- Email: [email protected]
Thoracoscopy for the diagnosis of the indeterminate solitary pulmonary nodule
Thoracoscopy for the Diagnosis of the Indeterminate Solitary Pulmonary Nodule Michael J. Mack, MD, Stephen R. Hazelrigg, MD, Rodney J. Landreneau, MD, and Tea E. Acuff, MD Section of Thoracic Surgery, Medical City Dallas, Dallas, Texas, St Luke’s Medical Center, Milwaukee, Wisconsin, and University of Pittsburgh, Pittsburgh, Pennsylvania
Traditional nonoperative diagnostic approaches to the indeterminate solitary pulmonary nodule include bronchoscopy and percutaneous needle biopsy. Although both methods are minimally invasive, the diagnosis of the small, peripheral nodule may remain elusive. Open thoracotomy is often required when these methods fail to obtain a diagnosis. Between January 1991 and June 1992, 242 patients with indeterminate solitary lung nodules underwent thoracoscopic excisional biopsy as the primary diagnostic method. Wedge excisions of the nodules were all performed by thoracoscopic techniques using an endoscopic stapler alone (72%), ne0dymium:yttriumaluminum garnet laser (18%), or both (10%). A definite diagnosis was obtained in all patients. Two patients required conversion to thoracotomy to locate the nodule (both malignant). A benign diagnosis was obtained in 127 patients (52%) and a malignant diagnosis in 115 (48%).Of the malignant nodules, 51 (44%)were primary lung cancer and 64 (56%)were metastases. All patients
diagnosed with primary lung cancer having adequate pulmonary reserve (n = 29) underwent formal open lung resection during the same procedure. There was no mortality, and significant morbidity was limited to atelectasis in 3 patients (1.2%), pneumonia in 2 patients (0.8%), and prolonged air leak more than 7 days in 4 patients (1.6%). Average hospital stay for patients having thoracoscopy only (n = 213) was 2.4 days (range, 1 to 12 days). Thoracoscopy offers a minimally invasive approach for the diagnosis of the indeterminate solitary nodule. It has advantages over traditional diagnostic methods of being virtually 100% sensitive and 100% specific with no mortality and minimal morbidity. Based on these results, as well as the benefits of cost and time effectiveness, we anticipate a larger role for thoracoscopy in the diagnostic management of the indeterminate solitary pulmonary nodule.
T
thoracoscopy as the primary diagnostic modality for indeterminate solitary pulmonary nodules.
he definitive diagnosis of the indeterminate solitary pulmonary nodule often remains elusive, despite the multiplicity of diagnostic tools available. Careful examination of the chest roentgenogram or computed tomogram may show signs of benignancy including lack of growth and particular calcification patterns [14].Bronchoscopy is helpful for large central lesions but has a low diagnostic yield, approximately 10% in the small, peripheral nodule [3, 5, 61. Transthoracic needle aspiration biopsy (TNAB) has reported diagnostic accuracy ranging from 43% to 97% [7-101 for malignant lesions but is less effective in yielding a definitive diagnosis in benign nodules. Thoracotomy is accurate for diagnosis but has significant morbidity [ll, 121. Recent wider application of thoracoscopic techniques has included pulmonary resection. By the use of endoscopic stapling devices and laser lung resection techniques, the majority of peripheral pulmonary nodules are now able to be removed by thoracoscopic resective techniques [13, 141. We report our experience with use of Presented at the Twenty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 2527, 1993. Address reprint requests to Dr Mack, 7777 Forest Ln, Suite C-202, Dallas, TX 75230.
0 1993 by The Society of Thoracic Surgeons
(Ann Thorac Surg 1993;56:825-32)
Material and Methods Between January 1991 and June 1992, 242 patients presented to one of three collaborating institutions with an indeterminate solitary pulmonary nodule. The participating institutions are Medical City Dallas, Dallas, Texas; University of Pittsburgh, Pittsburgh, Pennsylvania; and St Luke’s Medical Center, Milwaukee, Wisconsin. The definition of an indeterminate solitary pulmonary nodule for this study is a single nodule, less than 3 cm in largest diameter, containing no calcium and located in the outer one third of the lung parenchyma. There were 138 men (57%)and 104 women (43%)in the study population. The mean age was 69 years (range, 21 to 87 years). Nineteen patients (7%) were 80 years old or older. Most patients had no previous diagnostic procedure, although 9 patients referred from other institutions had undergone a nondiagnostic TNAB. All thoracoscopic resections for diagnosis are performed under general endotracheal anesthesia with a double-lumen endotracheal tube to obtain ipsilateral lung collapse. An arterial line is placed for hemodynamic monitoring. All patients are placed in the lateral position 0003-4975/93/$6.00
826
MACKETAL THORACOSCOPIC LUNG RESECTION
Fig 1 . Subtle pleural changes adjacent to nodule can be located at thoracoscopy.
and prepared fully as for a standard thoracotomy. A computed tomogram showing the location of the target nodule is always present in the operating room, and trocar placement is planned based on this study. Usually, the initial trocar is placed in the sixth or seventh intercostal space in the midaxillary line. However, this placement is changed to enhance visualization of the nodule by the thoracoscope. For example, for nodules in the anterior segment of the upper lobe, the initial trocar is placed in the fifth intercostal space in the anterior axillary line. Second and third trocars, for introduction of instruments, are placed in a triangular arrangement as far as possible from the initial trocar site to allow maximum maneuverability, ease of stapler insertion, and optimal retraction of adjacent lung. Initial exploratory thoracoscopy is performed to identify the lung nodule as well as to examine for associated pleural metastases. An initial assessment of the hilum and mediastinum is also performed at this time. If the nodule is not apparent on initial visual examination of the lung, the computed tomogram is again carefully assessed and landmarks identified to help locate the lesion including adjacent fissures and any pleural changes adjacent to subpleural lesions (Fig 1). Next, instrument palpation is performed, and a change in consistency of the lung containing the nodule can usually be appreciated by this method. If the nodule still cannot be located, then a digit is placed through the nearest trocar site and used to palpate the nodule. This technique is possible because carbon dioxide insufflation is usually not necessary for thoracoscopic procedures. If the nodule is not pleural based or immediately subpleural, or is less than 1 cm in size, then preoperative needle localization is used (Fig 2). When the nodule for resection has been identified, assessment is made regarding the best method of resection. If the nodule is near the pleural surface and adjacent to an acute margin of the lung, then the resection can be most expeditiously accomplished using an endoscopic stapler (EndoGIA-30, Autosuture; United States Surgical Corp, Norwalk, CT). If the nodule is located adjacent to a flat surface of the lung or is deep within the lung paren-
Ann Thorac Surg 1993;56:825-32
chyma so that the jaws of the stapler cannot be easily placed on the lung, then a neodymium:yttrium-aluminum garnet laser is used for resection alone or in conjunction with the laser. Once the nodule is successfully resected, an endoscopic bag is inserted through a trocar site, and the specimen is placed inside for removal from the chest cavity. This is done to prevent tumor seeding of the tract if the lesion is malignant. The specimen is sent for immediate frozen pathologic examination. A margin of at least 1 cm around the lesion is obtained. If the nodule is determined to be a primary lung malignancy and the patient has adequate pulmonary reserve, an immediate formal lobectomy is performed using either video-assisted or standard muscle-sparing techniques. If the lesion is benign or metastatic, or if the patient has primary lung cancer and poor pulmonary reserve, then the procedure is ended. The area of resected lung is carefully examined for pneumostasis by partially inflating the lung, and a small pleural drainage tube (20 to 24F) is placed. Trocar sites are examined for hemostasis, and incisions are closed with a subcuticular suture. Care in an intensive care unit is seldom necessary. Pleural drainage tubes are removed when no air leak is present and fluid drainage is less than 150 mL in a 24-hour period. Parenteral narcotics are administered by patient-controlled analgesia infusion and are seldom necessary after the chest tube is removed. Hospital discharge is usually within 12 hours of chest tube removal.
Results A definitive diagnosis was obtained in all patients. A diagnosis of benign disease was obtained in 127 patients (52%) and malignant disease in 115 patients (48%). Two patients (0.8%) required conversion to an open thoracotomy to locate the nodule and obtain a diagnosis, both lesions being malignant. The benign diagnoses are listed in Table 1. Of the 115 patients with a malignant diagnosis, 62 patients (56%) had a pulmonary metastasis from an
Fig 2 . Localizing wire placed preoperatively to help identify a small nodule.
MACKETAL THORACOSCOPIC LUNG RESECTION
Ann Thorac Surg 1993;56825-32
Table 1 . Benign Diagnoses ( n = 127) Diagnosis Interstitial fibrosis/pneumonia Granuloma
Pulmonary infarction/scar Hamartoma Lymphoid aggregates/subpleurallymph nodes
No. of
Patients 35 34 13 12 7
Cytomegalovirus/pneumocystis Sarcoid nodules
Rheumatoid nodules Cryptococcal Sequestration Norcardia
Anthrosilicosis Amyloid
Sclerosing hemangioma Histiolytosis X Chronic lung rejection Pseudoxanthoma Bronchiolitis obliterans Neurofibroma
1 1 1 1 1
1 ’ 1 1 1 1
extrathoracic primary neoplasm (Table 2). Fifty-one patients (44%) had primary lung cancer. Twenty-nine patients with primary lung cancer and adequate pulmonary reserve then underwent a formal open, muscle-sparing or video-assisted lung resection during the same procedure. The remaining 22 patients had poor lung function that precluded a formal procedure, and the thoracoscopic lung resection was the definitive treatment. The method of resection was the endoscopic stapler in 72%, the neodymium:yttrium-aluminum garnet laser in 18%, or combined resection using both stapler and laser in 10%. However, in the last 6 months of the study, 88% of the resections were performed using the stapler alone. There was no operative mortality and minimal morbidity (3.6%). Three patients (1.2%) suffered significant lobar atelectasis that required postoperative bronchoscopy.
Table 2. Malignant Diagnoses (n = 115) Diagnosis
Primary lung cancer Lymphoma Metastatic lesions Colon Liver Breast
Renal Melanoma
Adenocarcinoma, unknown primary Carcinoid Esophageal Bladder
No. of Patients 51 2 62 15 11 10 10 8 5
1 1 1
827
Pneumonia occurred in 2 patients, both of whom responded to antibiotics and respiratory therapy. A prolonged air leak of greater than 7 days occurred in 4 patients (1.6%). Three of these prolonged leaks occurred in patients who underwent laser lung resection and the fourth in a patient with extensive bullous emphysema who had resection of a nodule performed with a stapler. We now use fibrin glue any time we have a residual air leak in an area of resected lung tissue to diminish this problem. Chest tubes were used in all procedures. Average chest tube duration was 1.8 days (range, 0.5 to 11 days). Postoperative length of stay was 2.4 days (range, 1 to 12 days) with most patients being discharged shortly (6 to 12 hours) after chest tube removal.
Comment The solitary pulmonary nodule is a single intrapulmonary spherical lesion that is fairly well circumscribed. There are about 150,000 new nodules detected each year in the United States [4]. Currently, approximately 40% to 50%of solitary nodules are malignant, either bronchogenic carcinoma or metastatic from an extrathoracic site [3, 41. The standard management of the solitary pulmonary nodule is prompt surgical removal unless benignancy can be demonstrated. The probability that a nodule is malignant is related to the age of the patient, diameter of the nodule, amount of cigarette smoke inhalation, overall prevalence of malignancy in solitary pulmonary nodules, the nature of the edge of the lesion, and the presence or absence of occult calcification [4]. Standard definition of benign disease includes demonstration of no growth over 2 years by prior chest roentgenogram or a specific pattern of calcification. Other criteria for benign disease include a transthoracic needle aspiration biopsy demonstrating a specific benign diagnosis and patient age less than 30 to 35 years, unless a prior extrapulmonary malignancy exists. Many noninvasive or minimally invasive diagnostic procedures exist for the evaluation of the indeterminate solitary pulmonary nodule. The least invasive method of evaluation is clinical. If the patient is less than 30 to 35 years old, has no previous history of malignancy, and is a nonsmoker, the risk of malignancy is 1%or less [15]. A “wait and watch’ approach is appropriate in this clinical group. Evaluation of the chest roentgenogram or computed tomogram is also noninvasive and often helpful. If prior roentgenograms demonstrate a stable lesion over a 2-year period or if a particular pattern of calcification (diffuse, central, laminar, or “popcorn”) exists, a benign diagnosis can be affirmed and an “observation only” approach taken [16]. The next considerations are the size and configuration of the nodule. If the nodule is 3 cm or greater in diameter, the likelihood of malignancy is 93% to 99% [17, 181 and after appropriate computed tomographic or surgical staging, or both, surgical excision without further evaluation is appropriate. If the patient has limited pulmonary reserve, a video-assisted thoracoscopic resection is performed. If adequate lung function for a formal lung
828
MACKETAL THORACOSCOPIC LUNG RESECTION
resection exists, an open procedure is instead performed, although a case can be made for performance of an exploratory thoracoscopy first to rule out occult pleural metastasis and, therefore, avoid a needless thoracotomy. In lesions less than 3 cm in diameter, unless occult calcification can be demonstrated, further investigation is indicated. In one study of 431 nodules measuring 2 cm or less, 42% were malignant [17]. In another study of 177 nodules 1 cm or less in diameter, 15% were malignant [MI. As well as size, the edge of the nodule at the lung interface is important. Bronchogenic carcinoma tends to have lobulated or fuzzy margins, whereas benign and metastatic lesions tend to be smooth with sharp borders. If the nodule is less than 3 cm in diameter, noncalcified, and there is no current or past evidence for extrapulmonary malignancy in a patient 30 to 35 years old or older, further evaluation is necessary. Sputum cytology is a very nonproductive examination, having a very low yield for the demonstration of malignancy. Positive sputum cytology has been found in only 15% to 19% of patients with proven malignancy [4]. In a similar manner, sputum cultures seldom demonstrate benign disease with assurance. There appears to be little value of sputum cytology in evaluation of solitary pulmonary nodules. The diagnostic yield of bronchoscopy ranges from 20% to 80% [5, 61. Specific benign diagnoses can be made in about 10% of nodules subsequently proved to be benign [6]. As most benign nodules and solitary metastasis are not “bronchogenic” in origin, it is not surprising that the yield is this low [4]. Because the primary objective of bronchoscopy is to diagnose benign disease and avoid unnecessary surgical resection, these results are disappointing and, in our opinion, fiberoptic bronchoscopy offers little contribution to the management of the solitary peripheral nodule. Transthoracic needle aspiration biopsy is commonly used for evaluation of solitary pulmonary nodules, is simple to perform, and has a relatively low complication rate of pneumothorax of 15%to 30% (8, 19, 201. Transthoracic needle aspiration biopsy in the diagnosis of malignancy has a sensitivity ranging from 64% to 97% [18, 21-24]. There is a false-positive rate of 1.5%to 3% and a false-negative rate in the presence of malignancy ranging from 3% to 11% [8, 10, 19, 251. In lesions less than 2 cm in diameter, the yield of a positive diagnosis in the presence of malignancy is 60% [20]. There is also significant difficulty in making a specific benign diagnosis with TNAB. In a series of 137 confirmed benign lesions by Khouri and colleagues [21], a specific benign diagnosis was established in 67.8%.The diagnosis of nonspecific benign disease was established in another 19.7%. However, the absence of malignant cells in the aspirate and lack of a specific benign diagnosis was not considered to be sufficiently reliable to document benign disease and more importantly to prevent the need for thoracotomy. In Calhoun and associates’ [25] series of 397 consecutive TNABs, 132 patients had a “no cancer” diagnosis. Only 12% of these patients had a specific benign diagnosis. Furthermore, 38 of these patients (29%)were
Ann Thorac Surg 1993;5682.%32
subsequently found to have a malignancy. The question becomes, how often does TNAB yield a result that would avoid a thoracotomy. In Khouri’s series a specific benign diagnosis was obtained in 93 of 650 patients (14%). In Calhoun’s series, only 16 of 397 patients (4%) had a specific benign diagnosis. Is the avoidance of a thoracotomy or thoracoscopy in 4% to 14% of patients who undergo TNAB worth the effort, expense, and complications of this procedure? In addition, the requirement for up to three biopsies to obtain these results and the false-positive rate of 1.5%to 3% and false-negative rate in the presence of malignancy of 3% to 11% add further questions regarding the necessity of the procedure. Standard limited thoracotomy is the definitive procedure for diagnosing the indeterminate nodule being virtually 100% diagnostic. However, there is significant morbidity associated with the procedure. The mortality rate for wedge resection or segmentectomy is 1.4% [ll]. Ray and colleagues [12] reported no mortality in 179 thoracotomies for solitary pulmonary nodules; however, only 15% had malignancies and there was an 8% incidence of postoperative complications. Against this background of diagnostic modalities for determination of solitary pulmonary nodule, we report this series of thoracoscopic resections. It has the significant advantage over bronchoscopy and TNAB of being 100%sensitive and 100%specific. This same accuracy can be accomplished by thoracotomy; however, thoracoscopy has been demonstrated to have less postoperative pain, pulmonary dysfunction, and shortened postoperative hospitalization compared with open thoracotomy [26]. The lack of operative mortality and a complication rate of 3.6%in this series compare favorably with most thoracotomy results. Thoracoscopy has the additional advantage of often being the therapeutic procedure also. In this series, in 22 patients (9.1%)with primary bronchogenic carcinoma and poor pulmonary reserve, the thoracoscopy was diagnostic as well as therapeutic and other preliminary diagnostic procedures were avoided. An additional 29 patients (11%) with bronchogenic carcinoma and adequate pulmonary reserve also underwent definitive formal lung resection without requiring another procedure. A potential disadvantage of thoracoscopy compared with thoracotomy or sternotomy in the group of patients with metastases is the inability to perform careful bimanual palpation of the lung and diagnose the occult metastatic nodules not apparent on computed tomogram. In a radiology-surgery correlation series by Peuchot and Lebshitz [27], 20% of lesions labeled as solitary pulmonary nodules by computed tomographic scan preoperatively were in fact multiple at operation. It is possible, however, that many undetected lesions are in fact adjacent to the pleural surface and, therefore, detectable by thoracosCOPY. Based on our results and the results of other diagnostic modalities for solitary pulmonary nodules reviewed here, our present management of indeterminate solitary pulmonary nodules is as follows:
MACKETAL THORACOSCOPIC LUNG RESECTION
Ann Thorac Surg 1993;56825-32
If a solitary nodule has been demonstrated to have been present unchanged in size for 2 years, or if there is a benign pattern of calcification present, or if the patient is a nonsmoker with no history of malignancy and is less than 30 to 35 years old, then an ”observation only” approach is taken. In most other clinical settings, we proceed with thoracoscopy for diagnosis unless the patient is a prohibitive risk for general anesthesia due to general debilitation or associated cardiac disease. Limited pulmonary reserve does not prohibit thoracoscopy and, as the probability of malignancy is usually great, the diagnostic procedure is also therapeutic in these patients. Miller and others have demonstrated the role of wedge resection in patients with peripheral bronchogenic carcinomas and limited pulmonary reserve [28, 291. Thoracoscopy extends this management role further by having less morbidity than even a limited thoracotomy [26]. Because TNAB has a significant false-negative rate in the presence of malignancy as well as a significant failure to yield a definitive benign diagnosis in benign disease, we believe that this diagnostic modality should assume a less prominent role in the management of the solitary pulmonary nodule. In situations where the risk of malignancy is relatively high, ie, smokers, increasing age, nodule larger than 2 cm in diameter, TNAB merely adds another diagnostic step in a situation in which operation will be warranted anyway. Even in patients with a prior history of malignancy, the likelihood that a solitary pulmonary nodule will prove to be a metastasis is 33%to 50% [30]. The remainder are equally divided between primary lung tumors and benign lesions. This mitigates toward thoracoscopic resection rather than TNAB where differentiation may not always be possible. In patients with lower risk for malignancy but not low enough for observation only, ie, younger patients (35 to 50 years), nonsmokers, those with a nodule less than 2 cm in diameter, the question to be asked is: What steps will allow this patient to avoid a surgical excision? If the yield of a specific benign diagnosis in 4% to 14% of patients and a small but definite false-positive (1.5% to 3%) and false-negative rate (3% to 11%)can be accepted by the patient and the physician, then TNAB is a reasonable alternative. However, if this lack of specificity and sensitivity is not acceptable, then a thoracoscopic resection should be pursued. At present, we reserve TNAB for patients who are not surgical candidates, have unresectable malignancy in which a tissue diagnosis is needed, or have suspected nonlocalized small cell carcinoma. In patients who clinically are likely to have cancer (ie, older age, smoker, larger lesions, spiculated edge) and in the patient who is a candidate for a lobectomy, thoracokopy generally is bypassed for standard thoracotomy. Although thoracoscopy may have a role in staging for pleura’ involvementt this incidence is probably not enough to routinely perform a “staging thoracoscopy.”
829
In conclusion, thoracoscopy is both a diagnostic and therapeutic procedure in the management of the indeterminate solitary pulmonary nodule. It has the advantage, compared with open thoracotomy, of being a minimally invasive surgical procedure with no mortality, minimal morbidity, and shortened hospital stay. It is virtually 100% sensitive and 100% specific in the diagnosis of indeterminate nodules, offering significant advantages over other diagnostic modalities including transthoracic needle aspiration biopsy in which a significant incidence of false-positive and false-negative results occur. However, it is still a surgical procedure requiring a general anesthetic and, therefore, should not be used without careful consideration. For these reasons, we believe that thoracoscopy should assume a larger role in the management of the indeterminate solitary pulmonary nodule.
References 1. Nathan MH, Collins VP, Adams RA. Differentiation of benign and malignant pulmonary nodules by growth rate. Radiology 1962;79:221-32. 2. Steele JD, Buell P. Asymptomatic solitary pulmonary nodules. Host survival, tumor size, and growth rates. J Thorac Cardiovasc Surg 1973;65:14&51. 3. Midthun DE, Swensen SJ, Jett JR. Clinical strategies for solitary pulmonary nodule. Annu Rev Med 1992;43:195-208. 4. Lillington GA. Management of solitary pulmonary nodules. Dis Mon 1991;37271-318. 5. Richardson RH, Zavala DC, Mukerjee PK, Bedell GN. The use of fiberoptic bronchoscopy and brush biopsy in the diagnosis of suspected pulmonary malignancy. Am Rev Respir Dis 1974;109:63-6. 6. Fletcher EC, Levin DC. Flexible fiberoptic bronchoscopy and fluoroscopically guided transbronchial biopsy in the management of solitary pulmonary nodules. West J Med 1982;136: 477-83. 7. Levine MS, Weiss JM, Harrell JH, Cameron TJ, Moser KM. Transthoracic needle aspiration biopsy following negative fiberoptic bronchoscopy in solitary pulmonary nodules. Chest 1988;93:1152-5. 8. Westcott JL. Direct percutaneous needle aspiration of localized pulmonary lesions: results in 422 patients. Radiology 1980;13731-5. 9. Sinner WN. Pulmonary neoplasms diagnosed with transthoracic needle biopsy. Cancer 1979;43:153MO. 10. Zelch JV, Lalli AF, McCormack LJ, Belovich DM. Aspiration biopsy in diagnosis of pulmonary nodule. Chest 1973;63: 149-52. 11. Ginsberg RJ, Hill LD, Eagan RT, et al. Modern thirty-day operative mortality for surgical resections in lung cancer. J Thorac Cardiovasc Surg 1983;86:654-8. 12. Ray JF, Lawton BR, Mangin GE, et al. The coin lesion story. Update 1976-twenty years experience with early thoracotomy for 179 suspected malignant coin lesions. Chest 1976;70: 3324. 13. Mack MJ, Aronoff RJ, Acuff TE, Douthit MB, Bowman RT, Ryan WH. The present role of thoracoscopy in the diagnosis and treatment of diseases of the chest. Ann Thorac Surg 1992;54:402-9. 14. Landreneau RJ, Hazelrigg SR, Ferson PF, et al. Thoracoscopic resection of 85 pulmonary lesions. Ann Thorac Surg 1992:54:41520. 15. Fraser RG, Par6 JAP, Par6 PD, et al. The solitary pulmonary nodule. In: Fraser RG, Par6 JAP, eds. Diagnosis of diseases of the chest. Vol 2. Philadelphia: Saunders, 1989:1383-411. 16. Swenson SJ, Jett JR, Payne WS, Viggiano RW, Pairolero PC, Trastek VF. An integrated approach to evaluation of the solitary pulmonary nodule. Mayo Clin Proc 1990;65:173-86.
830
MACKETAL THORACOSCOPIC LUNG RESECTION
17. Zerhouni EA, Stitik FP, Siegelman SS, et al. CT of the pulmonary nodule: a cooperative study. Radiology 1986;160 319-27. 18. Westcott JL. Percutaneous transthoracic needle biopsy. Radiology 1988;169:593-601. 19. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle aspiration biopsy of benign and malignant lung lesions. Am J Roentgen01 1985;144:281-8. 20. Berquist TH, Bailey PB, Cortese DA, Miller WE. Transthoracic needle biopsy accuracy and complications in relation to location and type of lesion. Mayo Clin Proc 1980;55:475-81. 21. Khouri NF, Meziane MA, Zerhouni EA, Fishman EK, Siegelman SS. The solitary pulmonary nodule: assessment, diagnosis and management. Chest 1987;91:128-33. 22. Caskey CI, Templeton PA, Zerhouni EA. Current evaluation of the solitary pulmonary nodule. Radio1 Clin North Am 1990;28:511-20. 23. Viggiano RW, Swensen SJ, Rosenow EC. Evaluation and management of solitary and multiple pulmonary nodules. Clin Chest Med 1992;13:8>95. 24. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle
Ann Thorac Surg 1993;56825-32
25. 26. 27. 28.
29.
30.
aspiration biopsy of benign and malignant lung lesions. AJR 1985;144:281-8. 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;41:592-6. Landreneau RJ, Hazelrigg SR, Mack MJ, et al. Postoperative pain-related morbidity: video-assisted thoracic surgery versus thoracotomy. Ann Thorac Surg (in press). Peuchot M, Libshitz HI. Pulmonary metastatic disease: radiologic-surgical correlation. Radiology 1987;164:719-22. Miller JI, Hatcher CR. Limited resection of bronchogenic carcinoma in the patient with marked impairment of pulmonary function. Ann Thorac Surg 1987;44:340-3. Erret LE, Wilson J, Chiu RC, et al. Wedge resection as an alternative procedure for peripheral bronchogenic carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 1985;90: 65-1. Adkins PC, Wesselhoeft CW Jr, Newman W, et al. Thoracotomy on the patient with previous malignancy: metastases or new primary? J Thorac Cardiovasc Surg 1968;56:351-61.
DISCUSSION DR DAVID J. SUGARBAKER (Boston, MA): I think Dr Mack and associates are to be congratulated for presenting this important initial experience with thoracoscopic resection of indeterminate solitary pulmonary nodules. Their review of an initial experience with 242 patients from three collaborative institutions with thoracoscopic resection for diagnosis is impressive in the zero mortality and minimal morbidity they report. The clinical decision as to the disposition of patients presenting with small peripheral indeterminate nodules is a situation where clinical judgment is of paramount importance. Traditional clinical options include follow-up with repeated chest roentgenogram or computed tomographic scans, transthoracic fine-needle aspiration, or open resection. All of these options need to be considered in the light of risk factors for malignancy as well as medical risks of diagnostic intervention. Definitive diagnosis by thoracoscopic resection is an attractive alternative to prolonged follow-up. Indeed, we all as clinical surgeons have seen patients lost to follow-up, only to present late with unresectable disease. Mack and associates’ suggestion that thoracoscopic resection may replace fine-needle aspiration in the diagnosis of solitary pulmonary nodules is attractive given the 100% specificity and sensitivity of this technique. Fine-needle aspiration is, as stated, most useful clinically when results are positive for malignancy or a definitive benign diagnosis is obtained. Based on these factors, Mack and associates’ conclusion as to the potential importance of thoracoscopic nodulectomy as a diagnostic tool would seem valid. The therapeutic importance at this time, however, remains as yet unproven. I would like to ask Dr Mack two specific questions in that regard. First, you converted the procedure to formal lobectomy in 29 patients who had adequate pulmonary reserve at the time of nodulectomy. This decision was presumably based on a frozen section diagnosis and on your apparent conclusion that lobectomy remains the standard therapy for stage I lung cancer in good-risk patients, a conclusion that I would certainly concur with. However, you also removed 30 other solitary nodules that were read as metastatic adenocarcinoma secondary to colon cancer in 15 patients, secondary to breast cancer in 10 patients, and of unknown primary in 5 patients. Given that a diagnosis of primary versus metastatic disease is often impossible to make on a permanent section, particularly in adenocarcinoma, let alone
frozen section, did you return some patients to the operating room for standard therapy of a presumed new primary lung cancer in the setting of a previous or synchronous malignancy? My second question is, given that the presentation of metastatic disease is not often as a solitary lung lesion, if other metastatic disease or an uncontrolled primary lesion was present, how did nodulectomy help in the clinical management of these cases? If other disease was not present, how did you confirm your assumption that the indeterminate solitary lesion was metastatic and not a primary lung lesion? I believe Dr Mack and associates are to be, again, congratulated for presenting this very important initial experience. I certainly think that prospective, randomized trials will be needed to establish the therapeutic efficacy of this technique compared with current therapeutic procedures.
DR PHILIP A. TEMPLETON (Baltimore, MD): I would like to say that the needle biopsy data presented are old. We presented our data at this meeting in the poster session, and there are a couple of points to be made. One is that fine-needle aspiration is perfectly acceptable for diagnosing malignant lesions and is perfectly unacceptable for diagnosing benign lesions. We always use a 19-gauge needle to obtain a core specimen in addition to cytologic material so that those distinguishing features can be made. Doing that, we can detect 98% of all primary lung carcinomas with our needle, 95% of all metastases and primary lung carcinomas. Overall, in 83% of all lesions on which we performed biopsy we obtained a diagnosis, and we mean a specific histologic diagnosis, which is the only thing we accept; not benign cells, inflammatory cells, but a specific granuloma, hamartoma, amyloid, whatever it is. I think that we should expect that most people can get a diagnosis of benign disease in about 50% of patients overall. The indeterminate solitary pulmonary nodule is only indeterminate after a percutaneous needle biopsy fails to give a diagnosis, and I would take issue with the fact that thoracoscopy is a minimally invasive surgical procedure. I think needle biopsy is a much more minimally invasive procedure, one that does not require general anesthesia or hospitalization and does not cost $12,000. The other point I would like to make is that if you had done a
Ann Thorac Surg 1993;5682.5-32
needle biopsy in the patients for whom you had determined malignancy at thoracoscopic resection, you would have done the appropriate thoracotomy operation without doing a thoracoscopy, which was, in essence, an unnecessary procedure. Our algorithm at the University of Maryland is every patient gets a biopsy, most patients get a diagnosis from that biopsy, and then we use thoracoscopy as a very refined tool, usually using a computed tomographic needle-wire-guided approach to localize the nodule so it can be resected. If it is benign, the patient has been spared a thoracotomy; and if it is malignant, then the patient is treated as a malignant thoracotomy patient. DR THOMAS W. SHIELDS (Chicago, IL): I would like to make two comments. I do not think that a patient with a history of previous malignancy and a lesion in the lung, solitary or otherwise, should be included in the category of having an indeterminate pulmonary nodule. I think these patients should be handled entirely differently than a patient who has no history of previous tumor. It is a different category entirely, and I think it would make your presentation a lot cleaner if you dropped these patients out in the discussion of the indeterminate pulmonary nodule. I would like to object to the last discussion as to the value of percutaneous needle biopsy. It is probably one of the more overused, unnecessary procedures that are done. I believe it has a very good place in the management of the patient who has had a previous malignancy. But we really should not be talking about that. However, in the patient who comes in with a new solitary pulmonary nodule that is less than 3 cm in size, and that remains indeterminate after appropriate radiologic investigation, if the fine-needle aspiration is positive for tumor, it is going to be removed. If the fine-needle aspiration is equivocal as to its benign nature, the lesion is removed. The routine use of fine-needle aspiration does not stage the patient, and the knowledge that the lesion is a carcinoma before thoracotomy does not change one's standard operative approach to the management of the patient. DR WILLARD A. FRY (Evanston, IL): I am also from Northwestern. The prior discusser is the one thousand and first point of light. Doctor Mack, I enjoyed your report very much, and I certainly support your comments that solitary pulmonary nodules usually have to be excised, and some of my comments now change after you clearly defined that superficial lesions in the outer third of the lung were being discussed. But I would like to share a couple of comments and make a plea. I think we do need to hold on to what is good when we are evaluating a new technology. I think most of the time we have a pretty good idea of what a lesion is in the lung, particularly with an older person who has a smoking history, and with modern imaging. We usually have a pretty good hunch that we are dealing with a carcinoma. And I think that starting with videoassisted thoracoscopic surgery and then going on to thoracotomy to treat cancer is going to jack up the cost a lot. I think as we go into a new presidency and new government scrutiny of what we are doing and how much it is costing, we are going to have to be more careful than ever. Doctor Steve Shortel, the health economics guru at the Kellogg School of Business at Northwestern, keeps talking about valueenhanced technology, and when you keep pushing him on quality and value it really means the bottom line and how cheap it is. I think there is mounting indication that the government is going to stop listening to the doctors and going to start listening to some of the accountants. So my plea is for us to get together and share some of the cost information. I am part of the
MACKETAL THORACOSCOPIC LUNG RESECTION
831
video-assisted thoracoscopic surgery group, as you know, but I think this is desperately needed. I would just like to remind you that there is a technique that was presented before this group a few years ago of intraoperative fine-needle aspiration biopsy [l].It is highly accurate and can be done through a small axillary incision, and you do not really have to spread the ribs apart very far. It is a quick technique; we call it frozen section cytology. We recently reviewed a new series of our cases, and we found that its accuracy is more than 90%. Every lesion must be confirmed. However, if your cytopathologist calls it a hamartoma, it can be enucleated. We found that we were performing a lobectomy for indeterminate lesions that ended up being benign in less than 6% of our cases. I think we should not lose sight of some of the older techniques that have been pretty well established. DR SEPPO E. RAP0 (Cape Cod, MA): At times it is difficult to thoracoscopicallyidentify peripheral pulmonary nodules that are not visible on the surface of the lung. Could you share with us some tricks or methods for locating imbedded lesions in the outer third of the lung? DR ROBERT J. GINSBERG (New York, NY): I would just like to support Dr Templeton and the use of fine-needle aspiration biopsy, with which you can avoid lots of thoracotomies. Granulomas do not have to be excised, hamartomas do not have to be excised, and you can plan your operation by knowing it is a carcinoma when the accuracy is getting to be 98%. The only time you have to use a thoracoscopy is for the true indeterminate nodule after this simple outpatient test has been done. DR HOWARD S. BROWN (Atlanta, GA): I want to humbly support Dr Templeton, too. At our hospital in Atlanta we have a 60% positive benign diagnosis in one hospital and 50% in the other of computed tomography-guided aspiration needle biopsy of pulmonary nodules. In the one hospital the radiologist uses a 19-gaugeneedle, obtaining a core. At the other hospital, a type of needle biopsy gun is used. About 5 years ago my philosophy was, why needle it? If it is positive you take it out; if it is negative you still have to remove it. I had a 42-year-old female patient who was overweight and had a cast on one leg from a bunion operation. I wedged the granuloma out. She had thrombophlebitis postoperatively and was in the hospital about a month. At the time she was discharged she said, "You know, they took care of my sister at another hospital with a needle for the same disease." If, with a needle, you get a positive benign diagnosis-no patient-controlled analgesia pump, no double-lumen tube, no intensive care unit, no intermediate care. I agree 100% that we should obtain core biopsy specimens on these nodules before we consider them for any type of surgical procedure. DR MACK. I do not quite know where to start. I will start with Dr Shields and thank him for not making me feel quite so lonely up here. In terms of addressing the fine-needle aspiration biopsy first, I am at the end of the chain as far as referrals are concerned and I do not self-refer myself patients. When we initially started on this approach, we found an overwhelming support and referrals from our pulmonologists and our internists who expressed frustration with nondiagnostic needle biopsies over the years. Now, I commend Dr Templeton for his results, but frankly in our situation, and we have good radiologists and a high-volume thoracic surgery service, our subjective impression was that our results were not that good. Our results seemed more to mirror
832
Ann Thorac Surg 1993;56:825&32
MACKETAL THORACOSCOPIC LUNG RESECTION
series of Drs Calhoun and Daniels from Charlottesville in which a definitive benign diagnosis was obtained in only 12%of benign nodules. Now, the difference may be the larger needle that Dr Templeton is using, and I hope that if that holds true in multiple other series there will be an increased role for fine-needle aspiration biopsy. In addressing Dr Sugarbaker’s questions, our operation for a T1 peripheral lesion is still a lobectomy. Until the full data from the Lung Cancer Study Group are released, it is our belief that the current state of treatment for a peripheral T1 carcinoma of the lung is formal lobectomy. In addressing the other questions, most of the patients who were diagnosed as having adenocarcinoma had a previous history of malignancy and the excised nodule was believed to be the same histology by comparison. There were 5 patients in whom there was an unknown primary tumor. The pathologists believed, on frozen section and confirmed on permanent section, that it was metastatic from another site. We did not take any patients back for a second procedure, but would have been prepared to if they believed it was a primary lung cancer. Three of those patients, however, had very poor pulmonary function and were borderline, elderly patients, and those patients we probably would not have taken back. In terms of addressing the issue of metastasis, this clearly is the most controversial issue. I share some of the concerns that were mentioned earlier this afternoon. Hopefully, the placement of a finger in the chest to feel the lung, although not nearly as complete as bimanual palpation, still can serve some purpose. The addition of intraoperative ultrasound has been beneficial to us, but we are still in the learning stage of this technique. I agree with the comments that Dr Lewis made earlier. Why does a computed tomographic scan that is capable of detecting nodules down to 4 mm miss some nodules? This has been alluded to already: that the metastasis is on the surface of the lung, close to the hilum, or to the mediastinal structures, and therefore not detected by computed tomographic scan. Indeed, those are precisely the nodules that may be able to be appreciated at the time of thoracoscopy. What I failed to mention at the start of this discussion is that
10% of the referrals of these patients for thoracoscopy were actually from radiologists, and those involved nodules that were less than 1 cm in size, which the radiologist believed were too small to accurately diagnose, or were immediately beneath a rib or beneath the scapula, which they believed they could not comfortably access. In terms of the comments of Dr Fry, we certainly agree that clinical acumen does play a role here. If I have a 2.5-cm nodule in a 65-year-old smoker that has a ragged edge and in my mind is clearly a carcinoma clinically, I am not going to waste time and expense doing a video-assisted thoracoscopy but will proceed immediately to a thoracotomy in that instance. Taking the open approach, however, in the past year I did have 1 patient who had unrecognized T4 pleural disease that could have been detected by thoracoscopy. But I do not think that that is enough reason to change that overall approach. Regarding tricks to find nodules in the outer third of the lung, the index finger is the most helpful. Early in our experience, we did have 2 patients in whom we could not identify the nodule, and we had to perform an open thoracotomy to locate the nodule. Since that time, we have used techniques such as needle localization of n o d u l e s t h a t makes our radiologists not quite as unhappy about not performing fine-needle aspiration biops i e s t o help localize nodules less than 1 cm in size or greater than 1 cm beneath the pleura. We went through an intermediate learning stage in which we used that technique frequently. However, as we have become more proficient with thoracoscopy, we seldom use needle localization. By careful examination of the computed tomographic scan in the operating room, locating adjacent fissures by use of instrument palpation over the lung, and by use of a finger for digital palpation of the lung, I think in the overwhelming majority of those patients you should be able to find the nodule.
Reference 1. McCarthy WJ, Christ ML, Fry WA. Intraoperative fine needle aspiration biopsy of thoracic lesions. Ann Thorac Surg 1980; 30:24.
Traditional nonoperative diagnostic approaches to the indeterminate solitary pulmonary nodule include bronchoscopy and percutaneous needle biopsy. Although both methods are minimally invasive, the diagnosis of the small, peripheral nodule may remain elusive. Open thoracotomy is often required when these methods fail to obtain a diagnosis. Between January 1991 and June 1992, 242 patients with indeterminate solitary lung nodules underwent thoracoscopic excisional biopsy as the primary diagnostic method. Wedge excisions of the nodules were all performed by thoracoscopic techniques using an endoscopic stapler alone (72%), ne0dymium:yttriumaluminum garnet laser (18%), or both (10%). A definite diagnosis was obtained in all patients. Two patients required conversion to thoracotomy to locate the nodule (both malignant). A benign diagnosis was obtained in 127 patients (52%) and a malignant diagnosis in 115 (48%).Of the malignant nodules, 51 (44%)were primary lung cancer and 64 (56%)were metastases. All patients
diagnosed with primary lung cancer having adequate pulmonary reserve (n = 29) underwent formal open lung resection during the same procedure. There was no mortality, and significant morbidity was limited to atelectasis in 3 patients (1.2%), pneumonia in 2 patients (0.8%), and prolonged air leak more than 7 days in 4 patients (1.6%). Average hospital stay for patients having thoracoscopy only (n = 213) was 2.4 days (range, 1 to 12 days). Thoracoscopy offers a minimally invasive approach for the diagnosis of the indeterminate solitary nodule. It has advantages over traditional diagnostic methods of being virtually 100% sensitive and 100% specific with no mortality and minimal morbidity. Based on these results, as well as the benefits of cost and time effectiveness, we anticipate a larger role for thoracoscopy in the diagnostic management of the indeterminate solitary pulmonary nodule.
T
thoracoscopy as the primary diagnostic modality for indeterminate solitary pulmonary nodules.
he definitive diagnosis of the indeterminate solitary pulmonary nodule often remains elusive, despite the multiplicity of diagnostic tools available. Careful examination of the chest roentgenogram or computed tomogram may show signs of benignancy including lack of growth and particular calcification patterns [14].Bronchoscopy is helpful for large central lesions but has a low diagnostic yield, approximately 10% in the small, peripheral nodule [3, 5, 61. Transthoracic needle aspiration biopsy (TNAB) has reported diagnostic accuracy ranging from 43% to 97% [7-101 for malignant lesions but is less effective in yielding a definitive diagnosis in benign nodules. Thoracotomy is accurate for diagnosis but has significant morbidity [ll, 121. Recent wider application of thoracoscopic techniques has included pulmonary resection. By the use of endoscopic stapling devices and laser lung resection techniques, the majority of peripheral pulmonary nodules are now able to be removed by thoracoscopic resective techniques [13, 141. We report our experience with use of Presented at the Twenty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 2527, 1993. Address reprint requests to Dr Mack, 7777 Forest Ln, Suite C-202, Dallas, TX 75230.
0 1993 by The Society of Thoracic Surgeons
(Ann Thorac Surg 1993;56:825-32)
Material and Methods Between January 1991 and June 1992, 242 patients presented to one of three collaborating institutions with an indeterminate solitary pulmonary nodule. The participating institutions are Medical City Dallas, Dallas, Texas; University of Pittsburgh, Pittsburgh, Pennsylvania; and St Luke’s Medical Center, Milwaukee, Wisconsin. The definition of an indeterminate solitary pulmonary nodule for this study is a single nodule, less than 3 cm in largest diameter, containing no calcium and located in the outer one third of the lung parenchyma. There were 138 men (57%)and 104 women (43%)in the study population. The mean age was 69 years (range, 21 to 87 years). Nineteen patients (7%) were 80 years old or older. Most patients had no previous diagnostic procedure, although 9 patients referred from other institutions had undergone a nondiagnostic TNAB. All thoracoscopic resections for diagnosis are performed under general endotracheal anesthesia with a double-lumen endotracheal tube to obtain ipsilateral lung collapse. An arterial line is placed for hemodynamic monitoring. All patients are placed in the lateral position 0003-4975/93/$6.00
826
MACKETAL THORACOSCOPIC LUNG RESECTION
Fig 1 . Subtle pleural changes adjacent to nodule can be located at thoracoscopy.
and prepared fully as for a standard thoracotomy. A computed tomogram showing the location of the target nodule is always present in the operating room, and trocar placement is planned based on this study. Usually, the initial trocar is placed in the sixth or seventh intercostal space in the midaxillary line. However, this placement is changed to enhance visualization of the nodule by the thoracoscope. For example, for nodules in the anterior segment of the upper lobe, the initial trocar is placed in the fifth intercostal space in the anterior axillary line. Second and third trocars, for introduction of instruments, are placed in a triangular arrangement as far as possible from the initial trocar site to allow maximum maneuverability, ease of stapler insertion, and optimal retraction of adjacent lung. Initial exploratory thoracoscopy is performed to identify the lung nodule as well as to examine for associated pleural metastases. An initial assessment of the hilum and mediastinum is also performed at this time. If the nodule is not apparent on initial visual examination of the lung, the computed tomogram is again carefully assessed and landmarks identified to help locate the lesion including adjacent fissures and any pleural changes adjacent to subpleural lesions (Fig 1). Next, instrument palpation is performed, and a change in consistency of the lung containing the nodule can usually be appreciated by this method. If the nodule still cannot be located, then a digit is placed through the nearest trocar site and used to palpate the nodule. This technique is possible because carbon dioxide insufflation is usually not necessary for thoracoscopic procedures. If the nodule is not pleural based or immediately subpleural, or is less than 1 cm in size, then preoperative needle localization is used (Fig 2). When the nodule for resection has been identified, assessment is made regarding the best method of resection. If the nodule is near the pleural surface and adjacent to an acute margin of the lung, then the resection can be most expeditiously accomplished using an endoscopic stapler (EndoGIA-30, Autosuture; United States Surgical Corp, Norwalk, CT). If the nodule is located adjacent to a flat surface of the lung or is deep within the lung paren-
Ann Thorac Surg 1993;56:825-32
chyma so that the jaws of the stapler cannot be easily placed on the lung, then a neodymium:yttrium-aluminum garnet laser is used for resection alone or in conjunction with the laser. Once the nodule is successfully resected, an endoscopic bag is inserted through a trocar site, and the specimen is placed inside for removal from the chest cavity. This is done to prevent tumor seeding of the tract if the lesion is malignant. The specimen is sent for immediate frozen pathologic examination. A margin of at least 1 cm around the lesion is obtained. If the nodule is determined to be a primary lung malignancy and the patient has adequate pulmonary reserve, an immediate formal lobectomy is performed using either video-assisted or standard muscle-sparing techniques. If the lesion is benign or metastatic, or if the patient has primary lung cancer and poor pulmonary reserve, then the procedure is ended. The area of resected lung is carefully examined for pneumostasis by partially inflating the lung, and a small pleural drainage tube (20 to 24F) is placed. Trocar sites are examined for hemostasis, and incisions are closed with a subcuticular suture. Care in an intensive care unit is seldom necessary. Pleural drainage tubes are removed when no air leak is present and fluid drainage is less than 150 mL in a 24-hour period. Parenteral narcotics are administered by patient-controlled analgesia infusion and are seldom necessary after the chest tube is removed. Hospital discharge is usually within 12 hours of chest tube removal.
Results A definitive diagnosis was obtained in all patients. A diagnosis of benign disease was obtained in 127 patients (52%) and malignant disease in 115 patients (48%). Two patients (0.8%) required conversion to an open thoracotomy to locate the nodule and obtain a diagnosis, both lesions being malignant. The benign diagnoses are listed in Table 1. Of the 115 patients with a malignant diagnosis, 62 patients (56%) had a pulmonary metastasis from an
Fig 2 . Localizing wire placed preoperatively to help identify a small nodule.
MACKETAL THORACOSCOPIC LUNG RESECTION
Ann Thorac Surg 1993;56825-32
Table 1 . Benign Diagnoses ( n = 127) Diagnosis Interstitial fibrosis/pneumonia Granuloma
Pulmonary infarction/scar Hamartoma Lymphoid aggregates/subpleurallymph nodes
No. of
Patients 35 34 13 12 7
Cytomegalovirus/pneumocystis Sarcoid nodules
Rheumatoid nodules Cryptococcal Sequestration Norcardia
Anthrosilicosis Amyloid
Sclerosing hemangioma Histiolytosis X Chronic lung rejection Pseudoxanthoma Bronchiolitis obliterans Neurofibroma
1 1 1 1 1
1 ’ 1 1 1 1
extrathoracic primary neoplasm (Table 2). Fifty-one patients (44%) had primary lung cancer. Twenty-nine patients with primary lung cancer and adequate pulmonary reserve then underwent a formal open, muscle-sparing or video-assisted lung resection during the same procedure. The remaining 22 patients had poor lung function that precluded a formal procedure, and the thoracoscopic lung resection was the definitive treatment. The method of resection was the endoscopic stapler in 72%, the neodymium:yttrium-aluminum garnet laser in 18%, or combined resection using both stapler and laser in 10%. However, in the last 6 months of the study, 88% of the resections were performed using the stapler alone. There was no operative mortality and minimal morbidity (3.6%). Three patients (1.2%) suffered significant lobar atelectasis that required postoperative bronchoscopy.
Table 2. Malignant Diagnoses (n = 115) Diagnosis
Primary lung cancer Lymphoma Metastatic lesions Colon Liver Breast
Renal Melanoma
Adenocarcinoma, unknown primary Carcinoid Esophageal Bladder
No. of Patients 51 2 62 15 11 10 10 8 5
1 1 1
827
Pneumonia occurred in 2 patients, both of whom responded to antibiotics and respiratory therapy. A prolonged air leak of greater than 7 days occurred in 4 patients (1.6%). Three of these prolonged leaks occurred in patients who underwent laser lung resection and the fourth in a patient with extensive bullous emphysema who had resection of a nodule performed with a stapler. We now use fibrin glue any time we have a residual air leak in an area of resected lung tissue to diminish this problem. Chest tubes were used in all procedures. Average chest tube duration was 1.8 days (range, 0.5 to 11 days). Postoperative length of stay was 2.4 days (range, 1 to 12 days) with most patients being discharged shortly (6 to 12 hours) after chest tube removal.
Comment The solitary pulmonary nodule is a single intrapulmonary spherical lesion that is fairly well circumscribed. There are about 150,000 new nodules detected each year in the United States [4]. Currently, approximately 40% to 50%of solitary nodules are malignant, either bronchogenic carcinoma or metastatic from an extrathoracic site [3, 41. The standard management of the solitary pulmonary nodule is prompt surgical removal unless benignancy can be demonstrated. The probability that a nodule is malignant is related to the age of the patient, diameter of the nodule, amount of cigarette smoke inhalation, overall prevalence of malignancy in solitary pulmonary nodules, the nature of the edge of the lesion, and the presence or absence of occult calcification [4]. Standard definition of benign disease includes demonstration of no growth over 2 years by prior chest roentgenogram or a specific pattern of calcification. Other criteria for benign disease include a transthoracic needle aspiration biopsy demonstrating a specific benign diagnosis and patient age less than 30 to 35 years, unless a prior extrapulmonary malignancy exists. Many noninvasive or minimally invasive diagnostic procedures exist for the evaluation of the indeterminate solitary pulmonary nodule. The least invasive method of evaluation is clinical. If the patient is less than 30 to 35 years old, has no previous history of malignancy, and is a nonsmoker, the risk of malignancy is 1%or less [15]. A “wait and watch’ approach is appropriate in this clinical group. Evaluation of the chest roentgenogram or computed tomogram is also noninvasive and often helpful. If prior roentgenograms demonstrate a stable lesion over a 2-year period or if a particular pattern of calcification (diffuse, central, laminar, or “popcorn”) exists, a benign diagnosis can be affirmed and an “observation only” approach taken [16]. The next considerations are the size and configuration of the nodule. If the nodule is 3 cm or greater in diameter, the likelihood of malignancy is 93% to 99% [17, 181 and after appropriate computed tomographic or surgical staging, or both, surgical excision without further evaluation is appropriate. If the patient has limited pulmonary reserve, a video-assisted thoracoscopic resection is performed. If adequate lung function for a formal lung
828
MACKETAL THORACOSCOPIC LUNG RESECTION
resection exists, an open procedure is instead performed, although a case can be made for performance of an exploratory thoracoscopy first to rule out occult pleural metastasis and, therefore, avoid a needless thoracotomy. In lesions less than 3 cm in diameter, unless occult calcification can be demonstrated, further investigation is indicated. In one study of 431 nodules measuring 2 cm or less, 42% were malignant [17]. In another study of 177 nodules 1 cm or less in diameter, 15% were malignant [MI. As well as size, the edge of the nodule at the lung interface is important. Bronchogenic carcinoma tends to have lobulated or fuzzy margins, whereas benign and metastatic lesions tend to be smooth with sharp borders. If the nodule is less than 3 cm in diameter, noncalcified, and there is no current or past evidence for extrapulmonary malignancy in a patient 30 to 35 years old or older, further evaluation is necessary. Sputum cytology is a very nonproductive examination, having a very low yield for the demonstration of malignancy. Positive sputum cytology has been found in only 15% to 19% of patients with proven malignancy [4]. In a similar manner, sputum cultures seldom demonstrate benign disease with assurance. There appears to be little value of sputum cytology in evaluation of solitary pulmonary nodules. The diagnostic yield of bronchoscopy ranges from 20% to 80% [5, 61. Specific benign diagnoses can be made in about 10% of nodules subsequently proved to be benign [6]. As most benign nodules and solitary metastasis are not “bronchogenic” in origin, it is not surprising that the yield is this low [4]. Because the primary objective of bronchoscopy is to diagnose benign disease and avoid unnecessary surgical resection, these results are disappointing and, in our opinion, fiberoptic bronchoscopy offers little contribution to the management of the solitary peripheral nodule. Transthoracic needle aspiration biopsy is commonly used for evaluation of solitary pulmonary nodules, is simple to perform, and has a relatively low complication rate of pneumothorax of 15%to 30% (8, 19, 201. Transthoracic needle aspiration biopsy in the diagnosis of malignancy has a sensitivity ranging from 64% to 97% [18, 21-24]. There is a false-positive rate of 1.5%to 3% and a false-negative rate in the presence of malignancy ranging from 3% to 11% [8, 10, 19, 251. In lesions less than 2 cm in diameter, the yield of a positive diagnosis in the presence of malignancy is 60% [20]. There is also significant difficulty in making a specific benign diagnosis with TNAB. In a series of 137 confirmed benign lesions by Khouri and colleagues [21], a specific benign diagnosis was established in 67.8%.The diagnosis of nonspecific benign disease was established in another 19.7%. However, the absence of malignant cells in the aspirate and lack of a specific benign diagnosis was not considered to be sufficiently reliable to document benign disease and more importantly to prevent the need for thoracotomy. In Calhoun and associates’ [25] series of 397 consecutive TNABs, 132 patients had a “no cancer” diagnosis. Only 12% of these patients had a specific benign diagnosis. Furthermore, 38 of these patients (29%)were
Ann Thorac Surg 1993;5682.%32
subsequently found to have a malignancy. The question becomes, how often does TNAB yield a result that would avoid a thoracotomy. In Khouri’s series a specific benign diagnosis was obtained in 93 of 650 patients (14%). In Calhoun’s series, only 16 of 397 patients (4%) had a specific benign diagnosis. Is the avoidance of a thoracotomy or thoracoscopy in 4% to 14% of patients who undergo TNAB worth the effort, expense, and complications of this procedure? In addition, the requirement for up to three biopsies to obtain these results and the false-positive rate of 1.5%to 3% and false-negative rate in the presence of malignancy of 3% to 11% add further questions regarding the necessity of the procedure. Standard limited thoracotomy is the definitive procedure for diagnosing the indeterminate nodule being virtually 100% diagnostic. However, there is significant morbidity associated with the procedure. The mortality rate for wedge resection or segmentectomy is 1.4% [ll]. Ray and colleagues [12] reported no mortality in 179 thoracotomies for solitary pulmonary nodules; however, only 15% had malignancies and there was an 8% incidence of postoperative complications. Against this background of diagnostic modalities for determination of solitary pulmonary nodule, we report this series of thoracoscopic resections. It has the significant advantage over bronchoscopy and TNAB of being 100%sensitive and 100%specific. This same accuracy can be accomplished by thoracotomy; however, thoracoscopy has been demonstrated to have less postoperative pain, pulmonary dysfunction, and shortened postoperative hospitalization compared with open thoracotomy [26]. The lack of operative mortality and a complication rate of 3.6%in this series compare favorably with most thoracotomy results. Thoracoscopy has the additional advantage of often being the therapeutic procedure also. In this series, in 22 patients (9.1%)with primary bronchogenic carcinoma and poor pulmonary reserve, the thoracoscopy was diagnostic as well as therapeutic and other preliminary diagnostic procedures were avoided. An additional 29 patients (11%) with bronchogenic carcinoma and adequate pulmonary reserve also underwent definitive formal lung resection without requiring another procedure. A potential disadvantage of thoracoscopy compared with thoracotomy or sternotomy in the group of patients with metastases is the inability to perform careful bimanual palpation of the lung and diagnose the occult metastatic nodules not apparent on computed tomogram. In a radiology-surgery correlation series by Peuchot and Lebshitz [27], 20% of lesions labeled as solitary pulmonary nodules by computed tomographic scan preoperatively were in fact multiple at operation. It is possible, however, that many undetected lesions are in fact adjacent to the pleural surface and, therefore, detectable by thoracosCOPY. Based on our results and the results of other diagnostic modalities for solitary pulmonary nodules reviewed here, our present management of indeterminate solitary pulmonary nodules is as follows:
MACKETAL THORACOSCOPIC LUNG RESECTION
Ann Thorac Surg 1993;56825-32
If a solitary nodule has been demonstrated to have been present unchanged in size for 2 years, or if there is a benign pattern of calcification present, or if the patient is a nonsmoker with no history of malignancy and is less than 30 to 35 years old, then an ”observation only” approach is taken. In most other clinical settings, we proceed with thoracoscopy for diagnosis unless the patient is a prohibitive risk for general anesthesia due to general debilitation or associated cardiac disease. Limited pulmonary reserve does not prohibit thoracoscopy and, as the probability of malignancy is usually great, the diagnostic procedure is also therapeutic in these patients. Miller and others have demonstrated the role of wedge resection in patients with peripheral bronchogenic carcinomas and limited pulmonary reserve [28, 291. Thoracoscopy extends this management role further by having less morbidity than even a limited thoracotomy [26]. Because TNAB has a significant false-negative rate in the presence of malignancy as well as a significant failure to yield a definitive benign diagnosis in benign disease, we believe that this diagnostic modality should assume a less prominent role in the management of the solitary pulmonary nodule. In situations where the risk of malignancy is relatively high, ie, smokers, increasing age, nodule larger than 2 cm in diameter, TNAB merely adds another diagnostic step in a situation in which operation will be warranted anyway. Even in patients with a prior history of malignancy, the likelihood that a solitary pulmonary nodule will prove to be a metastasis is 33%to 50% [30]. The remainder are equally divided between primary lung tumors and benign lesions. This mitigates toward thoracoscopic resection rather than TNAB where differentiation may not always be possible. In patients with lower risk for malignancy but not low enough for observation only, ie, younger patients (35 to 50 years), nonsmokers, those with a nodule less than 2 cm in diameter, the question to be asked is: What steps will allow this patient to avoid a surgical excision? If the yield of a specific benign diagnosis in 4% to 14% of patients and a small but definite false-positive (1.5% to 3%) and false-negative rate (3% to 11%)can be accepted by the patient and the physician, then TNAB is a reasonable alternative. However, if this lack of specificity and sensitivity is not acceptable, then a thoracoscopic resection should be pursued. At present, we reserve TNAB for patients who are not surgical candidates, have unresectable malignancy in which a tissue diagnosis is needed, or have suspected nonlocalized small cell carcinoma. In patients who clinically are likely to have cancer (ie, older age, smoker, larger lesions, spiculated edge) and in the patient who is a candidate for a lobectomy, thoracokopy generally is bypassed for standard thoracotomy. Although thoracoscopy may have a role in staging for pleura’ involvementt this incidence is probably not enough to routinely perform a “staging thoracoscopy.”
829
In conclusion, thoracoscopy is both a diagnostic and therapeutic procedure in the management of the indeterminate solitary pulmonary nodule. It has the advantage, compared with open thoracotomy, of being a minimally invasive surgical procedure with no mortality, minimal morbidity, and shortened hospital stay. It is virtually 100% sensitive and 100% specific in the diagnosis of indeterminate nodules, offering significant advantages over other diagnostic modalities including transthoracic needle aspiration biopsy in which a significant incidence of false-positive and false-negative results occur. However, it is still a surgical procedure requiring a general anesthetic and, therefore, should not be used without careful consideration. For these reasons, we believe that thoracoscopy should assume a larger role in the management of the indeterminate solitary pulmonary nodule.
References 1. Nathan MH, Collins VP, Adams RA. Differentiation of benign and malignant pulmonary nodules by growth rate. Radiology 1962;79:221-32. 2. Steele JD, Buell P. Asymptomatic solitary pulmonary nodules. Host survival, tumor size, and growth rates. J Thorac Cardiovasc Surg 1973;65:14&51. 3. Midthun DE, Swensen SJ, Jett JR. Clinical strategies for solitary pulmonary nodule. Annu Rev Med 1992;43:195-208. 4. Lillington GA. Management of solitary pulmonary nodules. Dis Mon 1991;37271-318. 5. Richardson RH, Zavala DC, Mukerjee PK, Bedell GN. The use of fiberoptic bronchoscopy and brush biopsy in the diagnosis of suspected pulmonary malignancy. Am Rev Respir Dis 1974;109:63-6. 6. Fletcher EC, Levin DC. Flexible fiberoptic bronchoscopy and fluoroscopically guided transbronchial biopsy in the management of solitary pulmonary nodules. West J Med 1982;136: 477-83. 7. Levine MS, Weiss JM, Harrell JH, Cameron TJ, Moser KM. Transthoracic needle aspiration biopsy following negative fiberoptic bronchoscopy in solitary pulmonary nodules. Chest 1988;93:1152-5. 8. Westcott JL. Direct percutaneous needle aspiration of localized pulmonary lesions: results in 422 patients. Radiology 1980;13731-5. 9. Sinner WN. Pulmonary neoplasms diagnosed with transthoracic needle biopsy. Cancer 1979;43:153MO. 10. Zelch JV, Lalli AF, McCormack LJ, Belovich DM. Aspiration biopsy in diagnosis of pulmonary nodule. Chest 1973;63: 149-52. 11. Ginsberg RJ, Hill LD, Eagan RT, et al. Modern thirty-day operative mortality for surgical resections in lung cancer. J Thorac Cardiovasc Surg 1983;86:654-8. 12. Ray JF, Lawton BR, Mangin GE, et al. The coin lesion story. Update 1976-twenty years experience with early thoracotomy for 179 suspected malignant coin lesions. Chest 1976;70: 3324. 13. Mack MJ, Aronoff RJ, Acuff TE, Douthit MB, Bowman RT, Ryan WH. The present role of thoracoscopy in the diagnosis and treatment of diseases of the chest. Ann Thorac Surg 1992;54:402-9. 14. Landreneau RJ, Hazelrigg SR, Ferson PF, et al. Thoracoscopic resection of 85 pulmonary lesions. Ann Thorac Surg 1992:54:41520. 15. Fraser RG, Par6 JAP, Par6 PD, et al. The solitary pulmonary nodule. In: Fraser RG, Par6 JAP, eds. Diagnosis of diseases of the chest. Vol 2. Philadelphia: Saunders, 1989:1383-411. 16. Swenson SJ, Jett JR, Payne WS, Viggiano RW, Pairolero PC, Trastek VF. An integrated approach to evaluation of the solitary pulmonary nodule. Mayo Clin Proc 1990;65:173-86.
830
MACKETAL THORACOSCOPIC LUNG RESECTION
17. Zerhouni EA, Stitik FP, Siegelman SS, et al. CT of the pulmonary nodule: a cooperative study. Radiology 1986;160 319-27. 18. Westcott JL. Percutaneous transthoracic needle biopsy. Radiology 1988;169:593-601. 19. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle aspiration biopsy of benign and malignant lung lesions. Am J Roentgen01 1985;144:281-8. 20. Berquist TH, Bailey PB, Cortese DA, Miller WE. Transthoracic needle biopsy accuracy and complications in relation to location and type of lesion. Mayo Clin Proc 1980;55:475-81. 21. Khouri NF, Meziane MA, Zerhouni EA, Fishman EK, Siegelman SS. The solitary pulmonary nodule: assessment, diagnosis and management. Chest 1987;91:128-33. 22. Caskey CI, Templeton PA, Zerhouni EA. Current evaluation of the solitary pulmonary nodule. Radio1 Clin North Am 1990;28:511-20. 23. Viggiano RW, Swensen SJ, Rosenow EC. Evaluation and management of solitary and multiple pulmonary nodules. Clin Chest Med 1992;13:8>95. 24. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle
Ann Thorac Surg 1993;56825-32
25. 26. 27. 28.
29.
30.
aspiration biopsy of benign and malignant lung lesions. AJR 1985;144:281-8. 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;41:592-6. Landreneau RJ, Hazelrigg SR, Mack MJ, et al. Postoperative pain-related morbidity: video-assisted thoracic surgery versus thoracotomy. Ann Thorac Surg (in press). Peuchot M, Libshitz HI. Pulmonary metastatic disease: radiologic-surgical correlation. Radiology 1987;164:719-22. Miller JI, Hatcher CR. Limited resection of bronchogenic carcinoma in the patient with marked impairment of pulmonary function. Ann Thorac Surg 1987;44:340-3. Erret LE, Wilson J, Chiu RC, et al. Wedge resection as an alternative procedure for peripheral bronchogenic carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 1985;90: 65-1. Adkins PC, Wesselhoeft CW Jr, Newman W, et al. Thoracotomy on the patient with previous malignancy: metastases or new primary? J Thorac Cardiovasc Surg 1968;56:351-61.
DISCUSSION DR DAVID J. SUGARBAKER (Boston, MA): I think Dr Mack and associates are to be congratulated for presenting this important initial experience with thoracoscopic resection of indeterminate solitary pulmonary nodules. Their review of an initial experience with 242 patients from three collaborative institutions with thoracoscopic resection for diagnosis is impressive in the zero mortality and minimal morbidity they report. The clinical decision as to the disposition of patients presenting with small peripheral indeterminate nodules is a situation where clinical judgment is of paramount importance. Traditional clinical options include follow-up with repeated chest roentgenogram or computed tomographic scans, transthoracic fine-needle aspiration, or open resection. All of these options need to be considered in the light of risk factors for malignancy as well as medical risks of diagnostic intervention. Definitive diagnosis by thoracoscopic resection is an attractive alternative to prolonged follow-up. Indeed, we all as clinical surgeons have seen patients lost to follow-up, only to present late with unresectable disease. Mack and associates’ suggestion that thoracoscopic resection may replace fine-needle aspiration in the diagnosis of solitary pulmonary nodules is attractive given the 100% specificity and sensitivity of this technique. Fine-needle aspiration is, as stated, most useful clinically when results are positive for malignancy or a definitive benign diagnosis is obtained. Based on these factors, Mack and associates’ conclusion as to the potential importance of thoracoscopic nodulectomy as a diagnostic tool would seem valid. The therapeutic importance at this time, however, remains as yet unproven. I would like to ask Dr Mack two specific questions in that regard. First, you converted the procedure to formal lobectomy in 29 patients who had adequate pulmonary reserve at the time of nodulectomy. This decision was presumably based on a frozen section diagnosis and on your apparent conclusion that lobectomy remains the standard therapy for stage I lung cancer in good-risk patients, a conclusion that I would certainly concur with. However, you also removed 30 other solitary nodules that were read as metastatic adenocarcinoma secondary to colon cancer in 15 patients, secondary to breast cancer in 10 patients, and of unknown primary in 5 patients. Given that a diagnosis of primary versus metastatic disease is often impossible to make on a permanent section, particularly in adenocarcinoma, let alone
frozen section, did you return some patients to the operating room for standard therapy of a presumed new primary lung cancer in the setting of a previous or synchronous malignancy? My second question is, given that the presentation of metastatic disease is not often as a solitary lung lesion, if other metastatic disease or an uncontrolled primary lesion was present, how did nodulectomy help in the clinical management of these cases? If other disease was not present, how did you confirm your assumption that the indeterminate solitary lesion was metastatic and not a primary lung lesion? I believe Dr Mack and associates are to be, again, congratulated for presenting this very important initial experience. I certainly think that prospective, randomized trials will be needed to establish the therapeutic efficacy of this technique compared with current therapeutic procedures.
DR PHILIP A. TEMPLETON (Baltimore, MD): I would like to say that the needle biopsy data presented are old. We presented our data at this meeting in the poster session, and there are a couple of points to be made. One is that fine-needle aspiration is perfectly acceptable for diagnosing malignant lesions and is perfectly unacceptable for diagnosing benign lesions. We always use a 19-gauge needle to obtain a core specimen in addition to cytologic material so that those distinguishing features can be made. Doing that, we can detect 98% of all primary lung carcinomas with our needle, 95% of all metastases and primary lung carcinomas. Overall, in 83% of all lesions on which we performed biopsy we obtained a diagnosis, and we mean a specific histologic diagnosis, which is the only thing we accept; not benign cells, inflammatory cells, but a specific granuloma, hamartoma, amyloid, whatever it is. I think that we should expect that most people can get a diagnosis of benign disease in about 50% of patients overall. The indeterminate solitary pulmonary nodule is only indeterminate after a percutaneous needle biopsy fails to give a diagnosis, and I would take issue with the fact that thoracoscopy is a minimally invasive surgical procedure. I think needle biopsy is a much more minimally invasive procedure, one that does not require general anesthesia or hospitalization and does not cost $12,000. The other point I would like to make is that if you had done a
Ann Thorac Surg 1993;5682.5-32
needle biopsy in the patients for whom you had determined malignancy at thoracoscopic resection, you would have done the appropriate thoracotomy operation without doing a thoracoscopy, which was, in essence, an unnecessary procedure. Our algorithm at the University of Maryland is every patient gets a biopsy, most patients get a diagnosis from that biopsy, and then we use thoracoscopy as a very refined tool, usually using a computed tomographic needle-wire-guided approach to localize the nodule so it can be resected. If it is benign, the patient has been spared a thoracotomy; and if it is malignant, then the patient is treated as a malignant thoracotomy patient. DR THOMAS W. SHIELDS (Chicago, IL): I would like to make two comments. I do not think that a patient with a history of previous malignancy and a lesion in the lung, solitary or otherwise, should be included in the category of having an indeterminate pulmonary nodule. I think these patients should be handled entirely differently than a patient who has no history of previous tumor. It is a different category entirely, and I think it would make your presentation a lot cleaner if you dropped these patients out in the discussion of the indeterminate pulmonary nodule. I would like to object to the last discussion as to the value of percutaneous needle biopsy. It is probably one of the more overused, unnecessary procedures that are done. I believe it has a very good place in the management of the patient who has had a previous malignancy. But we really should not be talking about that. However, in the patient who comes in with a new solitary pulmonary nodule that is less than 3 cm in size, and that remains indeterminate after appropriate radiologic investigation, if the fine-needle aspiration is positive for tumor, it is going to be removed. If the fine-needle aspiration is equivocal as to its benign nature, the lesion is removed. The routine use of fine-needle aspiration does not stage the patient, and the knowledge that the lesion is a carcinoma before thoracotomy does not change one's standard operative approach to the management of the patient. DR WILLARD A. FRY (Evanston, IL): I am also from Northwestern. The prior discusser is the one thousand and first point of light. Doctor Mack, I enjoyed your report very much, and I certainly support your comments that solitary pulmonary nodules usually have to be excised, and some of my comments now change after you clearly defined that superficial lesions in the outer third of the lung were being discussed. But I would like to share a couple of comments and make a plea. I think we do need to hold on to what is good when we are evaluating a new technology. I think most of the time we have a pretty good idea of what a lesion is in the lung, particularly with an older person who has a smoking history, and with modern imaging. We usually have a pretty good hunch that we are dealing with a carcinoma. And I think that starting with videoassisted thoracoscopic surgery and then going on to thoracotomy to treat cancer is going to jack up the cost a lot. I think as we go into a new presidency and new government scrutiny of what we are doing and how much it is costing, we are going to have to be more careful than ever. Doctor Steve Shortel, the health economics guru at the Kellogg School of Business at Northwestern, keeps talking about valueenhanced technology, and when you keep pushing him on quality and value it really means the bottom line and how cheap it is. I think there is mounting indication that the government is going to stop listening to the doctors and going to start listening to some of the accountants. So my plea is for us to get together and share some of the cost information. I am part of the
MACKETAL THORACOSCOPIC LUNG RESECTION
831
video-assisted thoracoscopic surgery group, as you know, but I think this is desperately needed. I would just like to remind you that there is a technique that was presented before this group a few years ago of intraoperative fine-needle aspiration biopsy [l].It is highly accurate and can be done through a small axillary incision, and you do not really have to spread the ribs apart very far. It is a quick technique; we call it frozen section cytology. We recently reviewed a new series of our cases, and we found that its accuracy is more than 90%. Every lesion must be confirmed. However, if your cytopathologist calls it a hamartoma, it can be enucleated. We found that we were performing a lobectomy for indeterminate lesions that ended up being benign in less than 6% of our cases. I think we should not lose sight of some of the older techniques that have been pretty well established. DR SEPPO E. RAP0 (Cape Cod, MA): At times it is difficult to thoracoscopicallyidentify peripheral pulmonary nodules that are not visible on the surface of the lung. Could you share with us some tricks or methods for locating imbedded lesions in the outer third of the lung? DR ROBERT J. GINSBERG (New York, NY): I would just like to support Dr Templeton and the use of fine-needle aspiration biopsy, with which you can avoid lots of thoracotomies. Granulomas do not have to be excised, hamartomas do not have to be excised, and you can plan your operation by knowing it is a carcinoma when the accuracy is getting to be 98%. The only time you have to use a thoracoscopy is for the true indeterminate nodule after this simple outpatient test has been done. DR HOWARD S. BROWN (Atlanta, GA): I want to humbly support Dr Templeton, too. At our hospital in Atlanta we have a 60% positive benign diagnosis in one hospital and 50% in the other of computed tomography-guided aspiration needle biopsy of pulmonary nodules. In the one hospital the radiologist uses a 19-gaugeneedle, obtaining a core. At the other hospital, a type of needle biopsy gun is used. About 5 years ago my philosophy was, why needle it? If it is positive you take it out; if it is negative you still have to remove it. I had a 42-year-old female patient who was overweight and had a cast on one leg from a bunion operation. I wedged the granuloma out. She had thrombophlebitis postoperatively and was in the hospital about a month. At the time she was discharged she said, "You know, they took care of my sister at another hospital with a needle for the same disease." If, with a needle, you get a positive benign diagnosis-no patient-controlled analgesia pump, no double-lumen tube, no intensive care unit, no intermediate care. I agree 100% that we should obtain core biopsy specimens on these nodules before we consider them for any type of surgical procedure. DR MACK. I do not quite know where to start. I will start with Dr Shields and thank him for not making me feel quite so lonely up here. In terms of addressing the fine-needle aspiration biopsy first, I am at the end of the chain as far as referrals are concerned and I do not self-refer myself patients. When we initially started on this approach, we found an overwhelming support and referrals from our pulmonologists and our internists who expressed frustration with nondiagnostic needle biopsies over the years. Now, I commend Dr Templeton for his results, but frankly in our situation, and we have good radiologists and a high-volume thoracic surgery service, our subjective impression was that our results were not that good. Our results seemed more to mirror
832
Ann Thorac Surg 1993;56:825&32
MACKETAL THORACOSCOPIC LUNG RESECTION
series of Drs Calhoun and Daniels from Charlottesville in which a definitive benign diagnosis was obtained in only 12%of benign nodules. Now, the difference may be the larger needle that Dr Templeton is using, and I hope that if that holds true in multiple other series there will be an increased role for fine-needle aspiration biopsy. In addressing Dr Sugarbaker’s questions, our operation for a T1 peripheral lesion is still a lobectomy. Until the full data from the Lung Cancer Study Group are released, it is our belief that the current state of treatment for a peripheral T1 carcinoma of the lung is formal lobectomy. In addressing the other questions, most of the patients who were diagnosed as having adenocarcinoma had a previous history of malignancy and the excised nodule was believed to be the same histology by comparison. There were 5 patients in whom there was an unknown primary tumor. The pathologists believed, on frozen section and confirmed on permanent section, that it was metastatic from another site. We did not take any patients back for a second procedure, but would have been prepared to if they believed it was a primary lung cancer. Three of those patients, however, had very poor pulmonary function and were borderline, elderly patients, and those patients we probably would not have taken back. In terms of addressing the issue of metastasis, this clearly is the most controversial issue. I share some of the concerns that were mentioned earlier this afternoon. Hopefully, the placement of a finger in the chest to feel the lung, although not nearly as complete as bimanual palpation, still can serve some purpose. The addition of intraoperative ultrasound has been beneficial to us, but we are still in the learning stage of this technique. I agree with the comments that Dr Lewis made earlier. Why does a computed tomographic scan that is capable of detecting nodules down to 4 mm miss some nodules? This has been alluded to already: that the metastasis is on the surface of the lung, close to the hilum, or to the mediastinal structures, and therefore not detected by computed tomographic scan. Indeed, those are precisely the nodules that may be able to be appreciated at the time of thoracoscopy. What I failed to mention at the start of this discussion is that
10% of the referrals of these patients for thoracoscopy were actually from radiologists, and those involved nodules that were less than 1 cm in size, which the radiologist believed were too small to accurately diagnose, or were immediately beneath a rib or beneath the scapula, which they believed they could not comfortably access. In terms of the comments of Dr Fry, we certainly agree that clinical acumen does play a role here. If I have a 2.5-cm nodule in a 65-year-old smoker that has a ragged edge and in my mind is clearly a carcinoma clinically, I am not going to waste time and expense doing a video-assisted thoracoscopy but will proceed immediately to a thoracotomy in that instance. Taking the open approach, however, in the past year I did have 1 patient who had unrecognized T4 pleural disease that could have been detected by thoracoscopy. But I do not think that that is enough reason to change that overall approach. Regarding tricks to find nodules in the outer third of the lung, the index finger is the most helpful. Early in our experience, we did have 2 patients in whom we could not identify the nodule, and we had to perform an open thoracotomy to locate the nodule. Since that time, we have used techniques such as needle localization of n o d u l e s t h a t makes our radiologists not quite as unhappy about not performing fine-needle aspiration biops i e s t o help localize nodules less than 1 cm in size or greater than 1 cm beneath the pleura. We went through an intermediate learning stage in which we used that technique frequently. However, as we have become more proficient with thoracoscopy, we seldom use needle localization. By careful examination of the computed tomographic scan in the operating room, locating adjacent fissures by use of instrument palpation over the lung, and by use of a finger for digital palpation of the lung, I think in the overwhelming majority of those patients you should be able to find the nodule.
Reference 1. McCarthy WJ, Christ ML, Fry WA. Intraoperative fine needle aspiration biopsy of thoracic lesions. Ann Thorac Surg 1980; 30:24.