B ritish Journal of Neurosurgery 1999;13(1):46± 51

ORIGINAL ARTICLE

Anxiety and depression in patients with an intracranial neoplasm

before and after tumour surgery

A-M. PRINGLE, R. TAYLOR & I. R. WHITTLE

Department of Clinical Neurosciences,Western General Hospital, Edinburgh, UK

Abstract

The aims of this prospective study were to investigate levels of anxiety and depression in patients with a solitary intracranialneoplasm before and after surgery, and to determine if relationships exist between high levels of anxiety or depression andthe hemispheric location of the tumour, the tumour type or patient gender. Patients aged between 17 and 79 years with asolitary intracranial neoplasm completed the Hospital Anxiety and Depression Scale (HAD) before and after biopsy orresective tumour surgery. A control group of non-brain-damaged subjects also completed the HAD before and after lumbarspinal surgery. Of the 109 patients with a brain tumour 30 and 16% demonstrated the likely presence of anxiety anddepression, respectively, according to HAD scoring criteria. A greater proportion of females with a left hemisphere tumourreported higher levels of emotional disturbance than any other group of patients; relationships between dysphasia and levelsof anxiety or depression were not signi® cant. Patients with a meningioma had higher levels of anxiety and depression asmeasured by the HAD than those with any other tumour types. Levels of both anxiety and depression were signi® cantlylower after tumour surgery according to the HAD. There were no signi® cant differences in HAD scores between (a) left andright hemispheric tumour groups, and (b) the tumour and control (n = 20) groups. This study has found that anxiety anddepression as measured by the HAD are relatively uncommon in patients with an intracranial neoplasm, and that levels ofmood disturbance do not differ signi® cantly from those in patients undergoing lumbar spinal surgery. Levels of anxiety anddepression become lower after surgery in patients with a brain tumour. Patterns of anxiety and depression in patients with abrain tumour appear to differ from those reported in stroke.

Key words: B rain tumour, Hospital Anxiety and Depression Scale, mood disorders.

Introduction

Disturbances of mood or affect have been identi® ed

as common sequelae of stroke and other brain

diseases.1 ± 3 In some stroke patients such disorders

may be severe, often occurring 6 months after stroke

and persisting for up to 2 years.1 A relationship

between location of stroke and mood change has

been demonstrated by some workers: patients with

left anterior strokes were identi® ed as predominantly

depressed, while patients with right anterior strokes

were often inappropriately cheerful or apathetic.1 ,4

In contrast little has been published on the relation-

ship between brain tumours and mood disorders.3 As

there are fundamental differences in the pathophysi-

ology and epidemiology of stroke and intracranial

tumours, there are also likely to be different neuro-

physiological and psycholog ical reactions in the

different patient cohorts.3 In addition, most stroke

stud ies have fo cused on depression while the

com m onest sym ptom of m ood d isturbance in

patients with an intracranial tumour may be anxiety.5

The aims of the present study were therefore to

evaluate prospectively levels of anxiety and depres-

sion as measured by the Hospital Anxiety and Depres-

sion Scale (HAD)6 in patients presenting with a

solitary supratentor ial intracranial neoplasm ; to

investigate the acute effects of tumour surgery on

these levels; and to determine whether hemispheric

location, type of tumour or patient gender were associ-

ated with type or extent of mood disturbance.

Methods

Patient cohort

Patients aged between 17 and 79 years, with a solitary

supratentorial intracranial tumour demonstrated by

CT or MRI were recruited following admission to

the Department of Clinical Neurosciences (DCN).

Language function was assessed as par t of an

additional study. Exclusion criteria were: ® rst language

Correspondence to: Dr A-M. Pringle, Department of Clinical Neurosciences, Western General Hospital, Crewe Road, Edinburgh EH42XU, UK.

Received for publication 8th June 1998. Accepted 30th June 1998.

0268-869 7/99/010046 ± 06 $9.50 ½ The Neurosurgical Foundation

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.

not English; uncorrected impairments of hearing and

vision; or past medical history of psychiatric disorder,

alcoholism, head injury or other known brain disease.

Patien ts were a lso excluded if they were not

sufficiently well to cope with formal testing. The

cohort was essentially a consecutive series of brain

tumour patients with no apparent selection bias.

Tumour neuropathology was categorized according

to the WHO classi ® cation.

Twenty non-brain-damaged patients admitted for

elective lumbar spinal surgery were also assessed to

provide a measure of psychological reaction to the

stress of admission and operation. These control

subjects were matched for age and educational level.

The study was approved by the appropriate ethics

committee, and all patients and subjects gave written

consent to participate.

Mood assessment

The HAD was used to evaluate mood. It was designed

for use in non-psychiatric hospital departments to

screen for both anxiety and depression. By eliminating

questions relating to physical disorder the authors of

the HAD attempted to obtain a relatively `pure’ mood

score uncontaminated by pr imary symptoms of

physical illness (which can overlap with physical

symptoms of depression and anxiety). The instruc-

tion to patients to complete the questionnaire with

reference to how they have been feeling in the past

week is particularly appropriate to the patient group

in the study who have often been aware of impending

brain surgery and having a potentially life-threatening

condition for only a short period of time. Fourteen

items are included, seven of which relate to anxiety

and seven to depression. According to the authors’

guidelines a score of less than 8, between 8 and 10,

and higher than 10 on each scale were taken to

indicate, respectively, the probable absence, the

possible presence, and the probable presence of

anxiety or depression. Questionnaires were completed

prior to tumour biopsy or resective surgery when

language function was also being assessed, and before

and after lumbar spinal surgery in the control group.

Scores were recorded for both anxiety and depres-

sion. Data were not recorded for patients whose

language comprehension was found during testing to

be too impaired to permit reliable completion of the

scale. Where dyslexic difficulties prevailed each item

was read aloud to the patient and where language

difficulties were severe the HAD was not used. This

was not thought likely to skew results greatly as only

six patients were excluded on the grounds of severe

comprehension de® cit.

Statistical analysis

Paired and unpaired t-tests, and Pearson r correla-

tions were used as appropriate.

Results

Patients

Between September 1992 and June 1995, 109 patients

completed the HAD before biopsy or resection of an

intracranial tumour. Patients were assessed a mean of

6.7 days (median 7 days) after radiological diagnosis

of an intracranial neoplasm . The majority were

receiving dexamethasone (mean 8.1 mg/day). Of these

109 patients, 56 had a left and 49 had a right

hemisphere tumour.The HAD scores of four patients

which could not be localized by hemisphere were

omitted from the analysis of possible relationships

between emotional disturbance and hemispheric loca-

tion. Neuropathological diagnoses included 32 glio-

blastoma muliforme (GBM), 22 anaplastic astrocytoma

(AA) or anaplastic oligodendroglioma (AO), 14 astro-

cytoma or oligodendroglioma, 17 meningioma and

20 metastatic tumours.Two patients had a pineoblas-

toma and one each had a craniopharyngioma and

glioneuronal hamartom a. Following surgery, 94

patients were reassessed on the HAD. Of the 15 who

were not reassessed the clinical condition of ® ve

patients had deteriorated markedly after surgery; four

patients were lost to follow-up; three patients failed

to return their HAD questionnaires; two patients did

not have the planned surgical intervention and were

therefore assessed on only one occasion; and one

patient refused to complete the HAD after surgery.

The second assessment was conducted prior to

discharge, usually 7 days after initial evaluation.

Preoperative HAD scores in patients with a supratento-

rial intracranial neoplasm

The ® ndings on the HAD scale for anxiety are shown

in Table I. Fifty-one per cent of the cohort had scores

suggesting no signi® cant anxiety (i.e. 0 ± 7); 19% were

possibly anxious (8± 10); and 30% were probably

anxious (11 or more). Mean scores for depression are

shown in Table II. Sixty-six per cent of patients had

scores suggesting no signi® cant depression; 18%

obtained scores indicative of possible depression; and

16% had scores suggesting the likely presence of

depression.

A greater number of female patients obtained scores

indicating the probable presence of either anxiety or

depression, with the group of female patients with a

tumour of the left hemisphere containing the highest

percentage of patients demonstrating possible or likely

mood disturbance. As a group, female patients had

signi® cantly higher scores for anxiety, depression and

total HAD score (all p < 0.001) compared with males.

However, when tumour laterality was included in the

gender analysis only females with a left-sided tumour

had higher levels of anxiety, depression and total HAD

score (all p < 0.001) than males. The differences

between male and female patients with a tumour of

the right hemisphere did not reach signi ® cance for

any HAD score.

Anxiety and depression in intracranial tumours 47

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.

Patien ts w ith a m en ing ioma obtained higher

scores for both anxiety and depression than groups

of patients with other types of brain tumour (Table

III), desp ite there being no difference between

groups in the length of time between initial scans

and diagnosis (a mean of 6.6 days, median 7 days,

for patients with a glioma; mean 7.1 days, median 6

days, for patients with a meningioma), and preop-

erative assessment. The majority of patients with

a m en ing iom a were fem ale (1 2 out o f 17).

Mean anxiety and depression scores obtained by

patients with a glioblastoma were higher than those

obtained by patients with either anaplastic glioma,

low grade glioma or cerebral metastases, but were

not high enough to indicate probable emotional

disturbance.

The HAD scores did not correlate signi® cantly

with steroid dosage at initial assessment. There was

also no signi® cant correlation between any language

score (Aphasia Quotient and Language Quotient

derived from the Western Aphasia Battery7 or the

score for word ® nding obtained from the Boston

Naming Test8 and anxiety, depression or HAD total

at initial assessment.

TABLE I. Mean HAD scores and numbers of patients (also shown in percentages) whose scores indicate probable absence(score <8), possible presence (score between 8 and 10) or probable presence of anxiety (score >10) prior to surgery for asolitary supratentorial intracranial neoplasm. The scores of four patients with midline or central tumours have been omittedfrom the analysis according to side of lesion

Mean(SD)

Probable absence(% of total)

Possible presence(% of total)

Probable presence(% of total)

Whole group (n = 109) 8.3 55 21 33

(4.7) (51%) (19%) (30%)

All left hemisphere (n = 56) 8.7 28 10 18(5.0) (50%) (18%) (32%)

All right hemisphere (n = 49) 7.9 26 9 14

(4.6) (53%) (18%) (29%)All male patients (n = 62) 6.8 40 13 9

(4.0) (64%) (21%) (15%)

All female patients (n = 47) 10.3 15 8 24(4.9) (32%) (17%) (51%)

Male left hemisphere (n = 33) 6.7 22 8 3

(3.8) (67%) (24%) (9%)Female left hemisphere (n = 23) 11.7 6 2 15

(5.0) (26%) (9%) (65%)

Male right hemisphere (n = 28) 7.0 17 5 6(4.4) (61%) (18%) (21%)

Female right hemisphere (n = 21) 9.0 9 4 8

(4.8) (43%) (19%) (38%)

TABLE II. Mean HAD scores and numbers of patients (also shown in percentages) whose scores indicate probable absence(score <8), possible presence (score between 8 and 10) or probable presence of depression (score >10) prior to surgery fora solitary supratentorial intracranial neoplasm. The scores of four patients with midline or central tumours have beenomitted from the analysis according to side of lesion

Mean(SD)

Probable absence(% of total)

Possible presence(% of total)

Probable presence(% of total)

Whole group (n = 109) 5.6 72 20 17

(4.1) (66%) (18%) (16%)

All left hemisphere (n = 56) 6.0 36 8 12(4.4) (64%) (14%) (22%)

All right hemisphere (n = 49) 5.2 34 10 5

(3.9) (69%) (21%) (10%)All male patients (n = 62) 4.4 49 6 7

(3.8) (79%) (10%) (11%)

All female patients (n = 47) 7.3 23 14 10(4.0) (49%) (30%) (21%)

Male left hemisphere (n = 33) 4.5 26 2 5

(4.1) (79%) (6%) (15%)Female left hemisphere (n = 23) 8.2 10 6 7

(4.0) (44%) (26%) (30%)

Male right hemisphere (n = 28) 4.4 22 4 2(3.5) (79%) (14%) (7%)

Female right hemisphere (n = 21) 6.2 12 6 3

(4.2) (57%) (29%) (14%)

48 A-M . Pringle et al.

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.

Postoperative HAD scores in patients with a supratento-

rial intracranial neoplasm

At the time of second assessment following either

stereotactic biopsy, or craniotomy and resection of

intracranial tumour there were reductions in levels of

both anxiety and depression, in both males and

females, whether they had left (anxiety: p < 0.001;

depression: p = 0.001) or right (anxiety: p = 0.004;

depression: p = 0.028) hemispheric tumours. Scores

obtained by patients who had tumour biopsy did not

differ signi® cantly from those who underwent crani-

otomy and tumour resection. After surgery there were

reductions in the percentages of patients with likely

anxiety (30 to 16%) and depression (16± 6%), and in

mean daily dexamethasone dosage (preoperative:

8.1 mg/day; postoperative: 1.4 mg/day; p < 0.001).

Change in HAD scores (from ® rst to second assess-

ment) did not correlate signi® cantly with change in

steroid dosage.

HAD scores prior to elective lumbar spinal surgery

The presurgery HAD scores obtained by the control

group for anxiety and depression are shown in Tables

IV and V, respectively. The majority of subjects

obtained scores suggesting either probable absence

or possible presence of emotional disturbance before

surgery.There were no signi® cant differences in scores

between the control group and any of the patient

groups. The differences in both anxiety and depres-

sion that were present between male and female

tumour patients is not apparent in the control group,

although the size of the latter group is substantially

smaller. Postoperative HAD scores did not differ

signi® cantly from scores obtained at initial assess-

ment.

Discussion

This study has assessed anxiety and depression using

the HAD Scale in a cohort of inpatients awaiting

cranial tumour surgery. The HAD was chosen for the

reasons outlined in the methods section and is, in

addition, easy and quick to administer and can be

used in patients with a dominant hem ispher ic

neoplasm even though approximately half may have

some language impairment.9 ,1 0 It is an index of how

the patient felt in the past week: in many of these

patients the radiological or clinical diagnosis was not

known at the time of ® rst assessment, but the patients

wou ld undoubted ly have been aware of the ir

symptoms and the possible implications of these.

Using an extensive, structured interview to assess

mood would have been inappropriate in this cohort

as the mood disorder assessment was only a small

pa r t of a large r s tudy invest igat ing langua ge

disorders in brain tumour patients.9 ± 1 2 Addition-

ally, as items in the HAD assess fatigue or `feeling

slowed down’ , and inability to enjoy activities, which

may frequently be endorsed in medically ill patients,

one might have expected a slight bias towards high

rather than low HAD scores.

Given the above constraints and limitations, this

study has demonstrated that, according to HAD

scoring criteria, relatively low levels of either prob-

able anxiety (30%) or probable depression (16%) are

found in patients with a solitary supratentorial intrac-

ranial neoplasm admitted to a clinical neuroscience

depar tment and that their HAD scores did not differ

signi® cantly from a group of patients admitted to the

same department for elective lumbar spinal surgery.

It has been estimated that between 9 and 19% of the

general population could be diagnosed as actual or

borderline cases of psychiatric disorder.13 Although

TABLE III. Preoperative HAD scores for 109 patients with a solitary supratentorial tumour according to tumour type

GBM(n = 32)

AA or AO(n = 22)

Glioma(n = 14)

Mening(n = 17)

Metast(n = 20)

Others(n = 4)

Mean anxiety 8.7 7.6 5.7 11.2 7.4 9.8Range of scores 1± 19 11± 18 1± 9 2± 19 2± 17 6± 14Mean depression 5.8 4.2 4.1 7.9 5.2 7.0Range of scores 0± 15 0± 13 0± 10 0± 14 1± 14 1± 11

Abbreviations: GBM = glioblastoma; AA = anaplastic astrocytoma; AO = anaplastic obligodendroglioma; glioma= astro-cytoma, oligodendroglioma or mixed oligoastrocytoma; Mening = meningioma; Metast = cerebral metastasis; other =craniopharyngioma, glioneural hamartoma and pineoblastoma.2

TABLE IV. Mean HAD scores obtained prior to elective lumbar spinal surgery for 20 control subjects whose scores indicateprobable absence (score <8), possible presence (score between 8 and 10) or probable presence of anxiety (score >10)

Mean (SD) Probable absence(% of total)

Possible presence(% of total)

Probable presence(% of total)

All control group (n = 20) 7.9 10 6 4(3.9) (50%) (30%) (20%)

Male control group (n = 10) 7.4 4 4 2(3.1) (40%) (40%) (20%)

Female control group (n = 10) 8.3 6 2 2(4.7) (60%) (20%) (20%)

Anxiety and depression in intracranial tumours 49

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.

such ® gures will depend on the assessment measures

used, the patients in this study appeared to be remark-

ably well-adjusted emotionally. A plausible interpreta-

tion of these results is that since the diagnosis of

brain tumour was very recent at the time of initial

HAD assessment emotional reaction to the diagnosis

may not have had sufficient time to evolve.

Female patients with an intracranial tumour had

higher levels of anxiety and depression than male

patients, and females with a left-sided intracranial

tumour obtained higher HAD scores than females

with right sided tumours. Although these ® ndings are

similar to those reported by Irle and colleagues3 it is

surprising that such disparity should exist between

male and female patients with a tumour of the left

hemisphere when HAD scores were relatively similar

for males and females with a right hemispheric

tumour, and the male and female subjects in the

control group. It might be expected that dysphasic

patients would experience greater levels of emotional

distress than those with normal language. In this

study, however, using assessments administered at

the same time pre- and postoperatively as the HAD,

language ability in the female patients did not differ

signi® cantly from that of the males and, in fact, the

trend was for language function in the male group to

demonstrate slightly greater impairment. In addition,

correlational analysis did not identify signi® cant

relationships between HAD and language scores, in

contradiction of the ® ndings of one study of stroke

patients which demonstrated a relationship between

acute aphasia and depression.4 A number of patients

with a right-sided intracranial tumour had been

diagnosed as depressed by their general practitioner

prior to admission and tumour diagnosis. In view of

the low levels of depression identi® ed using the HAD,

it is likely that the preliminary clinical diagnosis of

depression had been in¯ uenced by observation of the

signs of disordered non-verbal communication, such

as reduced facial expression, few alterations in intona-

tion and poor eye contact, which often accompany

right hemisphere disease.14 ± 16 These clinical observa-

tions are also at variance with indications that patients

with a left-sided anterior stroke may be depressed,

while those with a right anterior stroke may be inap-

propriately cheerful or apathetic.1 ,4 Findings con-

cerning mood disorders in stroke are often

contradictory, however, as House and colleagues17 did

not ® nd any link between mood symptoms and

hemispheric location of the stroke and concluded in

another large study1 8 that undue emphasis had been

placed on mood disorders that largely resolved after

12 months. Variations in ® ndings of mood disorders

in brain disease may also be in¯ uenced by the applica-

tion of different assessment and diagnostic criteria,

where patients displaying episodic emotionalism or

lability are categorized as depressed in one series of

studies but not in another. Differences in patient

selection, particularly as regards ratios of hospital

inpatients to outpatients18 may also bias results in

different series.

The type and grade of tumour did not have as

much impact on preoperative HAD scores as might

be expected. Paradoxically, patients with the most

favourable prognosis, i.e. those with a meningioma,

appeared to demonstrate the highest levels of both

anxiety and depression. It is possible that the relatively

high ratio of female patients with a left hemispheric

meningioma might have had some impact on the

mean scores as there was no other obvious difference

between patients with a meningioma and those with

a glioma in the length of time intervening between

diagnosis and preoperative HAD assessment, or in

the neurosurgical and nursing staff involved in patient

care. The lack of correlation between language and

HAD scores does not suggest any systematic relation-

ship between level of distress and (i) frustration caused

by language disturbance or (ii) lack of insight as a

component of mental impairment.

Following surgery there were signi® cant reduc-

tions in the levels of anxiety and depression as

measured by the HAD in patients with an intracra-

nial tumour.This is somewhat surprising as over 67%

of patients in the brain tumour cohort would have

been informed that they had a malignant intracranial

tumour. Again, it is possible that the diagnosis of

brain tumour was too recent at the time of re-

assessment for emotions to have had sufficient time

to evolve. Irle et al.3 showed that depressive states

emerged in some oncological patients either during

radiotherapy or at home several weeks after diagnosis.

Nonetheless, in a separate study of mood disorders

in brain tumour patients undergoing chemotherapy,19

few patients obtained HAD scores suggestive of

possible or likely mood disorders despite their

questionnaires being completed in the knowledge that

TABLE V. Mean HAD scores obtained prior to elective lumbar spinal surgery for 20 control subjects whose scores indicateprobable absence (score <8), possible presence (score between 8 and 10) or probable presence of depression (score >10)

Mean (SD) Probable absence(% of total)

Possible presence(% of total)

Probable presence(% of total)

All control group (n = 20) 5.6 14 4 2(3.0) (70%) (20%) (10%)

Male control group (n = 10) 4.7 8 1 1(3.2) (80%) (10%) (10%)

Female control group (n = 10) 6.5 6 3 1(2.7) (60%) (30%) (10%)

50 A-M . Pringle et al.

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.

the disease had recurred. Patients with likely mood

disorders as assessed by the HAD were also in the

minority amongst a brain tumour group being treated

with radiotherapy.19 A link has been identi® ed in

studies of other cancer groups between physical status

and psychological distress.20 ,2 1 Persistent pain or

physical disability was relatively uncommon among

the patients in the present study, perhaps contributing

to the relatively low HAD scores.

Response bias is a recognized problem with self-

rating scales such as the HAD; therefore, it must be

acknowledged that some patients may have opted for

socially acceptable rather than accurate responses,

possibly in¯ uencing reported levels of mood distur-

bance in the whole group or in subgroups of patients.

Patients were usually well prepared by medical and

nursing staff in advance of treatment; it is likely that

explanation and discussion would have contributed

to a reduction in apprehension during the inpatient

stay. In general, however, the patients with an intrac-

ranial tumour in this study demonstrated very little

disturbance of mood as assessed by the HAD in the

perioperative period, with scores for anxiety and

depression very often within normal limits.

Acknowledgements

This study was supported by a grant from the

Disability and Continuing Healthcare Committee of

the Scottish Office.

References

1 Robinson RG, Kubos KL, Starr, LB, et al. Mooddisorders in stroke patients. Importance of location oflesions. B rain 1984;107:81 ± 93.

2 Pimental PA, Kingsbury NA. Neuropsychological Aspects

of Right B rain Injury. Austin, TX: Pro-Ed, 1989.3 Irle E, Peper M, Wowra B, et al. Mood changes after

surgery for tumours of the cerebral cortex. Arch Neurol

1994;51:164± 74.4 Robinson RG, Benson DF. Depression in aphasic

patients: frequency, severity and clinical-pathologicalcorrelations. B rain Lang 1981;14:282± 91.

5 Greer S, Moorey S, Baruch JDR, et al. Adjuvantpsychological therapy for patients with cancer: aprospective randomised trial. B MJ 1992;304:675 ± 80.

6 Zigmond AS, Snaith RP. The Hospital Anxiety and

Depression Scale. Acta Psychiat Scand 1983;67:361 ± 70.7 Kertesz A. The Wester n Aphasia B attery. San Antonio,

TX: The Psychological Corporation, 1982.8 Kaplan E, Goodglass H, Weintraub S. B oston Naming

Test. Philadelphia, PA: Lea & Febiger, 1983.9 Whittle IR, Pringle A-M, Taylor R. The effects of

resective surgery for left sided intracranial tumourson language function: a prospective study. Lancet

1998;351:1014 ± 18.10 Thomson A-M, Taylor R, Fraser D, Whittle IR. Stere-

otactic biopsy of nonpolar tumours in the dominanthemisphere: a prospective study of effects on languagefunctions. J Neurosurg 1997;86:923± 6.

11 Thomson A-M, Taylor R, Fraser D, Whittle IR. Theutility of the Right Hemisphere Language Battery inpatients with brain tumours. Eur J Disorders Comm

1997;32:325 ± 32.12 Pringle A-M, Taylor R, Whittle IR. Assessment of

communication impairment and the effects of resectivesurgery in solitary, right sided supratentorial intracra-nial tumours: a prospective study. B r J Neurosurg 1998;12:423± 429.

13 Freeman CP. Neurotic disorders. In: Kendall RE,Zealley AK, eds, Com panion to Psychiatr ic S tudies.Edinburgh: Churchill Livingstone, 1983.

14 Burns MS, Halper AS, Mogil SI. Clinical Management

of R igh t H em isphere Dysfunction . Rockville: AspenPublications, 1985.

15 Joanette Y, Goulet P. Right hemisphere and verbalcommunication: conceptual, methodological and clinicalissues. In: Prescott TE, ed., Clinical Aphasiology, Vol.22. Austin: Pro-Ed, 1992.

16 Tompkins CA. R igh t H em isphere C om mun ica tion

Disorde rs: Theor y and M anagement. USA: SingularPublishing Group, Inc, 1995.

17 House A, Dennis M, Warlow C, et al. Mood disordersafter stroke and their relation to lesion location. B rain

1990;113:1113± 29.18 H ouse A, Dennis M, M ogr idge L, et a l. Mood

disorders in the year after ® rst stroke. B r J Psych iatr y

1991;158:83± 92.19 Thomson A-M. Communication disorders in patients

with a hemispheric intracranial neoplasm. PhD Thesis,University of Edinburgh, 1997.

20 Cella DF, Oro® amma B, Holland JC, et al.The relation-ship of psychological distress, extent of disease andperformance status in patients with lung cancer. Cancer

1987;60:1661 ± 7.21 Kaasa S, Malt U, Hagen S, et al. Psychological distress

in cancer with patients with advanced disease. Radio-

ther Oncol 1993;27:193 ± 7.

Anxiety and depression in intracranial tumours 51

Br

J N

euro

surg

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

The

Uni

vers

ity o

f M

anch

este

r on

12/

02/1

4Fo

r pe

rson

al u

se o

nly.