the mind of a failing heart: a systematic review of the association between congestive heart failure...

Abstract

Background: Congestive heart failure (CHF) is afrequent complication of most diseases of the heart.CHF is associated with impairment in several aspectsof the quality of life of patients, including mood andcognitive performance. Early reports indicated thatpatients with CHF display deficits in memory andother intellectual abilities, although the clinical rele-vance of these findings remains unclear.

Aim: We reviewed systematically the medical litera-ture with the aim of clarifying the association betweenCHF and cognitive functioning.

Methods: Systematic review of Medline database forstudies published between 1966 and June 2000 usingthe following key words: congestive heart failure,cognition, cognitive disorders, memory, memorydisorders, short-term memory, attention.

Results: Thirteen studies reported cognitive informa-tion on patients with CHF, but only five met inclusioncriteria for systematic review.Three reports describedattention (total number of subjects = 369 patients and882 controls) and memory scores (total number of

subjects = 247 patients and 748 controls), two studiesreported measures of general cognitive functioning(total number of subjects = 203 patients and 704controls) and one reported the rate of cognitiveimpairment (total number of subjects = 88 patientsand 987 controls). Pooled analysis indicated thatCHF is associated with a pattern of generalized cogni-tive impairment that includes memory and attentiondeficits.

Conclusions: The results of the present review high-light the enormous paucity of systematic informationabout the association between CHF and cognitivefunctioning, with only five studies reporting datasuitable for analysis. We expect that new case-controland cohort studies will be designed to confirm thepresence of cognitive impairment in patients withCHF and trust that this information will improve themanagement of CHF patients and our understandingof the mechanisms associated with cognitive declinein later life. (Intern Med J 2001; 31: 290–295)

Key words: attention, cognitive disorder, congestiveheart failure, heart disease, memory, neuro-psychology, quality of life.

ORIGINAL ARTICLE

The mind of a failing heart: a systematic review of theassociation between congestive heart failure and cognitive functioning

O. P. ALMEIDA1 and L. FLICKER2

1Department of Psychiatry and Behavioural Science and 2Department of Medicine, University of Western Australia,Perth,Western Australia, Australia

INTRODUCTION

Congestive heart failure (CHF) is a common compli-cation of most diseases of the heart. Its prevalenceincreases exponentially from age 60 years,1 such that

CHF is now one of the leading causes of hospitaliza-tion, morbidity and mortality in western societies.2

The findings of several surveys indicate that physical,social, work and leisure activities are significantlyimpaired among CHF subjects.3 Psychologicaldistress is also frequent, although data on this impor-tant aspect of the quality of life of patients remainsparse and difficult to interpret. Depression rates,for example, seem to be high (up to 58%) in both inpatient and outpatient settings.4,5 Another impor-tant, but neglected, aspect of the quality of life of

Correspondence to: Associate Professor Osvaldo P. Almeida, UWADepartment of Psychiatry, Mental Health Unit for the Elderly, AinslieHouse, Level 6, Royal Perth Hospital, GPO Box X2213, Perth,WA6847, Australia. Email: [email protected]

Received 13 October 2000; accepted 30 March 2001.

Internal Medicine Journal 2001; 31: 290–295

291The mind of a failing heart


patients with CHF is cognitive functioning. Earlyreports indicated that up to 80% of patients withsevere CHF display deficits in memory and othercognitive abilities.6 The consequences of these deficitsare not clear, but it is conceivable that patients withcognitive impairment have even higher morbidity andmortality rates. For example, Cline et al.7 reportedthat 10 of their 22 patients were unable to name themedication they were receiving for the treatment ofCHF, 11 of 22 could not state the doses and 14 of 22failed to remember when to take their tablets. Allsubjects were surveyed 30 days after receiving detailedverbal and written information about their treatmentregimen. Failing to take the prescribed medication, insuch cases, may increase the frequency of clinicalcomplications associated with CHF.

We designed the present study with the aim ofreviewing all available medical information on theassociation between CHF and cognitive functioning.

METHODS

We searched Medline using the following strategy:#1. exp heart failure, congestive/; #2. esp cognitiondisorders/; #3. esp memory/or exp memory disorders/or exp memory, short-term/; #4 exp attention/; # 5. 2or 3 or 4; #6. 1 and 5; limit 6 to human. This searchstrategy returned a total of 35 papers for the periodbetween 1966 and June 2000. Another five paperswere retrieved through a manual search of referencesquoted by other papers. Criteria for inclusion ofstudies were: (i) case-control or cross-sectionalsurveys designed to investigate the associationbetween CHF and cognitive impairment; (ii) the useof valid measures of general cognitive functioning,memory or attention; and (iii) reported data suitablefor analysis (i.e. mean and SD, or proportions).Studies published in languages other than Englishwere excluded. Reports on heart transplant candi-dates were considered to indicate the presence ofCHF even when clinical criteria or physiologicalmeasures (such as ejection fraction) were not explic-itly described. In the case of general cognitivemeasures, the following priority sequence was used:Mini Mental State Examination (MMSE), BlessedDementia Scale and Wechsler Adult Intelligence Scale(WAIS) full IQ. Delayed memory scores had prece-dence over immediate recall or general memorymeasures; verbal memory had precedence over visualmemory. This hierarchical approach was used toincrease the homogeneity between studies and givepriority to the cognitive measures most commonlyused to describe impairment in general cognitive

functioning and memory. Cognitive impairment wasrecorded as present for cases with MMSE < 24,because this cut-off point has been used by otherstudies of patients with CHF.8

Data were analysed using the software Review Man-ager 4.1 (Cochrane Collaboration, Oxford, UK). Thestandardized mean difference (SMD = (xi – x)/SD)was estimated for cognitive ratings and the weightedtotal SMD calculated. A pooled odds ratio (OR) wascalculated for proportions (fixed-effect model).

RESULTS

Thirteen studies reported cognitive information onpatients with CHF, but only five met inclusion criteriafor systematic review. Table 1 describes the charac-teristics of studies included in the analyses. Threereports described attention9–11 and memory6,9,10

scores, two measures of general cognitive functioning6,9

and one the rate of cognitive impairment.9 Memoryscores represent the performance of patients on theRey Complex Figure Test9 and Logical Memorysubtest of the Wechsler Memory Scale.6,10 Attentionscores indicate results on tests of Attention Matrices9

and Trail Making B.10,11 Finally, overall cognitiveperformance was assessed with the MMSE8,11 andWAIS.6 One study8 described that 50 of 88 patientswith CHF and 197 of 987 controls had MMSE scoreslower than 24 (OR = 5.28; 95% confidence interval(CI) = 3.37–8.25), although CHF subjects were olderand more depressed than controls. Results of thepooled analyses are summarized in Fig. 1.

DISCUSSION

The results of the present review indicate that CHF isassociated with a pattern of generalized cognitiveimpairment that includes memory and attentiondeficits. They also highlight the enormous paucity ofsystematic information in the area, with only fivestudies reporting data suitable for analysis. Sevenother reports described the cognitive performance ofpatients with CHF, but did not use suitable compar-ison groups (Table 2). Overall, the published studiesseem to confirm that CHF is associated with cognitiveimpairment.

Attentional deficits are expected among patients witha severe and debilitating medical illness, such as CHF.In fact, the performance of subjects in other cognitivetasks is likely to be influenced by the presence ofattentional deficits, which may partly explain thefinding of memory deficits and generalized cognitive

Almeida & Flicker292


Tab

le1

Cha

ract

eris

tics

of

stud

ies

repo

rtin

g co

gnit

ive

func

tion

ing

of p

atie

nts

wit

h co

nges

tive

hea

rt f

ailu

re

Ref

eren

ceM

etho

dsP

arti

cipa

nts

Inte

rven

tion

sO

utco

mes

Not

es

9C

ase-

cont

rol s

tudy

;C

HF

×ot

her

hear

tdi

seas

es

CH

F=

183;

no C

HF

= 6

84N

one

MM

SE

,Lei

pad

(sel

f-m

emor

y),v

erba

lfl

uenc

y,sp

atia

l spa

n,di

git

span

,Rey

-im

med

iate

,Rey

-del

ayed

att

enti

onal

mat

rice

s

Sco

res

wer

e no

t ad

just

ed f

orco

nfou

ndin

g va

riab

les,

such

as a

ge,g

ende

r an

ded

ucat

ion

8C

ase-

cont

rol s

tudy

;C

HF

×no

CH

F,

CH

F (

MM

SE

<24

)×C

HF

(M

MS

E >

23)

Tot

al s

ampl

e si

ze =

107

5;C

HF

= 8

8;M

MS

E

< 2

4 =

247

Non

eM

MS

EL

ogis

tic

regr

essi

on w

ith

cogn

i-ti

ve im

pair

men

t us

ed a

sin

depe

nden

t va

riab

le:C

HF

OR

= 1

.96

(1.0

7–3.

58)

11C

ase-

cont

rol s

tudy

of

subj

ects

wit

h E

F

< 3

5%

NY

FC

-I v

ersu

s N

YF

C-I

I/II

IN

one

WA

IS v

ocab

ular

y,di

git

span

,tra

ilm

akin

g A

,tra

il m

akin

g B

10M

atch

ed c

ase-

cont

rol

stud

y44

hea

rt t

rans

plan

tca

ndid

ates

,44

heal

thy

cont

rols

rec

ruit

ed f

rom

the

com

mun

ity

Non

eG

roov

ed p

egbo

ard,

trai

l mak

ing

A,t

rail

mak

ing

B,l

ogic

al m

emor

yM

atch

ing

was

not

str

inge

nt(a

ge,g

ende

r,et

hnic

ori

gin,

educ

atio

n)

6 C

ase-

cont

rol s

tudy

(sub

ject

s us

ed a

s th

eir

own

cont

rols

),be

fore

and

afte

r su

rger

y

54 p

re-o

pera

tive

pat

ient

s,20

pre

/pos

t ca

rdia

ctr

ansp

lant

pat

ient

sre

crui

ted

from

the

Tem

ple

Uni

vers

ity

Hos

pita

l (U

SA

)

Car

dic

tran

spla

ntW

AIS

ful

l IQ

,WA

IS v

erba

l IQ

,WA

ISpe

rfor

man

ce I

Q,W

MS

logi

cal

mem

ory,

WM

S v

isua

l rep

rodu

ctio

n,H

R im

pair

men

t in

dex,

HR

cat

egor

y,H

R t

actu

al p

erfo

rman

ce,H

R t

otal

tim

e,H

R m

emor

y,H

R lo

cati

on,H

Rsp

eech

-sou

nd,H

R s

eash

ore,

HR

fing

er-o

scill

atio

n

Sub

ject

s w

ith

mor

e se

vere

impa

irm

ent

wer

e le

ss li

kely

to r

ecei

ve h

eart

tra

nspl

ant

CH

F,c

onge

stiv

e he

art

failu

re;M

MS

E,M

ini M

enta

l Sta

te E

xam

inat

ion;

WA

IS,W

echs

ler

Adu

lt I

ntel

ligen

ce S

cale

;EF

,eje

ctio

n fr

acti

on;N

YF

C,N

ew Y

ork

func

tion

al c

lass

;WM

S,

Wec

hsle

r M

emor

y S

cale

;HR

,Has

tead

–Rei

tan.



impairment among patients. Gorkin et al.11 showedthat subjects with more severe forms of CHF (func-tional class II/III) have greater difficulty than controls(functional class I) on digit span and trail making A tests. They also found that performance on these tasks is associated with decreased functional capacity,as measured by the 6 min walk test. These resultssuggest that cognitive deficits become more promi-nent with increasing severity of illness, althoughSchall et al.6 found that cardiac transplant fails toreverse the deficits of some cognitive skills, such asmemory.

The mechanisms that contribute to the developmentof cognitive impairment among patients with CHFremain unclear. Zuccalà et al.12 observed a linear rela-tionship between MMSE scores and left ventricularejection fraction rates for values lower than 40%.Similarly, Putzke et al.13 noted that Trail B, DigitSymbol Substitution and Stroop scores were allsignificantly associated with cardiac output. Cerebro-vascular disease is another likely cause of cognitiveimpairment, because many patients with CHF have widespread cardiovascular problems and are atincreased risk for strokes. Data derived from the

Figure 1 Summary data of studies investigating the cognitive performance of patients with congestive heart failure(CHF) and controls. Controls include subjects with no medical conditions, subjects with other cardiovascular diseasesand patients with mild CHF or CHF after cardiac transplant. The first comparison table displays the data summary forattention scores, followed by memory and general cognitive skills. SMD, standardized mean difference.

Almeida & Flicker294


Rotterdam study,14 for example, indicate that whitematter disease is associated with subjective memoryimpairment and lower scores on tests of cognitivefunction. Similarly, patients with strokes are morelikely to develop cognitive deficits and dementia.Other cardiovascular problems, such as low15 andhigh blood pressure,16 both frequent among patientswith CHF, are associated with cognitive impairment.Cognitive impairment in CHF may also be due to theabnormal hormonal response that characterizes thedisease, although no direct evidence is currently avail-able to support this hypothesis. Finally, there is stillthe possibility that some of the cognitive deficits ofpatients with CHF are secondary to the presence ofdepressive or other psychiatric symptoms.

In conclusion, the results of the present systematicreview suggest that CHF is associated with a patternof cognitive impairment that includes attention andmemory deficits. However, the number of studies isstill very small and neuropsychological informationis only available for fewer than 400 subjects withCHF. Generalization of these findings is furtherlimited by the heterogeneity of the samples selectedfor investigation and the absence of suitable controlgroups. In addition, most studies were exploratory

in nature and, as a consequence, used a large numberof neuropsychological tests without any a priorihypotheses.

We expect that new case-control and cohort studieswill be designed to confirm the presence of cognitiveimpairment in patients with CHF and trust that thisinformation will improve the management of CHFpatients and our understanding of the mechanismsassociated with cognitive decline in later life.

REFERENCES

1 Ho KK, Pinsky JL, Kannel WB, Levy D. The epidemiology ofheart failure: The Framingham study. J Am Coll Cardiol1993; 22: A6–13.

2 Watson RDS, Gibbs CR, Lip GYH. ABC of heart failure:Clinical features and complications. BMJ 2000; 320: 236–9.

3 Mayou R, Blackwood R, Bryant B, Garnham J. Cardiacfailure: Symptoms and functional status. J Psychosom Res1991; 35: 399–407.

4 Koenig HG. Depression in hospitalized older patients withcongestive heart failure. Gen Hosp Psychiatry 1998; 20:29–43.

5 Havranek EP, Ware MG, Lowes BD. Prevalence of depressionin congestive heart failure. Am J Cardiol 1999; 84: 348–50.

6 Schall RR, Petrucci RJ, Brozena SC, Cavarocchi NC, JessupM. Cognitive function in patients with symptomatic dilated

Table 2 Characteristics of excluded studies that reported cognitive functioning of patients with congestive heart failure

Reference Reason for exclusion

17 Sixty-two baseline assessments of cardiac transplant candidates; only 30 survived an average of 36 monthsfor reassessment; 11 were retested, of whom seven were operated on. Outcome is presented as change ofscores compared with baseline, but data are not available. The overall result was that the change of scoresbetween transplanted and not transplanted groups was significant (P = 0.03).

18 The authors reported an 18-month follow-up study of an undisclosed number of subjects with CHF andother medical conditions. They reported that CHF patients’ scores on the Dementia Rating Scaledeteriorated over time compared with the scores of adults with CVA. No hard data are available and theauthors have obviously made multiple comparisons and no statistical correction.

19 Abstract information only. Authors claimed that heart transplant was associated with improved IQ, memoryand attention.

20 The authors investigated the effect of age, gender, hip fracture, CVA and ethnic origin on mental status(Blessed Dementia Scale) of 212 older adults discharged from hospital. CHF was significantly associatedwith cognitive impairment, but no data are described in the paper.

13 Cross-sectional survey of 62 heart transplant candidates. The performance of 24% of subjects was 2SDbelow the mean for BNT, TMB, DSS, WMS logical memory, WMS paired associates. This study showed apositive correlation between the severity of heart failure and degree of cognitive impairment.

21 Study described in the paper by Acanfora et al.9 The number of subjects is unclear. Acanfora et al.9 reportedmean scores, but not SD; the MMSE score of patients with CHF was lower than for controls with othercardiovascular diseases (22.5 × 23.8).

12 Cross-sectional exploratory study of 57 subjects with CHF. Thirty of 57 patients had MMSE < 24. Therewas no comparison group. This study showed a positive correlation between the severity of heart failure anddegree of cognitive impairment.

CHF, congestive heart failure; CVA, cerebrovascular accident; MMSE, Mini Mental State Examination; WAIS, Wechsler AdultIntelligence Scale; WMS, Wechsler Memory Scale; BNT, Boston Naming Test; TMB, Trail Making B; DSS, Digit Symbol SubstitutionTest.



cardiomyopathy before and after cardiac transplantation. J AmColl Cardiol 1989; 14: 1666–72.

7 Cline CMJ, Björck-Linné AK, Israelsson BYA, WillenheimerRB, Erhardt LR. Non-compliance and knowledge ofprescribed medication in elderly patients with heart failure.Eur J Heart Failure 1999; 1: 145–9.

8 Cacciatore F, Abete P, Ferrara N et al. Congestive heartfailure and cognitive impairment in an older population. J AmGeriatr Soc 1998; 46: 1343–8.

9 Acanfora D, Trojano L, Iannuzzi GL et al. The brain incongestive heart failure. Arch Gerontol Geriatr 1996; 23:247–56.

10 Putzke JD, Williams MA, Daniel J, Foley BA, Kirklin JK, BollTJ. Neuropsychological functioning among heart transplantcandidates: A case control study. J Clin Exp Neuropsychol2000; 22: 95–103.

11 Gorkin L, Norvell NK, Rosen RC et al. Assessment of qualityof life as observed from the baseline data of the studies of leftventricular dysfunction (SOLVD) trial quality-of-lifesubstudy. Am J Cardiol 1993; 71: 1069–73.

12 Zuccalà G, Cattel C, Manes-Gravina E et al. Left ventriculardysfunction: A clue to cognitive impairment in older patientswith heart failure. J Neurol Neurosurg Psychiatry 1997; 63:509–12.

13 Putzke JD, Williams MA, Rayburn BK, Kirklin JK, Boll TJ.The relationship between cardiac function andneuropsychological status among heart transplant candidates.J Card Fail 1998; 4: 295–303.

14 Breteler MMB, van Swieten JC, Bots ML et al. Cerebralwhite matter lesions, vascular risk factors, and cognitivefunction in a population-based study: The Rotterdam study.Neurology 1994; 44: 1246–52.

15 Guo Z, Viitanen M, Winblad B. Low blood pressure and five-year mortality in a Stockholm cohort of the very old: Possibleconfounding by cognitive impairment and other factors. Am JPublic Health 1997; 87: 623–8.

16 Cacciatore F, Abete P, Ferrara N et al. The role of bloodpressure in cognitive impairment in an elderly population.J Hypertens 1997; 15: 135–42.

17 Bornstein RA, Starling RC, Myerowitz P, Haas GJ.Neuropsychological function in patients with end-stage heartfailure before and after cardiac transplantation. Acta NeurolScand 1995; 91: 260–5.

18 Brown JW, Chobor A, Zinn F. Dementia testing in the elderly.J Nerv Ment Dis 1993; 181: 695–8.

19 Phlaum S, Donohue TJ. Neuropsychological sequelae of hearttransplant patients. J Card Fail 1999; 5: 68.

20 Proctor EK, Morrow-Howell N, Chadiha L et al. Physical andcognitive functioning among chronically ill African–Americanand white elderly in home care following hospital discharge.Med Care 1997; 35: 782–91.

21 Rengo F, Acanfora D, Trojano L et al. Congestive heart failureand cognitive impairment in the elderly. Arch GerontolGeriatr 1995; 20: 63–8.

the mind of a failing heart: a systematic review of the association between congestive heart failure...

Documents