Rimonabant induces partial seizures in a patient with a history of generalized epilepsy

To the Editors: Rimonabant (N-piperidino-5-(4-chlorophenyl)-1-(2,4-dichlorophenyl)-4-methyl-3-pyrazole-carboxamide) was the first U.S. Food and Drug Administration (FDA)–approved cannabinoid CB1 receptor antagonist for the treatment of overweight. Rimonabant is effective in the treatment of obesity-related disorders, as it reduces food intake and body weight, and it is associated with a beneficial effect on different metabolic parameters and cardiovascular risk factors linked to overweight (Després et al., 2005; Pi-Sunyer et al., 2006; Scheen et al., 2006). We report herein on a 52-year-old obese man with a history of hypertension, diabetes mellitus, and idiopathic generalized epilepsy. Between the age of 4 and 15 years, he had absences and two generalized tonic–clonic seizures. Except for one generalized seizure triggered by the consumption of too much alcohol at the age of 30, he had been seizure free and discontinued antiepileptic drug treatment more than 20 years ago. Because of overweight and other cardiovascular risk factors, including hypertension and diabetes mellitus, he started rimonabant 20 mg/day in order to reduce weight. Two months later, he had nocturnal partial seizures consisting of a stereotypical feeling of falling into a deep hole. Often, this sensation was followed by right leg jerks lasting, on average, 3 min. These seizures were totally different from the generalized seizures he had in his youth. After rimonabant withdrawal the seizures disappeared. A few weeks later he restarted rimonabant and 3 days later, his partial seizures returned. While on rimonabant, routine and sleep-deprivation encephalography (EEG) studies were performed, showing frequent epileptiform paroxysms, consisting of generalized irregular slow waves intermingled with frontal and temporal spikes and focal epileptiform abnormalities in the left temporal lobe. Revision of older EEG studies (1982–1990) showed the same generalized and focal epileptiform abnormalities. Magnetic resonance imaging (MRI) of the brain and an ictal glucose level were normal. Because neither structural nor metabolic causes of these partial seizures could be revealed and a strong relationship with the use of rimonabant was supposed, he was strongly advised to discontinue rimonabant. Again the seizures disappeared immediately after rimonabant withdrawal. Three months after rimonabant withdrawal a sleep-deprived EEG showed mainly focal (left > right) temporally located epileptiform discharges and generalized irregular slow-wave paroxysms. This case report is the first to describe that rimonabant induces seizures in humans with a history of epilepsy and/or with epileptiform discharges on EEG. Mouse models have demonstrated that activation of the cannabinoid CB1 receptor is responsible for neuroprotection, as cannabinoids possess anticonvulsive properties (Wallace et al., 2002; Monory et al., 2006). Cannabinoid CB1 receptor blockade by rimonabant significantly reduced seizure threshold in mice (Wallace et al., 2002). Endocannabinoid signaling is impaired in patients with epilepsy, as the expression of CB1 receptor mRNA is downregulated and CB1 receptors are lost in key regions in epileptogenesis, including the dental gyrus (Ludányi et al., 2008). Cannabinoid CB1 receptor antagonists, such as rimonabant, are thus theoretically likely to also reduce seizure threshold in humans due to blocking of CB1 receptors. Hence, the information leaflet reads that rimonabant should be used with caution by patients with epilepsy, because it may aggravate existing seizures. However, seizure occurrence in humans using rimonabant has not been reported previously. In October 2008, the approval of rimonabant was withdrawn in the European Union owing to concerns over potential side effects, mainly its association with depressive disorders and suicide. In conclusion, cannabinoid CB1 receptor antagonists can provoke epileptic seizures in formerly seizure-free epileptic patients, especially in cases in which EEG shows epileptiform discharges. This observation adds to the already known side effects of depression and suicide, and to observations on the key role of cannabinoid CB1 receptors in epileptogenesis in mouse models. Because new cannabinoid receptor antagonists are now being developed and will enter clinical trials in the near future, it is very important to monitor the development of epileptic seizures. The authors did not receive any funding for this study. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose.