91 - Hydatid Disease of the Lung

Editors: Shields, Thomas W.; LoCicero, Joseph; Ponn, Ronald B.; Rusch, Valerie W.

Title: General Thoracic Surgery, 6th Edition

Copyright 2005 Lippincott Williams & Wilkins

> Table of Contents > Volume II > Section XVI - Carcinoma of the Lung > Chapter 106 - Surgical Treatment of Non Small Cell Lung Cancer

function show_scrollbar() {}

Chapter 106

Surgical Treatment of Non Small Cell Lung Cancer

Ronald B. Ponn

Joseph LoCicero III

Benedict D. T. Daly

Of the about 170,000 new cases of lung cancer discovered each year in the United States, many present with distant metastasis. Pulmonary resection as the primary or the sole treatment in such cases is not beneficial. For the minority of patients with non small cell lung cancer (NSCLC) limited to the lung, however, resection alone remains the most effective therapy. For those with more advanced locoregional involvement (mainly clinical N2 disease), many current protocols include operation as part of a multimodality approach. In addition, current clinical trials are in effect to assess combined-modality treatment of lower-stage patients (stages IB and II), patients who have traditionally been treated by primary resection. In general, most of the promising results occur when operation follows induction therapy, rather than vice versa. This chapter focuses on the role of primary resection in the management of NSCLC, the indications, surgical options, and results. The growing experience with combining operation with other treatment (adjuvant and induction therapy) is discussed in Chapter 113. A thorough understanding of staging, as presented in Chapter 105, is presumed.

Historical Aspects

Surgical resection for lung cancer began with the first successful pneumonectomy, reported by Graham and Singer (1933). Subsequent advances have led to smaller resections and improved operative mortality rates. Bronchoplastic procedures were developed during the late 1940s, culminating in Allison's successful sleeve lobectomy for a bronchial carcinoma in 1952, as reported by Price-Thomas (1960). Lesser resections such as lobectomy and segmentectomy were pioneered in the 1940s. Churchill and Belsey (1939) demonstrated the feasibility of segmentectomy in a patient with bronchiectasis. These techniques were developed and popularized by Overholt (1946) and Jensik (1973) and their colleagues, as well as by Shields and Higgins (1974). The introduction and refinement of surgical staplers have made lung resection safer, faster, and less traumatic, while maintaining surgical oncologic principles. All of the modern trials in lung cancer therapy use resection as the accepted standard therapy control arm for the study of new experimental treatments.

Overview of Resection for Non Small Cell Lung Cancer

Every patient with locoregional NSCLC should be approached as a potential candidate for resection. At present, most patients with clinical stage I and II NSCLC undergo resection as the definitive primary therapy. Most patients with clinical stage IIIA or IIIB should not be resected primarily, but should be considered for multimodality therapy, ideally as part of a clinical trial. Only under exceptional circumstances should patients with stage IV NSCLC be considered for resection. Certain anatomic or physiologic considerations may make an individual patient a poor candidate for resection.

Anatomic Considerations

Complete resection is the goal of all operations for lung cancer. There should be a logical progression from chest radiography to computed tomography (CT) of the chest and abdomen and other imaging [primarily fluorodeoxyglucose (FDG) positron emission tomography (PET) scanning], to invasive assessment, most commonly surgical mediastinal evaluation (Chapter 17), and occasionally biopsy of extrathoracic sites, and, if indicated, to thoracotomy. A detailed discussion of radiologic evaluation appears in Chapters 9,10,11,12 to 13. In patients in whom the tumor is deemed resectable, the appropriate procedure is carried out. If a tissue diagnosis is

P.1549


not made preoperatively, an operative biopsy is mandatory, particularly if more than a lobectomy is required.

When the clinical and imaging data indicate that a complete resection cannot be achieved or that the clinical tumor stage is associated with a poor long-term outcome even with complete resection, operation is not indicated, and induction therapy before an attempted resection should be considered, or resection should be abandoned as an option. Tumors that present as T4 lesions are almost uniformly unresectable. However, small numbers of carefully selected patients with involvement of the superior vena cava, aorta, and left atrial pulmonary vein confluence appear to benefit from operation, usually as part of a combined approach. Patients with phrenic or recurrent laryngeal nerve invasion, once considered an absolute contraindication to resection, are being included in multimodality protocols leading to operation. Some patients with T3N0 NSCLC due to chest wall invasion or carinal proximity are candidates for initial complete resection. In addition, a small number of centers have accumulated substantial experience with operation for selected cases of unresectable T4 lung cancer such as those with tracheal involvement. Operation in cases of T4 cancer due to malignant pleural effusion, in contrast, offers no survival advantage over less invasive therapies. It is imperative to note that reports of successful surgery in locally advanced cancers constitute a truly miniscule fraction of lung cancers, are generally performed in high-volume centers, and cannot be routinely extrapolated to the overall decision pathway for NSCLC.

The optimal initial management of NSCLC with clinically proven lymph node dissemination is also nonsurgical. N3 disease remains out of bounds for resection. This was pointed out by Watanabe and colleagues (1991a, 1991b), who support the position that even the anterior mediastinal nodes (station 3) have a dismal prognosis. In addition, little disagreement remains for patients with N2 disease documented before thoracotomy (clinical N2). In patients discovered to have N2 disease at thoracotomy following an appropriate negative evaluation (clinical N0 1, pathologic N2), in contrast, resection should usually proceed, but careful consideration should be given before performing a pneumonectomy in the setting of N2 disease with obvious extracapsular involvement. Resection as primary treatment, however, in clinical N0 1, pathologic N2 patients without bulky lymphadenopathy carries a better prognosis than for patients with prethoracotomy N2 status, with late survival rates between 17% and 28%, as reported by Pearson (1982), one of us (BDTD) (1993), and D.L. Miller (1994) and their colleagues, among many others. The experience with postoperative adjuvant therapy in these patients is detailed in Chapter 113.

Metastatic Disease

The presence of multiple extrathoracic metastases is an absolute contraindication to pulmonary resection. When a solitary metastasis is thought to be present, resection of the primary lung cancer should be considered only in occasional cases of cerebral metastasis and only after thorough imaging and invasive assessment have confirmed the absence of other sites of disease. Multiple series have shown compelling evidence of the efficacy of complete resection, when possible, of both the cranial and intrathoracic sites. In contrast, median survival with no treatment, steroids only, or cranial radiation is about 1 month, 2 months, and 6 to 9 months, respectively. Although a handful of synchronous solitary adrenal metastases have been treated by combined resections, experience with this approach is extremely limited, and an initial nonsurgical plan is preferable. Primary lung resection is not indicated in the presence of other sites of dissemination, even if clinically thought to be isolated. In all cases of suspected single metastasis associated with limited intrathoracic disease, in addition to ensuring that a distant lesion is solitary before embarking on combined resection, it is also important to reach a secure diagnosis that the abnormal focus is malignant before deciding against primary pulmonary resection. Brain metastases are usually diagnosed reliably by imaging and rarely require invasive confirmation. If an apparently solitary cerebral metastasis is identified by CT scan, however, it is often prudent to obtain a magnetic resonance imaging (MR) scan of the brain before resection, because of the superior sensitivity of the latter technique. Additional imaging is also often required after abnormalities are detected on bone scan, or after an adrenal, renal, hepatic, or other mass is identified by CT. In most instances, a synthesis of the data provided by an appropriate combination of plain radiographs, CT, MR, ultrasound, and PET will reliably confirm or rule out distant metastasis. In the few patients with solitary lesions that remain equivocal, percutaneous or open biopsy may be indicated before pulmonary resection is denied. The importance of securing a reliable radiographic or tissue diagnosis in this setting is exemplified by the findings of Porte and associates (1998), who reported that nearly half of all adrenal masses detected during the evaluation of NSCLC patients were benign. Ettinghausen and Burt (1991) also stressed the frequency of coexistent benign adenomas in patients with operable NSCLC. Other radiographic and scintigraphic abnormalities identified during staging are also frequently unrelated. Although its role in diagnosing and staging NSCLC remains under study, PET appears to be very helpful in assessing such findings.

Physiologic Considerations

Traditionally established physiologic barriers to surgical resection have been falling steadily over recent years. Centers with a strong commitment to aggressive preoperative and postoperative care are reporting mortality and morbidity rates equal to series with limited patient entry. Kirsch and associates (1976) noted that age per se is not a contraindication to resection but cautioned against pneumonectomy in patients older than 70 years of age. However,

P.1550


Naunheim and associates (1994) reported good survival among octogenarians having a pneumonectomy. Pagni and colleagues (1998) performed 24 pneumonectomies in patients older than 70 years with a 12.5% operative mortality rate, compared with a 4.3% mortality rate for pneumonectomies in younger individuals. Nugent and co-workers (1997) pointed out that they screen many more elderly patients than they eventually operate on (with only 6% of patients 80 years and older versus 32% of those 45 years and younger undergoing thoracotomy in their practice). Long-term survival following resection is not different for elderly patients, as pointed out by Harviel (1978) and Ishida (1990b) and their colleagues. However, there tends to be a decreased survival in patients younger than 40 years of age as noted by Pemberton and co-workers (1983). Jubelirer and Wilson (1991) attribute this difference to more advanced disease at the time of diagnosis rather than to a differential response to treatment.

Despite the fact that most patients with lung cancer are older and have a significant smoking history, which can lead to coronary artery disease, significant cardiac morbidity is rare in patients undergoing a thoracotomy for carcinoma. A myocardial infarction within 3 months before surgery carries some risk for reinfarction. However, Rao and associates (1983) found this rate to be only 5.7%. A full discussion of the appropriate cardiac evaluation for thoracotomy is covered in Chapter 20. If significant disease is identified, many patients can still undergo pulmonary resection, as reported by Piehler (1985) and Canver (1990) and their associates. Successful resections have been performed after angioplasty, concomitantly with coronary artery bypass surgery, or sequentially within 2 weeks after bypass.

Poor measured pulmonary function has traditionally been considered to be a formidable barrier to resection. Based on old, unconfirmed data, elaborate schemas were established to filter only the best candidates for resection. We have learned with the lung volume reduction surgery experience that nearly all patients will tolerate some type of pulmonary resection. In general, individuals who function daily at a normal activity level, regardless of their measured parameters, will do well. A report by Korst and co-workers (1998) showed that after 6 months, patients who underwent a lobectomy had measured pulmonary function that was not statistically different from their preoperative values. In fact, some patients who had upper lobectomies actually had better postoperative function. Nonetheless, pulmonary function testing (PFT) remains an important objective evaluation of a patient's ability to tolerate resection. However, no single parameter has proved to be reliably prognostic, and strict interpretation could deny resection to many physiologically eligible patients. As many as 60% of patients identified as high risk by PFT criteria did well with surgery according to Boushy and associates (1971). Several more recent prospective studies have found no correlation between pulmonary function testing and postoperative complications. Gerson and colleagues (1990) studying elderly patients and Milledge and Nunn (1975) studying patients with emphysema found that neither PFTs nor arterial blood gases correlated with postoperative problems. This lack of correlation led Kohman and colleagues (1986) to conclude that most postthoracotomy deaths are unpredictable and random. Ferguson and co-workers (1991) suggested that a postoperative diffusion capacity of less than 20% predicted was correlative with mortality, but we now routinely operate on such patients for lung volume reduction surgery with little trouble. It is to be noted, however, that Ninan and co-workers (1997) found a highly significant correlation between exercise desaturation and postpneumonectomy morbidity.

Algorithms for preoperative pulmonary evaluation are presented in Chapter 19. Briefly, assessment begins with the patient's history. An assessment of the patient's physical activity should be elicited. Specifically, attention should be paid to the ability to climb one or more flights of stairs, the ability to walk more than a block without stopping, and the ability to perform house or yard work without difficulty. Office assessment may include a modified 6-minute walk test. Standard PFTs should be performed and include the following: spirometry, including forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and forced expiratory flow rate (FEF); and lung volumes, including total lung capacity (TLC), residual volume (RV), functional residual capacity (FRC), and diffusion capacity. The predicted postoperative parameters (e.g., PPFEV1) can be estimated by multiplying the measured FEV1 by the expected number of segments remaining after resection, each of which is assigned a contribution to overall pulmonary function of 5%. A PPFEV1 of less than 30% of the patient's expected value may be cause for concern. Additional testing that may be helpful in equivocal situations includes quantitative ventilation-perfusion scanning and exercise testing. The stepwise approach suggested by Datta and Lahiri (2003) seems reasonable. These authors suggest that if the diffusing capacity of the lung for carbon monoxide (DLCO) and FEV1 are more than 60% of predicted, the patient is at low risk for pulmonary resection, including pneumonectomy, without further testing. If these values are less than 60%, a quantitative lung scan is performed. If the scan yields predicted postoperative FEV1 and DLCO greater than 40%, operation can proceed. When the predicted values are less than 40%, exercise testing is performed. In patients with maximum oxygen consumption ([V with dot above]O2max) greater than 15 mL/kg, surgery can be undertaken; if less than 15 mL/kg, the risk is high, and operation may be contraindicated. Such results must always be interpreted in conjunction with the patient's overall status.

Principles of Surgery for Non Small Cell Lung Cancer

The goal of surgical treatment of NSCLC is complete resection. Incomplete resection not only confers no therapeutic

P.1551


advantage but may also temporally postpone and physiologically limit any potential benefit of subsequent radiation therapy or chemotherapy as well as impair the patient's quality of life. With currently available staging modalities, the incidence of nonresective thoracotomy or grossly incomplete resections should be low but never totally unanticipated. In contrast, a microscopically positive resection margin (usually involving the bronchial margin), a finding of subclinical parenchymal sites of cancer, or, most commonly, the pathologic documentation of unsuspected malignant lymphadenopathy, is often unavoidable, despite appropriate preoperative assessment. In addition to the absence of gross residual tumor and microscopically negative surgical margins, some believe the definition of a complete resection should include negativity of the highest or most distant resected lymph node. However, this has never been proved.

Every operation for lung cancer has three essential parts: establishment or confirmation of the diagnosis and the intrathoracic stage, complete resection of the tumor, and the systematic sampling or complete dissection of every ipsilateral lymph node station potentially draining the primary tumor. In addition, when complex resections are performed, appropriate reconstruction may be necessary.

Intraoperative Diagnosis and Surgical Staging

Depending on the specifics of each case and the surgeon's preferred approach, a diagnosis of NSCLC will have been made preoperatively in many patients by bronchoscopy or transthoracic needle biopsy, less commonly by thoracoscopy, and rarely by mediastinoscopy in cases of primary resection. When a diagnosis of cancer has not been secured before thoracotomy, either intentionally or because of inconclusive results, it is advisable to establish a diagnosis intraoperatively before proceeding with resection. Stapled wedge resection and frozen section will most often accomplish this goal. Frozen section assessment for NSCLC is generally straightforward and accurate. For lesions that are central or otherwise not amenable to a nontraumatic wedge resection, sampling can be carried out by fine-needle aspiration (FNA) cytology or core needle biopsy. Incisional biopsy may occasionally be needed but is less desirable because it requires macroscopic violation of a potential neoplasm. For the same reason, stapling across abnormal tissue for diagnostic assessment is discouraged. In the rare instance of a suspicious tumor not amenable to wedge resection and in which aspiration or core biopsy is nondiagnostic, a lobectomy may be required. Pneumonectomy and extended lobar resections, however, should not be performed in the absence of a firm confirmation of malignancy.

In addition to confirming the diagnosis of cancer when necessary, thoracotomy for NSCLC includes an assessment of the hemithorax for other sites of disease, whether suggested by preoperative imaging or not. The entire lung is palpated and inspected for other masses. The hilar and mediastinal nodes are examined, as is the pleura and any fixation of the primary tumor to adjacent structures. Although it is currently rare to abandon a planned resection based on unexpected intraoperative findings, an occasional seemingly straightforward case may be found at thoracotomy to be unresectable, based on the intraoperative finding of multiple usually subpleural lesions, noneffusive pleural dissemination, or a malignant pleural effusion. In patients with unexpected, malignant small pleural effusions found at thoracotomy, Sawabata and associates (2002a) reported very poor survival. These authors noted a median survival time and 5-year survival rate of 13 months and 9%, respectively, even when the resection was complete. There is no current consensus, however, regarding how to proceed in this setting. Ideally, rapid cytology is performed on the fluid. If positive, a decision regarding resection is made based on the total clinical and oncologic synthesis. Extensive parenchymal resections should not be done if the fluid is known to be malignant. More often, nonresective thoracotomy occurs in the setting of suspected, but clinically unproven, invasion of vital structures that is confirmed and deemed unresectable at the time of operation. When unsuspected lymphadenopathy is found at operation, resection should generally proceed. As discussed subsequently, the presence of N2 disease discovered at operation following an appropriate invasive and noninvasive staging evaluation does not per se preclude a definitive resection. Exceptions include the unusual finding of unexpected extensive, fixed, or bulky adenopathy or the presence of positive interlobar nodes (or transfissural direct tumor extension) that would mandate a pneumonectomy in a marginal patient. In any case, there should be no hesitation to sample and assess by frozen section or cytologic analysis any nodal, pleural, or parenchymal tissue or pleural fluid that, if positive, would render resection inappropriate. Special care is warranted in cases that require a pneumonectomy. The potential for nonresective thoracotomy is generally known preoperatively and should be discussed clearly with the patient. The incidence of nonresective thoracotomy in older series was as high as 20%, but should currently be 5% or less, although there is no specific modern benchmark. Despite appropriate clinical staging, a small number of patients with locally advanced lesions still require exploratory operation to ascertain resectability with certainty.

In addition to standard surgical staging by gross inspection and frozen section, a number of authors, including Kondo (1993), Dressler (1999), Okada (1999), and Kotoulas (2001) and their associates, have suggested that intraoperative pleural lavage cytology (PLC) be performed in all cases without obvious pleural effusion, before resection or prior to chest closure, or both. The rationale for this procedure is to detect the presence of malignant cells in the pleural cavity and thereby identify patients at higher risk for recurrence. As noted by Kotoulas and co-workers (2001), positive cytology tends to be associated with a more advanced primary T factor. Although the volume of lavage fluid employed varies, all protocols require that the preresection

P.1552


sample be obtained immediately upon entry into the chest, before any manipulation of the tumor. When obtained, the postresection sample is taken just before closure of the thoracotomy, with some protocols requiring prior copious pleural irrigation. Higashiyama and associates (1997) found pleural recurrence in 26% of patients when both the initial and preclosure samples were positive. At present, it is not suggested that the results of pleural lavage be used to affect intraoperative decisions regarding resectability. In addition, positive lavage cytology does not currently alter the stage of the tumor or the definition of complete resection, nor have recommendations for adjuvant treatment been based on this finding. Saito and colleagues (2002) recently described, and one of us (RBP) (2002) commented on, the use of intraoperative pleural touch cytology (PTC). In this method, the visceral pleura adjacent to a peripheral cancer is gently touched with a glass slide in order to detect lesions that are likely to shed cells into the pleura. Among 100 patients, PTC was positive in 17, whereas PLC was positive in 7, all of which were detected by PTC. The clinical implications of PTC are unknown. Although worthy of further study, pleural lavage or touch cytology is not presently considered a standard component of resection for lung cancer.

Resection

After a diagnosis of cancer has been made and resectability established, the appropriate pulmonary or extended resection, along with systematic lymph node sampling or lymphadenectomy, is carried out. For patients with adequate lung function, the current standard cancer resections include lobectomy, bronchoplastic lobectomy, bilobectomy, and pneumonectomy, based on the extent of disease. In some cases, an anatomic segmentectomy may be appropriate. At present, nonanatomic or wedge resection should be considered as definitive therapy only in the minority of patients whose cardiac or pulmonary status mandates conservation of pulmonary parenchyma or in certain cases of synchronous or metachronous multiple tumors. As discussed later, however, there is increasing interest is assessing limited operations for NSCLC that is considered low-grade by preoperative imaging ( ground glass opacity on CT scan) (see Chapter 101). Although summarized subsequently, the morbidity and mortality of the various resections are discussed in detail in Chapters 28,29,30,31,32,33,34,35,36 and 37.

Lobectomy

Lobectomy is the ideal operation for resection of a lung cancer confined to the parenchyma of a single lobe. It permits removal of the tumor along with the associated peripheral (pleural) and central lymphatic drainage pathways. Lobectomy is generally well tolerated, usually leaves sufficient lung volume to fill the pleural void left by resection, and avoids some of the short-term and late complications of pneumonectomy. Lobectomy is associated with about half the operative mortality of pneumonectomy (about 2% versus 4%), as reported by Ginsberg (1983) and Wada (1998) and their associates. In elderly patients, the risk is higher, but in most recent series is acceptable, as demonstrated by Naunheim (1994) and Wada (1998) and their associates, as well as by Damhuis and Schutte (1996). Pagni and associates (1997, 1998) reported operative mortality rates for lobectomy of 2% in 293 patients older than 70 years of age and of 4% in 45 octogenarians.

A bilobectomy involves resection of the right upper and middle lobes or of the right lower and middle lobes. The former operation is indicated when a tumor located in the anterior segment of the right upper lobe or in the right middle lobe has spread across the minor fissure or approximates an incomplete fissure. Failure to perform a bilobectomy in this setting may result in a positive or unacceptably close parenchymal resection margin. When a tumor in the right lower lobe is central, a bilobectomy may be required because of the proximity of the origins of the superior segmental and middle lobe bronchi. Other indications may include certain cases of interlobar vascular or nodal involvement, but a pneumonectomy should be considered in most such instances. In a series of 166 bilobectomies reported by Keller and colleagues (1988), the indications for this procedure were tumor extending across a fissure in 45%, absent fissure in 21%, endobronchial tumor in 14%, external or nodal invasion of the bronchus intermedius in 10%, vascular invasion in 5%, and miscellaneous reasons in 5%. The operative mortality for bilobectomy is generally reported as higher than for lobectomy, but lower than the risk associated with pneumonectomy.

A sleeve lobectomy consists of the resection of a lobe along with a circumferential segment of the adjacent main-stem bronchus and is generally an alternative to pneumonectomy. Bronchial continuity is restored and lung parenchyma preserved by anastomosis of the proximal and distal bronchial resection edges. This operation is most often indicated for endobronchial tumors at the origins of the right upper or left upper lobe bronchi. Occasionally, sleeve lobectomy is suitable for patients with limited nodal disease affixed to the bronchial wall at the orifices of these lobes. Nodal disease of this type was the indication in 21% of the cases reported by Deslauriers and colleagues (1993). Overall, these authors achieved a complete resection in 87% of 142 sleeve lobectomies, with an operative mortality rate of only 2.5%. Five- and 10-year survival rates were 63% and 52%, respectively, for stage I tumors. Local recurrence ultimately occurred in 23% overall and in 17% of completely resected cases. The success of bronchoplastic resection in properly selected patients is also shown in the series reported by Watanabe and colleagues (1990), with late survival in 79% of stage I, 55% of stage II, 30% of stage III patients, and in 45% of patients overall.

A bronchoplastic resection is less often appropriate when bronchi other than those of the upper lobes are involved by NSCLC, but is occasionally undertaken in oncologically favorable

P.1553


situations or as an alternative to pneumonectomy in patients with limited lung function. Sleeve resection of the pulmonary artery can be accomplished with or without a bronchial sleeve resection, but most cases with this degree of local invasion are inoperable or are treated by pneumonectomy. Although the use of sleeve lobectomy varies widely among institutions, in most centers, it averages 5% or less of lung cancer resections.

Although concern has been raised that the local recurrence rate is higher following bronchoplastic lobectomy than after pneumonectomy, reports from centers with significant experience with this approach, including those of Faber (1987), and Vogt-Moykopf (1986), Watanabe (1990), Newton (1991), Tedder (1992), and Deslauriers (1993) and their associates, show an acceptable operative mortality, a high rate of complete resection, and a late survival that is generally comparable, stage-for-stage, with other types of complete resection in NSCLC.

Pneumonectomy

A pneumonectomy is required when a lobectomy or one of its modifications is not sufficient to remove all locoregional disease. It must be kept in mind that a pneumonectomy is a radical procedure that can result in the loss of more than 50% of a patient's lung function and pulmonary vascular bed. It is most often indicated in patients with central tumors that involve the main-stem bronchus, large parenchymal cancers that violate the fissures or invade the interlobar vessels or lymph nodes, and in some cases of lymph node involvement at the level of the main-stem bronchus. Pneumonectomy in the latter situation should be reserved for cases in which higher stations are benign and a complete resection is possible. The operative mortality for pneumonectomy is about twice that of lobectomy. Ginsberg and colleagues (1983) reported a 6% operative mortality rate among 2,220 cases of the Lung Cancer Study Group (LCSG), and Wada and associates (1998) noted a rate of 3% among 590 patients undergoing resection for lung cancer in Japan during 1994. Right pneumonectomy carries a higher risk than does removal of the left lung. An increasing number of patients with N2 disease or central, locally invasive cancer are now being treated by induction therapy. Because of the extent of their disease, a high percentage require pneumonectomy (23% to as high as 53%). Despite the frequent technical difficulty posed by postinduction peribronchial and perivascular fibrosis, operative mortality in this group can be as low as 5%, but ranges up to 15%, as reported by Faber (1989), Strauss (1992), Rusch (1993, 1994), Mathisen (1996), and Weiden (1994) and their associates.

There are three types of extended pneumonectomy. The most commonly employed variation is an intrapericardial pneumonectomy, necessitated by encroachment of a central tumor at or near the entry of the pulmonary vessels (most often the artery central to its branches) into the pericardium. Ligation within the pericardium may provide both a greater margin of resection and a longer segment for safe division of the vessel. Although this approach may be associated with a higher incidence of postoperative arrhythmias, the operative risk is not higher than for standard pneumonectomy. Another modification is a supraaortic pneumonectomy and involves transecting the left main-stem bronchus more proximal than in a standard left pneumonectomy (closer to the trachea, above the aortic arch). This approach is occasionally needed for tumors originating high in the bronchus. The third variation is the carinal or sleeve pneumonectomy, consisting of resection of the lower trachea, the carina, and a main-stem bronchus and its associated lung (usually the right) with a tracheobronchial anastomosis of the remaining lung. This procedure is indicated for central lesions approximating or involving the carina that appear totally resectable by this approach. Although some earlier series reported operative mortality in as many as one fourth to one third of cases, recent reports by Dartevelle and associates (1995) and by Mitchell and colleagues (2001) have achieved rates of 7% and 15%, respectively. The latter authors noted a decrease from 20% to 10% in operative mortality rates between the first and second halves of their series. Porhanov and colleagues (2002) reported an operative mortality rate of 16% among 231 carinal resections. Some cases in these series were pure carinal resections or lobar and carinal resections. The risk of the less often performed left carinal pneumonectomy is higher than that of the right lung.

Segmentectomy

It is generally agreed that a segmentectomy is an acceptable operation for NSCLC when a patient has limited pulmonary reserve and a small peripheral tumor confined to an anatomic segment. Whether segmental or wedge resection constitute adequate treatment for small peripheral cancers in general or for cancers in which the preoperative radiographic features suggest a low-grade tumor remains under investigation. Although any segment can be removed by anatomic dissection, resections of the upper lobe segments or the superior segments of the lower lobes are performed most commonly. Lingulectomy, although encompassing two segments, is also a form of segmentectomy and is often feasible for peripheral NSCLC. Jensik and colleagues (1973) reported the first large series of segmental resection for lung cancer. Among 123 patients, 5-year and 10-year survival rates were 56% and 27%, respectively. In an instructive analysis, Kodama and colleagues (1997) compared three groups of patients with T1N0 NSCLC: (a) 46 patients undergoing segmentectomy as an intentional procedure, (b) 17 patients in whom segmental resection was viewed as a compromise because of limited lung function, and (c) 77 patients treated by lobectomy and lymph node dissection. There was no significant difference in late survival between the lobectomy group (88%) and the intentional segmentectomy patients (93%). However, the difference in survival between these two groups and patients undergoing segmentectomy as a compromise procedure

P.1554


(48%) was significant. In a report by Warren and Faber (1994) comparing 68 patients with T1 2N0 tumors treated by segmental resection with 105 similar patients undergoing lobectomy, there was also an overall survival difference favoring the lobectomy group, but the differential was not significant for tumors 3 cm or less in size. However, for the total series, the rate of locoregional recurrence was 23% following segmentectomy, as contrasted with 5% after lobectomy. Similarly, the only prospective experience, collected by the LCSG and reported by Ginsberg and Rubinstein (1995), indicates that local recurrence following limited resection for T1N0 NSCLC (including both segmentectomy and wedge excision) is threefold higher than for lobectomy, although ultimate survival is less dramatically impacted. Despite an increased risk for local recurrence, anatomic segmental resection remains an appropriate option in patients with limited lung function and also in patients with small peripheral tumors.

Wedge Resection

In contrast to segmentectomy, wedge resection is a nonanatomic operation that should be considered as definitive therapy only in poor-risk patients. Despite a higher risk of local recurrence when compared with anatomic resection, however, wedge excision may still be preferable to alternative treatments when applied appropriately. In an early nonrandomized series reported by Errett and associates (1985), wedge resection was performed as a compromise operation in 97 patients with pulmonary impairment and compared with 100 patients treated by lobectomy. Despite higher predicted risk, the wedge resection group incurred only a 3% operative mortality rate, as compared with 2% in the lobectomy group, and late survival was not statistically different. In patients with T1N0 NSCLC, Landreneau and colleagues (1996) retrospectively compared 42 cases treated by open wedge resection, 60 by video-assisted thoracic surgery (VATS) wedge resection, and 117 by standard lobectomy. Despite reduced pulmonary function and older age, there was no mortality in the combined wedge groups, as compared with a 3% mortality rate in the lobectomy group. However, as in the LCSG experience, local recurrence rates were higher in the open and VATS wedge patients (24% and 16%, respectively) than in the cases treated by lobectomy (9%). Although the 5-year survival rate was significantly lower in the open wedge cohort than in the lobectomy patients (58% vs. 70%), the 5-year survival rate in the VATS patients was similar to that in the lobectomy group at 65%. The authors point out that the minimum requirements for an appropriate wedge resection for NSCLC include the following: a tumor less than 3 cm in diameter; a location in the outer third of the lung and technically amenable to adequate local excision, absence of endobronchial extension, clear margins by frozen section; and mediastinal and hilar lymph node sampling. When these criteria are met, wedge resection is an acceptable option in the few patients unable to tolerate an anatomic operation. Local excision by cautery or laser has been performed, but the late benefit is unknown, and this approach cannot be recommended as standard.

Video-assisted Thoracic Surgery Resection

VATS has been successfully employed for lobectomy, pneumonectomy, and local resection of NSCLC. For surgeons with experience and skill with this approach, VATS is an acceptable alternative to open operation. It is essential that the same principles of lung cancer surgery that guide standard resection be maintained when VATS is applied. This topic is addressed in Chapters 32 and 33.

Evaluation of the Lymph Nodes

The single most important prognostic surgical factor in NSCLC is the status of the mediastinal lymph nodes. Evaluation begins preoperatively with a CT scan of the chest. Although there is controversy regarding the need for routine mediastinoscopy, all accessible nodes that are enlarged should undergo biopsy before thoracotomy. Nodes that are positive on PET scan, whether or not enlarged on CT, need to undergo biopsy to rule out both metastasis and false-positive uptake. In addition, invasive mediastinal node assessment should be considered in all patients at high risk for metastasis, such as those with central tumors. In addition to mediastinoscopy, extended mediastinoscopy, and anterior mediastinotomy (see Chapter 17), transthoracic needle biopsy is often useful, especially for the subaortic and posterior subcarinal nodes. Roviaro (1995) and DeGiacomo (1997) and their associates, among others, have proposed thoracoscopy as a method of nodal and tumor staging.

At thoracotomy, lymph node assessment has been performed by examining only those nodes attached to the resected specimen, by sampling only nodes that appear abnormal, by systematic biopsy of each node station, and by complete lymph node dissection. It is clear that the first two approaches are insufficient because N2 nodes can be involved in the presence of benign N1 levels ( skip metastases), and even small, normal-appearing nodes may harbor metastasis, as shown by one of us (BDTD) and colleagues (1993).

The current minimum standard is a systematic sampling of each lymph node station draining a tumor. For right-sided resections, nodes should be taken from mediastinal levels 2, 3, 4, 7, 8, and 9, as well as from the tracheobronchial angle and interlobar area (levels 10 and 11). On the left, the subaortic and anterior mediastinal nodes (levels 5 and 6) should undergo biopsy as well. Systematic sampling is required because of the frequent finding of pathologic N2 disease that was unsuspected by clinical staging and the prognostic and therapeutic implications of this situation. The incidence of unsuspected N2 after various staging pathways using modern imaging, including PET, and either routine or selective mediastinoscopy is at least 10%. Goldstraw

P.1555


and colleagues (1994), for example, found pathologic N2 in 24% of clinical N0 1 cases. Further refinements of and experience with PET may decrease the gap between clinical and pathologic staging.

Some surgeons believe that complete mediastinal lymph node dissection (MLD) is indicated for diagnostic and therapeutic reasons in all resections for NSCLC. In the prospective randomized series reported by Sugi and associates (1998) and by Izbicki and colleagues (1998), however, MLD did not increase overall survival. Although also therapeutically unproven, we believe that it is reasonable to perform MLD in patients found at thoracotomy to have N2 disease. In addition, operations following induction therapy for known N2 NSCLC should include MLD in virtually all cases. Although improved survival with routine MLD has not been proven, this approach does minimize sampling error by identifying N2 metastases that otherwise might have been missed, as shown recently by Graham and associates (1999). In the aforementioned study by Izbicki and co-workers (1998), N2 disease in 5.5% of the MLD group would have been missed by sampling. This figure may be high because in their sampling cohort, only levels 4, 5, and 7 underwent routine biopsy, whereas nodes were taken from other mediastinal areas only when suspicious. In a subsequent report from the same group by Passlick and colleagues (2002) using immunohistochemical staining, no difference in survival was found between the sampling and MLD patients overall, nor in those with micrometastases. However, in patients whose staining was negative, survival was superior in the MLD group. The use of routine MLD remains controversial and is discussed further in Chapter 107. At present, either MLD or a complete systematic sampling is acceptable as a routine approach. However, there should be a low threshold to proceed to MLD.

Selection of the Operative Procedure

The appropriate operation depends on the clinical and surgical stage of the tumor and an accurate assessment of the structures involved. Unless all gross tumor can be encompassed in the resection, operation should not be undertaken. With some exceptions, therefore, T4 NSCLC is not suitable for resection.

Lung cancers have often been classified as occult, peripheral, or central. Occult lesions are not seen radiographically, but their presence is detected by sputum cytology or bronchoscopy. Central lesions are located radiographically within the central third of the hemithorax or bronchoscopically within or proximal to a segmental bronchus. Peripheral tumors are located beyond a segmental bronchus and in the outer two thirds of the lung.

Occult Tumors

Most cases of occult NSCLC are brought to attention in screening programs for high-risk people or present with hemoptysis, cough, or wheezing. Lobectomy is most often required [70% in the experience of Cortese and colleagues (1983)] because the lesion cannot be further localized anatomically or because of a documented location in a lobar or segmental bronchus. Because of a more central location, some require bilobectomy, pneumonectomy, or bronchoplastic resection. In a report of 94 cases by Saito and associates (1992), there were 58 lobectomies, 12 bilobectomies, 11 sleeve lobectomies, and 12 pneumonectomies. In some patients, when occult NSCLC is confined to the bronchial mucosa or is an in situ carcinoma covering less than 3 cm of the mucosal surface, or in medically inoperable cases, photodynamic therapy (PDT) has been used successfully as primary treatment, as reported by Lam (1994) as well as by Kato (1996), Cortese (1997), and Weigel (2000) and their associates. Alternatively, brachytherapy can be delivered through bronchoscopically placed catheters, as reported by Taulelle and colleagues (1998). Weigel and Martini (2000) recently reviewed these and other endobronchial approaches, such as laser, electrocautery, and cryoablation (see also Chapter 108). In most cases, however, the depth of invasion cannot be determined with certainty, and some have associated lymph node involvement. Resection, therefore, remains the most common approach. Although occult NSCLC has always constituted a small subset, improved imaging, the increasing incidence of parenchymal adenocarcinoma, as opposed to central airway squamous cancers, and the essential abandonment of sputum screening in the United States have decreased the number of cases still further. In Japan, in contrast, where sputum screening has persisted, this entity represents a small but continuing staging and treatment challenge.

Peripheral Tumors

The major considerations in the surgical evaluation of a peripheral tumor are its location within the lobe and its relationship to other structures. Lesions that are clearly surrounded by parenchyma and confined to a single lobe are treated by lobectomy and occasionally by lesser resections. If a tumor abuts the chest wall on CT, the possibility of pleural invasion should be entertained, especially if the patient has associated pain. If the tumor approximates an interlobar fissure, the possibility of extension into an adjacent lobe should be considered. In all such cases, the operation should be planned and discussed with the patient by including the possibility of chest wall resection, bilobectomy, or pneumonectomy, as appropriate. When peripheral tumors invade other structures, en bloc resection is often necessary.

Chest Wall.

Tumors invading the chest wall are often resectable. The involved ribs should be transected several centimeters beyond the margin of gross involvement. In most cases, one rib and intercostal tissue above and below the tumor should also be included in the resection. Chest

P.1556


wall reconstruction is carried out as needed to prevent physiologic impairment due to paradoxical chest wall function or for cosmetic reasons (see Chapter 47). For posterior defects, support by the remaining chest wall muscles and scapula is usually sufficient, whereas anterior and lateral defects more often require reconstruction. Although full-thickness resection is mandatory for tumors invading the osseous and muscular structures of the chest wall, there is controversy regarding the necessity of chest wall resection when invasion is confined to the parietal pleura. McCaughan and associates (1985) reported good results in such cases treated by the development of an extrapleural dissection plane when possible, stripping away the lung and parietal pleura from the endothoracic fascia, and proceeding to a full-thickness resection only if the margin was positive on frozen section. Albertucci and associates (1992), in contrast, found that a histologic complete resection was achieved in only one third of their patients treated by extrapleural dissection, as compared with all of those undergoing a standard chest wall resection. The experience of Trastek and colleagues (1984) also suggests that chest wall resection is preferable even when invasion is confined to the parietal pleura. We agree with Ratto and associates (1991), who suggest that when a tumor is firmly affixed to the parietal pleura, no attempt be made to strip it away, and that an en bloc resection be carried out. However, extrapleural dissection may be appropriate in patients in whom the risk for an extended resection is high, in those treated with preoperative radiation, or if intraoperative brachytherapy is applied, although the overall role of external and interstitial irradiation for fully resectable chest wall T3 NSCLC is unknown.

Diaphragm.

As noted by Rocco and associates (1999) and by Riquet and colleagues (2000), the diaphragm is rarely involved by direct extension of NSCLC, despite the large area of contact between this structure and the base of the lung. When invaded, it should be resected with a wide margin of normal tissue, without regard to the extent of the defect. Although unlikely to be helpful for NSCLC, it is feasible to resect and replace an entire hemidiaphragm. Unless the defect is small and can be closed primarily without tension, it should generally be replaced with a prosthetic material. Alternatively, a variety of muscle flaps can be used. When a large area of diaphragm has been resected or when the phrenic nerve has been sacrificed, it is important that the diaphragm be reconstructed near the position of full inspiration to avoid paradoxical motion. When the defect is peripheral, it may be possible to reinsert the remaining cut edge at a higher level on the chest wall and thereby obviate the need for prosthetic material, as described by Daly and Feins (1998).

Pericardium.

Total resection of the pericardium on the left can be performed without reconstruction. Partial defects should be closed to prevent herniation and strangulation of the left ventricle. On the right side, all pericardial defects, regardless of size, require repair. The potential problem if the pericardium remains open following right pneumonectomy is torsion of the heart into the hemithorax along the axis of the venae cavae, with consequent total occlusion of venous inflow. Large defects can be closed with the pericardial fat pad, a pleural flap, or nonautologous material such as bovine pericardium or polytetrafluoroethylene (PTFE). Many surgeons suggest that a small opening be left in the repair or that the prosthetic material be fenestrated to prevent intrapericardial fluid accumulation and subsequent cardiac tamponade.

Vertebrae.

Tumors invading the vertebral bodies are rarely cured and under most circumstances are considered unresectable. DeMeester and colleagues (1989) described a technique of partial vertebral resection for tumors fixed to the paravertebral fascia, involving a tangential osteotomy through the transverse process, costotransverse foramen, and superficial vertebral body (Fig. 106-1). The authors emphasize that this approach is not suitable for patients with radiographic evidence of bone destruction. Grunenwald and associates (1996), however, reported a small group of patients with radiographic evidence of osseous invasion who were treated by en bloc pulmonary resection and complete vertebrectomy with reconstruction by a combined thoracic and posterior approach. In a subsequent report (2002), this group had a 14% late survival in 19 patients treated by partial or total vertebrectomy. This technique should be limited to rare cases in which the tumor extent is completely delineated, node negative, and totally resectable, and, after careful evaluation with MR imaging, does not involve the spinal canal.

Superior Sulcus Tumors.

Resection of peripheral tumors involving the apex of the chest and the lower portion of the brachial plexus (superior sulcus or Pancoast tumors) is discussed in Chapters 35 and 36. Although they invade the chest wall, these challenging lesions are viewed as a distinct entity because of their unique clinical, anatomic, and surgical features. When a superior sulcus tumor is deemed potentially resectable at presentation, operation is preceded by induction therapy, traditionally by radiation therapy but increasingly by chemotherapy as well. Resection involves removal of the involved portions of the apical chest wall, typically including the first, second, and third ribs, and the T1 nerve root, along with lobectomy. A lesser pulmonary resection is now considered inadequate by most surgeons. Although invasion of the vertebral body or of the subclavian vessels, clinical N2 disease, and the presence of a Horner's syndrome generally contraindicate primary operation, a more aggressive approach may be indicated in selected cases, especially after induction therapy. These tumors can be resected through a high posterolateral thoracotomy or by an anterior approach using a cervical incision with extension down to the second intercostal space and resection

P.1557


of the first and second costal cartilages. Dartevelle and colleagues (1993) have used the anterior approach alone to perform extensive vascular resection along with lobectomy in this setting.

Fig. 106-1. Technique of removing part of the vertebral body when a tumor has become attached to the overlying parietal pleura. The costotransverse foramen must be free of disease. A C. Sequential steps in resection. From DeMeester TR, et al: Management of tumor adherent to the vertebral column. J Thorac Cardiovasc Surg 97:373, 1989.

Central Tumors

Central tumors are more likely to be associated with malignant lymphadenopathy and to involve mediastinal structures. Accordingly, careful imaging and invasive evaluation is mandatory before consideration of thoracotomy. By definition, complete resection requires at least a lobectomy and often requires a pneumonectomy or more extended procedure. Central bronchial T3 lesions and some T4 cancers involving the carina can be treated successfully by primary operation, in the latter situation usually by a carinal pneumonectomy. The selection, techniques, and results of resection in this setting are discussed in Chapters 28 and 30. The decreasing mortality rate associated with these extensive procedures has been noted previously. Infrequently, for a small lesion in the left main-stem bronchus, a localized bronchial sleeve resection with pulmonary conservation can be carried out, as reported by Newton and associates (1991). Localized tumors involving the pulmonary veins (even with extension into the pericardium and left atrium) may be amenable to resection by excision of a contiguous cuff of the atrium, using vascular clamps and sutured closure or vascular staplers. Central NSCLC with local invasion limited to the mediastinal pleura and adipose tissue, but not involving deeper structures, is also often suitable for total resection. With exceedingly rare exceptions, in contrast, tumors invading the superior vena cava or the aorta or its branches should not be addressed by primary operation, but should be considered for resection in a few cases only after postinduction reassessment. It cannot be overemphasized that, in all cases of locally invasive NSCLC considered for resection, the absence of N2 lymphadenopathy should be confirmed by rigorous preoperative staging.

Simultaneous Cardiac Operationand Pulmonary Resection

When a patient requires myocardial revascularization or other cardiac procedure and also has a resectable lung cancer, the question of simultaneous versus staged procedures arises. This clinical situation can arise during the physiologic assessment of a lung cancer patient being considered for resection or in the preoperative radiographic evaluation of a cardiac patient. When the lung cancer can be resected through a median sternotomy, the timing of the procedures is largely a matter of the surgeon's preference and patient-specific factors. The experiences reported by Canver (1990), Terzi (1994), and Danton (1998), along with their colleagues, support the safety and efficacy of simultaneous operations. The cardiac procedure should be performed first and without complications. Pulmonary resection is carried out after reversal of anticoagulation and confirmation of hemodynamic and hemostatic stability. Generally, a lobectomy is carried out, although pneumonectomy has been reported by Piehler (1985) and Danton (1998) and their associates. Because cardiac retraction required during transsternal left lower lobectomy may cause hemodynamic problems, most left lower lobe tumors should be resected at a separate session. Limited resections should be used only with the appropriate indications. In all cases except emergent cardiac surgery, a full staging evaluation of the lung tumor should be carried out before operation. Combined procedures have the advantage of a single operation and recovery as well as absence of delay in cancer treatment. However, concerns have been raised about the adverse oncologic effect of immunosuppression and the possibility of

P.1558


tumor dissemination associated with cardiopulmonary bypass, if the cardiac operation is performed before or concurrently with the pulmonary resection.

Synchronous Lung Cancers

Although the incidence of a second primary lung cancer in patients previously treated for NSCLC is more than 10%, as discussed by one of us (RBP) (2000), the prevalence of synchronous primary lung cancers has generally been placed at about 1% of surgical cases. There is evidence, however, that the incidence of this difficult therapeutic problem is increasing. In addition, the nature of multiple primary sites of involvement appears to have changed coincident with the rise in the proportion of adenocarcinomas among NSCLC patients. Multiple sites of parenchymal adenocarcinoma at presentation are now more commonly encountered than multiple sites of squamous cancer in the airways. Both the diagnosis of synchronous lesions, as opposed to metastasis, and the optimal surgical and nonsurgical treatment of such lesions represent a challenge. The criteria for differentiating between synchronous primary and metastatic lesions established by Martini and Melamed (1975) have traditionally been accepted by most surgeons. For lesions to be considered synchronous primaries, they must be physically distinct and separate. The histology must be different or, if similar, the neoplasms must have an origin from carcinoma in situ, or be located in different pulmonary segments, and have no carcinoma in the lymphatic vessels and nodes common to both lesions and no extrapulmonary metastasis. Antakli and associates (1995) modified the criteria for tumors of similar histology, regarding them as separate primaries if two or more of the following five conditions were met: (a) anatomically distinct, (b) presence of associated premalignant lesions, (c) absence of systemic metastasis, (d) no mediastinal disease, and (e) different DNA ploidy. Ichinose and colleagues (1991) reported on the use of flow cytometry to differentiate metastatic from synchronous lesions and recurrent NSCLC from a metachronous new primary.

As reviewed by Adebonojo and associates (1997), the incidence of multiple NSCLC in surgical series varies widely (from 0.8% to 14%). Ohada and associates (1998) noted a 3.2% incidence of synchronous cancers among 908 consecutive resections. McElvaney and colleagues (1989) reported the disturbing finding that, on careful pathologic examination, 12 of 62 consecutive resections for adenocarcinoma (19%) were found to have multiple sites of cancer that met the criteria for separate primary lesions. In only two cases were the additional lesions suspected preoperatively, despite CT scanning in all cases.

In practice, resection for synchronous primary NSCLC is largely confined to patients with two lesions. The presence of three or more sites strongly suggests metastatic disease or, at least, a patient unlikely to benefit from primary operation. Selection of the appropriate surgical approach depends on the size and location of the tumors and the patient's pulmonary function. In general, anatomic resection should be carried out, especially for the larger or more central tumor. The specifics of each case will dictate whether sternotomy with bilateral resections or staged thoracotomies are preferable when the lesions are bilateral. Although extensive anatomic parenchymal resection, including pneumonectomy, may be indicated when there is a high degree of certainty that the tumors are separate primaries without lymph node spread, the difficulty in distinguishing intrapulmonary metastasis often makes lesser resection of the smaller lesion prudent. Similar guidelines should be followed when dealing with metachronous NSCLC, although surgical options for new cancers following prior pneumonectomy may be limited, as noted by Westermann and associates (1993).

Incomplete Resection

Incomplete resection of NSCLC is variously defined, including macroscopic disease left behind, microscopic positive margins, involvement of the highest resected mediastinal lymph node, and the presence of remaining intrapulmonary or distant metastases. Grossly incomplete resection of NSCLC does not result in long-term survival. With current staging methods, grossly complete resection should be achieved in 95% or more of operated patients. When a patient is found at thoracotomy to have disease that cannot be removed completely, resection rarely should be carried out and only if it can be accomplished with low risk and with the expectation that it will prevent or reduce the likelihood of complications in the future. Except under very unusual circumstances, such as infection, bleeding, or pain that cannot be palliated by other means, which is rare with the therapeutic modalities available at the present time, should an incomplete resection be intentionally considered. As shown early on by Hara and associates (1984), patients with advanced NSCLC who have an incomplete resection (whether of the primary tumor or metastatic lymph nodes) have a clinical course identical to those who undergo nonresective exploration or no operation (Fig. 106-2). Furthermore, patients who undergo incomplete resection may have a poorer quality of remaining life than those treated by other modalities or by supportive care. Kimura and Yamaguchi (1994) found a range of 5-year survival when 279 incomplete resections were classified according to the site of residual cancer. When incomplete resection involved the chest wall, survival was 29%, versus only 5% for residual disease in lymph nodes. Others have found lower survival for incompletely resected chest wall cases. In addition, many of the patients in this study received adjuvant radiation, chemotherapy, and immunotherapy. Lacasse and colleagues of the Canadian Lung Oncology Group (1998) found that histologic positivity of surgical margins or of the highest resected

P.1559


lymph node did not predict clinical recurrence, but follow-up was only 3 years.

Fig. 106-2. Survival curves of patients with stage III carcinoma of the lung after palliative resection, exploratory thoracotomy only, or no surgical therapy. From Hara N, et al: Assessment of the role of surgery for stage III bronchogenic carcinoma. J Surg Oncol 25:153, 1984.

Much attention has been paid to the prognostic implications of residual tumor at the bronchial resection margin. Many such cases involve squamous histology. Shields (1974) described three categories of positive bronchial margins: (a) gross disease, (b) microscopic tumor in the peribronchial tissue, and (c) microscopic mucosal or submucosal residual cancer. Although the first two findings were found to portend a poor prognosis, late survival without adjuvant therapy occurred in about one fourth of patients in the third group. Similar findings were reported by Soorae and Stevenson (1979). More recently, however, Gebitekin and associates (1994) noted that a microscopically positive bronchial or peribronchial margin adversely affected survival in N2 NSCLC, no survival versus 17% 5-year survival for N2 with a clear margin. Snijder and associates (1998) found that for stage I NSCLC, either a peribronchial or mucosal margin positive for invasive cancer was associated with late survival of 27% versus 58% for carcinoma in situ or a negative margin. Massard and colleagues (2000) reported a 55% 5-year survival rate for in situ carcinoma, as compared with 20% for microscopic invasive mucosal or peribronchial residual cancer. Ghiribelli and co-workers (1999) likewise found no difference in the negative impact on survival between mucosal and extramucosal involvement. Passlick and associates (2001) emphasized the importance of lymphatic invasion as a negative factor regardless of the bronchial or extrabronchial location of positivity.

Because of these findings, intraoperative frozen section assessment of the airway margin should be employed frequently and a more proximal resection carried out if indicated. When an involved margin is detected only on final pathologic review (especially when lymphatic invasion and lymph node disease are absent), reoperation to achieve more proximal airway resection should be considered, as suggested by Kaiser (1989), Liewald (1992), Snijder (1998), and Ghiribelli (1999) and their associates. Many patients, however, will not be candidates for this aggressive approach. Although external radiation therapy has not been shown to prolong survival in the overall spectrum of incompletely resected NSCLC, adjuvant radiation therapy should be considered because local control may be enhanced. The role of brachytherapy is also unclear because there are no randomized studies and experience is limited. Although the risk associated with this method when isotopes are implanted in tissues such as the chest wall is minimal, its use in grossly normal bronchial margins and other vital structures for microscopic positivity may result in necrosis and its sequelae.

Results of Surgical Treatment

The results of surgical treatment of NSCLC are assessed based on the risk for death or significant complications and the long-term survival realized after resection, as measured by prolongation of life or by cure. The acute risk is dependent mainly on patient-specific factors, such as comorbidity and age. Tumor-related factors are important insofar as they dictate the extent of the required resection.

Long-term survival is determined by multiple factors. At present, the most valuable predictors are the tumor stage and the tumor node metastasis (TNM) subsets within each stage. As emphasized, the ability to accomplish a complete resection is also paramount and largely depends on the tumor stage. In assessing surgical results, the concept of clinical stage versus pathologic stage must be clearly appreciated (see Chapter 105). The clinical stage is based on a synthesis of all invasive and noninvasive studies short of resection. The pathologic stage is determined from the resected tissue and may or may not coincide with the clinical stage. Surgical decisions are made based on the clinical stage, but most surgical reporting is based on the pathologic stage. Bulzebruck and associates (1991) and Fernando and Goldstraw (1990) found that clinical and pathologic stage coincided in only 55% and 47% of cases, respectively. Clinically

P.1560


unsuspected node involvement was found in 23% of patients. In a prospective study comparing clinical staging by CT and mediastinoscopy with pathologic stage, Gdeedo and colleagues (1997) found that the T factor was overstaged in 27% and understaged in 18% of cases, whereas adenopathy was clinically overstaged by imaging in 45% and understaged in 20% of patients.

The vast literature on the surgical treatment of NSCLC is confusing. A few general principles for assessing individual reports may facilitate comparison. Of obvious importance is the time period encompassed, not only because of variations in imaging, but also because of evolution of surgical techniques and philosophy. It is similarly imperative to be clear regarding the clinical staging methods applied. Of utmost importance is recognition of the subset of patients on which the conclusions are based (whether the entire group, all operated patients, or only those who had a complete resection). Some series eliminate from survival calculations cases of operative mortality or late noncancer deaths. Some papers include only 30-day operative mortality, a clearly inadequate assessment of acute risk in the modern world of intensive care. Perspective may be enhanced by noting the actual number of long-term survivors with respect to the initial number of cases, in addition to the actuarial probability of survival. Although the latter is a valid figure for statistical reporting, the methodology is such that a numerically small surviving cohort may yield a surprisingly favorable actuarial probability of survival. Another variable requiring attention is the use of preoperative and postoperative adjuvant treatments that in many surgical reports are not factored into the conclusions. Similarly, it is very rare for surgical reports to include all patients who present with a given stage, especially when advanced, but rather only those selected for operation. Before sweeping conclusions can be drawn, analysis must view the presented experience in the overall spectrum of lung cancer. Reports may also not be useful because of lack of specificity about staging, combining various TNM subtypes, or presentation of the survival data in a nonstandard form. Also, the conclusions drawn by the authors of a report may not be supported by the data.

Table 106-1. Five-Year Survival by TNM Classification and by Stage According to the 1997 System

TNM Stage Mountain (1997) Van Rens et al (2000) Naruke et al (2001)
N 5-yr (%) N 5-yr (%) N 5-yr (%)
T1N0M0 IA 511 67 404 63 245 75
T2N0M0 IB 549 57 797 46 291 57
T1N1M0 IIA 76 55 83 52 66 52
T2N1M0 IIB 288 39 642 43 153 38
T3N0M0 IIB 87 38 337 19 106 33
T3N1M0 IIIA 55 25 NR NR 85 39
T1 3N2M0 IIIA 344 23 NR NR 368 15
T1 3N3M0 IIIB 572 3 NR NR 55 0
T4, any N, M0 IIIB 458 6 NR NR 104 8
Any T, any N, M1 IV 1427 1 NR NR 293 7
N, number of cases; 5-yr, percentage survival at 5 years: NR, not reported.
Note: Classification is postsurgical except in Mountain's IIIB and IV, which is clinical.
From the databases reported by Mountain CF: Revisions in the international system for staging lung cancer. Chest 111:1710, 1997; by van Rens MTM, et al: Prognostic assessment of 2,361 patients who under went pulmonary resection for non-small cell lung cancer Stage I, II, and IIIA. Chest 117:374, 2000; and by Naruke T, et al: Prognosis and survival after resection for bronchogenic carcinoma based on the 1997 TNM staging classification: the Japanese experience. Ann Thorac Surg 71:1759, 2001.

This discussion focuses on surgical results based on TNM staging and simple anatomic factors. Brief reviews of other prognostic factors, including biochemical and molecular biologic markers, patient demographics, and the recent interest in certain radiographic features as prognostic indicators, are also presented. Late results are presented as 5-year actuarial survival, unless otherwise specified. Although the new staging system proposed by Mountain (1997) remains the subject of minor controversy, and most extant data are reported according to the prior literature, the material is presented according to the new system. Table 106-1 presents survival by stage and TNM subsets from the reports of Mountain (1997), Naruke and associates (2001), and van Rens and colleagues (2000). These three series were selected as representative because of their large numbers of patients, their span of three decades and three continents, and their presentation of the material in a similar format. Many other valuable reports are noted in the text and the tables. Reports that are unclear or present the material in a nonstandard fashion, without a cogent alternative point, are not included.

Occult Non Small Cell Lung Cancer

Although occult NSCLC does not constitute a specific stage, many of the least advanced tumors are detected in this manner, and the results of resection in this group are excellent. Most occult cancers are squamous cell lesions, and many involve the large airways. Saito and colleagues (1992) reported 94 patients with occult squamous lung cancer who

P.1561


were treated by resection. Although all were clinical stage I, pathologic staging showed 16 TisN0 (stage 0), 72 T1N0 (stage IA), and 6 stage IIA and B (4 T1N1 and 2 T2N1). The 5-year actuarial survival rate overall was 80%, and when noncancer deaths were eliminated, it was 93%. Cortese and associates (1983) reported on 54 patients and postresection staging, including 19 cases of TisN0, 25 T1N0, 5 T1N1, 4 T2N1, and 1 T3N0. All were squamous cancers, some with large cell features. The overall 5-year survival rate was 74%, with a 90% survival rate when only lung cancer deaths were included. The survival rate was 91% for the TisN0 and T1N0 patients. Likewise, Martini and Melamed (1980), in a series of 47 cases, reported a high incidence of low-stage cancers and a high surgical cure rate. Watanabe and co-workers (1991c) treated 20 patients with occult central cancers, mainly by bronchoplastic resections, and noted a 10-year survival of 92%. Among 98 cases of centrally located occult cancers reported by Koike and associates (2000), most underwent lobectomy or bilobectomy, pathologic stage included 5 TxN0, 19TisN0, 63 T1N0, and 11 T2N0, and the late survival rate was 81% overall and 89% for in situ lesions. Sagawa and colleagues (2001) performed segmentectomy for suitable, more peripheral radiographically occult cancers in 58 patients. The 5-year survival rate in this series was 83% overall and 97% when noncancer deaths were excluded. Most reports, however, note a disturbing incidence of synchronous and metachronous multiple aerodigestive squamous cancers in this patient population. Saito and associates (1992) noted an incidence of 9.6% synchronous and 7.4% metachronous lesions; Martini and Melamed (1980) ultimately found new cancers in 45%; and Cortese and colleagues (1983) calculated an incidence of 5% per postoperative year, a rate that remained steady over time.

Table 106-2. Selected Series Reporting Postresection 5-Year Survival Rates for Pathologic Stage IA (T1N0) and Stage IB (T2N0) Non Small Cell Lung Cancer

Report Dates T1N0 IA T2N0 IB
N 5-yr (%) N 5-yr (%)
Williams et al (1981) 1972 1978 225 80 236 62
Little et al (1986)a 1974 1984 44 72 47 68
Martini et al (1986) 1973 1976 50 83 78 65
Roeslin et al (1987) 1977 1982 108 71 121 43
Read et al (1990) 1966 1988 214 73 327 57
Shimizu et al (1993) 1973 1989 288 total 74 54
Ichinose et al (1993) 1981 1988 71 85 80 67
Padilla et al (1997)a 1969 1993 109 76 45 78
Mountain (1997) 1975 1988 511 67 549 57
Inoue et al (1998) 1980 1993 480 80 271 65
Jassem et al (2000) 1991 1995 51 66 220 53
Tanaka et al (2001) 1980 1994 208 79 179 66
van Rens et al (2000) 1970 1992 404 63 797 46
Naruke et al (2001) 1961 1995 786 71 506 60
Fang et al (2001) 1961 1995 126 72 702 61
Rena et al (2002) 1993 1999 144 66 292 55
Thomas et al (2002) 1990 1999 147 74 368 55
N, number of cases; 5-yr, percentage survival at 5 years.
a Series limited to complete resection and small tumors.

Stage I: Localized Node-Negative Non Small Cell Lung Cancer

Because of an often striking and usually statistically significant difference in survival between tumors 3 cm or less that are entirely parenchymal (T1) as opposed to those that are larger, invade the visceral pleura, are located in a main bronchus more than 2 cm from the carina, or cause lobar atelectasis (T2), the 1997 system divides stage I into two subsets. Postsurgical 5-year survival rates in the T1N0 group (stage 1A) are uniformly excellent, whereas postresection survival in IB disease is less satisfactory. In both categories, however, the efficacy of resection is currently not matched by any other treatment modality. In addition, there is presently no evidence that additional therapy as induction or adjuvant offers benefit over resection alone, although clinical trials are in progress.

In the databases of Mountain (1997) and those of van Rens (2000) and Naruke (2001) and their colleagues, 5-year postsurgical survival rates for pathologic stage IA cases are 67%, 63%, and 75%, respectively; and in the latter two reports, 65% and 71%, respectively, for clinical IA tumors. Table 106-2 shows that surgeons worldwide have confirmed these results, and some have reported higher survival rates in cooperative and single-center series spanning three decades. The range of reported late survival is from 63% to 85%. The LCSG noted a median survival of 8 years for 907 pT1N0 cancers resected between 1977 and 1988, as reported by Thomas and colleagues (1990). Pairolero and associates (1984) reported their experience as rates of recurrence at 5 years (29% for T1N0 and 40% for T2N0).

Factors other than TNM descriptors that have variously been noted to affect survival in cases of limited, node-negative

P.1562


NSCLC include tumor size, cell type, and location. Among these variables, the size of the primary lesion has most consistently, although not universally, been found to influence the prognosis. Read and associates (1988, 1990) noted significantly better survival for T1N0 tumors 2 cm or smaller than for T1 lesions between 2 and 3 cm. Similarly, Ishida and co-workers (1990a) found a less favorable prognosis for tumors 2 to 3 cm as compared with those smaller than 1 cm, but no difference between T1N0 cancers 1 cm or less as compared with 1.1 to 2 cm (Fig. 106-3). Padilla and colleagues (1997) reported markedly better late survival for smaller diameter pT1N0 tumors (87% at 5 years and 74% at 10 years for 2 cm or less versus 65% at 5 years and 49% at 10 years for 2.1 to 3 cm cancers). Koike and associates (1998) have shown that the incidence of pathologic N2 adenopathy rises with increasing tumor diameter in cases clinically staged as T1N0. Suzuki and colleagues (2001) also noted a significantly higher rate of postresection upstaging for clinical T1 lesions equal to or less than 2 cm versus those 2 to 3 cm. Although the prognosis for patients in the latter two studies would be based on their postresection stage (pN2), Ohta and colleagues (2001) found an increase in nodal micrometastases with increasing size of squamous cancers in samples retrospectively analyzed by immunohistochemical staining that were negative by routine histology. No difference was found for nonsquamous lesions. In a study of 1,020 resected stage IA and IB cancers grouped by size, Lopez-Encuentra and colleagues (2002) noted significant survival differences among four groups. Five-year survival rates were 63%, 56%, 49%, and 38%, for tumors smaller than 2 cm, from 2.1 to 4 cm, from 4.1 to 7 cm, and larger than 7 cm, respectively. In contrast, Patz and co-workers (2000) found no difference in late results in T1N0 NSCLC based on tumor size divided either into quartiles or deciles.

Some series note a difference in late survival based on cell type, with T1N0 squamous histology usually conferring an improved prognosis over adenocarcinomas, large cell lesions, and mixed cancers of comparable extent. Gail and associates (1984), reporting for the LCSG, showed significantly lower recurrence and death rates per patient per year for squamous versus nonsquamous or mixed histologies, noting that at 3 years, 90% of resected patients with T1N0 squamous cancer were free of recurrence, as opposed to only 62% of those with nonsquamous histology. Similarly, Read and colleagues (1988) concluded that 5-year postresection survival is more often realized for T1N0 squamous tumors than for comparable adenocarcinomas. Macchiarini (1993) and Kodama (1997) and their associates, in contrast, did not detect a significant difference in outcome for T1N0 disease based on the subsets of non small cell histology. Many other series, including those reported by Shields (1975), Pairolero (1984), Little (1986), Ichinose (1993), Harpole (1995a, 1995b), and Fang (2001) and their associates, have also not established a statistically significant difference in survival based on squamous versus nonsquamous histology, although absolute long-term survival is often slightly higher in squamous cases. Occasional reports, such as that of Rena and colleagues (2002), note better survival for stage I adenocarcinomas and for squamous cancers. Given these conflicting findings, it is likely that within the common categories of NSCLC, histology does not directly affect surgical results. Certain less common cell types, however, appear to influence postresection prognosis. Although variously subclassified, large cell neuroendocrine carcinoma (LCNEC) and similar types of NSCLC are associated with a poorer prognosis than other histologies, as noted by Iyoda (2002) and Takei (2002) and their colleagues. In the former series, the 5-year survival rate was 67% for resected stage I cases versus 88% for stage I NSCLC of other histologies. Garcia-Yuste and co-workers (2000) noted 5-year survival in only 33% of resected stage I LCNEC patients. Hiroshima and associates (2002) found a lowered survival rate in resected stage I adenocarcinoma patients when 10% or more of the cells showed neuroendocrine differentiation. The LCSG, however, as reported by Linnoila and associates (1994), did not find a significant influence of such differentiation. Also, Zacharias and colleagues (2003) noted a survival comparable to other cell types (88%) in stage I LCNEC. Similarly, some authors, such as Riquet and colleagues (2001) and Ruffini and associates (2002b), report that adenosquamous cancers are associated with a poorer prognosis than other NSCLC following resection of stage I cases, whereas other series, such as those of Ishida (1992) and Hsia (1999) and their associates, found no difference. The surgical prognosis for patients with node-negative mixed small cell and non small cell lesions is considerably worse than for pure NSCLC. Hage and colleagues (1998) reported 5-year survival rates of 50% and 26% for mixed tumors of pathologic stage IA and IB, respectively, many of which were also treated with preoperative or postoperative chemotherapy. Resection for unifocal bronchoalveolar carcinoma (BAC) has generally been found to be associated with a favorable prognosis. Although the current World Health Organization

P.1563


definition of BAC has been narrowed to include only noninvasive lesions composed entirely of BAC, thoracic surgeons have long noted that patients with early-stage pure BAC, mixed BAC and adenocarcinoma, or adenocarcinoma with BAC features fare well with resection. Higashiyama and colleagues (1999) classified 206 resected peripheral adenocarcinomas smaller than 2 cm into four groups based on the extent of any BAC component and found a correlation between increased percentage of BAC and survival, with a 100% 5-year survival rate in stage I pure BAC patients. Carretta and co-workers (2001) also found a favorable effect of increasing BAC histology in mixed lesions, with 5-year survival rates of 80% in stage IA and 68% in stage IB patients. Combining both subsets and grouping tumors with more than 50% BAC versus those with 50% or less BAC yielded a significant difference, with 100% late survival in the former group as contrasted with 60% in the latter. Similarly, Breathnach and colleagues (2001) reported an 83% survival rate for stage I BAC versus 63% for adenocarcinoma, and noted a range of 56% to 83% in a literature review of stage I BAC. Ebright and associates (2002) found no difference in survival in a mixed-stage group for pure BAC, BAC with focal invasion, and adenocarcinoma with BAC features, but did find a significant impact of stage, with survival of 84% for stages I and II and 60% for stages III and IV.

Fig. 106-3. Survival curves of T1N0 lesions based on size of the lesion. From Ishida T, et al: Strategy for lymphadenectomy of lung cancer less than 3 cm in diameter. Ann Thorac Surg 50:708, 1990a.

In addition to cell type, other routinely reported histologic features of NSCLC likely correlate broadly with postresection prognosis in stage I cases. Increasing degrees of vascular invasion and perineural involvement appear to portend a less favorable prognosis. Higher levels of cellular differentiation or grade, lymphocytic response, and scar formation have often been found to be associated with longer survival following operation. As for basic histology, however, none of these factors appears to be determinative in low-stage cancers, and they are not currently used to determine stage or therapy. Brundage and associates (2002) have recently reviewed this area.

The location of the primary lesion may influence the results of resection. Peripheral T1N0 tumors have a better prognosis than central lesions. Although Read and colleagues (1990) found that nonsquamous T1N0 patients fared worse overall after resection than did squamous cases, the difference was largely confined to central cancers. When the tumor did not communicate with a segmental or subsegmental bronchus, the effect of cell type was negated. The presence of airway communication for both cell types conferred a significantly poorer outlook, dramatically so for nonsquamous histology (Fig. 106-4). Likewise, the influence of peripheral versus central location negated the effect of tumor size in the experience reported by Sagawa and colleagues (1990). Assessing only NSCLC arising in or peripheral to a subsegmental bronchus, they recorded similar 5-year survival rates for smaller or larger T1N0 lesions (80% for tumors 2 to 3 cm in diameter and 83% for smaller cancers). The favorable influence of parenchymal location is also emphasized by the report of Kodama and associates (1997), who achieved a 5-year survival rate of 93% in 77 patients with peripheral pT1N0 NSCLC treated by lobectomy and 88% in 46 cases following intentional lesser resection.

Fig. 106-4. T1N0 lesions that communicate with a segment of subsegmental bronchus have a worse prognosis than more peripheral tumors in both squamous and nonsquamous cases. Bronchial nonsquamous cancers do particularly poorly. From Read RC, Yoder G, Schaeffer RC: Survival after conservative resection for T1N0M0 non small cell lung cancer. Ann Thorac Surg 49:391, 1990.

In the databases of Mountain (1997), as well as those of van Rens (2000) and Naruke (2001) and their colleagues, postresection survival for pathologic T2N0 lung cancer is 57%, 46%, and 60%, respectively. Survival rates fall to 38% and 42% in the latter two reports for clinical stage IB NSCLC. It is striking that the most optimistic upper limit of late survival for stage IB disease in the series listed in Table 106-2 generally falls below the lower figures for stage IA lesions. As for stage IA, however, some authors report better results, but still only slightly more than 60% late survival. The 78% survival rate reported by Padilla and colleagues (1997) is confined to small lesions.

Specific variables within stage IB that influence prognosis have not been assessed as extensively as have those for IA disease because the subsets are new and most reports have focused on T1N0 cases. It seems reasonable to assume that the basic factors noted earlier for T1N0 NSCLC generally apply in higher stages and substages, but that their influence diminishes with advancing tumor extent. Larger tumor diameter and nonsquamous histology may correlate with a poorer prognosis, but the data are sparse. Although noting a highly significant adverse effect of increasing size within stage IA, Read and associates (1988) found no significant difference between cancers equal to or smaller than 5 cm versus those larger than 5 cm in stage IB cases. Harpole and colleagues (1995a), in contrast, found that larger tumors were associated with a more limited survival than smaller cancers, but the groupings were 2 to 4 cm versus tumors larger than 4 cm; hence, some cases were likely stage IA. Carbone and associates (2001) noted a statistically significant difference in 5-year survival rates for resected

P.1564


T2N0 cancers between 3 and 5 cm versus those larger than 5 cm: 62% and 51%, respectively. These authors suggested that NSCLC tumors larger than 5 cm be upgraded to T3.

Involvement of a main bronchus further than 2 cm from the carina or a lesion causing lobar atelectasis, both defining features of some T2 primaries, are not clearly negative prognostic indicators. In fact, central T2 cancers limited to the bronchial wall (often occult), as noted by Naruke (1988) and Watanabe (1991a) and their colleagues, have an excellent postresection survival. In contrast, direct invasion of the visceral pleura (another defining feature of some T2 lesions) clearly correlates with a poorer prognosis than comparable T2 cancers without pleural involvement, as noted early by Brewer (1977) and also by Merlier and colleagues (1985). Harpole and colleagues (1995a) found that visceral pleural invasion was a significant predictor of poor outcome, with 5-year survival rates of 44% for invasive lesions versus 67% for those without pleural transgression, and respective 10-year survival rates of 62% and 37%. Ichinose and colleagues (1993) reported similar findings. Manac'h and associates (2001) noted 5-year postresection survival in 61% of stage I patients without pleural involvement versus 46% of those with pleural invasion. The LCSG, reported by Gail and associates (1984), noted a 1.66-fold increase in recurrence rates in stage I patients when the visceral pleura was involved by cancer.

The significance or lack thereof of occult micrometastases in the blood, bone marrow, or intrathoracic lymph nodes in stage I disease, as determined on standard microscopic examination, is discussed in Chapter 101.

Table 106-3. Selected Series Reporting Postresection 5-Year Survival Rates for Pathologic Stage IIA (T1N1) and Stage IIB (T2N1) Non Small Cell Lung Carcinoma

Report Dates T1N1 IA T2N0T IIB
N 5-yr (%) N 5-yr (%)
Williams et al (1981) 1972 1978 30 50
Holmes (1987) NS NS 75 squamous
52 adenocarcinoma
NS 53 squamous
25 adenocarcinoma
Roeslin et al (1987) 1977 1982 50 45 All: 113
Lobar: 79
Hilar: 34
37
41
14
Martini et al (1992) 1973 1989 35 40 179 38
van Velzen et al (1996, 1997) 1977 1994 All: 57
Lobar: 25
Hilar: 32
46
57
23
All: 369
Lobar: 57
Hilar: 161
38
57
30
Mountain (1997) 1975 1988 76 55 288 39
Inoue et al (1998) 1980 1993 57 57 141 42
Riquet et al (1999) 1984 1993 64 49 125
Lobar: 102a
Hilar: 154a
47
54
38
van Rens et al (2000) 1970 1992 83 52 510 33
Jassem et al (2000) 1991 1995 6 17 64 27
Tanaka et al (2001) 1976 1997 32 62 63
Lobar 69a
Hilar 26a
56
66
39
Naruke et al (2001) 1961 1995 111 57 263 44
Fang et al (2001) 1961 1995 32 33 499 34
Marra et al (2003) 1990 1995 67 65 287 45
N, number of cases; 5-yr, percentage survival at 5 years; NS, not stated.
a Includes both T1N1 and T2N1.

Stage II: Non Small Cell Lung Cancer with N1 Adenopathy or Resectable Local Invasion

The 1997 system divides stage II lung cancer into two subsets. Stage IIA is limited to T1N1M0 lesions, whereas IIB includes both node-positive (T2N1M0) and node-negative (T3N0M0) cases. Stage IIA and T3N0 stage IIB include small numbers of patients. Stage IIA represents a very small proportion of patients based on clinical staging. Pathologic T1N1 is more common but remains a small group. The 5-year survival rate for patients with cT1N1 cancer is only 34% in the databases of both Naruke and associates (2001) and of Mountain (1997), and for pT1N1 cases, the rates are 53% and 55%, respectively. Survival for pT1N1 patients in the series of van Rens and associates (2000) is 52%. The Ludwig Lung Cancer Study Group (1987) reported their results as median survival time (4.8 years for resected T1N1 patients, significantly longer than the median of 2.3 years for T2N1). As noted in Table 106-3, about a half or more of patients operated on for pT1N1 NSCLC do not survive 5 years.

As in stage I, variables that may affect prognosis include tumor size, histology, and location. In addition, in stage II and higher, lymph node metastases come into play in most

P.1565


cases. The impact of any single nonnodal factor is less clear than in stage I. Holmes (1987), reporting the LCSG experience, confirmed a significant difference favoring squamous T1N1 cancers at 5 years (75% versus 52% for nonsquamous histology). In a mixed group of T1 2N1 cases, Ichinose and colleagues (1995) also found better results with squamous lung cancers. In contrast, Martini (1992) and Yano (1994) and their co-workers did not find a significant histologic variance, although squamous cell tumors had an 8% to 10% numerical survival advantage over other cell types. The data with respect to tumor size are also limited, conflicting, and not focused by stage II subset. Within stage II, however, Martini and associates (1992) found that tumors smaller than 3 cm had a higher survival rate than those larger than 5 cm, but other series have not found an impact of size. The influence of central versus peripheral location has not been studied sufficiently to allow meaningful conclusions.

The nature of the N1 lymphadenopathy in stage II is important. Martini and colleagues (1992) found that patients with a single malignant node had a 5-year survival rate of 45%, as opposed to 31% for those with multiple N1 metastases. Of note was that half of their patients had only a single cancerous node, and 86% had spread to only one N1 nodal level. In a study focused on N1 disease and including T1 to T3 cases (but only three with T3), Yano and colleagues (1994) achieved a postresection survival rate of 65% in patients with lobar nodal disease (levels 12 and 13), as compared with only 40% when the hilar nodes (levels 10 and 11) contained metastatic cancer. Van Velzen and colleagues (1996) reported a late overall survival rate of 46% and also noted that lobar N1 portended a better outcome than hilar N1 (57% vs. 23%). They also found that node involvement by direct extension carried a markedly more favorable surgical prognosis than noncontiguous metastasis (69% vs. 30%) (Fig. 106-5). Riquet and associates (1999) also noted improved survival rates for lobar versus hilar N1 (54% and 38%), but found no difference between direct extension or separate metastasis or based on the number of levels involved. Tanaka and co-workers (2001) divided N1 disease into three levels and reported postresection survival for combined T1 2N1 to be 72% when only levels 12 and 13 were affected, 62% for proximal extent limited to level 11, and 39% for level 10 metastasis. These authors found no significant effect of histology, number of involved nodes, or T1N1 versus T2N1. Although a few other reports also do not confirm a difference between stages IIA and IIB, most series confirm the 1997 system. Marra and colleagues (2003) also found a stepwise decrease survival based on the aforementioned three subgroups of N1 in a mixed group of T1 4N1, with two thirds being T1 2N1. This group also reported a markedly worse prognosis for multilevel versus single-level N1 disease

P.1566


(44% and 27%) and for increasing numbers of involved nodes (49% for one node and 30% for over five nodes). A more favorable outcome for N1 by direct extension was also reported in this study, as well as for T1N1 versus T2N1 (65% and 45%). A series reported by Sawyer and colleagues (1999), however, failed to detect an effect of lobar versus hilar N1 disease.

Fig. 106-5. Five-year survivals for T1N1 lung cancer with N1 by direct extension (solid line), lobar N1 (dash-dot line), and hilar N1 (dotted line). From van Velzen E, et al: Type of lymph node involvement influences survival rates in T1N1M0 non small cell carcinoma: lymph node involvement by direct extension compared with lobar and hilar metastases. Chest 110:1469, 1996. With permission.

Mountain (1997) and Naruke and associates (2001) reported postresection survival rates at 5 years for the T2N1 subset of clinical stage IIB of 23% and 39%, respectively, with corresponding outcomes for pathologic T2N1 of 38% and 44%. The 5-year surgical survival rate in the experience of van Rens and co-workers (2000) was 33% for pathologic T2N1. There is little information available specifically relating to prognostic factors in this group, likely for the reasons noted for IIA disease. Roeslin and associates (1987) found a dramatic difference between lobar and hilar N1 disease in patients with T2 tumors (41% vs. 14% late survival). In their experience, the latter group did as poorly as patients with N2 lymphadenopathy. Van Velzen and associates (1997) also detected a differential favoring lobar over hilar N1 and direct invasion over metastasis. It is interesting that the survival rate for lobar T2N1 was identical to that for lobar T1N1 in their earlier study (57%), whereas hilar T2N1 had a slightly better survival rate than hilar T1N1 (30% vs. 23%). In the presence of nodal spread, the prognostic significance of the T factor is diminished. The influence of visceral pleural extension, for example, remains unclear in the presence of N1 disease. Van Velzen (1997) and Inoue (1998) and their colleagues determined that such local invasion in T2N1 unfavorably affected outcome. Although the initial report of Martini and colleagues (1983) found that pleural invasion was associated with a significantly worse prognosis, their updated series (1992) negated this factor in mixed T1 2N1 disease.

Stage IIB also includes node-negative tumors that invade potentially resectable structures (T3N0M0). Naruke and colleagues (2001) reported a 22% 5-year survival for clinical T3N0 and 32% for pathologic T3N0. The corresponding rates calculated by Mountain (1997) were 22% and 38%, and van Rens and associates (2000) noted a survival rate of 33% for resected pT3N0 NSCLC. By far the largest surgical experience involves tumors invading the parietal pleura or chest wall. It must be stressed that a high proportion of patients in many surgical reports were treated with preoperative or postoperative radiation therapy, or both. Although a statistically significant effect of this adjuvant approach has not generally been demonstrated, cognizance of this factor is warranted in interpreting the published material. It is also important to note that, despite the theoretical resectability of direct chest wall invasion, many patients with chest wall invasion do not undergo resection because of the surgeon's judgment that the degree or location of the process combined with the patient's clinical status makes the risk-to-benefit ratio unfavorable. Data in T3 categories other than chest wall invasion are limited. Detterbeck and Socinski (1997) have recently provided a valuable comprehensive review of this subject.

The results of resection for chest wall T3 cancers are summarized in Table 106-4. The major correlates of long-term survival are node negativity and complete resection.

P.1567


When both criteria are met, late survival is realized in 29% to 56% of surgical cases, averaging 42%. When malignant lymphadenopathy accompanies chest wall invasion (T3N1 or T3N2), both subsets now stage IIIA, the efficacy of resection is markedly limited. Five-year survival rates in the presence of N1 metastasis range from 8% to 35%, with an average of 19%. When metastases reach N2 nodes, the outcome is dismal, with no survivors in many series to a maximum of 21% in the report of Magdeleinat and colleagues (2001). Most surgeons believe that clinically detected N2 malignant lymphadenopathy is a contraindication to primary resection, as emphasized by Pairolero (1999). Incomplete resection is also a strong negative prognostic factor. With both lymphadenopathy and incomplete resection, operation alone is associated with minimal late survival. The experience of Downey and colleagues (1999) is typical. They noted a 5-year survival rate of 49% for completely resected T3N0, 27% for T3N1, and 15% for T3N2. Complete resection yielded an overall survival rate of 32%, versus only 4% for either grossly or microscopically incomplete resections (Fig. 106-6).

Table 106-4. Selected Series Reporting Postresection 5-Year Survival Rates for Tumors Classified as T3 Due to Invasion of the Parietal Pleura or Chest Wall

Report Dates Chest Wall Overall Chest Wall T3N0 IIB Chest Wall T3N1/N2 IIIA Chest Wall CR/IR
N 5-yr (%) N 5-yr (%) N 5-yr (%) N 5-yr (%)
Piehler et al (1982)a 1960 1980 56 33 26 54 19 7
Allen et al (1991) 1973 1988 52 26 43 29 N1 9 11
Casillas et al (1989) 1969 1986 97 23 58 34 N1 16 8
Mountain (1990) 1965 1982 31 39
Watanabe et al (1991b)a 1973 1989 24 43
Ratto et al (1991) 1983 1988 14 47 N1 19 22 CR 45 16
  N2 22 0 IR 10 0
Albertucci et al (1992)a 1976 1988 37 30 21 41 N1 9 29
  N2 7 0
McCaughan (1994) 1974 1983 125 NS 45 46 N1 17 35 CR 77 40
  N2 42 16 IR 48 0
Downey et al (1999) 1974 1993 334 NS 100 49 N1 24 27 CR 175 IR 32
  N2 51 15 IR 94 4
Chapelier et al (2000)a 1981 1998 100 18 65 22 N1 28 9
  N2 7 0
Facciolo et al (2001)a 1990 1999 104 61 77 63 N2 14 18
Magdeleinat et al (2001) 1984 1998 201 21 116 25 N1 42 21 CR 167 24
  N2 21 20 IR 34 13
Elia et al (2001) 1980 1994 110 NS 83 47 N1 17 0
  N2 10 0
N, number of cases; 5-yr, percentage survival at 5 years; NS, not stated; CR, complete resection; IR, incomplete resection.
a Series limited to complete resections.

The extent of invasion, divided pathologically into parietal pleural involvement versus extension into muscle and bone, and surgically into those amenable to extrapleural dissection versus those requiring chest wall resection, may influence resectability. Deep invasion portends a more limited survival. In patients with T3N0 disease, McCaughan and associates (1985) achieved a 5-year survival rate of 62% for parietal pleural T3, as opposed to 35% for deeper invasion, but the difference was not statistically significant. Likewise, Casillas (1989), Elia (2001), and Akay (2002) and their colleagues noted no significant difference between patients treated by extrapleural dissection versus full-thickness chest wall resection. In contrast, Albertucci and associates (1992) stress the importance of en bloc chest wall resection for all peripheral tumors densely adherent to the chest wall. These authors found a significant incidence of incomplete resection and lower survival (33% versus 50%) for extrapleural dissection. Facciolo and associates (2001) also advocate en bloc chest wall resection and report a 67% late survival for T3N0 cases amenable to complete resection. It appears that NSCLC involving a limited area of the parietal pleura and only superficial invasion may be treated adequately by pulmonary resection combined with extrapleural dissection, but any degree of deeper invasion requires full-thickness chest wall resection. The practical challenge is the definitive identification of the extent of invasion by preoperative and intraoperative assessment. When there is doubt, a low threshold to proceed with en bloc resection is prudent.

Fig. 106-6. Survival after complete resection, incomplete resection, or nonresective thoracotomy in patients with chest wall involvement. From Downey RJ, et al: Extent of chest wall invasion and survival in patients with lung cancer. Ann Thorac Surg 68:188, 1999. With permission.

Superior sulcus (Pancoast) tumors have been the focus of much surgical attention. Vallieres and colleagues (2001)

P.1568


have provided a recent review of this area. Pancoast tumors may be staged locoregionally as stage IIB (T3N0), IIIA (T3N1 2), or IIIB (T4N1 3, T3N3). Chapters 35 and 36 review the techniques and results of resection of these challenging lesions. Apical lung cancer invading the thoracic inlet was considered unresectable and uniformly fatal, until Shaw and associates (1961) demonstrated that radiation followed by radical resection could yield long-term survival. Their contributions in this area were furthered by Urschel (1988). Although preoperative irradiation has been the standard since then, there is recent interest in induction chemoradiation therapy. A phase II trial evaluating this approach in clinical T3 4N0 1 cases appears encouraging. As reported by Rusch and coinvestigators (2001), about one third of patients had a complete histologic response, and another third had only microscopic residual disease following induction. Complete resection was possible in 91% of T3 cases and a surprising 87% of T4 lesions. The survival rate at 2 years was 55% overall and 70% for complete resections.

Table 106-5 presents selected surgical series of superior sulcus tumors. As for other chest wall lesions, the major surgical correlates of cure are the associated lymph node status and the completeness of resection. Complete resection in the absence of adenopathy offers late survival of about 25% to 45%. When N2 dissemination has occurred, in contrast, resection is not helpful, with most series reporting no late survivors. In addition to N2 status, T4 disease has historically often been inoperable. In the large series reported by Rusch and colleagues (2000), for example, 5-year survival was 46% for pT3N0 superior sulcus cancers, in contrast to no late survival for stage IIIA and 13% for stage IIIB. Gandhi and co-workers (1999) reported a 2-year overall survival rate of 54% in 17 patients who underwent partial or total vertebrectomy for T4 Pancoast tumors, but no survivors among those with positive margins. Of interest, Sartori and associates (1992) found that patients who experienced pain relief from radiation therapy fared better than those with ongoing symptoms. Anderson and associates (1986) noted that durable pain relief following resection portended a very favorable outlook (73% 5-year survival rate, in contrast to no long-term survivors among those with persistent discomfort). The presence of Horner's syndrome is also a negative prognostic factor, as emphasized by Hagen and colleagues (1999).

Table 106-5. Selected Series of Postresection 5-Year Survival Rates for Superior Sulcus Tumors

Report Dates Overall CR IR T3N0 IIB N2 IIIA T4 Vertebrae IIIB T4 Subclavian Vessels IIIB
N 5-yr (%) N 5-yr (%) N 5-yr (%) N 5-yr (%) N 5-yr (%) N 5-yr (%) N 5-yr (%)
Miller et al (1979) 1971 1977 26 31 20 40 5 0
Stanford et al (1980) 1962 1977 16 50
Paulson (1985) 1956 1983 78 31 56 44 17 0
Anderson et al (1986) 1956 1985 28 34
Wright et al (1987) 1976 1985 21 27
Sartori et al (1992) 1981 1999 42 25 37 30 5 0 NS 0 5 0
Dartevelle et al (1993)a 1980 1991 29 31
Ginsberg et al (1994)a 1974 1991 All: 124
Res cases:
100
26
30
69 41 55 9 All: 96

CR:60
33
46
All:
8
CR:9
4
15
22 9 NS 0
Maggi et al (1994) 1982 1990 60 17 36 24 24 0 NS 0 NS 0
Muscolino et al (1997) 1984 1988 15 27 11 40 4 0 CR: 6 57 4 0
Attar et al (1998) 1955 1997 67 26 59 42
Hagen et al (1999) 1975 1992 31 33
Rusch et al (2000)a 1974 1998 225 NS 117 46 33 0
N, number of cases; 5-yr, percentage survival at 5 years; NS, not stated; res, resected cases; CR, complete resection; IR, incomplete resection.
a Series includes use of interstitial brachytherapy in some cases.

Despite good results in favorable subsets, the large series reported by Ginsberg and co-workers (1994) emphasizes the challenging nature of superior sulcus cancers. Resection was not possible in 24 of 124 explored patients and was incomplete in 31 other cases (overall 44%). As in other NSCLC patients, incomplete resection was associated with the same dismal 5-year survival as nonresective thoracotomy. The extensive earlier experience of Paulson (1985) also highlights the limitations of surgery because

P.1569


only 60% of patients presenting with Pancoast tumors were deemed suitable for treatment pathways that included resection. As noted in Table 106 5, 5-year survival is realized in one third or fewer of all operated patients. The aforementioned chemoradiation induction trial noted earlier is encouraging with respect to resectability. Confirmation of a durable oncologic benefit, however, must await broader experience.

Data are limited regarding the results of resection in other types of T3 tumors (Table 106-6). Mediastinal invasion limited to the pleura is classified as T3, with deeper invasion being T4. It is often difficult, however, to determine precisely the depth of invasion. Even in T3 cases, the reported surgical outcome is inferior to that of chest wall T3 lesions, with late survival in one fourth to one third of completely resected cases, as compared with more than half of similarly staged and treated chest wall T3 cancers. An early large series from Memorial reported by Burt and colleagues (1987) found that complete resection was possible in only 49 of 225 patients (22%) and that even in this subset, the 5-year survival rate was only 9%, but this report included many cases now classified as T4. In an update, however, Martini and associates (1994) reviewed 102 surgical patients with mediastinal invasion, excluding N2 disease, superior vena cava obstruction, superior sulcus cancers, and malignant pleural effusion, and applying current criteria for T3 and T4. Of 58 patients with T3 lesions, complete resection was possible in 38 cases (66%), and the 5-year survival rate was 36%. In contrast, resectability for T4 tumors was 18%, and the 5-year survival rate was 12% following complete resection. Pitz and colleagues (1996) noted a 25% postoperative 5-year survival rate in patients with T3 cancers invading the mediastinum, but no late survival in those with N2 disease. Only cases of complete resection were used to determine survival. Riquet and co-workers (2002) reported 5-year survival in 35% of mediastinal T3 NSCLC patients and noted a high prevalence and negative impact of nodal disease in these cases. Watanabe and associates (1991b) reported 43% late survival in a small group of patients following complete resection in the presence of limited pericardial invasion.

Table 106-6. Selected Series Reporting Postresection 5-Year Survival Rates for Tumors Classified as T3 Due to Invasion of the Mediastinum or the Diaphragm or Due to Proximal Main Bronchial Location

Report Dates Mediastinum Proximal Bronchus Diaphragm
N 5-yr (%) N 5-yr (%) N 5-yr (%)
Burt et al (1987) 1974 1984 All: 225 7
CR: 49 9
N0: 57 10
N1: 30 9
N2: 38 8
Watanabe et al (1991b, 1991c)a 1973 1989 17 43 11 46
Martini et al (1994) 1974 1992 All: 102 19
T3 CR: 38 36
T4 CR: 8 12
Pitz et al (1996)a 1977 1993 All: 40 25 75 40
N2: 9 0
Vogt-Moykopf et al (1986) 97 12
Deslauriers et al (1993) 31 14
Weksler et al (1997)a 1974 1995 8 0
Inoue et al (1998) 1980 1993 5 0
Rocco et al (1999) 1976 1998 IR: 9 0
CR: 39 39
Yokoi et al (2000) 1986 1995 CR T3N0: 26 28
CR T3N1 2: 29 18
Riquet et al (2000) 1978 1997 15 20
Riquet et al (2002) 1984 1996 68 31 68 35
N, number of cases; 5-yr, percentage survival at 5 years; CR, complete resection; IR, incomplete resection.
a Series limited to complete resections.

Survival rates for main-stem bronchial cancers classified as T3 because of carinal proximity vary widely (from 12% to 46%). The highest reported survival, however, was based on only 11 patients, all of whom had complete resection, and was calculated exclusive of operative mortality. Detterbeck and Socinski (1997) determined from the literature an average of 24%, a figure similar to T3 mediastinal invasion. Although Pitz and associates (1996) reported a 40% survival rate for proximal bronchial T3 versus 25% for mediastinal T3, the difference was not statistically significant. Firm conclusions regarding the incidence and outcome of proximal bronchial T3 NSCLC are hampered by problems of definition, largely because lesions at the right upper lobe origin may variably be classified as T2, T3, or T4 based on subjective assessment of the radiographic and bronchoscopic

P.1570


features and interpretation of the staging criteria. By extrapolation from the considerable literature on bronchoplastic and carinal resection (see Chapters 28 and 30), however, it appears that complete resection of node-negative main bronchial T3 cancers yields late survival in one fourth to one third of cases. N2 disease is a strong negative factor, whereas the data with respect to N1 metastasis versus N0 are conflicting.

Despite the fact that the hemidiaphragms abut a large surface of lung parenchyma and that direct invasion in this area is potentially resectable, there is a paucity of reported surgical experience. Weksler and associates (1997) found only eight cases in a review spanning two decades at Memorial. All four patients with N2 disease died of their lung cancers, with a mean survival of only 92 weeks. Although a single N0 patient was alive at 70 weeks at the time of the report, the remaining 3 died of nonmalignant causes between 5 and 17 weeks after surgery. The diaphragmatic lesions also differed from the usual current spectrum of peripheral T3 NSCLC, in that seven were squamous cancers and the eighth had adenosquamous histology. Inoue and associates (1998) reported no 3-year survival in five operated patients with diaphragmatic invasion, despite complete resection and N0 1 status. For unclear reasons, more recent reports suggest an improved outcome. Rocco and associates (1999) and Riquet and colleagues (2000) achieved long-term survival in completely resected T3N0 cases of 39% and 27%, respectively. Yokoi and co-workers (2000) performed combined lung and diaphragmatic resections in 26 cases of T3N0 and 29 cases of T3N1 2 lung cancer. Complete resection of N0 and of N1 2 cases yielded 5-year survival rates of 28% and 18%, respectively. Invasion limited to the diaphragmatic pleura was associated with a survival rate of 33% versus 14% for muscle penetration. In all these series, there was no late survival following incomplete resection.

In contrast to lower-stage lesions, tumor size and histology are less predictive for T3 cancers. In general, invasive squamous cancer is more favorable than adenocarcinoma. Although most reports indicate no statistically significant difference, Mountain (1990) found a trend in favor of squamous cell histology, and Ratto and associates (1991) reported that patients with squamous cell cancer invading the chest wall fared better with both complete and incomplete resection than did those with adenocarcinoma. In T3 disease with either mediastinal invasion or carinal proximity, Pitz and co-workers (1996) also achieved better results in squamous cell cases. The experience reported by Martini and colleagues (1994) is unusual because T3 adenocarcinomas with mediastinal infiltration were more favorable.

Although complete resection of highly favorable T3N0 NSCLC yields late survival in one fifth to two thirds of cases (chest wall), depending on the structures invaded, the overall results are unsatisfactory. It remains unknown whether adjuvant therapy is helpful. Postoperative radiation therapy is used frequently, but sporadically. Interstitial brachytherapy and systemic chemotherapy are applied less often.

Stage IIIA: Non Small Cell Lung Cancer with Mediastinal Lymph Node Involvement

The T3N1 subset of stage IIIA represents a minority of cases. Results in this group have been summarized previously and are presented in Tables 106 4 to 106 6. N2 disease comprises the bulk of stage IIIA and has been the subject of much debate with respect to the role of primary resection. This debate appears to have largely ended, with most thoracic surgical and medical oncologists currently favoring combined-modality approaches for clinical N2 NSCLC.

The concept of resectable N2 disease is based on reports documenting late survival in 20% to 30% of patients with various combinations of favorable features, such as complete resection, single-node or single-level involvement, microscopic or intracapsular metastasis, N2 confined to lower stations, and left upper lobe cancers with N2 limited to the subaortic level. The problem with clinical application of this concept is that these results are based on retrospective pathologic staging in patients who were considered preoperatively by a variety of staging pathways either to be clinical N0 or N1 or to be good candidates for resection despite suspected or documented N2 metastasis. As pointed out by Shields (1990) and by Goldstraw (1992), the number of patients in the spectrum of N2 NSCLC who benefit from operation is low. Even with careful selection for minimal N2 disease, patients whose malignant N2 lymphadenopathy is diagnosed preoperatively rarely benefit from surgery. The rate of incomplete resection or nonresective exploration is high, whereas the 5-year survival rate is low. In contrast, those whose N2 disease is not apparent by clinical staging, but is diagnosed only by surgical or pathologic staging, have a better postoperative outlook. Retrospective outcome in the latter setting cannot be generalized to the large number of patients with clinical N2 lung cancer.

Table 106-7 summarizes selected reports of operation in N2 disease. The seminal study is the series of Pearson and associates (1982), who found that resection in mediastinoscopy-positive cases, along with adjuvant radiation in most instances, was associated with a 5-year survival rate of only 9%. In contrast, survival in mediastinoscopy-negative, pathologically positive cases (cN0 1, pN2) was almost threefold higher at 24%. The operated cN2 patients represented only one fifth of all such cases evaluated during the period and were deemed to have surgically favorable N2. Nonetheless, only 64% had a complete resection. It is noteworthy that complete resection in this setting increased the 5-year survival rate to only 15%. There were no late survivors in the 36% who had incomplete resections. These authors also noted that there was no survival with surgery alone in five prior reports of mediastinoscopy-positive patients. On analysis, no subsequent series has negated these conclusions.

More recently, Vansteenkiste and colleagues (1997) also reported a significant survival difference for mediastinoscopy-negative

P.1571


versus positive patients (32% and 15%, respectively), despite strict criteria for exploring only those thought to have minimal N2 metastasis. The authors emphasize that the few late survivors in the clinical N2 group had truly minimal N2 disease pathologically. In a distinctly unusual but instructive experience, Funatsu and colleagues (1992) performed mediastinoscopy in 619 patients and proceeded to thoracotomy in virtually all cases. Among 117 instances of positive mediastinoscopy, complete resection was possible in only 13 instances (11%). In the remainder, resection was either incomplete (78 cases, 67%) or not possible (26 nonresective explorations, 22%). Although late

P.1572


survival was 28% for the small group of completely resected patients, when incomplete resections are included, even eliminating nonresective operations, survival was only 6%.

Table 106-7. Selected Series Reporting Postsurgical Survival in Patients with Involvement of N2 Lymph Nodes

  Dates Staging All Operated pN2 cN0/1 pN2 cN2 pN2 Comments
N 5-yr (%) N 5-yr (%) N 5-yr (%)
Pearson et al (1982) 1964 1980 Med 62 24 All: 79 9 Preoperative and post-operative RT in most
CR: 51 15
IR: 28 0
Martini and Flehinger (1987) 1974 1981 Radiography 404 NS All: 224 NS All: 179 NS Presents only CR (18% of cN2); post-operative RT in 90%
Bronch CR: 119 34 CR: 32 9
Few med IR: 105 NS IR: 147 NS
Ishida et al (1990a) 1974 1988 Radiography All: 15 18 Nonresective thora-cotomies eliminated; postoperative RT in 56%; IR includes patients without MLND
Bronch CR: 63 27
CT scan
Selective med
IR: 52 9
Watanabe et al (1991d) 1980 1990 Mainly CT scan 153 17 All: 47 20 All: 106 16 Nonresective thora-cotomies eliminated (8% of cN2 pN2)
CR: 31 33 CR: 53 20
IR: 16 0 IR: 53 0
Regnard et al (1991) 1982 1988 CT scan All: 254 18 93% adj rx
Few med CR: 191 23 19 small cell
IR: 63 0
Funatsu et al (1992) 1970 1990 Imaging 46 14 All: 91 6 Nonresective thora-cotomies eliminated (22% of cN2 pN2)
Med CR: 13 28
IR: 78 2
Daly et al (1993) 1979 1991 CT scan All: 37 28 Adj rx in most
Bronch
Selective med
CR: 31 31
Maggi et al (1993) 1980 1990 CT scan All: 278 14
Bronch CR: 236 19
Selective med IR: 42 0
van Klaveren et al (1993) 1975 1985 Radiography
CT scan
Selective med
All: 48 10
Goldstraw (1994) 1979 1989 CT scan All: 149 17 No adj rx in CR cases
Bronch
Med: 48%
CR: 127 20
Miller et al (1994) 1982 1986 CT scan All: 167 21 Adj rx in 80%x
Bronch CR: 147 24
Med: 28% IR: 20 5
Vansteenkiste et al (1997) 1985 1993 CT scan 140 21 All: 121 22 Med: (+) 15 Postop RT in 50%
Bronch Med: (-) 32 19 CR (n = 113), 25% 5-yr
Selective med 68 IR (n = 27) 4% 5-yr
Mountain (1997) 1975 1988 Variable pT1N2: 53 38 344 23 471 13
pT2N2: 237 22
pT3N2: 56 12
Jassem et al (2000) 1991 1995 CT scan pT1N2: 11 18
Bronch pT2N2: 84 13
Few med pT3N2: 33 6
van Rens et al (2000) 1970 1992 Imaging pT1N2: 13 31
Med: 96% pT2N2: 187 18
pT3N2: 61 7
Naruke et al (2001) 1961 1995 Variable cT1 2N2: 299 22
pT1 2N2: 414 (see text)
N, number of cases; 5-yr, percentage survival at 5 years; NS, not stated; Med, mediastinoscopy, anterior mediastinoscopy/-otomy, or both; Bronch, bronchoscopy; RT, radiation therapy; adj rx, postoperative adjuvant therapy; CR, complete resection; IR, incomplete resection; MLND, mediastinal lymph node dissection; cN2, clinical N2; pN2, pathologic N2.

In addition to mediastinoscopy-proven N2 cases, true-positive N2 patients staged clinically by pathways employing only noninvasive modalities or selective mediastinal exploration also fare poorly with primary exploration. Martini and Flehinger (1987) reviewed a large experience with N2 cancers in which clinical staging was based on radiography and bronchoscopy, with few mediastinoscopies. Of 706 patients with N2 disease, 404 were considered operable. Of these, 224 were cN0 1 and 179, cN2. The survival analysis is limited to complete resections, which were possible in 53% of the former group but only 18% of the cN2 cases, with corresponding 5-year survival rates of 34% and 9%. Stated numerically, among 179 cN2 cases deemed operable, complete resection was possible in only 32 instances, and only a few of this highly selected group enjoyed late postoperative survival. Employing CT scan as the primary staging modality, Watanabe and associates (1991b, 1991d) similarly found that only half of cN2, pN2 cancers could be resected completely, with a late survival rate of 20%. In the remaining 50% of patients undergoing subtotal tumor removal, there were no late survivors. In contrast, complete resection was possible in two thirds of cN0 1, pN2 cases and yielded a 33% long-term survival rate. Andre and associates (2000) reported late survival in 18% of all pN2 cases, 23% of those who had a complete resection, and only 7% of those with clinical N2 disease. In the large series of Mountain (1997) and of Naruke and associates (2001), 5-year survival for pathologic N2 was about 22% overall. Van Rens and associates (2000) reported a 19% survival rate.

Numerous recent studies employing CT scan and selective invasive mediastinal staging have confirmed high complete resection rates and reasonable survival in cN0 1, pN2 lung cancer. Goldstraw and colleagues (1994) emphasize that clinically unsuspected N2 disease is common. These authors found that one fourth of cN0 1 patients staged by CT scan and 48% staged by mediastinoscopy were found to have pN2. Complete resection was possible in 85% and yielded late survival in 20%, with no adjuvant treatment. Although one of us (BDTD) and colleagues (1993), also using CT and selective mediastinoscopy, found only a 7% incidence of unsuspected N2, they likewise noted resectability and late survival in 89% and 28%, respectively. Similarly, Maggi (1993) and Miller (1994) and their associates were able to accomplish complete resection in 85% and 88% of such cases, with late survival in 19% and 24%, respectively. Incomplete resection in this group, as in cN2 patients, however, is of no benefit, with only a rare patient surviving 5 years. Although the actuarial survival for incomplete resection in unsuspected N2 in the series of Miller and colleagues (1994) is 5%, for example, this figure is based on a single case. The remaining patients all died within 3 years of operation. Andre and colleagues (2000) reported 8% postresection survival for clinical N2 involving one lymph node level and 3% when more than one level was positive. In cases of unsuspected N2 disease, the 5-year rate survival was 34% for single-level metastasis, but only 11% for multilevel N2 disease.

Despite the difficulty of prospectively identifying prognostic indices in individual cN2 cases, multiple factors have been extensively analyzed in postresection N2 disease. Incomplete resection, as in other subsets of NSCLC, is the most powerful negative prognostic factor. As noted, however, even complete resection yields little benefit. The extent of N2 metastases also appears to be a predictor of prognosis. Other factors that have been variably reported to be significant include the anatomic location of N2, the presence or absence of associated N1 lymphadenopathy, the T classification, and the cell type.

Watanabe and associates (1991d) found that with only one nodal station involved, the 5-year survival rate was 35%, as opposed to 9% when more than one level was affected. Regnard (1991) and Maggi (1993) and their colleagues confirmed this finding (24% vs. 9%, and 30% vs. 19%, respectively). Miller and associates (1994) reported a dramatic difference between the extremes of node station involvement (30% late survival in patients with single-level N2, but no survival when four or five levels harbored cancer). Andre and co-workers (2000) reported a long-term survival rate of 34% for patients with single-level N2 metastasis who had a negative preoperative CT scan (nodes smaller than 1 cm). With greater than one level of pathologic involvement, the survival rate decreased to 11% in the presence of a negative mediastinal CT scan. Goldstraw and associates (1994), however, reported that single-level N2 metastasis was associated with improved survival only for the first 3 postoperative years, with no advantage beyond that point. In the report by Vansteenkiste and colleagues (1997), the negative impact of multilevel metastases was limited to cases of nonsquamous histology.

In some series, such as that of Martini and Flehinger (1987), the number of involved nodes correlated directly with a worsened prognosis. Ishida (1990a) and Vansteenkiste (1997) and their associates, among others, have also shown that gross replacement of nodes by malignant cells, especially when there is disruption of the nodal capsule, presages a poorer postresection result than does the presence of intranodal or microscopic metastasis. Five-year survival rates in these studies of patients with gross versus intranodal disease were 16% versus 23%, and 11% versus 34%, respectively. Although also not universally confirmed, the anatomic site of N2 adenopathy may influence surgical curability. Metastasis to the subcarinal nodes (level 7) is often associated with lower survival than other sites, as reported by Pearson (1982), Regnard (1991), Watanabe (1991b, 1991d), and Miller (1994) and their co-workers. Also, when N2 disease is divided into inferior (levels 7, 8, and 9) and superior mediastinal categories (all other levels), survival in the former group is comparatively limited. In the series of Maggi and associates (1993), survival rates for N2 in an inferior or superior location were 8% versus 25%. Nakanishi and co-workers (1997) reported a more marked difference, with no late survivors when inferior stations

P.1573


were involved, in contrast to survival in about one third of patients with superior mediastinal N2. Miller and colleagues (1994) also identified metastasis to inferior mediastinal nodes as a negative prognostic index. In contrast to the finding that N1 disease defined by direct extension is more favorable than N1 anatomically separate from the primary site, Regnard and associates (1991) achieved better late survival in patients with separate N2 lymphadenopathy compared with those with direct invasion. This finding, however, has not been noted by others. The coexistence of N1 metastasis along with N2 disease is another node-specific factor that has occasionally been designated as a negative predictor. Although skip metastases (N2 disease in the presence of benign N1 stations) occurs in about one fourth of all pathologic N2 cases, the prognostic implication of this finding is not addressed in most analyses.

The primary tumor classification (T stage) in the presence of N2 disease overall appears less prognostically important than in lower stages. As noted earlier, however, the coexistence of N2 metastasis from a T3 NSCLC generally precludes surgical cure. Although there is a trend toward better results in T1N2 compared with T2N2 cases, most series do not report a statistically significant difference when other factors coincide. The effect of histology is similar, in that squamous cell cancers appear to have a slightly more favorable outlook, but the difference is usually not dramatic. In their series of patients with unexpected pathologic N2, Goldstraw and colleagues (1994) found a markedly more favorable outcome for squamous cancer compared with adenocarcinoma (29% and 5%, respectively).

Left upper lobe NSCLC with N2 metastasis confined to the subaortic (aortopulmonary, level 5) nodes has been suggested as a uniquely favorable subset (Table 106-8). Patterson and associates (1987) brought this area to attention in a review of 35 patients with isolated level 5 N2, in which they reported late survival in 28%. Two thirds underwent complete resection, with a 5-year survival rate of 42%. Watanabe (1991d) and Cybulsky (1992) and their collaborators noted 20% and 21% survival rates, respectively. In the latter series, the late survival rate was only 10% for the entire N2 group. Although half of the left upper lobe patients in this report had positive CT scans (cN2), survival in this subset was not reported separately. In the experience reported by Nakanishi and associates (1997), surgical survival was markedly better (80%) among patients with single-level N2 disease when localized to the subaortic area, but there were only six patients. In a study limited to skip metastasis (pathologic N2 with negative N1 nodes), Tateishi and associates (1994) also noted a numerically striking but statistically borderline better surgical prognosis in isolated level 5 N2 disease than in the total group (50% vs. 24%). The reports of Goldstraw (1994) and Vansteenkiste (1997) and their co-workers, however, do not confirm an advantage in this setting.

Table 106-8. Series Reporting Postsurgical Survival Rates in Patients with Left Upper Lobe Tumors Associated with N2 Lymphadenopathy Confined to the Subaortic Station (Level 5)

Report Dates Staging Operated Left Upper Lobe N2 Comments
N 5-yr (%)
Patterson et al (1987) 1968 1985 Med All: 35 28 Two cases of small cell; one a long-term survivor; radiation therapy in 57%
CR: 23 42
IR: 12 NS
Watanabe et al (1991d) 1973 1990 Variable 38 21 Limited to CR
Cybulsky et al (1992) 1982 1988 CT scan All: 32 20 Adj rx in most
cN2: 16 NS
Nakanishi et al (1997) 1979 1994 Radiography
CT scan
Bronch
Med
6 80 21% 5-year survival for total group of unsuspected pN2
N, number of cases; 5-yr, percentage survival at 5 years; Med, mediastinoscopy, anterior mediastinoscopy/-otomy, or both; Bronch, bronchoscopy; cN2, clinical N2; RT, radiation therapy; adj rx, postoperative adjuvant therapy; CR, complete resection; IR, incomplete resection.

In summary, very small numbers of patients have been identified to have a fair long-term prognosis based on postresection pathologic factors. It is clear, however, that cases with clinically evident N2, by imaging, invasive staging, or both, are often not amenable to complete resection and are associated with a dismal prognosis even with complete resection. For these reasons, primary operation in clinical N2 is considered contraindicated by most thoracic oncologists, despite some remaining respected disagreement. The issue of whether operation following induction therapy is beneficial remains under investigation (see Chapter 113). In addition, it is possible that the application of PET scanning and other new staging modalities will allow identification of truly favorable clinical N2 patients that has not been possible using prior approaches.

Stage IIIB: Non Small Cell Lung Cancer with Unresectable Local Invasion or N3 Lymphadenopathy

Although stage IIIB is defined as unresectable disease, an exceptional T4 case may be suitable for operation. In the

P.1574


overwhelming majority, however, primary operation is of no oncologic value, and operative mortality is high. Any benefit with respect to local control is offset by early distant metastasis. In the series of Mountain (1997), the 5-year survival rate was 6% for clinical T4 and 3% for N3 disease. In the database of Naruke and associates (2001), the late survival rate was 16% for all operated stage IIIB patients and 3% for N3 patients. The surgical series of van Rens and colleagues (2000) does not include NSCLC beyond stage IIIA. The role of multimodality therapy in stage IIIB NSCLC is discussed in Chapter 113. In addition to the problem of a degree of subjectivity in defining some cases of T4 versus T3 mediastinal invasion and carinal involvement, a new factor complicating surgical consideration of stage IIIB is that it represents a very inhomogeneous group because the 1997 staging system has added to T4 and N3 disease, already a diverse cohort, those patients with more than one parenchymal site of cancer within a single lobe. Although these cases overall have survival following clinical staging similar to T4 and N3, the place of operation in this setting is presently not at all as clear as in other stage IIIB cases. Multiple tumors are discussed later.

With respect to T4 lesions, the only patients who should be considered for primary operation are those few whose tumors, after rigorous imaging and invasive assessment, appear to be node negative and totally and safely resectable. Such patients are few in number. The extensive disappointing experience in T3N1 2 disease is compounded in the far lesser experience with T4 primaries associated with nodal metastasis. Initial exploration is never indicated in clinical T4N2 disease. Martini and colleagues (1994) noted that complete resection was possible in only 18%, 8 among 44 cases, of T4 tumors with mediastinal invasion that were selected for operation. Among the eight complete resections, there was only a single 5-year survivor. Applying a policy of more aggressive extended resection, Tsuchiya and associates (1994) noted a complete resection rate of 66% by gross inspection, but a histologic complete resection rate of 30%. The 5-year survival rate in this series was 13% for the entire group, 19% after complete resection, and 0% after incomplete resection. Although Fukuse and colleagues (1997a) reported complete resection in 35% of 42 cases with atrial or great vessel invasion, only the 3-year survival rate is specified (17%). Doddoli and associates (2001) were able to accomplish a complete resection in 25 of 29 cases of T4 NSCLC. The operative mortality rate was 7%, and nonfatal major complications occurred in 28%. Among the latter, half were due to chylothorax, bronchopleural fistula, or stroke. Although not further detailed, the potential magnitude of these problems in 14% of cases, coupled with a 7% mortality rate (total, 21%), raises therapeutic and quality-of-life questions about operation in T4 NSCLC that, in this series, was associated with median survival times of 16 months and 11 months for pN0 1 and pN2 cases, respectively. Bernard and co-workers (2001) reported a mixed cohort of resected T4 cases that they grouped into low-, intermediate-, and high-risk categories based on survival. Patients with N0 1, upper lobe or main bronchial T4 cancers had a 3-year survival rate of 36% and were deemed favorable. In the intermediate and high-risk groups, defined by various permutations of N disease and tumor location, 3-year survival rates were 4% and 0%, respectively. Although the authors concluded that certain carefully defined cases of T4 NSCLC are surgical, it is important to note that operation for T4 was performed in only 77 cases over 9 years, that intrapericardial pulmonary artery, carinal, and multiple parenchymal T4 NSCLC constituted 47 cases, and that only 3-year survival is reported.

As for other TNM subsets, there is variation in surgical results within the T4 category. The grave prognostic implication of T4 invasion in superior sulcus tumors has been noted. Likewise, T4 lesions, because of a pleural effusion containing malignant cells or diffuse pleural carcinomatosis, are not resectable. Although Ohta and associates (2000) reported a 5-year survival rate of 13% following lung resection and parietal pleurectomy, the number of survivors is very small, and adjuvant chemotherapy achieved borderline statistical significance in this experience. The authors recommend against pleuropneumonectomy, preferring partial lung resection and pleurectomy. Yokoi and colleagues (2002), in contrast, performed pleuropneumonectomy, including complete diaphragm resection on 11 patients. Although the actuarial 5-year survival rate was 54%, the five cases of T4N2 NSCLC remained disease-free for only 4 to 14 months, and one of the two T4N1 patients was disease- free for only 6 months. Because effective palliation of effusive symptoms can be achieved by a variety of minimally invasive means (see Chapter 68), local control is not a valid argument for operation in cases of malignant effusion. However, if the effusion is not malignant, but is caused by atelectasis secondary to an obstructing cancer or other etiology, the utility of operation is dependent on other clinical staging factors. Earlier, Decker (1978) noted that resectable cases in this setting were few in number (only 5% of operated patients with cytologically negative effusions had limited disease and were late survivors). Ruffini and associates (2002a) presented a series of small pleural effusions discovered intraoperatively. Cytology was positive in 53% of 45 cases, and negative in 47%. The median survival in the setting of a cytologically malignant effusion was only 6 to 9 months, whereas the negative cytology group had 3- and 5-year survival rates of 68% and 56%, respectively. Although the significance of cytologically negative effusions has been the subject of much debate, the clinical dilemma is largely resolved because a combination of modern imaging, thoracoscopy, and other modalities will establish the benign or malignant etiology of the pleural fluid preoperatively with a high degree of accuracy.

The largest surgical experience with T4 NSCLC encompasses tumors involving the carina (see Chapter 30). In properly selected cases treated at centers with experience, late survival can be realized in up to 40%, as reported by

P.1575


Faber (1987), and by Watanabe (1990), Deslauriers (1993), Roviaro (1994), and Mitchell (2001) and their colleagues.

Reports of resection in other T4 subsets are generally confined to small series. Tangential resection and primary closure or patching of the superior vena cava (SVC), as well as circumferential resection requiring graft reconstruction, has been reported by Dartevelle (1991), Nakahara (1989), and Tsuchiya (1994) and their colleagues. Late survival, however, is limited to a few cases [one reported by Inoue and associates (1990) and two of 30 cases in the series of Tsuchiya and associates (1994)]. In the experience of Burt and associates (1987), there was no late survival among 18 cases of NSCLC invading the SVC and treated by pulmonary resection, brachytherapy, or both. The rarity of lung cancers suitable for caval resection is underscored by the reports of Dartevelle and colleagues (1987, 1991). In their earlier paper, 5 of 13 SVC resections (31%) were performed for lung cancer. In the subsequent 4 years encompassed in the updated series, only two cases of NSCLC were added. All six patients had malignant adenopathy, and four received adjuvant radiation. Neither of the two cases with N2 disease survived beyond 8 months, whereas two of four N1 patients died at 1 month and 38 months, and two were alive at 16 and 52 months. A higher late survival rate of 25% was noted in the eight patients reported by Bernard and co-workers (2001). Spaggiari and colleagues (2002) presented their experience with 28 patients who underwent SVC resection and graft replacement for lung cancer. The overall 5-year survival rate was 15%. One-year and 2-year survival rates were 36% and 20%, respectively. The authors concluded that, although SVC T4 is not per se an absolute contraindication to operation, the disappointing oncologic results of surgery suggest that very few patients are candidates for this approach. The operative mortality rate in this series was 14%, and a literature review showed an acute mortality rate of 12% to 17%.

Operation for tumors invading the aorta has been performed in even fewer instances. Klepetko and associates (1999) reported five cases of combined left lung and aortic resection. Three patients with N2 disease died between 17 and 27 months, and two with pathologic T4N1 disease were alive at 14 and 50 months. Nakahara (1989), Horita (1993), and Tsuchiya (1994), along with their colleagues, also reported aortic resection for T4 lung cancer, but the long-term results of this aggressive approach, usually requiring cardiopulmonary bypass, are not specified. In the report of Burt and colleagues (1987), there were no late survivors among 19 patients with aortic involvement. Similarly, there were no 3-year survivors in the eight cases of Bernard and colleagues (2001). In summary, resection of the great vessels for NSCLC, even when surgically possible, is of unproven therapeutic value and should be considered only in exceptional circumstances.

T4 cancers involving the pulmonary artery trunk have rarely been approached surgically and should not be included with cases of invasion of the intrapericardial portion of the right or left pulmonary artery. In a report by Ricci and associates (1994), among 17 resections involving pulmonary arterial reconstruction, 14 had extrapericardial invasion. Among the three cases requiring cardiopulmonary bypass for intrapericardial repair, there was one operative death, and the other two patients died from metastases at 3 and 25 months after operation. Similarly, there were no late survivors among seven patients in the series of Tsuchiya and colleagues (1994) who underwent main pulmonary trunk resection. In contrast, Rendina (1999), Shrager (2000), and Bernard (2001) and their associates have clearly demonstrated that tangential or circumferential resection of the right or left proximal pulmonary artery is beneficial in appropriate cases and based on stage. Overall late survival rates in these reports were 38%, 48%, and 20%, respectively. The lower figure in the last instance is likely due to the inclusion of only patients with intrapericardial pulmonary artery invasion. Also, some localized cancers with direct invasion of the left atrium at the junction with the pulmonary veins can be resected and closed primarily, without the need for cardiopulmonary bypass. In the report of Tsuchiya and associates (1994), 5-year survival in this group was 22%, and 7 of 13 absolute late survivors in the entire series of 101 T4 cases of mediastinal invasion had undergone left atrial resection.

Although technically feasible in a few instances, as discussed earlier, the utility of total vertebrectomy is unknown. Grunenwald and colleagues (2002) reported 19 cases of total vertebrectomy or hemivertebrectomy. Two- and 5-year survival rates were 53% and 14%. Combined pulmonary and esophageal resection for invasive lung cancer, another occasionally reported surgical feat, must also be viewed in the global context of T4 disease.

Resection of N3 disease has been carried out through sternotomy, bilateral thoracotomy, or cervical dissection combined with thoracotomy and in a few instances has achieved complete removal of all known neoplasm. However, this approach has not been shown to result in long-term survival. Primary operation in clinical N3 lung cancer is not indicated. The utility of induction therapy followed by resection including all sites of original disease or addressing only ipsilateral metastasis, essentially a postinduction staging operation, as in the Southwest Oncology Group series reported by Rusch and colleagues (1993), remains unknown.

Stage IV: Non Small Cell Lung Cancer with Distant Metastases

Although dissemination of NSCLC can occur to virtually any organ, the most common sites are the brain, bone, liver, and adrenal glands. Only in very carefully selected cases of solitary cerebral metastasis does a combination of metastasectomy and lung resection appear to be beneficial. Adrenal metastases from NSCLC have been resected in a few instances,

P.1576


but the impact on survival is questionable. There is no long-term benefit by a surgical approach to the primary site in the presence of other areas of nonpulmonary M1 disease; although the 5-year survival rate for surgical patients with M1 disease in the series of Naruke and associates (2001) was 5% overall and 8% for M1 due to multiple parenchymal cancers. In Mountain's database (1997), the cumulative survival for patients with M1 disease was 1%.

Brain Metastasis

Several series have shown apparent improved survival of about 10% to 20% with removal of a synchronous or metachronous solitary brain metastasis (SBM) and pulmonary resection for otherwise limited NSCLC. As many as one third of patients with brain metastasis at presentation have a solitary lesion, but most also have extracerebral distant foci or advanced locoregional disease. Although resection of the cranial lesion may offer optimal neurologic palliation, pulmonary resection should be undertaken only in cases that would have been suitable for curative primary resection in the absence of M1 disease and only after a thorough evaluation to detect other sites of metastatic disease. The first large experience was reported by Magilligan and colleagues (1986), who found a 5-year survival rate of 21% and a 10-year survival rate of 15% in 41 synchronous and metachronous cases of SBM. Burt and associates (1992) reported 185 consecutive patients who underwent craniotomy for NSCLC (65 synchronous and 120 metachronous). The frequency of otherwise advanced NSCLC in the setting of cerebral M1 is emphasized by the fact that 37% of the 65 synchronous cases were not considered for pulmonary resection and that, among the 41 thoracotomies, incomplete or no resection occurred in 22%. The overall late survival rate was 13% at 5 years and 7% at 10 years. In a series of synchronous and metachronous cases, Read and associates (1989) reported a 21% 5-year survival rate following complete resection of both the SBM and the primary site in 27 cases as compared with a median survival rate of only 6.4 months in patients who underwent noncurative resection of either or both sites. Although a recent report by Mussi and colleagues (1996) notes only a 6.6% 5-year survival rate for combined resection in synchronous SBM, their figure was 19% for metachronous lesions. Granone and associates (2001) reported a 17% 3-year survival rate in 30 cases of synchronous and metachronous brain M1. In a series of 28 patients with synchronous metastasis, Billing and colleagues (2001) noted a 5-year survival rate of 21%, whereas Bonnette and colleagues (2001) reported a lower figure of 11% in 103 cases. Favorable factors in the setting of SBM include N0 disease, lower T factor, and adenocarcinoma. In summary, in selected patients with SBM and resectable primary sites of NSCLC, a surgical approach to both areas is warranted and can result in late survival in up to 20% of patients. In a randomized trial, Patchell and associates (1998) found that the addition of whole-brain radiation therapy (WBRT) following complete resection of SBM significantly decreased brain recurrences and death from neurologic causes. Further experience with stereotactic radiosurgery as an alternative to craniotomy or in cases of surgically inaccessible lesions may increase the proportion of patients suitable for pulmonary resection.

Adrenal Metastasis

The importance of documenting the nature of a solitary adrenal mass in patients with otherwise resectable lung cancer has been noted. Experience with resection of both synchronous and metachronous adrenal metastases from NSCLC is limited. Luketich and Burt (1991) reported on 14 patients with synchronous NSCLC and solitary adrenal metastases, all treated initially with cisplatin-based chemotherapy. Eight patients ultimately underwent resection and had a median survival of 31 months, as compared with 8 months for those treated with chemotherapy alone. Ayabe and associates (1995) presented three cases and summarized the literature. Among 12 cases up to 1995, there was documented late survival in three patients (at 6, 9, and 14 years). Two of these were synchronous and one metachronous. Porte and colleagues (1998) identified 11 cases of solitary adrenal metastases in 598 consecutive patients with otherwise operable or resected NSCLC at their institution. Among eight patients with synchronous disease treated by resection, the median survival time was only 10 months, although one patient remained cancer free at 66 months. Of the three cases of metachronous lesions, two died at 6 and 14 months, and one was alive at 6 months. Porte and colleagues (2001) later reported 43 patients from eight centers. The median survival time was about 16 months for both synchronous and metachronous disease. Two-, 3-, and 4-year survival rates were 29%, 14%, and 11%, respectively. Urschel and associates (1997) reported a patient who remained free of recurrence 9 years after resection of bilateral adrenal metastases and lobectomy. The patient was also treated with chemotherapy. Although these reports suggest that a tiny subset of patients demonstrate late survival after resection, most operated patients experience rapid dissemination of their disease.

Other Metastatic Sites

Late survival has rarely been achieved in any of the few reports of resection of other presumed solitary sites of metastatic NSCLC, including splenic resection as reported by Macheers and Mansour (1992) and hepatic resection, as reported by Hughes and Sugarbaker (1987), as well as by Lindell (1998) and Di Carlo (2003) and their associates (15.4 and 3 years, respectively). Burt (1996), however, reported a unique experience showing late survival in 14

P.1577


cases of metachronous solitary metastases from NSCLC to a variety of sites, 13 of whom were treated by resection.

Multiple Parenchymal Lung Tumors

Although the 1997 staging system classifies multiple sites of lung cancer as T4 or M1 (stage IIIB or IV), depending on relative location, the role of resection in these groups is not as clearly limited as in other subsets of T4 and M1 NSCLC. In addition, as noted earlier, it is often not possible to determine with certainty whether two lesions are separate primaries, or if one is metastatic from the other. The criteria introduced by Martini and Melamed (1975), as discussed earlier, are commonly applied. Deslauriers and associates (1989) noted that the presence of satellite nodules (malignant foci close to, but separate from, the dominant tumor) halved the late survival after resection. Nonetheless, the survival rate was 22% in patients with such nodules. A more favorable prognosis than for other T4 or M1 NSCLC treated by primary operation has been confirmed by others. Fukuse and associates (1997b) reported a 5-year survival rate of 26% in 41 cases of resected ipsilateral lesions, with no significant difference between unilobar or multilobar location. Yoshino and colleagues (1997) found a similar late survival in resected unilobar multifocal cases. In a series of 36 cases, Deschamps and colleagues (1990) reported survival rates of 16% and 14% at 5 and 10 years, respectively. If both tumors considered separately were stage I, the survival rate was 25% at both 5 and 10 years. Rosengart (1991), Pommier (1996), and Okada (1998) and their associates also noted varying but favorable results, with late survival rates of 44%, 24%, and 70%, respectively. These authors also emphasized that low stage per tumor correlates with improved results. In contrast, Ribet and Dambron (1995) and Adebonojo and colleagues (1997) had no 5-year survivors among synchronous cases, although each series consisted of only 15 patients, and each had a high proportion of bilateral resections. In addition, it is noteworthy that in the latter series, median survival was excellent (43 months) despite absence of 5-year survival. Okumura and co-workers (2001) reported a 34% late survival for multifocal pulmonary T4 and 11% for synchronous sites in different lobes. In a review of the literature, Urschel and colleagues (1998) determined that long-term survival following resection of multifocal T4 was about 20%. Battafarano and associates (2002) achieved a postresection 3-year survival rate of 66% for multifocal T4N0.

Although now classified with generally unresectable NSCLC, resection should be considered in patients thought to have limited, node-negative multifocal cancer, especially when unilateral. It is likely that some of the discrepancy among series of synchronous sites results from unavoidably inaccurate classification, thereby improving results for proposed metastasis and worsening results for true multiple primaries. More accurate differentiation by DNA studies and other assays, such as the p53 gene mutation described by Matsuzoe and colleagues (1999), is essential in future reports to clarify this important surgical issue. The role of systemic treatment as induction or adjuvant therapy requires further study.

Lung Transplantation for Diffuse Bronchioloalveolar Carcinoma

A small number of patients (about 10% to 15%) with BAC develop diffuse multicentric unilateral or bilateral involvement of the lungs. Moreover, in an additional 22% of cases in the surgical series reported by Daly and associates (1991), a pneumoniclike involvement of a lobe or an entire lung was present. The latter usually is pure BAC without invasion, but this variety, according to Daly (1991), Breathnach (1999), Okubo (1999), and Ebright (2002) and their associates, has the poorest prognosis for disease-free and long-term survival. In Ebright and colleagues' report (2002), the median disease-free survival was 18.7 months, and the overall median long-term survival was 48.7 months; in Daly and associates' series (1991), it was 36 months. Nonetheless, lobectomy and pneumonectomy have been done in these patients with unilateral disease for possible cure. Mediastinal lymph node involvement is rare (<1% to 5.7%), as is distant metastasis. Patients with stage IIIB and IV (multicentric) disease may also be surgical candidates when the disease is unilateral; the median disease-free survival time in Ebright and colleagues' study (2002) was 40.4 months, and the 5-year survival rate was just under 60%.

In both groups (multicentric and pneumonic), recurrences or a new primary BAC may occur in both the remaining ipsilateral and contralateral lung. A variable number of these patients will progressively proceed to a state of incapacitating respiratory failure with severe dyspnea.

In 1999, Garver and coinvestigators at the University of Alabama reported their initial experience with the use of bilateral lung transplantation (LTX) in five patients and a single LTX in two patients, both of whom had previously undergone a pneumonectomy, who had developed end-stage but still resectable multicentric BAC disease with severe respiratory failure. Rigorous evaluation excluded both the absence of mediastinal lymph node involvement and the presence of distant metastasis. Unfortunately, one patient was found to have an adenocarcinoma on final pathologic examination with involvement of a mediastinal lymph node in station 7. This patient died of distant metastatic disease within 38 months of the operation. Of the remaining six patients with BAC, four had recurrence in one or both of the donor lungs. The authors believed that the origin was from the recipients of the transplanted lungs in at least three of the four patients. Zorn and colleagues (2003) of the same group added two additional cases treated with bilateral LTX and reviewed the long-term results in the eight patients with end-stage BAC who were managed by LTX. Four of the patients (two without any recurrence and two with recurrence,

P.1578


one of whom was managed by subsequent left- and right-sided lobectomies 39 and 75 months after the transplantation) were alive at 53, 76, 87, and 89 months, respectively. The four other patients, despite a single retransplantation in two, had died at 15, 21, 33, and 82 months after the initial LTX. Thus, 50% of the patients remain alive, and 50% survived longer than 5 years after LTX. Nonetheless, because of the high incidence of recurrence (75%), the aforementioned group discontinued their study.

In an interesting editorial, Egan and Detterbeck (2003) noted, however, that the results obtained by Zorn and associates (2003) were as satisfactory as LTX for end-stage lung disease due to benign disease. Thus, pragmatically, appropriately evaluated patients with severe dyspnea due to widespread but locally confined pulmonary involvement due to BAC could be considered candidates for palliative lung transplantation. Whether or not this gains acceptance remains to be seen.

Other Prognostic Factors

Although the TNM stage is currently the most clinically useful determinant of outcome and the basis of most surgical decisions and reporting, numerous other factors have been variably shown to affect prognosis. Multivariate analysis including both clinical and biologic factors, as reported by Harpole and colleagues (1995a, 1995b), will likely become increasingly applied and prospectively more helpful as further data are accumulated. At present, however, the practical implications of individual factors other than TNM predictors and the obvious issue of performance status are limited. The impact of the extent of resection, with the exception of lesser resection versus lobectomy, as discussed previously, is related to the completeness of resection and the TNM stage. Without these specifics, general comparisons of lobectomy to pneumonectomy are not useful. Brundage and colleagues (2002) have provided a recent comprehensive review of prognostic indices in surgical and nonsurgical lung cancer.

Table 106-9. Reported Experience with Pulmonary Resection for Lung Cancer in Octogenarians

Report No. of Years (Dates) No. of Resections Postoperative Stay (days) Morbidity (Major) (%) Operative Mortalitya (%) 5-Year Survival in All Patients (%) 5-Year Survival, Stage I (%)
Ginsberg et al (1983) 3 (1979 1981) 37 NS NS 8.1 NS NS
Shirakusa et al (1989) 10 (1978 1987) 33 NS 51 13 55 79
Osaki et al (1994) 17 (1974 1991) 31 NS 67 21 32 38
Naunheim et al (1994) 10 (1981 1991) 37 14 8.8 45 (30) 16 30 NS
Riquet et al (1994) 6 (1984 1990) 11 NS NS 12.1 16b 30.1b
Harvey et al (1995) 9 (1985 1994) 17 NS NS 17.6 42c 65c
Pagni et al (1997) 16 (1980 1995) 54 8.1 3.5 45 (11) 3.7 43 57
NS, not stated.
a Operative mortality reflects all deaths within 30 days or same hospitalization for resected patients, as determined from the data reported.
b Long-term survival for all patients older than 75 years.
c Long-term survival for all patients older than 70 years.
From Pagni S, et al: Pulmonary resection for lung cancer in octogenarians. Ann Thorac Surg 63:785, 1997. With permission.

Gender, Age, and Race

Occasional reports have noted a better postresection prognosis for women than for men. Alexiou and colleagues (2002) found a significantly higher late survival for women with stage I cancer (56% versus 42% for men) but no difference in stages II and III. Gender is not a prognostic factor in most series. Albain (1998) has recently commented on the lack of current evidence of a differential outcome by gender in NSCLC. With respect to age, it has been claimed that patients older than 70 years of age have a poorer late outcome than younger patients. Ishida and associates (1990b), however, showed similar survival stage for stage. Current attention has shifted to the role of resection in octogenarians. The acute risk of operation is higher in people older than 80 years of age, but octogenarians who have undergone resection have an acceptable late outcome, with 5-year survival rates in stage I cases ranging from 30% to as high as 79% (Table 106-9), as summarized by Pagni and associates (1997). Focusing on younger patients, Gadgeel and co-workers (1999) found no difference in survival for similarly staged patients younger than 50 years versus older, although younger patients tended to present with more advanced disease.

Cancer statistics for the years 1989 to 1996, as reported by Greenlee and associates (2001), show a wide variation in lung cancer incidence based on race. The lowest rates were in American Indian men (14 of 100,000) and Japanese women (15 of 100,000). The highest rates were in African-American men (117 of 100,000) and Alaskan native women (50 of 100,000). African Americans were also found to have lower overall survival rates. Gadgeel and colleagues (2001) also showed a less favorable outcome for African Americans with lung cancer, including local disease. A study by Bach and associates (1999) of 860 African Americans and

P.1579


10,124 whites who underwent resection for stage I or II NSCLC, however, showed no difference in late survival (39% and 43%, respectively). There was also no difference in survival for those who did not undergo resection. The discrepancy in overall survival may be due to the finding that a significantly smaller proportion of blacks than whites were treated surgically (64% versus 77%). This area clearly needs further study.

Histologic Features

The influence of cell type has been discussed within the TNM stages. When matched for other variables, squamous histology may have a slightly more favorable prognosis than the collective group of nonsquamous NSCLC, but the differential is usually not statistically significant, especially in higher stages, and rarely affects therapeutic decisions. In fact, patients with adenocarcinoma in some series fare better than those with squamous cancer. Unifocal bronchoalveolar carcinoma is associated with postresection survival rates of about 90% and 55%, for T1N0 and T2N0, respectively. Large cell neuroendocrine and giant cell undifferentiated variants of NSCLC have often been noted to predict a poor surgical outcome, as have tumors made up of mixed cell types. Intermingling of small cell elements and NSCLC occurs in less than 1% of surgical cases. The surgical prognosis for these lesions is similar to that for pure small cell cancers, as discussed by Hage and colleagues (1998). Patients found to have mixed small cell histology should be treated with adjuvant chemotherapy.

Simple histopathologic features presumed to be associated with clinical neoplastic aggressiveness, such as lesser degree of differentiation, high mitotic index, and vascular or lymphatic invasion, have often been correlated with a poor prognosis. Other histopathologic findings, such as the presence of fibrosis and local infiltration by plasma cells or lymphocytes, have been suggested as influencing prognosis favorably. In summary, cell type and basic histologic features are not powerful prognostic indicators, and they rarely influence surgical decision pathways or the use of adjuvant therapy.

Molecular Biological and Other Markers

Despite its utility, postresection TNM staging of lung cancer does not predict which patients within each stage are likely to have recurrent cancer and which are unlikely. This problem is especially vexing in early-stage, surgically curable NSCLC. As noted, one fourth to one third of patients with T1N0 cancer will die of their disease, either from locoregional recurrence, distant metastases, or both. When more sophisticated assays are applied, clinical correlation is improved. This has led to a burgeoning interest in serum and biologic markers, alone or in groups, as prognostic factors. In addition, multivariate assessment of these indicators coupled with TNM factors and patient demographics appears promising. The molecular biologic aspects of lung cancer are discussed in Chapter 102. This section presents a few examples.

An example of this approach is the report by D'Amico and associates (1999), who assessed a panel of 10 markers in resected stage I NSCLC, reflecting all phases of tumor growth and spread (growth regulation, cell cycle regulation, apoptosis, angiogenesis, and cellular adhesion). Five markers were shown on multivariate analysis to be independent predictors of recurrence. In decreasing order of predictive value, with 5-year survival rates for positive versus negative cases in parentheses, these included p53 mutation (52% vs. 70%), hot-area angiogenesis factor VIII (56% vs. 70%), the protooncogene erb-b2 (47% vs. 67%), and the adhesion protein CD-44 (54% vs. 67%). Presence of the fifth marker (Rb, a cell cycle regulator) was a positive predictor, with a 63% late survival rate for positive cases and 55% when absent. The same group, in a report by Kwiatkowski and associates (1998), evaluated 24 patient, histologic, and molecular indices to assess their impact on survival in 244 resected stage I lung cancers. Nine factors were found individually to have a negative impact: male gender; age more than 60 years; wedge resection; tumor size 4 cm or more; solid mucinous adenocarcinoma; lymphatic invasion; p53 gene expression; ras mutation; and the absence of H-ras p21 expression. When all nine factors were taken into account, increasing numbers of negative predictors stratified survival (Fig. 106-7). When fewer than three were found, the 5-year survival rate was 91%. In cases with more than five factors present, the survival rate was 19%. In a subsequent report, D'Amico and co-workers (2001) raised the exciting possibility that certain patterns of molecular biologic markers might predict an increased risk for the development of brain metastases after resection of stage I lung cancer. If validated, this information could lead to recommendations for cranial radiation in such cases. The preliminary reports of D'Cunha and colleagues (2002) and of Ahrendt and associates (2002) noted that the application of molecular biological techniques to histologically negative lymph nodes may detect the presence of micrometastasis in a significant proportion of stage I patients.

At the opposite end of the spectrum from complex techniques and analyses, serum carcinoembryonic antigen (CEA) is believed by many to be a useful and simple single prognostic test. Many reports on this marker have appeared from Japan. Sawabata and colleagues (2002b), for example, found 5-year survival rates of 49% versus 72% for resected stage I lung cancer patients with a preoperative CEA level higher than 7 ng/mL (56 patients), as compared with those with a normal CEA level (241 patients). More striking was the finding that the survival rate fell to 18% for 15 patients whose CEA level remained elevated postoperatively, whereas the survival rate was 68% among 42 patients whose levels normalized. Buccheri and Ferrigno (2003) also found CEA useful in identifying patients at increased risk for recurrence and neoadjuvant trials in such cases.

Fig. 106-7. Cancer-free survival stratified according to number of risk factors: male gender; older than 60 years of age; wedge resection; lymphatic invasion; solid tumor with mucin production; tumor size greater than 4 cm; p53 expression; k-ras radon 12 mutation; and H-ras p21 lack of expression. From Kwaitkowski D, et al: Molecular pathologic substaging in 244 stage I non-small cell lung cancer patients: clinical implications. J Clin Oncol 16:2468, 1998. With permission.

P.1580


It is likely that, with further experience, information of this type will become part of the evaluation of all patients with lung cancer to identify those who might benefit from induction or adjuvant therapy or require special surveillance.

Radiographic Features

There has been recent interest in the potential ability of PET scan and CT scan to predict tumor behavior and to assess response to induction therapy before resection. Ahuja and associates (1998) found that an increased standard uptake value (SUV) for FDG correlated with a worse prognosis in lung cancer in general. In a surgical series reported by Dhital and colleagues (2000), there was a stepwise decreased 1-year survival rate with increasing SUV, with the extremes having 75% early survival for an SUV of less than 10 and 17% for an SUV of more than 20. Vansteenkiste and co-workers (1999) found a difference in postresection survival between patients with an SUV cutoff of 7. Both these series, however, combine stages I to III, making conclusions about early-stage NSCLC difficult. In the latter study, however, it is reported that among patients with tumors smaller than 3 cm, those with lower SUV experienced a 2-year survival rate of 86%, as compared with 60% for higher SUV. Validation of these findings and determination of whether they offer any advantage over basic histology and TNM stage must await further studies. Standard uptake volume may also indicate response to nonoperative treatment. The surgical implications of this might include decisions to proceed to resection or continue induction treatment.

The feature of CT scan that has been reported to correlate with low-grade lung cancers is the presence of ground-glass opacity (GGO). Suzuki and associates (2002) reported on 69 surgical patients who had a significant component of GGO on preoperative CT. All patients had stage I cancers, 47 of which were BAC and 22, adenocarcinoma. No patient had lymphatic invasion, and only two demonstrated vascular invasion. Watanabe and colleagues (2002) found no recurrence in 17 cases of BAC with pure GGO treated by local excision at a mean followup of 32 months and noted three cases of GGO that turned out to be atypical adenomatous hyperplasia (AAH). Matsuguma and co-workers (2002) classified 96 patients with clinical T1N0M0 adenocarcinoma into five groups based on increasing proportions of GGO. Seventeen cases turned out to have pathologic N1 or N2 disease, none of whom were in the upper two quintiles of GGO. Furthermore, none of the patients with GGO of more than 50% were found to have lymphatic invasion,

P.1581


and none recurred during medium-term follow-up. All lesions 2 cm or smaller and with more than 50% GGO were BAC. More work needs to be done in this area with respect to indications for limited resection and postoperative prognosis. Heitmiller (2002) and Rusch (2002) discuss the evolution of understanding of peripheral lung cancer and its transition from preinvasive to invasive forms.

Miscellaneous Prognostic Factors

Trotter and associates (1984) first reported an adverse effect on survival of perioperative blood transfusion in patients with stage I disease, and Hyman and colleagues (1985) supported this observation. Moores and colleagues (1989) also found a deleterious effect of blood transfusion in the data of the LCSG. Little and associates (1990) also provided supportive data. The series reported by Pastorino (1986) and Keller (1988b) and their associates, however, showed no correlation between blood transfusion or its amount and late survival in limited lung cancer. Elevation of the white blood cell count, serum lactic dehydrogenase (LDH), and CEA, as well as decreased hemoglobin level, are among basic blood tests that have been proposed to affect prognosis, but the data are sparse.

Recurrence After Resection

Much has been written about recurrence rates and locations after surgical treatment of NSCLC. In brief, surgical patients with NSCLC remain at risk for recurrence of the original tumor and the development of new primary lung cancers, as discussed earlier. Although the rate of recurrence falls over time, metastasis can appear many years after initial diagnosis and treatment. Among 62 patients who were found to have new lesions 5 or more years after resection, Martini and associates (1999) noted that 26 experienced recurrence of the original tumor, and 36 cases represented new primaries. Lifelong surveillance is mandatory. Most treatment failures occur within the first 3 years. These cases, especially those with very early clinical recurrence, likely represent instances of failure of the initial staging evaluation to detect the true extent of disease, unconfirmed incomplete resection, or both. In older series, between 10% and 30% of operated patients who died soon after surgery had unsuspected grossly evident metastases identified at autopsy. Although Stenbygaard and associates (1995) found no postmortem evidence of distant disease in eight patients who died within 30 days of resection for stage I or II adenocarcinoma performed between 1981 and 1985, data in this area are limited, and the incidence of micrometastasis in this group is unknown. Maruyama and colleagues (1997) found that, in patients with resected pathologic stage I NSCLC, unsuspected micrometastases to the N1 and N2 lymph nodes can often be found by monoclonal antibody stains, and that this finding correlates with early relapse.

Locoregional recurrence is less common than the appearance of distant metastasis in patients with lower-stage NSCLC following complete resection. The sites of first failure were distant metastases to the brain, bone, liver, or contralateral lung in most of the patients reported by Mountain (1980), Martini (1983), Feld (1984), Pairolero (1984), and Ichinose (1989, 1995) and their associates. In each, the brain was the most common site. In contrast, the Ludwig Lung Cancer Study Group (1987) reported the highest initial failure rate in patients with stage I disease to be in the ipsilateral hemithorax, as did Iascone and associates (1986).

Summary

Primary operation is the optimal current therapeutic approach for patients with clinical stage IA NSCLC. Although no other therapy and no combination of operation with other modalities has as yet been proved to surpass resection in stages IB, IIA and IIB, disappointment with primary surgical results has led to preliminary explorations of initial nonsurgical treatment. Higher-stage cases should undergo initial operation only in carefully selected instances. The role of operation as part of a planned multimodality approach in these patients remains under investigation. Accurate clinical staging is currently the most important factor for determining the place of operation as primary therapy in individual patients. It is likely, however, that the next several years will witness an expansion of understanding of novel preoperative and postresection means of assessing the risk for residual or recurrent lung cancer and thereby allow improved treatment and surveillance pathways.

REFERENCES

Adebonojo SA, Moritz DM, Danby CA: The results of modern surgical therapy for multiple primary lung cancers. Chest 112:693, 1997.

Ahrendt SA, et al: Molecular assessment of lymph nodes in patients with resected stage I non-small cell lung cancer: preliminary results of a prospective study. J Thorac Cardiovasc Surg 123:466, 2002.

Ahuja V, et al: The prognostic significance of fluorodeoxyglucose positron emission tomography imaging for patients with non-small cell lung carcinoma. Cancer 83:918, 1998.

Akay H, et al: Surgical treatment of peripheral lung cancer adherent to the parietal pleura. Eur J Cardiothorac Surg 22:615, 2002.

Albain K: Invited Commentary on Ouellette D, et al: Lung cancer in women compared with men: stage, treatment and survival. Ann Thorac Surg 66:1140, 1998.

Albertucci M, et al: Surgery and the management of peripheral lung tumors adherent to the parietal pleura. J Thorac Cardiovasc Surg 103:8, 1992.

Alexiou C, et al: Do women live longer than men following lung resection for carcinoma? Eur J Cardiothorac Surg 21:319, 2002.

Allen MS, et al: Bronchogenic carcinoma with chest wall invasion. Ann Thorac Surg 51:948, 1991.

Anderson TM, May PM, Holmes EC: Factors affecting survival in superior sulcus tumors. J Clin Oncol 4:1598, 1986.

Andre F, et al: Survival of patients with resected N2 non-small cell lung cancer: evidence for a subclassification and implications. J Clin Oncol 18:2981, 2000.

Antakli T, et al: Second primary lung cancer. Ann Thorac Surg 59:863, 1995.

P.1582


Attar S, et al: Superior sulcus (Pancoast) tumor: experience with 105 patients. Ann Thorac Surg 66:193, 1998.

Ayabe H, et al: Surgical management of adrenal metastasis from bronchogenic carcinoma. J Surg Oncol 58:149, 1995.

Bach PB, et al: Racial differences in the treatment of early stage lung cancer. N Engl J Med 341:1198, 1999.

Battafarano RJ, et al: Surgical resection of multifocal non-small cell lung cancer is associated with prolonged survival. Ann Thorac Surg 74:988, 2002.

Bernard A, et al: Risk analysis and long-term survival in patients undergoing resection of T4 lung cancer. Eur J Cardiothorac Surg 20:344, 2001.

Billing PS, et al: Surgical treatment of primary lung cancer with synchronous brain metastases. J Thorac Cardiovasc Surg 122:548, 2001.

Bonnette P, et al: Surgical management of non-small cell lung cancer with synchronous brain metastases. Chest 119:1469, 2001.

Boushy SF, et al: Clinical course related to preoperative and postoperative pulmonary function in patients with bronchogenic carcinoma. Chest 59: 383, 1971.

Breathnach OS, et al: Clinical features of patients with stage IIIB and IV bronchioloalveolar carcinoma of the lung. Cancer 86:1165, 1999.

Breathnach OS, et al: Bronchioloalveolar carcinoma of the lung: recurrences and survival in patients with stage I disease. J Thorac Cardiovasc Surg 121:42, 2001.

Brewer LA: Patterns of survival in lung cancer. Chest 71:644, 1977.

Brundage MD, Davies D, Mackillop WJ: Prognostic factors in non-small cell lung cancer: a decade of progress. Chest 122:1037, 2002.

Buccheri G, Ferrigno D: Identifying patients at risk of early postoperative recurrence of lung cancer: a new use of the old CEA test. Ann Thorac Surg 75:973, 2003.

Bulzebruck H, et al: Validation of the TNM classification for lung cancer: first results of a prospective study of 1086 patients with surgical treatment. Eur J Cardiothorac Surg 5:356, 1991.

Burt ME: Role of surgery in the treatment of patients with solitary metastasis from non-small cell lung cancer. In Aisner J, et al (eds): Comprehensive Textbook of Thoracic Oncology. Baltimore: Williams & Wilkins, 1996, p. 416.

Burt ME, et al: Results of surgical treatment of stage III lung cancer invading the mediastinum. Surg Clin North Am 67:987, 1987.

Burt ME, et al: Resection of brain metastases from non-small cell lung carcinoma: results of therapy. J Thorac Cardiovasc Surg 103:399, 1992.

Canver CC, et al: Pulmonary resection combined with cardiac operations. Ann Thorac Surg 50:796, 1990.

Carbone E, et al: T2 tumors larger than five centimeters in diameter can be upgraded to T3 in non-small cell lung cancer. J Thorac Cardiovasc Surg 122:907, 2001.

Carretta A, et al: Evaluation of radiological and pathological prognostic factors in surgically-treated patients with bronchoalveolar carcinoma. Eur J Cardiothorac Surg 20:367, 2001.

Casillas M, et al: Surgical treatment of lung carcinoma involving the chest wall. Eur J Cardiothorac Surg 3:325, 1989.

Chapelier A, et al: Factors affecting long term survival after en-bloc resection of lung cancer invading the chest wall. Eur J Cardiothorac Surg 18:513, 2000.

Churchill ED, Belsey R: Segmental pneumonectomy in bronchiectasis; lingula segment of left upper lobe. Ann Surg 109:481, 1939.

Cortese DA, Edell ES, Kinsey JH: Photodynamic therapy for early-stage squamous carcinoma of the lung. Mayo Clin Proc 72:595, 1997.

Cortese DA, et al: Roentgenographically occult lung cancer: ten-year experience. J Thorac Cardiovasc Surg 86:373, 1983.

Cybulsky IJ, et al: Prognostic significance of computed tomography in resected N2 lung cancer. Ann Thorac Surg 54:533, 1992.

Daly BDT, Feins NR: The diaphragm. In Kaiser LR, Kron IL, Spray TL (eds): Mastery of Cardiothoracic Surgery. Philadelphia: Lippincott-Raven, 1998, p. 196.

Daly BDT, et al: N2 lung cancer: outcome in patients with false-negative computed tomographic scans of the chest. J Thorac Cardiovasc Surg 105:904, 1993.

Daly RC, et al: Bronchoalveolar carcinoma: factors affecting survival. Ann Thorac Surg 51:368, 1991.

Damhuis RA, Schutte PR: Resection rates and postoperative mortality in 7,899 patients with lung cancer. Eur Respir J 9:7, 1996.

D'Amico TA, et al: A biologic risk model for stage I lung cancer: immunohistochemical analysis of 408 patients with the use of ten molecular markers. J Thorac Cardiovasc Surg 117:736, 1999.

D'Amico TA, et al: Predicting the sites of metastases from lung cancer using molecular biologic markers. Ann Thorac Surg 72:1144, 2001.

Danton MH, et al: Simultaneous cardiac surgery with pulmonary resection: presentation of a series and review of the literature. Eur J Cardiothorac Surg 13:667, 1998.

Dartevelle PG, et al: Replacement of the superior vena cava with polytetrafluoroethylene grafts combined with resection of mediastinal-pulmonary malignant tumors: report of thirteen cases. J Thorac Cardiovasc Surg 94:361, 1987.

Dartevelle PG, et al: Long-term followup after prosthetic replacement of the superior vena cava combined with resection of mediastinal-pulmonary malignant tumors. J Thorac Cardiovasc Surg 102:259, 1991.

Dartevelle PG, et al: Anterior transcervical-thoracic approach for radical resection of lung tumors invading the thoracic inlet. J Thorac Cardiovasc Surg 105:1025, 1993.

Dartevelle PG, et al: Tracheal-sleeve pneumonectomy for bronchogenic carcinoma: update 1995. Ann Thorac Surg 60:1854, 1995.

Datta D, Lahiri B: Preoperative evaluation of patients undergoing lung resection surgery. Chest 123:2096, 2003.

D'Cunha J, et al: Molecular staging of lung cancer: real-time polymerase chain reaction estimation of lymph node micrometastatic tumor cell burden in stage I non-small cell lung cancer preliminary results of Cancer and Leukemia Group B Trial 9761. J Thorac Cardiovasc Surg 123:484, 2002.

Decker DA: The significance of cytologically negative pleural effusion in bronchogenic carcinoma. Chest 74:640, 1978.

DeGiacomo T, et al: Thoracoscopic staging of IIIB non-small cell lung cancer before neoadjuvant therapy. Ann Thorac Surg 64:1409, 1997.

DeMeester TR, et al: Management of tumor adherent to the vertebral column. J Thorac Cardiovasc Surg 97:373, 1989.

Deschamps C, et al: Multiple primary lung cancers: results of surgical treatment. J Thorac Cardiovasc Surg 99:769, 1990.

Deslauriers J, et al: Carcinoma of the lung: evaluation of satellite nodules as a factor influencing prognosis after resection. J Thorac Cardiovasc Surg 97:504, 1989.

Deslauriers J, et al: Staging and management of lung cancer: sleeve resection. World J Surg 17:712, 1993.

Detterbeck FC, Socinski MA: IIB or not IIB: the current question in staging non-small cell lung cancer. Chest 112:229, 1997.

Dhital K, et al: Fluorodeoxyglucose positron emission tomography and its prognostic value in lung cancer. Eur J Cardiothorac Surg 18:425, 2000.

Di Carlo I, et al: Liver metastases from lung cancer: is surgical resection justified? Ann Thorac Surg 76:291, 2003.

Doddoli C, et al: Is lung cancer surgery justified in patients with direct mediastinal invasion? Eur J Cardiothorac Surg 20:339, 2001.

Downey RJ, et al: Extent of chest wall invasion and survival in patients with lung cancer. Ann Thorac Surg 68:188, 1999.

Dressler CM, Fratelli C, Babb J: Prognostic value of positive pleural lavage in patients with lung cancer resection. Ann Thorac Surg 67:1435, 1999.

Ebright MI, et al: Clinical pattern and pathologic stage but not histologic features predict outcome for bronchioloalveolar carcinoma. Ann Thorac Surg 74:1640, 2002.

Egan TM, Detterbeck FC: The ABC's of LTS for BAC [Editorial]. J Thorac Cardiovasc Surg 125:20, 2003.

Elia S, et al: Surgical treatment of lung cancer invading chest wall: a retrospective analysis of 110 patients. Eur J Cardiothorac Surg 20:356, 2001.

Errett LE, et al: Wedge resection as an alternative procedure for peripheral bronchogenic carcinoma in poor-risk patients. J Thorac Cardiovasc Surg 90:656, 1985.

Ettinghausen S, Burt ME: Prospective evaluation of unilateral adrenal masses in patients with operable non-small cell lung cancer. J Clin Oncol 9:1462, 1991.

Faber LP: Results of surgical treatment of stage III lung carcinoma with carinal proximity: the role of sleeve lobectomy versus pneumonectomy and the role of sleeve pneumonectomy. Surg Clin North Am 67:1001, 1987.

Faber LP, et al: Preoperative chemotherapy and irradiation for stage III non-small cell lung cancer. Ann Thorac Surg 47:669, 1989.

Facciolo F, et al: Chest wall invasion in non-small cell lung carcinoma: a rationale for en bloc resection. J Thorac Cardiovasc Surg 121:649, 2001.

Fang D, et al: Results of surgical resection of patients with primary lung cancer: a retrospective analysis of 1,905 cases. Ann Thorac Surg 72: 1155, 2001.

Feld R, et al: Site of recurrence in resected stage I non-small cell lung cancer: a guide for future studies. J Clin Oncol 2:1352, 1984.

P.1583


Ferguson MK, et al: Diffusion capacity predicts morbidity and mortality after pulmonary resection. J Thorac Cardiovasc Surg. 96:894, 1991.

Fernando HC, Goldstraw P: The accuracy of clinical evaluative intrathoracic staging in lung cancer as assessed by postsurgical pathologic staging. Cancer 65:2503, 1990.

Fukuse T, Wada H, Hitomi S: Extended operation for non-small cell lung cancer invading great vessels and left atrium. Eur J Cardiothorac Surg 11:664, 1997a.

Fukuse T, et al: Prognosis of ipsilateral intrapulmonary metastases in resected nonsmall cell lung cancer. Eur J Cardiothorac Surg 12:218, 1997b.

Funatsu T, et al: The role of mediastinoscopic biopsy in preoperative assessment of lung cancer. J Thorac Cardiovasc Surg 104:1688, 1992.

Gadgeel SM, et al: Lung cancer in patients < 50 years of age: the experience of an academic multidisciplinary program. Chest 115:1232, 1999.

Gadgeel SM, et al: Impact of race in lung cancer. Chest 120:55, 2001.

Gail MH, et al: Prognostic factors in patients with resected stage I non-small cell lung cancer. Cancer 54:1802, 1984.

Gandhi S, et al: A multidisciplinary surgical approach to superior sulcus tumors with vertebral invasion. Ann Thorac Surg 68:1778, 1999.

Garcia-Yuste M, et al: Prognostic factors in neuroendocrine lung tumors: a Spanish multicenter study. Ann Thorac Surg 70:258, 2000.

Garver RI Jr, et al: Recurrence of bronchioalveolar carcinoma in transplanted lungs. N Engl J Med 340:1071, 1999.

Gdeedo A, et al: Comparison of imaging TNM (iTNM) and pathological TNM (pTNM) in staging of bronchogenic carcinoma. Eur J Cardiothorac Surg 12:224, 1997.

Gebitekin C, et al: Fate of patients with residual tumor at the bronchial resection margin. Eur J Cardiothorac Surg 8:339, 1994.

Gerson MC, et al: Prediction of cardiac and pulmonary complications related to abdominal and noncardiac thoracic surgery in geriatric patients. Am J Med 88:101, 1990.

Ghiribelli C, et al: Treatment and survival after lung resection for non-small cell lung cancer in patients with microscopic residual disease at the bronchial stump. Eur J Cardiothorac Surg 16:555, 1999.

Ginsberg RJ, et al: Modern 30-day operative mortality for surgical resections in lung cancer. J Thorac Cardiovasc Surg 86:654, 1983.

Ginsberg RJ, et al: Influence of surgical resection and brachytherapy in the management of superior sulcus tumor. Ann Thorac Surg 57:1440, 1994.

Ginsberg RJ, Rubinstein LV: Randomized trial of lobectomy versus limited resection for T1N0 non-small cell lung cancer: Lung Cancer Study Group. Ann Thorac Surg 60:615, 1995.

Goldstraw P: Mediastinoscopy 1992: Is it still essential? Lung Cancer 8:79, 1992.

Goldstraw P, et al: Surgical management of non-small cell lung cancer with ipsilateral mediastinal node metastasis (N2 disease). J Thorac Cardiovasc Surg 107:19, 1994.

Graham ANJ, et al: Systematic nodal dissection in the intrathoracic staging of non-small cell lung cancer. J Thorac Cardiovasc Surg 117:246, 1999.

Graham EA, Singer JJ: Successful removal of the entire lung for carcinoma of the bronchus. JAMA 101:1371, 1933.

Granone P, et al: Non-small cell lung cancer with single brain metastasis: the role of surgical treatment. Eur J Cardiothorac Surg 20:361, 2001.

Greenlee RT, et al: Cancer statistics 2001. CA Cancer J Clin 51:15, 2001.

Grunenwald D, et al: Total vertebrectomy for en bloc resection of lung cancer invading the spine. Ann Thorac Surg 61:723, 1996.

Grunenwald DH, et al: Radical en bloc resection for lung cancer invading the spine. J Thorac Cardiovasc Surg 123:271, 2002.

Hage R, et al: Surgery for combined type small cell lung carcinoma. Thorax 53:450, 1998.

Hagen MP, et al: Superior sulcus lung tumors: impact of local control on survival. J Thorac Cardiovasc Surg 117:1086, 1999.

Hara N, et al: Assessment of the role of surgery for stage III bronchogenic carcinoma. J Surg Oncol 25:153, 1984.

Harpole DH, et al: Stage I non-small cell lung cancer: a multivariate analysis of treatment methods and patterns of recurrence. Cancer 76:787, 1995a.

Harpole DH, et al: A prognostic model of recurrence and death in stage I non-small cell lung cancer utilizing presentation, histopathology and oncoprotein expression. Cancer Res 55:51, 1995b.

Harviel JD, McNamara JJ, Straehley CJ: Surgical treatment of lung cancer in patients over the age of 70 years. J Thorac Cardiovasc Surg 75:802, 1978.

Harvey JC, et al: Surgical treatment of non-small cell lung cancer in patients older than seventy years. J Surg Oncol 60:247, 1995.

Heitmiller RF: Invited commentary on Watanabe S, et al: Results of wedge resection for focal bronchioloalveolar carcinoma showing pure ground glass attenuation on computed tomography. Ann Thorac Surg 73:1075, 2002.

Higashiyama M, et al: Pleural lavage cytology immediately after thoracotomy and before closure of the thoracic cavity for lung cancer without pleural effusion and dissemination: clinicopathologic and prognostic analysis. Ann Surg Oncol 4:409, 1997.

Higashiyama M, et al: Prognostic value of bronchiolo-alveolar carcinoma component of small lung adenocarcinoma. Ann Thorac Surg 68:2069, 1999.

Hiroshima K, et al: Prognostic significance of neuroendocrine differentiation in adenocarcinoma of the lung. Ann Thorac Surg 73:1732, 2002.

Holmes EC: Treatment of stage II lung cancer. Surg Clin North Am 67: 945, 1987.

Horita K, Itho T, Ueno T: Radical operation using cardiopulmonary bypass for lung cancer invading the aortic wall. Thorac Cardiovasc Surg 41: 130, 1993.

Hsia JY, et al: Adenosquamous carcinoma of the lung: surgical results compared with squamous cell and adenocarcinoma. Scand Cardiovasc J 33:29, 1999.

Hughes KS, Sugarbaker PH: Treatment of metastatic liver tumors. In Rosenberg SA (ed): Surgical Cure of Metastatic Cancer. Philadelphia: JB Lippincott, 1987.

Hyman NH, et al: Blood transfusions and survival after lung cancer resection. Am J Surg 149:502, 1985.

Iascone C, et al: Local recurrence of resectable non-small cell carcinoma of the lung. Cancer 57:471, 1986.

Ichinose Y, et al: Preoperative examination to detect distant metastasis is not advocated for asymptomatic patients with stages 1 and 2 non-small cell lung cancer. Chest 96:1104, 1989.

Ichinose Y, Hara N, Ohta M: Synchronous lung cancers defined by deoxyribonucleic acid flow cytometry. J Thorac Cardiovasc Surg 102:418, 1991.

Ichinose Y, et al: Is T factor of the TNM staging system a predominant prognostic factor in pathologic stage I non-small cell lung cancer? J Thorac Cardiovasc Surg 106:90, 1993.

Ichinose Y, et al: Prognostic factors obtained by a pathologic examination in completely resected non-small cell lung cancer: an analysis in each pathologic stage. J Thorac Cardiovasc Surg 110:601, 1995.

Inoue H, et al: Resection of the superior vena cava for primary lung cancer: 5 years' survival. Ann Thorac Surg 50:661, 1990.

Inoue H, et al: Prognostic assessment of 1310 patients with non-small cell lung cancer who underwent complete resection from 1980 to 1993. J Thorac Cardiovasc Surg 116:407, 1998.

Ishida T, et al: Strategy for lymphadenectomy of lung cancer less than 3 cm in diameter. Ann Thorac Surg 50:708, 1990a.

Ishida T, et al: Long-term results of operation for non-small cell lung cancer in the elderly. Ann Thorac Surg 50:919, 1990b.

Ishida T, et al: Adenosquamous carcinoma of the lung: clinicopathologic and immunohistochemical features. Am J Clin Pathol 97:678, 1992.

Iyoda A, et al: Pulmonary large cell carcinomas with neuroendocrine features are high-grade neuroendocrine tumors. Ann Thorac Surg 73:1049, 2002.

Izbicki JR, et al: Effectiveness of radical systematic mediastinal lymphadenectomy in patients with resectable non-small cell lung cancer: results of a prospective randomized trial. Ann Surg 227:138, 1998.

Jassem J, et al: Results of surgical treatment of non-small cell lung cancer: validation of the new postoperative pathologic TNM classification. J Thorac Cardiovasc Surg 119:1141, 2000.

Jensik RJ, et al: Segmental resection for lung cancer. A 15 year experience. J Thorac Cardiovasc Surg 66:563, 1973.

Jubelirer S, Wilson RA: Lung cancer in patients younger than 40 years of age. Cancer 67:1436, 1991.

Kaiser LR, et al: Significance of extramucosal residual tumor at the bronchial resection margin. Ann Thorac Surg 47:265, 1989.

Kato H, Okunaka T, Shimatani H: Photodynamic therapy for early stage bronchogenic carcinoma. J Clin Laser Med Surg 14:235, 1996.

Keller SM, Kaiser LR, Martini N: Bilobectomy for bronchogenic carcinoma. Ann Thorac Surg 45:62, 1988a.

Keller SM, et al: Blood transfusion and lung cancer recurrence. Cancer 62:606, 1988b.

Kimura H, Yamaguchi Y: Survival of noncuratively resected lung cancer. Lung Cancer 11:229, 1994.

Kirsch MM, et al: Major pulmonary resections for bronchogenic carcinoma in the elderly. Ann Thorac Surg 22:369, 1976.

Klepetko W, et al: T4 lung tumors with infiltration of the thoracic aorta: is an operation reasonable? Ann Thorac Surg 67:340, 1999.

P.1584


Kodama K, et al: Intentional limited resection for selected patients with T1N0M0 non-small cell lung cancer: a single institution study. J Thorac Cardiovasc Surg 114:347, 1997.

Kohman LJ, et al: Random versus predictable risks of mortality after thoracotomy for lung cancer. J Thorac Cardiovasc Surg 91:551, 1986.

Koike T, et al: Clinical analysis of small-sized peripheral lung cancer. J Thorac Cardiovasc Surg 115:1015, 1998.

Koike T, et al: Surgical results for centrally-located early stage lung cancer. Ann Thorac Surg 70:1176, 2000.

Kondo H, et al: Prognostic significance of pleural lavage cytology immediately after thoracotomy in patients with lung cancer. J Thorac Cardiovasc Surg 106:1092, 1993.

Korst RJ, et al: Lobectomy improves ventilatory function in selected patients with severe COPD. Ann Thorac Surg 66:898, 1998.

Kotoulas C, et al: Prognostic significance of pleural lavage cytology after resection for non-small cell lung cancer. Eur J Cardiothorac Surg 20: 330, 2001.

Kwiatkowski DJ, et al: Molecular pathologic substaging in 244 stage I non-small cell lung cancer patients: clinical implications. J Clin Oncol 16:2468, 1998.

Lacasse Y, et al: Incomplete resection in non-small cell lung cancer: need for a new definition. Ann Thorac Surg 65:220, 1998.

Lam S: Photodynamic therapy of lung cancer. Semin Oncol 21:15, 1994.

Landreneau RJ, et al: Wedge resection versus lobectomy for stage I (T1N0M0) non-small cell lung cancer. J Thorac Cardiovasc Surg 113: 691, 1996.

Liewald F, et al: Importance of microscopic residual disease at the bronchial margin after resection of non-small cell carcinoma of the lung. J Thorac Cardiovasc Surg 104:408, 1992.

Lindell G, et al: Liver resection of noncolorectal secondaries. J Surg Oncol 69:66, 1998.

Linnoila RI, Piantadosi S, Ruckdeschel JC: Impact of neuroendocrine differentiation in non-small cell lung cancer. The LCSG experience. Chest 106:367S, 1994.

Little AG, et al: Modified stage I (T1N0M0, T2N0M0) non-small cell lung cancer: treatment results, recurrence patterns and adjuvant immunotherapy. Surgery 100:621, 1986.

Little AG, et al: Perioperative blood transfusion adversely affects prognosis of patients with stage I non-small cell lung cancer. Am J Surg 160: 630, 1990.

Lopez-Encuentra A, et al: Staging in lung cancer: Is 3 cm a prognostic threshold in pathologic stage I non-small cell lung cancer? Chest 121: 1515, 2002.

Ludwig Lung Cancer Study Group: Patterns of failure in patients with resected stage I and II non-small cell carcinoma of the lung. Ann Surg 205:67, 1987.

Luketich JD, Burt ME: Does resection of isolated adrenal metastases in non-small cell lung cancer (NSCLC) improve survival? Lung Cancer 10:53, 1991.

Macchiarini P, et al: Blood vessel invasion by tumor cells predicts recurrence in completely resected T1N0M0 non-small cell lung cancer. J Thorac Cardiovasc Surg 106:80, 1993.

Macheers SK, Mansour KA: Management of isolated splenic metastases from carcinoma of the lung: a case report and review of the literature. Am Surgeon 58:683, 1992.

Magdeleinat P, et al: Surgical treatment of lung cancer invading the chest wall: results and prognostic factors. Ann Thorac Surg 71:1094, 2001.

Maggi G, et al: Results of surgical resection of stage IIIa (N2) non-small cell lung cancer, according to the site of the mediastinal metastases. Int Surg 78:213, 1993.

Maggi G, et al: Combined radiosurgical treatment of Pancoast tumor. Ann Thorac Surg 57:198, 1994.

Magilligan DJ, et al: Surgical approach to lung cancer with solitary cerebral metastasis: twenty-five years experience. Ann Thorac Surg 42:360, 1986.

Manac'h D, et al: Visceral pleura invasion by non-small cell lung cancer: an underrated bad prognostic factor. Ann Thorac Surg 71:1088, 2001.

Marra A, et al: Pathologic N1 non-small cell lung cancer: correlation between pattern of lymphatic spread and prognosis. J Thorac Cardiovasc Surg 124:543, 2003.

Martini N, Flehinger BJ: The role of surgery in N2 lung cancer. Surg Clin North Am 67:1037, 1987.

Martini N, Melamed MR: Multiple primary lung cancers. J Thorac Cardiovasc Surg 70:606, 1975.

Martini N, Melamed MR: Occult carcinomas of the lung. Ann Thorac Surg 30:215 223, 1980.

Martini N, et al: Prognostic significance of N1 disease in carcinoma of the lung. J Thorac Cardiovasc Surg 86:646, 1983.

Martini N, et al: Lobectomy for stage I lung cancer. In Kittle CF (ed): Current Controversies in Thoracic Surgery. Philadelphia: W.B. Saunders, 1986, pp. 171 174.

Martini N, et al: Survival after resection of stage II non-small cell lung cancer. Ann Thorac Surg 54:460, 1992.

Martini N, et al: Management of non-small cell lung cancer with direct mediastinal involvement. Ann Thorac Surg 58:1447, 1994.

Martini N, et al: Factors influencing ten-year survival in resected stages I to IIIA non-small cell lung cancer. J Thorac Cardiovasc Surg 117:32, 1999.

Massard G, et al: Prognostic implications of a positive bronchial resection margin. Eur J Cardiothorac Surg 17:557, 2000.

Mathisen DJ, Grillo HC: Carinal resection for bronchogenic carcinoma. J Thorac Cardiovasc Surg 102:16, 1991.

Mathisen DJ, et al: Assessment of preoperative accelerated radiotherapy and chemotherapy in stage IIIA (N2) non-small cell lung cancer. J Thorac Cardiovasc Surg 111:123, 1996.

Matsuguma H, et al: Proportion of ground glass opacity on high-resolution computed tomography in clinical T1N0M0 adenocarcinoma of the lung: a predictor of lymph node metastasis. J Thorac Cardiovasc Surg 124: 278, 2002.

Matsuzoe D, et al: Discrimination of double primary lung cancer from intrapulmonary metastasis by p53 gene mutation. Br J Cancer 79:1549, 1999.

Maruyama R, et al: Relationship between early recurrence and micrometastases in the lymph nodes of patients with stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 114:535, 1997.

McCaughan BC, et al: Chest wall invasion in carcinoma of the lung: therapeutic and prognostic implications. J Thorac Cardiovasc Surg 89:836, 1985.

McCaughan BC: Primary lung cancer invading the chest wall. Chest Surg Clin North Am 4:17, 1994.

McElvaney G, et al: Multicentricity of adenocarcinoma of the lung. Chest 95:151. 1989.

Merlier M, et al: The staging issue: unification of criteria. In Delarue NC, Eschapasse H (eds): International Trends in General Thoracic Surgery. Philadelphia: W.B. Saunders, 1985, pp. 27 36.

Milledge JS, Nunn JF: Criteria of fitness for anesthesia in patients with chronic obstructive lung disease. Br Med J 3:670, 1975.

Miller DL, et al: Results of surgical resection in patients with N2 non-small cell lung cancer. Ann Thorac Surg 57:1095, 1994.

Miller JI, Mansour KA, Hatcher CR: Carcinoma of the superior pulmonary sulcus. Ann Thorac Surg 28:44, 1979.

Mitchell JD, et al: Resection for bronchogenic carcinoma involving the carina: long-term results and effect of nodal status on outcome. J Thorac Cardiovasc Surg 121:465, 2001.

Moores DWO, Piantadosi S, McKneally MF: Effect of perioperative blood transfusions on the outcome in patients with surgically resected lung cancer. Ann Thorac Surg 47:346, 1989.

Mountain CF: Expanded possibilities for surgical treatment of lung cancer: survival in stage IIIA disease. Chest 97:1045, 1990.

Mountain CF, et al: Present status of postoperative adjuvant therapy for lung cancer. Cancer Bull 32:108, 1980.

Mountain CF: Revisions in the international system for staging lung cancer. Chest 111:1710, 1997.

Muscolino G, Valente M, Andreani S: Pancoast tumours: clinical assessment and long-term results of combined radiosurgical treatment. Thorax 52:284, 1997.

Mussi A, et al: Resection of single metastasis in non-small cell lung cancer: prognostic factors. J Thorac Cardiovasc Surg 112:146, 1996.

Nakahara K, et al: Extended operation for lung cancer invading the aortic arch and superior vena cava. J Thorac Cardiovasc Surg 97:428, 1989.

Nakanishi R, et al: Treatment strategy for patients with surgically discovered N2 stage IIIA non-small cell lung cancer. Ann Thorac Surg 64:342, 1997.

Naruke T, et al: The importance of surgery to non-small cell carcinoma of the lung with mediastinal lymph node metastasis. Ann Thorac Surg 46:603, 1988.

Naruke T, et al: Prognosis and survival after resection for bronchogenic carcinoma based on the 1997 TNM staging classification: the Japanese experience. Ann Thorac Surg 71:1759, 2001.

P.1585


Naunheim KS, et al: Adenosquamous lung carcinoma: clinical characteristics, treatment and prognosis. Ann Thorac Surg 44:462, 1987.

Naunheim KS, et al: Lung cancer surgery in the octogenarian. Eur J Cardiothorac Surg 8:453, 1994.

Newton JR, Grillo HC, Mathisen DJ: Main bronchial sleeve resection with pulmonary conservation. Ann Thorac Surg 52:1272, 1991.

Ninan M, et al: Standardized exercise oximetry predicts postpneumonectomy outcome. Ann Thorac Surg 64:323, 1997.

Nugent WC, et al: Non-small cell lung cancer at the extremes of age: impact on diagnosis and treatment. Ann Thorac Surg 63:193, 1997.

Ohada M, et al: Operative approach for multiple primary lung carcinomas. J Thorac Cardiovasc Surg 115:836, 1998.

Ohta Y, et al: Clinicopathological and biological assessment of lung cancers with pleural dissemination. Ann Thorac Surg 69:1025, 2000.

Ohta Y, et al: Can tumor size be a guide for limited surgical intervention in patients with peripheral non-small cell lung cancer? Assessment from the point of view of nodal micrometastasis. J Thorac Cardiovasc Surg 122:900, 2001.

Okada M, et al: Operative approach for multiple primary lung carcinomas. J Thorac Cardiovasc Surg 115:836, 1998.

Okada M, et al: Role of pleural lavage cytology before resection for primary lung cancer. Ann Surg 229:579, 1999.

Okubo K, et al: Bronchoalveolar carcinoma: clinical, radiographic and pathologic factors and survival. J Thorac Cardiovasc Surg 118:702, 1999.

Okumura T, et al: Intrapulmonary metastasis of non-small cell lung cancer: a prognostic assessment. J Thorac Cardiovasc Surg 122:24, 2001.

Osaki T, et al: Surgical treatment of lung cancer in the octogenarian. Ann Thorac Surg 57:188, 1994.

Overholt RH, et al: A new technique for pulmonary segmental resection. Its application in the treatment of bronchiectasis. J Thorac Surg 15:384, 1946.

Padilla J, et al: Surgical results and prognostic factors in early non-small cell lung cancer. Ann Thorac Surg 63:324, 1997.

Pagni S, Federico JA, Ponn RB: Pulmonary resection for lung cancer in octogenarians. Ann Thorac Surg 63:785, 1997.

Pagni S, et al: Pulmonary resection for malignancy in the elderly: is age still a risk factor? Eur J Cardiothorac Surg 14:40, 1998.

Pairolero P: Extended resections for lung cancer: How far is too far? Eur J Cardiothorac Surg 16:S48, 1999.

Pairolero P, et al: Postsurgical stage I bronchogenic carcinoma: morbid implications of recurrent disease. Ann Thorac Surg 38:332,1984.

Passlick B, et al: Significance of lymphangiosis carcinomatosa at the bronchial resection margin in patients with non-small cell lung cancer. Ann Thorac Surg 72:1160, 2001.

Passlick B, et al: Mediastinal lymphadenectomy in non-small cell lung cancer: effectiveness in patients with or without nodal micrometastases results of a preliminary study. Eur J Cardiothorac Surg 21:520, 2002.

Pastorino U, et al: Perioperative blood transfusion and prognosis if resected stage IA lung cancer. Eur J Clin Oncol 22:1375, 1986.

Patchell RA, et al: Postoperative radiotherapy in the treatment of single metastases to the brain. JAMA 280:1485, 1998.

Patterson GA, et al: Significance of metastatic disease in subaortic lymph nodes. Ann Thorac Surg 43:155, 1987.

Patz EF, et al: Correlation of tumor size and survival in patients with stage IA non-small lung cancer. Chest 117:1568, 2000.

Paulson DL: The superior sulcus lesion. In Delarue N, Eschapasse H (eds): International Trends in General Thoracic Surgery. Philadelphia: WB Saunders, 1985, pp. 121 131.

Pearson FG, et al: Significance of positive mediastinal nodes identified at mediastinoscopy in patients with resectable cancer of the lung. J Thorac Cardiovasc Surg 83:1, 1982.

Pemberton JH, et al: Bronchogenic carcinoma in patients younger than 40 years. Ann Thorac Surg 36:509, 1983.

Piehler JM, et al: Bronchogenic carcinoma with chest wall invasion: factors affecting survival following en bloc resection. Ann Thorac Surg 34: 684, 1982.

Piehler JM, et al: Concomitant cardiac and pulmonary operations. J Thorac Cardiovasc Surg 90:662, 1985.

Pitz CCM, et al: Results of resection of T3 non-small cell lung cancer invading the mediastinum or main bronchus. Ann Thorac Surg 62:1016, 1996.

Pommier RF, et al: Synchronous non-small cell lung cancers. Am J Surg 171:521, 1996.

Ponn RB: Lightning can strike twice: second primary lung cancers [Editorial]. Chest 118:1526, 2000.

Ponn RB: Invited commentary on Saito et al: Diagnosis of visceral pleural invasion by lung cancer using intraoperative touch cytology. Ann Thorac Surg 73:1556, 2002.

Porhanov VA, et al: Indications and results of sleeve carinal resection. Eur J Cardiothorac Surg 22:685, 2002.

Porte H, et al: Resection of adrenal metastases from non-small cell lung cancer: a multicenter study. Ann Thorac Surg 71:981, 2001.

Porte HL et al: Adrenalectomy for a solitary adrenal metastasis from lung cancer. Ann Thorac Surg 65:331, 1998.

Price-Thomas C: Lobectomy with sleeve resection. Thorax 15:9, 1960.

Rao TH, Jacobs KH, Err EL: Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 54:449, 1983.

Ratto GB, et al: Chest wall involvement by lung cancer: computed tomographic detection and results of operation. Ann Thorac Surg 51:182, 1991.

Read RC, et al: Diameter, cell type, and survival in stage I primary non-small cell lung cancer. Arch Surg 123:446, 1988.

Read RC, et al: Management of non-small cell carcinoma with solitary brain metastasis. J Thorac Cardiovasc Surg 98:884, 1989.

Read RC, Yoder G, Schaeffer RC: Survival after conservative resection for T1N0M0 non-small cell lung cancer. Ann Thorac Surg 49:391, 1990.

Regnard JF, et al: Results of resection for bronchogenic carcinoma with mediastinal lymph node metastases in selected patients. Eur J Cardiothorac Surg 5:583, 1991.

Rena O, et al: Stage I non-small cell lung carcinoma: really an early stage? Eur J Cardiothorac Surg 21:514, 2002.

Rendina EA, et al: Sleeve resection and prosthetic reconstruction of the pulmonary artery for lung cancer. Ann Thorac Surg 68:995, 1999.

Ribet M, Dambron P: Multiple primary lung cancers. Eur J Cardiothorac Surg 9:231, 1995.

Ricci C, et al: Reconstruction of the pulmonary artery in patients with lung cancer. Ann Thorac Surg 57:627, 1994.

Riquet M, et al: Operation for lung cancer in the elderly. What about octogenarians? [Letter to the Editor]. Ann Thorac Surg 58:916, 1994.

Riquet M, et al: Prognostic significance of surgical-pathologic N1 disease in non-small cell carcinoma of the lung. Ann Thorac Surg 67:1572, 1999.

Riquet M, et al: Resection of lung cancer invading the diaphragm. J Thorac Cardiovasc Surg 120:417, 2000.

Riquet M, et al: Do patients with adenosquamous carcinoma of the lung need a more aggressive approach? J Thorac Cardiovasc Surg 122:618, 2001.

Riquet M, et al: Characteristics and prognosis of resected T3 non-small cell lung cancer. Ann Thorac Surg 73:253, 2002.

Rocco G, et al: Prognostic factors after surgical treatment of lung cancer invading the diaphragm. Ann Thorac Surg 68:2065, 1999.

Roeslin N, et al: A better prognostic value from a modification of lung cancer staging. J Thorac Cardiovasc Surg 94:504, 1987.

Rosengart TK, et al: Multiple primary lung carcinomas: prognosis and treatment. Ann Thorac Surg 52:773, 1991.

Roviaro G, et al: Tracheal sleeve pneumonectomy for bronchogenic carcinoma. J Thorac Cardiovasc Surg 107:13, 1994.

Roviaro G, et al: Videothoracoscopic staging and treatment of lung cancer. Ann Thorac Surg 59:971, 1995.

Ruffini E, et al: The significance of intraoperative pleural effusion during surgery for bronchogenic carcinoma. Eur J Cardiothorac Surg 21:508, 2002a.

Ruffini E, et al: Lung tumors with mixed histologic pattern. Clinico-pathologic characteristics and prognostic significance. Eur J Cardiothorac Surg 22:701, 2002b.

Rusch VW: High-resolution computed tomography in clinical T1N0M0 adenocarcinoma of the lung [Editorial]. J Thorac Cardiovasc Surg 124: 221, 2002.

Rusch VW, et al: Surgical resection of stage IIIA and stage IIIB non-small cell lung cancer after concurrent induction chemoradiotherapy. J Thorac Cardiovasc Surg 105:97, 1993.

Rusch VW, et al: Neoadjuvant therapy: a novel and effective treatment for stage IIIB non-small cell lung cancer. Ann Thorac Surg 58:290, 1994.

Rusch VW, et al: Factors determining outcome after surgical resection of T3 and T4 lung cancers of the superior sulcus. J Thorac Cardiovasc Surg 119:1147, 2000.

Rusch VW, et al: Induction chemoradiation and surgical resection for non-small cell lung carcinomas of the superior sulcus: initial results of Southwest Oncology Group Trial 9416 (Intergroup Trial 0160). J Thorac Cardiovasc Surg 121:472, 2001.

Sagawa M, et al: Clinical and prognostic assessment of patients with resected small peripheral lung cancer lesions. Cancer 66:2653, 1990.

P.1586


Sagawa M, et al: Segmentectomy for roentgenographically occult bronchogenic squamous cell carcinoma. Ann Thorac Surg 71:1100, 2001.

Saito Y, et al: Results of surgical treatment for roentgenographically occult bronchogenic squamous cell carcinoma. J Thorac Cardiovasc Surg 104: 401, 1992.

Saito Y, et al: Diagnosis of visceral pleural invasion by lung cancer using intraoperative touch cytology. Ann Thorac Surg 73:1552, 2002.

Sartori F, et al: Carcinoma of the superior pulmonary sulcus: results of irradiation and radical resection. J Thorac Cardiovasc Surg 104:679, 1992.

Sawabata N, et al: Malignant minor pleural effusion detected on thoracotomy for patients with non-small cell lung cancer: is tumor resection beneficial for prognosis? Ann Thorac Surg 73:412, 2002a.

Sawabata N, et al: Serum carcinoembryonic antigen level in surgically resected clinical stage I patients with non-small cell lung cancer. Ann Thorac Surg 74:174, 2002b.

Sawyer TE, et al: Factors predicting patterns of recurrence after resection of N1 non-small cell lung carcinoma. Ann Thorac Surg 68:1171, 1999.

Shaw RR, Paulson DI, Kee JL: Treatment of the superior sulcus tumor by irradiation followed by resection. Ann Surg 154:29, 1961.

Shields TW: The fate of patients after incomplete resection of bronchial carcinoma. Surg Gynecol Obstet 139:569, 1974.

Shields TW: The significance of ipsilateral mediastinal lymph node metastasis in non-small cell lung cancer. J Thorac Cardiovasc Surg 99:48, 1990.

Shields TW, Higgins GA: Minimal pulmonary resection. Arch Surg 108: 420, 1974.

Shields TW, et al: Relationship of cell type and lymph node metastases to survival after resection of bronchial carcinoma. Ann Thorac Surg 20: 501 510, 1975.

Shimizu J, et al: Results of surgical treatment of stage 1 lung cancer. Nippon Geka Gakkai Zasshi 94:505, 1993.

Shirakusa T, et al: Results of resection for bronchogenic carcinoma in patients over the age of 80. Thorax 44:189, 1989.

Shrager JB, et al: Lobectomy with tangential pulmonary artery resection without regard to pulmonary function. Ann Thorac Surg 70:234, 2000.

Snijder RJ, et al: Survival in resected stage I lung cancer with residual tumor at the bronchial resection margin. Ann Thorac Surg 65:212, 1998.

Soorae AS, Stevenson JM: Survival with residual tumor on the bronchial margin after resection for bronchogenic carcinoma. J Thorac Cardiovasc Surg 78:175, 1979.

Spaggiari L, et al: Superior vena cava resection with prosthetic replacement for non-small cell lung cancer: long-term results of a multicenter study. Eur J Cardiothorac Surg 21:1080, 2002.

Stanford W, Barnes RP, Tucker AR: Influence of staging in superior sulcus (Pancoast) tumors of the lung. Ann Thorac Surg 29:406, 1980.

Stenbygaard LE, Sorensen JB, Olsen JE: Metastatic pattern of adenocarcinoma of the lung: an autopsy study from a cohort of 137 consecutive patients with complete resection. J Thorac Cardiovasc Surg 11:1130, 1995.

Strauss GM, et al: Neoadjuvant chemotherapy and radiotherapy followed by surgery in stage IIIA non-small cell carcinoma of the lung: report of a Cancer and Leukemia Group B phase II study. J Clin Oncol 10:1237, 1992.

Sugi K, et al: Systematic lymph node dissection for clinically diagnosed peripheral non-small cell lung cancer less than 2 cm in diameter. World J Surg 22:290, 1998.

Suzuki K, et al: Predictors of lymph node and intrapulmonary metastasis in clinical stage IA non-small cell lung carcinoma. Ann Thorac Surg 72:352, 2001.

Suzuki K, et al: Early peripheral lung cancer: prognostic significance of ground glass opacity on thin-section computed tomographic scan. Ann Thorac Surg 74:1635, 2002.

Takei H, et al: Large cell neuroendocrine carcinoma of the lung: a clinicopathologic study of 87 cases. J Thorac Cardiovasc Surg 124:285, 2002.

Tanaka F, et al: Surgery for non-small cell lung cancer: postoperative survival based on the revised tumor-node-metastasis classification and its time trend. Eur J Cardiothorac Surg 18:147, 2000.

Tanaka F, et al: Prognostic factors in patients with resected pathologic T1 2N1M0 non-small cell lung cancer. Eur J Cardiothorac Surg 19:555, 2001.

Tateishi M, et al: Skip mediastinal lymph node metastasis in non-small cell lung cancer. J Surg Oncol 57:139, 1994.

Taulelle M, et al: High dose rate endobronchial brachytherapy: results and complications in 189 patients. Eur Respir J 11:162, 1998.

Tedder M, et al: Current morbidity, mortality and survival after bronchoplastic procedures for malignancy. Ann Thorac Surg 54:387, 1992.

Terzi A, et al: Lung resections concomitant to coronary artery bypass grafting. Eur J Cardiothorac Surg 8:580, 1994.

Thomas P, Rubinstein L, and the Lung Cancer Study Group: Cancer recurrence after resection: T1N0 non-small cell lung cancer. Ann Thorac Surg 49:242, 1990.

Thomas P, et al: Stage I non-small cell lung cancer: a pragmatic approach to prognosis after complete resection. Ann Thorac Surg 73:1065, 2002.

Trastek VF, et al: En bloc (non-chest wall) resection for bronchogenic carcinoma with parietal fixation. J Thorac Cardiovasc Surg 87:352, 1984.

Trotter PI, Burrow L, Kirschner P: Perioperative blood transfusion adversely affects prognosis after resection of stage I non-oat cell lung cancer. J Thorac Cardiovasc Surg 88:659, 1984.

Tsuchiya R, et al: Extended resection of the left atrium, great vessels, or both for lung cancer. Ann Thorac Surg 57:960, 1994.

Urschel HC: Superior pulmonary sulcus carcinoma. Surg Clin North Am 68:497, 1988.

Urschel JD, Finley RK, Takita H: Long-term survival after bilateral adrenalectomy for metastatic lung cancer. Chest 112:849, 1997.

Urschel JD, et al: Prognostic implications of pulmonary satellite nodules: are the 1997 staging revisions appropriate? Lung Cancer 21:83, 1998.

Vallieres E, et al: Pancoast tumors. Curr Probl Surg 38:293, 2001.

van Klaveren RJ, et al: Prognosis of unsuspected but completely resectable N2 non-small cell lung cancer. Ann Thorac Surg 56:300, 1993.

van Rens MTM, et al: Prognostic assessment of 2,361 patients who underwent pulmonary resection for non-small cell lung cancer stage I, II, and IIIA. Chest 117:374, 2000.

Vansteenkiste JF, et al: Survival and prognostic factors in resected N2 non-small cell lung cancer: a study of 140 cases. Ann Thorac Surg 63:1441, 1997.

Vansteenkiste JF, et al: Prognostic importance of the standardized uptake value on (18) F-fluoro-2-deoxy-glucose positron emission tomography scan in non-small cell lung cancer: an analysis of 125 cases. J Clin Oncol 17:3201, 1999.

van Velzen E, et al: Type of lymph node involvement influences survival rates in T1N1M0 non-small cell lung carcinoma: lymph node involvement by direct extension compared with lobar and hilar metastases. Chest 110:1469, 1996.

van Velzen E, et al: Lymph node type as a prognostic factor for survival in T2N1M0 non small cell lung carcinoma: Ann Thorac Surg 63:1436, 1997.

Vogt-Moykopf I, et al: Bronchoplastic and angioplastic operation in bronchial carcinoma: long-term results of a retrospective analysis from 1973 to 1983. Int Surg 71:211, 1986.

Wada H, et al: Thirty-day operative mortality for thoracotomy in lung cancer. J Thorac Cardiovasc Surg 115:70, 1998.

Warren WH, Faber LP: Segmentectomy versus lobectomy in patients with stage I pulmonary carcinoma. J Thorac Cardiovasc Surg 107:1087, 1994.

Watanabe Y, et al: Results in 104 patients undergoing bronchoplastic procedures for bronchial lesions. Ann Thorac Surg 50:607, 1990.

Watanabe Y, et al: Results of surgical treatment in patients with stage IIIA non-small cell lung cancer. Thorac Cardiovasc Surg 39:44, 1991a.

Watanabe Y, et al: Results of surgical treatment of patients with stage IIIB non-small cell lung cancer. Thorac Cardiovasc Surg 39:50, 1991b.

Watanabe Y, et al: Early hilar lung cancer: its clinical aspect. J Surg Oncol 48:75, 1991c.

Watanabe Y, et al: Mediastinal nodal involvement and the prognosis of non-small cell lung cancer. Chest 100:422, 1991d.

Watanabe S, et al: Results of wedge resection for focal bronchioloalveolar carcinoma showing pure ground glass attenuation on computed tomography. Ann Thorac Surg 73:1071, 2002.

Weiden PL, et al: Preoperative chemotherapy (cisplatinum and fluorouracil) and radiation therapy in stage III non-small cell lung cancer: a phase II study of the LCSG 773. Chest 106[Suppl]:297s, 1994.

Weigel TL, Martini N: Occult lung cancer treatment. Chest Surg Clin North Am 10:751, 2000.

Weigel TL, et al: Photodynamic therapy: a curative surgical approach for early non-small cell lung carcinomas in high-risk patients. Proc Int Soc Optical Engineering 3909:235, 2000.

Weksler B, et al: Resection of lung cancer invading the diaphragm. Ann Thorac Surg 114:500, 1997.

Westermann CJJ, et al: Pulmonary resection after pneumonectomy in patients with bronchogenic carcinoma. J Thorac Cardiovasc Surg 106:868, 1993.

Williams DE, et al: Survival of patients surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg 82:70, 1981.

P.1587


Wright CD, et al: Superior sulcus lung tumors: results of combined treatment (irradiation and radical resection). J Thorac Cardiovasc Surg 94:69, 1987.

Yano T, et al: Surgical results and prognostic factors of pathologic N1 disease in non-small cell carcinoma of the lung. J Thorac Cardiovasc Surg 107:1398, 1994.

Yokoi K, et al: Results of surgical treatment of lung cancer involving the diaphragm. J Thorac Cardiovasc Surg 120:799, 2000.

Yokoi K, Matsuguma H, Anraku M: Extrapleural pneumonectomy for lung cancer with carcinomatous pleuritis. J Thorac Cardiovasc Surg 123:184, 2002.

Yoshino I, et al: Postoperative prognosis in patients with non-small cell lung cancer with synchronous ipsilateral intrapulmonary metastasis. Ann Thorac Surg 64:809, 1997.

Zacharias J, et al: Large cell neuroendocrine carcinoma and large cell carcinoma with neuroendocrine morphology of the lung: prognosis after complete resection and systematic nodal dissection. Ann Thorac Surg 75:348, 2003.

Zorn GL Jr, et al: Pulmonary transplantation in advanced bronchioloalveolar carcinoma. J Thorac Cardiovasc Surg 125:45, 2003.



General Thoracic Surgery. Two Volume Set. 6th Edition
General Thoracic Surgery (General Thoracic Surgery (Shields)) [2 VOLUME SET]
ISBN: 0781779820
EAN: 2147483647
Year: 2004
Pages: 203

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net