Depression is common among stroke patients, and the clinical importance of post-stroke depression is increasing. A recent systematic review indicated that post-stroke depression was present in 33% of stroke survivors [1]. Patients with post-stroke depression show more functional disability, worse rehabilitation outcomes, and increased morbidity and mortality in the first year after a stroke [2]. However, the exact pathophysiology of post-stroke depression remains unknown and several different mechanisms have been suggested [3,4].

Vitamin D deficiency has received increasing attention over the last decade. Lower levels of serum vitamin D have been associated with cardiovascular disease, hypertension, neurodegenerative diseases, diabetes, metabolic syndrome, and cancers [5]. Recently, interest in the potential role of vitamin D in mental health has increased, and it has been suggested that depression is dependent on vitamin D status. This is because vitamin D receptors are widely distributed in areas of the human brain that are involved in depression [6,7]. However, a direct relationship between vitamin D deficiency and depression in stroke patients has not been established. Thus, the aims of this study were to compare depressive symptoms between a vitamin D-deficient group and a normal vitamin D level group, and to investigate the correlation between serum 25-hydroxyvitamin D (25-OHD) levels and the degree of depressive symptoms in stroke patients.

All participants were informed of the procedure and objectives of this study. Only patients who provided written, informed consent were selected. Ethical approval was obtained from the Institutional Human Research Ethics Committee (No. SCHBC2013-02-004-001).

In total, 126 stroke patients were analyzed, and 77.8% of patients were classified into the vitamin D-deficient group (<20 ng/mL). Baseline characteristics of the patients are shown in Table 1. Mean serum 25-OHD levels in the vitamin D-deficient group and the normal group were 10.8±4.2 ng/mL and 28.4±9.1 ng/mL, respectively. When comparing the BDI-II and PHQ-9 levels, both scores were significantly higher in the vitamin D-deficient group than in the normal group (Table 1).

There was a significant negative correlation between the serum 25-OHD levels and the BDI-II score (r=0.205, p=0.021; Fig. 1A), but no correlation between serum 25-OHD levels and PHQ-9 (r=0.156, p=0.081; Fig. 1B) or K-MBI scores (r=0.034, p=0.705; Fig. 1C).

In our study, BDI-II and PHQ-9 scores were significantly higher in the vitamin D-deficient group, showing that depressive symptoms were associated with lower serum 25-OHD levels. A relationship between vitamin D deficiency and depression has been described in several studies. Ganji et al. [17] reported that the prevalence of depression was significantly higher in a population of young adults with vitamin D deficiency. Kjaergaard et al. [18] evaluated vitamin D deficiency and depression in a general population, and established that low serum 25-OHD levels were a significant predictor of depression.

The mechanism linking vitamin D deficiency and depression in stroke patients remains unclear, but there are several plausible biochemical mechanisms by which vitamin D deficiency could change the manifestations of depression in stroke patients. First, depression could be manifested by the inflammatory cytokines associated with vitamin D deficiency. Inflammatory cytokines, such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α), activate the hypothalamic-pituitary-adrenal axis, resulting in the release of cortisol, which is closely related to manifestations of depression [19]. These inflammatory cytokines can also impair neuroplasticity due to neurotoxicity, which may also be associated with the pathophysiology of depression [19,20]. Vitamin D is a potent anti-inflammatory agent and reduces inflammation in the brain [21,22]. Lefebvre d'Hellencourt et al. [23] described the anti-inflammatory activity of vitamin D, and showed a negative correlation between the levels of IL-6 and TNF-α and vitamin D in microglial cells. Thus, the diminution of the anti-inflammatory effect of vitamin D in the vitamin D-deficient group might increase depressive symptoms in stroke patients.

Another possible mechanism is through the regulation of neurotransmitters. Vitamin D has been reported to regulate several neurotransmitters. Vitamin D promotes the gene expression of tyrosine hydroxylase, an essential enzyme involved in the synthesis of norepinephrine and dopamine, which are involved in mood regulation and depression [24]. Vitamin D also upregulates serotonin synthesis by activating the transcription of the serotonin-synthesizing enzyme tryptophan hydroxylase 2 in the brain [25]. Norepinephrine, dopamine, and serotonin are associated with the pathophysiology of depression, and these are target neurotransmitters for antidepressants, such as selective serotonin re-uptake inhibitors and serotonin-norepinephrine re-uptake inhibitors [26]. Several lines of experimental evidence have shown that dietary manipulation or direct addition of 25-OHD to the brain alters these neurotransmitter systems. Tekes et al. [27] reported that 25-OHD administered to rodents elevated dopamine and norepinephrine contents in the brainstem, and Smith et al. [28] reported that adult animals treated with high doses of 25-OHD showed increased basal striatal dopamine content. However, we did not evaluate whether the supplementation of vitamin D could improve the symptoms of depression in the stroke patients with vitamin D deficiency. Further studies are required on this.

Several reports have shown that serum 25-OHD levels are negatively correlated with scales of depression [29,30,31]. Polak et al. [30] reported that serum 25-OHD status was significantly negatively associated with a depression scale in young adults. Lee et al. [31] revealed a significant negative correlation between 25-OHD levels and BDI-II scores in their multi-center trial. In this study, a significant inverse correlation was seen between BDI-II scores and serum 25-OHD levels, consistent with the results of previous studies conducted on general populations. And, to our knowledge, this is the first report showing a negative linear correlation between serum 25-OHD levels and symptom scales of depression in stroke patients.

This study has some limitations. First, this study was cross-sectional, so the change in vitamin D levels or progression of depression was not evaluated; thus, we could not determine a temporal or causal association between vitamin D deficiency and depression in stroke patients. Second, we did not classify our participants by stroke severity or duration after stroke onset. Stroke severity or duration could have an influence on diet or level of activity, which may affect vitamin D metabolism. However, K-MBI score and duration after stroke onset did not differ between the two groups, and all of our participants were admitted patients in a university hospital engaged in a conventional rehabilitation treatment program. Thus, daily diet or physical activity would not differ largely between participants and would not have influenced our results. Third, the study sample was comprised of admitted patients who were motivated to engage in a rehabilitation treatment program. Thus, these results may not generalize to all stroke patients and the favorable outcome of our study could have resulted, in part, from patient selection bias. Fourth, free fraction of serum 25-OHD enters through an energy-independent passive mechanism, while protein bound 25-OHD is restricted and only a part crosses the brain-blood barrier through active transport [13,14]. Therefore, the concentration of serum 25-OHD is higher than those found in human CSF, and serum 25-OHD might not exactly reflect CSF and brain 25-OHD level. However, studies have shown a positive correlation between serum and CSF 25-OHD concentration, and concluded that CSF concentration is determined by the serum concentration, thus measuring serum 25-OHD is reasonable in this study [15,32]. Fifth, there were relatively few study participants (n=126) and they came from a single institution. Further research will be needed considering these limitations.

In conclusion, we showed that BDI-II, PHQ-9 scores differed significantly between a vitamin D-deficient group and a normal group in stroke patients. Additionally, the correlation between serum 25-OHD levels and BDI-II scores was significant. We suggest that vitamin D levels may be a clinical marker for post-stroke depression, and physicians should pay attention to whether stroke patients with vitamin D deficiency have depression. Future studies are needed to confirm our findings and determine whether vitamin D supplementation could influence post-stroke depression.