Predictors of early post ischemic stroke apathy and depression: a cross-sectional study

We reviewed the charts of 270 patients hospitalized for an acute stroke between December 2010 and December 2011. After detail evaluation with inclusion and exclusion criteria, 75 patients were used for this study. Inclusion criteria were: patients with a diagnosis of ischemic stroke (IS) according to valid ICD-10 codes [11], the stroke had occurred within the previous 7 days, computerized tomography (CT) was performed to exclude cerebral hemorrhage, patients were conscious and their Glasgow Coma Scale (GCS) scores were 15, patients were able to cooperate with the testing, and their admission NIH Stroke Scale (NIHSS) score was ≤8. All patients had acute lacunar infarctions.

Exclusion criteria were: patients with a diagnosis or history of atrial fibrillation, those with a history of schizophrenia, depression, anxiety or other mental illness, including dementia, Parkinson’s disease and other neurodegenerative diseases, the development of symptoms of severe aphasia so that patients were unable to complete the evaluation, severe heart, liver or kidney dysfunction, malignancies or thyroid diseases, contraindications to magnetic resonance imaging (MRI) such as pacemaker implantation and claustrophobia, drug abuse or drug dependence, and refusal to participate in the study. Smoking was defined as: Non-smoker, Ex-smoker who used to smoke but had quit, Current smoker was defined as those who continued to smoke 1–19 cigarettes per day during the study period [12]. Drinking was defined as: (1) Non-drinker who did not have a history of drinking in the past or was an occasional drinker (< 1 drink, where 1 unit equals to 8 gram of alcohol per week), (2) Light drinker of 1–15 units per week, (3) Moderate drinker of 16–42 units per week, (4) Heavy drinker of > 42 units per week, (5) Ex-drinker who used to drink but had quit [13].

Patients who met the inclusion criteria were recruited and their clinical information was recorded in detail. Data were recorded on hypertension, diabetes, a history of cerebral vascular disease and other chronic diseases. The NIHSS score had been recorded on admission. The evaluation of apathy was carried out using the Apathy Evaluation Scale, clinician version (AES-C) within 2 weeks after onset of the stroke. The AES was developed by Marin et al. and its reliability has been shown to be satisfactory [14]. The diagnostic criteria for apathy in this study were based on the Marin scale [14] and Robert et al. [15]. These criteria were (1) a lack of motivation relative to the patient’s previous level of functioning or the standards of his/her age and culture, (2) the presence of at least one symptom from the following three domains: (i) diminished goal-directed behavior (lack of effort, dependency on others to structure activity); (ii) diminished goal-directed cognition (lack of interest or concern about one’s personal problems), or (iii) diminished concomitant to goal-directed behavior (unchanging affect, lack of emotional responsiveness), (3) the symptoms cause clinically significant distress or impairment in social and occupational functioning, and (4) the absence of a diminished level of consciousness or administration of substances such as narcotics or medications [15].

Within 2 weeks after the stroke, the Montreal Cognitive Assessment (MoCA), mini-mental state examination (MMSE), Hamilton Depression Scale (HAMD), Mattis Dementia Rating Scale initiation/perseverance subscale (MDRS I/P), Frontal Assessment Battery (FAB) and Stroop Color Word Test were employed to evaluate emotion, cognitive function and executive function. The MDRS can be used to evaluate attention, initiation/perseveration (I/P), construction, conceptualization and memory. The reliability of the MDRS Chinese version has been validated [16]. The MDRS I/P is one part of the MDRS and includes language production and coherent motion, and is used to assess the ability to initiate and sustain motion, as these are closely related to executive function of the frontal lobe. In the Stroop Color Word Test, two cards were used for testing. Card A was the Stroop color card and Card B the color word card. The time to complete the evaluation with each card was recorded and the difference between them was calculated (B-A_time). The number of correct answers was recorded for each card and the difference was calculated (B-A_number). On the HAMD, >35 points was considered to be an indication of serious depression and >20 points as mild or moderate depression. A MoCA score of <25 but ≥14 was defined as mild cognitive impairment (MCI). The scale assessments for the FAB were performed by a qualified psychiatric specialist and the testing was confirmed by a psychologist and a neurologist.

According to the Diagnostic and Statistical Manual of Mental Disorders (DSM IV), the criteria for major depression area depressed state of mind for most of the day, and at least four of the following: (1) loss of interest in or no pleasant feelings about activities, (2) the loss of energy or fatigue, (3) psychomotor retardation or agitation, (4) feelings of worthlessness, (5) diminished ability to think or concentrate, (6) recurrent thoughts of death, (7) sleep disorders such as insomnia or sleeping too much, and (8) significant weight loss or weight gain. The criteria for severity include impaired social functioning and pain or adverse consequences for the depressed individual. The symptom criteria and the severity criteria must have lasted for at least 2 weeks. The symptoms cannot be due to substances or a general medical condition.

According to the findings on T2-weighted images, the infarct lesions were located in the bilateral frontal lobes, temporal lobes, parietal lobe, occipital lobe, basal ganglia, corona radiata or brain stem.

The AES-C consists of 18 items which provide an index of overt goal-directed behavior, goal-related cognitions, and goal-related emotional responses. Factor analysis using principal component analysis was performed. The Kaiser-Meyer-Olkin (KMO) score for the AES-C was 0.84, indicating sampling adequacy (≥.50). Bartlett’s test of sphericity was statistically significant. The overall Cronbach’s alpha for internal consistency of the 18-item instrument was 0.87. HbA1c shows the average level of blood glucose over the previous three months and < 5.6% is considered normal.

In this study, we compared 25 stroke patients who had apathy with 50 stroke patients who did not. The two groups were similar except that the apathy group was significantly older, had significantly fewer years of education, had a significantly higher proportion of frontal lobe lesions, and had a significantly lower level of HbA1c. The patients were assessed within 2 weeks of their stroke. Multivariate logistic regression analysis revealed three risk factors for apathy: a history of cerebrovascular disease, a low HbA1c level and a low MDRS I/P score. We also compared 12 patients who had post-stroke depression with 63 who did not. The groups were similar except that a significantly higher proportion of patients with depression had temporal lobe lesions.

Our results differed from those of previous studies that had attempted to identify risk factors for post-stroke apathy. Sagen et al. [17] assessed patients within the first 2 weeks and at 4 months after the stroke, and, using multivariate logistic regression analysis, found that at 4 months after the stroke, somatic comorbidity was significantly associated with apathy and there was a borderline association with an HADS-D score ≥8 at admission. Caeiro et al. [3] assessed patients during the acute phase of a stroke (≤ 4 days) and stepwise regression analysis showed that cerebral hemorrhage, a low level of education and a right hemispheric lesion were predictors. Brodaty et al. [4] assessed patients at 3–6 months after a stroke and performed univariate analysis which showed that older age, more functional dependency and a lower score on the MMSE were associated with apathy. Mayo et al. [18] assessed apathy in 408 stroke patients by telephone interviews with caregivers at 1, 3, 6 and 12 months after the stroke and analyzed the data using the group-based trajectory method; they found that greater apathy was predicted by older age, poor cognitive status and low functional status. One possible reason for the conflicting findings is that assessments were carried out at different periods after the stroke. In our study, patients were assessed within 2 weeks of the stroke during the subacute phase. Different variables were assessed and different assessment instruments were used.

We found that a decrease in the MDRS I/P score was a risk factor for both post-stroke apathy and post-stroke depression. Patients with one or both of these scoured lower than did the control group. Patients with post-stroke apathy without depression also had lower scores on the MMSE and the MoCA than did controls; however, they had lower HAMD scores than those without apathy but with depression, suggesting that these groups are really different.

In our study, abnormalities were not found in B-A_time and B-A_number. The Stroop Color Word Test can be used to evaluate the degree of impairment of the lateral and inner-upper frontal lobe [19], but fails to specifically assess the impairment of the whole frontal lobe.

The frontal-subcortical circuit has been found to connect the frontal cortex and the striatum, globus pallidus, substantia nigra and thalamus, and functions as an important effector mechanism and interacts adaptively [20, 21]. Dysfunction of the frontal-subcortical circuit is characterized by a reduction in executive function, apathy and impulsive behavior. Studies have revealed that most stroke patients with depression and executive dysfunction had infarct lesions in the frontal lobe [22, 23]. A study of participants in the Age Gene/Environment Susceptibility-Reykjavik Study found that infarct lesions involving both cortex and subcortex were related to compromised processing and executive function [24]. Another study found that among patients with frontotemporal dementia, the severity of apathy was associated with atrophy in the right dorsolateral prefrontal cortex [25]. Brodaty et al. found that among stroke patients who underwent MRI there was a trend for apathy to be associated with the extent of MR signal abnormalities (hypertintensities) in the deep white matter and in the right hemisphere and right frontal-subcortical circuit [4].

We also found that a history of cerebrovascular disease was a risk factor for post-stroke apathy. Angelelli et al. found that apathy was more frequently found at 6 months after a stroke than during the subacute phase [26]. Following cerebral infarction, the nerve fibers distant from the lesion may experience secondary damage. Animal studies have shown that connectivity remodeling was present in the cortex, hypothalamus, striatum, hippocampus and tissues surrounding the lesions following a stroke [27‐30]. We therefore speculate that one possible explanation for a history of cerebrovascular disease as a risk factor for post-stroke apathy is that infarcts and secondary damage to nerve fibers following a stroke may impair the frontal-subcortical circuit and connectivity remodeling. Alexopoulos et al. [31] postulated two broad hypotheses for the mechanism of vascular depression. The first hypothesis was that small lesions disrupting critical neural pathways might precipitate vascular depression. The second was that an accumulation of lesions exceeding a threshold predisposes to depression. The second hypothesis, which is a “threshold hypothesis,” is most applicable to patients who have neurologically silent lesions or an old stroke. Hama et al. [32] investigated the correlation between damage to the basal ganglia or frontal lobe and depression (both affective and apathetic dimensions) in a study of 243 stroke patients examined with CT, and suggested that affective depression was associated with left frontal lobe damage whereas apathetic depression was associated with damage to the basal ganglia in both hemispheres. Our finding that there was a tendency for a larger proportion of patients in the apathy group to have a history of cerebrovascular disease compared with the non-apathy group suggests that there was a larger proportion of recurrent stroke cases in the apathy group compared with the non-apathy group, and that our finding supports the threshold hypothesis. Cognitive function was lower in the apathy group compared with the non-apathy group. These findings related to a history of cerebrovascular disease and cognitive function suggest that repeated ischemic strokes damage brain function bit by bit, leading to both apathy and cognitive dysfunction.

We found that the level of HbA1c was also a risk factor for post-stroke apathy. Padala et al. [33] found that apathy was prevalent in patients with diabetes without depression. The mean HbA1c level was 0.66% lower for apathetic patients as compared to non-apathetic ones. Apathy may have a negative impact on self-care behaviors and diabetes control [33]. To the best of our knowledge, this is the first time that this variable has been associated with post-stroke apathy. Because of small sample size, however, it is very likely that the statistically significant relationship between HbA1c and post-stroke apathy that we found was due to chance. Therefore, this possible relationship needs to be explored further in larger studies before conclusions can be drawn.

Notably, results from our additional logistic regression analysis on patients separated into four categories, using patients without apathy and depression as the reference category, showed that most findings are largely similar; however, this additional analysis is limited by the lack of multivariate analysis due to small sample size. In this limited analysis, HbA1c is not a risk factor for post-stroke apathy, and new brain areas had also shown significant difference such as brain stem for post-stroke apathy and cerebellum for post-stroke depression. Thus, this highlights the importance of having more studies with larger sample size in order to confirm our findings.

Our study had several limitations. The sample size of 75 stroke patients was small and the study was carried out at only one medical center with the testing instruments used being selected based on the availability of department and hospital team support. Also, there was no further investigation of the influence of damage in the cerebral regions on apathy or on the influence of impaired connectivity in each cerebral region on apathy. In addition, we excluded patients with severe aphasia because they could not complete the evaluation, and this might limit the generalization of our findings. Moreover, as pointed out by Mayo et al. [18], based on modern psychometric standards, there is no available measurement of apathy that would be considered optimal. Lastly, we attempted to identify “risk factors” for apathy or depression alone; however, results of multivariate multinomial logistic regression were not estimable due to small sample size.