The startle reflex in alcohol-dependent patients: changes after cognitive-behavioral therapy and predictive validity for drinking behavior. A pilot study.

BACKGROUND: Previous studies demonstrated an attenuation of the affect-modulated startle reflex when alcohol-dependent patients were viewing alcohol-associated pictures. This indicates an appetitive valence of these stimuli. We used the affect-modulated startle reflex to assess the effects of behavioral treatment on the emotional processing of alcohol-associated stimuli. Further, we examined whether the affect-modulated startle reflex is a predictor of treatment success.

METHODS: Forty-three alcohol-dependent patients (21 females, mean age 45.67 years, SD 9.45) were recruited consecutively from an inpatient alcohol detoxification facility where patients attended a 3-week detoxification program including cognitive-behavioral treatment to successfully handle high-risk situations. The eye blink component of the affect-modulated startle response, self-reported cue-induced craving and skin conductance responses to alcohol-associated and control slides were assessed before and after treatment. Changes were analyzed using repeated measures analysis of variance. Drinking behavior was assessed in the 6 months following treatment, and a regression analysis was performed to evaluate the predictive validity of the affect-modulated startle response for drinking behavior.

RESULTS: Drinking behavior as well as craving and skin conductance responses decreased significantly over time. The pattern of the affective modulation of the startle reflex was not altered over time. However, startle modulation and relapse were related, and within the group of relapsers, startle modulation was a significant predictor of drinking behavior.

CONCLUSIONS: Our results suggest that the modulation of the startle reflex may reflect more enduring and permanent processes of emotional responding to alcohol-related cues than autonomic arousal and self-reported craving, and that startle modulation by alcohol-associated cues may be a better predictor of drinking behavior for relapsers than other measures. Further studies including a control condition are necessary to validate these findings.