Post-Stroke Apathy: An Exploratory Longitudinal StudyCaeiro L.a · Ferro J.M.b · Pinho e Melo T.b · Canhão P.b · Figueira M.L.c
aInstitute of Molecular Medicine, bStroke Unit, Neurology Service, and cPsychiatry Service, Department of Neurosciences, Hospital de Santa Maria, University of Lisbon, Lisbon, Portugal Corresponding Author
Introduction: Post-stroke apathy is a disturbance of motivation evidenced by low initiative, difficulties in starting, sustaining or finishing any goal-directed activity, low self-activation or self-initiated behavior and emotional indifference. Apathy is a common behavioral disturbance in stroke survivors. We aimed to analyze the relationship between post-stroke apathy at 1 year after stroke and (1) apathy in acute phase; (2) demographic, pre-stroke predisposing conditions (previous mild cognitive impairment, alcohol abuse, mood/anxiety disorder) and clinical features (stroke type and location, neurological symptoms); (3) post-stroke depression and post-stroke cognitive impairment, and (4) post-stroke functional outcome, quality of life and the perception of health. Methods: Consecutive stroke (infarct/intracerebral hemorrhage) patients without aphasia or consciousness disturbances were included in the acute phase of stroke and assessed at 1 year after stroke. We assessed apathy with the clinically rated version of the Apathy Evaluation Scale. We also assessed post-stroke depression (Montgomery Asberg Depression Rating Scale) and post-stroke cognitive impairment (attention, mental flexibility, verbal, motor and graphomotor initiative, and non-verbal and verbal abstract reasoning, and Mini-Mental State Examination), functional outcome (Barthel Index), quality of life and perception of health (EuroQol). Data were analyzed using bivariate associations (χ2 and t test) and stepwise multivariate analysis. Results: We included 76 stroke patients [32.9% women, mean age 62.9 years (SD = 10.9) and a mean of 6.9 (SD = 4.3) years of education]. Apathy was present in 17 patients in the acute phase and in 18 (23.7%) patients at 1 year after stroke. At 1 year after stroke, 41% of the acute apathetic patients remained apathetic. Sixty-one percent of new cases of post-stroke apathy were detected. Post-stroke apathy was associated only with previous cognitive impairment, apathy in acute stroke, post-stroke cognitive impairment, verbal abstract reasoning and with worse Barthel Index scale scores. In the multivariate logistic regression model, verbal abstract reasoning (odds ratio, OR = 7.03) and apathy in acute stroke (OR = 3.8) were identified as independent factors for post-stroke apathy at 1 year. Apathetic patients did not report worse quality of life or health. Conclusion: Apathy in acute stroke phase was a reliable indicator of post-stroke apathy. Apathy should be assessed in both phases. Verbal abstract reasoning impairment was also an independent factor for post-stroke apathy impairing patients' ability to reason about goal-directed activity. Even though apathetic patients had worse post-stroke functional outcome, they did not report losing quality of life or having worse health.
© 2013 S. Karger AG, Basel
Post-stroke apathy is a disturbance of motivation evidenced by low initiative, difficulties in starting, sustaining or finishing any goal-directed activity, low self-activation or self-initiated behavior and emotional indifference [1,2]. Post-stroke apathy can be included in the DSM-IV-TR  criteria of Personality Change due to Stroke, characterized by apathy and indifference.
Recent studies on post-stroke apathy were performed using validated apathy scales [4,5,6,7,8]. Post-stroke apathy is a disabling symptom present in 20-55% of stroke survivors [6,7,9,10,11,12,13,14,15]. Post-stroke apathy has been associated with post-stroke cognitive impairment [6,10,12,15,16], and specifically with executive functioning impairment [6,17,18,19,20], but not all studies supported that association [7,11,13,14,21,22,23].
In previous publications, post-stroke apathy and post-stroke depression overlapped at 3, 6 and 12 months' follow-up [6,12,13,24] but not at 15-months' follow-up . A review of the literature  and a systematic review  concluded that post-stroke apathy may be associated with post-stroke depression, but both can arise separately.
Several studies found an association between post-stroke apathy and poor functional outcome [9,10,12,16,22] or inability to return to previous occupational and social activities. However, a systematic review  did not confirm this association. Nonetheless, the quality of life of apathetic stroke patients and their perception about their own health was not previously studied.
Only two longitudinal studies [6,10] reported the relationship between apathy in the acute phase of stroke with post-stroke apathy. In one of these, the follow-up was performed by telephone interview .
For this study, we operationalized the following research questions: (1) Is post-stroke apathy related to apathy in acute phase of stroke? (2) Is post-stroke apathy associated with age or with right-sided and subcortical lesion? (3) Is post-stroke apathy associated with post-stroke depression? (4) Is post-stroke apathy associated with post-stroke cognitive impairment? (5) Do post-stroke apathetic patients have a worse functional outcome, low quality of life and the perception of poor health in the post-stroke phase?
Materials and Methods
Between 2006 and 2008, we studied prospectively consecutive stroke patients (infarct or intracerebral hemorrhage). Patients in the acute phase of stroke were admitted to the stroke unit of the Neurology Department of a University Hospital. All patients performed a post-stroke neuropsychiatric and neuropsychological evaluation at 1-year follow-up. We obtained informed consent from all patients.
Following a previously described methodology , we excluded acute stroke patients who had a severe communication disturbance, defined as a score of 2-3 on the Neurological Institute Health Stroke Scale , items ‘Best language' or ‘Dysarthria', and patients with a Glasgow Coma Scale  score below 9. We also excluded patients who did not perform the evaluation at 1-year follow-up.
In the acute phase, a neurologist (T.P.M., P.C. or J.M.F.) collected clinical and neuroimaging data. All patients had a CT or MRI, including DWI, performed in the acute phase of stroke to define the location of the lesion. The location of stroke was categorized as (1) brainstem/cerebellum or hemispheric; (2) hemispheric left or right; (3) hemispheric subcortical, cortical or corticosubcortical, and (4) thalamic or striatocapsular . If the stroke lesion could not be identified by neuroimaging, we used clinical criteria (symptoms and signs) to classify the stroke as brainstem/cerebellum or hemispheric, and hemispheric left or right. Ischemic stroke was classified according to the Oxford Community Stroke Project classification (PACI, TACI, POCI and LACI) .
The following pre-stroke predisposing conditions for post-stroke apathy were considered: (1) mild cognitive impairment, defined as a medical diagnosis of mild cognitive impairment or a history of memory and another cognitive domain disorder with no functional impairment in daily living activities, confirmed by a proxy, (2) alcohol abuse, defined as 5 or more drinks daily, and (3) previous mood or anxiety disorder , defined if the patient had a previous diagnosis of a mood or anxiety disorder or if the patient had been either prescribed specific medication for these conditions and used it for more than a month.
At discharge, functional outcome was assessed with the modified Rankin Scale (mRS) . Unfavorable outcome was defined as an mRS score ≥3 (death or dependency).
At 1-year follow-up, we used the Barthel Index (range 0-100 points)  to assess the post-stroke functional outcome in daily living activities. To assess post-stroke quality of life, we used the EuroQol (range 0-10 points). The EuroQol includes a self-rated perception of heath (health self-rating; range 0-100 points) .
A trained psychologist (L.C.) performed an interview in the acute phase of stroke, whenever possible during the first 7 days after stroke onset. This interview aimed to collect pre-stroke predisposing conditions and evaluate apathy in acute stroke. The same psychologist performed a neuropsychiatric and neuropsychological evaluation at 1-year follow-up.
Apathy was evaluated with the 18-item clinically rated version of the Apathy Evaluation Scale (AES-C/Clinical-Rated Apathy) . This evaluation was performed in the acute phase of stroke and at 1-year follow-up. The AES-C cutoff points were used to define the presence of apathy, taking into consideration Portuguese educational levels . An evaluation of self-rated apathy was performed using the self-rated version of the AES (AES-S/Self-Rated Apathy) . The AES-S was used only for a comparison between clinical-rated and self-rated differences of post-stroke apathy; no other analyses were performed with the AES-S.
At 1-year follow-up, patients were diagnosed as having post-stroke depression if they fulfilled the DSM-IV-TR criteria for Mood Disorder due to Stroke (post-stroke depression) , i.e. if they reported or displayed depressive mood and anhedonia, and scored ≥7 points on the Montgomery-Asberg Depression Rating Scale [36,37].
At 1-year follow-up, to assess post-stroke cognitive impairment, we used the Mini-Mental State Examination . The normative data of the Portuguese validation defined cutoff points taking educational levels into consideration . We also assessed post-stroke executive functioning including attention (Trail making test A), mental flexibility (Trail making test B), verbal initiative (Verbal fluency-food products), motor initiative (Luria's alternate hand sequences), graphomotor initiative (Luria's alternating series), verbal abstract reasoning (Proverbs) and non-verbal abstract reasoning (Raven's Progressive Matrices Ab). We used normative data of the Portuguese validation for each test [40,41]. Patients were classified as having impairment in a particular test when they scored 1.5 standard deviation [42,43,44] below the mean (for their age and educational level).
The Ethics Committee of the Faculty of Medicine, University of Lisbon, approved the study.
Data were analyzed using IBM SPSS Statistics version 19. For categorical variables, we used the χ2, with continuity correction if necessary, and for continuous variables we used the independent t test to test bivariate associations. These statistical associations were performed between clinically rated post-stroke apathy and gender, age (<65 years old or ≥65 years old), educational level (0-9 or ≥10 years of education), previous mild cognitive impairment, previous alcohol abuse, and previous psychiatric disorder, stroke type, stroke location and lateralization, apathy in acute stroke, post-stroke depression, post-stroke cognitive impairment, post-stroke executive dysfunctioning, mRS grade at discharge (0-2 or ≥3), Barthel Index, EuroQol and health scores.
In order to find which variables were strongly associated with post-stroke apathy, we performed a multivariate analysis stepwise logistic regression model [odds ratio (OR) and Nagelkerke R2 (R2)] for post-stroke apathy at 1-year follow-up, entering significant variables with a p ≤ 0.15  on the bivariate analysis.
A p value ≤0.05 was statistically significant. No correction for multiple comparisons was performed .
Ninety-eight patients were included in the acute phase of stroke. Of these, 76 patients (77.7%) were assessed at 1-year follow-up (table 1). There were no statistical differences in gender (χ2 = 0.00, p = 0.92), age (t = 0.60, p = 0.55), educational level (t = 0.10, p = 0.93), previous mild cognitive impairment (χ2 = 0.52, p = 0.77) or apathy in acute stroke (χ2 = 0.02, p = 0.90) between those patients who participated at 1-year follow-up and those who did not. Twenty-two patients were not assessed because 1 had died at the 6-month follow-up, 3 asked to be excluded from the study at 6 months, 4 asked to be excluded at 1-year follow-up, 14 missed the four appointments made for them to come to the reevaluation at 1 year.
|Table 1. Characteristics of the sample (n = 76) and results of bivariate analysis|
The sample of 76 patients included 25 (32.9%) women, with a mean age of 62.9 years (SD = 10.9) and a mean of 6.9 (SD = 4.3) years of education. The remaining characteristics of the sample are depicted in table 1.
In the acute phase, 17 patients were clinically apathetic, and of these 5 also self-reported apathy. Only one patient of a total of 6 self-reporting apathy was not clinically rated as apathetic. Acute clinically rated apathy and acute self-rated apathy were significantly associated (χ2 = 10.2; p = 0.001).
At 1-year follow-up, post-stroke apathy was present in 18 (23.7%) patients. Twelve (15.8%) patients self-reported post-stroke apathy. Only 1 patient self-reporting post-stroke apathy was not clinically rated as apathetic (χ2 = 32.1; p = 0.000). Of the 18 apathetic patients, 17 had severe apathy (upper 5th quintile of the AES-C score distribution: AES-C ≥37 points). Seven (41.2%) acute apathetic patients remained apathetic at 1-year follow-up (fig. 1). Eleven new cases of post-stroke apathy were identified at 1-year follow-up. Comparing the 7 patients who remained apathetic at 1-year follow-up with the 11 new cases of post-stroke apathy, we did not find any significant differences in demographic, clinical, acute or post-acute variables.
|Fig. 1. Changes in the frequencies of apathy in the acute phase of stroke and at 1-year follow-up (76 patients).|
Post-stroke apathy was associated with previous mild cognitive impairment, but not with demographic or other predisposing conditions. No associations were found between post-stroke apathy and lesion location in spite of a trend association with hemispherical location (88.9% of apathetic patients at 1 year had hemispheric lesion vs. 65.5% in non-apathetic patients; table 1).
At 1-year follow-up, no association was found between post-stroke apathy and post-stroke depression. Post-stroke apathy was associated with post-stroke verbal abstract reasoning disturbances, and a trend association was found with global post-stroke cognitive impairment (table 1).
Post-stroke apathetic patients scored lower in the Barthel Index when compared with non-apathetic patients. Apathetic post-stroke patients did not report lower quality of life or health (table 1).
We performed exploratory stepwise logistic regressions entering significant variables with a p ≤ 0.15 on the bivariate analysis, to assess if they were independently associated with post-stroke apathy.
This model revealed that post-stroke verbal abstract reasoning (OR = 7.03, 95% CI = 1.4-33.3), and apathy in acute stroke (OR = 3.8, 95% CI = 0.97-14.9) were independently associated with post-stroke apathy at 1-year follow-up (R2 = 23.3%, area under the curve = 71%; specificity = 83.8%; sensitivity = 100%; positive predictive value = 21.4%; negative predictive value = 100.0%).
At 1-year follow-up, 11 new cases of post-stroke apathy were identified. The bivariate analysis with these 11 new cases (vs. 65 non-new cases of post-stroke apathy) identified an association with post-stroke verbal abstract reasoning disturbances (χ2 = 4.1; p = 0.04). When we compared the 11 new cases of post-stroke apathy with the 48 patients who never presented apathy, either in the acute phase or in post-stroke phase, we only found an association with post-stroke verbal abstract reasoning disturbances (χ2 = 3.9; p = 0.05). Further analyses were performed comparing 4 groups of patients (7 who were apathetic in the acute and post-acute phase; 11 presenting only post-stroke apathy; 10 presenting only apathy in the acute phase; 48 who never had apathy), but the results were similar, although results should be interpreted with caution because each of the subgroups had a modest size (online suppl. table 1; see www. karger.com/doi/10.1159/000350202).
We aimed to study the temporal evolution of apathy at 1 year after stroke and its relationship with apathy in acute stroke, demographic, pre-stroke predisposing conditions and clinical features (stroke type and location), post-stroke depression and cognitive impairment, and with functional outcome, quality of life and health. We found a frequency of 23.7% of post-stroke apathy. Of the patients with apathy in the acute phase of stroke, 41% remained apathetic at 1-year follow-up. Verbal abstract reasoning (OR = 7.03) and apathy in acute stroke (OR = 3.8) were independent factors for post-stroke apathy at 1-year follow-up. Post-stroke apathetic patients reported worse functional outcome.
In this study, we had some limitations and bias inducers: (1) inclusion of a sample of stroke patients of a modest size; (2) exclusion of 7 acute stroke patients with language disturbances (severe aphasia or dysarthria) may lead to a selection bias in relation to stroke lesion lateralization; (3) loss of 22 patients during the follow-up, and (4) non-visualization of the acute stroke lesion in two CT scans in 26 patients. These patients did not undergo an MRI scan, which limits the analysis of the role of the stroke lesion location in post-stroke apathy. The lack of a second MRI scan at follow-up to search for new stroke events was also a limitation.
The first goal of our study was to identify if apathy in acute stroke predisposes to post-stroke apathy. We found a high frequency of post-stroke apathy, which is within the range of recent publications (20-48.8% [7,9,10,16,47]). Forty-one percent of the patients with apathy in acute stroke remained apathetic at 1-year follow-up. Apathy in acute stroke increases the risk of post-stroke apathy by almost 4 times. This confirms previous findings from the study by Mayo et al. . Previously, Angelelli et al.  had identified a three-fold risk of development of post-stroke apathy at 6-month/1-year follow-up in patients presenting apathy at 2-month follow-up. Eleven (61%) new cases of post-stroke apathy were identified at 1-year follow-up. No demographic or acute clinical variables were identified as risk factors for new cases of post-stroke apathy. No associations were found between post-stroke apathy and acute demographic features or pre-stroke predisposing conditions.
Post-stroke apathy was not associated with any particular acute stroke location. Previously, 3 studies had associated post-stroke apathy with frontal lobe areas [13,48,49]. Considering the lateralization of the stroke, other publications found a relationship between post-stroke apathy and right-sided cortical and subcortical lesions [12,13,50], left basal ganglia  and bilateral basal ganglia lesions .
We aimed to identify the relationship between post-stroke apathy and post-stroke depression. As in some previous studies [10,12], we did not find a statistical relationship between post-stroke apathy and post-stroke depression. The occurrence of depression is relatively the same between the apathetic and non-apathetic groups, corroborating previous studies . Aybek et al.  reported that 14% of patients with post-stroke apathy and depression share neuropsychological and neuroimaging findings. Jorge et al.  reported 8% of coexistence of post-stroke apathy and depression. In our sample, 27.8% of the patients with post-stroke apathy also had post-stroke depression, which is higher than the previous studies.
We did not identify any relationship between post-stroke apathy and post-stroke cognitive impairment, which had been previously reported [51,52]. Other studies reported a relationship between post-stroke apathy and cognitive impairment [6,10,12,16]. At 1 year, post-stroke apathy was associated only with post-stroke verbal abstract reasoning impairments. In previous studies, post-stroke apathy was related to cognitive impairments, which included concentration, information processing speed and executive functioning impairments [6,10,12,16,18,20,22,50].
Finally, the last goal was to find a relationship between post-stroke apathy and post-stroke poor functional outcome, quality of life and perception of health. Patients with post-stroke apathy had worse functional outcome when compared with non-apathetic patients. Nevertheless, apathetic post-stroke patients did not report lower quality of life or of health when compared with non-apathetic post-stroke patients. In 2 previous studies [6,9,10], post-stroke apathy had already been related to functional dependency. In a recent systematic review , it was shown that post-stroke apathy did not appear to have a negative impact on functional outcome. Additionally, from our study we may conclude that post-stroke apathetic patients did not evaluate worse functional outcome as a worsening in quality of life or of health, when compared with non-apathetic patients.
Post-stroke apathy can be identified in 2 out of every 5 apathetic acute stroke patients. The assessment of apathy should be included in the evaluation performed in the acute and post-acute stroke phases. Post-stroke apathy is a neuropsychiatric disturbance independent of post-stroke depression. Post-stroke apathetic patients present verbal abstract reasoning impairments which reduce their ability to find reasons for starting, sustaining or finishing any goal-directed activity. Although post-stroke apathy was related to poor functional outcome, apathetic patients did not see it as a loss in the quality of life or health.
This research was partly supported by the Fundação para a Ciência e a Tecnologia (PhD scholarship, ref. SFRH/BD/22282/ 2005, attributed to L.C.).
All authors declare no conflicts of interest.
Serviço de Neurologia, Hospital de Santa Maria
Av. Professor Egas Moniz
PT-1649-035 Lisbon (Portugal)
Received: October 18, 2012
Accepted: February 20, 2013
Published online: June 5, 2013
Number of Print Pages : 7
Number of Figures : 1, Number of Tables : 1, Number of References : 52
Additional supplementary material is available online - Number of Parts : 1
Vol. 35, No. 6, Year 2013 (Cover Date: July 2013)
Journal Editor: Hennerici M.G. (Mannheim)
ISSN: 1015-9770 (Print), eISSN: 1421-9786 (Online)
For additional information: http://www.karger.com/CED