INTERNATIONAL JOURNAL OF IMMUNOPATHOLOGY AND PHARMACOLOGY

Vol. 26, no. 4, 965-971 (2013)

LETTER TO THE EDITOR PREOPERATIVE ASSESSMENT OF SALIVARY GLAND NEOPLASMS WITH FINE NEEDLE ASPIRATION CYTOLOGY AND ECHOGRAPHY: A RETROSPECTIVE ANALYSIS OF 357 CASES A. PASTORE l , M. BORINl, N. MALAGUTTP,A. DI LAORAl, D. BECCATIl, A.L. DELAZER2 , C. BIANCHINI!, F. STOMEO!, A. CIORBA 1 and S. PELUCCHP lENT Department, University Hospital ofFerrara, Ferrara, Italy; 'Pathology Department, University Hospital ofFerrara, Ferrara, Italy Received June 11, 2013 - Accepted October 11, 2013 Fine-needle aspiration cytology (FNAC) is a minimally invasive procedure usually well tolerated, easy to perform, quick, cheap and easy to repeat in case of doubts or non-diagnostic results. Echography is also a fast, cheap and non-invasive tool; however, the role of FNAC and echography in the diagnosis of salivary gland pathology is not universally recognised. Three hundred and fifty-seven patients with a cytological diagnosis at FNAC, and 247 of these who were also studied with echography, were enrolled for this retrospective study. The final histopathological diagnoses, obtained after surgery, were then compared to the preoperative FNAC diagnoses and echographic findings. From the analysis of our data, the overall FNAC specificity resulted 93%, sensitivity 83%, and diagnostic accuracy 92%. Echography sensibility was 57.1% specificity 98.2%, while positive and negative predictive value were respectively 80% and 94.8%. While echography can be useful in order to provide a better characterization of salivary gland lesions, FNAC can then be considered a safe diagnostic tool with reliable sensitivity and specificity for the assessment of salivary gland pathology and thus for selecting patients and indicating the best surgical treatment. It has been reported that major salivary gland tumours represent approximately 3% of all head and neck tumours; 80% involve parotid gland and 75% are benign neoplasms (l). According to the most recent WHO histological classification (2005) there is a broad spectrum of different histotypes of major salivary gland tumours (2), thus it is necessary to make a correct preoperative diagnosis in order to decide the best surgical/therapeutical approach. In this way, in order to perform a correct preoperative assessment, we used fine needle aspiration cytology (FNAC) and echography. FNAC is a minimally invasive method that does not require anaesthesia; it is well tolerated

by the patient, easy to perform, quick, with rare complications, cheap and can be easily repeated in case of doubts or non-diagnostic results in order to reach a more accurate diagnosis (3). Echography is also a fast, cheap and non invasive tool. However, the role of FNAC and echography in the diagnosis of salivary gland pathology is not universally recognised (4, 5). Since its use is still controversial, most ENT surgeons prefer intraoperative frozen section examination to preoperative FNAC (6-7). The aim of this study is to show the accuracy and reliability of FNAC for the diagnosis of benign and malign tumours of major salivary glands through the evaluation of its diagnostic accuracy-sensibility,

Key words: salivary gland tumours, fine needle aspiration cytology, FNAC, echography, diagnosis Mailing address: Francesco Stomeo, MD ENT Department, University Hospital of Ferrara Via A. Moro 8, 44100 Ferrara (Cona),ltaly Tel.: +39 0532 237447 email: [email protected]

0394-6320 (2013)

965

Copyright © by BIOLIFE, s.a.s. This publication and/or article is for individual use only and may not be further reproduced without written permission from the copyright holder. Unauthorized reproduction may result in financial and other penalties DISCLOSURE: ALL AUTHORS REPORT NO CONFLICTS OF INTEREST RELEVANT TO THIS ARTICLE.

966

A. PASTORE ET AL.

specificity, negative predictive value (NPY), and positive predictive value (PPY) on a large group of patients. We also evaluated the contribution that echography could provide for the assessment of these lesions. MATERIALS AND METHODS This is a retrospective study of 357 patients, with a cytological diagnosis after FNAC, who were surgically treated at the Otolaryngology Department of the University Hospital of Ferrara from January 2001 until December 2011. Of these, 3 16 underwent a parotidectomy while 41 underwent a submandibular gland excision in cervicotomy. The preoperative staging of the lesions was performed by clinical examination, ultrasound examination, FNAC, and in some cases by a head and neck CT scan and/or a head and neck MRI. FNAC

The FNAC cytology was usually performed by the ENT specialist in the outpatient clinic. The procedure was carried out by a vacuum system (CAMECO, 23-25G needles) and without ultrasound guidance, in case of a palpable lesion. Samples were immediately processed, at the Institute of Pathology, University Hospital of Ferrara, (May-Grunwald Giemsa staining), and the diagnosis was made by an experienced pathologist. Cytology results were then divided into six categories: benign cytological sample = SO; non-diagnostic cytology sample = S1; cytologic sample addressing a benign neoplasm = S2; cytologic sample with atypical cytological findings = S3; cytologic sample suspicious for malignancy = S4; cytologic sample addressing a malignant neoplasm = S5. The cases SO, S 1 and S2 were considered as negative (non-malignant) while S3, S4, S5 as positive (malignant). The final histological diagnoses were divided into three categories: I) non-neoplastic lesion; 2) benign neoplasm; 3) malignant neoplasm. The final histological diagnosis of non-neoplastic lesions and benign neoplasm were combined in order to have only two preoperative sets: benign and malignant tumours. The final histopathological diagnoses were then compared to the preoperative FNAC diagnoses, in order to evaluate FNAC accuracy-sensibility, specificity, negative predictive value (NPY), and positive predictive value (PPY). Echography Two hundred and eighteen of the 316 patients with parotid tumours and 29 of 41 patients with submandibular

gland tumours, underwent an echographic ultrasound examination (247 of 357 patients underwent ultrasound, 70% of the studied group). Each ultrasound report described the following characteristics of the lesion: I) morphology; 2) margins; 3) dimensions; 4) echogenicity; 5) echostructure. The reports were then divided in order to assess: I) a suspected benign lesion; 2) a suspected malignant lesion. Parameters to suspect a malignant lesion were an irregular morphology, margins not defined, an non-homogeneous echostructure and size with a major axis greater than 2 cm. We then compared the final histopathological diagnosis with the corresponding diagnostic categories obtained from the ultrasound evaluation. Statistical analysis An SPSS software (vers.12.0) for Windows operating systems was utilized. The analysis of biological and clinical variables was conducted using the Chi-square tests.

RESULTS FNAC Of the 357 lesions that underwent FNAC (and definitive histological examination), showed that 285 cases (79.8%) were benign tumour, 37 cases (10.4%) were non-neoplastic lesions, and 35 cases (9.8%) were malignant pathologies. In particular, of the latter, 8 of 35 resulted metastatic squamous carcinoma cells, 7 of 35 resulted carcinoma with primitive squamous cells, 4 of 35 adenoidocystic carcinoma cells, 3 of 35 malignant lymphoma cells, 2 of 35 anaplastic carcinoma. Among the benign lesions, 146 of 285 were pleomorphic adenoma cells, 112 of 285 were cistoadenolinfoma cells, 17 of 285 basal cell adenoma cells. Non-neoplastic lesions identified at FNAC were, chronic sclerosing scialoadenitis (lO cases), cystic neoplasms (8 cases), chronic scialoadenitis (6 cases), 4 abscess lesions and few minor injuries. The correlation between FNAC and definitive histological examination was 82.9% when considering malignant histotypes. Particularly, the FNAC results 00/35 (8.6%) cases did not agree with the final histological result; one case was epidermoid

cysts at FNAC, and resulted a squamous cell carcinoma at the final histology; another case was a pleomorphic adenoma at FNAC, while it resulted an

967

Int. J. Immunopathol. Pharmacol.

Table I. Malignant histotypefrequency and cyto-histological (FNAC) concordanc es.

FREQ 1 1 1 7 4 2 1 1 1 1 3 8 1 1 1 11 35

M ALIGNANT HISTOTYPE !Acinar cell carcinoma Basal cell adenocarcinoma [NAS adenocarcinoma Squamous cell carcinoma [Adenoido-cystic carcinoma !Anaplastic carcinoma lDuctal carcinoma ~pi the l i a l -myoepi the l i a l carcinoma Small Merkel-like cells neuroendocrine carcinoma Carcinosarcoma (true malignant mixed tumor) lLymphoma Me tastatic squamous cell carcinoma Metastatic angiosarcoma Metastatic melanoma Metastatic kidney clear cell carcinoma rror AL METASTASES IrOT

% CONC 1 3 1 3 3 1 20 5 11 3 2 6 3 1 3 1 1 3 3 1 9 2 23 7 3 1 3 0 1 3 34 9 100 29

% 100 100 100 71 75 100 100 100 100 100 67 88 100 0 100 75 83

Table II. Benign histotype frequency and cyto-histological (FNAC) concordances. ~EN IGN HISTOTYPE

Pleomorphic adenoma Cystadenolymphoma Basal cells adenoma u.rpoma Oncocytoma M ioepitelioma Papillary Cystadenoma [I'otal

FREQUENCY 146 11 2 17 3 3 2 2 285

adenoidocystic carcinoma at the final histology; the latter case was a granulocytic inflammation at FNAC, and the definitive histology disclosed a metastatic squamous cell carcinoma. The correlation between FNAC and benign lesions was 85.4% (overall specificity); the highest correlation was demonstrated for pleomorphic adenoma (97%, 141/146), followed by Warthin tumour (88%, 98/112). There were only 3 false positive cases. When dividing FNAC diagnosis into benign and malignant, 299 of 302 benign cases were confirmed histologically (while 3 were malignant,

% 51 39 6 1 1 1 1 100

CONC 141 98 7 1 2 2 0 251

% 97 88 41 33 67 100 0 88

false negative), while 29 of32 malignant cases were confirmed histologically (3 were then false positive). Thus, the overall FNAC specificity was 93% (299/322), the sensitivity 83% (29/35), and the diagnostic accuracy 92% (299 +29/357). The positive predictive value was then 100% for categories S4 and S5, while it was 50% for category S3, with an overall PPY of90.6% (29/32). The negative predictive value was 99% (299/302). Echography

We also evaluated the echography reliability, comparing the ultrasound appearance with the final

A. PASTORE ET AL.

968

Table III. FNAC accuracy in relation to the final histology.

C YTO LO GY

NON-NEOPLA ST ]C LES ]ON

SO beni gn lesion S I inadequate S2 beni gn neopl asia S3 malignant abnomalities/doubts

24

S4 mali gn suspect S5 malign Tota l

HISTOLOGY BEN]GN M ALIGN NE O PLAS]A NEOPLASI A 3 I

47 23 255

3

21 II 253 0

3

6

0 0 37

0 0 285

7 19 35

7 19 357

9 I

2

Tot

Table IV. FNAC accuracy in relation to the final histology.

CYTOLOGY linadeguate benign/negative neoformation Im align/ positive I . n eo f ormation !:.

!fotal

BENIGN NEOFO RMAnON 20

HISTOLOG Y 1 MALIGN N EO FORMATION 3

Tot 23

I

299

3

302

3 322

29

32 357

histological results. Of 58 cases (92%) of benign suspects, only 55 actually had a benign neoformation confirmed by the histological examination (true negative). Among the 5 cases of malignant suspects (8%), 4 showed a malignant report following the histological examination (true positive). The false negative and positive were respectively 3 and 1. On the basis of these data, echography sensibility was 57.1% (417) and the specificity 98.2% (55/56). The positive and negative predictive value were respectively 80% and 94.8%. We also considered the diagnostic accuracy of FNAC combined with the ultrasound investigation. The cytological diagnoses associated with the

35

echography were divided into two groups: S2 (suspected benign) and S4 (suspected malignancy). The first group included 56 cases, 55 of which were confirmed histologically as benign lesions (true negatives), while one was a malignant lesion (false negatives). The malignant results were 7 in total, 4 of which were confirmed at the histological verification (true positives) while 3 were benign lesions (false positives), with an accuracy of93.7% (55+4/63). DISCUSSION It has been reported that FNAC is a minimally invasive method usually well tolerated by the patient,

969

Inl. J. Immunopalhol. Pharmacol.

easy to perform, quick, with rare complications, cheap and easy to repeat in case of doubts or nondiagnostic results. Also echograpy is a fast, cheap and non-invasive tool; however, the role of FNAC and echography in the diagnosis of salivary gland pathology is still not universally recognised (3-8). In the present study, FNAC had a sensibility, a specificity and a diagnostic accuracy of 83%, 93% and 92%, respectively. Positive and negative predictive values were 91% and 99%, respectively. The value of false negative rates of our study is satisfactory (8.6%) if compared to the value of other studies available in the literature (9-19). The value of the false positive rates is 0.9%. In the literature (from 1993 to 2012), the sensibility of the FNAC Table V. Overall FNA C accuracy.

Absolute sensibility S5+S4 Complete sensibility S5+S4+S3 VPPS5 VPPS4 VPP S3 VPP S3/S4/S5 VPN Complete Specificity (SO+S2) Specificity S2 Specificity SO FN FP(S3+S4+S5) FP (S4+S5) nadequate rate nadeq.zmalignitv rate Doubts rate S3 NB/NM ACCURACY

74.3 82.9 100.0 100.0 50.0 90.6 99.0 92.9 88.8 64.9 8.6 0.9 0 6.4 8.6 1.7 9.2 91.9

shows a variability that ranges from 52% to 97.6%, a specificity range of 53%-100%, while the accuracy for the identification of benign and/or malignant lesions ranges from 56% to 98%; these values are well correlated with the results of our study. Thus, the results from the present study show that FNAC is a helpful high-quality tool for the diagnosis of the major salivary gland lesions (20-22). This technique allows a high degree of histological identification rate (85.2% of histotype concordance) for both benign and malignant lesions. In the paper by Kechagias et al. (23), the diagnostic accuracy of specific types of tumours was about 80% and was higher for benign tumours. The present study also shows that histotype concordance is higher for benign neoplasm than for malign (83%). The accurate diagnosis of malign histotypes is still a challenge, and it is probably due to the extremely various morphology of presentation of the major salivary gland pathology, representing a highly complex cytological interpretation and great difficulty in gaining the necessary adequate experience (24-27). Some carcinomas of the salivary glands are characterized by malign cells of different nature and generally their proportions in their representations and their architectural organization define the specific malignant histotype. Also when considering benign tumours, the cellular variability of the lesions may generate some difficulties for the FNAC diagnosis. This problem is also present within pleomorphic adenomas diagnosis, but, despite these difficulties, in our study the diagnostic concordance for this histotype resulted very high 96.6% (141 of 146 cases), while in the literature it ranges between 78% and 94%. Also for Warthin tumour histotype the diagnostic sensibility has a wide reported range of variability, from 33 to 100%, while in our experience the diagnostic accuracy was very good (88%) (25).

Table VI. Accuracy ofecography in relation to the final histology.

fECO morphology+margins+echostructure) BENIGN SUSPECT MALIGN SUSPECT frotal

r

BENIGN NEOFORM ATION

55 1 56

HISTOLOGY MALIGN NEOFORMAnON 3 4 7

TOT

58 5 63

970

A. PASTORE ET AL.

It is also necessary to emphasize that our data represent the result of a close cooperation between ENT specialists and skilled pathologists; this type of collaboration is a fundamental requisite in order to optimize FNAC diagnostic accuracy (26, 27). There are many factors that can influence FNAC adequacy, including experience in the procedures, the positioning of the tiny needle within the target tissue, artefacts when preparing the histological samples. In non-diagnostic cases, the FNAC repetition reduced the inadequate cytological case rate (S I) from 36 to 23, thus increasing FNAC sensibility from 71.4% to 82.9%, FNAC specificity from 89.8 to 92.9%, and FNAC diagnostic accuracy from 88% to 92%. Also echography resulted a good diagnostic tool in order to provide a better characterization of salivary gland lesions, with a sensibility of 57.1%, and a specificity that resulted even higher than FNAC's. Nonetheless, due to its lower sensibility when compared to FNAC echography cannot be considered a method as reliable as FNAC. Moreover, the total cost of echography and FNAC is 123.50 euros, against 249.45 euros of a contrast MRI; therefore, the application of echography and FNAC as first level examinations allows a costs saving. In conclusion, the present study indicates that FNAC is a safe diagnostic tool that has a reliable sensitivity and specificity for the assessment of salivary gland pathology. While echography can be useful in order to provide a better characterization of the lesions, FNAC can then be considered a useful tool for selecting patients and for indicating surgical treatment. Moreover it could be useful to select patients that need further investigations such as CT and/or MRI scans, in order to reach a better compromise between costs and benefits for the preoperative stadiation of major salivary glands pathology.

4.

5.

6.

7.

8.

9.

10.

II.

12.

13.

REFERENCES I.

2. 3.

masses cervicales. Schweiz Med Wochenschr 2000; 116:47-49. Kuvezdic KG, Aurer I, Ries S, Sucic M, Glamocak MM, Hic I, Basic-Kinda S, Radman I, Labar B. FNA based diagnosis of head and neck nodal lymphoma. Coli Antropol2010; 34:7-12. Mobley DL, Wakely PE Jr, Frable MA. Fine-needle aspiration biopsy: application to pediatric head and neck masses. Laryngoscope 1991; 101:469-72. Layfield LI, Tan P, Glasgow Bl. Fine needle aspiration of salivary gland lesions. Arch Pathol Lab Med 1987; III :346-53. Schwarz R, Chan NH, MacFarlane lK. Fine needle aspiration cytology in the evaluation of head and neck masses. Am 1 Surg 1990; 159:482-5. Raymond MR, Yoo lH, Heathcote io, McLachlin CM, Lampe HB. Accuracy of fine-needle aspiration biopsy for Warthiri's tumours. Otolaryngol 1 2002; 31:263-70. Sengupta S, Roy A, Mallick MG, Kundu B, Das SK, Das S. FNAC of Salivary Glands. Indian 1 Otolaryngol Head Neck Surg 2002; 54:3-10. Contucci AM, Corina L, Sergi B, Fadda G, Paludetti G. Correlation between fine needle aspiration biopsy and histologic findings in parotid masses. Personal experience. Acta Otorhinolaryngol Hal2003; 23:31418. Cohen EG, Patel SG, Lin 0, Boyle 10, Kraus DH. Fine-needle aspiration biopsy of salivary gland lesion in a selected patient population. Arch Otolaryngol Head Neck Surg 2004; 130:773-78. Elagoz S, Gulluoglu M, Yilmazbayhan D, Ozer H, Arslan I. The value of fine-needle aspiration cytology in salivary gland lesions, 1994-2004. ORL 2007; 69:51-56. Mihashi H, Kawahara A, Kage M, Kojiro M, Nakashima T. Comparison of preoperative fine - needle aspiration cytology diagnosis and histopathological diagnosis of salivary gland tumors. Kurum Med 12006; 53:23-27. Zbaren P, Guelat D, Loosli H, Stauffer E. Parotid tumors: fine-needle aspiration and/or frozen section. Neck Surg Capo Otolaryngol2008; 139:811-15.

Holland IF, Bast RC, Morton DL, Frei E, Kufe DW, Weichselbaum RR. Neoplasms of the head and neck region. Cancer Medicine. Fourth edition, Vol I. Baltimore. MD: BC Decker 1997; 1674-8. WHO Classification of Head and Neck Tumors

14.

IARC Press Lione 2005. Waridel F, Mihaescu A, Lang FJ. Apport de la cytoponction a laguille fine dans la diagnostic des

15. Tandon S, Shahab R, Benton 11, Ghosh SK, Sheard 1. Fine-needle aspiration cytology in a regional head and neck cancer center: comparison with a

Int. J. Immunopathol. Pharmacol.

16.

17.

18.

19.

20.

systematic review and meta-analysis. Head Neck 2008; 30: 1246-52. Tahoun N, Ezzat N. Diagnostic accuracy and pitfalIs of preoperative fine needle aspiration cytology in salivary gland lesions. J Egyptian Nat Cancer Inst 2008; 20:358-68. Daneshbod Y, Daneshbod K, Khademi B. Diagnostic difficulties in the interpretation of FNAC in salivary lesions: diagnostic pitfalIs revisited. Acta cytological 2009; 53 :53-70. Gobic MB, Pedisic D, Bekafigo IS, Cerovic R, Starcevic R. Fine needle aspiration cytology in the evaluation of parotid gland tumors. ColI Antropol 2010; 34:345-48. Paris J, Facon F, Pascal T, Chrestian MA. Valori diagnostici preoperatori di citologia con ago sottile w di MRI nei tumori delIa ghiandola parotide. Eur Arch Otorhinolaryngol 2005; 262:27-31. Verma K, Kapila K. Role of fine needle aspiration cytology in diagnosis of pleomorphic adenomas. Cytopathology 2002; 13:121-27.

21. Lukas J, Duskova J. Fine needle aspiration biopsy

22.

23.

24.

25.

26.

971

in the diagnostic of the tumors and non- neoplastic lesions of salivary glands. Bratisl Lek Listy 2006; 107:12-15. Kechagias N, Ntomouchtsis A, Valeri R, et al. Fineneedle aspiration cytology of salivary gland tumours: a 10-year retrospective analysis. Oral Maxillofacial Surg 2012; 16:35-40. Riley N, AlIison R, Stevenson S. Fine-needle aspiration cytology in parotid masses: our experience in Canterbury, New Zealand. ANZ J Surg 2005; 75:144-46. Singh Nanda KD, Mehta A, Nanda 1. Fine-needle aspiration cytology: a reliable tool in the diagnosis of salivary gland lesions. J Oral Pathol Med 2012; 41: 106-12. Raymond MR, Yoo JH, Heathcote JG, McLachlin CM, Lampe HB. Accuracy of fine-needle aspiration biopsy for Warthin's tumours. Otolaryngol J 2002; 31:263-70. Jandu M, Webster K. The role of operator experience in fine needle aspiration cytology of head and neck masses. Int J Oral MaxilIofac Surg 1999; 28:441-44.