Post-stroke depressive mood (PSDM) is a serious and common complication that occurs in one-third of the total number of stroke patients [1]. As PSDM has been found to significantly affect functional outcome, resumption of social activities, and quality of life [2], numerous studies have been conducted to evaluate and manage PSDM in patients in the early phase of stroke [3,4,5,6,7,8]. Although the studies showed inconsistent results, factors such as sex, stroke severity, functional impairments, family, and social support were the most frequently mentioned risk factors [9]. However, most of the studies included patients who had a supratentorial stroke.

Recent bodies of evidence from studies have revealed that the cerebellum might play an important role in the regulation of emotion [10,11]. The study by Shah et al. [12] showed that patients with a major depressive disorder have reduced cerebellar volume. Leroi et al. [13] found that patients with cerebellar degeneration had high prevalence of depressive mood. In cerebellar stroke patients, Frank et al. [14] observed significantly high depression scores. However, there are a few studies on PSDM after isolated cerebellar stroke.

Therefore, the aim of this study was to analyze the association between the clinical data of isolated cerebellar stroke patients and the occurrence of depressive mood in order to identify predictive factors of depressive mood after isolated cerebellar stroke and help early detection.

Among all the patients, 137 had cerebellar stroke, of whom 52 satisfied both the inclusion and exclusion criteria, as shown in Fig. 1. We classified 23 subjects into the non-depressive group and 29 subjects into the depressive group (Fig. 1).

In this study, older age, history of hypertension, ischemic lesion rather than hemorrhagic lesion, and lesion of the right posterior cerebellar hemisphere were correlated with depressive mood in isolated cerebellar stroke patients. When the effects of the other factors were assessed, lesion of the right posterior cerebellar hemisphere was the only factor that correlated with depressive mood.

Previous research studies identified female sex [19], severe functional impairment [20,21], previous history of depression [22], aphasia [23], and absence of family and social support as the risk factors for PSDM [24]. However, our study did not show a relationship between these factors and depressive mood after cerebellar stroke. As the subjects of our study were limited to those with isolated cerebellar stroke, it may be difficult to apply the risk factors reported in previous studies.

In this study, older patients were more depressed than younger patients. In a normal population, the aging process itself increases the risk of depressive mood [25]. In stroke patients, the tendency for developing depressive mood after stroke is more frequently observed at an older age [26]. The atherosclerotic, inflammatory, endocrinologic and immunologic changes related to the aging process might form the pathophysiological mechanism that is responsible for increasing the vulnerability to depression [27].

The risk of depressive mood was higher in the patients who had a history of hypertension. Previous history of hypertension could cause the change of periventricular white matter hyperintensity [28], which is one of the risk factors for depressive mood in the elderly [29]. Due to this reason, the comorbid condition of previous hypertension can increase the risk of depressive mood in patients with isolated cerebellar stroke, although we did not measure the lesion of periventricular white matter change. Regarding another probable explanation, a study conducted by Meurs et al. [30] demonstrated that the patients with the combination of major depressive disorder and hypertension showed decreases of brain volumes in the areas associated with emotional regulation, such as anterior and middle cingulate cortices and cerebellum. The alterations of brain volume in these emotional regulatory areas might be an explanation for the comorbidity between hypertension and depressive mood. Further studies should be performed to confirm our results.

The patients with ischemic cerebellar stroke were more vulnerable to depressive mood than those with hemorrhagic stroke; however, previous studies showed controversial results [20,24,31]. The present study showed that ischemic stroke was bivariately related to depressive mood, but it failed to demonstrate the relationship in a multivariate logistic regression model. Similar results were observed in a study conducted by van de Port et al. [31]. In this study, ischemic stroke patients were significantly older than hemorrhagic stroke patients (59.9±14.0 vs. 47.4±21.5 years; p=0.014). Thus, older age could cause higher vulnerability to depressive mood in ischemic stroke patients than in hemorrhagic stroke patients.

Our study showed that lesion of the right posterior cerebellar hemisphere was a risk factor for depressive mood, which was consistent with the findings of prior research studies [32,33]. The researchers argued that the posterior lobe, not the anterior lobe, contributed to higher-level processes such as mood regulation [32]. Besides, Damasio et al. [33] demonstrated that the right posterior cerebellar hemisphere was significantly activated by emotions such as anger and fear. A neuroanatomical study has shown that cerebellar hemispheres project to the contralateral dorsolateral prefrontal cortex (DLPFC) through dentatothalamic fiber tracts and receive cortical input back to the cerebellum, forming a closed prefrontal-cerebellar circuit [34]. Imbalance between the left and right DLPFC was demonstrated in major depressive disorder and it was linked to negative emotional judgment [35]. In our study, a lesion in the right posterior cerebellar hemisphere might have affected the balance between the left and right DLPFCs via the prefrontal-cerebellar circuit, causing depressive mood. Although prior studies showed that the cerebellar vermis and paravermian area are the most important anatomical locations in the cerebellum responsible for emotional control [10,36,37], our results showed no statistical correlation between cerebellar vermian lesion and depressive mood. The reason for this might be the small number of patients with a cerebellar vermian lesion alone. Most of the subjects had mixed hemispheric and vermian lesions, or no vermian lesion at all. This might have affected the results. Further study is needed to localize the lesion associated with depressive mood in patients with isolated cerebellar stroke by using functional neuroimaging for brain connectivity.

This study has some limitations. First, the subjects' cognitive impairment could have influenced the development of depressive mood, as they were patients who had mild to moderate cognitive impairment. As GDS is a screening tool for subjective depressive mood, reliable data could not be obtained from patients with moderate cognitive impairment. Second, the development of depressive mood could be affected by the other functional factors in subjects such as balance [38] and fatigue [39]. In our study, the subjects' functional status was evaluated only with the FIM score, which does not correlate well with balance function or energy expenditure in stroke patients [40]. Third, the number of subjects was small. Further study with a prospective cohort design, large number of subjects, and considerations of cognitive and other functional factors is needed.

In conclusion, older age, history of hypertension, ischemic stroke, and especially lesion of the right posterior cerebellar hemisphere in patients with cerebellar stroke could confer a higher risk of developing depressive mood. During rehabilitation, the patients with these factors should undergo careful observation of emotional changes for early detection and management of depressive mood in patients with isolated cerebellar stroke.