Use of varenicline in smokeless tobacco cessation influences sleep quality and dream recall frequency but not dream affect

Objective: Varenicline (VCL) treatment has become popular as a part of smoking-cessation therapies, even though its possible implications in neuropsychiatric adverse events include abnormal sleep and nightmares. Our study is the first aimed at prospectively investigating changes in sleep and dream measures across such treatment by using a one week sleep diary and collecting dream recall (DR). Methods: Thirty-eight smokers participated in two different effective smoking-cessation programs: 25 subjects received VCL, 13 subjects received nicotine replacement therapy (NRT). Participants were requested to fill out a sleep diary and audio-record dreams for one week, both at baseline assessment (T0) and after two weeks from the beginning of treatment (T1). Differences in sleep and dream measures were tested using mixed Analysis of Covariance with time (T0 vs T1) as a within-subject factor and group (VCL vs NRT) as a between-subject factor, considering age as a covariant. Results: Sleep measures showed a significant increase in the number of awakenings in the VCL group at T1 compared to T0. The VCL group reported an increase in DR rate at T1. These sleep and dream measures were intercorrelated, but only at T1 and during VCL treatment. Conclusions: The correlations between DR rate and sleep fragmentation measures underlined that VCL cannot be considered as a direct cause of abnormal dreams. Comparisons between groups cast some doubts on the actual existence of abnormal dreams, at least as compared to the NRT group. Our results are consistent with the 'activation models', showing that numerous DRs in the VCL group may be related to lighter sleep