Cortisol response under low intensity exercise during cognitive-behavioral therapy is associated with therapeutic outcome in panic disorder–an exploratory study

OBJECTIVES: Patients with Panic Disorder (PD) show an abnormal stress-induced functioning of the Hypothalamic-Pituitary-adrenal (HPA)-axis. Different protocols for stress induction are of rather low relevance for the psychotherapeutic treatment. In practice, interoceptive exposure is often realized as Low Intensity Exercise (LIE), as compared to an incremental cycle exercise test to exhaustion. Currently, it is not known, whether LIE displays an effective interoceptive stressor 1.) leading to a significant anxiety induction; 2.) a comparable HPA- and Sympathetic-Adreno-Medullar (SAM)-axis response in both patients and healthy controls; 3.) stress responses under LIE are associated with treatment outcomes. PATIENTS AND METHODS: N = 20 patients with PD and n = 20 healthy controls were exposed to ten minutes of LIE on an exercise bike. LIE was applied as part of the interoceptive exposure, during an intensive Cognitive-Behavioral Therapy (CBT) in a day clinic. Heart rate was monitored and salivary cortisol samples collected. Before and after the LIE, state anxiety/ arousal were assessed. In order to evaluate psychopathology, the Panic and Agoraphobia Scale, Mobility Inventory, Agoraphobic Cognitions Questionnaire and Body Sensations Questionnaire were applied, before (T1) and after five weeks (T2) of an intensive CBT. RESULTS: LIE led to a significant and similar heart rate increase in both groups. Cortisol decreased over time in both groups, especially in male patients. A higher psychopathology before, and after CBT, was associated with a significantly lower cortisol response under LIE. CONCLUSIONS: In the present study, LIE led to a divergent stress response: while there was a significant heart rate increase, cortisol decreased over time, particularly in male patients. A lower reactivity of the HPA-axis seems to be associated with a lower treatment outcome, which may affect extinction based learning. The findings suggest, that interoceptive stimuli should be designed carefully in order to be potent stressors.