Journal of Cardiac Failure Vol. 20 No. 9 2014

Cognitive Dysfunction in Older Adults Hospitalized for Acute Heart Failure CYNTHIA ARSLANIAN-ENGOREN, PhD, RN, ACNS-BC, FAHA, FAAN,1 BRUNO J. GIORDANI, PhD,2 DONNA ALGASE, PhD, RN, FGSA, F-NGNA, FAAN,1 AMANDA SCHUH, MS, RN, PMHNP-BC,1 CORINNE LEE, DNP, RN,3 AND DEBRA K. MOSER, DNSc, RN, FAHA, FAAN4 Ann Arbor, Michigan; and Lexington, Kentucky

ABSTRACT Background: Few studies have measured cognitive dysfunction in older adults during acute exacerbations of heart failure (HF), even though 25% of patients are readmitted within 30 days. The aims of this study were to examine cognitive dysfunction and acute HF symptoms in older adults hospitalized for HF and to evaluate the relationship between cognitive dysfunction and 30-day rehospitalization rates for acute HF. Methods and Results: A cross-sectional descriptive design was used to characterize cognitive function in 53 older adults hospitalized for acute HF with the use of Cogstate computerized neuropsychologic tests. Demographic characteristics, HF symptoms (dyspnea, fatigue, pain, and depressed mood), comorbidity, and 30-day readmission HF rates were also measured. Dyspnea was measured with the use of the Parshall Brief Clinical Dyspnea Rating Questionnaire while fatigue was measured with the use of the Chalder et al Brief Fatigue Scale. We measured pain with the use of the Short-Form McGill Pain Questionnaire and depressed mood with the use of the depression subscale of the Hospital Anxiety and Depression Scale. Comorbid conditions were measured with the use of the Charlson comorbidity index. With the use of linear regression, dyspnea (b 5 .281; P 5 .030), pain (b 5 .323; P 5 .011), and depressed mood (b 5 .406, P 5 .003) were associated with reduced attention and working memory speed, and pain (b 5 .372; P 5 .005) and fatigue (b 5 .275; P 5 .033) were associated with reduced accuracy of attention and working memory. Ten patients were readmitted within 30 days for HF. According to Mann-Whitney U analysis, cognitive dysfunction measures (P 5 .090e.803) failed to show differences in HF readmission. Conclusions: Participants with more and worse symptoms had decreased speed and decreased accuracy in the cognitive domains tested. Cognitive dysfunction measures did not differentiate participants who were readmitted versus those who were not readmitted within 30 days for acute HF. (J Cardiac Fail 2014;20:669e678) Key Words: Cognition, heart failure, hospital readmission, symptoms.

Approximately 25%e50% of heart failure (HF) patients experience cognitive dysfunction1 in basic (attention and memory)2 and higher-order domains (decision making and executive function).3 Although cognitive dysfunction as a

common sequela of chronic stable HF has been studied,1,4e8 few investigators have measured cognitive dysfunction sequelae during acute exacerbations of HF and its effects on 30-day readmission rates even though w25% of HF

From the 1University of Michigan School of Nursing, Ann Arbor, Michigan; 2University of Michigan, Ambulatory Psychiatry, Ann Arbor, Michigan; 3Professional Development and Education University of Michigan Health Systems, Ann Arbor, Michigan and 4University of Kentucky College of Nursing, Lexington, Kentucky. Manuscript received January 25, 2014; revised manuscript received May 16, 2014; revised manuscript accepted June 6, 2014. Reprint requests: Cynthia Arslanian-Engoren, PhD, RN, ACNS-BC, FAHA, FAAN, University of Michigan School of Nursing, 400 North

Ingalls, Room 2176, Ann Arbor, Michigan 48109. Tel: þ 734-647-0182; Fax: þ 734-935-5525. E-mail: [email protected] Funding: Donald and Karin Allen Faculty Fund, University of Michigan School of Nursing, Sigma Theta Tau International, Rho Chapter, University of Michigan, School of Nursing See page 677 for disclosure information. 1071-9164/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cardfail.2014.06.003

669

670 Journal of Cardiac Failure Vol. 20 No. 9 September 2014 patients are readmitted within 30 days of HF hospitalizations.9 The little evidence available suggests that cognitive dysfunction noted during periods of acute HF and hospitalization is associated with increased mortality for older adults10 and previously unrecognized cognitive dysfunction.11 Cognitive dysfunction is common in hospitalized elders; 35%e45% of older adults had cognitive dysfunction as measured by Mini-Mental Status Examination (MMSE),10,11 and 24% of those cognitively impaired older adults also had HF.11 In addition, 18% of patients with cognitive dysfunction had delirium based on the Confusion Assessment Method diagnostic algorithm.11 Yet only 35% of hospital staff recognize symptoms of cognitive dysfunction during acute illness of older adults, underscoring the need for better recognition, assessment, and management of cognitive dysfunction.11 The acute phase of HF, which often requires hospitalization for medical management, is associated with new or worsening symptoms of HF, including dyspnea, fatigue, depression, and pain.12 Acute HF and its associated hospital admission is a time of important health care decision making and extensive patient education for symptom management.13e15 These common symptoms (dyspnea, fatigue, depression, and pain) may impair the ability of HF patients to direct their attention, remember new information, and engage in self-care management decisions. A better understanding of these HF symptoms and their relationship to cognitive dysfunction is important not only because they may lead to poor learning and self-care performance, but also because they may add to the burden of illness during the acute phase and persist as an added disability in the long term. Indeed, 26% of HF patients are discharged to extended care facilities,16 and 21% of adults aged $65 years are discharged to long-term care institutions,17 which may represent, in part, the inability to engage in self-care activities. It is estimated that 50% of HF hospitalizations are preventable18e21 and are most commonly caused by poor self-care,22,23 the inability to maintain health and manage illness and disease using positive health practices.24 Therefore, the primary purpose of the present study was to examine cognitive dysfunction and acute HF symptoms in adults aged $65 years who were hospitalized for HF and secondarily to determine if cognitive dysfunction was associated with hospital readmission. More specifically, we aimed to characterize in-hospital cognitive dysfunction (specifically attention and memory) and common HF symptoms (dyspnea, fatigue, depression, and pain) during the acute phase of HF, evaluate the relationship among cognitive dysfunction and common HF symptoms and evaluate the relationship between cognitive dysfunction and 30-day rehospitalization rates for acute HF. Methods The University of Michigan Institutional Review Board Committee approved this study, and written informed consent was obtained from each subject. A cross-sectional descriptive design was used to

characterize cognitive dysfunction in older adults hospitalized for acute HF and to evaluate the relationship between acute HF symptoms and cognitive dysfunction. Participants were recruited from a university-affiliated medical center. Data were collected within 48 hours of admission 3 days per week (Monday, Wednesday, and Friday) between the hours of 12:00 and 16:00 to minimize effects of fatigue from morning care, diagnostic tests, or procedures on the assessment of cognitive function. We strove to test at the same time each day, within a 2-hour window, for consistency in procedures with the use of established questionnaires25e33 and Cogstate computerized neuropsychologic tests.34 The inclusion criteria included patients who were $65 years old, hospitalized for acute HF (New York Heart Association functional class IV), able to understand and read English, and had visual ability, normal hearing at a conversational tone, and the physical ability to use a computer keyboard. Excluded were acute HF patients who were prisoners, unable to provide verbal responses or describe their acute HF experience, currently being treated in the intensive care unit, or had a current medical diagnosis of psychosis, terminal cancer, dementia, or encephalopathy. Research assistants (RAs) were hired and trained in all data collectionerelated procedures (recruitment, consent, minimization of environmental distractions, cognitive testing, and extracting chart review data) and completed the University of Michigan Program for Education and Evaluation in Responsible Research and Scholarship. A joint chart review session was conducted by the lead author with each RA to demonstrate interrater reliability with the use of 10 charts and trained to 100% agreement. A random 10% of charts were reviewed to ensure project-related data collection procedures and to assess completeness of collected data. Procedures We contacted the Clinical Nurse Specialist (CNS) or senior staff nurse on the inpatient acute HF units to identify potential participants. These nurses screened patients with the use of the daily census, study inclusion/exclusion criteria, and patient interest in participation. Screeners briefly explained the study purpose and asked potential participants about their interest and willingness to be approached by the study team. Those who agreed had their names and contact information telephoned to the lead author. Before entering patient rooms, we confirmed with staff nurses that it was a good time to approach the patient. We then described the study to potential participants, verified their eligibility, and addressed all study-related questions. Once eligibility was determined, each participant completed an MMSE as a screening for cognitive function. All participants scored $21 on the MMSE and were therefore able to provide their own written informed consent to participate. After written consent was obtained, face-to-face structured interviews were conducted in patient rooms without visual or auditory distractions. To minimize distractions, a do-not-disturb sign was posted on the outside of the patient’s door indicating testing was in progress. Demographic information was obtained with the use of an instrument developed by the lead author, and other data were collected with the use of established questionnaires. Measures Cognitive function was measured with the use of neuropsychologic tests from the Cogstate34 battery in the domains of attention and memory. Cogstate measures are validated and reliable computerized neuropsychologic tests34 that correlate with

Cognitive Dysfunction in Acute HF conventional neuropsychologic measures (r 5 0.49e0.83).35 These measures were selected to reduce test fatigue and simplify test administration and for their strong test-retest reliability (0.81e0.89)35 and sensitivity to cognitive changes in older adults.36 The Cogstate Detection Test was used to detect simple reaction time, followed by the Cogstate Identification Task to measure visual attention/vigilance and then the Cogstate One Back Test to measure attention/working memory. Minimal learning effects are noted with these measures; patients can be repeatedly tested as often as needed, even multiple times within short periods of time (eg, 1 day, 1 week).37,38 Patients were encouraged to complete the task as quickly as they could; time to complete each measure is 2 minutes. For the Cogstate Detection Task (simple reaction time), participants viewed a computer screen and a playing card that appeared face down. As soon as the playing card flipped over, they were asked to press the ‘‘yes’’ key. Lower scores equal better performance. For the Cogstate Identification Task (visual attention/vigilance), participants viewed a computer screen and a playing card that appeared face down. After the playing card flipped over, participants were asked to indicate if the playing card color was red or not. If it was red, they were instructed to press the ‘‘yes’’ key. If the card was not red, they were instructed to press the ‘‘no’’ key. Lower scores equal better performance. For the Cogstate One Back Test (working memory and attention), participants viewed a computer screen where a playing card appeared face up in the center of the screen. They were asked if the presented card was identical to the one just presented. If yes, they pressed the ‘‘yes’’ key, if not the same they pressed the ‘‘no’’ key. Higher scores equal better performance. The cognitive performance of HF patients was compared with normative data of healthy older adults without HF.39 Abnormal scores were defined as O2 standard deviations above the mean of the transformed scores for similarly aged normal healthy test subjects.39 All participants completed a ‘‘practice run’’ of the Cogstate measures to familiarize themselves with the program and how to complete the measure. Measurement of Symptoms Next, acute HF symptoms were measured with the use of questionnaires with documented validity and reliability. Questionnaires consisted of Likert scales with acceptable values of validity and reliability and that were feasible to use in the clinical setting with minimal respondent burden. Dyspnea was measured with the use of the Parshall Brief Clinical Dyspnea Rating Questionnaire,25,26 which was based on the descriptor list originally developed by Simon and Schwartzstein.40e42 Participants were asked to rate their overall breathing distress with the use of a 0e10 scale (0 5 not at all bothered; 10 5 as bothered as I could possibly be), compare today’s dyspnea with yesterday’s dyspnea, and to rate the intensity of their dyspnea with the use of 7 dyspnea descriptors and to assign a numeric intensity score from 1 to 10 (1 5 just barely noticeable; 10 5 intense (severe sensation). Potential scores range from 7e70, with higher scores indicating more severe dyspnea. The 16-item questionnaire has established reliability (Cronbach a 5 0.88e0.94)43 and content validity25,43 and is sensitive to change in dyspnea.43 In the present sample, the Cronbach a was 0.895. Fatigue was measured with the use of the Chalder et al Brief Fatigue Scale.27 Participants were asked to self-rate the severity of their physical (8 questions) and mental (6 questions) fatigue along



Arslanian-Engoren et al

671

a continuum of 4 options from better than usual to much worse than usual. Higher scores indicate worse fatigue. The Chalder et al measure is a reliable (Cronbach a 5 0.89) 27and valid (sensitivity 66.6%, specificity 84.6%)27 fatigue scale instrument. In the present sample, the Cronbach a was 0.890. Pain was measured with the use of the Short-Form McGill Pain Questionnaire (SF-MPQ),28,29 a reliable (test-retest 0.88-0.95)44 and validated measure that is sensitive to change in pain29 and measures pain intensity with the use of sensory/affective descriptors. Pain intensity is rated from none to worst possible pain. A visual analog scale measures current pain, and an ordinal scale is used to measure overall intensity of the total pain experience. The sensory, affective, and total scores of the SF-MPQ are significantly correlated with the McGill Pain Questionnaire.28,29 The SF-MPQ can be self-administered or interviewer administered and takes 2e5 minutes to complete.28 The SF-MPQ has been used in acute and chronic conditions and is sensitive to clinical changes in pain.29 The Hospital Anxiety and Depression Scale32,33 (depression subscale only) was used to measure depressive symptoms of hospitalized older adult acute HF patients. Cronbach a for the depression subscale is 0.67e0.9045 with sensitivity and specifity of 0.80.45 A score of O11 on the depression subscale is indicative of depressed mood. In the present sample, the Cronbach a was 0.758. Comorbid conditions were measured with the use of the Charlson comorbidity index,30,31 a validated and reliable measure with established interrater reliability46 that predicts hospitalization and 1-year mortality.30,31 The Charlson comorbidity index assigns weights from 1 to 6 for presence of specific comorbidities that estimate relative risk of death. Each decade of life over age 40 years adds to the total risk score. Total comorbidity category scores are 1e5; higher scores are associated with higher estimated relative risk of death.30,31 To standardize data collection, minimize missing data, and reduce patient fatigue, all questionnaires were loaded on the computer and read to participants. Participants were also allowed to read the questionnaires without assistance from the study team, if preferred. The administration of all of the measures took w30 minutes to complete. Measures could be completed in 1 session or in divided sessions with rest periods if the patient became too fatigued to continue. After completing the cognitive dysfunction and symptoms measures, participants’ medical records were reviewed for comorbid data and to obtain clinical variables (laboratory values, diagnostic studies, and pain medication history). For participants who scored O11 on the depression scale of the Hospital Anxiety and Depression Scale32,33 or who showed $2 standard deviations longer than expected on a normative population of the Cogstate tests, a standardized letter was sent to the primary physician at the completion of data collection to inform him/ her of this finding. The attending physician on the inpatient HF unit was also notified via pager/e-mail of the high depressed mood score and/or impairment score. Statistical Analyses Descriptive statistics were performed to characterize the sample. Spearman rank correlation coefficients were conducted to analyze the correlations between HF symptoms and cognitive dysfunction. Per Cogstate recommendations to linearize the data, we used the base 10 logarithm transformation for the

672 Journal of Cardiac Failure Vol. 20 No. 9 September 2014 Cogstate Identification and Detection tests and the arcsine of the square root of the proportion of correct responses for the Cogstate One Back Test. Both the accuracies of the test and the log-transformed speeds in milliseconds were entered as continuous variables into the regression models. Backward multiple linear regression models, adjusting for age, education, and comorbidities were used to determine the independent associations of pain, dyspnea, fatigue, and depressed mood with the 6 measures of cognition. In creating the models, we used Akaike information criteria to select the variables in each model that produced the minimum Akaike value when variables were sequentially removed.47 Akaike information criteria have the advantage of permitting comparison of models with different error distributions and avoid the necessity to adjust P values for multiple comparisons.47 Variance inflation factors were used to assess multicollinearity in the regression coefficients. Values of !2.0 were considered to be acceptable. To determine the factors associated with readmission, first Mann-Whitney U analyses were performed to compare nonparametric continuous data, including cognitive scores, and Fisher exact or chi-square analyses were used to compare categoric data. Next, all factors that were univariably associated with readmission with P ! .20 were entered into a binary logistic regression using forward selection and confirmed by backwards selection. A P value of !.05 and an odds ratio that excluded 1 were considered to be statistically significant. All analyses were performed with the use of SPSS version 21 (IBM, Armonk, New York).

with 1 participant requesting a rest period of !1 minute during questionnaire completion. Characterization of In-Hospital Cognitive Dysfunction

Speed and accuracy are the primary Cogstate measures. Increased completion time speeds indicate that participants took longer to complete the test compared with similarly aged normal healthy adults. Abnormally slow completion times were noted in all 3 cognitive measures: simple reaction time (n 5 10; 19%); visual attention/vigilance times (n 5 15; 28%); and working memory and attention (n 5 11; 21%). Response accuracy on the 3 cognitive measures ranged from 17% to 100%. Approximately one-third of participants (n 5 18, 35%) scored 100% accuracy on the simple reaction time measure, whereas only 6 participants (11%) had 100% accuracy on the visual attention/vigilance measure. None of the participants had 100% accuracy on the working memory and attention measure (Table 2). Among those participants with 100% accuracy on the simple reaction measure, nearly one-fourth (n 5 4; 22%) had abnormally slow simple reaction scores. Similarly, 3 of the 6 participants with 100% accuracy scores on the visual attention/vigilance measure had abnormally slow visual attention/vigilance scores. Characterization of In-Hospital Common Heart Failure Symptoms

Power Analysis To achieve 90% power to find an association between HF symptoms and cognitive dysfunction by linear regression, with 2-tailed a of 0.05, assuming a variance inflation factor of 1.1, beta of 0.33, standard deviation of 15, and error standard deviation of 10 for 12 predictors, 50 participants were needed to participate in the study.

Results A total of 115 acute HF patients were screened for participation, of which 62 potential participants were excluded because they were deemed to be ineligible or because they refused to participate in the study. Reasons for ineligibility or refusal are presented in Fig. 1. Fiftythree older adult HF patients participated in this study. Participants were 72 6 5 (mean 6 SD) years old and primarily male (66%), white (93%), and right-handed (87%). Handedness was collected as part of the baseline measures required by Cogstate. MMSE scores of participants ranged from 23 to 30 (median 29; interquartile range [IQR] 28e29). Nearly one-fifth (19%) of the participants had Charlson comorbidity index total scores of $5 (Table 1). All but 1 participant completed all study questionnaires. Owing to a 1-time computer malfunction 1 participant did not complete the visual attention/vigilance measure, and 2 participants opted to complete the questionnaires without having them read to them by a member of the research team. All study questionnaires were completed in 1 session,

Dyspnea. Participants were nearly equally divided in their description of how bothered they were by their breathing today. Approximately one-third (n 5 17; 32%) were ‘‘not bothered at all’’ by it. When asked to describe how their breathing feels today compared to the way it usually feels, more than one-half (n 5 28; 52%) of respondents described it as ‘‘about the same as usual.’’ Only 2 participants (4%) indicated that they were ‘‘as bothered as they could be’’ by it. The median score for this measure of dyspnea was 3.0 (IQR 0e5). Participants most commonly described their breathing as ‘‘required effort’’ with a median intensity of 4 (IQR 3e5.75) (Table 3). Fatigue. Total fatigue scores ranged from 14 to 41 (median 30), with the physical symptoms subscale scores ranging from 8 to 28 (median 17) and the mental symptoms subscale scores ranging from 6 to 17 (median 12). Approximately 40% of participants indicated feeling weak that was ‘‘worse’’ or ‘‘much worse’’ than usual, and 30% rated their tiredness as ‘‘worse’’ or ‘‘much worse than usual.’’ Regarding mental fatigue, O1 in 5 participants (23%) rated their ability to find the correct word as ‘‘worse than usual,’’ and almost 10% rated their memory as ‘‘worse than usual’’ (Table 4). Depressed Mood. Depressed mood scores ranged from 0 to 15 (median 4). Six (11.3%) participants had scores O11 on the HAD, which were indicative of depressed mood. Pain. Approximately one-third of participants reported having pain (n 5 13; 34%). When present, it was most often reported as moderate in intensity and described as: sharp (n 5 7; 13%); gnawing (n 5 6;

Cognitive Dysfunction in Acute HF



Arslanian-Engoren et al

673

Fig. 1. Inclusion and exclusion strategies. MRSA, methicillin-resistant Staphylococcus aureus.

11%); tender (n 5 6; 11%); aching (n 5 5; 9%); tiringexhausting (n 5 5; 9%); and punishing-cruel (n 5 3; 6%). Only 1 participant reported pain as excruciating.

Table 1. Participants Characteristics (n 5 53), n (%) Participant Characteristics Age, y, mean (SD; range) Sex (male) Race Black or African American White Handedness (right) Highest education High school and less College degree (Associate, Bachelor, Master) Professional Degree (MD, DDS, DO) Other Charlson comorbidity category scores 1 3 5

72.16 (5.33; 65e90) 35 (66) 3 (5.7) 50 (94.3) 46 (86.8) 36 (67.9) 15 (28.3) 1 (1.9) 1 (1.9) 19 (35.8) 24 (45.3) 10 (18.9)

Nearly one-half (n 5 25; 48%) of participants reported pain (range 0e71). Relationship Between Cognitive Dysfunction and Heart Failure Symptoms

Participants with higher fatigue scores had more cognitive dysfunction with slower speeds and less accuracy (Table 5). Patients with higher visual analog pain scores had slower times and reduced accuracy in the simple reaction times and working memory and attention scores. Participants with higher depressed mood scores had abnormally slow simple reaction times and working memory times, and had reduced accuracy in completing the simple reaction time measure. However, dyspnea was not correlated with cognitive dysfunction. Using multiple linear regressions to determine the independent association of dyspnea, fatigue, pain, or depressed mood with cognitive dysfunction, adjusting for the demographic characteristics and comorbidities listed in Table 1, we found depressed mood, pain, and fatigue to be associated with worse cognitive function. Specifically,

674 Journal of Cardiac Failure Vol. 20 No. 9 September 2014 Table 2. Characterization of In-Hospital Cognitive Dysfunction Simple Reaction Time (Cogstate Detection) (n 5 52)

Visual Attention/Vigilance (Cogstate Identification) (n 5 52)

Working Memory and Attention (Cogstate One Back) (n 5 53)

262e1,336 ms 390 2.41e3.12 2.59 562 ms

434e1,272 ms 603 2.64e3.10 2.78 692 ms

533e1,979 ms 972 2.72e3.29 2.98 1,318 ms

44.6%e100% 94.7% 0.732e1.57 1.34

45.50%e100% 90.9% 0.740e1.57 1.26

17.10%e96.9% 78.0% 0.42e1.32 1.08

Speed Raw score Median Standardized score Median Cutoff score Accuracy Raw score Median Standardized score Median

participants with higher depressed mood scores had abnormally slow simple reaction and attention/working memory speeds. Those with higher pain scores had slower visual attention/vigilance and attention/working memory speeds. They were also less accurate in their simple reaction time measures and in their attention/working memory measures. Fatigued participants had normal speeds on each cognitive measure, but reduced accuracy on all 3 measures, whereas dyspneic participants had abnormally slow attention/working memory speeds (Table 6). Cognitive Dysfunction and Rehospitalization for Heart Failure

A total of 12 participants were readmitted within 30 days of hospital discharge. Of these, 10 (19%) were readmitted for HF, of whom 9 (90%) had cognitive dysfunction according to abnormal Cogstate test scores on their initial hospitalization. Cognitive dysfunction was not associated with readmission. HF patients who were younger or who had at least some college education were more likely to be readmitted (Table 7). After adjusting for factors, only having some college remained associated with readmission for HF (odds ratio 7.70, 95% confidence interval 1.673e35.43; P 5 .009). Discussion We found that older adults with more and worse HF symptoms had cognitive dysfunction in attention and Table 3. Dyspnea Descriptors and Intensity of Sensation (n 5 53) Descriptor My breathing required effort My breathing required work I felt hunger for air Chest felt constricted Chest felt tight I felt smothering I felt that I was suffocating IQR, interquartile range.

n (%) 16 13 10 8 7 5 4

(30%) (25%) (19%) (15%) (13%) (9%) (8%)

Intensity of Sensation, Median (IQR) 4.0 3.0 6.0 5.0 5.0 7.5 6.5

(3e5.75) (3e6) (3.75e7) (3.25e5.75) (4e5) (7e9) (3.5e8)

working memory. Participants with more and worse symptoms had decreased speed and decreased accuracy in the cognitive domains tested. Dyspnea, pain, and depressed mood were associated with reduced attention and working memory speed, and pain and fatigue were associated with reduced accuracy of attention and working memory. We did not find cognitive dysfunction to be different in participants who were readmitted for acute HF within 30 days of hospital discharge than in participants who were not readmitted within 30 days for acute HF. Instead, our data revealed that younger patients and patients who were better educated were more likely to be readmitted for HF. Our results are similar to 2 recent studies48,49 that assessed cognitive function in older adults hospitalized for acute HF with the use of measures of cognitive impairment, defined as the cognitive state between normal cognitive aging and dementia50 and pseudodementia.51 Although the measures differed in the 2 studies (Dodson et al49 used the Folstein Mini-Mental State measure,51 although Hajduk et al48 used the Montreal Cognitive Assessment Tool), both Dodson et al49 and Hajduk et al48 reported that cognitive impairment was a common occurrence (47%49 to 79%48) in older adults hospitalized for acute HF.49 Hajduk et al48 found that 79% of older adults (mean age 6 SD, 71 6 12.8 years) had cognitive impairment in $1 domain (memory, processing speed, executive function), and Dodson et al.49 noted that almost one-half (47%) of older adults (mean age 6 SD, 80 6 8 years) hospitalized for acute HF had cognitive impairment. Yet, despite its common occurrence, only 23% of acute HF patients with cognitive impairment had it documented by hospital staff physicians at hospital discharge.49 This lack of documentation was associated with increased likelihood of 6-month hospital readmission or death49 and may reflect poor recognition of the potentially reversible causes of cognitive dysfunction by health care providers that may have contributed to the poor patient outcomes. Cognitive dysfunction is associated with symptoms common to HF (dyspnea, depression, fatigue, pain). Dyspnea is a complex interaction of signals occurring within the central nervous system52 and various areas of the brain, including the anterior insula and the amygdala.53,54

Cognitive Dysfunction in Acute HF



Arslanian-Engoren et al

675

Table 4. Fatigue Scores (n 5 53), n (%) Better Than Usual Physical fatigue subscale Tiredness Needing to rest more Feeling drowsy or sleepy Having problems starting things Starting things without difficulty, but getting weak as they go on Lacking energy Less strength in muscles Feel weak Mental fatigue subscale Having difficulty concentrating Problems with thinking clearly Making slips of the tongue Finding the correct word Evaluating their memory Evaluating if they lost interest in things they used to do

10 8 9 6 4

No More Than Usual

(18.9) (15.1) (17.0) (11.3) (7.5)

27 27 34 33 32

(50.9) (50.9) (64.2) (62.3) (60.4)

Worse Than Usual 14 17 7 13 16

Much Worse Than Usual

(26.4) (32.1) (13.2) (24.5) (30.2)

2 1 3 1 1

(3.8) (1.9) (5.7) (1.9) (1.9)

9 (17.0) 7 (13.2) 5 (9.4)

24 (45.3) 20 (37.7) 27 (50.9)

15 (28.3) 20 (37.7) 16 (30.2)

5 (9.4) 6 (11.3) 5 (9.4)

5 5 4 4 9 8

41 42 43 37 39 29

7 6 6 12 5 15

0 0 0 0 0 1

(9.4) (9.4) (7.5) (7.5) (17) (15.1)

Although no studies were found that specifically examined dyspnea and cognitive dysfunction in HF, studies in patients with other chronic conditions (chronic obstructive pulmonary diseases55 and emphysema56) show that dyspnea, cognitive function (verbal memory, sequential skills, visual memory, and visuomotor sequential skills), and depression improved after surgical intervention (lung volume reduction surgery)55 and medical treatment.57 Acute58 and chronic56 pain is also associated with deficits in working memory and an inverse relationship has been reported between pain and executive function.59 Participants in our study with higher pain scores had slower attention/working memory speeds and slower simple reaction speeds. Fatigue, the 2nd most common symptom reported by HF patients,60,61 is also common in other conditions with coexisting cognitive deficits (eg, chronic fatigue syndromes,62,63 chronic hepatitis C infection,64 and postoperative states65). Fatigued patients with chronic fatigue syndrome required activation of more bilateral extensive regions of the brain than healthy control subjects, and perceptions of fatigue correlated with the extensiveness of brain activation.63

(77.4) (79.2) (81.1) (69.8) (73.6) (54.7)

(13.2) (11.3) (11.3) (22.6) (9.4) (28.3)

(0) (0) (0) (0) (0) (1.9)

This raises the possibility that if HF, with decreased cerebral perfusion (low cardiac index) or decreased oxygen delivery (low cardiac delivery, anemia, not fully saturated hemoglobin) is superimposed on fatigue, there may not be enough oxygen available for this increased brain processing, leading to worse cognition in HF patients with fatigue compared with those without fatigue. This is particularly salient because the symptoms of fatigue reduce the capacity to attend and learn,66 including acquiring new information and engaging in self-monitoring behaviors.67 Because directed attention is severely limited in acute illnesses,67 it may impair learning and memory in HF patients needed for self-care decisions. The influence of acute illnesses on cognitive function in 460 adults age $70 years admitted for acute illnesses, including HF (n 5 54), showed that despite being elderly and frail and having at least a moderate degree of illness severity and comorbidities, 39% of older adults demonstrated recoverable cognitive function by hospital discharge, as measured by the MMSE.68 Similarly, Zuccala et al8 reported improved cognitive performance at hospital discharge for 1,511 hospitalized HF patients when

Table 5. Spearman Rho Correlations: Relationship Between Cognitive Dysfunction and Heart Failure Symptoms

Cognitive Measures Heart failure symptoms Dyspnea Rating Fatigue Chalder physical subscale Chalder mental subscale Chalder total Depressed mood Pain McGill PPI PPI, present pain intensity. *P # .05; **P # .01.

Simple Reaction Time (n 5 53)

Visual Attention/Vigilance (n 5 52)

Speed

Accuracy

Speed

Accuracy

0.085

0.004

0.184

0.002

Working Memory and Attention (n 5 53) Speed

0.069

Accuracy

0.260

0.258 0.203 0.268 0.328*

0.364** 0.364** 0.402** 0.417*

0.093 0.120 0.118 0.233

0.130 0.246 0.157 0.233

0.255 0.309* 0.324* 0.495**

0.366** 0.370** 0.364** 0.253

0.268

0.423**

0.231

0.035

0.321*

0.357**

676 Journal of Cardiac Failure Vol. 20 No. 9 September 2014 Table 6. Multiple Linear Regression of Heart Failure Symptoms and Cognitive Dysfunction Predictor HADS (Constant) McGill pain VAS Chalder fatigue total (Constant) McGill pain VAS (Constant) Chalder fatigue mental (Constant) HADS Dyspnea rating McGill pain VAS (Constant) McGill pain VAS Chalder fatigue total (Constant)

Outcome: Cognitive Dysfunction Test

b

Simple reaction time speed

0.239 0.404 0.238

Simple reaction time accuracy Visual attention/vigilance speed

0.269

Visual attention/vigilance accuracy

0.239

Attention/working memory speed

0.406 0.281 0.323

Attention/working memory accuracy

0.372 0.275

b

SE

0.010 20.568 0.004 0.009 1.658 0.001 2.788 0.020 1.512 0.014 0.013 0.002 2.932 0.004 0.010 1.391

0.006 .037 0.001 0.005 0.137 0.001 0.016 0.012 0.147 0.004 0.006 0.001 0.028 0.001 0.005 0.143

t

P

VIF

1.759

0.084

1.0

3.227 10.899

0.002 .063

1.079 1.079

1.978

0.054

1.000

1.742

0.088

1.000

3.161 -2.232 2.634

0.003 0.030 0.011

1.171 1.128 1.070

2.965 2.186

0.005 0.033

1.079 1.079

HADS, Hospital Anxiety and Depression Scale (depressed mood); SE, standard error; VAS, visual analog scale; VIF, variance inflation factor.

potentially treatable comorbid conditions of abnormal glucose, potassium, and hemoglobin levels were restored during hospitalization.

Given the negative consequences associated with cognitive dysfunction in older adults with HF, routine screening by health care providers is needed to identify at-risk patients

Table 7. Characteristics of Participants Readmitted (n 5 10) Versus Not Readmitted (n 5 43) Within 30 Days for HF Characteristics Demographics Age, y Race African American/Black White Sex Male Female Highest education High school or less At least some college Heart failure symptoms Dyspnea rating 0e3 4e6 7e10 Fatigue Chalder physical Chalder mental Chalder total Pain McGill VAS HADS Charlson comorbidity total score Charlson comorbidity category 1 3 5 Cognitive measures* Simple reaction time Speed Accuracy Visual attention/vigilance Speed Accuracy Working memory and attention Speed Accuracy

Readmitted 67 (66.75e72.25)

Not Readmitted 72 (68e76)

P Value .039 (MW)

0 10

3 40

1.000 (Fisher)

8 2

27 16

.464 (Fisher)

3 7

33 11

.011 (MW)

4 4 2

27 14 2

.184 (Prs c2)

19 (15.5e22.25) 12 (11.75e14) 31 (27.5e36.25) 0 (0e10) 5.5 (2e8.25) 3 (2.5e4.25)

17 (16e21) 12 (12e13) 29 (28e34)

.493 (MW) .757 (MW) .741 (MW)

0 (0e16) 4 (2e7) 3 (2e4)

.548 (MW) .599 (MW) .926 (MW)

2 6 2

17 18 8

.482 (Prs c2)

2.66 (2.47e2.80) 1.34 (0.912e1.36)

2.58 (2.49e2.69) 1.41 (1.29e1.57)

.481 (MW) .090 (MW)

2.78 (2.73e2.95) 1.18 (1.12e1.39)

2.78 (2.73e2.87) 1.26 (1.16-1.39)

.536 (MW) .460 (MW)

2.95 (2.89e3.05) 1.13 (0.925e1.23)

2.99 (2.90e3.08) 1.05 (0.854e1.23)

.803 (MW) .585 (MW)

Values are presented as n or median (interquartile range). MW, Mann-Whitney U; Prs, Pearson’s chi-square; other abbreviations as in Table 6. *Transformed scores.

Cognitive Dysfunction in Acute HF

and to manage potentially reversible causes of cognitive dysfunction. Patient-centered interventions (eg, fatigue, dyspnea, and pain management, social engagement) are needed to optimize cognitive function, specifically attention and memory, during the acute HF phase. Optimizing cognitive function may enhance self-care activities and improve health-related quality of life for older adults with acute HF. This study has several limitations. We collected data from a single site, our sample was homogeneous regarding race, which may decrease the generalizability of our findings, and we did not collect complete medication information. Finally, we had a relatively small sample size. Although we were unable to find differences in cognitive dysfunction between participants readmitted and not readmitted within 30 days for acute HF, our study was not powered for this, but was powered to detect the relationship between cognitive dysfunction and HF symptoms. Our readmission results will provide data for power analysis for future studies evaluating cognitive dysfunction and readmission. The present study has several strengths. It is one of only a few to characterize in-hospital cognitive dysfunction among older HF patients and to evaluate the relationship between cognitive dysfunction and common HF symptoms. In conclusion, we found that older adults with more and worse HF symptoms had cognitive dysfunction in attention and working memory. Future studies are needed to determine if addressing these areas of cognitive dysfunction while hospitalized will improve self-care behaviors, reduce the burden of illness, and improve health-related quality of life.

Disclosures None.

References 1. Pressler SJ. Cognitive functioning and chronic heart failure. J Cardiovasc Nurs 2008;23:239e49. 2. Vogels RL, Scheltens P, Schroeder-Tanka JM, Weinstein HC. Cognitive impairment in heart failure: a systematic review of the literature. Eur J Heart Fail 2007;9:440e9. 3. Pressler SJ, Kim J, Riley P, Ronis DL, Gradus-Pizlo I. Memory dysfunction, pscyhomotor slowing and decreased executive function predict mortality in patients with heart failure and low ejection fraction. J Card Fail 2010;16:750e60. 4. Zuccala G, Onder G, Marzetti E, Monaco MR, Cesari M, Cocchi A, et al, GIFA Study Group. Use of angiotensin-converting enzyme inhibitors and variations in cognitive performance among patients with heart failure. Eur J Heart Fail 2005;26:226e33. 5. Antonelli Incalzi R, Trojano L, Acanfora D, Crisci C, Tarantino F, Abete P, et al, CHF Italian Study Investigators. Verbal memory impairment in congestive heart failure. J Clin Exp Neuropsychol 2003;25:14e23. 6. Riegel B, Bennett JA, Davis A, Carlson B, Montague J, Robin H, et al. Cognitive impairment in heart failure: issues of measurement and etiology. Am J Crit Care 2002;11:520e8.



Arslanian-Engoren et al

677

7. Vogels RL, Oosterman JM, van Harten B, Scheltens P, van der Flier WM, Schroeder-Tanka JM, et al. Profile of cognitive impairment in chronic heart failure. J Am Geriatr Soc 2007;55:1764e70. 8. Zuccala G, Marzetti E, Cesari M, lo Monaco MR, Antonica L, Cocchi A, et al. Correlates of cognitive impairment among patients with heart failure: Results of a multicenter survey. Am J Med 2005;118:496e502. 9. Dharmarajan K, Hsich AF, Lin Z, Bueno H, Ross JS, Horwitz LI, et al. Diagnoses and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA 2013;309:355e63. 10. Rodrıguez-Pascual C, Vilches-Moraga A, Paredes-Galan E, FerreroMarinez AI, Torrente-Carballido M, Rodrıguez-Artalejo F. Comprehensive geriatric assessment and hospital mortality among older adults with decompensated heart failure. Am Heart J 2012;164:756e62. 11. Naylor MD, Stephen C, Bowles KH, Bixby MB. Cognitively impaired older adults: From hospital to home. AJN 2005;105:52e61. 12. Allen LA, O’Connor CM. Management of acute decompensated heart failure. Can Med Assoc J 2007;176:797e805. 13. Flacker J, Park W, Sims A. Hospital discharge information and older patients: Do they get what they need? J Hosp Med 2007;2:291e6. 14. Naylor MD, Brooten D, Campbell R, Jacobsen BS, Mezey MD, Pauly MV, et al. Comprehensive discharge planning and home follow-up of hospitalized elders. JAMA 1999;281:613e20. 15. Str}omberg A. The crucial role of patient education in heart failure. Eur J Heart Fail 2005;7:363e9. 16. Fang J, Mensah GA, Croft JB, Keenan NL. Heart failureerelated hospitalization in the U.S. 1979 to 2004. JACC 2008;52:428e34. 17. Hall MJ, Levant S, DeFrances CJ. Hospitalization for congestive heart failure: United States, NCHS Data Brief 2000e2010. 2012;108:1e8. 18. Phelan D, Smyth L, Ryder M, Murphy N, O’Loughlin C, Conion C, et al. Can we reduce preventable heart failure eradmission in patients enrolled in a disease management programme? Ir J Med Sci 2009;178:167e74. 19. Braunstein JB, Anderson GF, Gerstenblith G, Weller W, Niefeld M, Herbert R, et al. Noncardiac comorbidity increase preventable hospitalizations and mortality among medicare beneficiaries with chronic heart failure. JACC 2003;42:1226e33. 20. Krumholtz HM, Chen Y-T, Wang Y, Vaccarino V, Radford MJ, Horwitz RI. Predictors of readmisson among elderly survivors of admission with heart failure. Am Heart J 2000;139:72e7. 21. Michalsen A, K€onig G, Thimme W. Preventable causative factors leading to hospital admission with decompensated heart failure. Heart 1998;80:437e41. 22. Moser DK, Doering LV, Chung ML. Vulnerabilities of patients recovering from an exacerbation of chronic heart failure. Am Heart J 2005; 150:984. 23. Formiga F, Chivite D, Manito N, Casas S, Llopis F, Pujol R. Hospitalization due to acute heart failure: role of participating factors. Int J Cardiol 2007;120:237e41. 24. Moser DK, Dickson V, Jaarsma T, Lee C, Stromberg A, Riegel B. Role of self-care in the patient with heart failure. Curr Cardiol Rep 2012;14:265e75. 25. Parshall MB. Psychometric characteristics of dyspnea descriptor ratings in emergency department patients with exacerbated chronic obstructive pulmonary disease. Res Nurs Health 2002;25:331e44. 26. Parshall MB, Welsh JD, Brockopp DY, Heiser RM, Schooler MP, Cassidy KB. Reliability and validity of dyspnea sensory quality indicators in herat failure patients treated in an emergency department. Heart Lung 2001;30:57e65. 27. Chalder T, Berelowitz G, Pawlikowska T, Watts L, Wessely S, Wright D, et al. Development of a fatigue scale. J Pscyhosom Res 1993;37:147e53. 28. Melzack R. The Short-Form McGill Pain Questionnaire. Pain 1987; 30:191e7. 29. Katz J, Melzack R. Measurement of pain. Surg Clin North Am 1999; 79:231e52. 30. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol 1994;47:1245e51.

678 Journal of Cardiac Failure Vol. 20 No. 9 September 2014 31. Charlson ME, Pompei P, Alex KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987;40:373e83. 32. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361e70. 33. Mykletun A, Stordal E, Dahl AA. The Hospital Anxiety and Depression (HAD) Scale: factor structure, item analyses and internal consistency in a large population. Br J Psychiatry 2001;179:540e4. 34. Cogstate clinical trials 2010. Available at: http://www.cogstate.com/. Accessed October 22, 2010. 35. Maruff P, Thomas E, Cystique L, Brew B, Collie A, Snyder P, et al. Validity of the Cogstate Brief Battery: relationship to standardized tests and sensitivity to cognitive impairment in mild traumatic brain injury, schizophrenia and AIDS dementia complex. Arch Clin Neuropsychol 2009;24:165e78. 36. Cysique LAJ, Maruff P, Darby D, Brew BJ. The assessment of cognitive function in advanced HIV-1 infection and AIDS dementia complex using a computerised cognitive test battery. Arch Clin Neuropsychol 2006;21:185e94. 37. Darby D, Maruff P, Collie A, McStephen M. Mild cognitive impairment can be detected by multiple assessments in a single day. Neurology 2002;59:1042e6. 38. Falleti MG, Maruff P, Collie A, Darby DG, McStephen M. Qualitative similarities in cognitive impairment associated with 24h of wakefulness and a blood alcohol concentration of 0.05%. J Sleep Res 2003;12: 265e74. 39. Fredrickson J, Maruff P, Woodward M, Moore L, Fredrickson A, Sach J, et al. Evaluation of the usability of a brief computerized cognitive screening test in older people for epidemiological studies. Neuroepidemiology 2010;34:65e75. 40. Mahler DA, Harver A, Lentine T, Scott JA, Beck K, Schwartzstein RM. Descriptors of breathlessness in cardiorespiratory diseases. Am J Respir Crit Care Med 1996;154:1357e63. 41. Simon PM, Schwartzstein RM, Weiss JW, Fencl V, Teghtsoonian M, Weinberger SE. Distinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis 1990;142:1009e14. 42. Simon PM, Schwartzstein RM, Weiss JW, Lahive K, Fencl V, Teghtsoonian M, Weinberger SE. Distinguishable sensations of breathlessness induced in normal volunteers. Am Rev Respir Dis 1989;140:1021e7. 43. Parshall MB. Brief Clinical Dyspnea Rating. National Institute of Nursing Research: Testing a Measurement Model of Acute Dyspnea (NIN-NINR: 1 R15 NR008883-01A1). Project dates April 14, 2005, to March 31, 2008. 44. Grafton KV, Foster NE, Wright CC. Test-retest reliability of the Short-Form McGill Pain Questionnaire: assessment of intraclass correlation coefficients and limits of agreement in patients with osteoarthritis. Clin J Pain 2005;21:73e82. 45. Bjelland I, Dahl AA, Tangen Haug T, Neckelman D. The validity of the hospital anxiety and depression scale an updated literature review. J Psychosomatic Res 2002;52:69e77. 46. Hall SF, Groome PA, Streiner DL, Rochon PA. Interrater reliability of measurements of comorbid illness should be reported. J Clin Epidemiol 2006;59:926e33. 47. Brockwell PJ, Davis RA. Time series: theory and methods. 2nd ed. New York: Springer; 1999. 48. Hajduk AM, Lemon SC, McManus DD, Lessrd DM, Gurwitz JH, Spencer FA, et al. Cognitive impairment and self-care in heart failure. Clin Epidemiol 2013;5:407e16. 49. Dodson JA, Truong TT, Towle VR, Kerins G, Chaudhry SI. Cognitive impairment in older adults with heart failure: prevalence,

50.

51.

52. 53.

54.

55.

56.

57.

58. 59.

60.

61.

62.

63.

64.

65.

66. 67. 68.

documentation and impact on outcomes. Am J Med 2013;126: 120e6. Nasreddine ZS, Phillips NA, Bedirian V, Charbonneau S, Whitehead V, Collin S, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005;53:695e9. Folstein MF, Folstein SE, McHugh PR. ‘‘Mini-Mental State’’ a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189e98. Manning HL, Schartzstein RM. Pathophysiology of dyspnea. N Engl J Med 1995;333:1547e53. Mountcastle VB, Darian-Smith I, American Physiological Society. The nervous system: handbook of physiology. Ed. Brookhart, JM, 1981. von Leupoldt A, Sommer T, Kegat S, Baumann HJ, Klose H, Dahme D, B€uchel C. The unpleasantness of perceived dyspnea is processed in the anterior insula and amygdala. Am J Respir Crit Care Med 2008;177:1026e32. Kozora E, Emery CF, Ellison MC, Wamboldt FS, Diaz PT, Make B. Improved neurobehavioral functioning in emphysema patients following lung volume reduction surgery compared with medical therapy. Chest 2005;128:2653e63. Berryman C, Stanton TR, Bowering KJ, Tabor A, McFarlane A, Moseley GL. Evidence for working memory deficits in chronic pain: a systematic review and meta-analysis. Pain 2013;154:1181e96. Kozora E, Emery CF, Zhang L, Make B. Improved neurobehavioral functioning in emphysema patients following medical therapy. J Cardiopulm Rehabil Prev 2010;30:251e9. Hood A, Pulvers K, Spady TJ. Timing and gender determine if acute pain impairs working memory performance. J Pain 2013;14:1320e9. Abeare CA, Cohen JL, Axelrod BN, Leisen JC, Mosley-Williams A, Lumley MA. Pain, executive functioning and affect in patients with rheumatoid arthritis. Clin J Pain 2010;26:683e9. Falk K, Granger BB, Sweberg K, Ikman I. Breaking the vicious circle of fatigue in patients with chronic heart failure. Qual Health Res 2007;17:1020e7. Falk K, Patel H, Swedberg K, Ikman I. Fatigue in patients with chronic heart failure  a burden associated with emotional and symptom distress. Eur J Cardiovasc Nurs 2009;8:91e6. Cook DB, O’Connor PJ, Lange G, Steffener J. Functional neuroimaging correlates of mental fatigue induced by cognition among chronic fatigue syndrome patients and controls. NeuroImage 2007;36:108e22. Lange G, Stennener J, Cook DB, Bly BM, Christodoulou C, Liu WC, et al. Objective evidence of cognitive complaints in chronic fatigue syndrome: a bold fMRI study of verbal working memory. Neuroimage 2005;26:513e24. Kramer L, Hofer H, Bauer E, Funk G, Formann E, Steindl-Munda P, et al. Relative impact of fatigue and subclinical cognitive brain dysfunction on health-related quality of life in chronic hepatitis C infection. AIDS 2005;19(Suppl 3):S85e92. Wan Y, Xu J, Ma D, Zeng Y, Cibelli M, Maze M. Postoperative impairment of cognitive function in rats: a possible role for cytokine-mediated inflammation in the hippocampus. Anesthesiology 2007;106:436e43. Kaplan S, Kaplan R. Cognition and environment: functioning in an uncertain world. New York: Praeger; 1982. Cimprich B. A theoretical perspective on attention and patient education. Adv Nurs Sci 1992;14:39e51. Inouye SK, Zhang YA, Han L, Leo-Summers L, Jones R, Marcantonio E. Recoverable cognitive dysfunction at hospital discharge in older persons during acute illness. J Gen Intern Med 2006;21:1276e81.

Cognitive dysfunction in older adults hospitalized for acute heart failure.

Few studies have measured cognitive dysfunction in older adults during acute exacerbations of heart failure (HF), even though 25% of patients are read...
470KB Sizes 0 Downloads 3 Views