fNIRS evidence of abnormal frontotemporal cortex activation and functional connectivity in depressed patients after stroke: neuromodulatory mechanisms from mild to moderate depression

Post-stroke depression (PSD) is a prevalent psychiatric complication following a stroke, significantly delaying neurological recovery. The assessment of scales in clinical diagnosis often lacks objectivity, while functional near-infrared spectroscopy (fNIRS) has been recognized as an adjunctive diagnosis of depression. This research was designed to evaluate whether fNIRS signals can differentiate different degrees of PSD and explore the pathogenesis behind PSD. We recruited 56 stroke patients treated at the Wenzhou TCM Hospital of Zhejiang Chinese Medical University and stratified them into three groups according to PSD severity: non-PSD (n = 18), mild-PSD (n = 19), and moderate-PSD (n = 19). fNIRS was employed to monitor frontotemporal cortical activity while administering a verbal fluency task across all participant groups. Differences in hemodynamic activity and functional connectivity across six frontotemporal cortex subregions were examined in three patient groups, and their correlations with 17-item Hamilton Depression Rating Scale (HAMD-17) scores were evaluated. In terms of brain activation, the moderate-PSD group demonstrated significantly diminished activation in four particular brain regions in comparison to the non-PSD group (p < 0.05): the bilateral medial prefrontal cortex (mPFC), the ipsilateral dorsolateral prefrontal cortex (DLPFC), and the contralateral temporal lobe (TL), and the activation intensity within these regions was negatively associated with HAMD-17 scores (L-mPFC: rs = -0.315, p = 0.018; R-mPFC: r = -0.377, p = 0.004; L-DLPFC: r = -0.323, p = 0.015; R-TL: r = -0.401, p = 0.002). Mild-PSD exhibited lower activation only in CH42 but higher in CH6 than moderate-PSD (p < 0.05). Regarding brain functional connectivity, the strength of connectivity between the DLPFC~mPFC on the ipsilesional side was positively correlated with the HAMD-17 scores (rs = 0.405, p = 0.002), with significant disparities in the moderate-PSD versus the non-PSD groups. In contrast, the mild-PSD group displayed no notable connectivity differences between the two groups. This study presents distinct patterns of frontotemporal cortex activation and functional connectivity alterations associated with varying severity levels of PSD. In contrast with patients with stroke alone, PSD patients showed decreased activation levels and abnormally increased functional connectivity, and this change was more pronounced in moderate-PSD patients. These findings indicate that functional features of the frontotemporal cortex may serve as a neural indicator for identifying high-risk cases of PSD. https://www.chictr.org.cn/showproj.html?proj=249555, ChiCTR2400093089.