agitation and aggression become more significant and frequent as dementia becomes more advanced. Lyketsos and colleagues6 studied patients with symptoms of aggression and agitation and found that 13% had mild dementia, 24% had moderate dementia, and 39% had severe dementia.

The large expected increase in patients with dementia and aggression will produce significant burden for psychiatric hospitals and nursing homes.  Psychiatric care and management of aggressive symptoms must be obtained before the patient can return to the nursing home, even after the hostile behaviors have been ameliorated. Thus, length of hospital stay for general medical purposes is expected to increase.

This article examines the use of gabapentin in a traditional inpatient setting, including patients ≥65 years of age who were both demented and aggressive. Gabapentin, a relatively nontoxic, nonmetabolized, nonplasma-bound antiepileptic drug, was used in addition to a minimal amount of atypical antipsychotics. The results indicate that gabapentin offers a safe alternative to metabolized, plasma-bound antiepileptic agents.


Recent treatment of behavioral disturbance with aggression in the elderly has included anticonvulsants, traditional antipsychotics, and novel antipsychotics. The use of anticonvulsants has a substantial advantage over antipsychotics; anticonvulsants are less anticholinergic, thus they are less likely to contribute to increasing dementia.8-16 Anticholinesterase medications have been used to decrease the enzyme acetylcholinesterase to preserve acetylcholine (ACH) and increase mental acuity. Anticonvulsants have less impact on ACH and may be less harmful to memory, attention, and concentration in demented patients. There have been more reports of the use of gabapentin in the treatment of behavioral disturbance in the elderly.17-22 Gabapentin has a unique advantage because it does not plasma bind, displace other medications, or cause drug-medication interactions. Gabapentin is not metabolized in the body and 95% of the drug is excreted in the urine. This obviates problems associated with liver toxicity or other metabolic concerns in the cytochrome P450 system. Because it is excreted in the urine, excessive quantities of gabapentin can be accumulated in those patients with renal failure. Gabapentin doses must be reduced in such patients.


Patients treated with gabapentin over 3 years (N=210) through a small community hospital service were retrospectively reviewed. Gabapentin blood levels were obtained from a small number of patients during the course of this study (BJ Wilder, MD, oral communication, 1996). All patients who underwent treatment with gabapentin were selected from this pool. Only patients ≥65 years of age were identified and those with dementia and behavioral disturbance were included in the study. Of the patients >65 years of age, 48 were identified and 13 were excluded. Although the 13 patients excluded from the study did indeed meet criteria for a diagnosis of dementia, they did not display sufficiently aggressive or disturbing behaviors to result in expulsion from a nursing home. Several of the patients had frequently experienced paranoid ideations, suspiciousness/distrust of others, and cognitive psychotic disturbance, but were not overtly physical or disruptive in their behavior. However, 35 patients were identified as having significant behavioral problems resulting in their expulsion from nursing homes. All patients in the study were treated with gabapentin throughout the course of hospitalization. During the course of treatment, ancillary medications were used (Table). Eleven patients had small-to-moderate dosages of risperidone, up to a maximum of 6 mg/day, added to their course of treatment. Most had much more modest dosages. Many of the medications patients were taking before hospitalization were withdrawn for the study.

Patient charts were reviewed by an independent research assistant who recorded frequency of the following behaviors: yelling, moaning, screaming, crying, and verbal or physical threats of aggression. Sexually inappropriate behaviors (grabbing, fondling, or sexually provocative comments) were also recorded.

Length of hospital stay was divided into the first and second halves of hospitalization. Each patient served as his or her own control. Charts were reviewed on each patient, and the number of aggressive events that occurred during each patient’s first and second half of hospitalization was recorded (Table).


The average age of all 35 patients was 78 years, and the average length of stay in the hospital was 14.37 days. The number of aggressive events occurring in the first half of the hospitalization was 102; in the second half there were 34. Three patients accounted for 61.8% of the aggressive behavior in the second half of the hospitalization.

The data were analyzed by pairing each observation in the second half of hospitalization with an observation in the first half. The mean difference in aggressive events between the first and second samples was 1.94, with a standard error of 0.518. The probability of observing such a difference in aggressive behaviors by chance alone between the first and second observation period is less than .001. The value of the t statistic for this test was 3.747, thus we can say with 99.9% confidence that the behavioral change exhibited between the first and second half of the hospital stay was not a result of chance.

Although 17 patients accounted for 100% of the disturbing behaviors in the first half of the hospitalization, 11 patients accounted for all of the aggressive behaviors in the second half of hospitalization. Both frequency and intensity of aggressive acts diminished during the course of hospitalization for 16 patients. Only one patient had more events in the second portion of the hospitalization than in the first. Although 17 patients (48%) had aggressive behaviors that continued during hospitalization, 18 patients who had aggressive behaviors before hospitalization had no aggressive behaviors during either their first or second half of the hospitalization. This result is possibly associated with a good response to medication management or the result of hospitalization itself. All of the patients were managed with gabapentin throughout the entire course of hospitalization. Risperidone was the most common medication given as an add-on throughout hospitalization, although one patient received haloperidol. The addition of risperidone was employed in the 11 patients exhibiting aggressive behaviors. Other antipsychotics, antidepressants, and benzodiazepines were avoided. The use of risperidone does not account for the positive results in this study because only 11 of the 35 patients took risperidone during the study. Six of the 11 patients had no aggression in both halves of the hospitalization, whereas 5 exhibited aggressive behaviors in the first half and 4 continued to be aggressive in the second half. The total number of aggressive episodes for the risperidone- and gabapentin-treated group was 30 in the first half and 11 in the second. These figures represent a 36% reduction in aggressive behaviors, whereas in the gabapentin-treated group there was a 30% reduction (102 aggressive events occurred in the first half of hospitalization and 34 in the second).


It is highly likely that the removal of a patient from his or her environment and the placement of that patient in a hospital with staff who are highly trained to manage aggressive behaviors does have a salutary effect on the diminishing aggression in a patient with dementia. This may contribute to the fact that 18 of the patients had no disturbing behaviors in the first half of their hospitalization. It is also possible, however, that medication management at the inception of treatment in the first hospital stay could account for some of the diminishment in aggression.

This was a retrospective, open-label study of gabapentin. As such, it is limited to noncontrolled conditions. Although the data were retrospectively examined, results must be replicated in a controlled, blind experiment.


The use of gabapentin in demented patients with aggressive behaviors has been shown to be effective in the management and control of aggressive, hostile symptoms. Gabapentin had no substantial side effects other than mild sedation in one patient,4 who tolerated only 300 mg/day secondary to renal insufficiency with elevated serum urea nitrogen levels and increased creatinine clearance, both of which were in the abnormal range. One other patient could tolerate only 600 mg due to sedation. Both patients were noted to have sedation on higher doses. No adverse events (eg, falling) were noted in the treatment cohort. Aggressive behaviors have a substantial impact on caregivers, and can lead to expulsion from nursing homes and mandatory psychiatric hospitalization. Gabapentin represents a safe medication for elderly patients with dementia and aggressive behaviors. This study employed an average gabapentin dosage of 1,400 mg/day in an elderly population (mean age=78.8 years), demonstrating the drug’s effectiveness in high dosages without any deleterious side effects other than mild sedation. The use of antipsychotics, such as risperidone, did not substantially improve aggressive behaviors more than gabapentin.  PP


1.     Aging and the Oldest Old. Geneva, Switzerland: United Nations, Population Division, Department of Economics and Social Affairs; 1998.
2.     Ernst RL, Hay JW. US economic and social costs of Alzheimer’s disease revisited. Am J Public Health. 1994;84:1261-1264.
3.     Weiner M, Powe NR, Weller WE, Shaffer TJ, Anderson GF. Alzheimer’s disease under managed care: implications from Medicare utilization and expenditures patterns. J Am Geriatr Soc. 1998;46:762-770.
4.     Lyketsos CG, Sheppard JM, Rabins PV. Dementia in elderly persons in a general hospital. Am J Psychiatry. 2000;157:704-707.
5.     Kolbeinsson H, Jonsson A. Delirium and dementia in acute medical admissions of elderly patients in Iceland. Acta Psychiatr Scand. 1993;87:123-127.
6.     Lyketsos CG, Steinberg M, Tschanz J, Norton MC, Steffens DC, Breitner JC. Mental and behavioral disturbances in dementia: findings from the Cache County study on Memory in Aging. Am J Psychiatry. 2000;157:708-714.
7.     Mega MS, Cummings JL, Fiorello T, Gombein J. The spectrum of behavioral changes in Alzheimer’s disease. Neurology. 1996;46:130-135.
8.     Finkel S. Research methodologic issues in evaluating behavioral disturbances of dementia. Int Psychogeriatr. 1996;8(suppl 2):149-150.
9.     Tariot PN, Erb R, Podjorski CA, et al. Efficacy and tolerability of carbamazepine for agitation and aggression in dementia. Am J Psychiatry. 1998;155:54-61.
10.     Tariot PN, Frederiksen K, Erb R, et al. Lack of carbamazepine toxicity in frail nursing home patients: a controlled study. J Am Geriatr Soc. 1995;43:1026-1029.
11.     Cooney C, Mortimer A, Smith A, Newton K, Wrigley M. Carbamazepine use in aggressive behavior associated with senile dementia. Int J Geriatr Psychiatry. 1996;11:901-905.
12.     Sandborn WD, Benfeldt F, Hamdy R. Valproic acid for physically aggressive behavior in geriatric patients. Am J Geriatr Psychiatry. 1996;3:239-242.
13.     Herrmann N. Valproic acid treatment of agitation in dementia. Am J Psychiatry. 1998;43:69-72.
14.     Mazure CM, Druss BH, Cellar JS. Valproate treatment of older psychotic patients with organic mental syndromes and behavioral dyscontrol. J Am Geriatr Soc. 1991;42:906-909.
15.     Mellow A, Solano-Lopez C, Davis S. Sodium valproate in the treatment of behavioral disturbance in dementia. J Geriatr Psychiatry Neurol. 1993;6:205-209.
16.     Raskind MA. Evaluation and management of aggressive behavior in the elderly demented patient. J Clin Psychiatry. 1999;60:45-49.
17.     Herrmann N, Lanctot K, Myszak M. Effectiveness of gabapentin for the treatment of behavioral disorders in dementia. J Clin Psychopharmacol. 2000;20:90-93.
18.     Roane DM, Feinberg TE, Meckler L, Miner CR, Scicutella A, Rosenthal RN. Treatment of dementia-associated agitation with gabapentin. J Neuropsychiatry Clin Neurosci. 2000;12:40-43.
19.     Dallocchio C, Buffa C, Mazzarello P. Combination of donepezil and gabapentin for behavioral disorders in Alzheimer’s disease [letter]. J Clin Psychiatry. 2000;61:64.
20.     Goldenberg G, Kahaner K, Bassavaraju N, Rangu S. Gabapentin for disruptive behaviour in an elderly demented patient [letter]. Drugs Aging. 1998;13:183-184.
21.     Sheldon LJ, Ancill RJ, Holliday SG. Gabapentin in geriatric psychiatry patients [letter]. Can J Psychiatry. 1998;43:422-423.
22.     Regan WM, Gordon SM. Gabapentin for behavioral agitation in Alzheimer’s disease [letter]. J Clin Psychopharmacol. 1997;17:59-60.