Fine-needle aspiration cytology of extranodal lymphoma

Limited Tissue Biopsies and Hematolymphoid Neoplasms

OBJECTIVES

Use of fine-needle aspiration/needle core biopsy (FNA/CNB) in evaluating hematolymphoid processes has been debated. We investigate its applicability in various clinicopathologic settings.

METHODS

We retrospectively analyzed 152 cases of FNA/CNB.

RESULTS

Of 152 FNA/CNBs, 124 (81.6%) resulted in diagnoses without excisional biopsies, while 28 required subsequent excisional biopsies. Of these, 43 FNA/CNBs performed for suspected lymphoma relapse demonstrated 95.4% diagnostic rate (41/43), which was significantly better than those without history of lymphoma (77/109, 71%; odds ratio [OR], 8.5; confidence interval, 1.9-37.4). Patients with immunodeficiency also showed a high rate of diagnosis by FNA/CNB (100%). When stratified by types of disease, diffuse large B-cell lymphoma/high-grade B-cell lymphoma demonstrated a higher success rate (92.7%) than small B-cell lymphoma (79.2%), though the difference was not statistically significant (OR, 3.3; P value = .07). A subsequent excisional biopsy was required in 28 cases, 23 of which resulted in diagnoses concordant with the FNA/CNB. Five cases showed diagnostic discordance, reflecting pitfalls of FNA/CNB in unusual cases with complex pathology.

CONCLUSIONS

FNA/CNB is practical in evaluating most hematolymphoid lesions, with high efficacy in recurrent disease and some primary neoplasms with homogeneous/ aggressive histology, or characteristic immunophenotype.

Skeletal Muscle Involvement in Six Cases of Non-Hodgkin's Lymphoma of the Head and Neck Region

Extra-nodal involvement of non-Hodgkin's lymphoma (NHL) is well described. Skeletal muscle involvement is rare with the muscles of the extremities being most commonly affected. We present a series of six patients that have attended our Joint Head and Neck Ultrasound and Cytopathology clinic. In these cases, skeletal muscle involvement by NHL was demonstrated in the head and neck region. The ultrasound findings in these cases are presented and discussed.

Thyroid Lymphoma: A Single Institution's Experience

BACKGROUND: Primary lymphoma of the thyroid gland is uncommon.

METHODS: The tumor registry at the Penn State Milton S. Hershey Medical Center was searched for all patients carrying a diagnosis of thyroid lymphoma. All available records for these patients were examined and information regarding patient and tumor characteristics, clinical presentation, interventions, and survival were tabulated.

RESULTS: The average age of patient in our series was 67.5 years. There was a 2.7:1 female to male preponderance. Thirteen (60%) patients presented with thyroid nodule(s). Eleven (50%) presented with aerodigestive tract obstructive symptoms. Sixteen (73%) patients had no pre-existing history of thyroid disease; none had known pre-existing Hashimoto's thyroiditis.

CONCLUSIONS: The typical thyroid lymphoma patient is female and elderly with painless thyroid enlargement. Compressive symptoms of the aerodigestive tract are common at presentation and may require urgent intervention. Treatment modalities (XRT, CHOP chemotherapy) and outcomes are distinct from other thyroid malignancies.

EBM RATING: C

A primary nipple lymphoma diagnosed by a modified fine-needle aspiration method

Primary breast lymphomas are uncommon. All reported primary breast lymphomas were in the breast parenchyma. Here we reported the first case of primary nipple lymphoma in a 76-year-old woman initially diagnosed using a modified fine-needle aspiration method. The aspirated material by this method had yielded adequate material for both cytomorphologic and flowcytometric analysis, as well as for molecular analysis of light chain rearrangement. In smears the atypical lymphocytes were predominantly middle-sized with irregular nuclei. Scattered large centroblast or immunoblast-like cells, a few reactive lymphocytes,histocytes, and few plasma cells were also observed. These findings suggested a low-grade lymphoma that was further confirmed by flow cytometry (CD19+/CD3-, positive for cytoplasmic kappa light chain but negative for lambda light chain) and molecular analysis (monoclonal rearrangement of immunoglobulin kappa chain). Immunohistochemical stains performed on the excised specimen showed that the tumor cells were positive forCD20 and CD79a but negative for cytokeratin, CD3, CD5,CD10, CD23, CD43, CD45RO, bcl-6, and cyclin-D1. The Ki-67 proliferation index was less than 20%. Taking these together, the case was diagnosed as a primary MALT lymphoma of the nipple.FNA usually provides a better cell morphology than tissue sections, but pathologists have to face the sampling error and lack of immunophenotype information when subtyping lymphoma issues using FNA. With the help of flow cytometry and molecular analysis, more and more trials have proved the accuracy of FNA in diagnosis of lymphomas. Therefore, FNA could play an informative and diagnostic role in diagnosis of lymphoma

Ultrasound-guided fine needle aspiration cytology in the diagnosis and management of thyroid nodules

OBJECTIVE

To examine the accuracy of fine needle aspiration cytology (FNAC) in the diagnosis of thyroid nodules and compare the inadequacy rates for ultrasound-guided and freehand FNAC.

METHODS

A retrospective study of 434 patients with thyroid nodules who underwent diagnostic FNAC over a 2-year period. Cytological diagnoses have been compared with the histological assessment of resection specimens in 69 cases.

RESULTS

The inadequacy rate was significantly lower from ultrasound guided FNAC (24/373 cases, 6.4%) than from freehand FNAC (8/61 cases, 13.1%) (P = 0.043). Seventy-six percentage of patients had a non-neoplastic cytological diagnosis and, after multidisciplinary review, the patients were reassured and assigned to clinical follow-up. Sixty-seven patients had a resection for cytological appearances consistent with non-neoplastic disease (n = 34), suspicious of follicular neoplasia (n = 23), or suspicious of malignancy (n = 10), and two patients had resections following inadequate cytology with ultrasound appearances suspicious of a neoplasm. The overall accuracy of FNAC analysis for malignancy was 97.0%, with sensitivity 83.3%, specificity 98.0%, positive predictive value 71.4% and negative predictive value 98.4%. The overall accuracy of FNAC analysis for the prediction of neoplasia was 97.5%, with sensitivity 80.5%, specificity 97.8%, positive predictive value 89.2% and negative predictive value 95.9%. Difficulties in cytological diagnosis were associated with lymphoid infiltrates and with degenerative changes in follicular adenomas.

CONCLUSION

Ultrasound-guided FNAC has a significantly lower yield of inadequate aspirates than palpable FNAC. The ability of FNAC to predict neoplasia in 89% patients and to exclude neoplasia in 95.9% patients makes an important contribution to the multidisciplinary assessment of patients.

Non-Hodgkin's lymphoma of the breast: a review of 18 primary and secondary cases

We evaluated the clinical features, morphology, and incidence of 18 cases of breast lymphoma over 10 years at the University of Alabama at Birmingham. Fine needle aspiration was performed in 7 of 18 patients, and tissue biopsy/resection was available for all 18 cases. Patients were 33 to 91 years old (median, 61); 17 were women and 1 was a man. Fine needle aspiration was consistent with the tissue diagnosis in 6 of the 7 cases (86%). One case was diagnosed by fine needle aspiration as atypical cells, favor benign; the biopsy revealed diffuse large B-cell lymphoma. Tissue diagnoses revealed that 11 cases (61%) represented diffuse large B-cell lymphoma and 3 (17%) were follicular lymphomas. The remaining 4 cases (22%) were plasma cell neoplasm, T-cell neoplasm, Burkitt's lymphoma, and precursor B-cell lymphoblastic lymphoma. Flow cytometry and/or gene rearrangement supported the diagnosis of lymphoma in 8 cases. Although rare, lymphoma should be considered in the differential diagnosis of a breast mass. Fine needle aspiration can facilitate appropriate triage of patients for further management.

Bilateral primary lymphoma of the breast: A case report initially diagnosed by FNAC

Primary lymphoma of the breast (PLB) is a rare disease, representing 0.04-0.5% of all malignant breast neoplasms. We present a patient with bilateral breast involvement by a high-grade diffuse large B-cell lymphoma, which was diagnosed initially by fine-needle aspiration cytology (FNAC). Mammography revealed a diffuse increase in density of the right breast and a large solitary mass on the left breast, suggestive of an inflammatory carcinoma. The patient underwent FNAC and the diagnosis of a non-Hodgkin's lymphoma (NHL) was suggested. Physical examination revealed palpable bilateral axillary lymph nodes but no evidence of concurrent widespread disease. The patient underwent complete staging evaluation. The only positive findings were an elevated lactate dehydrogenase (LDH) and evidence of axillary lymphadenopathy on CT. Excisional biopsy was performed on the left breast. The morphological and immunohistochemical analysis confirmed the diagnosis of a high-grade diffuse large B-cell lymphoma with an immunophenotype suggestive of a germinal center cell origin.

Pitfalls in Salivary Gland Fine-Needle Aspiration Cytology: Lessons From the College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology

Context.—We use data from the College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology to identify common diagnostic errors in salivary gland fine-needle aspiration (FNA).

Objective.—To identify salivary gland FNA cases with poor performance characteristics in the Nongynecologic Cytology Program surveys, so that the most common diagnostic pitfalls can be avoided.

Design.—A retrospective review of the College of American Pathologists Nongynecologic Cytology Program's cumulative data from 1999 to 2003 revealed the most common false-positive and false-negative interpretations on FNA for common salivary gland lesions. Slides that performed poorly were then reviewed to identify the cytologic characteristics that may have contributed to their poor performance.

Results.—A total of 6249 participant responses with general interpretations of benign (n = 4642) or malignant (n= 1607) were reviewed. The sensitivity and specificity of the participant responses for correctly interpreting the cases as benign or malignant were 73% and 91%, respectively. Benign cases with the highest false-positive rates were monomorphic adenoma (53% false-positive), intraparotid lymph node (36%), oncocytoma (18%), and granulomatous sialadenitis (10%). Malignant cases with the highest false-negative rates were lymphoma (57%), acinic cell carcinoma (49%), low-grade mucoepidermoid carcinoma (43%), and adenoid cystic carcinoma (33%). Selected review of the most discordant individual cases revealed possible explanations for some of the interpretative errors.

Conclusions.—These data confirm the difficulty associated with interpretation of salivary gland FNA specimens. Cytologists should be aware of the potential false-positive and false-negative interpretations that can occur in FNAs from this organ site in order to minimize the possibility of diagnostic errors.

Determination of light chain restriction in fine-needle aspiration-type preparations of B-cell lymphomas by mRNA in situ hybridization

The purpose of this study was to investigate whether B-cell clonality could be demonstrated in fine-needle aspiration (FNA)-type preparations by using automated and manual in situ hybridization (ISH) systems. FNA-like preparations were made from 10 cases of B-cell lymphoma and 5 cases of reactive lymphoid hyperplasia. Kappa/lambda expression was determined using an automated mRNA ISH assay or a manual ISH system. Other variables tested included type and length of fixation, protease digestion, and time in chromogen. Clonality data were corroborated by either flow cytometry or tissue-based analysis. Optimal conditions required formalin fixation, strong protease digestion, and prolonged hybridization and chromogen times; under these conditions, monoclonality was demonstrated by in situ in 8 of 10 cases. Each of the five cases of reactive lymphoid hyperplasia showed polyclonal light chain expression by automated mRNA ISH. In situ kappa/lambda mRNA analysis of FNA-type specimens allows direct determination of monoclonality in cytologic preparations.

Molecular diagnostic cytopathology: definitions, scope and clinical utility

Molecular diagnosis is the application of molecular biology techniques and knowledge of the molecular mechanisms of disease to diagnosis, prognostication and treatment of diseases. Although it is not widely used in routine molecular cytological practice, some examples are presented here of the application of molecular techniques to the routine cytopathological diagnosis of solid tumours and lymphoreticular malignancies. The term 'molecular diagnostic cytopathology' is proposed to define the application of molecular diagnosis to cytopathology, and the challenges of the introduction of molecular diagnosis into routine diagnostic histopathology and cytopathology are discussed. Finally, the importance of a combined morphological, immunophenotypic and molecular approach to maintain the diagnostic pathologist at the heart of the clinical decision-making process is emphasized.

Utility of Fine-Needle Aspiration As a Diagnostic Technique in Lymphoma

Purpose

To evaluate, from a clinician's perspective, the sensitivity and specificity of fine-needle aspiration (FNA) as a technique for the diagnosis of lymphoma.

Patients and Methods

Medical records of 470 new patients seen in one lymphoma specialist's clinic from January 1998 through December 2002 were reviewed. Ninety-nine (21%) of the 470 patients underwent a total of 115 FNA procedures, which were assessed by more than 70 different pathologists in 32 different pathology departments. Subsequent excisional biopsies were performed in 67 of these patients and interpreted by a single hematopathology group without independent review.

Results

Of 115 FNA procedures, 93 were completed for the initial evaluation of lymphoma and 22 were done for assessment of relapsed disease. Of the 93 FNA attempts at initial diagnosis, only 27 (29%) were given a specific and complete histologic diagnosis using an accepted classification system (Working Formulation, Revised European-American Classification of Lymphoid Neoplasms, WHO). For the 22 FNAs done for recurrent disease, only nine (41%) were classified using an accepted system. Sixty-seven (72%) of the 93 FNAs performed for the evaluation of initial disease had subsequent excisional biopsies. Among these paired comparisons, only eight (12%) of 67 FNA diagnoses were correlated with the subsequent excisional biopsy diagnosis. Immunophenotyping was completed on 24 of the 67 paired FNAs. Seven of the 24 FNAs with immunophenotyping (29%) were correlated with subsequent histology on excisional biopsy. Only one (2%) of 43 FNA diagnoses, based on morphology alone, was correlated with subsequent excisional biopsy diagnosis.

Conclusion

Overall, FNA for lymphoma diagnosis is not helpful, not cost effective, and in addition may misguide treatment.

Role of fine-needle aspiration cytology in breast lymphoma

Lymphomas of the breast are rare and may mimic carcinoma clinically. We investigated the ability of fine-needle aspiration (FNA) biopsy combined with adjunctive flow cytometry (FC), immunofluorescence microscopy (IFM), and immunocytochemistry (ICH) to diagnose and eventually subclassify lymphomas of the breast according to the Revised European American Lymphoma/World Health Organization classification. We retrieved 21 breast aspirates from 19 patients with a cytologic diagnosis of lymphoma or plasmacytoma over a 10-year period (1992-2002), excluding 98 benign intramammary lymph nodes and 1 atypical lymphohistiocytic proliferation (Rosai Dorfman disease). FC was performed in 15/21 aspirates, IFM in 1/21, ICH in 3/21. Histologic follow-up (HF) was obtained for 10 patients, most of them with primary lymphoma. For the remaining nine patients without HF, flow cytometric analysis, comparative morphology, or remission after chemotherapy regimens supported the cytologic diagnosis. Of 19 patients, 11 patients had a secondary lymphoma (SL) and 8 patients had a primary lymphoma (PL). FNA and FC/IFM/ICH classified 7/8 PLs as B-cell lymphomas and 1/8 PLs as plasmacytoma. However, FNA could only subclassify 3 of 8 PLs. FNA and/or FC subclassified accurately 10/11 SLs. All cases were accurately immunophenotyped as B-, T-cell non-Hodgkin's lymphomas or plasmacytoma. World Health Organization classification was achieved in 3/8 PLs (42%) and 10/11 SLs (91%; P = 0.04). Subclassification (which has an impact on long-term management and prognosis) was significantly better in SL, when a previous histologic diagnosis had already been made, when compared to PL, of which 5/8 cases (62.5%) could not be accurately classified.

Immunocytochemistry versus molecular fingerprinting of metastases

Examination of cytological samples of cancer to suggest a possible primary site of origin is one of the commonest and most difficult tasks of diagnostic cytopathologists. Currently, both cytomorphology and immunocytochemistry are the main approaches to this diagnostic dilemma. We report the application of microsatellite analysis in cytological samples in a patient with a primary colonic tumour and two subsequent lung nodules, which were suspected on CT scans of the chest, and compared the findings with those obtained with conventional immunocytochemistry. The molecular results were in agreement with the radiological impression and conflicted with the immunocytochemistry. We conclude that immunocytochemical and molecular biology approaches to the diagnosis of tumours may give rise to contradictory results.

Cytologic findings of marginal zone lymphoma

BACKGROUND

Low-grade lymphomas can be difficult to diagnose in cytology specimens, especially marginal zone B-cell lymphomas (MZLs). MZL is a low-grade lymphoma characterized by a heterogeneous lymphoid population, which can be difficult to distinguish from reactive processes in cytology specimens.

METHODS

Fourteen cytology specimens of MZL from 11 patients with histologically confirmed MZL (n = 10) or flow cytometry (FC)-confirmed MZL (n = 1) were reviewed, including 13 fine-needle aspiration biopsy (FNAB) specimens (salivary gland, n = 6; lung, n = 3; lymph node, n = 2; breast, n = 1; and soft tissue, n = 1) and 1 pleural effusion specimen. Cytologic preparations included air-dried and alcohol-fixed direct smears, ThinPrep(R) slides, and cell blocks. FC studies were available in six specimens.

RESULTS

All 13 FNAB specimens were composed predominantly of intermediate-sized lymphoid cells with interspersed small, round lymphocytes and transformed cells. The intermediate-sized cells displayed a moderate amount of cytoplasm, slight nuclear membrane irregularities, and inconspicuous-to-absent nucleoli. The pleural fluid contained mostly small-to-intermediate-sized, round lymphocytes. In 10 specimens, the intermediate-sized cells often showed plasmacytoid morphology, which were seen best in Diff-Quik (Mercedes Scientific Co., Inc., Sarasota, FL)-stained slides. Monocytoid cells (n = 6 specimens), plasma cells (n = 7), lymphohistiocytic aggregates (n = 11), and tingible body macrophages (n = 5) were variably identified. Lymphoepithelial lesions were not observed. Three specimens with FC studies showed a phenotype compatible with MZL, and three specimens were nondiagnostic.

CONCLUSIONS

Cytologic features suggestive of MZL included abundant intermediate-sized lymphoid cells with mild atypia in a background of small lymphocytes and transformed cells, often with plasmacytoid morphology. Flow cytometry was beneficial only in selected MZL cytology specimens. Surgical correlation may be necessary to confirm the diagnosis.

High frequency of incidental diagnosis of extrathyroidal neoplastic diseases at the fine-needle aspiration biopsy of laterocervical lymph nodes in patients with thyroid nodules

This study was undertaken to evaluate the frequency of the incidental diagnosis of extrathyroidal lymph node diseases at ultrasound-guided fine-needle aspiration biopsy/cytology (FNAB/C) being done to check the presence of metastatic thyroid cancer in 30 subjects with thyroid nodule (TN) and enlarged cervical lymph nodes (CLN). The patients in whom cytology suggested the presence of malignancy in the TN or in the CLN underwent surgical removal for histologic diagnosis. The spectrum of diseases revealed by this survey included: (1) 10 benign diseases including 1 case of Piringer-Kuchinka lymphadenitis with benign TN; (2) 10 metastatic thyroid cancers (2 anaplastic and 8 papillary cancers); (3) 3 benign TN associated with metastatic invasion of cervical lymph nodes from lung (2 cases) and breast (1 case) cancer; (4) 1 Hodgkin's lymphoma of the cervical lymph nodes with hyperplastic TN; (5) 3 nodal lymphomas with benign thyroid nodule and 2 cases of thyroid lymphoma with nodal invasion; and (6) 1 nodal sarcoidosis with benign TN. The results of this study demonstrate that important neoplastic and hematologic diseases affecting the cervical lymph nodes may frequently be incidentally detected using ultrasonography (US) and FNAB/C in the diagnostic procedure for thyroid nodule.

The diagnostic value of fine needle aspiration cytology in Waldeyer's ring lymphomas

Fine needle aspiration cytology (FNAC) is a well-established diagnostic technique for tumours in the head and neck area. In recent years FNAC has been established as an accurate and useful method for the diagnosis of nodal malignant lymphoproliferative disease. The purpose of the present study was to determine and evaluate the accuracy of FNAC in the diagnosis of primary malignant lymphoma of Waldeyer's ring. The cases of 29 patients suffering from tumours of the oro- and nasopharynx, in which the diagnosis of lymphoma was established by FNAC during the years 1991-1998, were collected and analysed. Twelve of the patients had a previous history of lymphoma, and FNAC was used to establish the diagnosis of recurrent disease. In 17 patients with no previous history of malignancy FNAC was used to diagnose primary extranodal non-Hodgkin's lymphomas (NHLs). In two patients FNAC failed to diagnose NHL. In all patients cytological findings were complemented and compared with those of a histopathological examination after open biopsy. In two cases a difference in the specific histological type of the lymphoma was noted. The findings from the present study (sensitivity 93.10% and positive predictive value 100%) indicated that FNAC is a useful and accurate method in establishing diagnosis of Waldeyer's ring lymphomas.

Renal lymphoma. The diagnostic and therapeutic roles of fine-needle aspiration

This study focused on 19 patients with renal lymphoma (RL) from whom 20 initial (1 patient with fine-needle aspiration [FNA] specimens of masses in both kidneys) and 1 repeated FNA specimen were obtained. Of the 19 patients, 10 had secondary RL, 8 primary RL, and 1 transplant RL. The FNA samples were studied by smears (all cases), tissues (11), phenotyping by immunostaining (13) or flow cytometry (4), and gene rearrangement (3). The final diagnoses included 1 T-cell lymphoma and 18 B-cell lymphomas. Of the 20 original specimens, 14 were reported as positive for lymphoma, 3 suggestive of lymphoma, 1 positive for transitional cell carcinoma, and 2 unsatisfactory. The follow-up specimen showed reactive changes. Tissue correlation, available in 11 cases, confirmed a positive cytodiagnosis (7), provided a final diagnosis in the cytologically inconclusive cases (3), or revised the misdiagnosis of transitional cell carcinoma from smears (1). The phenotyping elucidated the B vs T lineage of the lymphoma in all tested cases, confirmed the positive cytodiagnosis in 10 cases, confirmed the reactive cytodiagnosis in 1 case, and helped achieve a conclusive diagnosis in 2 cases suggestive of lymphoma. Gene rearrangement studies showed light chain restriction in the 2 tested cases. FNA has an essential role in treatment planning for RL. Although FNA usually is diagnostically conclusive, a high index of suspicion and awareness of atypical or misleading cytomorphologic features are important for a correct interpretation, especially for primary RL. Ancillary testing is essential for the diagnosis in problematic cases and lays the foundation for the differential diagnosis.

Fine-needle aspiration of extranodal and extramedullary hematopoietic malignancies

There is relatively little information concerning the use of fine-needle aspiration (FNA) to diagnose extranodal and extramedullary hematopoietic malignancies. Seventy-one such cases diagnosed by FNA form the basis of this study. Seventy-one cases of FNAs performed between 1988 and 1998 on extranodal and extramedullary hematopoietic malignancies were reviewed in order to evaluate the usefulness of this technique in diagnosing these entities as well as to assess patterns of relapse. There were 45 male and 26 female patients ranging in age from 29-86 years (mean, 68 years). Sixty-six patients had a previous history of a hematopoietic malignancy. Aspirates from 65 of these patients were consistent with the patient's known primary. One aspirate of a paravertebral mass from a multiple myeloma patient showed extramedullary hematopoiesis. The remaining five aspirates were cases of multiple myeloma that first presented as soft tissue masses. The most common malignancies were lymphoma: 52 cases (73%), 48 large cell lymphomas, four mixed small and large cell lymphoma; followed by multiple myeloma: 12 cases (17%); leukemia: four cases (5.4%); Hodgkin disease: two cases (2.8%); and one case of extramedullary hematopoiesis. The aspirate sites were soft tissue: 23 cases (32%); bone: 17 cases (24%); kidney: 14 cases (20%); liver: 11 cases (15%); lung: three cases (4%); adrenal: two cases (3%); and eye: one case. The interval between primary diagnosis and FNA was 1-36 months (mean, 13 months). In conclusion, 98% of the aspirates were neoplastic in patients with a known history of hematopoietic malignancies. The most common site of involvement was soft tissue in 23 (32%) cases. In five patients with multiple myeloma, the FNA diagnosis prompted a work-up to find the primary site of involvement. FNA is a useful technique in assessing extranodal and extramedullary hematopoietic malignancies.

Fine-needle aspiration cytology of salivary gland: a review of 341 cases

Three hundred and forty-one salivary gland fine-needle aspiration (FNA) cytology specimens taken over a 6-yr period were reviewed and correlated with clinical and/or histological findings. The aspirates were derived from parotid gland (212 cases), submandibular gland (124 cases), and minor salivary gland (5 cases). The major diagnostic categories were unsatisfactory (10 cases), normal (100 cases), sialadenitis (74 cases), cyst (34 cases), lipoma (5 cases), pleomorphic adenoma (55 cases), Warthin's tumor (36 cases), and malignancy (27 cases). The latter included 14 primary salivary neoplasms (4 lymphomas of mucosa-associated lymphoid tissue (MALT) type, 3 adenocarcinomas, 2 squamous carcinomas, 2 adenoid cystic cacinomas, and one case each of carcinoma ex pleomorphic adenoma, undifferentiated carcinoma, and high-grade mucoepidermoid carcinoma), and 13 metastases, 9 of which were derived from squamous carcinomas of head and neck origin. Clinicopathological review showed that 88 of 91 (97%) benign epithelial tumors and 27 of 31 (87%) malignant neoplasms with adequate FNA sampling were accurately diagnosed cytologically. False-negative results were caused by sampling error (7 cases), most notably in cystic tumors, or were due to misinterpretation of uncommon neoplasms (3 cases). The overall sensitivity, specificity, and accuracy were 92%, 100%, and 98%, respectively. FNA cytology provides accurate diagnosis of most salivary gland lesions and contributes to conservative management in many patients with nonneoplastic conditions.