Intraoperative Fine Needle Aspiration Biopsy of Thoracic Lesions

Transcription

1 Intraoperative Fine Needle Aspiration Biopsy of Thoracic Lesions Walter J. McCarthy, M.D., Miriam L. Christ, M.D., and Willard A. Fry, M.D. ABSTRACT Forty-one intraoperative fine needle aspiration biopsies were performed on 35 patients during exploratory thoracotomy (33 patients) or mediastinoscopy (2 patients). Each biopsy was done with a 22 gauge needle. Smears were prepared at the operating table, air-dried, sent directly to the laboratory, stained, and interpreted immediately by the pathologist. Preparation and reporting time averaged ten minutes. Surgical decisions were made on the basis of the pathologist s reports. Intraoperative fine needle aspiration biopsy was 100% accurate in differentiating inflammatory from neoplastic lesions. Ninety-five percent diagnostic accuracy for malignancy (39 out of 41 specimens) was obtained. It permitted quick biopsy of lesions deep within the lung parenchyma without the need to cut across uninvolved tissue, thus permitting appropriate resection in each patient. There were no deaths related to the procedure. At exploratory thoracotomy for suspected lung neoplasm, the surgeon is faced first with the need for an accurate diagnosis. However vigorous the preoperative evaluation, some patients come to operation with an undiagnosed mass or with disease more advanced than anticipated. Surgical judgment varies on the definition of a resectable lesion and the extent of resection indicated, but every surgeon desires exact determination of the diagnosis and a confirmation of the presence or absence of central lymph node involvement at the time of thoracotomy. Over the past three years, besides using the standard intraoperative frozen-section techniques, we have used fine needle aspiration biopsy with rapid cytological analysis to help in From the Departments of Surgery and Pathology, Evanston Hospital and Northwestern University Medical School, Evanston, IL. Presented at the Sixteenth Annual Meeting of The Society of Thoracic Surgeons, Jan 21-23, 1980, Atlanta, GA. Address reprint requests to Dr. Fry, 2500 Ridge Ave, Evanston, IL decision-making during thoracotomy. The purpose of this presentation is to report our experience with this intraoperative technique in 35 patients and to discuss its merits and limitations. Material and Methods Fine needle aspiration biopsies were taken at thoracotomy from 33 patients and at mediastinoscopy from 2 patients. The biopsy technique was used under three general circumstances. First, 28 lung lesions and 1 mediastinal tumor not diagnosed preoperatively were aspirated as soon as the chest was opened in order to establish or clarify a primary diagnosis. Second, suspicious lymph nodes in the hilum or mediastinum were biopsied ten times in patients with a diagnosis of lung cancer, in order to determine the extent of disease and to define resectability. Third, in 2 patients with lung cancer undergoing staging mediastinoscopy, firm, fixed masses, which we hesitated to biopsy with standard cutting forceps, were biopsied by fine needle aspiration technique to minimize any chance of hemorrhage. Conventional frozen sections were done when convenient or appropriate. Fine needle aspiration biopsy of lung masses was used preferentially for lesions deep within the lung parenchyma that we considered technically difficult to biopsy with ease and safety by standard wedge resection techniques. Our operative sampling technique and smear preparation are performed by the surgeon (Fig 1). A 22 gauge needle is attached to a 20-ml disposable plastic syringe, which is then fitted into a syringe pistol.* Those items as well as the glass slides to be used are all sterile. If a syringe pistol is not available, manual retraction works well, using either a 10 ml or a 20 ml syringe. The *Cameco syringe pistol, Precision Dynamics Corp, 3031 Thornton Ave, Burbank, CA by The Society of Thoracic Surgeons

2 25 McCarthy, Christ, and Fry: Fine Needle Aspiration Biopsy Fig 1. Performing the aspiration. The syringe pistol enables the surgeon to fix the mass with one hand while exerting a vacuum to the syringe with the other. needle is guided into the lesion, and the piston is retracted to create a vacuum in the syringe. Several short, jabbing motions are made to sample different parts of the lesion. The piston is then relaxed and the needle withdrawn. Most of the specimen is contained within the needle at this time. The needle is then disconnected from the syringe. The syringe is filled with air, reconnected with the needle, and the needle's contents are blown onto a glass slide. Another slide is pressed to or drawn over the first to make a thin layer of cells. These slides are allowed to air-dry, although sometimes the pathologist will request additional alcohol-fixed or formalin-fixed smears to be prepared as well. A quick hematoxylin and eosin stain as used for traditional frozen section and a rapid, modified Wright-Giemsa stain, are made. Those stains can be ready for viewing in several minutes and, in the interest of speed, are used in preference to a Papanicolaou stain. While the hematoxylin and eosin stain gives the best cellular detail, the Wright-Giemsa stain provides metachromatic staining of stroma and mucin not well seen with the former. Cytological interpretation can be reported back to the surgeon within about ten minutes of the actual biopsy. *"Diff-Quik," Harleco, 480 Democrat Rd, Gibbstown, NJ Results Intraoperative fine needle aspiration was performed forty-one times in 35 patients: 39 lesions in 33 patients at thoracotomy and 2 lesions in 2 patients at mediastinoscopy (Table 1). Neoplastic cells were identified in each instance of tumor by needle aspiration. Of 30 malignant lesions, 23 had available histology. In 22 of those 23 malignant lesions (96%), the fine needle aspiration biopsy agreed with the final histological diagnosis regarding the presence of malignancy (Table 2), and in 16 of those 23 Table 1. Clinical Material for lntraoperative Fine Needle Aspiration Biopsy No. of No. of Patients Lesions Malignant Benign Thoracotomy Lung lesion Hilar or mediastinal lymph node Mediastinal tumor Mediastinoscopy Total 'Six patients had aspiration of 2 different lesions. Table 2. Results of Fine Needle Aspiration Biopsy in 30 Malignant Lesions in 24 Patientsa Operation Mediastinoscopy Thoracotomy Lung lesion Hilar or mediastinal node Result /Sb "Accuracy was 97%. bdiagnostic error in 1 patient: breast metastasis misread as carcinoid.

3 26 The Annals of Thoracic Surgery Vol30 No 1 July 1980 Table 3. Correlation of Cytology of lntraoperative Fine Needle Aspiration in 23 Malignant Lesions with Available Histology Final Histological Diagnosis LUNG LESIONS Squamous cell carcinoma-2 Adenocarcinoma-7 Large cell undifferentiated carcinoma-1 Oat cell carcinoma-2 Mucoepidermoid carcinoma-1 Metastatic endometrial carcinoma-2 Metastic breast carcinoma-la LYMPH NODE LESIONS Adenocarcinoma-5 Squamous cell carcinoma-1 Oat cell carcinoma-1 "Diagnostic error in 1 patient. I Cytological Diagnosis Squamous cell carcinoma-2 Adenocarcinoma-4 Carcinoma-3 Carcinoma-1 I Oat cell carcinoma-2 Atypical cells-1 Metastatic endometrial carcinoma-2 Carcinoid-la Adenocarcinoma4 Carcinoma-1 Squamous cell carcinoma-1 Oat cell carcinoma-1 (70%) the cytology report from the aspiration biopsy provided the correct neoplastic cell type as well as a diagnosis of malignancy (Table 3). Histology was not obtained from 7 lesions in patients with unresectable carcinoma because the definitive cancer diagnosis had been established by other means and diagnosis of the specific lesion by intraoperative fine needle aspiration was considered definitive. There was no false positive or false negative report for tumor as such. However, there was one false negative report for carcinoma. Two lesions misdiagnosed by fine needle aspiration were both problems of tumor classification: a metastatic breast carcinoma misdiagnosed as a carcinoid and a carcinoid misdiagnosed as lymphoma. Lobectomy was performed in each of those 2 patients. In no instance was a resection performed that, in light of the final histological diagnosis, was not indicated. With some granulomas, only necrotic debris was reported by intraoperative fine needle aspiration, but the lack of malignant cells was helpful in deciding that a limited resection should be performed (Table 4). The overall diagnostic accuracy in identifying malignancy in our series was 95%. The technique was essentially complication free. In only Table 4. Results of lntraoperative Fine Needle Aspiration in 11 Benign Lesions Final Surgical Diagnosis Granuloma-5 (Histoplasmoma in 2) Hamartoma-3 Abscess-1 Carcinoid tumor-la Dermoid cyst (mediastinumtl adiagnostic error in 1 patient. Cytological Diagnosis at Operation No tumor ceb-2 Necrosis and inflammation-1 Amorphous debris1 Granuloma-1 No malignant cells-2 I Possible hamartoma-1 Necrosis and inflammation-1 Lymphoma-la Squamous cells and hair shaft-1 1 patient, a hematoma developed at the site of aspiration biopsy of a carcinoid tumor and was so noted in the resected lobectomy specimen. The following two reports demonstrate the usefulness of intraoperative fine needle aspiration. Patient 1 A 56-year-old female nurse was found on periodic routine health evaluation to have a new lung lesion on a chest roentgenogram (Fig 2). It was located in the anterior segment of the right upper lobe. Laminagraphy showed no signs of benignancy, and previous roentgenograms were available for comparison. She was a cigarette smoker and had a negative tuberculin test. Physical examination and standard laboratory workup were unremarkable. Sputum cytology findings were negative. It was thought that exploratory thoracotomy was indicated. At operation a firm mass of 2 cm was palpated deep within the substance of the upper lobe. Wedge resection would have been impossible, and a segmental resection for diagnosis would have given a very narrow margin of resection at the anterior segmental bronchus, too close for tumor and sacrificing too much lung for benign disease. Intraoperative fine needle aspiration was performed. The report was "necrotic debris-no tumor seen" (Fig 3A). On the basis of that report, a local excision of the mass was performed, and barely any lung tissue was resected at the margins of the lesion. The histological diagnosis was a caseating His-

4 27 McCarthy, Christ, and Fry: Fine Needle Aspiration Biopsy A B Fig 2. (Patient 1. ) ( A )Chest roentgenogram demonstrutes a n e w lesion in the right upper lung field. ( B ) Laminagraphy of the lesion shows that it is in the anterior segment of the right upper lobe and contains no calcification. toplasma granuloma (Fig 3B). The patient made an uneventful recovery. COMMENT. Preoperative transthoracic needle biopsy was not performed because of the size and location of the lesion and because we thought that a 10% false negative rate for tumor diagnosis was not acceptable for a patient in such good general condition. Prior to our use of intraoperative fine needle aspiration, this patient would have undergone lobectomy with the sacrifice of a considerable amount of lung tissue. Fig 3. (Patient 1.) (A)Cytological study of intraoperative fine needle aspiration shows only necrotic debris. T h e failure to demonstrate tumor cells i n suck material has, in our experience, meant that there is no tumor present. ( B ) Photomicrograph of the resected specimen shows a typical granuloma. A f e w Histoplasma organisms were identified on special stains. Cultures for f u n g i gave no growth. (A: Diff-Quik; B: Ht3.E.) Patient 2 A 58-year-old retired male police officer had an episode of hemoptysis. He was a smoker. Chest A B

5 28 The Annals of Thoracic Surgery Vol30 No 1 July 1980 A Fig 4. (Patient 2.) Sagittal section of the pneumonect o m y specimen. T h e small arrow indicates the primary tumor. T h e large arrow indicates the metastatic tumor adherent to the pulmonary artery in the fissure. This tumor was biopsied b y fine needle aspiration. roentgenogram demonstrated a 3 cm mass in the left lower lobe. Flexible fiberoptic bronchoscopy with fluoroscopic control enabled biopsy of the mass, which was diagnosed as adenocarcinoma. Physical examination, routine blood and urine studies, and a workup for a primary tumor site other than the lung were all unremarkable. At operation the lower lobe mass was obvious and not fixed. However, within the major fissure there was a firm mass adherent to the pulmonary artery, which, if tumor, would preclude a lobectomy (Fig 4). Fine needle aspiration of the mass was performed, and the aspiration cytology diagnosis was adenocarcinoma (Fig 5A). A left pneumonectomy was performed. The final histological diagnosis on the pneumonectomy specimen was well differentiated peripheral adenocarcinoma of the lung with metastasis to lymph nodes in the fissure B Fig 5. (Patient 2. ) (A)Cytological preparation of the aspirated material from the l y m p h node in the fissure demonstrates adenocarcinoma. ( B ) Histological preparation from the same area of the pneumonectomy specimen shows h o w well the cytological and histological techniques correlate. (A: Diff-Quik; B: H D E. ) (Fig 5B). The patient s postoperative recovery was satisfactory. It is our practice to document by COMMENT. surgical pathology the specific indication for pneumonectomy. In this instance, although a small scalpel biopsy could have been taken and submitted for frozen tissue section, the lesion was adherent to the pulmonary artery in the fissure, and our experience with needle aspiration is that it is just as reliable but quicker and perhaps safer in such a situation. Comment Martin and Ellis [9] at Memorial Hospital in New York pioneered the use of needle aspira-

6 29 McCarthy, Christ, and Fry: Fine Needle Aspiration Biopsy tion biopsy in the evaluation of various neoplasms fifty years ago. The technique did not gain wide acceptance in the United States, however, until recent years, in spite of many reports from abroad [l, 111 documenting its safety, diagnostic accuracy, and technical ease. Recent reports in the literature indicate an increasing experience with fine needle aspiration biopsy in this country. Kline and Neal [71 reported an experience of 3,267 cases and a diagnostic accuracy rate of 90%. Other large series include those of Hajdu and Melamed [41 and Frable [3] who reported accuracy rates consistently over 80%. The application of fine needle aspiration to lung lesions has received considerable recent attention and finally appears to have gained acceptance in the United States. Lalli and coworkers [8] reported 1,223 transthoracic needle aspiration biopsies with a diagnostic yield of 84%, and Sagel and colleagues [lo] reported a positive diagnostic rate of 87% in 1,153 patients. Our own experience with transthoracic needle biopsy, spanning nine years and 181 patients, gave an accuracy rate of 84%. The use of fine needle aspiration biopsy for rapid diagnosis in an operating room setting, in place of conventional frozen section, was described by Kline and colleagues [6] in a series of 28 pancreatic lesions diagnosed by this method at the time of celiotomy. They achieved a high degree of accuracy and encountered no complication. The practice of rapid staining and immediate reporting of cytological findings of needle aspirations of breast masses was reported by Duguid and co-workers [2] with favorable results. Those reports together with the known diagnostic accuracy of transthoracic needle biopsy led us to apply fine needle aspiration biopsy to thoracic lesions at the time of operation. The results of our study indicate that intraoperative fine needle aspiration is a very valuable tool at thoracotomy for obtaining an initial diagnosis in exploratory cases and for determining whether or not lymph nodes are involved with tumor. It is particularly useful in diagnosing lesions deep within the lung parenchyma where its use can lead to conservative resection of benign lesions and confirm the appropriateness of resection for malignancy. It also appears to be highly accurate and safe at mediastinoscopy for biopsy of lesions in close proximity to major blood vessels. Ninety-six percent of the malignant lesions were diagnosed as such by this method, and 70% were even correctly diagnosed down to specific cell type. The technique requires a cytopathologist experienced in fine needle aspiration biopsy [51, and that may be a limiting factor in some clinical settings. In competent hands, a cytological diagnosis based on fine needle aspiration is as definitive as a histological diagnosis based on frozen section. In summary, needle aspiration biopsy with immediate interpretation is a valuable addition to the operative diagnostic armamentarium of the surgeon. References 1. Dahlgren S, Nordenstrom B: Transthoracic Needle Biopsy. Chicago, Year Book, Duguid HLD, Wood RAB, Irving AD, et al: Needle aspiration of the breast with immediate reporting of material. Br Med J 2:185, Frable WJ: Thin-needle aspiration biopsy. Am J Clin Pathol 65:168, Hajdu S, Melamed MR: The diagnostic value of aspiration smears. Am J Clin Pathol59:350, Johnston WW, Frable WJ: Diagnostic Respiratory Cytopathology. Paris, Masson, Kline TS, Abramson J, Goldstein F, et al: Needle aspiration biopsy of the pancreas at laparotomy. Am J Gastroenterol 68:30, Kline TS, Neal HS: Needle aspiration biopsy: a critical appraisal eight years and 3,267 specimens later. JAMA 239:36, Lalli AF, McCormack LJ, Zelch M, et al: Aspiration biopsies of chest lesions. Radiology 127:35, Martin HE, Ellis EB: Biopsy by needle puncture and aspiration. Ann Surg 92:169, Sagel SS, Ferguson TB, Forrest JV, et al: Percutaneous transthoracic aspiration needle biopsy. Ann Thorac Surg 26:399, Zajicek J: Aspiration Biopsy Cytology: Part I. Cytology of Supradiaphragmatic Organs. New York, Karger, 1974