Cluster and meta-analyses on factors influencing stress-induced alcohol drinking and relapse in rodents.

Numerous preclinical studies have focused on the identification of biological and environmental factors that modulate stress and alcohol interactions. Although there is a good qualitative description of the determinants of alcohol consumption in rodents, the magnitude of the variables influencing stress-induced ethanol intake and its dynamics are still poorly understood. We therefore carried out a clustered meta-analysis on stress-induced alcohol consumption in 1520 rats. Two-step clustering of the literature-derived dataset suggests a strong dependency of the experimental outcome on the method used to measure alcohol intake. Free-choice home cage drinking versus operant self-administration is the most critical determinant of stress-induced increases in alcohol consumption in rats. Stress does not typically result in enhanced alcohol consumption in operant self-administration paradigms, whereas it leads to increased home cage drinking. Stress-induced alcohol consumption is age dependent, with adults being more sensitive than adolescents. In addition, foot shock and forced swim stress enhance alcohol intake, while restraint stress does not. In contrast, a meta-analysis of 327 rats on stress-induced reinstatement of alcohol-seeking behavior shows less influence of those modulating factors, and usually foot shock or yohimbine leads to a reinstatement of approximately 300 percent of extinction level responding. Via accurate characterization of the significant factors in the interplay of alcohol consumption, relapse and stress, our quantitative description not only improves the understanding of underlying mechanisms, but also provides an appropriate framework for the optimal experimental design of preclinical studies that more accurately translates to the human condition.