Cold Pressor Task Reactivity: Predictors of Alcohol Use Among Alcohol-Dependent Individuals With and Without Comorbid Posttraumatic Stress Disorder Kathleen T. Brady, Sudie E. Back, Angela E. Waldrop, Aimee L. McRae, Raymond F. Anton, Himanshu P. Upadhyaya, Michael E. Saladin, and Patrick K. Randall Background: The association between stress and alcohol dependence has been well established. Abnormalities in stress reactivity and hypothalamic–pituitary–adrenal axis (HPA) function may be involved in the mechanistic connection between stress and the initiation, development, and/or main- tenance of alcohol dependence. Posttraumatic stress disorder (PTSD) commonly co-occurs with alcohol dependence and is characterized by HPA axis abnormalities. This study investigated the relationship between subjective and neuroendocrine stress reactivity to the cold pressor task (CPT) and prospective alcohol use among individuals with alcohol dependence, with and without comorbid PTSD. Methods: Participants were 63 individuals with (a) alcohol dependence only (n 5 35) or (b) comorbid alcohol dependence and PTSD (n 5 28). Participants completed the CPT, a widely used physical laboratory stressor. Subjective stress, craving, adrenocorticotrophin (ACTH), and cortisol were measured before, immediately after, and at 5, 30, 60, and 120 minutes after the CPT. Alcohol use during 1 month following testing was also assessed. Results: For the alcohol-only group, change in craving immediately following the CPT and crav- ing during the 120-minute recovery phase were predictive of follow-up alcohol use. For the alcohol/ PTSD group, change in craving was not predictive of follow-up use. Baseline drinking was, however, predictive of followup alcohol use for the alcohol/PTSD group. For the alcohol-only group, a blunted ACTH response coupled with a higher change in craving following the CPT was associated with sig- nificantly greater frequency and intensity of drinking during the follow-up phase. Conclusions: These preliminary findings demonstrate significant differences between the alcohol- only and the alcohol/PTSD group in predictors of relapse. For the alcohol-only group, reactivity to an acute laboratory stressor may be predictive of subsequent alcohol use. This was not true for the alcohol/PTSD group. Although preliminary, the findings may help shed light on the mechanistic relationship between stress reactivity and increased risk for alcohol relapse and dependence in indi- viduals with and without other Axis I comorbidity. Key Words: Alcohol, Stress Reactivity, ACTH, Craving, PTSD. T HE RELATIONSHIP BETWEEN stress and sub- stance use disorders has received increasing attention over the past several years. Clinical observations, animal models, and prominent theories of addiction implicate stress as a contributing factor in the initiation, develop- ment, or maintenance of substance use disorders (Brady and Sinha, 2005; Goeders, 2003; Khantzian, 1985; Marlatt and Gordon, 1985; Shaham et al., 2000), but the mecha- nism underlying this connection has not been extensively explored. Animal studies have consistently demonstrated that a number of stressors are associated with reinstate- ment of drug-seeking behavior (Stewart, 2003), and preclinical evidence suggests that corticotrophin releasing factor (CRF) and the hypothalamic–pituitary–adrenal (HPA) axis are involved in this process (Shalev et al., 2002). Chronic alcohol consumption is associated with abnor- malities of the HPA axis response (Adinoff et al., 2005; Bernardy et al., 1996; Errico et al., 1993; Gianoulakis et al., 2003), while increased adrenocorticotrophin (ACTH)/cortisol secretion is typically seen during acute alcohol withdrawal (Kirkman and Nelson, 1988). Alcohol- dependent individuals who have been abstinent for up to 1 month demonstrate an attenuated ACTH and cortisol response to a variety of stressors, including chemical challenge, mental arithmetic, cold pressor, and isometric handgrip (Adinoff et al., 2005; Bernardy et al., 1996; From the Medical University of South Carolina, Charleston, South Carolina. Received for publication October 24, 2005; accepted January 11, 2006. This work was supported by NIAAA Grant P50 AA1076, NIH Grant M01RR01070, and NIDA Grant K24 DA 00435. Reprint requests: Dr. Kathleen Brady, Division of Clinical Neurosci- ence, Medical University of South Carolina, 67 President Street, PO Box 250861, Charleston, SC 29425; Fax: 843-792-4817; E-mail: bradyk @musc.edu Copyright r 2006 by the Research Society on Alcoholism. DOI: 10.1111/j.1530-0277.2006.00097.x Alcohol Clin Exp Res, Vol 30, No 6, 2006: pp 938–946 938 ALCOHOLISM:CLINICAL AND EXPERIMENTAL RESEARCH Vol. 30, No. 6 June 2006