Abstract
To the Editor
Patients with schizophrenia have significantly reduced life expectancy compared to the general population. Much of this increased mortality could possibly be attributed to smoking related illness as rates of smoking are high (up to 75%) in this population (Hennekens et al., 2005).
Varenicline is a pharmacologic agent used to help with smoking cessation. It has partial agonist and antagonist effects at α4β2 nicotinic acetylcholine receptors (Stahl, 2008). As an agonist with high affinity it counters the smoking cessation-induced reduction in mesolimbic dopamine levels, potentially relieving withdrawal symptoms. As an antagonist it prevents nicotine-induced dopaminergic activation on reward circuits.
However there has been a reported association between varenicline use and neuropsychiatric events in patients with a history of mental illness (Lorenz et al., 2010). Due to the strength of this relationship varenicline has been given an FDA ‘boxed warning’ label advising caution in prescribing it for patients with a known history of mental illness (US Food and Drug Administration, 2009).
We present a report of a 42-year-old woman, Ms Z, with a history of Schizophrenia commencing at the age of 35 years. Ms Z’s maintenance medication to prevent relapse was risperidone depot preparation 25mg given fortnightly. On risperidone depot she had no hospital admissions in the preceding 12 months.
Ms Z was a chronic heavy smoker and consulted her GP about quitting. She was prescribed varenicline on a normal dose regime (0.5 mg for 3 days and 1mg/d for next 4 days followed by 2 mg/day). Within 5-6 days of varenicline initiation she started becoming unwell, characterised by sleep disturbance, irritability, disorganisation and increasing paranoia. A week after commencing varenicline therapy, Ms Z was detained by the Community Mental Health Team. On admission, Ms Z presented with florid psychosis warranting closed ward management. On day two of admission, varenicline was ceased, prn benzodiazepines were administered and risperidone depot was continued. With these interventions Ms Z’s mental state quickly improved over the next four days and she was discharged after one week in hospital. On discharge her mental state was back to a baseline state. She remained in remission the following 12 months.
This symptom presentation was given a Naranjo score of 6 (Naranjo et al., 1981), suggesting a probable adverse drug reaction. We would speculate that Ms Z’s psychotic episode described herein was precipitated by her varenicline use given the temporal association, the resolution of symptoms with withdrawal of the drug, the neurobiological plausibility (Stahl, 2008) and previous similar case reports (Lorenz et al., 2010). Our patient developed breakthrough psychosis even though she was on risperidone depot 25 mg fortnightly. It is of note that a recent study (Hong et al., 2011) mentions lack of neuropsychiatric side effects when varenicline was administered to patients with schizophrenia (n=69) on slow dose titration and moderate final daily dosing regime (1 mg/day). Doses at this level can still maintain a reasonable anti-smoking efficacy (Oncken et al., 2006).
Varenicline may induce psychosis by producing prolonged bursts of action potentials on mesolimbic dopamine neurons, leading to supraphysiological dopamine release (Stahl, 2008).
Clinicians should be aware of the potential for varenicline to induce psychosis in predisposed individuals. However, the efficacy of varenicline and overall health benefits of smoking cessation in psychiatric population is well documented. We hypothesize that slower dose titration and a moderate final dose regime (1mg/day) could potentially allow the receptors to adapt to the surge of dopamine, thereby minimising emergence of psychosis in the longer term whilst maintaining reasonable anti-smoking efficacy.