CANCER 299 CYTOPATHOLOGY Ultrasound-Guided Fine-Needle Aspiration Biopsy of the Thyroid Rosemary Tambouret, M.D. Wanda M. Szyfelbein, M.D. Martha B. Pitman, M.D. BACKGROUND. We reviewed the Massachusetts General Hospital experience with Department of Pathology, Massachusetts General Hospital, Boston, Massachusetts. METHODS. All radiologically guided FNABs of the thyroid from January 1993 ultrasound-guided fine-needle aspiration biopsies (FNABs) of the thyroid to determine the indications, rate of unsatisfactory smears, correlation with excisional biopsy results, and verification of efficient use of personnel time. through June 1997 were reviewed. As a measure of efficient use of technologist time, a sample of times spent by the technologist during the procedure for 20 cases in 1993 and 1997 was compared with that of an equal number of random nonthyroid image guided FNABs. RESULTS. Two hundred-ninety FNABs were identified in 251 patients, representing 12% of all thyroid FNABs and 11% of all radiologically guided FNABs. Indications in the 251 patients included multiple nodules (78), solitary nodules (61), complex nodules (39), prior failed FNAB (39), thyroid bed abnormalities post-thyroidectomy (21), difficult access (7), and investigation of recurrent tumor in residual thyroid lobe (6). Available records indicated 118 lesions were palpable and 45 were nonpalpable; the physical examination characteristics of the remainder (88) were not stated. Diagnoses included 44 unsatisfactory cases (15%), 103 macrofollicular lesions, 20 microfollicular lesions, 26 mixed macro/ microfollicular lesions, 5 oxyphilic lesions, 1 trabecular pattern, 15 nonspecific follicular cell pattern, 9 follicular cell atypia, 30 cysts, 11 thyroiditis, 23 malignant tumors, and 3 other (1 parathyroid, 2 lymph node). Eighty-nine FNABs from 76 patients had subsequent surgical biopsy. Excisional biopsies in 14 unsatisfactory FNABs were benign. In the remaining 75 FNABs from 67 patients, 18 malignancies on FNAB were correctly diagnosed, but 3 other papillary carcinomas were only qualified as atypical follicular cells on cytology. No false-positive cases occurred. Of 15 macrofollicular lesions on cytology, 10 were adenomas on excision, only 2 of which were microfollicular adenomas, and 4 were adenomatous nodules. An aspirate of a parathyroid adenoma was misinterpreted as a macrofollicular lesion of the thyroid. Three microfollicular lesions on FNAB proved to be nodular hyperplasia on excision, and the other 11 were adenomas, 5 of them microfollicular. Average technologist time was significantly longer for thyroid FNABs than nonthyroid FNABs in 1993, but in the 1997 sample no significant difference was identified. CONCLUSIONS. Radiologically guided FNAB of the thyroid is a clinically useful Presented in abstract form at the 87th Annual Meeting of the United States and Canadian Academy of Pathology, Boston, MA, March, 1998. procedure with a high correlation between benign lesions not needing excision (macrofollicular), and lesions that need excision (microfollicular/oxyphilic cell or malignant). Technologist time needed for immediate evaluation tends to Address for correspondence: Rosemary Tambouret, M.D., Department of Pathology, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114. decrease with increasing operator experience. Cancer (Cancer Cytopathol) Received December 31, 1998; revision received April 21, 1999; accepted April 28, 1999. KEYWORDS: thyroid, parathyroid, cytology, ultrasound-guided fine-needle aspiration biopsy. © 1999 American Cancer Society 1999;87:299 –305. © 1999 American Cancer Society. 300 CANCER (CANCER CYTOPATHOLOGY) October 25, 1999 / Volume 87 / Number 5 hyroid nodularity is a frequent clinical problem.1,2 Although the estimated risk of developing a palpable thyroid nodule is 5 to 10%,3 autopsy and ultrasound studies of individuals with clinically normal thyroids have demonstrated the presence of nodules in 25 to 40% of cases.4 –10 However, only 5 to 15% of surgically excised nodules prove to be malignant. Fine-needle aspiration biopsy (FNAB) has become a standard initial step in the work-up of thyroid nodules, although protocols for the triage of thyroid lesions vary somewhat from one institution to another. The goal of the procedure is to identify for excision clinically significant lesions, either malignant or with the potential for malignancy, and observe all others. At our institution, most thyroid FNABs are performed by a clinician or cytopathologist on palpable masses (approximately 700 per year). In certain cases, however, such as a nonpalpable mass discovered by radiological techniques during surveillance of patients at high risk for thyroid cancer or during radiological study of nearby structures (e.g., the carotid artery), real-time imaging studies guide the procedure, which aids in obtaining adequate material.11,12 A radiologist performs ultrasound (US)-guided thyroid FNABs with immediate evaluation by a cytotechnologist and a cytopathologist to ascertain adequacy, a procedure that demands considerable personnel time. In addition, because US-guided FNAB of the thyroid is perceived by our cytology staff as frequently yielding suboptimal or nondiagnostic results while being excessively time consuming, we undertook a review of the Massachusetts General Hospital (MGH) experience to determine the indications, rate of unsatisfactory smears, and correlation with subsequent biopsy diagnosis. We also wished to evaluate whether the technologist’s time was being efficiently used in preparing and staining smears for rapid interpretation. T METHODS Pathology and clinical records were reviewed on all cases of radiologically guided FNABs of the thyroid from January 1993 through June 1997, for indication, clinical history, cytological diagnosis, and histological follow-up. An individual FNAB was defined as the ensemble of samples or passes taken from one lesion in a patient at one specific time. The data were compared with the number of all thyroid FNABs and with all radiologically guided FNABs performed during the same period. The FNABs were obtained by a radiologist, freehand, under continuous real-time sonographic guidance with 7- to 10-MHz high-frequency transducers. The patient was supine with neck extended. With the operator at the patient’s head, the neck was trans- versely scanned, the lesion was located, and local anesthetic was administered adjacent to the medial or lateral edge of the transducer. Needle size generally ranged from 22 to 25 gauge, although rarely an 18- or 20-gauge needle was used, and 1 mL of suction was applied. The exact number of passes varied from 1 to 12, but frequently the radiologist simply noted that multiple samples had been obtained from the lesion. Adequacy was determined during the procedure by a cytotechnologist who made smears and immediately fixed the smears in 95% alcohol. Smears for rapid interpretation were stained with hematoxylin and eosin. The cytotechnologist and the pathologist determined if the sampling was adequate or if additional sampling was advised. The remainder of the alcoholfixed smears were later Papanicolaou stained. Needle rinsings were evaluated and submitted for cytological preparation (Cytospin or ThinPrept) or for cell block if tissue fragments were visible. Cytological reports contained a statement of adequacy and a cytological diagnosis. Adequacy was designated as “unsatisfactory,” “satisfactory but limited by . . . [reason],” and “satisfactory” according to the judgment of the individual cytopathologist; uniform guidelines were not used. For the cases of “satisfactory but limited by . . . ,” the slides were reviewed, and the number of cell groups present were tabulated to compare with published guidelines for adequacy.13–15 The cases were divided into three groups: 1) 10 or more groups per case with 20 or more cells per group, 2) 6 or more groups per case with 10 to 20 cells per group, and 3) 0 to 9 groups of 10 or more cells per group. The main cytological diagnoses of the thyroid included follicular lesions, subdivided into macrofollicular, microfollicular, mixed macro/microfollicular, and nonspecific types. Macrofollicular lesions are those exhibiting more than 90% of their cells as groups or sheets of at least 6 to 12 uniform, round, nonoverlapping follicular cells in a background of colloid or cyst debris. A microfollicular lesion is composed predominantly of small groups of few follicular cells in a ring or rosette arrangement. A mixed lesion contains variable proportions of microfollicles and macrofollicles. This categorization is intended to give an estimate of the risk that a given lesion is follicular carcinoma. For example, a macrofollicular lesion—whether a colloid nodule, an adenomatous nodule of nodular goiter, or a macrofollicular adenoma— has almost no risk of malignancy, whereas a microfollicular lesion is likely to be neoplastic and carries a risk of approximately 20 to 25% of being a follicular carcinoma, somewhat higher or lower depending on the presence or absence of associated risk factors.3,16 An estimate of technologist’s time needed during US-Guided FNAB Thyroid/Tambouret et al. TABLE 1 Indications for Radiologically Guided Biopsy 301 TABLE 2 Unsatisfactory Cases Indication Number of patients Multiple nodules Solitary nodule Prior failed FNAB Complex nodules Thyroid bed abnormality post-thyroidectomy Difficult access Recurrent carcinoma in residual lobe of thyroid Total patients 78 (31.4%) 61 (24.1%) 39 (15.5%) 39 (15.5%) 21 (8.3%) 7 (2.8%) 6 (2.4%) 251 (100%) the procedure was obtained by averaging the time spent by the technologist at the FNAB from 20 consecutive US-guided thyroid FNABs in 1993 and 20 in 1997, compared with similarly selected nonthyroid FNABs. Student t-test was used to compare the two groups. RESULTS Patient Demographics The study population consisted of 290 FNABs in 251 patients (79 men and 172 women) aged 17 to 85 years (average 52.3 years), constituting 12% of all thyroid FNABs and 11% of all radiologically guided FNABs performed during this time period. Two hundred sixteen patients had one FNAB, 31 patients had two FNABs (25 synchronously of different lesions in the thyroid and 6 metachronously of the same lesion), and 4 patients had three FNABs, all at the same time of three separate thyroid lesions. The lesion size, recorded in 113 FNABs in 100 patients (43%), ranged from 0.3 to 4 cm (1.8 cm average); 17 were less than 1 cm in diameter. The lesion type was a solitary nodule in 122 patients, multiple nodules in 108 patients, and a thyroid bed lesion postthyroidectomy in 21 patients. The indications for use of US guidance (Table 1), other than the simple presence of a solitary nodule in 61 patients, included multiple nodules (78), complex nodules (39), prior failed FNAB performed without guidance (39), thyroid bed abnormalities post-thyroidectomy (21), difficult access (7), and abnormalities in a residual thyroid lobe worrisome for recurrent carcinoma (6). Examples of difficult access included substernal dominant nodules in a multinodular goiter and nodules near large vessels. Palpable masses or diffuse glandular enlargements were noted in 118 glands. Nonpalpable lesions were found in 45 glands, and the quality of the remainder (88) on physical examination could not be determined from the records examined. Thirteen lesions were found incidentally during Total Too few follicular cells Excision or biopsy (n 5 11) Mixed adenoma Microfollicular adenoma Follicular adenoma, NOS Macrofollicular lesion, NOS Hashimoto’s thyroiditis Thyroid bed fibrosis No follicular cells or colloid Excision or biopsy (n 5 2) Benign thyroid tissue, NOS Thyroid bed fibrosis Colloid or debris only Excision (n 5 1) Benign thyroid tissue N 5 44 (15%) 29 (65%) 6 1 1 1 1 1 9 (20%) 1 1 6 (14%) 1 NOS: not otherwise specified. scans of other sites, such as a carotid or parathyroid scan. Of the 13 incidental lesions, 8 were solitary with all but 1 nonpalpable. Five glands contained multiple nodules. Size was recorded in four and ranged from 1 to 3 cm. Ultrasound follow-up post-FNAB of the incidental lesions in two patients demonstrated a decrease in the size of the nodules. Five patients had no further recorded follow-up, and six had excisions revealing benign lesions except for one papillary carcinoma. Sixty patients had one or more significant associated clinical conditions, including prior radiation therapy (27), prior history of thyroid carcinoma (26), carcinoma at other sites (10), and a family history of medullary carcinoma (1). Cytological Diagnoses Initially, 49 cases were categorized as unsatisfactory; however, 5 had accompanying adequate cell blocks, 2 of which showed a microfollicular lesion, 1 a macrofollicular lesion, and 2 a nonspecific follicular pattern. Excluding the cases with adequate cell block, 44 cases were truly unsatisfactory (15%) (Table 2). Of these, five were FNABs of the thyroid bed, post-thyroidectomy (for medullary carcinoma in one case and papillary carcinoma in four cases) and were all deemed unsatisfactory because of absent or too few cells. Of the remaining 39 unsatisfactory cases, 11 samples were from 7 patients who had multiple sites sampled during the procedure. In five of the seven patients, one or more samples were adequate. Inadequate smears were obtained in 25% of patients having multiple sites aspirated, as opposed to only 14.8% of those undergoing 302 CANCER (CANCER CYTOPATHOLOGY) October 25, 1999 / Volume 87 / Number 5 TABLE 3 Satisfactory FNABs in 246 Patients* Histology Cytological findings Follicular adenomas Follow-up biopsy Total 246 FNABs N 75 (30.6%) NOS Micro Mixed Adenomatous nodule Macro Micro Mixed Nonsp foll patt Oxy Atyp foll cells Trab Coll nod/cyst Ch thy/Hashi Pap ca Med ca Anapl ca LN Parath Totals 103 20 26 15 5 9 1 30 11 20 2 1 2 1 (41.9%) (8.1%) (10.6%) (6.1%) (2.0%) (3.6%) (0.4%) (12.2%) (4.5%) (8.1%) (0.8%) (0.4%) (0.8%) (0.4%) 15 14 10 0 2 9 1 4 1 16 1 1 0 1 75 (14.7%) (70%) (38%) (0%) (40%) (100%) (100%) (13%) (9%) (80%) (50%) (100%) (0%) (100%) 5 4 2 5 3 2 2 4 3 5 1 1 1 1 3 1 1 2 Benign NOS Diffuse hyperplasia Pap ca Med ca Ana ca Parath 1 2 1 3 2 16 1 1 27 7 14 2 1 19 1 1 1 2 * Including 5 by cell block only and 36 “evaluation limited by . . .”. Macro: macrofollicular lesion; Micro: microfollicular lesion; Mixed: mixed follicular lesion; Nonsp foll patt: nonspecific follicular pattern; Oxy: oxyphilic lesion; Atyp foll cells: atypical follicular cells; Trab: trabecular lesion; Coll nod: colloid nodule; Ch thy/Hashi: chronic thyroiditis, Hashimoto’s disease; Pap ca: papillary carcinoma; Med ca: medullary carcinoma; Ana ca: anaplastic carcinoma; LN: lymph node; Parath: parathyroid; NOS: not otherwise specified. an FNAB of a single site; multiple site samples seemed to predispose to more unsatisfactory specimens. Out of the 246 satisfactory cases (241 on cytology and 5 on cell block alone), the evaluation of 36 was limited by various factors, including obscuring blood; obscuring chronic inflammation; lack of follicular architecture, with single follicular cells predominating; or few to no follicular cells. Size was noted in 11 of these 36 cases: 2 were 1 cm, 4 were 1 to 2 cm, and 5 were more than 2 cm. Five of the paucicellular cases were consistent with cysts. Slides were available in 29 of the 36 cases. On review of these 29 cases, 11 had at least 10 well formed follicular groups (up to 54), with more than 20 cells per group. Fifteen cases had 6 to 40 groups of follicular cells with each group containing 10 to 20 cells, as well as lesser numbers (0 to 9) of groups with more than 20 cells. Finally, three cases had 0 to 6 cell groups of more than 10 cells/groups, and their diagnoses included two cysts, one of which had no follicular cell groups and one with only three cell groups. The third consisted of a vast amount of lymphocytes but only two groups of follicular cells, consistent with chronic thyroiditis. Thus, depending on which of the published guidelines were used, either 11 of 29 cases (33%) or 26 of 29 (90%) were fully adequate. The cytological diagnoses of the 246 fully satisfactory cases (by cytology or cell block) are listed in Table 3. A macrofollicular lesion was the most common diagnosis, rendered in 103 FNABs (41.9%). Microfollicular or mixed follicular patterns were identified in 46 FNABs. Nonspecific or atypical follicular cells were found in 24 FNABs (10%), and 5 FNABs (2%) contained predominantly oxyphilic cells. Colloid cysts or nodules were recognized in 30 FNABs (12.2%). Twenty-three FNABs (9.3%) were diagnosed as malignant and included papillary carcinoma (20), medullary carcinoma (2), and anaplastic carcinoma (1). Two lymph nodes and one parathyroid gland were diagnosed as such on FNAB; however, another FNAB was recognized to be from the parathyroid gland only on retrospective review. Follow-Up Excisional Biopsies Seventy-six (30%) patients with 89 FNABs had subsequent surgical excision or biopsy. However, 14 of the 89 FNABs were unsatisfactory, leaving 75 FNABs for correlation. Follow-up excision in the 14 unsatisfactory FNABs were all benign (Table 2). In three of the five unsatisfactory thyroid bed aspirates, only fibrosis was identified on excisional biopsy, which correlated with the absent or scant cellularity seen on FNAB. Of the 246 satisfactory FNABs, 75 had tissue follow-up in 67 patients (Table 3). In 75% of the excisions neoplasms were found, 30% of which were malignant. Included in this group were lobectomies in two pa- US-Guided FNAB Thyroid/Tambouret et al. tients after FNABs in which only the cell blocks were adequate, showing a microfollicular pattern. Histology in both cases identified microfollicular adenomas. Also included in this group were six surgical biopsies in the group of 36 cases characterized as “evaluation limited by . . .”. All six biopsy samples were benign, including two follicular adenomas and one adenomatous nodule. The cytological diagnoses in these three cases had been macrofollicular lesion, mixed follicular lesion, and atypical follicular cells. No false-positive cases occurred. Three cases of papillary carcinoma were given only descriptive diagnoses noting the presence of atypical follicular cells with various abnormal features such as excessive nuclear grooves. One of these cases was a part of a pair of FNAs, the second of which was diagnostic of papillary carcinoma. Of follicular lesions, those qualified as macrofollicular and nonspecific follicular pattern tended to be excised less frequently (14.7% and 0%, respectively) than those identified as mixed follicular or microfollicular lesions (38% and 70%, respectively). Similarly, the diagnosis of an oxyphilic, trabecular, or atypical follicular lesion often was followed by excision (80%). Of those designated macrofollicular on cytology, 2 of 15 (13%) were microfollicular adenoma on excision as opposed to 5 of 14 (36%) microfollicular lesions. None of the mixed macro/microfollicular lesions proved to be microfollicular adenomas on excision. No follicular carcinoma was identified. The unrecognized parathyroid FNA was from a 1-cm, palpable cystic nodule in the right lower pole of the thyroid in a 44-year-old euthyroid female. A macrofollicular lesion was diagnosed cytologically. When removed, the nodule was found to be a parathyroid adenoma. The patient’s serum calcium level, which had been 10.2 mg/dL preoperatively, decreased to 8.3 mg/dL immediately postoperatively. Serum parathyroid hormone level had not been obtained because the adenoma was an unexpected finding. Review of the cytologic smears showed a low to moderately cellular sample composed predominantly of naked nuclei and few cells in groups with abundant delicate to oncocytic cytoplasm. The nuclei showed moderate anisonucleosis, chromatin stippling, and nuclear overlap. Microacini were identifiable, as were fragments of capillaries. Rare histiocytes were seen. The negative predictive value of a cytological diagnosis of a macrofollicular or mixed follicular lesion—that is, the ability to predict lesions that need not be excised (adenomatous nodules, macrofollicular and mixed follicular adenomas)—was 88%. The positive predictive value of a cytological diagnosis of microfollicular lesion, oxyphilic lesion, or le- 303 TABLE 4 Technologist Time in Performing FNAB Procedure 1993 average 1997 average Overall average P value Thyroid FNAB Nonthyroid FNAB 74 min 57 min 66 min 58 min 70 min 57 min 0.0319 0.1435 sion with atypical follicular cells or malignant cells— that is, the ability to predict lesions that need to be excised (microfollicular adenoma and thyroid carcinoma)—was 65%. Technologist Time On average, the technologist time for US-guided thyroid FNAs was 70 minutes (74 minutes in 1993 and 66 minutes in 1997), whereas for other radiologically guided FNABs, the average time was 57 minutes (57 minutes in 1993 and 58 minutes in 1997). The difference in time for thyroid FNABs as opposed to other radiologically guided FNABs was significantly different for the 1993 group (P 5 0.0319) but not for the 1997 group (P 5 0.1435), (Table 4). This may reflect a learning curve for the radiologist or simply a difference in personnel. DISCUSSION Although most FNABs of the thyroid can be accomplished without US localization, in certain situations image guidance may be necessary. Indications for US include indeterminate physical examination, nonpalpable lesions identified in patients at high risk for thyroid malignancies, and prior failed or nondiagnostic FNAB using direct palpation.17 In one series, about 50% of nondiagnostic FNABs without guidance were caused by cyst contents only, and subsequent repeat FNAB under US guidance helped correctly place the needle in the solid portion of the mass.18 Nondiagnostic FNABs are associated with thyroid malignancy in an estimated 2 to 9% of cases.19 –21 The percentage is greater in high-risk categories such as in men, in those exposed to radiation, and in those with larger nodules. Recent studies have questioned the utility of FNAB in small nonpalpable masses (less than 1.5 cm) because small masses, even if malignant, rarely cause clinical disease.22,23 Review of our material showed only 5.2% of US-guided FNABs were performed on lesions identified incidentally during scans of other sites. The majority were performed for reasons of difficult access, complexity, multiplicity of nodules, prior failed FNAB, or investigation of possible recurrent carcinoma. Our 15% unsatisfactory rate is at the limit of the 304 CANCER (CANCER CYTOPATHOLOGY) October 25, 1999 / Volume 87 / Number 5 acceptible range advised by the Papanicolaou Society Guidelines,13 which suggest less than 15% for the unsatisfactory rate and, although borderline, is heartening in view of the impression of poor performance held within the department. Cochand-Priollet et al.24 reported an unsatisfactory rate of only 3.8% in their group of US-guided thyroid FNABs, but all their cases were palpable masses. Our cases deemed “satisfactory but limited by . . .” were surprisingly cellular. By applying the broader published guidelines, only 3 of 29 cases were limited by scant cellularity. Thus, US per se does not seem to engender less than optimal smears. This review also demonstrated the usefulness of our terminology regarding follicular lesions. Microfollicular adenomas were found more frequently after cytological diagnosis of a microfollicular lesion than after diagnosis of a macrofollicular or mixed lesion. This terminology may be more clinically useful than attempting to employ the surgical pathology diagnostic categories for follicular lesions in cytology specimens because overlap exists in their use in cytology.25 A well recognized pitfall in interpretation of aspirates from the region of the thyroid gland is failure to consider the possibility of a parathyroid lesion unless suspicion is raised by the clinical information.26 Features that are found in FNABs of parathyroid adenoma and that were present in our case included nuclear chromatin stippling; delicate, clear, or granular cytoplasm easily stripped from cells, leaving naked nuclei; nuclear overlap; three-dimensional clusters; and microacini.27,28 Although a helpful clue is perivascular cell arrangement, only rare capillaries without clinging epithelial cells were seen in our case. Finally, with experience, technologist time does not seem to be significantly longer for thyroid FNAB compared with FNAB of other body sites. In conclusion, US guidance is useful for certain types of thyroid nodules. At our institution, US guidance of the thyroid is employed for a variety of situations. Only a limited number of incidental nodules are aspirated, in accordance with published recommendations. REFERENCES 1. 2. 3. 4. Mazzaferri EL. Management of a solitary thyroid nodule. N Engl J Med 1993;328:553–9. Rojeski MT, Gharib H. Nodular thyroid disease. N Engl J Med 1985;313:428 –36. Mazzaferri EL. Thyroid cancer in thyroid nodules: finding a needle in the haystack. Am J Med 1992;93:359 – 62. Vander JB, Gaston EA, Dawber TR. The significance of nontoxic thyroid nodules: final report of a 15 year study of the incidence of thyroid malignancy. Ann Intern Med 1968;69: 537– 40. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Mortenson JD, Woolner LB, Bennett WA. Gross and microscopic findings in clinically normal thyroid glands. J Clin Endocrinol Metab 1955;15:1270 – 80. Caruso D, Mazzaferri EL. Fine-needle aspiration biopsy in the management of thyroid nodules. Endocrinologist 1991; 1:194 –202. Oertel JE, Klinck GH. Structural changes in the thyroid glands of healthy young men. Med Ann District Columbia 1965;34:75–7. Ezzat S, Sarti DA, Cain DR, Braunstein GD. Thyroid incidentalomas. Prevalence by palpation and ultrasonography. Arch Intern Med 1994;154:1838 – 40. Brander A, Viikinkoski P, Nickels J, Kivisaari L. Thyroid gland: US screening in a random adult population. Radiology 1991;181:683–7. Woestyn J, Afschrift M, Schelstraete K, Vermeulen A. Demonstration of nodules in the normal thyroid by echography. Br J Radiol 1985;58:1179 – 82. Boland GW, Lee MJ, Mueller PR, Mayo-Smith W, Dawson SL, Simeone JF. Efficacy of sonographically guided biopsy of thyroid masses and cervical lymph nodes. A Am J Roentgenol 1993;161:1053– 6. McIvor NP, Freeman JL, Salem S, Elden L, Noyek AM, Bedard YC. Ultrasonography and ultrasound-guided fine-needle aspiration biopsy of head and neck lesions: a surgical perspective. Laryngoscope 1994;104:669 –74. The Papanicolaou Society of Cytopathology Task Force on Standards of Practice: Guidelines of the Papanicolaou Society of Cytopathology for examination of fine-needle aspiration specimens from thyroid nodules. Diagn Cytopathol 1996;15:84 –9. Hamburger JI, Husain M. Semiquantitative criteria for fineneedle biopsy diagnosis: reduced false-negative diagnoses. Diagn Cytopathol 1988;4:14 –7. Nguyen GK, Ginsberg J, Crockford PM. Fine-needle aspiration biopsy cytology of the thyroid. Its value and limitations in the diagnosis and management of solitary thyroid nodules. Pathol Annu 1991;26:63–91. DeMay RM. The art and science of cytopathology. Chicago: ASCP Press, 1996:725–9. Reading CC, Gorman CA. Thyroid imaging techniques. Clin Lab Med 1993;13:711–24. Goellner JR, Gharib H, Grant CS, Johnson DA: Fine needle aspiration cytology of the thyroid, 1980 –1986. Acta Cytol 1987;31:587–90. Schmidt T, Riggs MW, Speights VO. Significance of nondiagnostic fine-needle aspiration of the thyroid. South Med J 1997;90:1183– 6. McHenry CR, Walfish PG, Rosen IB. Non-dignostic fine needle aspiration biopsy: a dilemma in management of nodular thyroid disease. Am Surg 1993;59:415–9. MacDonald L, Yazdi HM. Nondiagnostic fine needle aspiration biopsy of the thyroid gland. A diagnostic dilemma. Acta Cytol 1996;40:423– 8. Khurana KK, Richards VI, Chopra PS, Izquierdo R, Rubens D, Mesonero C. The role of ultrasonography-guided fine needle aspiration biopsy in the management of non-palpable and palpable thyroid nodules. Thyroid 1998;8: 511–5. Tan GH, Gharib H. Thyroid incidentalomas: management approaches to nonpalpable nodules discovered incidentally on thyroid imaging. Ann Intern Med 1997;126:226 –31. US-Guided FNAB Thyroid/Tambouret et al. 24. Cochand-Priollet B, Guillausseau PJ, Chagnon S, Hoang C, Guillausseau-Scholer C, Chanson P, et al. The diagnostic value of fine-needle aspiration biopsy under ultrasonography in nonfunctional thyroid nodules: a prospective study comparing cytologic and histologic findings. Am J Med 1994; 97:152–7. 25. Sidawy MK, Del Vecchio DM, Knoll SM. Fine-needle aspiration of thyroid nodules, correlation between cytology and histology and evaluation of discrepant cases. Cancer (Cancer Cytopathol) 1997;81:253–9. 305 26. Layfield LJ. Fine needle aspiration cytology of cystic parathyroid lesions; a cytomorphologic overlap with cystic lesions of the thyroid. Acta Cytol 1991;35:447–50. 27. Kini U, Shareff S, Thomas JA. Ultrasonically guided fine needle aspiration of the parathyroid, a report of two cases. Acta Cytol 1993;37:747–51. 28. Abati A, Skarulis MC, Shawker T, Solomon D. Ultrasoundguided fine-needle aspiration of parathyroid lesions: a morphological and immunocytochemical approach. Hum Pathol 1995;26:338 – 43.