Tilburg University

Cognitive status and quality of life in patients with suspected versus proven low-gradegliomasReijneveld, J.C.; Sitskoorn, M.M.; Klein, M.; Nuyen, J.; Taphoorn, M.J.B.

Published in:Neurology

Publication date:2001

Link to publication

Citation for published version (APA):Reijneveld, J. C., Sitskoorn, M. M., Klein, M., Nuyen, J., & Taphoorn, M. J. B. (2001). Cognitive status andquality of life in patients with suspected versus proven low-grade gliomas. Neurology, 56(5), 618-623.

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

- Users may download and print one copy of any publication from the public portal for the purpose of private study or research - You may not further distribute the material or use it for any profit-making activity or commercial gain - You may freely distribute the URL identifying the publication in the public portal

Take down policyIf you believe that this document breaches copyright, please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 13. Jul. 2020

Cognitive status and quality of life inpatients with suspected versus

proven low-grade gliomasJ.C. Reijneveld, MD; M.M. Sitskoorn, PhD; M. Klein, PhD; J. Nuyen; and M.J.B. Taphoorn, MD, PhD

Article abstract—Background: The preferred management of patients with suspected low-grade gliomas (S-LGG) re-mains controversial. The benefits of resection or radiotherapy early in the course of the disease have not been proven interms of survival. Little is known about the effects of early therapy on quality of life (QOL) and cognitive status. Theauthors compared functional status, QOL, and cognitive status of patients suspected of having a LGG, in whom treatmentwas deferred, and patients with proven LGG (P-LGG), who underwent early surgery. Methods: The authors recruited 24patients suspected of having an LGG. These patients were matched with 24 patients with a histologically proven LGG andhealthy control subjects for educational level, handedness, age, and gender. The two patient groups were also matched fortumor laterality, use of anticonvulsants, and interval between diagnosis and testing. Functional status was determined inboth patient groups. QOL and cognitive status were compared between the three groups. Results: Matching criteria andfunctional status did not differ significantly between groups. Both patient groups scored worse on QOL scales thanhealthy control subjects. Unoperated patients with S-LGG scored better on most items than patients with P-LGG.Cognitive status was worse in both groups than in healthy control subjects, but, again, patients with S-LGG performedbetter than patients with P-LGG. Conclusion: These data suggest that a wait-and-see policy in patients with S-LGG hasno negative effect on cognitive performance and QOL.

NEUROLOGY 2001;56:618–623

A recurring dilemma arises when a patient presentswith late-onset epilepsy and brain imaging shows alesion suggestive of a low-grade glioma (LGG).1-3 Al-though several retrospective analyses have sug-gested a survival benefit in patients with earlyradiotherapy, these studies are confounded by pa-tient selection.4,5 Preliminary results of a large ran-domized trial indicate that early radiotherapy failsto prolong survival in LGG patients.6 Also, in thedose range evaluated, a randomized clinical trialfailed to demonstrate a dose–response relationshipfor radiation in LGG patients.7 The benefits of surgicalresection have not been proven either, as retrospectivestudies show conflicting results.5,8 Therefore, deferringintervention and even biopsy may be justified in thesepatients.9

Health-related quality of life (QOL) has become anincreasingly important measure in studies of braintumor patients.10 Assessment of QOL in glioma pa-tients is particularly relevant, as these patients can-not be cured and nearly all will eventually die oftheir disease. The effects on QOL of all treatmentmodalities in patients with glioma have to be bal-anced against their possible survival benefits.10,11

QOL of patients with glioma is often impaired by adecreased general and neurologic performance statusand decreased cognitive functioning.12 Patients with

LGG do not have a significantly affected perfor-mance status, but they do show serious cognitiveimpairment.13 Modern radiotherapy to moderatedoses does not lead to significant cognitive impair-ment within a median of 3 years of treatment, butthe results of longer-term studies are yet to be re-ported.14 Therefore, cognitive disturbances shouldprobably be attributed to damage by the tumor itselfor surgical procedures or to psychological effects.14,15

To our knowledge, no studies have assessed QOLand cognitive performance status (CPS) in patientswith suspected but unconfirmed LGG (S-LGG). Un-certainty about the diagnosis may negatively influ-ence CPS and QOL, as has been shown in otherpatient groups.16 If this also applies to S-LGG pa-tients, it would be a reason to aim at a definitediagnosis at an early stage. On the other hand, sur-gical procedures (stereotactic biopsy or resection)necessary to reach this goal may also have a nega-tive impact on CPS and QOL.

The present study was undertaken to determinethe CPS and QOL of patients with S-LGG and inves-tigate whether there is a difference in CPS and QOLbetween S-LGG patients and patients with a provenLGG (P-LGG). We therefore assessed CPS and QOLin 24 S-LGG patients in whom all intervention forthe tumor was deferred. These results were com-

From the Departments of Neurology (Drs. Reijneveld and Taphoorn) and Psychiatry (Dr. Sitskoorn and J. Nuyen), University Medical Center Utrecht; andDepartment of Medical Psychology (Dr. Klein), “Vrije Universiteit” Medical Center, Amsterdam, the Netherlands.Received July 27, 2000. Accepted in final form November 8, 2000.Address correspondence and reprint requests to Dr. J.C. Reijneveld, Department of Neurology, G03.228, University Medical Center Utrecht, P.O. Box 85500,3508 GA Utrecht, the Netherlands; e-mail: J.Reijneveld@neuro.azu.nl

618 Copyright © 2001 by AAN Enterprises, Inc.

pared with data of carefully matched healthy controlsubjects and with data of patients with P-LGG whounderwent subtotal resection or biopsy.

Methods. Patients and control subjects. Patients withS-LGG were recruited from 16 hospitals in the westernand central part of the Netherlands. Recruitment of thisgroup took approximately 1 year. The subjects were stud-ied between June 1998 and July 1999. All patients were atleast 18 years old and had presented with epileptic sei-zures without neurologic deficits. All patients showed non-enhancing supratentorial lesions on MRI or CT, withoutedema or mass effect. Furthermore, none of the patientshad radiologic or clinical signs of progression for at least 6months from the moment of the presumed diagnosis. Noneof the patients used corticosteroids. All patients gave writ-ten informed consent before assessment, and the ethicalcommittee of the University Medical Center Utrecht ap-proved all aspects of the study.

Patients with P-LGG were recruited from a database of200 patients with histologically proven supratentorialLGG from the “Vrije Universiteit” Medical Center (Amster-dam, the Netherlands). All patients had also presentedwith epileptic seizures, and none of them had neurologicdeficits. They had undergone biopsy or resection at least 6months previously and had been without progression sincethen. None of them had been treated with radiotherapy.They were selected to match S-LGG patients on tumorlaterality, handedness, educational level, number of yearsof full-time education, interval between diagnosis and test-ing, gender, and age. Healthy control patients were re-cruited from a database of the University of Maastricht(the Netherlands; MAAS study; CPS data)17 and the Neth-erlands Organization for Applied Scientific Research(TNO; Medical Outcomes Study [MOS] SF-36 Short FormHealth Survey QOL data). They were matched with thepatients for handedness (CPS data only), educational level,gender, and age.

Performance status. Functional impairment was as-sessed with the Karnofsky Performance Status (KPS)18

and the Barthel Index.19 Neurologic functioning of patientswas rated by means of the Neurologic Functional StatusScale (NFSS).20

Quality of life. The QOL was assessed with the MOSSF-36 Short-Form Health Survey, which has been trans-lated into the Dutch language.21,22 It is a self-report ques-tionnaire, composed of 36 items, organized into eightmulti-item scales assessing physical functioning, physicalrole functioning, emotional role functioning, pain, vitality,social functioning, mental health, and general health per-ceptions. A supplementary questionnaire module, theBrain Cancer Module 20 (BCM20), was employed to assessbrain tumor-specific QOL issues. The BCM20 contains fourmulti-item scales assessing visual disorder, communica-tion deficit, motor dysfunction, and uncertainty about thefuture and seven single items about headache, seizures,drowsiness, hair loss, itching, weakness of the legs, anddifficulty in bladder control.23 To assess the self-perceivedlevel of cognitive functioning, we used the Cognitive Func-tioning Scale (CFS), a six-item scale reflecting problemsolving, concentration, confusion, forgetfulness, sustainedattention, and slowness.21 The summary “cognitive com-

plaints” score of this scale was used for statistical analysis.Filling in of all questionnaires took '45 minutes.

Cognitive performance status. A neuropsychologicaltest battery was administered to each patient by trainedpsychometricians, who were supervised by a neuropsychol-ogist. This test battery consisted of the following tests:Visual Verbal Learning Test,24 Working Memory Task,25

Letter–Digit Substitution Test,26 Categoric Word Fluen-cy,27 Concept Shifting Test,28 Stroop Color Word Test,29 andWorking Memory Task.25 Summary measures were calcu-lated to detect possible deficits in the domains of attentionand concentration, visual–verbal memory, executive func-tion, and psychomotor function. Administration of thesetests, which were administered in a fixed order for eachpatient, took an average of 90 minutes. Because of missingdata, 21 subjects of the S-LGG group and 23 subjects of thehealthy control group were used in the final analysis of theCPS data.

Statistical analysis. All statistical analyses were car-ried out with the Statistical Package for the Social Sci-ences (SPSS version 9.0 for Windows). All tests were twosided, and significance was accepted at the 5% level. Sta-tistical analysis of the matching criteria was done with x2

tests (sex, handedness), Fisher’s exact test (tumor location,anticonvulsants), analysis of variance (ANOVA; age, yearsof full-time education, interval between diagnosis and test-ing), and Kruskal–Wallis H test (educational level). Statis-tical analysis of tumor diameter, KPS, Barthel Index, andNFSS was done with Mann–Whitney U tests.

Considerable obliquity of distribution was present in allgroups for most of the QOL variables. This obliquity alsoseemed to account for the inequality of variances amongthe groups for several QOL variables. Transformationswere not helpful to overcome this problem. Therefore, theQOL scales were treated independently and analyzed bymeans of nonparametric tests. In case of the MOS SF-36,Kruskal–Wallis H tests were performed to assess whetherthere were overall differences between patients and controlsubjects. We used post hoc Mann–Whitney U tests withBonferroni adjustment to determine which of the threegroups differed from each other. In case of the BCM, weused Mann–Whitney U tests to demonstrate possible dif-ferences between both patient groups. Performances of thepatient groups on the CFS (i.e., summary “cognitive com-plaints” scores) were compared with a one-way univariateANOVA.

For analysis of the CPS data, all test scores wererescaled to obtain z scores (standard equivalents) by usingthe means and the standard deviations of a matched con-trol group as norm scores. The advantage of the use ofmeans and standard deviations of control subjects to ob-tain z scores, instead of ratings of the general population,is that the control subjects in our study were matched fordemographic characteristics of the patient groups (i.e., sexdistribution, handedness, age at test, and educational lev-el). The four summary functional domain scores for eachsubject were calculated by averaging the z scores for asimilar domain. A multivariate ANOVA was performed,with the four functional domain scores serving as depen-dent variables and group (i.e., S-LGG patients, P-LGG pa-tients, and healthy control subjects) as independentvariables. In case of a significant multivariate effect, uni-variate analyses were performed to examine overall group

March (1 of 2) 2001 NEUROLOGY 56 619

differences for a specific functional domain. These werefollowed by t-tests with Bonferroni adjustment to deter-mine differences between the three groups.

To investigate possible relationships between cognitiveimpairment and tumor size, Spearman correlation coeffi-cients were used to correlate individual CPS and QOL dataand tumor diameter. Bonferroni corrections were appliedto adjust for multiple testing.

Results. Demographic and clinical characteristics.Matching criteria did not differ significantly between thetwo patient groups. Moreover, we found no significant dif-ferences in performance on the KPS, Barthel Index, andNFSS between the two patient groups. High scores for theKPS, Barthel Index, and NFSS were found in both groups,indicating that patients were functionally independent,were able to carry on normal activity, and showed onlyminor signs or symptoms of disease. Seventeen patientswith P-LGG had undergone (sub-)total resection (71%) andsix patients stereotactic biopsy (25%); data on the type ofoperation were not available in one patient. The histologicdiagnosis was low-grade astrocytoma (WHO classification)in all patients. Tumor diameter on the latest MR or CTscan before surgery (P-LGG patients) or before testing (S-LGG patients) did not differ significantly between the twogroups. Data on tumor diameter were not available inthree S-LGG and two P-LGG patients. Eight patients withP-LGG noticed a change in the pattern or frequency of

seizures after biopsy or resection. In four of them, seizurefrequency decreased; in two of them, there was an in-crease; and in the other two, the type of seizures changed.At the time of testing, 19 of 24 S-LGG patients (79%) and15 of 24 P-LGG patients (63%) were using anticonvulsants.The interval between diagnosis and testing did not differsignificantly between the two groups (S-LGG patients, 4.4 64.5 years; P-LGG patients, 5.5 6 3.7 years). Since the endof the study, two patients with S-LGG have been operatedon because of clinical deterioration in combination with anenhancing and mass-occupying lesion on MRI; the histo-logic diagnosis was anaplastic astrocytoma in both. Thetable shows the demographic, clinical, and radiographiccharacteristics of the two patient groups. The two healthycontrol groups did not differ significantly from the twopatient groups on the matching criteria (QOL healthy con-trol subjects: 13 male, mean age 42.8 6 12.0 years [range19 to 71 years], mean educational level 4.0; CPS healthycontrol subjects, 13 male, mean age 42.8 6 11.6 years[range 25 to 71 years], mean educational level 4.5, 22right-handed).

Quality of life. Kruskal–Wallis U tests showed overallgroup differences for the domains of general health percep-tions (p , 0.05), vitality (p , 0.005), social functioning(p , 0.05), and mental health (p , 0.05) of the MOS-SF 36.Post hoc Mann–Whitney U tests with Bonferroni adjust-ment indicated that patients with P-LGG scored worsethan normal control subjects on all four scales (i.e., general

Table Characteristics of patients with suspected (S-LGG) and proven (P-LGG) low-grade glioma

Characteristic S-LGG patients P-LGG patients Significance

No. of persons 24 24 —

Sex, M/F 10/14 15/9 NS*

Age at test, y; mean 6 SD (range) 42.8 6 12.1 (18–71) 38.2 6 10.6 (20–55) NS‡

Handedness, R/L 21/3 22/2 NS*†

Tumor location, R/L 11/13 11/13 NS†

Frontal 10 4

Temporal 10 7

Parietal 3 7

Occipital 1 3

Midline — 3

Tumor diameter, cm 4.24 6 2.32 (n 5 21) 3.59 6 1.59 (n 5 22) NS\

Treatment, stereotactic biopsy/surgery/unknown — 6/17/1 —

Educational level¶ 5.0 3.5 NS§

Education, y 12.5 6 3.6 12.9 6 4.0 NS‡

Anticonvulsants, N/Y 5/19 9/15 NS†

Diagnosis to test interval, y; mean 6 SD 4.4 6 4.5 5.5 6 3.7 NS‡

Barthel Index, mean 6 SD 20.0 6 0.2 19.9 6 0.4 NS\

Neurologic Functional Status Scale, mean 6 SD 1.0 6 0.0 1.0 6 0.0 NS\

Karnofsky Performance Scale, mean 6 SD 92.5 6 9.0 88.7 6 9.7 NS\

* x2 test.† Fisher’s exact test.‡ Analysis of variance.§ Kruskal–Wallis H test.\ Mann–Whitney U test.¶ Educational level was measured by a Dutch scoring system ranging from unfinished primary education (level 1) to university educa-

tion (level 8).

620 NEUROLOGY 56 March (1 of 2) 2001

health perceptions, p , 0.05; vitality, p , 0.01; social func-tioning, p , 0.05; mental health, p , 0.05). Patients withS-LGG scored worse on the general health perceptions do-main than normal control subjects (p , 0.05). With theexception of the physical functioning domain, we found aconsistent tendency for patients with P-LGG to score loweron all domains than S-LGG patients, as illustrated in fig-ure 1A. The difference was significant only for the vitalitydomain. BCM20 data show the same tendency of betterscores for patients with S-LGG (figure 1B). Mann–WhitneyU tests showed worse scores for patients with P-LGGthan for those with S-LGG on the scale motor dysfunc-tion (p , 0.001) and the single item measuring difficultyin bladder control (p , 0.05). We found no significant

difference in the score for the dimension future uncer-tainty between both groups. The CFS data are shown infigure 1C. ANOVA revealed no significant differencesbetween the total sum scores of both patient groups(S-LGG, 18.87 6 12.61; P-LGG, 22.11 6 5.57). We foundno correlation between tumor diameter and the scoreson the different QOL subscales.

Cognitive performance status. The multivariateANOVA showed an overall difference between the groups(F [2.65] 5 2.51, p , 0.05). Univariate analysis indicatedgroup differences for psychomotor functioning (F [2.65] 57.47, p , 0.005). No significant group differences werefound for the other three functional domains. Post hoc ttests with Bonferroni adjustment showed that patients

Figure 1. Quality of life of patientswith proven (P-LGG) and suspected (S-LGG) low-grade glioma. (A) Scores ofpatients with P-LGG (filled columns)and S-LGG (hatched columns) on theeight domains of the Medical OutcomesStudy SF-36 Short Form Health Surveycompared with those of healthy controlsubjects (open columns) (mean andSEM; higher score means fewer com-plaints). *Lower score for P-LGG pa-tients compared with control subjects;**lower score for S-LGG patients com-pared with control subjects; ***lowerscore for P-LGG compared with S-LGG.(B) Scores of patients with P-LGG(filled columns) and S-LGG (hatchedcolumns) on the subscales of the BrainCancer Module 20 (mean and SEM;higher score means more complaints).*Worse scores for P-LGG comparedwith S-LGG patients. (C) Scores of pa-tients with P-LGG (filled columns) andS-LGG (hatched columns) on the itemsof the self-rating Cognitive FunctioningScale (mean and SEM; higher scoremeans fewer complaints).

March (1 of 2) 2001 NEUROLOGY 56 621

with P-LGG performed worse on the psychomotor function-ing than S-LGG patients (p , 0.005) and than healthycontrol subjects (p , 0.01) (figure 2). We found no correla-tion between tumor diameter and the scores on the fourCPS domains.

Discussion. This study demonstrates that pa-tients with P- or S-LGG have a worse QOL and cog-nitive status than healthy control subjects and thatpatients with LGG who underwent resection or bi-opsy are worse on these measures than patients withS-LGG who were managed conservatively. Patientswith S-LGG show significantly better scores for theQOL domains vitality, motor dysfunction, and diffi-culty in bladder control than patients with P-LGG.Psychomotor functioning was significantly worse forpatients with P-LGG.

QOL has become an increasingly important mea-sure in glioma studies.10,30 The measurement of QOLin recent studies, however, was more or less re-stricted to the patient’s functional status18 or cogni-tive functioning.31 Nowadays, it is generally acceptedthat a more comprehensive set of measures reflect-ing health status, CPS, and QOL is needed for amore complete assessment.10,13,14,32 This is again illus-trated by our study: Despite an excellent perfor-mance status and the absence of neurologic deficits,both groups demonstrate deficits in CPS as well as areduction in perceived QOL, when measured with anextensive battery of validated tools.

Neuropsychological assessment showed a signifi-cantly worse performance on the domain of psy-chomotor functioning for P-LGG patients, althoughthey had fewer cognitive symptoms on the self-ratingCFS. This discrepancy between subjective com-plaints and objective test results has been describedbefore14 and indicates that P-LGG patients may notbe aware of their cognitive deficits.

The results of our study may have been biased bythe selection of patients. Initial treatment of patientswith LGG depends on the local neurologist’s andneurosurgeon’s view about the benefit of early, ag-gressive treatment, the clinical status of the patient,the site and extent of the lesion, and the patient’spreferences. Presumably, patients with more exten-sive neurologic deficits or nondominant lesions willmore frequently undergo early treatment. All ourpatients presented with epileptic seizures but wereotherwise clinically normal. We found no significantdifference in the number of patients taking anticon-vulsants between both groups. The patients in ourstudy were matched on tumor laterality, but thegroups were too small to use tumor location as asupplementary matching criterion. Tumor size, how-ever, did not differ significantly between the twogroups and was not correlated to performance on theQOL or CPS subscales. Furthermore, patients whocannot deal with uncertainty about their diagnosismay insist on early treatment. Premorbid patientcharacteristics determine, among other factors, thecapability of patients to deal with this uncertainty.Therefore, age and educational level, which are arelatively accurate indication of premorbid function-ing level, were used as matching criteria.

We do not know whether the decreased QOL andcognitive dysfunctioning of patients with P-LGGshould be attributed to the surgical procedures. Pre-vious reports at least have not shown clear benefitsof the surgical approach in terms of survival timeor progression-free interval.4,7,8,33 Hope of improv-ing epilepsy may be another reason to operate onpatients with LGG. In our group of P-LGG pa-tients, however, epilepsy improved in 4 of 24 pa-tients (17%) after the operation but got worse in atleast 2 other patients (8%).

Another reason for operation could be that know-ing the definite diagnosis would diminish feelings ofuncertainty about the future. Patients with gliomaare faced with uncertainty about several issues suchas length of survival and anxiety about neurologicimpairment with disability.15 Uncertainty about thefuture, however, did not differ between the two pa-tient groups in our study. Therefore, we concludethat a definite diagnosis does not lead to a largereduction of these feelings.

AcknowledgmentThe authors thank Dr. J.E.C. Bromberg (Department of Neurol-ogy) and D.R. van Rij, S.P.W.M. Bouts, and M. van der Linden(Department of Psychiatry, University Medical Center Utrecht,the Netherlands) for assistance in different parts of this study.They also thank Prof. Dr. J. van Gijn (Department of Neurology,University Medical Center Utrecht), Prof. Dr. N. Aaronson (De-partment of Psychosocial Research and Epidemiology, Nether-lands Cancer Institute, Amsterdam), Prof. J.J. Heimans(Department of Neurology, “Vrije Universiteit” Medical Center,Amsterdam), and Prof. H.M. van der Ploeg (Department of Medi-cal Psychology, “Vrije Universiteit” Medical Center, Amsterdam,the Netherlands) for valuable comments on previous versions ofthe manuscript.

Figure 2. Mean z scores for patients with proven (P-LGG;filled columns) and suspected (S-LGG; hatched columns)low-grade glioma on the four cognitive domains comparedwith the scores of a matched group of healthy control sub-jects. The mean z scores of the healthy control subjects(open columns) were rescaled to zero for all four domainsto visualize the relative performance of the two patientgroups (mean and SEM; higher score means better perfor-mance). *Lower score for P-LGG patients compared withhealthy control subjects and S-LGG patients.

622 NEUROLOGY 56 March (1 of 2) 2001

References1. Cairncross JG, Laperriere NJ. Low-grade glioma: to treat or

not to treat? Arch Neurol 1989;46:1238–1239.2. Levin VA. Controversies in the treatment of low-grade astro-

cytomas and oligodendrogliomas. Curr Opin Neurol 1996;8:175–177.

3. Cairncross JG. Understanding low-grade glioma: a decade ofprogress. Neurology 2000;54:1402–1403.

4. Medbery CA, Straus KL, Steinberg SM, et al. Low-grade as-trocytomas: treatment results and prognostic variables. Int JRadiat Oncol Biol Phys 1988;15:837–841.

5. Shaw EG, Daumas-Duport C, Scheithauer BW, et al. Radia-tion therapy in the management of low-grade supratentorialastrocytomas. J Neurosurg 1989;70:853–861.

6. Karim ABMF, Cornu P, Bleehen N. Immediate postoperativeradiotherapy in low-grade glioma improves progression freesurvival but not overall survival: preliminary results of anEORTC/MRC randomized phase III study. J Clin Oncol 1998;17:400A.

7. Karim ABMF, Maat B, Hatlevoll R, et al. A randomized trialon dose–response in radiation therapy of low-grade cerebralglioma: European Organization for Research and Treatmentof Cancer (EORTC) study 22844. Int J Radiat Oncol Biol Phys1996;36:549–556.

8. Piepmeier JM. Observations on the current treatment of low-grade astrocytic tumors of the cerebral hemispheres. J Neuro-surg 1987;67:177–181.

9. Recht LD, Lew R, Smith TW. Suspected low-grade glioma: isdeferring treatment safe? Ann Neurol 1992;31:431–436.

10. Kiebert GM, Curran D, Aaronson NK, et al. Quality of lifeafter radiation therapy of cerebral low-grade gliomas of theadult: results of a randomised phase III trial on dose response(EORTC trial 22844). Eur J Cancer 1998;34:1902–1909.

11. Choucair AK, Scott C, Urtasun R, et al. Quality of life andneuropsychological evaluation for patients with malignant as-trocytomas: RTOG 91–14. Radiation Therapy Oncology Group.Int J Radiat Oncol Biol Phys 1997;38:9–20.

12. Osoba D, Aaronson NK, Muller M, et al. Effect of neurologicaldysfunction on health-related quality of life in patients withhigh-grade glioma. J Neurooncol 1997;34:263–278.

13. Taphoorn MJB, Heimans JJ, Snoek FJ, et al. Assessment ofquality of life in patients treated for low-grade glioma; a pre-liminary report. J Neurol Neurosurg Psychiatry 1992;55:372–376.

14. Taphoorn MJB, Klein Schiphorst A, Snoek FJ, et al. Cognitivefunctions and quality of life in patients with low-grade glio-mas: the impact of radiotherapy. Ann Neurol 1994;36:48–54.

15. Newton C, Mateo MA. Uncertainty: strategies for patientswith brain tumor and their family. Cancer Nurs 1994;17:137–140.

16. Kash KM, Holland JC, Osborne MP, et al. Psychological coun-seling strategies for women at risk of breast cancer. JNCIMonogr 1995;17:73–79.

17. Jolles J, van Boxtel MP, Ponds RW, Metsemakers JF, HouxPJ. The Maastricht Aging Study (MAAS). The longitudinalperspective of cognitive aging. Tijdschr Gerontol Geriatr 1998;29:120–129.

18. Sachsenheimer W, Piotrowski W, Bimmler T. Quality of life inpatients with intracranial tumors on the basis of Karnofsky’sperformance status. J Neurooncol 1992;13:177–181.

19. Brazil L, Thomas R, Laing R, et al. Verbally administeredBarthel Index as functional assessment in brain tumour pa-tients. J Neurooncol 1997;34:187–192.

20. North CA, North RB, Epstein JA, et al. Low-grade cerebralastrocytomas: survival and quality of life after radiation ther-apy. Cancer 1990;66:6–14.

21. Stewart AL, Ware JE. Measuring functioning and well-being:the medical outcomes study approach. Durham: Duke Univer-sity Press, 1992.

22. Aaronson NK, Muller M, Cohen PD, et al. Translation, valida-tion, and norming of the Dutch language version of the SF-36Health Survey in community and chronic disease populations.J Clin Epidemiol 1998;51:1055–1068.

23. Osoba D, Aaronson NK, Muller M, et al. The development andpsychometric validation of a brain cancer quality-of-life ques-tionnaire for use in combination with general cancer-specificquestionnaires. Qual Life Res 1996;5:139–150.

24. Brand N, Jolles J. Learning and retrieval rate of words pre-sented auditorily and visually. J Gen Psychol 1985;112:201–210.

25. Brand N, Jolles J. Information processing in depression andanxiety. Psychol Med 1987;17:145–153.

26. Watson CG, Gasser B, Schaefer A, et al. Separation of brain-damaged from psychiatric patients with ability and personal-ity measures. J Clin Psychol 1981;37:347–353.

27. Butters N, Granholm E, Salmon DP, et al. Episodic and se-mantic memory: a comparison of amnesic and demented pa-tients. J Clin Exp Neuropsychol 1987;9:479–497.

28. Eson ME, Yen JK, Bourke RS. Assessment of recovery fromserious head injury. J Neurol Neurosurg Psychiatry 1978;41:1036–1042.

29. Houx PJ, Jolles J, Vreeling FW. Stroop interference: agingeffects assessed with the Stroop Color-Word Test. Exp AgingRes 1993;19:209–224.

30. Weitzner MA, Meyers CA. Cognitive functioning and qualityof life in malignant glioma patients: a review of the literature.Psychooncology 1997;6:169–177.

31. Trojanowski T, Peszynski J, Turowski K, et al. Quality of sur-vival of patients with brain gliomas treated with postoperativeCCNU and radiation therapy. J Neurosurg 1989;70:18–23.

32. Mackworth N, Fobair P, Prados MD. Quality of life self-reportsfrom 200 brain tumor patients: comparisons with Karnofsky per-formance scores. J Neurooncol 1992;14:243–253.

33. Medbery CA, Straus KL, Steinberg SM, et al. Low-grade as-trocytomas: treatment results and prognostic variables. Int JRadiat Oncol Biol Phys 1988;15:837–841.

March (1 of 2) 2001 NEUROLOGY 56 623