Introduction: Near-infrared spectroscopy (NIRS) has attracted attention as a simple brain function imaging method. However, few longitudinal studies have focused on the relationship between depressive symptoms and prefrontal cortical findings. The purpose of this study was to investigate the relationship between treatment outcome and prefrontal cortex activation in outpatients with major depression longitudinally measured using NIRS. Methods: We performed NIRS at about two-month intervals for outpatients with major depression. At Time 1 and Time 2, we evaluated changes in [oxy-Hb] during the verbal fluency task, depression symptoms and psychosocial functioning. There were 3 regions of interest in the prefrontal cortex: the right prefrontal cortex (channels 1-4), frontal pole (channels 7-10), and left prefrontal cortex (channels 13-16). Participants were divided into the non-remission group and partial remission group based on treatment outcome. Results: In the non-remission group, there was a significant increase in [oxy-Hb] in the frontal pole (P < 0.05), and a decreasing trend in [oxy-Hb] in the left prefrontal cortex (P < 0.10) during the measurement period. In contrast, in the partial remission group, there were no [oxy-Hb]changes during the measurement period. Furthermore, in the non-remission group, there was a significant positive correlation between changes of depression symptoms (Time 2-Time 1) and changes of [oxy-Hb] in the frontal pole (Time 2-Time 1) (r=0.48, P < 0.05), and a marginal negative correlation with changes of [oxy-Hb] in the left prefrontal cortex (Time 2-Time 1) (r=-0.45, P < 0.10). Meanwhile, in the partial remission group, there was no significant correlations. Conclusion: This longitudinal study suggests that the change in prefrontal cortical function related to prolonged depression symptoms could be evaluated using NIRS measurement. In future studies, strict criteria for the selection of research participants and controlling for extraneous variables are required.
 <Authors' abstract>