Heart failure (HF) affects 1% to 2% of the adult population in developed countries.1 The incidence of HF increases with age, with more than 80% of patients being older than 65 years.2 Among these patients, HF is a leading cause of hospitalization, early readmissions, and mortality.3,4
Cognitive impairment is highly prevalent among the HF population, where the risk of cognitive impairment is twice as high among patients with HF than in age-matched adults without HF.5,6 The cognitive domains adversely affected in HF include working memory, learning memory and delay recall, attention, processing speed, and executive function.7 Cognitive impairment has been found to interfere with self-care abilities pivotal to HF management.8,9 Cognitively impaired patients with HF were less likely to seek help to cope with changes in HF symptoms and adhere to medication regimen,10,11 leading to poor clinical outcomes. Cognitive impairment has been reported to predict mortality, above and beyond HF alone.12
Despite its complications in the treatment of HF, guidance on the management of cognitive impairment remains sparse. A recent systematic review that investigated change in cognitive outcomes in HF yielded studies of standard surgical and pharmacological HF therapy, including heart transplantation, left ventricular assist device implantation, and medications.13 Several of these interventions were hypothesized to be effective based on postulated mechanisms responsible for the pathogenesis of HF and cognitive decline: (1) via the mediation of cerebral hypoperfusion and impaired autoregulation of cerebral perfusion pressures, (2) cerebral emboli, leading to cardiac insufficiency, and (3) shared myocardial and cerebral pathology with Alzheimer's disease.6,14,15 Although improvements in cognition were reported, none of the standard HF therapies found significant correlations between cardiac and cognitive parameters, suggesting that cognition may not be directly targeted via these interventions and the posited mechanistic links between HF and cognitive impairment remain unclear.
Recently, research on a nonpharmacological intervention—computerized cognitive training (CCT)—has generated a strong evidence base to support its effectiveness in enhancing cognition or ameliorating the risk of cognitive decline. Computerized cognitive training involves guided repetitive practice of standardized tasks designed to target specific cognitive skills or processes, with the aim of improving cognition via reinforcement of neural pathways. Accessibility to cognitive training is enhanced with the use of technology, particularly for older adults whose mobility may be limited because of their HF conditions. The computerized features also enable a more individualized approach, as opposed to their traditional counterparts, where activities may be tailored to the users' ability levels, thereby improving user engagement. The efficacy of CCT has been rigorously reviewed in a recent meta-analysis, which reported cognitive benefits in both cognitively healthy older adults and patients with mild cognitive impairment.16,17 Although CCT has been found to be viable in enhancing the cognitive function in other populations, its effectiveness in HF remains unclear. This review thus seeks to examine the feasibility and efficacy of CCT on the cognition of patients with HF.
This review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.18
Types of Studies
Randomized controlled trials (RCTs) published in the English language from January 1, 2000, to December 31, 2017, were included.
Types of Participants
Studies included adults (≥18 years old) with a current clinical diagnosis of HF with preserved ejection fraction or HF with reduced ejection fraction. Participants with any major neurological and/or psychiatric conditions were excluded.
Types of intervention
Eligible studies compared the effects of 4 hours or more of practice on standardized CCT tasks against an active or passive control condition.
Types of Outcome Measures
Primary outcome measures included (1) feasibility of intervention, measured through adherence rates and patient satisfaction, and (2) cognitive function (memory, working memory, attention, processing speed, and executive function), measured using standardized tests.
The secondary outcome measure was adaptive functioning, including instrumental activities of daily living.
Information Sources and Search Strategy
Searches were conducted in PubMed, Embase, and PsycINFO using the search terms “heart failure or congestive heart failure,” “cognition or neuropsychological tests,” and “cognitive training or brain training or memory training or computerized training” to identify relevant studies. The reference lists of retrieved articles were reviewed to identify studies that were overlooked in the electronic search.
Eligibility assessment was performed independently in a standardized manner via screening of titles and abstracts by 2 authors (Z.K. and Y.D.). Disagreements in retrieved full-text records were resolved by consensus.
Risk of Bias in Individual Studies
After the final selection of studies, risk of bias of individual studies was assessed using items recommended in the Cochrane Collaboration's risk of bias tool.19 The following aspects were assessed in the included studies to account for any heterogeneity of results: sequence generation, allocation concealment, blinding of outcome assessors, extent of loss to follow-up, and selective outcome. Trials that failed to include assessor blinding or were not compliant to intention-to-treat analyses were considered high or unclear risk of bias.
Quality of Evidence
The Grades of Recommendation, Assessment, Development and Evaluation system20 was used to assess the overall quality of evidence for the primary outcome of cognitive functioning and secondary outcome of adaptive functioning.
The initial search provided a total of 443 records. After adjusting for duplicates and reviewing of titles and abstracts, the authors discarded 434 studies. The full text of the remaining 9 citations was examined and 5 articles were excluded because (1) no intervention was documented or the intervention was not a form of cognitive training or (2) no cognitive outcomes were measured (see Figure 1). A total of 4 articles were retained for the final review.21–24
Characteristics of Included Studies
The included studies encompassed 138 participants (intervention group, n = 66; control group, n = 72) (Table 1). Mean age ranged between 55.6 and 75.1 years. Heart failure inclusion criteria consisted of (1) New York Heart Association classification I to IV21,23,24 and (2) left ventricular ejection fraction (LVEF) lower than 40%.21,23 One study included patients with documented HF regardless of LVEF.24 Another study assessed the presence of HF through self-report.22
Two studies included an active control group where controls were asked to read cardiovascular health magazines.23,24 One study had a wait-listed control group that received standard HF care during the study and were offered the cognitive intervention when the study period ended.21 Another study had a no-contact control group.22 Two of the included studies were pilot trials.21,23
Three cognitive interventions were described in the 4 included RCTs (Table 1).
Athilingam et al (2015) assessed the feasibility and efficacy of auditory cognitive training (ACT) on auditory processing speed and working memory. The ACT program used in the study was the Brain Fitness Program software developed by Posit Science Corporation.21 The use of ACT was supported by the information degradation theory, which postulates that processing errors in basic perceptual systems led to higher-order cognitive difficulties.25 Thus, ACT aimed to improve the cognitive efficiency of the auditory processing system to enhance higher-order cognitive function. Consisting of 6 listening exercises, the ACT required participants to identify, discriminate, and remember speech tones, words, and instructions. These exercises are continually adapted in speed and task complexity to participants' ongoing performance. Participants were instructed to use the program daily for 30 to 60 minutes, with the goal of completing up to 40 hours of ACT in 16 weeks. Computers were loaned to participants who do not have a home computer or Internet access.
Pressler et al (2011) also tested the Brain Fitness Program on the memory, working memory, processing speed, and executive function of patients with HF. Guided by the theories of neuroplasticity, this CCT sought to enhance sensory integration of auditory and visual information to strengthen neural networks necessary for memory encoding and recall.23 The program incorporated novel visual and auditory exercises to engage the user's learning and recall of information. Participants were asked to practice 40 hourly sessions of the program over 8 weeks. Computers were placed in the homes of all participants throughout the intervention. A unique feature of this CCT was the inclusion of a nurse-enhanced component. Advanced practice nurses visited and demonstrated use of the program to participants weekly at their homes to assess HF-related symptoms and intervention adherence. The same intervention and experimental design was examined again in Pressler et al (2015), but the nurses made only weekly contact with participants through telephone calls.
Ellis et al (2014) examined the effects of the speed of processing training on HF participants who were part of a longitudinal multisite RCT known as the Advanced Cognitive Training for Independent and Vital Elderly.26 This training targeted the improvement of processing speed in visual attention tasks. Computerized visual targets were presented via systematic reductions in display speeds. Task complexity was modified by changing stimulus detection, identification, or by adjusting the locations of peripheral targets. The training was delivered in ten 60–75-minute sessions by certified trainers in a group setting over 6 weeks.
Risk of Bias in Included Studies
Two studies were rated as having a low risk of bias across all criteria23,24 (Table 2). One study used cognitive measures that were novel to the study of CCT.21 Although these outcome measures lacked validity, the context of his RCT being a pilot study, in addition to the clear reporting of selection bias, performance bias and detection bias warrant the study to be classified as low risk of bias. Another study was considered high risk of bias owing to reporting bias and failure to comply with the intention-to-treat analysis.22
Effects of Intervention
Feasibility was measured in the pilot studies by the percentage of participants who completed the intervention and the number of training hours completed.21,23 Of the total participants assigned to the nurse-enhanced Brain Fitness intervention group, 85% completed the training. Seventy-seven percent were considered adherent to the intervention for meeting at least 90% of the required training hours. Intervention adherence was also met in the ACT study, with 78% of participants (completion criterion of 75%) completing at least 90% of the training hours.21 Although other studies did not measure feasibility, the reported completion rates were high (>80%). Common reasons cited for withdrawal included being ill or busy.
Because of the variability of intervention features, the following results are presented as a narrative synthesis of significant intervention effects across cognitive domains.
The most commonly measured cognitive domain was processing speed. Improvements in auditory processing speed were observed for participants who completed the ACT intervention at 16 weeks compared with wait-listed controls who demonstrated worsening performance over time.21 Large effect sizes were observed for this improvement (d = 0.78). Ellis et al (2014) found significant improvements in everyday speed of processing compared with the no-contact control group post training. Pressler et al (2015) reported that after controlling for age and LVEF, participants in the Brain Fitness group showed significantly less decline in their processing speed post intervention than active controls did. The above evidence suggests that processing speed may be enhanced by CCT.
Memory was assessed as a primary outcome in the nurse-enhanced CCT studies. Pressler et al (2011) found a statistically significant improvement in delayed recall memory compared with age-matched controls immediately post intervention and at follow-up at 4 weeks. Pressler et al (2015), however, did not report similar intervention effects on delayed memory. Inability to replicate findings was attributed to (1) the study's small sample size and (2) participants' higher baseline memory performance, that is, the Hopkins Verbal Learning Test-Revised scores, relative to the first study.24 The effects of cognitive training on memory function in patients with HF remain unclear given the limited evidence.
Working memory was assessed in 3 studies. A moderate effect size was found for improvements in working memory for participants in the ACT intervention (d = 0.44) compared with the control group, who had worsening scores over time.21 Pressler et al (2015) also found significant intervention effects in the improvement of working memory from baseline to 12 weeks post intervention for participants in the CCT group compared with active controls, whose scores declined over time. Even though a group-by-time interaction was not observed by Pressler et al (2011), a trend toward improvement in working memory was reported for the intervention and control groups over time. This suggests that cognitive training programs may benefit working memory in patients with HF; however, further studies are required.
Executive function was assessed in the nurse-enhanced CCT studies, but no significant intervention effects were found.
Attention was not assessed in any of the studies.
Quality of Evidence
A judgement was made to include all trials in the analysis as most of the evidence for the primary outcome of cognitive function came from RCTs with few limitations. Given considerable clinical heterogeneity in terms of CCT features and cognitive outcome measures, the evidence was rated down from high to moderate quality for inconsistency (Table 3). In addition, the total sample size of the review was less than the optimal information size (based on the control event rate of 0.2 and relative risk reduction of 25%). The evidence was therefore further downgraded by another level based on imprecision.
Adaptive Functional Outcomes
Instrumental Activities of Daily Living (IADLs): HF functional outcomes were measured in 3 of the studies. Athilingam et al (2015) found that participants who underwent ACT training had lesser decline in their HF self-care scores (d = 0.34) compared with wait-listed controls. Moreover, the ACT group tended to be more accurate and faster in their performance of Timed Instrumental Activities of Daily Living post intervention, whereas controls showed worsening speed and accuracy over time.21 Although no significant intervention effects on IADLs were reported in either of the nurse-enhanced CCTs, there was a trend for improved performance from baseline to 12 weeks post training in tasks of medications and telephone use for both training and active control groups.23,24 Overall, CCT seemed to enhance the adaptive functions of patients with HF.
The quality of evidence for adaptive functional outcomes was mainly downgraded for imprecision because of the small sample size of the current review, which limited the ability to detect a precise estimate of effect.
Despite growing interest in cognitive impairment and its impacts on self-care in HF, intervention research that targets and manages these cognitive problems is limited. In this review, we have reported preliminary evidence for the feasibility and efficacy of cognitive training on patients with HF. Four studies describing 3 CCTs—the ACT, the nurse-enhanced CCT, and the speed of processing training—were reviewed. Improvements in processing speed and working memory were found for participants with HF who had undergone the CCTs as compared with controls. No effects on executive function were found. Computerized cognitive trainings also improved adaptive functioning, including IADLs. Overall, adherence to interventions was high and participants were highly satisfied with the interventions, indicating the willingness of older patients with HF to receive cognitive training. This brief review is the first, to our knowledge, that synthesizes evidence showing that cognitive domains typically adversely affected in HF may be amenable to change via these targeted cognitive interventions.
The findings of improvements in processing speed and working memory are encouraging, given that these 2 cognitive domains were commonly identified to be adversely affected in patients with HF.6 Previous meta-analyses on CCTs in other populations have also found these domains to be highly responsive to training, demonstrating moderate to large effect sizes.16,17,27 Whereas improvements in these lower-order cognitive processes have been found, there seemed to be a lack of intervention effects on higher-order cognition, including delayed memory and executive functioning. As these domains were assessed in only 2 studies with small sample sizes, it is premature to draw conclusions about the efficacy of CCT on the memory and executive functions of patients with HF. However, the lack of translational benefits from lower-order to higher-order cognitive function raises doubts about the hypothesized mechanisms of the information degradation and neuroplasticity theories that underlie the Brain Fitness programs. Only 1 of the reviewed studies sought to examine the pathways through which CCT may improve cognition. An exploratory analysis conducted by Pressler et al (2015) found that their 8-wk CCT was significantly associated with increased levels of a brain-derived neurotropic factor involved in long-term memory formation and neuronal growth.24 This provides preliminary mechanistic insight to the effects of CCT and supports future studies in using biomarkers and neuroimaging to ascertain the mechanisms through which cognitive interventions may benefit cognition in HF.
The lack of intervention effects in memory and executive function could be explained by the training content of the CCTs. Gains in cognitive training have been found to be reflective of training domains.28 The reviewed CCTs consisted primarily of processing speed and working memory tasks instead of delayed recall, attention, and executive processes, thereby possibly accounting for a paucity of benefits in these domains. Future studies could thus consider developing CCTs that directly target higher-order processes given their impacts on ability to manage HF self-care. Aside from training modalities, intervention designs have shown to influence cognitive outcomes.17 Delivering CCT in a group-based setting was superior to home-based training owing to direct supervision to ensure treatment adherence, support technological issues, and motivate patients through increased social interaction.17 This could have contributed to the success of group-based speed of processing training.22 While the Brain Fitness interventions were administered within the homes of participants, they were visited or contacted weekly by nurses, who offer training support and motivation, thereby potentially contributing to higher treatment adherence and effects.23,24 Interventions involving advanced practice nurses have shown consistent efficacy in improving functional outcomes such as HF-related hospitalizations and quality of life.29,30 The unique contribution of supervision for home-based CCT can be considered for future studies.
The current review has also demonstrated preliminary benefits of CCTs on daily functional outcomes in patients with HF. However, improvement in self-care abilities remains unclear. In research of cognitive training, transfer effects have been observed in the training of executive control processes, where improvements were extended beyond a specific task to structurally similar new task settings.31–34 Transfer of training effects was also observed across different modalities, from cognition to improved physical benefits, including improved balance and postural control in healthy older adults.33,35 Such findings are encouraging when considered in the context of self-care in HF. Future studies could thus explore the efficacy of targeted cognitive training on the uptake of self-care skills.
Overall, the quality of evidence was assessed to be low for cognitive outcomes and moderate for adaptive functional outcomes. This is primarily because of the inconsistency and imprecision of results across the small number of reviewed studies. With only 4 studies included, there may be large potential impact if the average effect of a study differs in size or direction. The generalizability of the current findings was also limited because of the small sample sizes. Future RCTs involving larger sample sizes are required to ascertain the stability of intervention effects on the cognition of patients with HF. One of the studies relied only on self-report instead of clinical measures to determine the presence of HF. Hence, participants were not stratified by HF severity when allocated to treatment conditions and it is unclear whether HF status might have an impact on treatment effects. Furthermore, a recent meta-analysis has found that the cognitive function in patients with HF declined significantly within 1 year.13 Hence, it is questionable as to whether the effects of short-term interventions may be retained over a longer duration. Caution should therefore be taken when interpreting these results; they are considered indicative rather than definitive.
This review highlights potentially promising results when using cognitive training programs to improve the cognitive functions of patients with HF. Further RCTs with larger sample sizes are required to confirm intervention effects. Nonetheless, this finding is encouraging considering the association between cognitive impairment and suboptimal self-care among patients with HF. It will be important to examine whether any putative cognitive benefits translate into improved self-care management.
What's New and Important
- Cognitive domains adversely affected in HF may be amenable to change by nonpharmacological interventions.
- Preliminary evidence was found for the efficacy of CCT on improving the processing speed and working memory in patients with HF. Computerized cognitive training also appeared to improve the adaptive functioning of patients with HF.
- Quality of the evidence was rated to be of moderate to low quality and hence should be treated with caution. The stability of current findings would need to be evaluated in future RCTs involving larger sample sizes.
1. Ponikowski P, Anker SD, AlHabib KF, et al. Heart failure
: preventing disease and death worldwide. ESC Heart Fail
2. Bui AL, Horwich TB, Fonarow GC. Epidemiology and risk profile of heart failure
. Nat Rev Cardiol
3. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics—2011 update: a report from the American Heart Association. Circulation
4. Ross JS, Chen J, Lin Z, et al. Recent national trends in readmission rates after heart failure
hospitalization. Circ Heart Fail
5. Gure TR, Blaum CS, Giordani B, et al. Prevalence of cognitive impairment in older adults with heart failure
. J Am Geriatr Soc
6. Leto L, Feola M. Cognitive impairment in heart failure
patients. J Geriatr Cardiol
7. Vogels RL, Scheltens P, Schroeder-Tanka JM, Weinstein HC. Cognitive impairment in heart failure
: a systematic review of the literature. Eur J Heart Fail
8. Dickson VV, Tkacs N, Riegel B. Cognitive influences on self-care decision making in persons with heart failure
. Am Heart J
9. Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure
: the Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure
of the European Society of Cardiology (ESC)Developed with the special contribution of the Heart Failure
Association (HFA) of the ESC. Eur Heart J
10. Currie K, Rideout A, Lindsay G, Harkness K. The association between mild cognitive impairment and self-care in adults with chronic heart failure
: a systematic review and narrative synthesis. J Cardiovasc Nurs
11. Harkness K, Heckman GA, Akhtar-Danesh N, Demers C, Gunn E, McKelvie RS. Cognitive function and self-care management in older patients with heart failure
. Eur J Cardiovasc Nurs
12. Zuccalà G, Pedone C, Cesari M, et al. The effects of cognitive impairment on mortality among hospitalized patients with heart failure
. Am J Med
13. Hajduk AM, Kiefe CI, Person SD, Gore JG, Saczynski JS. Cognitive change in heart failure
: a systematic review. Circ Cardiovasc Qual Outcomes
14. Alagiakrishnan K, Mah D, Ahmed A, Ezekowitz J. Cognitive decline in heart failure
. Heart Fail Rev
15. Cannon JA, McMurray JJ, Quinn TJ. 'Hearts and minds': association, causation and implication of cognitive impairment in heart failure
. Alzheimers Res Ther
16. Hill NT, Mowszowski L, Naismith SL, Chadwick VL, Valenzuela M, Lampit A. Computerized cognitive training
in older adults with mild cognitive impairment or dementia: a systematic review and meta-analysis. Am J Psychiatry
17. Lampit A, Hallock H, Valenzuela M. Computerized cognitive training
in cognitively healthy older adults: a systematic review and meta-analysis of effect modifiers. PLoS Med
18. Moher D, Shamseer L, Clarke M, et al. Preferred Reporting Items for Systematic Review and Meta-analysis Protocols (PRISMA-P) 2015 statement. Syst Rev
19. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ
20. Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck-Ytter Y, Schünemann HJ. Rating quality of evidence and strength of recommendations: what is “quality of evidence” and why is it important to clinicians? BMJ
21. Athilingam P, Edwards JD, Valdes EG, Ji M, Guglin M. Computerized auditory cognitive training
to improve cognition
and functional outcomes in patients with heart failure
: results of a pilot study. Heart Lung
22. Ellis ML, Edwards JD, Peterson L, Roker R, Athilingam P. Effects of cognitive speed of processing training among older adults with heart failure
. J Aging Health
23. Pressler SJ, Therrien B, Riley PL, et al. Nurse-enhanced memory intervention in heart failure
: the MEMOIR study. J Card Fail
24. Pressler SJ, Titler M, Koelling TM, et al. Nurse-enhanced computerized cognitive training
increases serum brain-derived neurotropic factor levels and improves working memory in heart failure
. J Card Fail
25. Schneider BA, Pichora-Fuller MK. Implications of perceptual deterioration for cognitive aging research. In: Craik FIM, Salthouse TA, eds. The Handbook of Aging and Cognition
. 2nd ed. Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers; 2000:155–219.
26. Jobe JB, Smith DM, Ball K, et al. ACTIVE: a cognitive intervention trial to promote independence in older adults. Control Clin Trials
27. Leung IH, Walton CC, Hallock H, Lewis SJ, Valenzuela M, Lampit A. Cognitive training
in Parkinson disease: a systematic review and meta-analysis. Neurology
28. Ball K, Berch DB, Helmers KF, et al. Effects of cognitive training
interventions with older adults: a randomized controlled trial. JAMA
29. Naylor MD, Brooten DA, Campbell RL, Maislin G, McCauley KM, Schwartz JS. Transitional care of older adults hospitalized with heart failure
: a randomized, controlled trial. J Am Geriatr Soc
30. Sochalski J, Jaarsma T, Krumholz HM, et al. What works in chronic care management: the case of heart failure
. Health Aff
31. Bherer L, Kramer AF, Peterson MS, Colcombe S, Erickson K, Becic E. Transfer effects in task-set cost and dual-task cost after dual-task training in older and younger adults: further evidence for cognitive plasticity in attentional control in late adulthood. Exp Aging Res
32. Karbach J, Kray J. How useful is executive control training? Age differences in near and far transfer of task-switching training. Dev Sci
33. Li KZ, Roudaia E, Lussier M, Bherer L, Leroux A, McKinley PA. Benefits of cognitive dual-task training on balance performance in healthy older adults. J Gerontol A Biol Sci Med Sci
34. Bherer L. Cognitive plasticity in older adults: effects of cognitive training
and physical exercise. Ann N Y Acad Sci
35. Li H, Li J, Li N, Li B, Wang P, Zhou T. Cognitive intervention for persons with mild cognitive impairment: a meta-analysis. Ageing Res Rev