Proceedings S.Z.P.G.M.I vol: 12(1-2) 1998, pp. 28-36.

Fine Needle Aspiration Cytolog·y : Does It Have a Role in the Management of Clinical Solitary

Thyroid Nodules? Bilquies A. Soleman*, Abdul Latif Khan**

Departments of Histopathology, *Sheikh Zayed PGMI, Lahore and **Royal Liverpool University Hospital, Liverpool, UK

SUMMARY

Fine needle aspiration ·cytology is a safe and cost effective diagnostic method for thyroid abnormalities. It is recognised as the first line investigation for solitary thyroid nodules. We analysed 443 thyroid aspirates, carried out in this Postgraduate Medical Institute. Aspirations were done in an outpatient FNA clinic by the pathologist, under local anaesthesia. Slides were stained with May - Grunwald Giemsa and Papanicolaou methods. On cytological examination these aspirates were categorised as follows: Unsatisfactory for diagnosis = 22, non Neoplastic = 155, Indeterminate but benign 31, .Follicular adenoma I Follicular neoplasm = 67, Hyperplastic colloid nodule Vs Follicular adenoma = 111, Suspicious malignant = 40 and malignant = 17. _Where appropriate, excision biopsy was suggested. in 97 (21. 9 % ) cases, surgical excision specimens were available for comparative analysis. Although a degree of overlap was observed in benign thyroid lesions including hyperplastic colloid nodules and follicular adenomas and one microscopic sized papillary carcinoma was misdiagnosed as benign due to sampling error, the remaining aspirates were correctly labelled as either, benign or malignant. Most of the suspicious malignant aspirates turned out to be malignant on histology. No significant complications occurred after the FNA procedure. Despite, the overlap of the spectrum of cytological features of different thyroid lesions, FNA cytological examination of solitary thyroid nodules is strongly recommended.

INTRODUCTION

F ine needle aspiration (FNA) cytology is a safeand expedient diagnostic method for thyroid

abnormalities l . The procedure is of great importance in the examination of thyroid swellings because of its convenience, accuracy, sensitivity and utility in obviatipg the need for open tissue biopsies of this richly vascular organ.2. It is not only useful in selecting patients for surgery, thereby avoiding un-necess.µ-y operations, bu_t it may also be therapeutic for cystic lesions3. Although FNA is now reco_gnised to be the first line investigation for a solitary thyroid nodule, however, the principal

area of difficulty in thyroid aspiration cytology _ the distinction· between cellular non-neoplasn.: nodules, follicular adenomas and follicU:Z: carcinomas4 .

Our experience with fine needle aspirati cytology of palpable breast lesionsS ,6 and space -occupying lesions in liver7 has been reviewec previously. The present study is cytology in £bf management of solitary thyroid nodules Cytomorphological features of different thyroi!: lesions are described. In cases where follow-up e�cision biopsies were carried out, the histologica diagnosis are compared with the original cytologi� diagnosis and pitfalls in thyroid aspiration cytology are discussed.

28

Fine Needle Aspiration Cytology

MATERIAL AND METHODS

From 1991 to 1995, 443 aspirates of the thyroid gland were carried out in Shaikh Zayed Postgraduate Medical Institute, Lahore. These aspirates from solitary thyroid nodules, were initially detected by the clinicians. With ra?a exception, most of these were "cold" nodules on radio-isotope uptake studies. The aspirates were performed by the pathologist in an FNA outpatient clinic, after careful palpation of the thyroid gland and examining the accompanying thyroid scan. After a brief explanation to the patient, the procedure was performed under local anaesthesia. The patients were either sitting upright or were in a supine position with neck extended, supported by a small pillow under the shoulder. The skin was cleaned with an antiseptic and using· a 21 or 23 gauge needle attached to 10 nil disposable syringe, aspirations were performed by applying constant suction pressure on the syringe plunger. In most patients two passes were made. However, depending on the material aspirated, upto 4 passes were carried out in some cases. Where the lesions were cystic, most of the cyst fluid was aspirated first and then a second aspirate was taken from the cyst wall to obtain an adequate, cellular aspirate. The patients were asked not to talk or swallow during the aspiration procedure, to avoid movements of the thyroid gland.. After the procedure, a piece of cotton swab pressed firmly against the needle puncture site(s) prevented any haematoma formation. Slides were prepared in a manner similar to that for blood smears. On average 4-8 slides were prepared in each case. Half of theslides were air dried, and the remaining wereimmediately fixed in 95 % alcohol. Any cyst fluidaspirated was transferred to the laboratory as such,for subsequent centrifugation and slide preparationfrom the sediment. In addition to cytology smears,aspirates with sufficient tissue particles were used tomake conventional paraffin embedded blocks.

The labelled unstained slides, accompaniedwith

. the cytology request forms were taken to the

main laboratory. The .air dried smears were stained with May - Grunwald Giernsa methoc.i8 , and the alcohol fixed were stained with Papanicolaou method9. 3 - 4 um sections were cut from the paraffin blocks and stained with conventional haematoxylin and eosin stain. The stained slides were examined and screened by the pathologist and

when the aspirate . was considered adequate, a diagnosis was made, as specific as possible. In most cases, reports were available to the clinician within 1 to 4 days.

As recommended by Lioe et al10, all reports of the thyroid aspirates performed were retrieved and categorised into the following groups:

*

*

*

*

*

*

*

Unsatisfactory for diagnosis. Non-neoplastic; including cysts, colloid goitre and inflammatory conditions. Indeterminate but benign. Follicular adenoma I Follicular neoplasm.

Hyperplastic colloid nodule vs. · Follicular adenoma. Suspicious malignant. Malignant.

An attempt was made to subclassify malignant lesion. Out of these 443 aspirates, in 97 (21.9%) cases, subsequently performed excision biopsies of the thyroid lesions were available for comparison with the cytodiagnosis.

Statistical analysis Correlation. between cytology and

histopathology was evaluated using Chi-square analysis. Phi co�fficient was used to increas� the strength of correlation. A P value � 0.05 was considered significant.

RESULTS

The 443 aspirates comprised of 379 (85.55%) fem4le and 64 (14.45%) male patients. The breakdown of different diagnostic category was as follows (Table - 1).

1. Unsatisfactory for diagnosisTwenty two (4.9%) aspirates were consideredinadequate for diagnosis. Majority of thesewere either acellular or contained v_ery fewepithelial cells. Reaspiration was done duringthe subsequent visit of the patient or excisionbiopsy was suggested.

2. Non-neoplastic

29

One hundred and fifty five (34. 9 % ) non­neoplastic lesions . were diagnosed. Theseincluded 101 (22. 8 % ) patients of rriultinodulargoitre I hyperplastic colloid nodule, 47

Suleman et al.

Table 1: FNA cytodiagnoses of the thyroid gland aspirates included in this study ( n = 443).

Categories

Inadequate for diagnosis Indeterminate but benign i:ollicular adenoma Vs. Hyperplastic colloid nodule. Follicular adenoma/follicular neoplasm Suspicious malignant Malignant

Follicular carcinoma Papillary carcinoma Anaplastic carcinoma Carcinoma NOS

Non neoplastic Lesions. Benign cyst MNG /HPCN Abscess Hashimoto's Thyroiditis NOS Granulomatous

TOTAL:

No. Of

Patienls

(%)

22(4.97) 31(6.99)

111(25.05)

67(15.12) 40(9.03) 17(3.84)

3 4 6 4

155(34.99)

101 2 l 2 2

443(100%)

Male

2 3

13

10 6

0 1 3 3

23 47

64(14.45)

SEX Age (Years)

Female Range

20 17 - 70 28 14 -70 98 13 - 90

57 16 - 70 34 10 - 75

3 25 - 50 3 24 - 35 3 34 - 80 l 52 - 60

132 6 - 75

379(85.55) 6 - 90

MNG, Multinodular goitre; HPCN, Hyperplastic colloid nodule; NOS, Not otherwise specified.

Average

35.2 36.9 32.2

37.4 40.0

41.7 29.0 57.0 57.3

36.3

Table 2: Analysis of the thyroid aspirate cytodiagnoses compared with the histological diagnosis of excision specimens (n=97).

FNA Cytology

Diagnosis No. OJCases Cyst

Inadequate 3 Cyst 6 FA /FN 19 HPCN /MNG 23 FA Vs. HPCN 26

Benign NOS 5 Suspicious Malignant 10

Malignant 5

TOTAL: 97

F.A.

3 4

16 4

10 1 2

40

Histological Diagnosis

MNGIHPCN

2 3

18

15

3

42

Malignant

l**

7 5

13

Others

l*

FA I FN, Follicular adenoma I Follicular neoplasm; HPCN I MNG = Hyperplastic colloid nodule I Multinodular goitre; NOS, Not otherwise specified *Non-specific inflammation, fibrosis and calcification.**Microscopic sized papillary carcinoma.

30

Fine Needle Aspiration Cytology

( 10. 6 % ) benign cysts, 2 (0 .45 % ) inflammatory lesions NOS, 2 (0.45%) non-specific abscesses (AFB negative), 2 (0.45%) granulomatous thyroiditis (AFB negative) and a single patient of Hashimoto' s thyroiditis.

3. Indeterminate but benign (Benign nototherwise specified, NOS)

4.

5.

6.

7.

Thirty one (6.99%) aspirates contained benignfollicular epithelial cells of non-descript typewith no associated or background features tosupport a specific disease entity. Theseaspirates were labelled as benign thyroidlesions with the concluding remarks that nomalignant cells were present.

Follicular adenoma I Follicular neoplasmIn 67 (15.12%) cases the diagnosis of follicularadenoma I Follicular neoplasm was made. Inthese patients the cytomorphological featureswere characteristic enough for rendering thisdiagnosis.

Hyperplastic colloid nodule Vs. FollicularadenomaThis group comprised of 111 (25. 05 % )patients. The lesions were grouped as benignand, due to the inherent inability to distinguishfollicular adenoma from a well differentiatedfollicular carcinoma cytologically, a diagnosisof "follicular neoplasm" was made in thepresence of mild cellular and nuclearpleomorphism or marked cellularity of theaspirate. An excision biopsy was stronglysuggested in these cases. On balance one or theother diagnosis was suggested as moreprobable. However, no specific diagnosis waschosen among these two entities.

Suspicious malignantForty (9. 03 % ) aspirates were labelled assuspicious malignant based on equivocalcellular and nuclear pleomorphism or scantyatypical cells. Excision biopsies were stronglysuggested in these cases.

MalignantIn 17 (3.84%) patients a definite diagnosis ofmalignancy was made. These comprised of 3follicular carcinomas, 4 papillary carcinomas,6 anaplastic carcinoma and 4 malignant lesionsNOS.

Sensitivity: Out of 13 malignant cases examined histologically, 12 were detected cytologically, · suggesting an overall sensitivity of 92.3%. Specificity: Out of 84 histologically benign cases, the cytological diagnosis of benign lesions was given in 81 cases with 3 false positive cases. The specificity therefore was 96.4%. Positive predictive value for malignancy on FNA thyroid is = 80%. Negative predictive value for malignancy on FNA thyroid is = 98.8%. False negative rate = 7. 7 % . False positive rate = 3.6%.

Significant correlation .was found between cytology and histopathology evaluation of FNA and biopsy results using Chi-square analysis (P<0.001).

A Phi coefficient of O. 836 indicated a very strong correlation between FNA cytology and histology of thyroid lesions.

CYTOMORPHOLOGICAL FEATURES

Clinical and radio-isotope scan findings were always correlated with the cytological picture in each case. The cytological features found helpful in the diagnosis of multinodular. goitre I hyperplastic collo,id nodule were epithelial cells with regular nuclei, oxyphilic cells, large follicles, honeycomb pattern, moderate to abundant colloid material and ·foamy macrophages (Figs. 1, 2) as also describedby Harach et ail I .

The cysts were usually obvious at the time ofprocedure when large quantities of brown, colloidstained . fluid was aspirated. The smears preparedfrom the centrifuged deposit were mostly cellularcontaining few degenerate cells, necrotic debris andabundant brown pigment laden macrophages withdirty, colloid stained background.

Follicular adenomas (Fig. 3) werecharacterised by good cellularity, large sheets ofcells with prominent microfollicular pattern andthree dimensional tissue fragments.. Smallernumbers of loose follicular cells and scanty colloidmaterial was also seen12

. Nuclear enlargement wasfrequent finding in follicular adenomas (Fig. 4).

31

Suleman et al.

'" '

Fig. I: Multinodular colloid goitre: Uniform cells are arranged in monolayers with vague, microfollicular pattern. Giemsa stain,

Fig. 2: Multinodular colloid goitre: In this field from the same case as Fig I, there are foamy �acrophages and colloid material. Giemsa stain.

Aspjrates from papillary carcinomas showed p.i.pillary tissue fragments, intranuclear inclusions and nuclear grooves (Fig .. 5)13, 14, 15 . Psammoma

· bodies and "chewing - gum'.'. colloid were not seenin our cases, although these features were detectedby other authors4 .

Fig. 3: Follicular adenoma. Hypercelluffiar � containing 3-dimentional tissue ffiiag,uemk. Papanicolaou stain.

Fig. 4: Follicular neoplasm I Adenoma. 'lllbmle iii; millil nuclear pleomorphism. The follicul:ur � is striking in this example. Giemsa stwn.

Follicular carcinomas yielded hi�,,

dyscohesive aspirates, showing obvious; mllhmm amnrl! nuclear pleomorphism. The nuclei wemi= JlamF \\Wtlbi irregular contours and the chromatin w.m; �­Many prominant nucleoli were seen amrdl � '1WcJB;

minimal colloid and follicular different-miliiomF<ii ..

32

Fine Needle Aspiration Cytofpgy

�ic carcinomas wei:e characterised by uru:ourlkreml codllhnlar and nuclear pleomrophism with �ia. Spindle shaped and multinucleated, � tumour giant cells were present. JP'n:m111miim,mml rne or more nucleoli were seen and there w.m; t1m1m:11mu[ 1diathesis in the background (Figs. 6,. . 71)'!7' ..

�ary carcinoma. The aspirate reveals lfDQliDary architecture. The cells are round to �e shaped and are moderately pleomorphic. Oacasional cells with intranuclear inclusions and IIDldtear grooves are present. Note the clumped dmmnatin. Giemsa stain.

lJno. '9J71 (�1. 9 % ) cases, surgical excision biopsies � �le for histological examination and <omm,g,i:aJrn'Sl)lUl with the cytodiagnosis, as shown in T�-l ..

� of the 97 surgically excised specimens � ttllimt \We did make mistakes in reaching a � almitgnosis on FNA cytology. In the 47· ((DJID .. (6,1%)) ccyltiologically diagnosed cysts, 6 were � excised. In these, 4 turned out to be WiliiomlbDr an!lenomas and 2 multinodular goitres � qstic degeneration. In the 26 cases from ttlme � 1of "follicular adenoma vs. hyperplastic COllillkllidl ll!IllXlftl!llle", after histological examination 15 � IIIJJIJlll)ttjjdular goitres, 10 follicular adenomas mrdl n lbxamii.gn cyst. To highlight further the � in the cytomorphology of these Jkesfui:ms:,,

3 W'licular adenomas ( on cytology) were in 1lim:tt,, �astic colloid nodules in mutlinodular

Fig. 6: Anaplastic carcinoma. Markedly pleomorphic, discohesive and hyperchromatic malignant cells. Note the large tumour giant cells. Giemsa stain.

Fig. 7: Anaplastic carcinoma. In this example, there is cellular pleomorphism anii nuclear hyperchromasia associated with striking cellular discohesion. Tumour diathesis can be seen in the background. Giemsa stain.

. . .

goitres on histology and 4 inultinodular goitres (on cytology) were follicular adenomas when surgical excision specimens were e?(amined. Out of 40 cases

33

Suleman et al.

of cytologically suspicious malignant lesions, in 10 patients surgical excision specimens were available. In these 2 were follicular adenomas and 1

rnultinodular goitre while the remaining 7 were confirmed as malignant. Five cytologically diagnosed definite malignant lesions were confirmed histologically as carcinomas.

DISCUSSION

Fine Needle Aspiration (FNA) cytology is an in expensive, atraumatic technique for the diagnosis of disease sites. The most obvious advantages of FNA cytology over surgical and large needle biopsy are that it is quicker to perform and report, less painful, less technically demanding and easily repeatable. It is far more convenient and may be set up in any clinical situation LS. There is little doubt that FNA cytology has in part replaced frozen sections and tissue biopsies, particularly for the diagnosis of cancer5 ,6 . However, the great danger of the method lies in erroneous interpretation of the sample by inexperienced observer without sufficient cognisance of the many diagnostic pitfalls19 .

(;arcinoma of the thyroid is relatively uncommon, however thyroid nodules occur frequently20. It is the clinically solitary, non-toxic (cold) thyroid nodule· which has usually been a strong indication for surgery because it is considered to be a highly suspicious sign of carcinoma21,22 . As FNA cytology has a higher sensitivity for the detection of malignancy compared with ultrasound and radio-isotope scan23, and with experienced aspirator in a well organised service, an FNA cytology clinic is highly cost effective, resulting in considerable savings in resourcesI0 ,21 , it has become an important aid in the diagnosis and management of palpable thyroid nodules 15 . The technique is now recognised to be the first line investigation for a solitary thyroid nodul&.

In the pr�sent study only 4.9% aspirates were considered inadequate for diagnosis. A review of the literature revealed a wide range of inadequate aspirates from 2. 2 % , 3 to as high as 25 % 10 in thyroid aspirates. Various authors have· stressed the importance, of multiple aspirates in one patient and an expensed cytopathologist making the aspirate him/herself 24. The lower figure of unsatisfactory .aspirates in 9ur study supports these suggestions as multiple passes were done ·in each case and the aspirates were carried out by the cytopathologist in

an FN.A outpatient clinic. . . 34.9% lesions ·were specifically_ categorised as.

non-neoplastic based on the characteristic cytomorphological features. Multinodular colloid goitre/hyperplastic colloid nodµle was the most frequently made di_agnosis, GOmp.ri.sing of 65 .2 % of the non-neoplastic category. Nodular hyperplastic goitre is one of the frequent cytologically diagnosed thyroid lesions in all series3 ,2 1.24 and our results are in agreement with these. This no surprise, as multinodular goitre is the most common cause of thyroid enlargement and a dominant nodule can

. present clinically as solitary nodule25 . It is mostly in this area, where a correct diagnosis, on FNA cytology can avoid unnecessary or excessive surgery.

Unfortunately, the aspirate from a multinodular goitre can show a spectrum of cytomorphological changes, so that erroneous diagnosis of benign cyst or neoplasia may be made4 ,11. In the 23 aspirates diagnosed as goitre and where histology was available afterwards, 18 were diagnosed aorrectly, the remaining being 4 follicular adenomas and 1

carcinoma. The missed" carcinoma was a microscopic sized papillary carcinoma ( < 1 cm diameter). Similar "missed" papillary ·carcinomas are frequently reported in the literature 26 .. Most of these lesion�., are .small . ( < 3 cmL in size . and sampling ,;�r_ron is, :�Onsipered ·. 9ne , Of the main reasons for .these·. false .negatiye ,, results. It is recommende1 that s .\lc::h �f�fcik�� ��y .be reduced by careful palpation, r,epeating .;�.aspiJation when necessary and possibly by the use of ultrason0g1;_aphy., in, §�lected c;a,�e�o _ _. ._

Nuol�?.X, ;pJ<;omorp.ll,isrp. �ani ocfµr ii;i benign endocri11e µeopl�.sm,s"�U;!�t.t.h!! 9p,fii;ii�e, criterion for the diagnosis of follicular carcinoma is _capsular and/or ws,.m.l�r1 ;inv�iPQ., :fl1ese f�f\tl;li;es, obviously, cannot .b� assess.eo iJ].' -er:tolog�cal , preparations. Therefo�e� ijne ,p.��dle,,, aspir,ation,q of follicular lesions sery�s_;as a .. sci;eening tq(ill for,the selection of cases . re,guii;ing exc1s1on and histological examinatjon ttPLmake the definitjve diagnosis of an adenoma O! �arcinoma on the basis of the architectural featl:lfes4. , Some au.fuors recommend

• .:. •• j • - • � �., � • '

that aspirat�s.. £rqm foll:icular adenomas_ and carcinomas sh(i)uld . be gr_<_>_uped and named as "follicular neop��m" on I FNA. cytology and this cyto diagnosis should be an indication for removal of the nodut�0,27. The distinctjon of a follicular adenoma is ,particularly qifficult from a well

34

Fine J:leedle Aspiration Cytology

differentiated follicular carcinoma. Consequently, follicular neoplasm reported by FNA that subsequently prove by histologic examination to be follicular carcinomas should not be regarded as false negatives 3. Certain features are helpful in making a diagnosis of follicular carcinoma on FNA cytology in comparatively less well differentiated lesions. These include good cellularity, cellular dyscohesion, three dimentional tissue fragments and more importantly cellular and nuclear pleomorphism, hyperchromasia, clumped chromatin and prominent and frequent nucleoli l2 , t6 . Based on these features we picked up 3 follicular carcinomas.

Due to its typical features, papillary carcinomas should be diagnosed even cytologically with high accuracy26 . Similar is the case with anaplastic carcinoma where the aspirate contains very pleomorphic obviously malignant cells17 .

Minor degrees of cellular pleomorphism is common in benign thyroid lesions including multinodular goitre and follicular adenomas 28. In our suspicious malignant: category both benign and malignant lesions were represented in cases where histology of the excision specimens was available. Interestingly, · one of the suspicious" lesion was Hurthle cell adenoma on histology, in which cellular and nuclear pleomorphism is a well known and common feature to the ex,tent that it can be misinterpreted as carcinoma on cytological examination3.

"Cysts" are well recognised diagnostic pitfall in FNA cytology of thyroid, as a diversity of thyroid lesions can exhibit cystic degeneration which appear as solitary cold nodules on radio­isotope scans. These include multinodular colloid goitre, follicular adenomas and more importantly papillary carcinomas4

•13

. It is due to .this reason that it is recommended to obtain an adequate cellular aspirate from cyst wall, after emptying the cyst contents, for proper cytological assessment of these lesions. This certainly is the case in recurrent cysts requiring repeated aspirations and these should be excised surgically for proper diagnosi�.

Excluding papillary and anaplastic carcinomas, the diagnostic accuracy of FNA cytology of the thyroid gland is limited in distinguishing between aspirates from cellular colloid nodule and follicular adenoma, because of similar and overlapping cytologic findings28 . A substantial proportion (25.05%) of this study belonged to this quivocal category. However in all cases a diagnosis of

benignity was made and one of the two diagnosis was considered more probable in each patient. In cases where histology was available, significant overlap between these two diagnostic categories was noted.

There are very occasional reports of complications arising due to the FNA thyroid including haematoma formation!, partial or complete infarct of the tumour22 and needle tract implantation of papillary carcinoma of the thyroid29. We did not encounter any of these complications in our series, apart from very few patients developing transient vasovagal shock when aspiration was attempted in an upright sitting position. These patients quickly recovered when laid down in supine position and the FNA procedure was successfully completed afterwards.

CONCLUSION

Because of the overlap of the spectrum of cytological features, FNA cytology of the thyroid gland is considered a difficult area. Despite this difficulty analysis of 97 cases where both FNA cytology and histology results were available, revealed that we were able to separate benign from malignant lesions apart from a single case of "missed" papillary carcinoma, due to the inherent sampling error. Most of the "suspicious malignant" lesions on cytology did turn out to be malignant on histological examination. If performed by an experienced cytopathologist, non-neoplastic lesions can be confidently separated from neoplastic conditions in most cases. The cytological features of papillary and anaplastic carcinomas are characteristic enough to make these straight forward diagnosis on FNA cytology. The techniqµe is certainly cost effective and is a valuable diagnostic tool in countries with limited resources, reducing greatly the number of unnecessary surgeries in thyroid gland, The use of FNA cytology in the evaluation of solitary thyroid nodules is, therefore, strongly recoQ1.Il1.ended. Due to tfie frequent occurrence of cystic degeneration in a diversity of thyroid lesions, both benign and malignant any "recurrent cyst" should be excised surgically for correct diagnosis.

. ACKNOWLEDGEMENT

We wish to thank Dr. Asrshad Kamal Butt for

35

Suleman et al.

his help in statistical analysis of the. results. We are also particularly thankful to the Surgeons and Physicians of Shaikh Zayed· Postgraduate Medical Institute for referring the cases for FNA cytology of thyroid.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

REFERENCES

Goellner JR, Gharib H, et al. Fine needle aspiration cytology of the thyroid, 1980 to 1986. Acta Cytologica 1987; 31: 587-90. Bhambhari S, Kashyap V, Das DK. Nuclear grooves in thyr:oid aspirates. Acta Cytologica 1990; 34: 809-12. Hsu C, Boey J. Diagnostic pitfall in fine needle aspiration of rhyroid nodules. Acta Cytologica 1987; 31: 699�703. Buley ID. Problems in fine needle' aspiration of the rhyroid. Curr Diagn Pathol 1995; 2: 23-31. Suleman BA, Latif A, Chughtai NM. Role of fine needle aspiration cytology in diagnosis of Phylloides tumour of the breast. Proceeding S.Z.P.G.M.I 1991; 8(3-4): 125-9. Suleman BA, Latif A. Evaluation of diagnostic accuracy of fine needle aspiration cytology in palpable breast lesions. Proceeding S.Z.P.G.M.I. 1996; 10(1-2): 55-62. Suleman BA, Latif A, Chughtai N, et al. Evaluation of ultrasound guided fine needle aspiration cytology in diagnosis of space occupying lesions of liver. The professional 1995; 2(4): 276-88. Inwood MJ, Thomson S, Bryant NJ. Practice of h�ematology. In: Lynch : Medical Laboratory Technology (ed. Raphael SS), 4th edition 1983; pp. 686. Igaku - Shoin Saunders, Tokyo. Bourne LD. Exfoliative cytology. In: Theory and practice of histological techniques (eds. Bancroft J, Stevens A), 2nd ed. 1987; pp. 429-30. Edinburgh: Churchill, Livingstone. Lioe TF, Elliott H, Allen DC, et al. A 3 year audit of thyroid fine needle aspirates. Cytopathology 1998; 9: 188-92. Harach HR, Zusman SB, Day ES. Nodular goitre: a histocytological study with some emphasis on pitfalls of fine needle aspiration cytology. Diagn Cytopathol 1992; 8: 409-19. Kaur A, Jayaram G. Thyroid tumours: Cytomorphology of follicular neoplasms. Diagn Cytopathol 1991; 7: 469-72.

13. Kulacoglu S, Ashton - Key M, Buley I. Pitfalls in thediagnosis of papillary carcinoma of thyroid.Cytopathology 1998; 9: 193-200.

14. Kini SR, Miller JM, Hamburger JH, et al. Cytopathologyof papillary carcinoma of the thyroid by fine needleaspiration. Acta Cytologica 1980; 6: 511-21.

15. Miller TR, Bottles K, Holly EA, et al. A step-wiselogistic regression analysis of papillary carcinoma of thethyroid. Acta Cytologica 1986; 3: 285-93.

16. Suen KC. How does one separate cellular follicularlesions of the thyroid by fine needle aspiration biopsy ?

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

Diagn Cytopathol 1988; I: 78-81 Guarda LA, Peterson CE, Hall W, et al. Anaplastic thyroid carcinoma: Cytomorphology and clinical implications of fine needle aspiration. Diagn Cytopathol 1991; 7: 63-7. Lever JV, Trott PA, Webb AJ. Fine needle aspiration cytology. J Clin Pathol 1985; 38: 1 -11. Koss LG. Thin needle aspiration biopsy. Acta Cytologica 1980; I: 1-3. Klemi PJ, Joensuu HJ, NyLamo E. Fine needle aspiration biopsy in the diagnosis of thyroid nodules. Ac.ta Cytologica 1991; 4: 434-38. Cagneten CB et al. The role of fine needle aspiration biopsr cytology in the evaluation of the clinically solitary thyroid nodules. Acta Cytologica 1987; 5: 595-98. Rosai I. Thyroid gland. In: Ackerman's Surgical Pathology (ed. Rosai J), 8th ed. 1996; pp 544. Mosby -Year Book, Inc St. Louis. Fon LJ,. Deans GT, Lioe TF, et al. An audit of thyroidsurgery m a general surgical unit. Ann R Coll Surg Engl 1996; 78: 192-96. Altavilla G, Pascale M, Nenci I. Fine needle aspiration cytology of thyroid gland diseases. Acta Cytologica 1990; 2: 251-56. The endocrine System In: Pathologic Basis of disease (eds. Cortran RS, Kumar V, Robbins SL), 5th ed. 1994; pp. 1131-33. Schmid KW, et al. Papillary carcinoma of the thyroid gland. Acta Cytologica 1987; 5: 591-94. Lowhagen T, Williams JS, Lundell G, et al. Aspiration biopsy cytology in diagnosis of thyroid cancer. World J Surg 1981; 5: 61-73. Ravinsky E, Safneck JR. Fine needle aspirates of follicular lesions of the thyroid gland. The intermediate type smear. Acta cytologica 1990; 6: 813-20. Hales MS, Hsu FSF. Needle tract implantation of papillary carcinoma of the thyroid following aspiration biopsy. Acta Cytologica 1990; 6: 801-4.

The Authors:

Bilquies A. Suleman Assistant Professor Department of Histopathology· Sheikh Zayed Postgraduate Medical Institute Lahore, Pakistan.

Abdul Latif Khan Specialist Registrar Histopathology Royal Liverpool University Hospital Liverpool, United Kingdom

Address For Correspondence:

Dr. Bilquies A. Suleman Assistant Professor Department of Histopathology Shaikh Zayed Postgraduate Medical Institute Lahore, Pakistan.

36