Post stroke depression (PSD) is a common psychiatric complication caused by stroke. PSD always leads to a poor prognosis such as cognitive impairment, severe disability, high suicide rate (Paolucci et al., 2019). The etiology of PSD is multifactorial and included stress-induced activation of hypothalamic-pituitary-adrenal (HPA) axis, glutamate-mediated excitotoxicity, inflammation, and neurogenesis, etc. (Villa et al., 2018) Notably, chronic inflammation plays an important role in the pathogenesis of PSD.

A large body of evidence suggests that circulating level of interleukin (IL)-18, IL-10, IL-6, TNF-α and C-reactive protein levels were closely associated with PSD at different time points (Yang et al., 2010; Chi et al., 2021; Chen et al., 2020; Yang et al., 2022). A prospective stroke cohort showed that systemic immunoinflammatory indices at admission was associated with PSD at 1 month (Hu et al., 2021). Then, in animal models, suppression of inflammatory responses, including microglial polarization and release of pro-inflammatory factors such as IL-1β and IL-18, can reduce the severity of PSD (Wang et al., 2019; Li et al., 2021; Zhang et al., 2021). In addition, inflammation is closely related to other pathological processes involved in PSD. For example, pro-inflammatory factors stimulate the release glucocorticoids from HPA axis and reduce neurotransmitters in the brain by enhancing the expression of genes encoding indoleamine 2,3-dioxygenase (IDO) (Szczudlik et al., 2004; Maes et al., 2011). These studies suggest that a sustained inflammatory response promotes the development of PSD. However, depressive symptoms after stroke is fluctuating, and the pathogenesis of PSD may vary in different periods after stroke. To our knowledge, most clinical studies have focused on the prediction of single cytokine at baseline for different periods of PSD, and few studies have examined the dynamic changes of cytokines network in PSD.

Furthermore, the diagnosis of PSD currently relies on subjective tools such as Hamilton Depression Scale (HDRS) and Patient Health Questionnaire (PHQ-9) and lacks effective objective biomarkers (Das and G KR., 2018). Oxidative stress is characterized by endogenous production of oxygen species such as reactive oxygen species (ROS) and subsequent harmful effects on lipids, proteins and DNA (Forman and Zhang, 2021). Previous studies showed that two oxidative stress markers including malondialdehyde (MDA) (Liu et al., 2017a) and 8-hydroxy-2′-deoxyguanosine (8-OHdG) (Liu et al., 2018) have high area under curve (AUC) value when predicting PSD. However, some oxidative markers including SOD, CAT and GSH-PX showed no predictive effect on 1-month PSD (Liu et al., 2017a; Liu et al., 2018). In addition to oxidative stress, the predictive role of inflammation markers in PSD has been gradually recognized. For example, serum C-reactive protein (CRP) levels within 24 h after stroke was an indicator for prediction of, with the AUC value at 0.765 (Yang et al., 2016). Moreover, oxidative stress and inflammation are mutually cause and effect, promoting a series of injury reactions after cerebral ischemia (He et al., 2021). Thereafter, combination of oxidative stress and inflammation is meaningful for predicting PSD. In addition, no studies have yet explored the predictive role of early inflammation and oxidative stress on 2-week PSD, a timepoint at which many stroke patients have been discharged from hospital and often miss diagnosis and treatment of PSD.

In our current study, we examined peripheral cytokines and oxidative stress markers in patients with PSD. The main objectives of this study included: 1) to compare the dynamic changes of peripheral cytokines network between non-PSD patients and PSD patients at 3 days, 2 weeks and 3 months; 2) to confirm the predictive role of cytokines and oxidative stress markers at 3 days for PSD at 2 weeks.