Neuropsychological impairments fatigue syndrome, multiple ... · The Beck depression inventory (BDI)15 and the state-trait anxietyscale (STAI)16werealso given. In addition, a metamemory

3Journal ofNeurology, Neurosurgery, and Psychiatry 1995;58:38-43

Neuropsychological impairments in chronicfatigue syndrome, multiple sclerosis, anddepression

John DeLuca, Susan K Johnson, Dawn Beldowicz, Benjamin H Natelson

AbstractTo examine the degree and nature ofcog-nitive impairments in chronic fatiguesyndrome, a comprehensive neuropsy-chological battery was given to patientswith chronic fatigue syndrome, multiplesclerosis, depressed patients, and healthycontrols. The battery included tests ofattention and concentration, informationprocessing speed, verbal and visualmemory, intellectual ability, and conceptformation. Measures of depression andanxiety were also obtained. The chronicfatigue syndrome group did not differfrom the depressed group in overall neu-ropsychological performance, but dif-fered from the multiple sclerosis andcontrol groups. The most significantimpairment was in information process-ing speed in the chronic fatigue syn-drome group. Depression and anxietywere not related to neuropsychologicalperformance. The influence of reducedinformation processing on other areas ofcognition is discussed.

(_NeurolNeurosurg Psychiatry 1995;58:38-43)

Keywords: neuropsychological impairments; multiplesclerosis; depression; fatigue syndrome

Kessler Institute forRehabilitation, WestOrange, New Jersey,USADepartment ofResearch andPsychologyJ DeLucaS K JohnsonUMDNJ-New JerseyMedical School,Newark, New Jersey,USADepartment ofPhysical Medicine andRehabilitationJ DeLucaS K JohnsonDepartment ofNeurosciencesJ DeLucaD BeldowiczB H NatelsonCorrespondence to:Dr John DeLuca, KesslerInstitiute for Rehabilitation,1199 Pleasant Valley Way,West Orange, NJ 07052USA.Received 15 March 1994and in revised form13 June 1994.Accepted 20 July 1994

Chronic fatigue syndrome is a disabling illnessthat primarily afflicts middle aged, whitewomen. The illness is defined as new onset ofsevere fatigue without any known medicalcause'; accompanying symptoms are rheuma-tological (for example, myalgias, joint pain),infectious (for example, sore throat, fever,swollen lymph nodes), and neuropsychiatric(for example, headache, weakness, depres-sion, cognitive disturbance).

There have been few published reportsaimed at examining cognition in patients withchronic fatigue syndrome. Early studies hadmethodological drawbacks, particularly theabsence of a control group. More recent stud-ies have generally yielded subtle, but inconsis-tent findings. For instance, verbal memoryimpairments in chronic fatigue syndrome or

myalgic encephalomyelitis (chronic fatiguesyndrome as defined in the United Kingdom)have been found by some2' but not others,45or have been related to depression.6 Similarly,both impaired4 and unimpaired2 visual mem-ory have been reported in chronic fatigue syn-drome and myalgic encephalomyelitis.Discrepancies have also been found in multi-

modal sensory evoked potentials and auditoryevent related cognitive potentials. One study7found impaired auditory cognitive potentials(N2 and P3) compared with controls, whichwere interpreted as reflecting impairments inattention and speed of information processingin myalgic encephalomyelitis. Another group,however, did not find differences betweenchronic fatigue syndrome and controls in thevisual domain.8 Reaction times were signifi-cantly prolonged in the chronic fatigue syn-drome and myalgic encephalomyelitis groupsof both studies.DeLuca et a19 examined patients with

chronic fatigue syndrome, multiple sclerosis,and healthy controls on the paced auditoryserial addition test (PASAT), a test of com-plex auditory information processing. Theyfound that both the chronic fatigue syndromeand multiple sclerosis groups scored signifi-cantly below controls on the PASAT, but didnot differ from each other. Performance onthe PASAT was not significantly correlatedwith depression. DeLuca et a19 suggested thatpatients with chronic fatigue syndrome havedifficulty in the simultaneous processing ofmultiple elements of complex information.They hypothesise that the perceived memoryimpairments in those with chronic fatiguesyndrome may actually be a reflection ofimpaired information processing rather thanmemory itself. Partial support for this con-tention has been provided in recent reports.35

Given the paucity of objective data and theinconsistencies in the data that do exist, thepresent study was conducted to assess a broadrange of neuropsychological functioning inpersons with chronic fatigue syndrome.Patients with chronic fatigue syndrome werecompared with patients who had multiplesclerosis, patients with major depressive dis-order or dysthymia, and healthy controls. Themultiple sclerosis group was chosen so thatperformance of the chronic fatigue syndromegroup could be compared with a populationpresenting with a symptom cluster similar inmany ways to patients with chronic fatiguesyndrome (for example, impairments in infor-mation processing, learning and memory,fatigue, depression), but of known organicaetiology. The addition of a depressed groupis critical as patients with chronic fatigue syn-drome often manifest symptoms of depres-sion, and given the current controversy overwhether cognitive symptoms in chronicfatigue syndrome are due solely to depres-sion.6 Finally, patients with chronic fatiguesyndrome in the present study were selected

38

on August 23, 2019 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.58.1.38 on 1 January 1995. D

ownloaded from

http://jnnp.bmj.com/

Neuropsychological impairments in chronic fatigue syndrome, multiple sclerosis, and depression

more stringently regarding premorbid psychi-atric state, disease duration, and severity thanin our prior work.5 9

MethodsPATIENTSPatients who met the published case defini-tion for chronic fatigue syndrome'0 wereincluded in the study. Additional inclusioncriteria were: (1) patients had to have beendiagnosed within the past four years; (2)symptoms of at least moderate severity had tobe present at the time of intake; (3) there wasno history of a psychiatric disorder in the fiveyears before the diagnosis of chronic fatiguesyndrome or history of substance abuse at anytime as determined by the diagnostic inter-view schedule; and (4) there had to be nohistory of loss of consciousness greater thanfive minutes.Twenty six patients with chronic fatigue

syndrome were compared with age, sex, andeducation matched groups consisting of 12patients with multiple sclerosis, 14 depressedpatients, and 20 healthy controls. No signifi-cant differences were found among thegroups in mean age (chronic fatigue syndromegroup = 34-3 (SEM 1.8); healthy group =37-0 (2-4); multiple sclerosis group = 36-0(3 0); depressed group = 39 7 (2.1); F (3,68)= 1-02, p < 0 38) or in mean years of educa-tion (chronic fatigue syndrome group = 14-6(0-4); healthy group = 15-5 (0 6); multiplesclerosis group = 14-2 (0 8); depressed group= 15-8 (0-6); F (3,68) = 1-5, p < 0.20).Patients with multiple sclerosis were all clini-cally stable, were diagnosed with clinicallydefinite multiple sclerosis," and had anexpanded disability status scale (EDSS) of 2-5or less. This cut offwas specifically selected tocontrast patients with multiple sclerosis and"mild" physical symptoms with patients withchronic fatigue syndrome. The mean EDSSfor the multiple sclerosis group was 1-2 (SEM0-22). Current affective or anxiety disorderswere not a cause for exclusion for patientswith chronic fatigue syndrome or multiplesclerosis.'0 Healthy subjects and patients withdepression were recruited from local thera-pists and by advertising in the local commu-nity, and were paid for their participation.Healthy subjects and depressed patients whoreported no medical problems and were tak-ing no medication other than birth controlpills were included. Depressed patients had tobe diagnosed with either major depression ordysthymia according to the DIS III-R criteria.Fatigue was assessed by the Krupp fatiguescale.'2 The mean level of fatigue in thechronic fatigue syndrome group (57-2 (SEM1-0)) was significantly increased relative toall other groups (p < 0-001). The depressed(32-5 (3-8)) and multiple sclerosis (35-6(6*1)) groups did not significantly differ fromeach other, but both differed significantlyfrom controls (17-1 (2 3); p < 0 01).

PROCEDURESubjects were given a battery of standard neu-

ropsychological tests, administered andscored in accordance with published proce-dures. The tests consisted of the digit spansubtest of the WAIS-R13 -a test of "auditoryspan" of attention; paced auditory serial addi-tion test (PASAT)-a test of information pro-cessing speed; trail making test-a test ofvisuomotor scanning and concentration;booklet category test (BCT)-a test of com-plex problem solving and conceptual reason-ing; vocabulary, arithmetic, similarities, andblock design, subtests of the WAIS-R-testsof intellectual functions; Rey complex figuretest-a test of visual memory; California ver-bal learning test (CVLT)-a test of verbal listlearning and memory; and the logical memorysubtest of the Wechsler memory scalerevised'4-a test of verbal paragraph recall.The Beck depression inventory (BDI)15 andthe state-trait anxiety scale (STAI)16 were alsogiven. In addition, a metamemory question-naire'7 and a cognitive problems checklistwere given to all subjects. Based on factoranalysis, the metamemory questionnaire isdivided into four primary areas of memory:attention/prospective memory, retrogradememory, historic/overload memory, andanterograde memory. The cognitive problemschecklist was used to quantify the cognitivecomplaints that patients with chronic fatiguesyndrome often have. It consisted of a 1-5Likert scale, where subjects rated their diffi-culties in the areas of attention, concentra-tion, memory, and reasoning. The tests andquestionnaires were given in one session lastingabout 150 minutes.

DATA ANALYSISPlanned comparisons of non-orthogonal con-trasts were performed between the clinicalgroups and the healthy controls. With thesetests, it is not necessary to perform an overallomnibus test of significance before testing theplanned comparisons.'8 Dunnett's test wasused to compare each of the clinical groupswith the healthy group. Contrasts between thevarious clinical groups were performed withthe Dunn's multiple comparisons procedure.Fatigue data were analysed by Wilcoxon ranktest.18

Because cognitive impairments in multiplesclerosis are well documented,'9 one tailedDunnett's tests comparing multiple sclerosiswith controls were performed. Also, becauseof the previous findings of impaired PASATperformance in patients with chronic fatiguesyndrome and multiple sclerosis,9 a one tailedDunnett's test for contrasts involving a con-trol mean with all clinical groups was per-formed for the PASAT results. All othercontrasts were analysed by a two tailed proce-dure.

ResultsSUBJECTIVE COMPLAINTSOverall, patients with chronic fatigue syn-drome had the broadest degree of subjectivecomplaints of cognitive impairment comparedwith the other groups (table 1). Subjective

39



j.com/

J Neurol N


ownloaded from


DeLuca, J7ohnson, Beldowicz, Natelson

Table I Subjective data for the clinical and control groups (meaan (SEM))

CFS Healthy MS DEP

CPC questionnaire data:Attention 3-1 (0-21)**t 1.1 (0 08) 1-9 (0 35) 2-3 (0 29)**Concentration 3-2 (0-19)t 1.1 (0-18) 1-9 (0 23) 2-6 (0-25)tMemory 3-2 (0 20)** 1-2 (0-12) 2-2 (0 25)** 2-2 (0 30)*Reasoning 3-1 (0 22)4 1 1 (0 08) 1-6 (0-26) 1-8 (0 23)

Metamemory questionnaire data:Attention/prospective 2-7 (0 2)fi 1-4 (0-1) 1-7 (0-1) 2-1 (0 2)tRetrograde 1-8 (0 2)* 1-2 (0-1) 1-3 (0-1) 1-6 (0 2)Anterograde 2-3 (0 2)t 1-2 (0-1) 1-6 (0 2) 1-8 (0 2)Biographical 1-3 (0-2) 1 1 (0.0) 1 1 (0-1) 1-1 (0-1)

*p < 0 05; **p < 0-01 v healthy group.tP < 0 05 v healthy and MS groups.#p < 0 05 v DEP group.CFS = chronic fatigue syndrome; MS = multiple sclerosis; DEP = depressed.

Table 2 Performance ofclinical and control groups on psychological measures (mean(SEM))

CFS Healthy MS DEP

Beck depression inventory 14-5 (1-6)* 2-1 (0.6) 8-7 (2-3)t 18-8 (2.7)*State anxiety inventory 39-8 (2 7)* 26-8 (1-4) 29-4 (1-8)* 43-3 (3 8)*

*p < 0-01 v healthy group.t-p < 0 05 v DEP group.1p < 0 05 v CFS and DEP groups.Abbreviations as for table 1.

Table 3 Performance of clinical and control groups on tests of attention, concentration,and information processing (mean (SEM))

CFS Healthy MS DEP

PASAT total 124 (6-1)* 144 (5-1) 130 (6-7) 136 (8-1)Trails A 27-5 (1-8) 27-5 (2-1) 27-9 (7-8) 27-8 (2-4)Trails B 55-8 (3-0) 52-8 (3-7) 68-6 (7 8) 62-0 (5 8)Digitspan 17-0 (1-0) 18-2 (08) 16-5 (1-3) 17-5 (1 1)

* p < 0-05.PASAT = Paced auditory serial addition test. Other abbreviations as for table 1.

Table 4 Performance of clinical and control groups on tests ofmemory (mean (SEM))

CFS Healthy MS DEP

Rey figure:Copy 66-3 (1 1) 69-0 (06) 68-2 (1-2) 67-2(1-2)Immediate recall 34-9 (2.8) 43-7 (1-9) 37-1 (3-4) 33-5 (4 0)Delay recall 34-5 (2-8) 40-9 (2-1) 37-7 (3 6) 31-8 (4-0)Forgetting 0-99 (0 04) 0-89 (0 05) 1-01 (0 04) 0-98 (0 06)

Logical memory:Immediate recall 24-3 (1-2) 28-5 (1-5) 22-8 (1-4)* 25-1 (2 0)Delay recall 20-4 (1-3) 25-0 (1-8) 18-0 (2-1)* 20-3 (2-1)Forgetting 0-83 (0 04) 0-86 (0 03) 0 79 (0 06) 0-78 (0 04)

CVLT:T score(trials 1-5) 34-6 (2.8)** 49 0 (2.2) 42-2 (3 2) 38-1 (3.6)List A trial 1 6-8 (0.4)** 8-8 (0 4) 7 9 (0 5) 7-4 (0 4)List A trial 5 12-1 (0.4)* 14-0 (0 4) 12-5 (0-5) 12-1 (0.7)*ListB 6-8(0-5) 7-5(04) 70(04) 7-1 (07)SD free recall 9-8 (0.6)* 12-3 (0 6) 9.7 (0 7)* 10-8 (0-9)SDcuedrecall 11-3(0-5) 13-1 (06) 12-1 (07) 12-2(0-7)LD free recall 10-6 (0.6)* 12-9 (0 6) 11-9 (0 8) 11-0 (0 8)LD cued recall 11-8 (0-5) 13-5 (0 6) 12-4 (0-6) 12-0 (0 7)Recognition 13-5 (0-6) 15-0 (0 3) 14-1 (0 6) 14-4 (0-5)

* p < 0-05; ** p < 0-01 v healthy controls.CVLT = California verbal learning test. Other abbreviations as for table 1.

complaints of attention differed significantlyacross groups (F (3,57) = 15-7, p < 0-001),with the chronic fatigue syndrome groupexpressing a significantly greater degree ofimpairment than either the multiple sclerosisor healthy control groups. The depressedgroup differed only from controls. For con-centration (F (3,57) = 27-0, p < 0-001), thechronic fatigue syndrome and depressedgroups did not differ from each other, butboth differed from the multiple sclerosisgroup and healthy controls. All three clinicalgroups showed increased complaints of mem-ory relative to controls (F (3,57) = 18-0, p <

0-001). The chronic fatigue syndrome groupexpressed significantly greater difficulties, inreasoning ability than any other groups (F(3,57) = 20-0, p < 0-0001).On the metamemory questionnaire, the

chronic fatigue syndrome group expressedsignificantly more subjective difficulties thanany of the other three groups on the atten-tion/prospective memory factor (F (3,68) =

16-1, p < 0-001). The depressed group dif-fered from both the multiple sclerosis andhealthy groups. For retrograde memory, theonly significant difference was between thechronic fatigue syndrome v healthy group (F(3,68) = 3-0, p < 0-04). On the anterogradememory factor, the chronic fatigue syndromegroup differed significantly from the multiplesclerosis and healthy groups, but not from thedepressed group (F (3,68) = 8-5, p < 0-001).No differences were seen across groups for thehistoric/overload memory factor (F (3,68) =1-4, p = 0-24).

DEPRESSION AND ANXIETYOn the STAI (table 2), the chronic fatiguesyndrome and depressed groups showed sig-nificantly higher scores than either the multi-ple sclerosis or healthy control groups, whichthemselves did not differ significantly (F(3,68) = 16-4, p < 0-001). The mean perfor-mance by the chronic fatigue syndrome groupwas not, however, in the clinically impaired

22range. The mean Beck depression inventoryscores for the chronic fatigue syndrome anddepressed groups were significantly higherthan the healthy control group (F (3,68) =16-4, p < 0-001). Whereas the multiple scle-rosis group scored significantly lower than thedepressed group on the Beck depressioninventory, they did not differ statistically fromcontrols.

ATTENTION AND CONCENTRATIONTable 3 gives the mean scores for attentionand concentration. There were no statisticallysignificant differences across the four groupsin digit span performance (F (3,68) = 0-46, p<0-71), trails A (F (3,68) = 0-007, p = 0-99)or trails B (F (3,68) = 1-9, p = 0-12). On thePASAT, the mean number of correctresponses was significantly lower in thechronic fatigue syndrome than in controls butdid not differ significantly from the multiplesclerosis or depressed groups. Mean PASATperformance in both the multiple sclerosisand depressed groups did not differ statisti-cally from healthy controls.To examine the potential influence of anxi-

ety on PASAT performance in patients withchronic fatigue syndrome, a median split ofthe STAI of the group was performed. Themean total PASAT score for the high and lowSTAI groups was 131-1 (SEM 9-2) and 1 17-0(SEM 8-0). This difference between the twogroups did not differ statistically (F (1,24) =

1-35, p = 0-26). To examine the influence ofdepression on PASAT performance, meanPASAT scores of the chronic fatigue syn-drome group were divided into high (> 13)and low (< 12) Beck depression inventory

40



j.com/

J Neurol N


ownloaded from



scores (n = 13 for each group). Mean PASATtotal scores for the high (1 19-3 (8 5)) and low(128-7 (9 0)) Beck depression inventorygroups did not differ and significantly (F(1,24) = 0-57, p = 0-45).

MEMORYTable 4 provides the mean scores on tests ofmemory. No statistically significant differ-ences were found between any of the groupsin the amount of material recalled on the Reycomplex figure in either the immediate (F(3,68) = 2-4, p = 0 07) or delayed (F (3,68) =1-5, p = 0.20) conditions. Thus visual mem-ory performance was not impaired in any ofthe clinical groups evaluated.

For verbal memory, the chronic fatiguesyndrome and depressed groups did not differfrom each other or from the healthy controlgroup in immediate and delayed recall onlogical memory. Only the multiple sclerosisgroup recalled significantly fewer elements onlogical memory in both the immediate anddelayed conditions than controls. Whendelayed recall was measured as a percentageof the amount of information initially learned,no statistically significant difference wasfound between the multiple sclerosis and con-trol conditions. This suggests that the mem-ory difficulty found in the multiple sclerosisgroup was in initial learning of material, andthat what information was initially learnedwas adequately retrieved.On a measure of the amount of information

learned across the five trials (trials 1-5) of theCVLT, the chronic fatigue syndrome grouprecalled a mean of 34-6 words, which was sig-nificantly below the mean of 49 0 words forthe healthy controls, suggesting difficulty inacquiring the material (F (3,68) = 5-2, p =0.01). Compared with controls, the chronicfatigue syndrome group recalled significantlyfewer words both during the initial learningtrial (list A trial 1) (F (3,68) = 4 9, p = 0-01)and the final learning trial (list A trial 5) (F(3,68) = 3.7, p = 0-01). Although thedepressed group and healthy controls did notdiffer in the amount of words recalled duringlist A trial 1, the depressed group recalled sig-nificantly fewer words (1 2 0) than controls bylist A trial 5. The multiple sclerosis group didnot differ from the healthy control group inthe amount of information learned.With respect to recall from long term stor-

age, the chronic fatigue syndrome group

Rating of overallneuropsychologicalimpairmentfor each group(mean (SEM)). SIR =

summed impairmentrating. CFS = chronicfatigue syndrome; MS =

multiple sclerosis; DEP =

depressed group.

C,,enCc

a

11

10

9

8

7

6

5

4

3

2

T

T

Controls MS EP

recalled significantly fewer words during boththe immediate (F (3,68) = 3-2, p = 0 02) anddelayed (F (3,68) = 2-6, p < 005) free recallconditions than the healthy controls. No dif-ference between the chronic fatigue syndromeand control groups were found in the cuedrecall conditions. The difference betweenchronic fatigue syndrome and controls in freebut not cued recall may suggest difficulties inretrieval from long term storage in the chronicfatigue syndrome group. Given that thepatients with chronic fatigue syndromelearned significantly fewer words than con-trols, a measure of the amount of informationrecalled relative to what was initially learnedmight be a more representative analysis of freeand cued recall effects. Thus free and cuedrecall were analysed as a percentage of theinformation recalled on list A trial 5. In thisanalysis, no differences were found betweenpatients with chronic fatigue syndrome andhealthy controls in the relative amount ofinformation recalled during both the free(t =-1-4, p<0.15) and cued (t=-1-2,p < 0 25) recall conditions. During free recall,the chronic fatigue syndrome group recalled81% of that initially learned compared with87% in healthy controls. During cued recall,87% and 92% recall was observed for thechronic fatigue syndrome group and healthycontrols respectively. The chronic fatigue syn-drome and healthy control groups did not dif-fer during recognition memory testing.

Overall, the multiple sclerosis group didnot differ from healthy controls on the CVLTmeasures of acquisition, recall, or recognition.The only exception was the significantlyreduced short delay free recall in the multiplesclerosis group compared with controls.

HIGHER COGNITIVE FUNCTIONSThere were no statistically significant differ-ences on the four tests of intellectual func-tions from the WAIS-R, or on the bookletcategory test. Thus higher level cognitivefunctioning did not differ between any of thefour groups.

OVERALL GROUP PROFILESAs well as contrasting the mean performanceon each neuropsychological measure acrossgroups, it was also of interest to obtain a sum-mary measure of overall neuropsychologicalimpairment. An impairment rating (IR) wascalculated based on SDs from the mean per-formance of the healthy control group. TheIR was calculated as follows: scores for eachtest measure within 1 SD of the control groupwere assigned an IR of 0; scores between 1and 2 SDs below the control group wereassigned an IR of 1; scores between 2 and 3SDs below that of controls were assigned anIR of 2; and scores greater than 3 SDsreceived an IR of 3. Raw scores for each of the21 test measures were transformed into an IRscore for each subject. The IRs were thensummed to obtain a summed IR (SIR) ofoverall performance for each subject. The SIRserved as the dependent variable for a 1 wayanalysis of variance (ANOVA) contrasting the

41



j.com/

J Neurol N


ownloaded from


DeLuca, Johnson, Beldowicz, Natelson

four groups. The mean SIR (figure) for thechronic fatigue syndrome group was 9 5(SEM = 1'6), for the depressed group 9-2(SEM 0 9), for the multiple sclerosis group6-8 (SEM 2-3), and for the healthy controls3-2 (SEM 2 0) (F (3,71) = 3 45, p = 0-021).Significant differences (F (3,68) = 3-4, p =0-02) were found between the healthy con-trols and both the chronic fatigue syndrome(p < 0-05) and depressed (p < 0-05) groupsin overall neuropsychological performance.

DiscussionThe results of this study replicate and extendour earlier work9 suggesting that patients withchronic fatigue syndrome show a selectiveimpairment in information processing of com-plex auditory material. One difficulty with thisinterpretation is the likelihood that the audi-tory (PASAT) and visual (trail making test)tasks used in the present study may not beequally difficult. Only auditory and visualtasks equated for difficulty will satisfactorilyanswer whether there is a general or selective(auditory) information processing deficit inchronic fatigue syndrome.

Another potential explanation for the dif-ference in performance between the PASATand the trail making test among patients withchronic fatigue syndrome may reflect differ-ences in the specific component of "attention"assessed. Intact performance on the trail mak-ing test suggests that patients with chronicfatigue syndrome show intact visuomotorscanning ability,20 whereas impaired PASATreflects impaired sustained selective process-ing of information.2' Future work must begeared towards specifically defining the natureof the "attentional" disorder in chronicfatigue syndrome.The second major finding was that patients

with chronic fatigue syndrome were generallynot impaired on tests of memory relative tocontrols. That is, the chronic fatigue syn-drome group did not differ from any of theother groups in visual memory and verbalparagraph recall. This general lack of a mem-ory impairment in chronic fatigue syndrome isconsistent with most recent work486 but notall.3The only exception to the lack of memory

impairment among patients with chronicfatigue syndrome was on the CVLT. The lackof a CVLT impairment in our earlier study5 isprobably attributed to our stringent inclusioncriteria and the resulting more homogeneoussample studied here. In the present study, thechronic fatigue syndrome group displayed sig-nificantly more difficulty in the acquisition ofinformation relative to healthy controls on theCVLT; retrieval of the information initiallylearned did not differ between the chronicfatigue syndrome group and any of the othergroups. Taken together, the memory datafrom the CVLT suggest that patients withchronic fatigue syndrome may have a subtledifficulty in learning verbal material (acquisi-tion), but show intact retrieval from long termstorage.

The reduced acquisition of information inthe chronic fatigue syndrome group on theCVLT likely reflects a reduced ability toprocess auditory information, rather than adeficit in storage, consolidation, or retrieval ofthe memory trace. Another potential explana-tion lies in the significantly higher depressionand anxiety among the chronic fatigue syn-drome group relative to healthy controls; thechronic fatigue syndrome and depressedgroups did not differ statistically in level ofdepression or anxiety. Interestingly, the over-all pattern of performance of the chronicfatigue syndrome group seemed more similarto the depressed group than any other groupon the CVLT. The possibility that depressionmay have influenced the CVLT performanceis supported by recent work, which found thatafter adjusting for increased depression, thechronic fatigue syndrome group no longer dis-played a significant verbal memory impair-ment.6The third major finding of the present

study was that the chronic fatigue syndromegroup did not differ from any of the othergroups on tasks of "higher order" cognitivefunctioning. Thus despite their subjectivecognitive complaints and increased level offatigue, depression, and anxiety, patients withchronic fatigue syndrome can perform higherlevel cognitive tasks at the level of healthycontrol subjects.The fourth major finding was that despite

the lack of memory or intellectual deficits onneuropsychological tests, the patients withchronic fatigue syndrome had the highestdegree of subjective complaints of cognitiveimpairments. One likely possibility for theperceived level of impaired higher cognitivefunctions among patients with chronic fatiguesyndrome is that they are confusing the diffi-culty in speed and efficiency of informationprocessing with memory and intellectual orproblem solving capability. This hypothesishas important implications for rehabilitation.Proper education and treatment may increasethe understanding of the nature of the cogni-tive difficulties in chronic fatigue syndrome,which may result in improved functional abil-ity.Our results replicate earlier work22 in show-

ing that the depressed group did not differ sig-nificantly from controls on the PASAT, trailmaking test, and digit span. These data sug-gest that depression alone does not signifi-cantly influence performance on tests ofcomplex attention, concentration, and infor-mation processing. Similar conclusions havebeen made from studying depression in othermedical populations including those withspinal cord injury23 and HIV.24 Therefore, thesignificant difference between chronic fatiguesyndrome and controls on the PASAT foundin the present study cannot simply be attrib-uted to depression. Thus consistent with thesuggestion of Wessely,25 the present data showthat whereas depression may play a part inexplaining the cognitive dysfunction found inchronic fatigue syndrome, it cannot explain allthe variance.

42



j.com/

J Neurol N


ownloaded from



Although performance on individual neu-ropsychological tests produced few significantdifferences across groups, the measure ofoverall neuropsychological performance didshow significant differences between thechronic fatigue syndrome and depressedgroups and controls. The multiple sclerosisgroup did not differ statistically from anyother group. These data suggest that "subtle"cognitive deficits can be seen in subjects withchronic fatigue syndrome and depressionwhen overall performance of a comprehensivetest battery is taken into account. Overall per-formance shows that chronic fatigue syn-drome and depressed patients share milddeficits in effortful cognition.

Difficulties in attention and speed of pro-cessing have been correlated with white mat-ter changes in healthy elderly subjects,26 andpatients with multiple sclerosis27 and late stageLyme disease.28 Interestingly, the only threeMRI studies examining patients with chronicfatigue syndrome to date29"3 show that whenMRI findings are present in chronic fatiguesyndrome, white matter changes predomi-nate. Taken together, it can be hypothesisedthat the impaired information processing abil-ity in at least a subgroup of patients withchronic fatigue syndrome may be due tochanges in cerebral white matter.

Impaired attention and information pro-cessing speed, however, have also been associ-ated with depression and anxiety disorderssuch as post-traumatic stress disorder.32 Yet,heightened "state" or "test" anxiety itself isnot necessarily associated with decreased neu-ropsychological performance.32 Further,patients with major affective disorder requir-ing admission to hospital show similar MRIfindings to those observed in chronic fatiguesyndrome.33 Clearly, further studies arerequired to clearly delineate the nature of theinformation processing deficit in chronicfatigue syndrome, and to determine its simi-larity to or difference from that seen indepression.We thank Drs John Ottenweller and Steven Ellis for statisticalconsultation and Mr John Duryee for help in data manipula-tion and manuscript preparation. This study was supported inpart by grant AI-32247 from NIH, establishing a ChronicFatigue Syndrome Research Center at UMDNJ-New JerseyMedical School, Newark NJ; and NIH grant H133P10002-92to UMDNJ-New Jersey Medical School, and Departnent ofEducation, Rehabilitation Services Administration grantH129J10006 to Kessler Institute.

1 Holmes GP, Kaplan JE, Gantz NM, et al. Chronic fatiguesyndrome: a working case definition. Ann Intern Med1988;108:387-9.

2 Riccio M, Thompson C, Wilson B, Morgan DJ, LarientAF. Neuropsychological and psychiatric abnormalities inmyalgic encephalomyelitis: A preliminary report. Br 7Clin Psychol 1992;31:1 11-20.

3 Sandman CA, Barron JL, Nackoul K, Goldstein J, FidlerF. Memory deficits associated with chronic fatigueimmune dysfunction syndrome. Biol Psychiatry 1993;33:618-23.

4 Grafman J, Schwartz V, Dale JK, Scheffers M, Houser CStraus SE. Analysis of neuropsychological functioning in

patients with chronic fatigue syndrome. J NeurolNeurosurg Psychiatry 1993;56:684-9.

5 Johnson SK, DeLuca J, Fiedler N, Natelson BH.Cognitive functioning in chronic fatigue syndrome. ClinInfect Dis 1994;18(suppl 1):S84-5.

6 Krupp LB, Sliwinski M, Masur DM, Friedberg F, CoylePK. Cognitive functioning and depression in patientswith chronic fatigue syndrome and multiple sclerosis.Arch Neurol 1994;51:705-10.

7 Prasher D, Smith A, Findley L. Sensory and cognitiveevent related potentials in myalgic encephalomyelitis.J NeurolNeurosurg Psychiatry 1990;1:253-7.

8 Scheffer MK, Johnson R, Grafinan J, Dale JK, Straus SE.Attention and short-term memory in chronic fatiguesyndrome patients: An event-related potential analysis.Neurology 1992;42:1667-75.

9 DeLuca J, Johnson SK, Natelson BH. Information pro-cessing in chronic fatigue syndrome and multiple sclero-sis. Arch Neurol 1993;50:301-4.

10 Schluederberg A, Straus SE, Peterson P, et al. Chronicfatigue syndrome: Definition and medical outcomeassessment. Ann Intern Med 1992;117:325-31.

11 Poser CM, Paty DW, Scheinberg L, McDonald WI, DavisFA, Toutellotte WW. New diagnostic criteria for multiplesclerosis: guidelines for research protocols. Ann Neurol1983;13:227-31.

12 Krupp LB, LaRocca NG, Muir-Nash J, Steinberg AD.The fatigue severity scale: Application to patients withmultiple sclerosis and systemic lupus erythematosus.Arch Neurol 1989;46:1121-3.

13 Wechsler D. Wechsler adult inteUigence scale-revised, manual.New York: Psychological Corporation, 1981.

14 Wechsler D. Wechsler memory scale-revised, manual. NewYork: Psychological Corporation, 1987.

15 Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J.An inventory for measuring depression. Arch GenPsychiatry 1961;4:561-71.

16 Spielberger CD, Gorsuch R, Lushene R. The state traitanxiety inventory (STAI) test manual. Palo Alto,California: Consulting Psychologists Press, 1983.

17 Mateer CA, Sohlberg MM, Crinean J. Focus on clinicalresearch: Perception of memory function in individualswith closed head injury. Journal of Head TraumaRehabilitation 1987;2:78-84.

18 Kirk A. Experimental design: procedures for the behavioralsciences. 2nd ed. Belmont, CA: Brooks-Cole.

19 DeLuca J, Johnson SK. Cognitive impairments in multiplesclerosis: implications for rehabilitation. NeuroRehabilitation 1993;3:9-16.

20 Shum DHK, McFarland KA, Bain JD. Construct validity ofeight tests of attention: comparison ofnormal and closedhead injured samples. The Clinical Neuropsychologist1990;4: 151-62.

21 Gronwall D. Paced auditory serial addition task: A mea-sure of recovery from concussion. Perceptual Motor Skills1977;44:367-73.

22 Channon S, Baker J, Robertson M. Working memory inclinical depression: an experimental study. Psychol Med1993;23:87-91.

23 Davidoff GN, Roth EJ, Richards JS. Cognitive deficits inspinal cord injury: epidemiology and outcome. Arch PhysMed Rehabil 1992;73:275-84.

24 Bomstein RA, Pace P, Rosenberger P, et al. Depressionand neuropsychological performance in asymptomaticHIV infection. AmJPsychiatry 1993;150:922-7.

25 Wessely S. Chronic fatigue syndrome. Jf Neurosurg NeurolPsychiatry 1991;54:669-71.

26 Ylikoski R, Ylikoski A, Erkinjuntti T, Sulkava R, RaininkoR, Tilvis R. White matter changes in healthy elderly per-sons correlate with attention and speed of mental pro-cessing. Arch Neurol 1993;50:818-24.

27 Rao SR Neurobehavioral aspects of multiple sclerosis. NewYork: Oxford University Press, 1990.

28 Halperin nJ, Pass HL, Anand AK, Luft BJ, Volkman DJ,Dattwyler RJ. Nervous system abnormalities in Lymedisease. Ann NYAcad Sci 1988;539:24-34.

29 Natelson BH, Cohen JM, Brassloff I, Lee HJ. A controlledstudy of brain magnetic resonance imaging in patientswith fatiguing illnesses. J7 Neurol Sci 1993;120:213-7.

30 Buchwald DPR, Cheney DI, Peterson B, et al. A chronicillness characterized by fatigue, neurologic and immuno-logic disorders and active herpesvirus type 6 infection.Ann Inten Med 1992;116:103-13.

31 Schwartz RB, Garada BM, KomaroffAL, et al. Detection ofintracranial abnormalities in patients with chronicfatigue syndrome: Comparison of MRI imaging andSPECT. American Roentgen Ray Society 1994;162:935-41.

32 Puente E, McCaffrey RJ. Handbook of Neuropsychologicalassessment: a biopsychosocial perspective. New York:Plenum Press, 1992.

33 Swayze VW, Andreasen NC, Alliger RJ, Ehrhardt JC, YuhWC. Structural brain abnormalities in bipolar affectivedisorder. Arch Gen Psychiatry 1990;47:1054-9.

43



j.com/

J Neurol N


ownloaded from


Documents

Neuropsychological impairments fatigue syndrome, multiple ... · The Beck depression inventory (BDI)15 and the state-trait anxietyscale (STAI)16werealso given. In addition, a metamemory