The Role Of Fine Needle Aspiration Cytology In Palpable Head And Neck Masses.Correspondence Address :
Dr Hilda Fernandes Professor,Dept of
pathology Fr Muller Medical college
Background And Objective: This study is done to evaluate the role of FNAC in palpable head and neck masses and also to study their distribution. A correlation was done between cytology and histopathology whenever surgical specimens were available and to assess the accuracy, sensitivity, specificity, positive predictive value and negative predictive value in various head and neck lesions.
Materials And Methods: From 620 cases, FNA smears were taken and stained with PAP, MGG and special stains whenever required. FNA results were interpreted and analysed according to the anatomical sites and the lesions were categorized into inflammatory and neoplastic conditions. The cytological findings were compared with those of histopathology wherever available. With the help of K value, the agreement between the methods was de-termined.
Results: Among 620 cases, histopathological correlations were available only in 129 cases. The sensitivity, specificity, predictive value of the positive test, predictive value of the negative test and false negatives of the thyroid lesions which were being detected were 83.33%, 100%, 100%, 97% and 16.66%, respectively. There were no false positives. The diag-nostic accuracy of the salivary gland, lymph node and soft tissue lesions were 100%.
Interpretation And Conclusion: There was perfect agreement in a majority of the lesions. The technique is simple, safe, convenient and an accurate method for tissue di-agnosis. Hence, FNAC is an effective diagnostic tool in the diagnosis of head and neck masses.
FNAC, histopathology, thyroid, lymph nodes, salivary glands, soft tissue.
Fine needle aspiration cytology (FNAC) is one of the most valuable tests available in the initial assessment of the patient who presents with a mass in the head and neck region or where a re-currence is suspected after previous treatment1. It is accurate, inexpensive and quick. The tissues which are most frequently sampled are lymph nodes, thyroid and major salivary glands. The present study was done to evaluate the role of FNAC in palpable head and neck masses.
FNAC was done on 624 patients who presented with palpable head and neck masses in a tertiary hospital for a period of 2 years from March 2005 to March 2007. Prior to FNAC, the patients were examined in detail, which included the recording of their pertinent clinical history and significant clinical findings. Relevant investigations were carried out as per requirements. After a brief explanation of the technique, an informed con-sent of the patient was obtained.
FNA was done using a 23 gauge needle fitted to a 10ml disposable syringe. An average of 2 passes was performed and some slides were air dried and stained by the May-Grunwald Giemsa stain. The rest of the slides were fixed in metha-nol and stained by the Papanicolaou stain. The Zeihl-Neelsenâ€™s stain for AFB was done in those cases with lymph node swelling, where the clinical suspicion or diagnosis was tuberculosis and /or in those cases where purulent or cheesy material was aspirated. Surgically excised specimens were available in 129 cases, which were routinely processed and stained with Haematoxylin and Eosin stains.
Among the 624 patients, 4 were excluded from the study as the smears were unsatisfactory. The distribution of the 620 cases are given in [Ta-ble/Fig 1].The accuracy of FNAC was verified by histological examination in the 129 patients. Among the 129 patients, 24 were males and 105 were females (M:F ratio =1:4.41). Thyroid gland (71.31%) was the commonest site aspirated, fol-lowed by lymph node (22.48%), salivary gland (3.87%) and soft tissue lesions (2.32%).
Patients with a thyroid swelling comprised of 13 males and 79 females and their ages ranged from 18- 73 years. The commonest lesion encountered in the thyroid gland was Nodular goitre, fol-lowed by Hashimotoâ€™s thyroiditis. Among the malignant neoplasms, Papillary carcinoma was the most common lesion noted. Fifty two cases of histologically proven nodular goitres were correctly diagnosed by the FNA cytology. Out of the fifty two cases, smears from fifty cases were moderately cellular and two were hypocel-lular. The smears showed follicular cells in clus-ters which were scattered singly. Forty eight cases had a few cyst macrophages and thick col-loid in the background. Four cases showed nu-merous cyst macrophages in a background of thick and thin colloid.
Ten cases of adenomatous hyperplasia were con-firmed by histological diagnosis. The smears were moderately cellular and showed follicular cells in clusters and tissue fragments in nine cases. Fire flares were seen in seven cases. Papillaroid fragments were seen in three cases. The colloid was either scanty or absent in all these cases.
Hashimotoâ€™s thyroiditis was detected by the FNA cytology in eleven out of twelve cases. The smears showed large clusters of Hurthle cells and lymphocytes in nine cases. Multi-nucleate giant cells and epitheloid histio-cytes were seen in three cases. Cyst macro-phages were seen in one case. The colloid was scant to absent in ten cases. One case showed the presence of thick colloid. Three cases showed fire flares.
Thyroglossal cysts yielded a clear yellow col-oured fluid. The smears were hypocellular and showed follicular cells, lymphocytes and cyst macrophages.
Papillary carcinoma of the thyroid was detected by the FNA cytology in nine out of eleven cases. The smears were hypercellular and showed pap-illary tissue fragments (Table/Fig 2) and syncytial aggregates in six cases. Nuclear grooves were seen in seven cases and intranu-clear cytoplasmic inclusions were seen in three cases. Cyst macrophages were seen in three cases. Psammoma bodies were seen in one case. Thick colloid was seen in five cases.
One case, which was suspected for malignancy, showed moderately cellular smears comprising of small papillaroid fragments and numerous cyst macrophages. Few cells with dense cyto-plasm and nuclear grooves were seen. The histopathological diagnosis made, was that of papillary carcinoma.
Four follicular adenomas in the series were des-ignated as follicular neoplasms in cytology. Three were moderately cellular and one was hy-pocellular. The smears showed follicular cells in a repetitive manner and scant colloid in a haem-orrhagic background. One case showed syncytial aggregates and microfollicles.
One case of Hurthle cell neoplasm showed hy-percellular smears (Table/Fig 3) with predomi-nant Hurthle cells in sheets and clusters and a few follicular cells. Cyst macrophages and col-loid were also seen in the background.
One case of medullary carcinoma was correctly diagnosed by cytology. The smears were hyper-cellular and showed tumour cells dispersed sin-gly and in sheets. The tumour cells exhibited fine granular nuclear chromatin and moderate amounts of granular cytoplasm. A few of them had a plasmacytoid appearance. Fragments of hyaline material, lymphocytes, cyst macro-phages, focal calcification and scant colloid were seen in the background. The overall accu-racy rate for the thyroid series was 96.3%.
Patients with lymph node swelling com-prised of 8 males and 21 females and their ages ranged from 2-76 years. Reactive lym-phadenopathy was the commonest cause of lymphadenopathy, followed by tuberculous and granulomatous, metastatic lymphade-nopathy. The present study did not make any attempt to categorize the type of Reactive lymphadenopathy. All ten cases of tubercu-lous lymphadenopathy showed moderately cellular smears. There were epithelioid granulomas (Table/Fig 4) in seven cases and caseation necrosis in three cases. The special stain for acid fast bacilli was positive in four cases.
One case of suppurative lymphadenitis showed neutrophils in a necrotic back-ground. No granulomas/giant cells were seen. The special stain for AFB showed negative re-sults. This case was confirmed by histological diagnosis.
Four cases of granulomatous lymphadenitis were correctly diagnosed by cytology. The smears showed focal collections of epithelioid cells in association with reactive lymphoid cells. The special stain for AFB showed negative results.
One case of metastatic carcinoma showed tu-mour cells in sheets and they were dispersed singly. These cells exhibited pleomorphism and had hyperchromatic nuclei with dense cytoplasm in a necrotic background. This was subsequently confirmed by histopathology.
Smears from Langerhans cell histiocytosis re-vealed histiocytic cells in clusters and they were dispersed singly. Binucleate and multinucleate forms were of cells were seen along with eosi-nophils in the background. A subsequent biopsy confirmed the cytological diagnosis.Patients with salivary gland lesions comprised of 3 males and 2 females and their ages ranged from 20-49 years. The commonest benign and malignant tumours reported were Pleomorphic adenoma and Mucoepidermoid carcinoma, respectively. One case of benign cystic lesion of the salivary gland showed cyst macrophages, lymphocytes and stromal fragments in a proteinaceous back-ground. Three cases of pleomorphic adenoma showed moderately cellular smears. Epithelial cells in clusters and sheets were seen in two cases. Fibrillary chondromyxoid ground sub-stances were seen in all the three cases. One case showed plasmacytoid cells with well defined cell borders. Cyst macrophages were seen in one case. One case of mucoepidermoid carcinoma was confirmed by histological diagnosis. The smears were hypercellular and showed cohesive clusters and sheets of tumour cells. Some of the cells had cytoplasmic vacuolization. The back-ground was dirty, with mucus and necrotic de-bris. The diagnostic accuracy for salivary gland lesions was 100%.
Patients with soft tissue lesions comprised of 3 females and their ages ranged from 13-60 years. One case each of Lipoma, Hamartoma and Spindle cell tumour were cytologically diag-nosed and confirmed by histopathology.
The correlation between cytological diagnosis and subsequent histological studies in all the 129 cases is shown in (Table/Fig 5). In the 129 cases, the sensitivity was 87.5%, the specificity was 100%, the positive predictive value was 100%, the negative predictive value was 98.26% and false negatives were 12.5%.
Head and neck masses often pose a challenging diagnostic problem to the clinician. Malignancy remains an important differential diagnosis and neck mass is often the first or the only symptom of this disease. Although surgical biopsy is the commonest method of tissue diagnosis, FNAC is in practice since the 1930s.This method has be-come popular as a diagnostic step in the evalua-tion of a head and neck mass (2).
The results of 620 aspirates from head and neck masses have been categorized into inflamma-tory, benign and malignant lesions. Four aspi-rates (0.64%) were excluded, as they were in-adequate. The incidence of inadequate or unsat-isfactory samples in various studies has ranged from 0-25% (3). Unsatisfactory aspirates were the result of poor handling of the aspirated mate-rial and the lack of trained cytopathologists. In-adequacy was also attributable to the small size of the lesions (4).
Two cytologically diagnosed nodular goitres turned out to be papillary carcinoma after they were studied histopathologically. The slides on review revealed follicular cells in clusters and singles in a background of thick colloid. How-ever, no papillary fragments/ nuclear grooves/ intranuclear cytoplasmic inclusions were seen. The causes of false negative results were the poorly cellular sample in a large cystic papillary carcinoma and the thick fibrous capsule. Gag-neten stressed the importance of doing multiple aspirations in a thyroid swelling in order to ob-tain representative samples (1).
One case of thyroglossal cyst was correctly di-agnosed and the accuracy was similar to that in the study done by Hsu and Boey (5). In our study, the patient was 40 years old. It is well known that thyroglossal cyst presents clinically in children, but lesions can also be seen in adults even late in life (4).
The diagnostic errors were most commonly due to inadequate specimens and cystic lesions. One must be careful in committing a false negative diagnostic error in cystic lesions that contain macrophages and scanty material, since these features do not exclude malignancy. Repeat FNAC or thyroidectomy is advised for persistent nodules (5),(6). Cystic thyroid lesions pose di-agnostic difficulties. Cystic change and/or haemorrhage in neoplasms is seen in upto 25% of primary Papillary carcinomas, in 20% of Fol-licular neoplasms and in 26% of Follicular car-cinomas (1). Recurrent cysts, incompletely de-compressed lesions, lesions greater than 3-4 cm in diameter in which aspiration of several areas does not give good evidence of the colloid nod-ule and lesions in young males, have all been recommended as indications for surgical exci-sion.Intranuclear cytoplasmic inclusions and psammoma bodies detected in up to 83% and 24% of cases of Papillary thyroid carcinoma,(7) were seen in only three cases (33.3%) and one case (11.1%) respectively, in the present study.
One case of Nodular goitre turned out to be Hashimotoâ€™s thyroiditis after studying it histopa-thologically. The slides on review revealed hy-pocellular smears comprising of few follicular cells and scanty thick colloid. No Hurthle cells/lymphocytes were seen. Jayaram (7), in her study, misdiagnosed two cases as colloid goiters. Hurthle cells and lymphocytes were not be pre-sent in the smears, which probably are examples of non representative sampling. Hashimotoâ€™s thyroiditis was sometimes a problem, since aspi-rates of that lesion, not uncommonly show a marked proliferation of Hurthle cells or follicu-lar cells in nodular masses. A needle sampling of such an area would yield a cytological picture similar to that of Follicular or Hurthle cell tu-mour (7).
Follicular and Hurthle cell neoplasms posed no diagnostic problems. Aspirates from hyperplas-tic areas of some goitres presented a picture similar to that of Follicular neoplasms. This po-tential cause of misdiagnosis was counteracted in most cases by the sampling of two to three different areas of the thyroid nodule (7).The ma-jor limitation of FNAC in cases of tumour of the thyroid, is in the evaluation of the nature of the neoplasm, which is done by histopathology.
Literature survey has shown that the FNAC di-agnostic accuracy rate in tuberculous lymphade-nitis is as high as 90-100%. FNAC coupled with ZN staining for AFB is a very useful diagnostic tool in the diagnosis of tuberculous lymphadeni-tis.There are problems in arriving at a definitive diagnosis in certain cases of Tuberculous lym-phadenitis, when the aspirate shows a polymor-phous picture with occasional epithelioid cells, with an absence of Langhanâ€™s giant cells or cas-eous necrosis, making it necessary to resort to excisional biopsy for a definitive diagnosis. This is particularly true in children, in whom a similar picture may be seen in cases of reactive hyper-plasia due to viral or Toxoplasma infection, since the mere presence of the epithelioid cells is not diagnostic of any specific condition (8).
Two cases of Reactive lymphadenopathy diag-nosed cytologically, proved to be Kikuchiâ€™s lym-phadenitis after histopathological studies. The slides were reviewed to look for the typical cy-tomorphological features. There were no ne-crotic debris or plasmacytoid monocytes. Few clusters of histiocytes were seen. The special stain for AFB showed negative results. Tong et al. (9), in their study, had a similar problem.
Smears from Catscratch disease did not show typical cytological features. Literature survey showed that the cytology in early lesions of Cat scratch disease may be nonspecific, with a mix-ture of lymphocytes, immunoblasts, macro-phages, plasma cells and neutrophils (10).Characteristic granulomas with peripherally palisading epithelioid histiocytes and centrally located neutrophils and an associated polymor-phic cell population, are observed in Cat scratch disease. Cytological appreciation of the suppura-tive granulomas can be expected in FNA of the intermediate and the advanced lesions. The cyto-logical differential diagnosis includes any dis-ease in which epithelioid cells and/or granulo-mas occur. (10).
One case diagnosed by cytology as Reactive lymphadenopathy proved to be Castlemanâ€™s disease after histopathological studies. The plasma cell lesion represents an earlier, more active stage and the hyaline vascular type represents a later stage. The cytological characterization of Castlemanâ€™s disease may be difficult(11).
Among the salivary gland lesions, the pa-rotid was the most commonly involved gland. Our observation is similar to that of Cristallini et al.(12) and Cajulis et al.,(13). Among benign tumours, pleomorphic ade-noma was the commonest tumour and among the malignant tumours mucoepider-moid carcinoma was the most common one. This finding is similar to that of Fernandes et al.(14). The overall diagnostic accuracy was 100%. Review of literature shows that the accuracy has ranged from 80.4 to 98% (14).
One case of Lipoma was correctly diagnosed by cytology. One case of Hamartoma diag-nosed by cytology proved to be Capillary haemangioma after histological studies. One case of Spindle cell tumour which was sus-pected to be malignant turned out to be ma-lignant peripheral nerve sheath tumour. FNAC usually faces no problem in distin-guishing high grade soft tissue sarcomas from benign lesions. However, borderline and low grade lesions are susceptible to be missed. Accurate typing and grading of the tumour is not possible in many cases by FNAC alone. Almost all studies on soft tis-sue tumours have reported this limitation of FNAC (15).
The present study highlights certain limita-tions of the procedure in the head and neck region, namely:
- Typing the various Reactive lymphade-nopathies (9),(10),(11).
- To categorize the borderline and malig-nant soft tissue tumours (15).
- To differentiate Colloid goiter from the Follicular variant of papillary carcinoma (1).
On comparing the results of the present se-ries with other workers (Table/Fig 6), it can be said that the results of this study are fa-vourable with those published in literature and are fairly accurate.
Fine Needle Aspiration Cytology is a rapid, convenient and accurate method of tissue diag-nosis that can be done on an out patient basis. FNAC offers a simple method of diagnosis of neoplastic and non neoplastic lesions of the head and neck. The procedure is safe and free from complications and is well tolerated by the pa-tients. There is no need of anaesthesia and speedy results are obtained. It serves as a com-plementary diagnostic procedure to histopa-thological examination. There was almost per-fect agreement between the cytological and his-tological findings and there was fairly good ac-curacy. There were only two false negatives and no false positives in our study. Hence, we con-clude that Fine Needle Aspiration Cytology is a highly effective diagnostic procedure in the di-agnosis and management of palpable head and neck masses.
FNAC is an effective diagnostic tool in the diagnosis of head and neck masses.
- Emerging Sources Citation Index (Web of Science, thomsonreuters)
- Index Copernicus ICV 2017: 134.54
- Academic Search Complete Database
- Directory of Open Access Journals (DOAJ)
- Google Scholar
- HINARI Access to Research in Health Programme
- Indian Science Abstracts (ISA)
- Journal seek Database
- Popline (reproductive health literature)