Gabapentin Withdrawal: Case Report in an Older Adult and Review of the Literature

To the Editor: Gabapentin is approved by the Food and Drug Administration for the treatment of epilepsy and postherpetic neuralgia, but its wide therapeutic margin and good tolerability1 have led to aggressive off-label use for other medical and psychiatric conditions,2 including more recently, behavioral and psychological symptoms of dementia.3 Yet there is potential for abuse and dependence. Gabapentin's U.S. sales alone in 2004 were $2.4 billion.1 Here we present a case of protracted withdrawal after gabapentin taper in an older adult. AB was a 75-year-old woman admitted for assessment of recurrent falls. She had a 20-year history of fibromyalgia and postherpetic neuralgia, for which she was taking gabapentin 1,800 mg/d, and recurrent depression treated with sertraline 300 mg/d and lorazepam 3 mg/d. Side effects from psychotropic agents were thought to contribute to AB's falls. Gabapentin was tapered and discontinued completely over a 10-day period. Lorazepam was slowly tapered by 1 mg/d over a week, beginning 1 day after initiation of gabapentin taper. AB tolerated the 10-day taper schedule well, but on Day 11, she complained of mild abdominal pain and headache. Her blood pressure was 190/95. The following day, she reported chills, cold sweats, nausea, and insomnia. She became increasingly agitated and developed anorexia. Benzodiazepine withdrawal was considered. Over the next 5 days, symptoms failed to improve despite re-initiation of lorazepam or switching to a longer-acting benzodiazepine. Abdominal X-ray and computed tomography were normal. Laboratory tests were unremarkable except for hypokalemia (2.6 mEq/L) and mildly elevated liver enzymes. Potassium replacement was initiated. On Day 18, psychiatry was consulted. The constellation of agitation, chills, diaphoresis, and severe abdominal pain described as “worse than childbirth” were suspected to be due to gabapentin withdrawal. Gabapentin 400 mg was given immediately and re-initiated at 600 mg at bedtime. Blood pressure normalized. She was no longer diaphoretic but had only partial relief of anxiety and abdominal pain. She continued to report insomnia. Gabapentin was increased to a maximum of 1,400 mg/d due to renal impairment. Her withdrawal symptoms completely resolved within 3 days. A survey of the literature identified 19 cases of withdrawal after gabapentin discontinuation (Table 1). Five were aged 60 or older. Indications for gabapentin were chronic pain, psychiatric illness, and alcohol abuse or dependence. Across cases, withdrawal symptoms occurred within a few days after abrupt discontinuation of gabapentin, except for two patients, for whom withdrawal occurred despite gradual reduction of gabapentin. Overall, the clinical presentation of gabapentin withdrawal appears to be similar to that of benzodiazepines: agitation and anxiety, diaphoresis, somatic pain, confusion, tremulousness, gastrointestinal distress, and tachycardia or palpitations. Hypertension, which was observed in AB, was reported in only two other cases. The atypical constitutional symptoms of shivering and cold sweats that AB experienced 1–2 days after complete discontinuation of gabapentin were reported in an 81-year old woman who experienced “flu-like symptoms” one day after discontinuing gabapentin. Typically, withdrawal symptoms did not respond to benzodiazepines but resolved completely within a few days after restarting gabapentin. The mechanism of gabapentin withdrawal is unclear (see 1). Gabapentin was designed as a structural analogue of the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) that would penetrate the blood–brain barrier yet retain GABA properties as much as possible. Yet, gabapentin does not bind to GABA receptors, nor does it alter GABA transport. Gabapentin is considered a calcium channel α2-δ ligand, which binds to calcium channels at the α2-δ subunit, leading to inhibition of neurotransmitter release and attenuation of postsynaptic excitability, yet gabapentin administration in human subjects leads to an increase in central GABA levels.4 Gabapentin's effect on neuropathic pain symptoms may result from inhibition of glutamate release,5 so withdrawal may be GABA-mediated, similar to that of benzodiazepines,6 or through upregulation of glutamatergic excitatory neurotransmission, which is partly responsible for acute ethanol withdrawal.7 Gabapentin withdrawal has been reported infrequently, suggesting there may be predisposing conditions for a withdrawal syndrome. AB and an older adult in a previous case report experienced withdrawal despite gabapentin taper. Premorbid psychiatric illness may be a risk factor for gabapentin withdrawal. Advanced age may increase risk of withdrawal from gabapentin because of age-related reduction of GABA-mediated cortical inhibition8 or alterations in expression of glutamate receptors.9 The mechanism of action of gabapentin, particularly in older adults, is poorly understood. A conservative approach to off-label use of gabapentin with clearly defined endpoints for dosage titration and discontinuation is suggested. In older adults and in the presence of psychiatric illness, gabapentin discontinuation should follow a slow tapering schedule, as for benzodiazepines (e.g., biweekly reduction of 10–25% of dose),10 monitoring carefully for withdrawal symptoms. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: All authors contributed to the drafting and approval of the final manuscript. Sponsor's Role: Not applicable.