Background: In healthy volunteers, exposure to antidepressants increases the recognition of positive face emotions and decreases recognition of negative emotions. It has been proposed that this may underlie therapeutic effects of antidepressants, but to date this has not been tested in clinical populations.
Method: Recognition of facial emotions was measured at baseline (N=108) and after 2 (N=59) and 6 weeks (N=69) of treatment in depressed primary care patients who had been randomised to treatment with either citalopram (SSRI) or reboxetine (NaRI) in an open-label study. Changes in emotional processing were compared to clinical outcome after 6 weeks of treatment.
Results: Significant increases in recognition accuracy of disgust, happiness and surprise occurred by two-weeks of treatment with both antidepressants, and did not further change at 6 weeks. There was a significant correlation between the increased accuracy in recognition of happy faces over the first two-weeks of treatment and the clinical improvement after six-weeks of treatment.
Limitations: There was no control group and changes over time may be due to practice effects.
Conclusions: Antidepressants altered emotional processing in depressed patients with some similarities to the effects seen in healthy volunteers. The largest effect seen was increased recognition of disgust that may be specific to depressed patients. The correlation between increased accurate recognition of happy faces at two-weeks of treatment and clinical outcome at six-weeks of treatment suggests that early changes in emotional processing may underlie clinical response to antidepressants.