csfvsmrmedullo

Comparison of CSF Cytology and Spinal MagneticResonance Imaging in the Detection of Leptomeningeal

Disease in Pediatric Medulloblastoma or PrimitiveNeuroectodermal Tumor

By Maryam Fouladi, Amar Gajjar, James M. Boyett, Andrew W. Walter, Stephen J. Thompson, Thomas E. Merchant,Jesse J. Jenkins, James W. Langston, Aiyi Liu, Larry E. Kun, and Richard L. Heideman

Purpose: Leptomeningeal disease (LMD) significantlyaffects the prognosis and treatment of pediatric patientswith medulloblastoma or primitive neuroectodermaltumor (PNET). Examination of CSF for malignant cells,detection of LMD on spinal magnetic resonance imaging(MRI), or both are the methods routinely used to diag-nose LMD. A recent study suggested 100% correlationbetween CSF and MRI findings in children with medullo-blastoma. To determine the validity of this hypothesis,we compared the rate of detection of LMD betweenconcurrent lumbar CSF cytology and spinal MRI per-formed at diagnosis in patients with medulloblastomaor PNET.

Patients and Methods: As a part of diagnostic stag-ing, 106 consecutive patients newly diagnosed with me-dulloblastoma or PNET were evaluated with concurrentlumbar CSF cytology and spinal MRI. CSF cytology wasexamined for the presence of malignant cells and spinalMRI was reviewed independently for the presence of LMD.

Results: Thirty-four patients (32%) were diagnosedwith LMD based on CSF cytology, spinal MRI, or both.There were 21 discordant results. Nine patients (8.5%)with positive MRI had negative CSF cytology. Twelvepatients (11.3%) with positive CSF cytology had nega-tive MRIs. The exact 95% upper bounds on the propor-tion of patients with LMD whose disease would havegone undetected using either CSF cytology or MRI as theonly diagnostic modality were calculated at 14.4% and17.7%, respectively.

Conclusion: With the use of either CSF cytology orspinal MRI alone, LMD would be missed in up to 14% to18% of patients with medulloblastoma or PNET. Thus,both CSF cytology and spinal MRI should routinely beused to diagnose LMD in patients with medulloblastomaor PNET.

J Clin Oncol 17:3234-3237. r 1999 by AmericanSociety of Clinical Oncology.

LEPTOMENINGEAL disease (LMD) occurs in up to32% of children with primary CNS tumors at diagno-sis.1 Metastatic spread along the leptomeninges has beenreported for virtually all types of CNS tumors but is mostfrequent in embryonal tumors such as medulloblastoma andprimitive neuroectodermal tumor (PNET).2-4

Patients with CNS tumors who have LMD at diagnosishave a worse prognosis and require more aggressivetherapy.5-8 Consequently, the definitive diagnosis and careful

staging of metastatic disease are of great prognostic impor-tance in childhood brain tumors and are crucial for theselection of effective treatment.

At present, the diagnosis of LMD is based on the use ofgadolinium (Gd)-enhanced spinal magnetic resonance imag-ing (MRI) and the cytologic examination of CSF formalignant cells, which were obtained 2 to 3 weeks aftersurgery. In retrospective analyses of adult patients with avariety of primary or secondary CNS tumors and positivelumbar CSF cytology, MRI revealed abnormalities consis-tent with LMD in up to 70% of patients at some point in theirclinical course.9-13 Although a recent pediatric study re-ported 100% correlation between MRI and lumbar CSFcytology in patients with medulloblastoma,14 the number ofpatients was small and the studies were not contemporane-ous. To determine the relative roles of cytology and MRI forthe diagnosis of LMD, we investigated the impact of usingonly one of the two modalities to diagnose LMD in a largecohort of patients with PNET and medulloblastoma who hadundergone both diagnostic procedures.

PATIENTS AND METHODSBetween December 1989 and August 1998, 106 consecutively

accrued pediatric patients with primary CNS PNET or medulloblastomahad concurrent lumbar punctures and Gd-enhanced spinal MRI as partof their initial diagnostic work-up for the presence of LMD.

From the Departments of Hematology-Oncology, Radiation Oncol-ogy, Pathology and Laboratory Medicine, Biostatistics and Epidemiol-ogy, and Radiology, St Jude Childrens Research Hospital; and Depart-ments of Pediatrics, Radiation Oncology, and Radiology, University ofTennessee, College of Medicine, Memphis, TN.

Submitted February 2, 1999; accepted May 27, 1999.Supported in part by Cancer Center support grants no. PA30CA

21765 and P01 CA 23099 from the National Cancer Institute and by theAmerican Lebanese Syrian Associated Charities (ALSAC).

Address reprint requests to Maryam Fouladi, MD, St Jude ChildrensResearch Hospital; Department of Hematology-Oncology, St JudeChildrens Research Hospital, 332 N Lauderdale, Memphis, TN 38105;email [email protected].

r 1999 by American Society of Clinical Oncology.0732-183X/99/1710-3234

3234 Journal of Clinical Oncology, Vol 17, No 10 (October), 1999: pp 3234-3237

Downloaded from jco.ascopubs.org on September 26, 2013. For personal use only. No other uses without permission.Copyright 1999 American Society of Clinical Oncology. All rights reserved.




Patients eligible for the study were younger than 21 years of age, witha histologically proven diagnosis of medulloblastoma or PNET. Two tothree weeks after the initial surgery, CSF cytology by lumbar puncturewas obtained up to 48 hours postGd-enhanced MRI as part of theinitial staging evaluation.

CSF samples were mixed with one drop of 22% bovine albumin(Organon Teknika, Durham, NC), placed in a cytospin sample chamber,and centrifuged at 1,000 revolutions per minute for 5 minutes (Shandoncentrifuge; Shandon, Cheshire, United Kingdom). A pathologist re-viewed all slides for the detection of malignant cells. Both cytologicstudies and spinal MRIs were interpreted independently of each other.

Spinal MRI examinations were performed on three different SiemensMRI machines over the 10-year period of this study, with softwareprovided by the same vendor. The software underwent a slow evolutionover this period. During the first 9 years, the examinations reliedprimarily on postcontrast imaging, with contiguous 4- or 5-mm sagittalimages of the entire spine followed by routine transverse imaging belowthe conus. During the last year, the examination consisted of 3- or 4-mmoverlapping sagittal images at 1- and 1.5- or 2-mm intervals, withoutroutine transverse images.

Statistical ConsiderationsExact test and 95% confidence upper bounds, for the proportion of

patients in whom LMD would be missed using either MRI or CSF as theonly modality for diagnosis of LMD, were calculated using StatXact 3(Statistical software for exact nonparametric Inference; CYTEL Soft-ware Corporation, Cambridge, MA). Logistic regression was used toinvestigate the relationship between CSF positivity and volume of theCSF test sample. SAS Release 6.12 (SAS Institute, Cary, NC) was usedfor this analysis.

RESULTS

Among the 106 patients, there were 68 males and 38females. The median age was 7.3 years (range, 0.3 to 19.8years). Twenty-five patients had positive CSF cytology formalignant cells at diagnosis, and 81 had negative cytology.There were 22 patients with MRI evidence of LMD and 84without. As listed in Table 1, 34 children (32%) werediagnosed with LMD based on CSF cytology, spinal MRI, orboth. Overall, among the 106 patients, there were 21discordant results. Nine patients (8.5%) with positive MRIhad negative CSF cytology. The exact 95% upper bound onthe proportion of patients with LMD who would have goneundetected using CSF cytology alone is 14.4%. Twelvepatients (11.3%) with positive CSF cytology had negative

MRIs. The exact 95% upper bound on the proportion ofpatients who would have gone undetected using only MRI ofthe spine is 17.7%.

CSF volume was available in 103 of the 106 patients inthe study; three patients did not have CSF volume dataavailable. Only three patients had CSF volumes less than 0.5mL. The median CSF volume was 1.3 mL. The quartilerange for CSF volume (ie, 25% to 75%) was 1 mL to 2 mL.

There was no evidence of an association between thevolume of the CSF sample and the probability that thesample was positive (P . .7). Furthermore, the discordantrates between CSF cytology and MRI did not differ signifi-cantly when CSF volume was greater than 0.5 mL and lessthan 2.0 mL versus a CSF volume greater than or equal to2.0 mL.

DISCUSSION

Up to 32% of patients with PNET or medulloblastomahave metastatic disease at diagnosis.1 The diagnosis of LMDgenerally requires the detection of malignant cells in lumbarCSF or the presence of metastatic disease by Gd-enhancedMRI or computed tomography (CT) myelogram. The sensi-tivity and specificity of each diagnostic method is difficult toestimate because there is no diagnostic standard short ofdetection of LMD at autopsy.

Some series have suggested that CSF cytology obtainedfrom cisternal or shunt taps may be more sensitive than CSFobtained from lumbar punctures in detecting LMD.15,16 Incontrast to largely retrospective studies of adults with avariety of malignancies, we recently completed a prospec-tive study that compared the findings of cytologic examina-tions of CSF obtained from concurrent lumbar and ventricu-loperitoneal shunt taps in 52 pediatric patients with primaryCNS tumors as part of an initial diagnostic work-up orduring follow-up testing to detect LMD.17 A total of 90paired shunt and lumbar CSF samples were analyzed.Among the 90 paired samples in that study, malignant cellswere detected at a significantly higher rate in lumbar CSFthan in shunt CSF (P 5 .0018, one-sided). When the analysiswas confined to only those patients with embryonal tumors,malignant cells continued to be detected at a higher rate inlumbar CSF samples than in shunt samples (P 5 .0008,one-sided). Thus, we concluded that lumbar CSF shouldremain the specimen of choice for the routine cytologicdetection of malignant cells in the CSF of children withLMD.

In recent studies, Gd-enhanced MRI has been found to bemore sensitive than CT myelogram in the detection ofLMD.18,19 Heinz et al18 reported the results of a prospective

Table 1. Lumbar CSF Cytology and Spinal MRI Correlation in 106 PatientsWith PNET/Medulloblastoma at Diagnosis

PositiveMRI

NegativeMRI

TotalNo. ofPatients

Positive CSF 13 12 25Negative CSF 9 72 81Total 22 84 106

NOTE. Total number of patients diagnosed with LMD by MRI only, CSFcytology only, or both: 9 1 12 1 13 5 34.

LEPTOMENINGEAL DISEASE IN MEDULLOBLASTOMA/PNET 3235

study of 33 patients with primary CNS tumors who hadGd-enhanced MRI within 2 weeks of a CT myelogram.Seven patients were found to have metastatic disease.Although both MRI and CT myelogram detected metastasesin patients, the number of lesions detected by MRI was 24compared with 15 detected by CT myelogram. These resultslead to the conclusion that MRI may be more sensitive thanCT myelogram in detecting LMD.18 This data has also beenconfirmed in a larger study of 61 patients by Chamberlain.19

Most studies of the role of neuroimaging and CSFcytology in the diagnosis of LMD have been retrospectiveand have primarily reported the incidence of MRI or CTabnormalities in patients with positive CSF cytologic re-sults. In these early studies, which involved adults with avariety of metastatic solid tumors, MRI revealed abnormali-ties in 23% to 71% of patients with positive results on CSFcytology.9-13

The only study to address the correlation between CSFcytology and demonstration of spinal metastases by Gd-enhanced MRI in medulloblastomas is by Harrison et al.14Of 26 patients with medulloblastoma, 10 patients had beeninvestigated with both CSF cytology and MRI of the spineup to 8 months after diagnosis. Seven of these samples werelumbar, seven were obtained intraoperatively from cisternamagna, and three were from both. When comparing lumbarCSF with spinal MRI, three cases demonstrated involvementby both routes, and four were negative by both routes. Theauthors concluded that there was a 100% correlation be-tween lumbar CSF and spinal MRI. A significant limitationto Harrisons study is that the sample size is too small todraw a meaningful conclusion regarding the sensitivity ofthese two modalities; among the 26 patients, only seven hadboth lumbar CSF cytology and spinal MRI. Furthermore, theCSF samples and spinal MRIs were not concurrent. Theinterval between the two studies ranged from 6 to 198 days(average, 65 days). Finally, the timing of the MRI was notuniform, ranging from preoperative to postoperative (range,8 days to 8 months).

In contrast, the current study involved 106 consecutivepatients with PNETs or medulloblastomas who had bothCSF cytology and Gd-enhanced MRI of the spine performedas part of their initial staging evaluation in the earlypostoperative period. There was no selection of patients bysymptoms of leptomeningeal disease, and the two diagnosticmethods were performed concurrently (within 48 hours ofeach other), 2 to 3 weeks postsurgery. Based on our findings,if CSF cytology had been the only diagnostic procedureused, we estimate that up to 14.4% of patients with LMDwould have been missed. If only MRI of the spine had beenused to diagnose LMD, up to 17.7% of patients with LMDwould have been missed. Thus, the use of either diagnosticmodality alone could result in missing the diagnosis of LMDin up to 14% to 18% of patients, which, in turn, could lead tothe potential under-treatment of a patient.

There was no association noted in this study between thevolume of CSF and the probability that the sample waspositive (P . .7). Furthermore, the discordant rates betweenCSF cytology and MRI did not differ significantly whenCSF volume was greater than 0.5 mL and less than 2.0 mLcompared with volumes greater than or equal to 2.0 mL.However, it must be pointed out that the CSF volume wasnot standardized because the data were collected retrospec-tively. A prospective study that uses larger volumes of CSFneeds to be performed to assess this question more accu-rately and in a more standardized fashion.

In conclusion, this study strongly suggests that neitherspinal MRI nor CSF cytology alone is sufficient for thedefinitive diagnosis of LMD. We conclude that both CSFcytology and spinal Gd-enhanced MRI are necessary tomaximize the potential identification of the presence ofLMD in patients with medulloblastoma or PNET.

ACKNOWLEDGMENTThe authors thank Jennifer Havens and Annemarie Fraga for

assistance in data collection and Wei Wang for assistance with thestatistical analysis.

REFERENCES1. Packer RJ, Siegel KR, Sutton LN, et al: Leptomeningeal dissemi-

nation of primary central nervous system in childhood. Ann Neurol18:217-221, 1985

2. Deutsch M, Riegel DH: The value of myelography in themanagement of childhood medulloblastoma. Cancer 45:2194-2197,1980

3. Dorwat RH, Wara WM, Norman D, et al: Complete myelographicevaluation of spinal metastases from medulloblastoma. Radiology139:403-408, 1981

4. Packer R, Sutton L, Goldwein J, et al: Improved survival with theuse of adjuvant chemotherapy in the treatment of medulloblastoma. JNeurosurg 74:433-440, 1991

5. Packer RJ, Sutton LN, Elterman R, et al: Outcome of childrenwith medulloblastoma treated with radiation, cisplatin, CCNU, andvincristine chemotherapy. J Neurosurg 81:690-698, 1994

6. Boyett J, Zetlzer P, Finlay J, et al: Progression free survival andrisk factors for primitive neuroectodermal tumors (PNET) of theposterior fossa (PF)[medulloblastoma] in children: Report of theChildrens Cancer Study Group (CCG) randomized trial, CCG 921.Proc Am Soc Clin Oncol 14:147, 1995 (abstr 283)

7. Kuhl J, Berthold F, Bode U, et al: Preradiation chemotherapy ofchildren with poor prognosis medulloblastoma: Response rate andtoxicity of the ifosfamide-containing multidrug regimen HIT 88/89.Am J Pediatr Hematol Oncol 15:567-571, 1993 (suppl A)

3236 FOULADI ET AL

8. Bailey C, Gnekow A, Wellek S, et al: Prospective randomised trialof chemotherapy given before radiotherapy in childhood medulloblas-toma: International Society of Pediatric Oncology (SIOP) and theGerman Society of Pediatric Oncology (GPO)SIOP II. Med PediatrOncol 25:166-178, 1995

9. Freilich RJ, Krol G, DeAngelis LM: Neuroimaging and cerebro-spinal fluid cytology in the diagnosis of leptomeningeal metastasis. AnnNeurol 38:51-57, 1995

10. Krol G, Sze G, Malkin M et al: MR of cranial and spinalmeningeal carcinomatosis: Comparison with CT and myelography.AJNR Am J Neuroradiol 9:709-714, 1988

11. Chamberlain MC, Sandy AD, Press GA: Leptomeningeal metas-tasisa comparison of Gadolinium enhanced MR and contrast enhancedCT of the brain. Neurology 40:435-438, 1990

12. Davis PC, Friedman NC, Fry SM, et al: Leptomeningealmetastases: MR imaging. Radiology 163:449-454, 1987

13. Youssem DM, Patrone PM, Grossman RI: Leptomeningealmetastases: MR evaluation. J Comput Assist Tomogr 14:255-261, 1990

14. Harrison SK, Ditchfield MR, Walters K: Correlation of MRI andCSF cytology in the diagnosis of medulloblastoma spinal metastsases.Pediatr Radiol 28:571-574, 1998

15. Wasserman WR, Glass JP, Posner JB: Diagnosis and treatment ofleptomeningeal metastases from solid tumors: experience with 90patients. Cancer 49:759-772, 1982.

16. Rogers LR, Duchesneau PM, Nunez C, et al: Comparison ofcisternal and lumbar CSF examination in leptomeningeal metastases.Neurology 42:1239-1241, 1992

17. Gajjar A, Fouladi M, Walter A: Comparison of lumbar and shuntcerebrospinal fluid specimens for cytologic detection of leptomen-ingeal disease in pediatric patients with brain tumors. J Clin Oncol 17:1825-1828, 1999

18. Heinz R, Wiener D, Friedman H, et al: Detection of cerebrospi-nal fluid metastasis: CT myelography or MR? AJNR Am J Neurora-diol16:1147-1151, 1995

19. Chamberlain MC: Comparative spine imaging in leptomeningealmetastases J Neurooncol 23:233-238, 1995

LEPTOMENINGEAL DISEASE IN MEDULLOBLASTOMA/PNET 3237

csfvsmrmedullo

Documents