Longitudinal disturbances of objective sleep architecture in cocaine use disorder: A translational systematic review
I. Medigue, S. Catoire, C. Peyron, P.-A. Geoffroy, T. Bernabeu, L. Peter-Derex, B. RollandIntroduction
Cocaine Use Disorder (CUD) is frequently associated with severe and persistent sleep disturbances, affecting both the initiation and maintenance of abstinence. Despite their clinical relevance, the nature, timeline, and underlying neurobiological mechanisms of these disturbances remain insufficiently understood. In particular, objective sleep measures have not been systematically explored across different stages of cocaine use and withdrawal. Improved knowledge of these disruptions is critical for identifying new therapeutic targets and reducing relapse risk.
Objectives
This systematic review aimed to characterize objective alterations in sleep architecture associated with cocaine use and its withdrawal phases. A secondary aim was to compare findings between human and animal studies to identify converging patterns and highlight potential translational mechanisms.
Methods
A systematic search was conducted across PubMed, PsycInfo, and Google Scholar databases, following PRISMA guidelines. Inclusion criteria covered original studies reporting objective sleep data (PSG, EEG, or actigraphy) in the context of cocaine use or withdrawal, in both humans and animal models. Studies involving significant psychiatric comorbidities or other substance use (excluding nicotine) were excluded. Risk of bias was assessed using NIH and ARRIVE tools.
Results
A total of 19 studies were included (12 human, 7 animal). In both populations, cocaine use was associated with reduced total sleep time (TST), decreased sleep efficiency (SE), prolonged sleep onset latency (SOL), and a marked reduction in REM sleep. In early withdrawal, a REM rebound was frequently observed, along with a transient improvement in sleep quantity. However, during later stages of withdrawal, sleep quality deteriorated again, with persistent fragmentation, reduced REM, and prolonged latencies (Image 1). These changes were more consistently captured in human studies. Animal models provided additional insight into underlying mechanisms but showed methodological heterogeneity and limited longitudinal data.
Image 1: Long description.