diagnostic value of image-guided needle aspiration cytology in the assessment of vertebral and...

ORIGINAL ARTICLES

Diagnostic Value of Image-Guided Needle AspirationCytology in the Assessment ofVertebral and IntervertebralLesionsRaj K. Gupta, M.D., F.I.A.C.,* Y.K. Cheung, F.R.C.R., F.R.A.C.R.,Abdul G. Al Ansari, M.D., F.F.R. (R.C.P.), Sarla Naran, B.Sc., C.F.I.A.C.,Sharda Lallu, B.Sc., C.F.I.A.C., and Robert Fauck, C.T. (I.A.C.)

The aim of this study was to evaluate the diagnostic value of thenoninvasive method of image-guided needle aspiration cytology(NAC) in the assessment of radiologically detected vertebral andintervertebral space-occupying lesions. NAC was performed undercomputed tomographic (CT) guidance on 70 patients suspected ofhaving a vertebral or intervertebral lesion. Cytologic examinationwas performed on site after staining smears with the Papanicoloaumethod. In addition, air-dried smears, fixed smears, filter prepa-rations from needle washings, and cell blocks were studied. TheNAC diagnosis was supported by examining cell blocks, whichshowed the reliability of histologic architecture; further supportwas obtained with a tissue biopsy in some cases. Additionally,immunoperoxidase and/or histochemical studies were done. Tencases were diagnosed as inflammatory/degenerative lesions, andin 2 cases a diagnosis of neurofibroma was made. Twelve casesconsisted of normal cellular elements, 7 cases were unsatisfactory/inadequate for diagnosis, and 4 cases were suspicious for malig-nancy. In 32 cases, a variety of metastatic tumors were diagnosed,while in 3 cases, a diagnosis of chordoma was made. Based on ourstudy, we feel that NAC, as a first line of investigation, is not onlyuseful in the diagnosis of space-occupying lesions of vertebral andintervertebral regions, but can also help in deciding on appropri-ate management. Also, histologic architecture from cell blocks canbe correlated with cytology, and such material can be used forhistochemical and immunomarker studies. Diagn. Cytopathol.2002;27:191–196. © 2002 Wiley-Liss, Inc.

Key Words: needle aspiration cytology; vertebral; intervertebral;lesion

In recent years, image-guided needle aspiration cytology(NAC) has assumed considerable importance as a preferredfirst line of investigation in the preoperative assessment ofradiologically detected vertebral and intervertebral le-sions.1–23 Since in a number of such cases a known historyof malignancy is present, NAC is performed to confirm orexclude the presence of metastatic disease. However, alllesions in these locations need not necessarily be metastatic,and can vary from nonneoplastic benign to primary neo-plasms. Also, in patients with an unknown clinical historyof malignancy, a positive NAC diagnosis may be the firstindication of malignancy, and may be helpful in furtheringthe clinical investigation to search for an occult neoplasm.Since a decision on conservative vs. radical managementdepends on the type of tumor or nontumorous lesion, it is ofparamount importance that NAC diagnosis of the space-occupying lesion be as accurate as possible. Despite the factthat views on the reliability of NAC diagnosis have recentlymet with some skepticism,24 the diagnostic accuracy ofNAC has continued to improve, not only due to bettertraining of cytopatholgists, but also due to the fact that inrecent years there has been a trend toward a teamworkapproach with radiologists for the diagnosis of such lesions.Furthermore, on-site evaluation of aspirated material foradequacy, procuring additional material in paucicellularsamples, cell blocks, and/or biopsy, application of immu-nocytochemistry, and rarely ultrastructural study on aspi-rates in difficult cases have provided useful information, notonly in enhancing the diagnosis but also in the differentialdiagnosis.

In this paper, we report on our total experience in con-tinuation of our preliminary study17 with image-guided

Department of Cytology and Radiology, Wellington Hospital andSchool of Medicine, Wellington, New Zealand

*Correspondence to: Raj K. Gupta, M.D., F.I.A.C., Cytology Unit,Laboratory Services, Wellington Hospital, Wellington, New Zealand.

Received 15 February 2002; Accepted 3 June 2002DOI 10.1002/dc.10169Published online in Wiley InterScience (www.interscience.wiley.com).

© 2002 WILEY-LISS, INC. Diagnostic Cytopathology, Vol 27, No 4 191

NAC diagnosis of vertebral and intervertebral space-occu-pying lesions that were sampled over a period of 21 yr witha teamwork approach.

Materials and MethodsDuring the period 1981–2001, NAC was performed on 70patients (31 males and 39 females between the ages of18–88 yr) who were suspected of having a vertebral-inter-vertebral lesion following clinical and investigative assess-ments. In all cases, the material was obtained under CTguidance; a maximum of three passes was consideredenough to obtain diagnostic material.

The cytologic examination was performed on site imme-diately following aspiration of at least one or two 95% ethylalcohol-fixed smears from a total of 4–6 smear prepara-tions, which were made by expelling the material directlyonto clean glass slides. Half the remaining smears werefixed in 95% ethyl alcohol, and the others were air-dried.Following the smear preparation, the syringe and needlecontents in all cases were immediately washed in a cytologycontainer containing 30% ethyl alcohol in physiologic sa-line. This was accomplished by withdrawing the 30% ethyl

alcohol into the syringe barrel with the needle attached, andgently flushing the contents back into the cytology con-tainer. From these washings, filter preparations were madeon 25-mm Schleicher and Schuell filters, pore size 3 �m.Additionally, cell block preparations were made in all casesfrom the washings, and sections of these were cut afterroutine processing and stained by hematoxylin-eosin. Allsmears fixed in 95% ethyl alcohol and filter preparationswere stained by the Papanicolaou method, whereas theair-dried smears were stained by the Diff-Quik method. Theon-site NAC diagnosis was confirmed by permanent cyto-logic preparations and cell blocks. Further support was alsogiven with a tissue biopsy which was done in some cases(Tables I II III IV).

For immunoperoxidase and/or histochemical studies, sec-tions of cell blocks and/or tissue were used. Immunostain-ing for cytokeratin (CAM 5.2), keratin, vimentin, B 72.3prostatic-specific antigen (PSA), prostatic-specific acidphosphatase (PSAP), carcinoembryonic antigen (CEA),neuron-specific enolase (NSE), epithelial membrane antigen(EMA), L 26, S100, glial fibrillary acid protein (GFAP),HMB 45, and alpha-fetoprotein was performed using spe-

Table I. Summary of Findings in 10 Cases Diagnosed as Inflammatory/Degenerative Lesions and Two Benign Tumorsa

Case no.Age

(yr)/sex Clinical presentation NAC diagnosis Tissue diagnosisImmunochemical �histochemical stains

1 26/M Destuctive process L1–L2,previous lungtuberculosis.

Tuberculosis Tuberculosis (C � B) Acid-fast bacilli-P

2 38/M Destructive process T12–L1,previous lungtuberculosis.

Tuberculosis Tuberculosis (C � B) Acid-fast bacilli-P

3 51/M Back pain. Destructiveprocess L4–L5.

Osteomyelitis Osteomyelitis (C � B) N.D.

4 58/M Back pain. Suspiciouslesions in L3–L4.

Inflammatory degenerativeprocess

Inflammatory degenerativeprocess (C � B)

N.D.

5 75/F Back pain. Extensivedestructive process T12–L1.


Inflammatory degenerationprocess (C � B)

N.D.

6 66/M Severe back pain. Extensivedestructive process T12.



N.D.

7 85/F Back pain several months.Destructive process in L2region, with collapseddisc.



N.D.

8 42/F Back pain. Destructiveprocess T10–11, withcollapsed disc.



N.D.

9 69/M Back pain. Destructiveprocess in L1–L2.



N.D.

10 28/F Back pain. Destructiveprocess in T9–T10.

Osteomyelitis Osteomyelitis (C � B) N.D.

11 40/F Mild back pain, intermittent,4 � 2.5 cm mass incoccyx region.

Neurofibroma Neurofibroma (C � B) S-100 protein-P

12 32/M Back pain, 3.5 � 3 cm massin sacrococcygeal region.

Neurofibroma Neurofibroma (C � B) S-100 protein-P

aC, cell block; B, biopsy; P, positive; N.D., not done.

GUPTA ET AL.

192 Diagnostic Cytopathology, Vol 27, No 4

cific commercial preparations. The histochemical stains in-cluded mucicarmine, Alcian blue (pH 1 and 2.5), oil- red-O,and periodic acid-Schiff (PAS), both with and without dia-stase digestion. Pertinent details are summarized in TablesI–IV. Known negative and positive controls were usedduring all immunostaining procedures.

ResultsTen cases (Table I) were diagnosed as inflammatory/degen-erative lesions, and in these cases, correlation with a tissuediagnosis was done. In 2 cases (cases 11 and 12, Table I),the NAC diagnosis of neurofibroma correlated with thetissue diagnosis. No elaborate or special stainings wereconsidered necessary (Table I), except for staining for acid-fast bacilli (cases 1 and 2) and S-100 protein (cases 11 and12).

In the 4 cases with NAC suspicion of malignancy (TableII), the diagnosis of the primary and of the site was subse-quently confirmed and was supported by appropriate immu-nostains (Table II), which had been done on the cell blockprior to the biopsy diagnosis.

In 32 cases (Table III), a variety of metastatic tumors onNAC were suggested from the previously known primarysite, and the cell blocks provided adequate material forcorrelation of the tissue diagnosis. Further immunohisto-chemical staining was done in some cases (Table III), andthe results of these were considered useful in supporting thecytohistologic diagnosis (Table III).

In 3 cases (Table IV), the NAC diagnosis of chordomawas made. The smears and filter preparations in all thesecases were cellular, and the tumor cells were found asdissociated cells or loose clusters; the cells were uniform,round to polygonal, mononucleate or binucleate, and rarely

spindly, with occasional cells displaying vacuolated to bub-bly cytoplasm (physaliphorus cells) against a backgroundmatrix that showed marked metachromasia on Diff-Quik.Also, on PAS staining, intense intracytoplasmic PAS posi-tivity was noted in the tumor cells, while Alcian blue stainshowed positivity of the background matrix. Furthermore,tumor-cell immunopositivity for keratin, cytokeratin, EMA,vimentin, and S-100 protein, and negativity for all othermarkers (Table IV), strongly supported the NAC diagnosisof a chordoma. Similar findings in NAC samples weredescribed in other studies.4,5,8–10,14,17

In the 12 cases with NAC findings of normal cellularelements, no further workup was considered necessary, ex-cept for a follow-up with no untoward outcome. However,2 of the 7 cases with NAC diagnosis of unsatisfactory/inadequate material underwent a biopsy due to a persistentand enlarging lesion, and were found, to have a plasmacy-toma (1 case) and a chondrosarcoma (1 case). In the re-maining 5 cases, no untoward outcome was found onfollow-up.

DiscussionNAC of radiologically detected lesions of the vertebral andintervertebral region is now being used with somewhatincreasing frequency in the diagnosis of neoplastic andnonneoplastic conditions in view of the fact that it is aneasy, rapid, relatively safe, and cost-effective method.15,16,18

A frequent indication for NAC of a vertebral column lesionis to exclude a metastasis, especially in a patient with aknown history of a primary tumor.19–21 However, in thislocation a number of benign nontumorous processes canalso occur, and it is important that a distinction between

Table II. Summary of Findings in Four Cases Suspicious for Malignancy on NACa

Case no.Age

(yr)/sex NACImmunochemical �

histochemical stains (C) Biopsy diagnosis

1 58/M ? Lymphoma L 26, N Small-cell carcinoma, lung? Small-cell carcinoma EMA, P? Other Cytokeratin, P

2 88/F ? Breast carcinoma B 72.3, P Carcinoma, breast? Lung carcinoma CEA, P? Other EMA, P

Cytokeratin, P3 35/F ? Breast carcinoma B 72.3, T Non-Hodgkins lymphoma

? Other CEA, NEMA, NCytokeratin, NL 26, T

4 84/M ? Carcinoma, prostate PSA, T Carcinoma prostate? Lung carcinoma PASP, N? Other EMA, T

Cytokeratin, NL 26, N

aC, cell block; P, positive; N, negative; T, traces.

NAC OF VERTEBRAL AND INTERVERTEBRAL LESIONS

Diagnostic Cytopathology, Vol 27, No 4 193

tumors and nontumorous processes be obtained, to permitappropriate management decisions as to which cases needsurgery and which do not. Furthermore, it is quite feasiblethat a diagnosis of the offending lesion be done, as far aspossible, by the least invasive and cost-effective method,since the difficulties of exploring the vertebral column arewell-known.

Despite the fact that the awareness and use of NAC in thediagnosis of vertebral column lesions is being practiced, theexperience with NAC for such lesions seems somewhat lim-ited.1–12,14,15,17,18 In our detailed study as presented here, wecontinued to use NAC on a variety of lesions of the vertebralcolumn, and our results further indicate that image-guidedNAC as the first line of investigations has merit in the diag-nosis, especially because of the well-known clinical difficultiesof diagnosis. The problems of differential diagnosis werestressed in some recent studies,3,5–11,13,14 and it was suggested

that some of these difficulties may be resolved by applicationof immunocytochemical and histochemical stains on cytologicspecimens. We fully concur with these studies;3,5–11,13,14 like-wise, in this study we further showed that the application ofimmunocytochemical and histochemical stains is indeed veryuseful (Tables I–IV).

In conclusion, we are of the opinion that NAC as a firstline of investigation is useful for space-occupying lesions ofthe vertebral and intervertebral region in view of its sim-plicity, low cost, and noninvasiveness. Also, the use ofNAC in diagnostic protocols could save 2 number of pa-tients from an unnecessary radical surgical procedure. Wealso think that in cases of nonneoplastic lesions and neo-plastic lesions, which have to be typed on the basis ofhistologic architecture, the cytologic interpretation can becorrelated with cell blocks of the aspirate. Such material canalso be used for histochemical and immunomarker studies

Table III. Summary of Findings in 32 Cases Diagnosed on NAC and Cell Blocks as Consistent With Metastatic Tumors With a Clinical History ofKnown Malignancya

Case no.Age

(yr)/sexKnown primary site of

malignancySite of

aspiration Immunohistochemical � histochemical stains

1 63/F Cervix T12 EMA-P; CK-P2 67/F Lung T9 ND3 66/M Prostate T6 PSA-P; PSAP-P4 80/F Breast L3 ND5 69/F Breast T4 ND6 81/F Colon L4 MC-P; CEA-P7 61/M Rectum S1 ND8 67/M Sigmoid S2 ND9 48/F Kidney L2 Oil-red-O-P; keratin-T; CK-T; EMA-T10 49/F Large bowel L1 MC-P; CEA-P; B72.3-P; AB-P11 58/F Large bowel T12 ND12 52/F Rectum T9 ND13 73/F Breast T5 ND14 60/F Breast L5 CEA-P; B 72.3-P15 68/M Lung T4 ND16 18/M Leg (fibrosarcoma) L2 ND

17 43/MArm(chondrosarcoma) T11 ND

18 79/M Prostate T8 PSA-P; PSAP-N19 34/M Skin (melanoma) T11 HMB 45-P; S 100-P; vimentin-P; CK-N20 71/M Larynx L5 ND21 88/M Lung S1 ND22 56F Cervix L1 ND23 73/F Rectum T8 ND24 82/M Kidney L11 ND25 58/F Colon L4 CEA-P; B 72.3-P26 41/M Skin (melanoma) L2 HMB 45-P; S 100-P; CK-N; vimentin-P

27 46/FLymph node(lymphoma) L6 L 26-P; EMA-N; CK-N

28 49/F Breast L4 ND29 72/M Lung S4 ND30 83/M Prostate T8 PSA-T; PSAP-P31 78/F Breast L1 ND32 86/M Prostate T9 PSA-P; PSAP-TaCK, cytokeratin (CAM 5.2); ND, not done; L, lumbar; T, thoracic; S, sacrum; MC, mucicarmine; AB, Alcian blue (pH 1 and 2.5).

GUPTA ET AL.


in difficult cases, and such an investigation further enhancesthe cytodiagnosis and provides useful information for thedifferential diagnosis.

References1. Angelopulos N, Ceppi M, Serio G. Sacrococcygeal chordoma: a case

report: fine needle aspiration cytologic and histopathologic findings.Pathologica 1990;82:167–171.

2. Das DK, Francis IM, AbuZeidan FM. FNA cytology diagnosis of asacrococcygeal chordoma and its recurrence. Diagn Cytopathol 1994;10:194–195.

3. Finley JL, Silverman JF, Dabbs DJ, West RL, Dickens A, FeldmanPS, Frable WJ. Chordoma: diagnosis by FNAC with histologic andimmunocytochemical and ultrastructural confirmation. Diagn Cyto-pathol 1986;2:330–337.

4. Hughes DE, Lamb J, Salter DM, Al-Naffusi A. Fine needle aspirationcytology in a case of chordoma. Cytopathology 1992;3:129–133.

5. Kontozoglolu T, Qizibash AH, Sianos J, Stead R. Chordoma: cytologicand immunocytochemical study of four cases. Diagn Cytopathol 1986;2:55–61.

6. Nijhawan VS, Rajwanshi A, Das A, Jayaram N, Gupta SK. Fine-needle aspiration cytology of sacrococcygeal chordoma. Diagn Cyto-pathol 1989;5:404–407.

Table IV. Summary of Findings in Three Cases of Chordomaa

Case no. Age (yr)/sex Clinical presentationNAC

diagnosis Tissue diagnosisImmunochemical �histochemical stains

1 65/F Low back pain. Destructivesacrococcygeal mass.

Chordoma Chordoma (C � B) Keratin, P

Cytokeratin, PEMA, PVimentin, PGFAP, NCEA, NHMB 45, NS 100 protein, PB 72.3, NAlpha-fetoprotein, NNSE, NPAS, PAlcian blue (pH1— 2.5),

P

2 57/M Low back pain. Destructivesacrococcygeal mass.


Cytokeratin, PEMA, PVimentin, PGFAP, NCEA, NHMB 45, NS 100 protein, PB 72.3, NAlpha-fetoprotein, NNSE, NPAS, PAlcian blue (pH1—2.5),

P

3 63/M Back pain. Destructivesacral mass.


Cytokeratin, PEMA, PVimentin, PGFAP, NCEA, NHMB 45, NS 100 protein, PB 72.3, NAlpha-fetoprotein, NNSE, NPAS, PAlcian blue (pH1—2.5),

P

aC, cell Block; B, biopsy; P, positive; N, negative.

NAC OF VERTEBRAL AND INTERVERTEBRAL LESIONS

Diagnostic Cytopathology, Vol 27, No 4 195

7. O’Dowd GJ, Schumann GB. Aspiration cytology and cytohistochem-istry of coccygeal chordoma. Acta Cytol 1983;27:178–183.

8. Perasole A, Infantolino D, Spigariol F. Aspiration cytology and im-munocytochemistry of sacral chordoma with liver metastases. DiagnCytopathol 1991;7:277–281.

9. Plate KH, Bitinger A. Value of immunohistochemistry in aspirationcytology of sacrococcygeal chordoma: a report of two cases. ActaCytol 1992;36:87–90.

10. Plaza JA, Ballestin C, Periz-Barrios A, Martinez MA, de Agustin P.Cytologic, cytochemical, immunocytochemical and ultrastructural di-agnosis of a sacrococcygeal chordoma in a fine needle aspirationbiopsy specimen. Acta Cytol 1989;33:89–92.

11. Rone R, Ramzy I, Duncan D. Anaplastic sacrococcygeal chordoma:fine needle aspiration cytologic findings and embryologic consider-ations. Acta Cytol 1986;30:183–188.

12. Thompson SK, Callery RT. Cytologic diagnosis of a chordoma with-out physaliferous cells. Diagn Cytopathol 1988;4:144–147.

13. Yang GCH, Coleman B, Daly JM, Gupta PK. Presacral myelolipoma:report of a case with needle aspiration cytology and immunohisto-chemical studies. Acta Cytol 1992;36:932–936.

14. Walaas L, Kindblom LG. Fine needle aspiration biopsy in the preop-erative diagnosis of chordoma: a study of 17 cases with application ofelectron microscopic, histochemical immunocytochemical examina-tion. Hum Pathol 1991;22:22–28.

15. Carson HJ, Castelli MJ, Reves CV, Gattuso P. Fine-needle aspirationbiopsy of vertebral body lesions: cytologic, pathologic, and clinicalcorrelation of 57 cases. Diagn Cytopathol 1994;11:348–351.

16. Civardi G, Livraghi T, Colombo P, Fornari F, Cavanna L, Buscarini L.Lytic bone lesions suspected for metastasis: ultrasonically guidedfine-needle aspiration biopsy. J Clin Ultrasound 1994;22:307–311.

17. Gupta RK, Al Ansari AG. Value of image-guided needle aspirationcytology in the assessment of thoracolumbar and sacrococcygealmasses. Acta Cytol 1996;40:215–221.

18. Mondal A, Misra DK. CT-guided needle aspiration (FNAC) of 112vertebral lesions. J Pathol Microbiol 1994;37:255–261.

19. Kang M, Gupta S, Khandelwal N, Shankar S, Gulati M, Suri S.CT-guided fine-needle aspiration biopsy of spinal lesions. Acta Radiol(Stockh) 1999;40:474–478.

20. Alder O, Rosenberger A. Fine needle aspiration biopsy of osteolyticmetastatic lesions. AJR 1979;133:15–18.

21. Agarwal PK, Goel MM, Chandra T, Agarwal S. Predictive value offine needle aspiration cytology of bone lesions. Acta Cytol 1997;41:659–665.

22. Jorda M, Rey L, Hanly A, Ganjei-Azar P. Fine-needle aspirationcytology of bone: accuracy and pitfalls of cytodiagnosis. CancerCytopathol 2000;90:47–54.

23. Roy M, Bhattacharyya A, Sanyal S, Roy A, Dasgupta S. Fine needleaspiration cytology of bone and joint lesions: an assessment. J IndianMed Assoc 1995;93:259–261.

24. Hajdu SI, Melamed MR. Limitations of aspiration cytology in thediagnosis of primary neoplasms. Acta Cytol 1984;28:337–345.

GUPTA ET AL.


diagnostic value of image-guided needle aspiration cytology in the assessment of vertebral and...

Documents