Transthoracic Fine-Needle Aspiration vs Concurrent Core Needle ...

Anatomic Pathology / DIAGNOSTIC ACCURACY OF FNA AND CORE BIOPSY

Transthoracic Fine-Needle Aspiration vs Concurrent Core Needle Biopsy in Diagnosis of Intrathoracic Lesions A Retrospective Comparison of Diagnostic Accuracy Yun Gong, MD,1 Nour Sneige, MD,1 Ming Guo, MD,1 Marshall E. Hicks, MD,2 and Cesar A. Moran, MD1 Key Words: Fine-needle aspiration; Core needle biopsy; Accuracy; Intrathoracic lesions; Malignant; Benign DOI: 10.1309/FK9XMJ19T036LJ70

Abstract To assess the value and limitations of fine-needle aspiration (FNA) and core needle biopsy (CNB) in the diagnosis of intrathoracic lesions, we retrospectively compared the diagnostic accuracy of 362 FNA and concurrent CNB procedures performed on 350 patients. Based on the final diagnoses that were determined based on combined information from biopsy, resection, clinical, radiologic, and microbiologic findings, the study cases were grouped into 188 malignant, 161 benign, and 13 inconclusive lesions. FNA and CNB yielded similar diagnostic accuracy for malignant tumors (85.1% vs 86.7%) and epithelial malignant neoplasms (86.4% vs 85.2%), whereas CNB yielded better diagnostic accuracy (96%) than FNA (77%) for nonepithelial malignant neoplasms. Combined FNA and CNB substantially improved the rate of malignancy diagnosis (95.2%). Of 161 benign cases, 50 were proven to be benign-specific lesions; FNA provided specific diagnosis in 20 (40%) and CNB in 46 (92%). The remaining 111 benign lesions yielded benignnonspecific findings on both specimens. These results indicate that CNB should be obtained when clinical or radiologic findings do not match the cytologic findings or nonepithelial lesions and benign lesions are considered likely.

438 438

Am J Clin Pathol 2006;125:438-444 DOI: 10.1309/FK9XMJ19T036LJ70

Transthoracic fine-needle aspiration (FNA), a simple and cost-effective diagnostic tool, is an important integral part in the evaluation of intrathoracic lesions and often is used as the initial diagnostic procedure. FNA provides valuable information that enables clinicians to make optimal therapeutic strategies, including planning preoperative management for patients with operable tumors and choosing adequate medical therapy for patients with nonresectable lesions or for those who are poor surgical candidates owing to advanced age or impaired cardiopulmonary function. However, interpretation of FNA specimens sometimes is limited by a small sample, a sampling error, or the lack of a histologic pattern.1,2 Core needle biopsy (CNB) is another commonly used but more invasive procedure for evaluating intrathoracic lesions. It has the advantage of preserving the tissue architecture for histologic evaluation. Although CNB is, in general, safe when used judiciously,3-7 it may be associated with increased complications.8-10 Controversy exists about the value and limitations of FNA and CNB, and it is unclear whether both techniques are indicated in the evaluation of intrathoracic lesions.11-14 Although several studies have compared the diagnostic accuracy of FNA and CNB for thoracic lesions, the value of these studies is limited by small series13,15-18 or by design, in which 2 sets of lesions from different patients were compared.11,14,19-21 We retrospectively reviewed our experience with these 2 techniques that were performed sequentially during the same procedure for intrathoracic lesions from 350 patients. The large case number in this study and its direct comparison of these 2 techniques in the same lesions allow us to obtain reliable data without bias in regard to the nature of the lesions, such as their location, consistency, and size. © American Society for Clinical Pathology

Anatomic Pathology / ORIGINAL ARTICLE

Materials and Methods From January 2001 to July 2003, 350 patients at The University of Texas M.D. Anderson Cancer Center, Houston, underwent 362 sequential FNA and CNB procedures during the same visit for evaluation of intrathoracic lesions. The study was conducted with the approval of the M.D. Anderson Cancer Center Institutional Review Committee. Twelve lesions were rebiopsied because of failure to obtain diagnostic specimens during the previous biopsies (ie, FNA and concurrent CNB). Clinical and radiologic information for all patients was available for review. The patient population included 171 males (48.9%) and 179 females (51.1%). The mean age at presentation was 62 years (range, 9-88 years). The lesions ranged from 0.5 to 12 cm in diameter (mean, 2.3 cm); 332 of them were pulmonary lesions, 3 were pleural lesions, and 15 were mediastinal lesions. Transthoracic, computed tomography–guided FNA was performed using a 20- or 22-gauge aspirating needle through a coaxial 18-gauge guide needle (Cook, Bloomington, IN). Usually, 2 or 3 FNA passes were made. Direct smears were air-dried for Diff-Quik staining (StatLab, Lewisville, TX) or fixed in modified Carnoy fixative (a 6:1 ratio of 70% ethanol to glacial acetic acid) for Papanicolaou staining. Smears of each case were assessed immediately by an onsite cytopathologist for specimen adequacy. An aspirate specimen was considered inadequate if the sample was nonrepresentative (ie, contained only blood, normal lung parenchyma, fibrosis, histiocytes, necrotic debris, nonspecific inflammations, or poorly preserved cells or had scant cellularity incompatible with clinical and/or radiologic findings). Accordingly, additional aspirates were obtained provided the patient remained asymptomatic and no pneumothorax or hemoptysis had occurred. The on-site cytopathologist triaged the specimens for appropriate study, including requesting cell block material, especially when immunochemical workup likely was needed, obtaining additional aspirates for flow cytometric immunophenotyping when non-Hodgkin lymphoma was suspected, and sending fresh samples to the microbiology laboratory for staining and culture when an infectious process needed to be ruled out. For cell block sections, cells from aspiration or needle rinses collected in RPMI medium were centrifuged, and the sediment was fixed in a 50:50 mixture of 95% ethanol and 10% formalin, embedded in paraffin, sectioned, and stained with H&E. The cytologic diagnosis was made based on the cytomorphologic features of the smears, cell block preparations, and the results of immunocytochemical stains, special stains (for mucin or microorganisms), flow cytometric immunophenotyping, cytogenetics, and molecular study. Staining was performed on cell block sections or Papanicolaou-stained smears. © American Society for Clinical Pathology

CNB was indicated when the FNA smears, on immediate assessment, were considered suboptimal or nonrepresentative of clinical and/or radiologic findings or when the lesions were small and local bleeding that would reduce visibility for subsequent aspirations was imminent. The latter were determined by the radiologist who performed the procedures. CNB also was performed if FNA smears showed features suggestive of malignant lymphoma or sarcoma to facilitate histologic classification, particularly in cases of first-time diagnosis. CNB was performed using a 20-gauge, coaxial, automated cutting needle biopsy system (Cook). Usually, 2 to 4 cores were obtained. Before the core biopsy specimen was transferred to 10% formalin fixative, cytologic imprints of CNB specimens sometimes were made (available in 12% of cases) by gently touching the CNB tissue to a glass slide. The air-dried slides were stained and immediately assessed by the on-site cytopathologist to determine whether the core tissues contained material representative of the target lesion. The interpretations of the touch imprints, along with the imprint slides, were submitted to the pathologist signing out the concurrent CNB. The formalin-fixed core tissues were embedded in paraffin, sectioned, and stained with H&E. Immunohistochemical stains and special stains were performed when indicated. FNA and CNB specimens were interpreted separately by different pathologists in 92% of cases and by the same pathologist in 8% of cases. The major diagnostic categories for FNA and CNB were malignant, atypical or “suspicious” (when atypical cells had quantitatively and/or qualitatively fallen short of the criteria for malignancy), benign (ie, negative for malignancy), and inadequate. For cases classified as positive for malignancy, an attempt was made to determine the histologic type, whether it was a primary or metastatic lesion, and, if it was a metastasis, its origins. Within the group of poorly differentiated non–small cell carcinomas, minor differences between FNA and CNB diagnoses with regard to favoring squamous vs glandular differentiation was not considered a typing discrepancy because dual differentiation is common and the difference is not of therapeutic relevance.22 In the cases of benign lesions, if the pathologic findings were compatible with the radiologic and clinical impressions, they were defined as benign-specific lesions; this group included lesions such as granuloma, fungal or mycobacterial infection, organizing pneumonia, and benign tumors. If biopsy material comprised only normal parenchyma, fibrosis, histiocytes, necrotic debris, and nonspecific inflammation, it was classified as a benign-nonspecific lesion or inadequate sampling. The diagnostic accuracy of each modality and the combination of both was judged by comparing the diagnosis with the final diagnosis of the lesion. The latter was determined based on the combined information from definitive biopsy diagnosis (FNA and/or CNB), diagnosis of resected specimens (available in 53% of primary tumors and 23% of Am J Clin Pathol 2006;125:438-444 439 DOI: 10.1309/FK9XMJ19T036LJ70

439 439

Gong et al / DIAGNOSTIC ACCURACY OF FNA AND CORE BIOPSY

metastatic tumors), clinical and radiologic impressions, microbiologic findings, follow-up data, and therapeutic response. All cases with final diagnoses of malignancy had unequivocal pathologic findings. For final diagnosis of a benign-nonspecific lesion without surgical tissue confirmation, a follow-up of at least 2 years in which the lesion was stable or regressed was required.

Results The final diagnosis in 188 cases was malignant tumor and in 161 cases was benign lesion. The remaining 13 cases were inconclusive because of the lack of pathologic, radiologic, or clinical evidence to determine the nature of the lesions and, thus, were excluded from further analysis. Among the 349 cases that constituted the final cohort, FNA and CNB findings agreed completely in 279 cases (79.9%), 148 of which were malignant and 131 were benign.

❚Table 2❚ Diagnostic Accuracy of FNA, CNB, and Both in 188 Malignant Tumors* Diagnosis Malignant FNA CNB FNA and CNB

Atypical or “Suspicious”†

160 (85.1) 163 (86.7) 179 (95.2)

13 (6.9) 9 (4.8)

NFM or Inadequate‡ 15 (8.0) 16 (8.5)

CNB, core needle biopsy; FNA, fine-needle aspiration; NFM, negative for malignancy. * Data are given as number (percentage). † Atypical cells had quantitatively and/or qualitatively fallen short of the criteria for malignancy. ‡ Encompassing specimens containing only blood, fibrosis, histiocytes, necrotic debris, normal lung parenchyma, or nonspecific inflammation.

Among the 188 malignant cases, 131 were primary tumors (129 carcinomas and 2 mesotheliomas), 44 were metastatic tumors (31 carcinomas, 10 sarcomas, and 3 melanomas), and 13 were lymphomas (9 non-Hodgkin lymphoma and 4 Hodgkin lymphoma) ❚Table 1❚. Epithelial malignant neoplasms (encompassing carcinoma and mesothelioma) accounted for 86.2% (162/188) of the malignant tumors, and nonepithelial malignant neoplasms (encompassing sarcoma, lymphoma, and melanoma) constituted the remaining 13.8% (26/188). The indeterminate diagnosis (ie, atypical or suspicious category) was made in 6.9% (n = 13) of FNA cases and in 4.8% (n = 9) of CNB cases; the false-negative rates were 8.0% (n = 15) and 8.5% (n = 16), respectively. The overall diagnostic accuracy rates of FNA and CNB for malignant tumors were 85.1% and 86.7%, respectively ❚Table 2❚. For epithelial malignant neoplasms, definitive diagnosis was made in 86.4% (140/162) of the cases by FNA, similar to the 85.2% (138/162) by CNB. Within this group, FNA and CNB had similar typing accuracy rates for small cell carcinoma, adenocarcinoma, squamous cell carcinoma, and non–small cell carcinoma. For nonepithelial malignant neoplasms, however, CNB yielded better diagnostic accuracy than FNA, especially for sarcoma and lymphoma. The diagnostic yield for nonepithelial malignant neoplasms was 77% (20/26) by FNA and 96% (25/26) by CNB. Notably, 34 (18.1%) of the malignant tumors were diagnosed by only one of the modalities, 16 by FNA only ❚Table 3❚ and 18 by CNB only ❚Table 4❚. When the FNA and CNB results were reviewed together, diagnostic accuracy for malignancy was improved substantially, to 95.2%. Of 31 metastatic carcinomas, the FNA diagnosis was correct in 26 cases (84%) and the CNB diagnosis in 20 cases (65%); among them, 14 FNA and 12 CNB samples had available

❚Table 1❚ Diagnostic Accuracy of FNA, CNB, and Both in 188 Epithelial and Nonepithelial Malignant Neoplasms* Final Diagnosis Epithelial tumors Poorly differentiated non–small cell carcinoma (n = 41) Adenocarcinoma (n = 77) Squamous carcinoma (n = 33) Small cell carcinoma (n = 3) Neuroendocrine carcinoma (n = 6) Mesothelioma (n = 2) Subtotal (n = 162) Nonepithelial tumors Sarcoma (n = 10) Non-Hodgkin lymphoma (n = 9) Hodgkin lymphoma (n = 4) Melanoma (n = 3) Subtotal (n = 26) Total

FNA

CNB

FNA and CNB

36 (88) 66 (86) 30 (91) 3 (100) 4 (67) 1 (50) 140 (86.4)

34 (83) 64 (83) 29 (88) 3 (100) 6 (100) 2 (100) 138 (85.2)

40 (98) 71 (92) 31 (94) 3 (100) 6 (100) 2 (100) 153 (94.4)

7 (70) 7 (78) 3 (75) 3 (100) 20 (77) 160 (85.1)

9 (90) 9 (100) 4 (100) 3 (100) 25 (96) 163 (86.7)

10 (100) 9 (100) 4 (100) 3 (100) 26 (100) 179 (95.2)

CNB, core needle biopsy; FNA, fine-needle aspiration. * Data are given as number (percentage).

440 440


© American Society for Clinical Pathology


immunostains. Of 10 metastatic sarcomas, 7 (70%) and 9 (90%) were recognized by FNA and CNB, respectively. All 3 metastatic melanomas were classified correctly by both techniques. Among 161 benign cases, 50 were proven to be benignspecific lesions. FNA was diagnostic in 20 cases and CNB in 46 cases, accounting for 12.4% and 28.6% of all benign cases, respectively. The diagnostic accuracy rates for benign-specific lesions were 40% for FNA and 92% for CNB. The combination of both recognized 48 benign-specific cases (96%). In 2 cases in which the initial interpretations from both FNA and CNB were benign-nonspecific, the subsequent resection specimens showed granulomatous inflammation. FNA and CNB yielded discrepant diagnoses in 30 cases of benign-specific lesions; 28 of them were diagnosed by CNB only (24 granulomas, 1 hamartoma, 2 cases of organizing pneumonia, and 1 neurofibroma), and 2 were identified on FNA smears only, which revealed organisms consistent with Cryptococcus species) ❚Table 5❚. ❚Image 1❚ showed a granuloma that was diagnosed on CNB only. The remaining 111 benign cases showed benign-nonspecific features on both specimen types. Although the pathologic nature of lesions could not be defined specifically, they were considered benign because all of the patients had benign clinical and radiologic courses (stability or regression of the lesions during at least a 2-year follow-up period). Reactive epithelial atypia with associated inflammation was found in 2 FNA samples (both proven to be granulomas by CNB) and in 3 CNB samples (2 with granulomatous inflammation and 1 with a benign-nonspecific lesion). In this study, 53.4% of 131 primary tumors and 23% of 44 metastatic tumors had follow-up resections. No false-positive diagnosis of malignancy was observed in these tumors with either technique.

Discussion Therapy for thoracic lesions depends on an accurate pathologic diagnosis, which relies on the skill of the operator,

❚Table 3❚ Malignant Cases Diagnosed by FNA Only Core Needle Biopsy FNA Poorly differentiated non–small cell carcinoma (n = 6) Adenocarcinoma (n = 7) Squamous carcinoma (n = 2) Sarcoma (n = 1) Total (N = 16)

Atypical or “Suspicious”*

NFM or Inadequate†

2

4

1 0 0 3

6 2 1 13

FNA, fine-needle aspiration; NFM, negative for malignancy. * Atypical cells had quantitatively and/or qualitatively fallen short of the criteria for malignancy. † Encompassing specimens containing only blood, fibrosis, histiocytes, necrotic debris, normal lung parenchyma, or nonspecific inflammation.

❚Table 4❚ Malignant Cases Diagnosed by CNB Only Fine-Needle Aspiration Atypical or NFM or “Suspicious”* Inadequate†

CNB Poorly differentiated non–small cell carcinoma (n = 4) Adenocarcinoma (n = 5) Squamous carcinoma (n = 1) Neuroendocrine carcinoma (n = 2) Sarcoma (n = 3) B-cell lymphoma (n = 2) Hodgkin lymphoma (n = 1) Total (N = 18)

1

3

3 0 1 2 0 0 7

2 1 1 1 2 1 11

CNB, core needle biopsy; NFM, negative for malignancy. * Atypical cells had quantitatively and/or qualitatively fallen short of the criteria for malignancy. † Encompassing specimens containing only blood, fibrosis, histiocytes, necrotic debris, normal lung parenchyma, or nonspecific inflammation.

experience of the interpreter, and nature of the lesion.22-25 The ease with which most lung cancers can be diagnosed by FNA justifies the ready acceptance and wide use of this technique. Several studies have documented the reliability of FNA in the diagnosis of intrathoracic malignant tumors, with most reported

❚Table 5❚ Diagnostic Accuracy of FNA, CNB, and Both in 50 Benign-Specific Lesions* Final Diagnosis Granuloma (n = 39)† Fungal infection (n = 4)‡ Organizing pneumonia (n = 2) Hamartoma (n = 2) Solitary fibrous tumor (n = 1) Neurofibroma (n = 1) Thymoma (n = 1)

FNA 13 (33) 4 (100) 0 (0) 1 (50) 1 (100) 0 (0) 1 (100)

CNB 37 (95) 2 (50) 2 (100) 2 (100) 1 (100) 1 (100) 1 (100)

FNA and CNB 37 (95) 4 (100) 2 (100) 2 (100) 1 (100) 1 (100) 1 (100)

CNB, core needle biopsy; FNA, fine needle aspiration. * Data are given as number (percentage). † Encompassing caseating and noncaseating granulomas. ‡ Two cases were Aspergillus species that were diagnosed on FNA and CNB materials, and the other 2 were Cryptococcus species that were diagnosed only on FNA smears.


Am J Clin Pathol 2006;125:438-444 441 DOI: 10.1309/FK9XMJ19T036LJ70

441 441


A

B

❚Image 1❚ A, Fine-needle aspiration smears showing poorly defined aggregates of histiocytes and inflammatory cells (Papanicolaou, ×200). B, Concurrent core needle biopsy demonstrating a well-formed granuloma (H&E, ×200).

accuracy rates at 80% to 95%.11,12,16-18,20,21,24-28 However, controversy exists as to the overall value and limitations of FNA and CNB in the evaluation of intrathoracic lesions. Direct comparative data on the use of these 2 techniques performed in the same lesions also are limited. Our results demonstrate that for malignant tumors, the diagnostic accuracy of FNA (85.1%) generally was comparable to that of CNB (86.7%). For histologic typing, FNA and CNB were comparable in the diagnosis of epithelial malignant neoplasms (86.4% vs 85.2%), whereas for nonepithelial malignant neoplasms, FNA yielded a lower diagnostic rate than CNB. This could result, in part, from difficulty in aspirating mesenchymal neoplasms because of their high intercellular adherence and the fact that CNB specimens facilitate the assessment of histologic architecture and the performance of ancillary studies, which often are essential for classifying mesenchymal tumors and lymphomas.18,21,29,30 We were successful in diagnosing non-Hodgkin lymphoma from FNA specimens with the help of flow cytometric immunophenotyping, cytogenetics, and molecular studies.31,32 However, cytologic diagnosis may be challenging when flow cytometric results are inconclusive because of low cellularity or poor cellular preservation. For these cases and some cases of Hodgkin lymphoma in which cytologic features are equivocal, CNB can provide additional diagnostic clues derived from the histologic architecture and immunostaining results. Our findings are in accordance with those from other series in which CNB was found to improve diagnostic accuracy for mediastinal masses, such as lymphoma and thymoma.21,33,34 Bocking et al21 recommended that mediastinal, pleural lesions and presumed mesenchymal tumors 442 442


be sampled by CNB, in contrast with pulmonary lesions, which can be sampled reliably by FNA. FNA and CNB were complementary in the diagnosis of malignant tumors.12,17,18 Despite similar diagnostic yields of FNA and CNB in our study, a substantial number of malignant tumors were diagnosed by only one of the modalities, whereas the combining of FNA and CNB substantially increased the diagnostic accuracy. These results indicate that the combination of FNA and CNB can obviate, in most cases of malignancy, the need for open biopsy, mediastinoscopy, and thoracoscopy, all of which need to be performed under general anesthesia, with associated high morbidity and cost. Considering the simplicity and cost-effectiveness of FNA and the fact that most thoracic malignant neoplasms are epithelial, FNA should be the initial procedure of choice for evaluating intrathoracic lesions, especially for staging disease in patients known to have a history of malignancy. Furthermore, although tissue sampled by CNB allows better classification of lymphomas and mesenchymal tumors, an FNA sample is ideal for flow cytometric phenotyping and cytogenetic analysis. These ancillary tests often are important integral parts in the diagnosis of these tumors. Finally, FNA material might be preferable to that obtained by CNB for microorganism culture. Technical difficulty in obtaining representative material (eg, small lesions, subpleural locations, lesions close to large vessels or right behind ribs, or lesions with poor FNA yield) accounted for the majority of false-negative results and indeterminate diagnoses. Poor cytomorphologic preservation was another contributory factor. Immediate assessment by an onsite cytopathologist improves the overall diagnostic yield by © American Society for Clinical Pathology


providing information about whether the FNA or core sample contains material representative of the target lesion and by allowing triage of cases for appropriate workup.26,27 In the present study, the 2 sampling techniques yielded similar frequencies of false-negative results for malignant tumors and similar rates of indeterminate diagnosis. Although FNA can diagnose malignant lesions with high accuracy, its ability to establish the diagnosis of a benign lesion is less efficacious. Only 1 study has reported that FNA and CNB yielded similar diagnostic rates for benign-specific lesions.12 Other studies showed that specific diagnoses of benign conditions made by an initial cytologic analysis varied from 12% to 50%15,18,24,35-37 and that CNB greatly improved the ability to provide a specific diagnosis for benign lesions.15,17,18,35 In the present study, we showed similar findings, and the substantial difference occurred in identifying granuloma. The difference between the 2 procedures lies in the different capabilities in analyzing benign lesions. Unlike the diagnosis of malignant tumors, for which individual cell morphologic features provide substantial diagnostic information, the diagnosis of benign-specific lesions often requires knowledge of their histologic architecture, which frequently is lost or altered in FNA specimens. It is noteworthy that only positive findings (of malignant or benign-specific diseases) provide valuable information and might preclude the need for more invasive diagnostic procedures. Nonspecific negative results do not equate with benign pathology unless they are supported fully by negative clinical, radiologic, and other findings (eg, negative for microorganism).24,36,38 Therefore, each case should be considered in a multidisciplinary manner to avoid false-negative results. Clinically and radiologically benign lesions with benign cytologic and/or core biopsy findings might require follow-up only without surgical intervention, whereas a lesion with benign findings on FNA or CNB but clinical and/or radiologic evidence suggesting malignancy warrants rebiopsy or surgical excision. In the present study, all benign-nonspecific findings were classified as such based on benign pathologic findings and benign clinical and radiologic courses. The lesions that initially were classified as negative (n = 31) or atypical or suspicious (n = 22) but that had clinical and radiologic features suggesting malignancy were pursued via additional biopsy or surgical resection; all the lesions eventually proved to be malignant (Table 2). Regenerative, reactive, degenerative epithelial cells and histiocytes can give rise to cellular atypia to a degree indistinguishable from that of cancer cells.21,24,25 The presence of a necrotic background in caseating granulomatous lesions also might lead to misdiagnosis. In a study by Khouri et al,24 5 (3.6%) of 137 benign lesions initially were classified as malignant tumors by FNA. No false-positive diagnoses were observed in our study, although we encountered cases with © American Society for Clinical Pathology

reactive atypia. In our experience, the reactive atypical cells usually are scant, with a morphologic spectrum ranging from benign to atypical, and often are associated with background inflammation. FNA is comparable to CNB in the diagnosis of malignant lesions, whereas CNB has better diagnostic accuracy for nonepithelial malignant neoplasms and benign-specific lesions. Combining FNA and CNB greatly improves diagnostic accuracy for malignant lesions. Histologic confirmation should be obtained when clinical or radiologic findings do not correlate with the cytologic findings or nonepithelial lesions and/or benign lesions are considered likely so that the frequency of inadequate material and false-negative and indeterminate results can be minimized. From the Departments of 1Pathology and 2Diagnostic Radiology, The University of Texas M.D. Anderson Cancer Center, Houston. Presented at the 94th Annual Meeting of the United States and Canadian Academy of Pathology; February 26-March 4, 2005; San Antonio, TX. Address reprint requests to Dr Gong: Dept of Pathology, Unit 53, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030.

References 1. Hajdu SI, Melamed MR. Limitations of aspiration cytology in the diagnosis of primary neoplasms. Acta Cytol. 1984;28:337345. 2. Cagle PT, Kovach M, Ramzy I. Causes of false results in transthoracic fine needle lung aspirates. Acta Cytol. 1993;37:16-20. 3. Niden AH, Salem F. A safe high-yield technique for cutting needle biopsy of the lung in patients with diffuse lung disease. Chest. 1997;111:1615-1621. 4. Haaga JR, LiPuma JP, Bryan PJ, et al. Clinical comparison of small- and large-caliber cutting needles for biopsy. Radiology. 1983;146:665-667. 5. Clore F, Virapongse C, Saterfiel J. Low-risk large-needle biopsy of chest lesions. Chest. 1989;96:538-541. 6. Gleeson F, Lomas DJ, Flower CD, et al. Powered cutting needle biopsy of the pleura and chest wall. Clin Radiol. 1990;41:199-200. 7. Harrison BD, Thorpe RS, Kitchener PG, et al. Percutaneous Trucut lung biopsy in the diagnosis of localised pulmonary lesions. Thorax. 1984;39:493-499. 8. Meyer JE, Ferrucci JT Jr, Janower ML. Fetal complications of percutaneous lung biopsy: review of the literature and report of a case. Radiology. 1970;96:47-48. 9. Norenberg R, Claxton CP Jr, Takaro T. Percutaneous needle biopsy of the lung: report of two fatal complications. Chest. 1974;66:216-218. 10. Knight RK. How to biopsy the lung and pleura. Br J Hosp Med. 1981;26:160-161, 165. 11. Arakawa H, Nakajima Y, Kurihara Y, et al. CT-guided transthoracic needle biopsy: a comparison between automated biopsy gun and fine needle aspiration. Clin Radiol. 1996;51:503-506.

Am J Clin Pathol 2006;125:438-444 443 DOI: 10.1309/FK9XMJ19T036LJ70

443 443


12. Kim HK, Shin BK, Cho SJ, et al. Transthoracic fine needle aspiration and core biopsy of pulmonary lesions: a study of 296 patients. Acta Cytol. 2002;46:1061-1068. 13. Yazdi HM, MacDonald LL, Hickey NM. Thoracic fine needle aspiration biopsy versus fine needle cutting biopsy: a comparative study of 40 patients. Acta Cytol. 1988;32:635640. 14. Skamel HJ, Hanusch A, Mathias K. CT-guided punctures: fine-needle biopsy vs automated co-axial cutting biopsy [in German]. Aktuelle Radiol. 1998;8:273-277. 15. Greif J, Marmor S, Schwarz Y, et al. Percutaneous core needle biopsy vs fine needle aspiration in diagnosing benign lung lesions. Acta Cytol. 1999;43:756-760. 16. Greif J, Marmur S, Schwarz Y, et al. Percutaneous core cutting needle biopsy compared with fine-needle aspiration in the diagnosis of peripheral lung malignant lesions: results in 156 patients. Cancer. 1998;84:144-147. 17. Boiselle PM, Shepard JA, Mark EJ, et al. Routine addition of an automated biopsy device to fine-needle aspiration of the lung: a prospective assessment [published correction appears in AJR Am J Roentgenol. 1997;169:1755]. AJR Am J Roentgenol. 1997;169:661-666. 18. Moulton JS, Moore PT. Coaxial percutaneous biopsy technique with automated biopsy devices: value in improving accuracy and negative predictive value. Radiology. 1993;186:515-522. 19. Zardawi IM. Fine needle aspiration cytology vs core biopsy in a rural setting. Acta Cytol. 1998;42:883-887. 20. Laurent F, Latrabe V, Vergier B, et al. Percutaneous CT-guided biopsy of the lung: comparison between aspiration and automated cutting needles using a coaxial technique. Cardiovasc Intervent Radiol. 2000;23:266-272. 21. Bocking A, Klose KC, Kyll HJ, et al. Cytologic versus histologic evaluation of needle biopsy of the lung, hilum and mediastinum: sensitivity, specificity and typing accuracy. Acta Cytol. 1995;39:463-471. 22. Spira A, Ettinger DS. Multidisciplinary management of lung cancer. N Engl J Med. 2004;50:379-392. 23. Layfield LJ, Coogan A, Johnston WW, et al. Transthoracic fine needle aspiration biopsy: sensitivity in relation to guidance technique and lesion size and location. Acta Cytol. 1996;40:687-690. 24. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle aspiration biopsy of benign and malignant lung lesions. AJR Am J Roentgenol. 1985;144:281-288. 25. Thornbury JR, Burke DP, Naylor B. Transthoracic needle aspiration biopsy: accuracy of cytologic typing of malignant neoplasms. AJR Am J Roentgenol. 1981;136:719-724.

444 444


26. Conces DJ Jr, Schwenk GR Jr, Doering PR, et al. Thoracic needle biopsy: improved results utilizing a team approach. Chest. 1987;91:813-816. 27. Austin JH, Cohen MB. Value of having a cytopathologist present during percutaneous fine-needle aspiration biopsy of lung: report of 55 cancer patients and metaanalysis of the literature. AJR Am J Roentgenol. 1993;160:175-177. 28. Johnston WW. Percutaneous fine needle aspiration biopsy of the lung: a study of 1,015 patients. Acta Cytol. 1984;28:218224. 29. Ben-Yehuda D, Polliack A, Okon E, et al. Image-guided coreneedle biopsy in malignant lymphoma: experience with 100 patients that suggests the technique is reliable. J Clin Oncol. 1996;14:2431-2434. 30. Pappa VI, Hussain HK, Reznek RH, et al. Role of imageguided core-needle biopsy in the management of patients with lymphoma. J Clin Oncol. 1996;14:2427-2430. 31. Sneige N, Dekmezian RH, Katz RL, et al. Morphologic and immunocytochemical evaluation of 220 fine needle aspirates of malignant lymphoma and lymphoid hyperplasia. Acta Cytol. 1990;34:311-322. 32. Gong Y, Caraway N, Gu J, et al. Evaluation of interphase fluorescence in situ hybridization for the t(14;18)(q32;q21) translocation in the diagnosis of follicular lymphoma on fineneedle aspirates: a comparison with flow cytometry immunophenotyping. Cancer. 2003;99:385-393. 33. Morrissey B, Adams H, Gibbs AR, et al. Percutaneous needle biopsy of the mediastinum: review of 94 procedures. Thorax. 1993;48:632-637. 34. Heilo A. Tumors in the mediastinum: US-guided histologic core-needle biopsy. Radiology. 1993;189:143-146. 35. Klein JS, Salomon G, Stewart EA. Transthoracic needle biopsy with a coaxially placed 20-gauge automated cutting needle: results in 122 patients. Radiology. 1996;198:715-720. 36. Calhoun P, Feldman PS, Armstrong P, et al. The clinical outcome of needle aspirations of the lung when cancer is not diagnosed. Ann Thorac Surg. 1986;41:592-596. 37. Greene R, Szyfelbein WM, Isler RJ, et al. Supplementary tissue-core histology from fine-needle transthoracic aspiration biopsy. AJR Am J Roentgenol. 1985;144:787-792. 38. Savage C, Walser EM, Schnadig V, et al. Transthoracic imageguided biopsy of lung nodules: when is benign really benign? J Vasc Interv Radiol. 2004;15:161-164.
