Abstract. Anatomic Pathology / PREOPERATIVE SELECTION OF PALPABLE THYROID NODULES

Transcription

1 Anatomic Pathology / PREOPERATIVE SELECTION OF PALPABLE THYROID NODULES Large-Needle Aspiration Biopsy for the Preoperative Selection of Palpable Thyroid Nodules Diagnosed by Fine-Needle Aspiration as a Microfollicular Nodule or Suspected Cancer Angelo Carpi, MD, 1 Andrea Nicolini, MD, 2 Andrea Sagripanti, MD, 2 Claudia Righi, PhD, 1 Fabrizio Menchini Fabris, MD, 1 and Giancarlo Di Coscio, PhD 3 Key Words: Fine-needle aspiration; FNA; Large-needle aspiration biopsy; LNAB; Microfollicular; Suspected diagnosis; Thyroid nodules Abstract The palpable thyroid nodules with a fine-needle aspiration (FNA) diagnosis of microfollicular nodule or suspected cancer usually are excised; however, most of them are proved benign by postoperative histologic examination. We reviewed the clinical and pathologic data for patients with thyroid nodules with an FNA diagnosis of microfollicular nodule or suspected cancer; nodules also were examined by large-needle aspiration biopsy (LNAB) to assess whether the distinction achieved by LNAB into pure microfollicular or mixed microfollicular-macrofollicular nodules could be used preoperatively to better predict malignancy. One hundred fourteen nodules of this type were excised. The prevalence of cancer was 22% (14/63) among the microfollicular and 4% (2/51) among the microfollicular-macrofollicular nodules at LNAB. These data indicate that histologic examination of the LNAB specimen can be used for preoperative selection of thyroid nodules diagnosed by FNA as a microfollicular nodule or suspected cancer. Needle aspiration techniques are the most efficient tools for preoperative selection of palpable thyroid nodules. In fact, the proportion of patients treated surgically fell markedly after introduction of the techniques. 1 However, their effectiveness is limited by the occurrence of some fine-needle aspiration (FNA) cytologic diagnoses that are not sufficiently accurate for defining the benign or malignant nature of the nodule. The most common indefinite FNA cytologic diagnoses are those of follicular neoplasm (or microfollicular nodule) and suspected malignant neoplasm. 1-3 These diagnoses occur in about 10% to 30% of all aspirates. 1-3 Although most of these preoperative cytologic findings are shown postoperatively to be benign, the probability of malignant neoplasm in these categories is considered high enough for surgical excision to be recommended for these patients. 2,3 The experience of the few authors, including us, who have used large-needle biopsy (LNB) in addition to FNA for the preoperative selection of palpable thyroid nodules is that LNB can be used to avoid surgery in a large proportion of patients with these cytologic diagnoses. 1,4-8 Previous studies suggested that the thyroid nodules diagnosed as microfollicular by FNA can be differentiated after histologic examination of the LNB specimen according to whether the pure microfollicular structure was confirmed or a macrofollicular component was associated. 1,7,8 Furthermore, preliminary findings indicated that LNB is useful for increasing the accuracy of the preoperative cytologic FNA diagnosis of suspected cancer. 1 The aim of the present study was to review the clinical and pathologic data for patients with thyroid nodules with an FNA diagnosis of microfollicular nodule or suspected 872 Am J Clin Pathol 2000;113: American Society of Clinical Pathologists

2 Anatomic Pathology / ORIGINAL ARTICLE cancer; nodules also were examined by large needle aspiration biopsy (LNAB), to assess whether the distinction achieved by LNAB into pure microfollicular or microfollicular-macrofollicular nodules could be used preoperatively to better predict malignancy. Materials and Methods Patients A total of 6,630 patients were examined between April 1980 and December 1997, including 760 men (11.5%) and 5,870 women (88.5%) with an age range of 14 to 87 years. A single nodule was palpated in 4,549 subjects (68.6%), while 2,081 (31.4%) had multiple nodules. Hyperfunctioning nodules were depicted by thyroid scanning in 5% of cases. All patients underwent cytologic examination by FNA; 1,875 (28.3%) also were examined by LNAB to provide additional preoperative histologic information. One hundred eighty-seven patients with an FNA cytologic diagnosis of microfollicular nodule (n = 132) or suspected cancer (n = 55) who were given an LNAB diagnosis of microfollicular or microfollicular-macrofollicular nodule were selected. The FNA cytologic diagnoses of microfollicular nodule included the following features in all patients: high cellularity with microfollicular or rarely trabecular architecture, minimal colloid content, and small and regular nuclei. The FNA cytologic diagnoses of suspected cancer were grouped and specified as follows: suspected papillary cancer (10 patients), presence of rare papillae-similar structures or occasional pseudonucleoli, or monomorphic cells with large nuclei and granular widespread chromatin; or suspected follicular cancer (45 patients), absent colloid, very high cellularity, microfollicular and/or trabecular architecture, large and occasionally pleomorphic nuclei, nucleoli, and rare mitoses. The nodules described as having Hürthle cells were not included in the study. The postoperative pathologic reports of the nodules that were excised were collected and compared with the preoperative FNA and LNAB diagnosis. The 5 patients who were not operated on in our hospital were invited to undergo a clinical examination and to have the slides reviewed by our pathology department or to provide the official postoperative pathologic report. Fine-Needle Aspiration FNA has been described in many reports; we describe the technical aspects typical of our technique that are not used by most other investigators. A total of 1 ml or less of local anesthetic (lidocaine 2%) was injected subcutaneously to avoid discomfort from multiple punctures and from the subsequent insertion of the large needle for LNAB and histologic evaluation in some patients. All punctures were made through a single skin puncture by sliding the skin over the nodule to the various sites to be aspirated. Relatively large needles (22 to 20 gauge) were preferred for FNA, as they can provide relatively large tissue fragments that can be examined macroscopically at the time of smear preparation, thus indicating the adequacy of the specimens. The preparation of smears from several aspirations also allowed us to evaluate the different cytologic features that can be shown to advantage by air-dried smears stained with Giemsa stain or fixed smears stained with the Papanicolaou method. Large-Needle Aspiration Biopsy The most common LNB method used for diagnosis of thyroid nodules is a core (or cutting) needle biopsy performed with Tru-cut needles (Baxter, Deerfield, IL) This technique requires a skin incision for needle insertion and has been used as an alternative to FNA. 5 We adopted a simpler method of LNB using smaller hypodermic needles (16 or 18 gauge) in combination with FNA. The needle, inserted into the nodule through the skin puncture already made for the previous FNA punctures, is rotated within the nodule, so that the sharp end of the needle severs the tissue fragments, which then are aspirated into the barrel of the syringe (hence, the term large-needle aspiration biopsy). A simpler way to obtain tissue fragments is to perform the same procedure as for FNA, only more vigorously. This method was first described in and more recently in greater detail. 7-9,13 The patient usually does not experience more pain or discomfort than with the FNA method. Therefore, without any planning, LNAB can be performed at the end of the FNA procedure in the case of an easily palpable nodule. As the tissue fragments are usually smaller than those obtained with the core needle biopsy method, we used the celloidin bag technique to optimize tissue fragment collection. 14 We performed LNAB on the nodules that were readily palpable, solid, and larger than 1.5 cm in diameter. Statistical Analysis The chi-square test was performed to assess whether the prevalence of cancer postoperatively was significantly different in the 2 nodule groups (microfollicular or microfollicular-macrofollicular at preoperative LNAB). Results As reported in a previous more comprehensive series, 1 no morbidity or important complication occurred following American Society of Clinical Pathologists Am J Clin Pathol 2000;113:

3 Carpi et al / PREOPERATIVE SELECTION OF PALPABLE THYROID NODULES Table 1 Principal Features for 187 Patients With Thyroid Nodules * Postoperative Findings Preoperative Diagnostic Sex Mean Age Mean Nodule No. (%) No. (%) No. (%) Combination FNA/LNAB (F/M) (y) Size (cm) Excised Benign Malignant Histotype Microfollicular/microfollicular 51/ (58) 31 (84) 6 (16) Follicular, 4; (n = 64) papillary; 2 Microfollicular/microfollicular- 49/ (53) 35 (97) 1 (3) Papillary, macrofollicular (n = 68) follicular variant, 1 Suspected 28/ (74) 1 8 (69) 8 (31) Follicular, 3; cancer/microfollicular (n = 35) papillary, 5 Suspected 19/ (75) 14 (93) 1 (7) Papillary, 1 cancer/microfollicularmacrofollicular (n = 20) Total or mean 147/ (61.0) 98 (86.0) 16 (14.0) Follicular, 7; papillary, 8; papillary, follicular variant, 1 * Patients had a preoperative cytologic diagnosis by fine-needle aspiration (FNA) of a microfollicular nodule or suspected cancer that was differentiated by large-needle aspiration biopsy (LNAB) findings; comparison is between preoperative and postoperative findings of 114 excised nodules. the LNAB. Table 1 shows the principal clinical and postoperative pathologic features of the 187 patients operated on for palpable thyroid nodules with 1 of the aforementioned preoperative FNA and LNAB diagnostic combinations. All except 5 patients were operated on in our hospital, and their pathologic diagnosis was determined in our Department of Pathology. For 64 patients, the preoperative diagnosis was pure microfollicular nodule by both FNA and LNAB. Thirty-seven (58%) underwent surgery (29 women and 8 men; mean age, 44.2 years [range, years]; mean nodule size, 3.2 cm [range, 1-6 cm]). The results of subsequent histologic examination were benign in 31 (84%; 25 women and 6 men; mean age, 43.3 years [range, years]; mean nodule size, 3.0 cm [range, 1-6 cm]) and malignant in 6 (16%; 4 follicular cancer, 2 papillary cancer; 5 women and 1 man; mean age, 45.1 years [range, years]; mean nodule size, 3.4 cm [range, 2-5 cm]). For 68 patients, the preoperative diagnostic combination was microfollicular nodule by FNA and microfollicular-macrofollicular nodule by LNAB. Thirty-six (53%) underwent surgery (24 women and 12 men; mean age, 43.0 years [range, years]; mean nodule size, 2.7 cm [range, 1-6 cm]). The results of subsequent histologic examination were benign in 35 (97%; mean age, 43.0 years [range, years]; mean nodule size, 2.7 cm [range, 1-6 cm]) and malignant in 1 (3%; papillary cancer follicular variant; 1 woman; 43.0 years old; nodule size, 2.5 cm). For 35 patients, the preoperative diagnostic combination was suspected cancer by FNA and pure microfollicular by LNAB. Twenty-six (74%) underwent surgery (20 women and 6 men; mean age, 45.4 years [range, years]; mean nodule size, 2.5 cm [range, 1-5 cm]). Results of the subsequent histologic examination were benign in 18 (69%; 14 women and 4 men; mean age, 45.1 years [range, years]; mean nodule size, 2.4 cm [range, 1-5 cm]) and malignant in 8 (31%; 3 follicular cancer, 5 papillary cancer; 6 women and 2 men; mean age, 45.7 years [range, years]; mean nodule size, 2.6 cm [range, cm]). For 20 patients, the diagnostic combination was suspected cancer by FNA and microfollicular-macrofollicular nodule by LNAB. Fifteen (75%) underwent surgery (14 women and 1 man; mean age, 46.0 years [range, years]; mean nodule size, 2.3 cm [range, 1-4 cm]). Results of the subsequent histologic examination were benign in 14 (93%; 13 women and 1 man; mean age, 50.0 years [range, years]; mean nodule size, 2.2 cm [range, 1-4 cm]) and malignant in 1 (6%; papillary cancer; 1 woman; 42.0 years old; nodule size, 2.5 cm). The chi-square test showed that prevalence of cancer postoperatively was significantly different in the 2 nodule groups (microfollicular or microfollicular-macrofollicular at preoperative LNAB) (chisquare, 6.4; P <.025). Table 2 shows the principal clinical features of 73 patients with the aforementioned preoperative FNA and LNAB diagnostic combinations who were not operated on and specifies for each category of diagnostic combinations the number of patients who were followed up or not available for follow-up. There was no important difference in the principal clinical features among the various patient groups. The testing results for these patients showed no malignancy. 874 Am J Clin Pathol 2000;113: American Society of Clinical Pathologists

4 Anatomic Pathology / ORIGINAL ARTICLE Table 2 Principal Features for 73 Patients With Thyroid Nodules Who Did Not Undergo Surgery * Followed Up Preoperative Diagnostic Mean Nodule Size Combination FNA/LNAB Sex (F/M) Mean Age (y) (cm) Yes/No. (%) No/No. (%) Microfollicular/microfollicular 22/ (41) 16 (59) (n = 27) Microfollicular/microfollicular- 25/ (50) 16 (50) macrofollicular (n = 32) Suspected cancer/microfollicular 8/ (17) 5 (83) (n = 6) Suspected cancer/microfollicular- 5/ (40) 3 (60) macrofollicular (n = 5) Total or mean 60/ (43) 40 (57) * Patients had a cytologic fine-needle aspiration (FNA) diagnosis of microfollicular nodule or suspected cancer that was differentiated by the large-needle aspiration biopsy (LNAB) findings. Discussion Two important studies examined the factors that help to predict malignancy in palpable thyroid nodules that have been assigned an indeterminate cytologic diagnosis of follicular neoplasm or suspected malignancy by FNA. 15,16 The first article reported the follow-up data for 187 patients with these FNA diagnoses and concluded that no clinical, scintigraphic, or ultrasonographic characteristic predicted the presence of a malignant lesion. 15 This study also included 71 cases with a cytologic FNA diagnosis of Hürthle cell neoplasm, while we did not include our few cases with this diagnosis. The second article more specifically analyzed the data for the nodules that had been diagnosed by FNA as being suggestive of follicular neoplasm. 16 The investigators found 26 malignant lesions (11.9%) among the nodules examined in 219 patients who underwent surgery and that the clinical findings associated with a malignant index nodule were fixation of the mass, larger diameter, and younger age. Fixation of the mass was a specific predictor; however, it was rare, as it occurred in only 3 patients. Malignancy was observed in 29% of the nodules larger than 4 cm, but only 9% of the smaller nodules were cancers. This last correlation can be explained readily because malignant nodules usually grow faster than benign nodules; however, it also is clear that cancers are small before becoming large and that we do not need to wait for them to grow before diagnosing their malignant nature. In the same study, the risk of cancer was decreased by a factor of 0.7 for every 10-year increase in age. 17 The relationship between age and risk of thyroid cancer is controversial 17 ; however, age alone cannot be used for diagnosis. These 2 studies 15,16 did not report or comment on definite pathologic criteria or on the use of LNB. Miller et al, 18 while describing their experience with 1,005 patients with thyroid nodules examined by FNA and LNB and who subsequently underwent surgery, reported that of all the nodules with an FNA diagnosis of cellular adenoma and suspected cancer also studied by LNB, 50% and 20%, respectively, received a revised diagnosis of benign. Carpi et al 1,8 and Lo Gerfo 5 specified that about half of all the nodules diagnosed by FNA as microfollicular or follicular neoplasm were hyperplastic benign mixed microfollicular-macrofollicular nodules by LNB. Carpi et al 1 also reported that by the addition of LNAB to the nodules diagnosed as suspected cancer by FNA, it was possible to distinguish groups of nodules with significantly different probabilities of malignancy. Lo Gerfo 19 clearly expressed the hypothesis that the LNB finding of macrofollicles in the nodules diagnosed as microfollicular by FNA cytology could be useful to differentiate between possible follicular cancer and benign hyperplastic nodules. To verify this hypothesis, Carpi et al 20,21 reported in 2 preliminary studies the postoperative pathologic findings of palpable thyroid nodules, all preoperatively diagnosed by FNA cytology as microfollicular but distinguished by the LNAB finding of pure microfollicles or mixed microfollicles and macrofollicles. These studies showed that malignancy was absent in the mixed microfollicular-macrofollicular nodules, while it was rather frequent in the pure microfollicular nodules. In the present study, we expanded the study of the nodules that were microfollicular by FNA, and we extended the observation to the nodules with the FNA cytologic diagnosis of suspected cancer and compared the role of the LNAB findings in predicting malignancy with that of selected clinical factors (sex, age, nodule size). We confirmed that the clinical factors are not helpful for predicting the malignancy of the nodules examined. LNAB effectively distinguished the nodules found to be microfollicular at FNA. Similarly effective was the distinction made by American Society of Clinical Pathologists Am J Clin Pathol 2000;113:

5 Carpi et al / PREOPERATIVE SELECTION OF PALPABLE THYROID NODULES LNAB among the nodules with the cytologic diagnosis of suspected malignancy. We observed only 1 case of malignant neoplasm among the excised nodules that were diagnosed preoperatively as microfollicular by FNA and as microfollicular-macrofollicular by LNAB. This was 1 of the 5 cases not operated on in our hospital; our expert pathologist who reviewed these slides stated that there was no cancer present. Nevertheless, we considered it as malignant because the slides of the remaining 4 patients operated on elsewhere had not been reviewed. Two cancers were found among the nodules studied that showed a macrofollicular component at LNAB, and the diagnosis for one was questionable. In a previous smaller series, 21 we found no cancer after excision of nodules that were microfollicular at FNA and microfollicular-macrofollicular at LNAB. However, the series reported in the present article is too limited to fully assess the prevalence of cancers missed using LNAB as we suggest. Sixteen cancers were found postoperatively (7 follicular and 9 papillary), corresponding to 14% of all excised nodules. Among the cancers excised with a preoperative FNA diagnosis of microfollicular nodule, the follicular type was more frequent (57%). However, as in other studies, 15,16 a significant number of papillary cancers was observed among nodules of this type at FNA. It is easily understandable why follicular cancers can occur much more frequently or exclusively in pure microfollicular nodules. To explain why papillary cancer also can be more frequent in pure microfollicular than in microfollicular-macrofollicular nodules, we hypothesize that the tissue sampled by the needle biopsy in these cancers is the follicular component of a mixed cancer or a follicular variant of papillary thyroid carcinoma that usually is composed of small follicles. 22 Whatever the case, in the present study we selected a simple, objective, well-defined histologic parameter to select the nodules defined by FNA as microfollicular or as suspected cancer. In a previous study, Carpi et al 1 showed that the LNAB histologic diagnosis of suspected cancer was useful to select the nodules that had been diagnosed by FNA cytology as suspected cancer. However, the LNAB diagnosis of suspected cancer depended on multiple morphologic elements 1 and was, to some extent, subjective. The technique of LNB we used, LNAB, is simpler than the core needle biopsy technique used by Lo Gerfo 5 and Hamburger et al. 9 LNAB uses smaller needles and usually obtains smaller tissue fragments; however, both LNB techniques seem to obtain similar results when distinguishing the nodules found to be microfollicular by FNA into pure microfollicular and mixed microfollicular-macrofollicular nodules by histologic examination. 1,5,8 At the present time, no other laboratory test has proved as useful as LNAB or has a similar cost-effectiveness ratio for predicting malignancy in the nodules with an FNA cytologic diagnosis of microfollicular nodule or suspected malignancy Therefore, we believe that LNAB should be used for this purpose. From the Departments of 1 Reproduction and Aging, 2 Internal Medicine, and 3 Oncology-Pathology, University of Pisa, Pisa, Italy. Address reprint requests to Dr Carpi: Dipartimento di Medicina della Procreazione e dell Età evolutiva, Ospedale Santa Chiara, Via Roma, Pisa, Italy. Acknowledgment: We thank B. Shapiro, MD, University of Michigan, Ann Arbor, for careful revision of the manuscript. References 1. Carpi A, Ferrari E, Toni MG, et al. Needle aspiration techniques in preoperative selection of patients with thyroid nodules: a long term study. J Clin Oncol. 1996;14: Gharib H, Goellner J. Fine-needle aspiration biopsy of the thyroid: an appraisal. Ann Intern Med. 1993;118: Gharib H. Fine-needle aspiration biopsy of thyroid nodules: advantages, limitations, and effect. Mayo Clinic Proc. 1994;69: Carpi A, Di Coscio G, Toscano S. Needle aspiration of thyroid nodule: long-term control of its efficiency in preoperative selection. Thyroidol Clin Exp. 1988;1: Lo Gerfo P. The value of coarse needle biopsy in evaluating thyroid nodules. Thyroidol Clin Exp. 1994;6: Carpi A, Iervasi G, Nicolini A, et al. One-year experience with aspiration of thyroid nodules. In: Andreoli M, Robbins J, Monaco F, eds. Advances in Thyroid Neoplasia. Rome, Italy: Field Educational; 1981: Carpi A, Toni MG, Nicolini A, et al. Progress in the management of thyroid nodule patients. In: Carpi A, Sagripanti A, Mittermayer C, eds. Progress in Clinical Oncology. Munich, Germany: Sympomed; 1992: Carpi A, Ferrari E, De Gaudio C, et al. The value of aspiration needle biopsy in evaluating thyroid nodules. Thyroidol Clin Exp. 1994;6: Hamburger JI, Miller JM, Kini SR. Clinical Pathological Evaluation of Thyroid Nodules Handbook and Atlas. Private publication; Wang C, Vickery AL Jr, Maloof F. Needle biopsy of the thyroid. Surg Gynecol Obstet. 1976;143: Lo Gerfo P, Colacchio T, Caushaj F, et al. Comparison of fine needle and coarse needle biopsies in evaluating thyroid nodules. Surgery. 1982;92: Martin HE, Ellis EB. Biopsy by needle puncture and aspiration. Ann Surg. 1930;92: Crile G Jr, Hawk WA Jr. Aspiration biopsy of thyroid nodules. Surg Gynecol Obstet. 1973;136: Bussolati G. A celloidin bag for histological preparation of cytologic material. J Clin Pathol. 1982;35: Cersosimo E, Gharib H, Suman SJ, et al. Suspicious thyroid cytologic findings: outcome in patients without immediate surgical treatment. Mayo Clin Proc. 1993;68: Am J Clin Pathol 2000;113: American Society of Clinical Pathologists

6 Anatomic Pathology / ORIGINAL ARTICLE 16. Schlinkert RT, van Heerden JA, Goellner JR, et al. Factors that predict malignant thyroid lesions when fine-needle aspiration is suspicious for follicular neoplasm. Mayo Clin Proc. 1997;72: Mazzaferri EL, De Los Santos ET, Rofagha-Keyhani S. Solitary thyroid nodule: diagnosis and management. Med Clin North Am. 1988;72: Miller MJ, Kini SR, Hamburger JI. The diagnosis of malignant follicular neoplasm of the thyroid by needle biopsy. Cancer. 1985;55: Lo Gerfo P. Coarse needle biopsy of the thyroid. In: Hamburger JI, ed. Diagnostic Methods in Clinical Thyroidology. New York, NY: Springer; 1989: Carpi A, Ferrari E, Sagripanti A, et al. Aspiration needle biopsy refines preoperative diagnosis of thyroid nodules defined at fine needle aspiration as microfollicular nodule. Biomed Pharmacother. 1996;50: Carpi A, Fabris FM, Ferrari E, et al. Aspiration needle biopsy in preoperative selection of thyroid nodules defined at fine needle aspiration as microfollicular lesions. Am J Clin Oncol. 1999;22: Albores-Saavedra J, Gould E, Vardaman C, et al. The macrofollicular variant of papillary thyroid carcinoma: a study of 17 cases. Hum Pathol. 1991;22: Komatsu M, Kuroda T. Flow cytometric nuclear DNA analysis of follicular thyroid nodules. Res Surg. 1995;7: Karmakar T, Dey P. Role of AgNOR in diagnosis of thyroid follicular neoplasms on fine-needle aspiration smears. Diagn Cytopathol. 1995;12: Delbridge L, Lean CL, Russel P, et al. Proton magnetic resonance and human thyroid neoplasia: a potential avoidance of surgery for benign follicular neoplasm. World J Surg. 1994;18: Yamashita H, Noguchi S, Muracami N, et al. Immunohistological differentiation of benign thyroid follicular cell tumors from malignant ones: usefulness of antiperoxidase and JT-95 antibodies. Acta Pathol Jpn. 1993;43: Carpi A, Nicolini A, Sagripanti A. Protocols for the preoperative selection of palpable thyroid nodules: review and progress. Am J Clin Oncol. 1999;22: American Society of Clinical Pathologists Am J Clin Pathol 2000;113: