Outcomes of Sublobar Resection Versus Lobectomy for Stage I Non Small Cell Lung Cancer: A 13-Year Analysis

Transcription

1 GENERAL THORACIC SURGERY: The Annals of Thoracic Surgery CME Program is located online at To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal. Outcomes of Sublobar Resection Versus Lobectomy for Stage I Non Small Cell Lung Cancer: A 13-Year Analysis Amgad El-Sherif, MD, William E. Gooding, MS, Ricardo Santos, MD, Brian Pettiford, MD, Peter F. Ferson, MD, Hiran C. Fernando, MD, Susan J. Urda, BS, James D. Luketich, MD, and Rodney J. Landreneau, MD Department of Surgery, Heart, Lung, and Esophageal Surgery Institute, University of Pittsburgh Medical Center, Biostatistics Facility, and Cancer Registry, University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania, and Department of Cardiothoracic Surgery, Boston Medical Center, Boston, Massachusetts Background. The appropriate use of sublobar resection versus lobectomy for stage I non small cell lung cancer continues to be debated. A long-term analysis of the outcomes of these resections for stage I non small cell lung cancer in a high-volume tertiary erral university hospital center was performed. Methods. The outcomes of all stage I non small cell lung cancer patients (n 784) undergoing resection were analyzed from our lung cancer registry from 1990 to Lobectomy was the standard of care for patients with adequate cardiopulmonary reserve. Sublobar resection was reserved for patients with cardiopulmonary impairment prohibiting lobectomy. Predictors of overall survival and disease-free survival were evaluated. Statistical analyses included Kaplan Meier estimates of survival, log-rank tests of survival differences, and multivariate Cox proportional hazards models. Results. Lobectomy was used for 577 patients and sublobar resection for 207 patients. The median age was 70 years (range, 31 to 107 years). The median follow-up of patients remaining alive was 31 months. Compared with lobectomy, sublobar resection had no significant impact on disease-free survival, with a hazard ratio of 1.20 (95% confidence interval, 0.90 to 1.61; p 0.24). Sublobar resection had a statistically significant association with overall survival when compared with lobectomy, with an increased hazard ratio of 1.39 (95% confidence interval, 1.11 to 1.75; p 0.004). Twenty-eight percent of lobectomy patients experienced disease recurrence in that time compared with 29% of the sublobar patients. Seventytwo percent of the recurrences in the lobectomy cohort were distant metastasis versus 52% of the sublobar group recurrences (p ). Conclusions. Although sublobar resection is thought to be associated with increased incidence of local recurrence when compared with lobectomy, we found no difference in disease-free survival between the two types of resection for stage IA patients but slightly worse disease-free survival for stage IB. (Ann Thorac Surg 2006;82:408 16) 2006 by The Society of Thoracic Surgeons Surgical resection remains the mainstay of therapy for stage I non small cell lung cancer (NSCLC), providing the best opportunity for cure. The extent of pulmonary resection required to achieve complete eradication of the malignancy has been a hotly debated issue [1]. Many patients presenting with resectable early stage disease are unable to tolerate pulmonary resection because of compromised cardiopulmonary function. It has Accepted for publication Feb 4, Presented at the Forty-first Annual Meeting of The Society of Thoracic Surgeons, Tampa, FL, Jan 24 26, Address correspondence to Dr Landreneau, Division of Thoracic and Foregut Surgery, Department of Surgery, Shadyside Medical Bldg, 5200 Centre Ave, Suite 715, Pittsburgh, PA 15232; landreneaurj@upmc.edu. been suggested that more than 20% of patients who are diagnosed with stage I or II NSCLC do not undergo operation because of comorbid health factors [2]. The use of sublobar resection with adequate surgical margins in this setting may provide a comparable survival advantage to such patients along with preservation of pulmonary function [3]. Additionally, computed tomographic scan surveillance efforts directed to patients with a higher risk for lung cancer have identified small peripheral cancers amena- Dr Luketich discloses that he has a financial relationship with US Surgical, Stryker and Berchtold by The Society of Thoracic Surgeons /06/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg EL-SHERIF ET AL 2006;82: SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC ble to sublobar resection with generous margins of parenchymal resection. Many surgeons have questioned the necessity of total lobectomy for the management of such lesions [4 6]. The purpose of this investigation was to evaluate the patterns of recurrence, postoperative morbidity differences, and survival differences between patients with stage I NSCLC treated by either sublobar resection or anatomic lobectomy. This period (from 1990 through 2003) of surgical experience with stage I NSCLC was chosen for analysis as computed tomographic clinical staging and follow-up were routinely applied. Additionally, this was also a period of increased surgical experience with patients having significant impairment in cardiopulmonary reserve as our group was actively involved in the initial efforts with minimally invasive thoracic surgery, lung-volume reduction surgery, and the National Emphysema Treatment Trial [7]. These experiences influenced the nature of our lung cancer erral; accordingly, many patients found to have peripheral small lung cancers with significant cardiopulmonary impairment were considered for surgical resection. Many of these cancer patients were thought to be at risk for unacceptable pulmonary functional loss associated with lobectomy, and thus sublobar resection was chosen as the primary resectional therapy. The evaluation of this mature experience with surgical resection of stage I NSCLC by either lobectomy or sublobar resection follows. Material and Methods The study was approved by the Institutional Review Board at the University of Pittsburgh (IRB ), and patient consent was waived. We retrospectively analyzed the outcomes of all stage I NSCLC patients undergoing resection between January 1990 and December Using the data from the Lung Cancer Registry at the University of Pittsburgh, 817 patients were identified, of whom 784 had adequate recurrence and follow-up information. Lobectomy and mediastinal sampling was the Fig 1. Small peripheral lesion where substantial surgical margin can be obtained with sublobar resection. Fig 2. Larger, more central lesion where compromised margin may occur with sublobar resection. 409 standard of care for patients with adequate cardiopulmonary reserve. Sublobar resection was used depending on assessment of patient cardiopulmonary impairment prohibiting lobectomy (Figs 1, 2). All patients were staged postsurgically according to the TNM classification of the American Joint Committee for Cancer Staging and Revised International System for Staging Lung Cancer [8]. All patients were staged as stage I on the basis of clinical, radiologic, and pathologic staging. Mediastinoscopy was not used in preoperative evaluation of this group of patients. No patients required neoadjuvant or adjuvant chemotherapy or radiation therapy. The demographic data of patients age and sex were collected as well as tumor histology, size, number of lymph nodes sampled, length of stay, and operative and perioperative mortality data. Perioperative mortality included patients who died within the first 30 days after surgery. End points of the analysis included the incidence of disease recurrence (locoregional and distant), diseasefree survival, and overall survival. Locoregional recurrence was defined as any recurrence within the same lobe of the lung, or interlobar and hilar lymph nodes (N1 nodes). All other metastases were classified as distant recurrence. The two cohorts, lobectomy and sublobar resection, were compared on the basis of clinical, demographic, and pathologic data. Both the Student s t test and the Wilcoxon test [9] were used to compare the distributions of continuous data (age, tumor size), and the 2 test or Fisher s exact test [10] was used to compare the frequencies of categorical measures (sex, histology, stage, vascular pleural invasion, and number of lymph nodes removed). In cases with low cell counts, the exact 2 test was used in place of the usual (asymptotic) test. All comparisons were two-tailed. Disease-free survival was defined as the time from surgery to the first diagnosis of local, regional, or distant GENERAL THORACIC

3 410 EL-SHERIF ET AL Ann Thorac Surg SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC 2006;82: Table 1. Patient Characteristics by Type of Procedure Characteristic Lobar Resection n 577 (73.6%) Sublobar Resection n 207 (26.4%) p Value Age (y) Mean b Range Sex Male 298 (51.6%) a 104 (50.2%) c Female 279 (48.4%) 103 (49.8%) Histology Adenocarcinoma 192 (33.3%) 75 (36.2%) c Squamous cell 194 (33.6%) 52 (25.1%) Bronchoalveolar 69 (12.0%) 26 (12.6%) Adenosquamous 23 (4.0%) 7 (0.5%) Large cell 22 (3.8%) 7 (0.5%) Other 77 (13.3%) 40 (19.3%) Stage IA 288 (49.9%) 161 (77.8%) c IB 289 (50.1%) 46 (27.2%) VPI Positive among IB 108 (37.3%) 35 (76.1%) c Negative among IB 181 (62.7%) 11 (23.9%) Tumor size (median) 2.8 cm 1.8 cm b a Column percentages. b Student s t test. c 2 test. VPI visceral pleural involvement. disease recurrence, or until last follow-up. Overall survival was defined as the time from surgery to death or last follow-up. Disease-free and overall survivals were estimated with the Kaplan Meier method [11] with the Greenwood formula for estimation of standard errors. Individual predictors such as sex, histology, stage tumor size, and type of surgery (lobe versus sublobe) were compared with the log-rank test and Cox regression. To fully assess the effect of sublobar resection, Cox proportional hazards models [12] were constructed to evaluate the independent effect of sublobar resection while adjusting for other prognostic variables, including age, stage, tumor size, and number of resected lymph nodes. Results Clinical and Pathologic Description of the Cohorts A total of 577 patients received lobectomy and 207 patients were treated with sublobar resection. Of these, there were 122 wedge resections and 85 segmentectomies. Mean ages were slightly different between the two groups, and the sex ratio was equal (Table 1). Lung cancer histologic diagnosis was also well balanced between the cohorts (Table 1, Fig 3). Patients receiving sublobar resection tended to have smaller tumors (1.8 cm versus 2.8 cm; p ) and were more likely to be stage IA rather than IB (78% stage 1A versus 27% stage IB; p ), but among the 46 stage IB patients having sublobar resection, 35 of them (76%) had visceral pleural invasion (p ). Surgical Outcomes Patients receiving sublobar resection were much more likely to have limited lymph node sampling (Table 2). Forty-three percent (83 of 191 known lymph node samplings) of the sublobar patients did not have a single node sampled compared with only 2.7% of lobectomy patients. More than half of the lobectomy patients had extensive (8 or more) nodes removed, whereas only 38 of 191 (20%) of sublobar patients had the same. The difference in lymph node sampling between the two cohorts was strongly significant ( 2, p ). There was no significant difference between the two groups in length of stay or perioperative mortality. The median length of stay was 6 days in each group (p Fig 3. Distribution of tumor histologic diagnosis.

4 Ann Thorac Surg EL-SHERIF ET AL 2006;82: SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC Table 2. Number of Nodes Sampled by Type of Surgery a Type of Surgery Number of Nodes Sampled or more Total Lobectomy Sublobar resection Total a 2 p , Wilcoxon test). The perioperative mortality for the lobectomy group was 2.6% versus 1.4% for the sublobar resection group (p 0.42, Fisher s exact test). Disease-Free Survival All 784 patients were followed until disease recurrence or death or until loss of follow-up in the tumor registry (Table 3). The median follow-up for patients who were alive and were disease-free at their last review was 2.7 years, with a range of less than 1 month to 14 years. Twenty-eight percent of lobectomy patients experienced disease recurrence in that time compared with 29% of the sublobar patients. Although the proportion of patients found with disease recurrence was the same in both groups, there is a suggestion that the recurrence patterns differ, as 72% of the recurrences in the lobectomy cohort were distant metastasis versus 52% of the sublobar group recurrences. This difference was significant (exact 2 test, p ). Sublobar resection was evaluated for association with disease-free survival along with age, sex, histology, stage, tumor size, visceral pleural invasion, and number of sample lymph nodes. Of these, tumor size as a continuous variable and stage were the only significant predictors (Table 4). Because visceral pleural invasion was not a predictive factor, disease stage was substituted for tumor size. Table 4 shows individual proportional hazards regression coefficients and confidence intervals for these individual covariates. Stage IB conferred a 50% increase in the risk of disease recurrence (hazard ratio, 1.51; p ). The estimated hazard ratio for sublobar resection compared with lobectomy was 1.20, with a 95% confidence interval of 0.90 to The hazard ratios were also estimated with joint proportionate hazards models that included stage and surgery type. The hazard ratio for sublobar resection increased to 1.64, with a 95% confidence interval of 1.25 to On further analysis this increase in risk as a result of Table 3. Disease Recurrences Recurrence Lobar Resection Sublobar Resection Disease-free 415 (71.9%) a 147 (71.0%) Local 24 (4.2%) 15 (7.2%) Regional 22 (3.8%) 14 (6.8%) Distant 116 (20.0%) 31 (15.0%) Total a Column percentages. Table 4. Individual Predictors of Disease-Free Survival Covariate Hazard Ratio 95% CI p Value Age Stage IA Stage IB Lobectomy Sublobar resection VPI no a VPI yes a No. of nodes a VPI analysis limited to 335 stage IB patients. CI confidence interval; VPI vascular plueral invasion. 411 sublobar resection was caused solely by the greater than expected number of recurrences among the 46 stage IB patients with sublobar resection. To illustrate the effects of sublobar resection by stage, Kaplan Meier plots of disease-free survival are shown in Figure 4 for stages IA and IB. As shown in Figure 4A, stage IA patients had identical disease-free survival in both cohorts for 7 years (log-rank test, p 0.308). Stage IB patients, although more likely to have lobectomy, did exhibit slightly worse disease-free survival after sublobar resection, with an decrease in 5-year disease-free survival from 58% (lobectomy) to 50% (sublobar, log-rank test, p ). Thus, the risk of recurrence for sublobar resection in our cohort was equivalent to that of lobectomy among stage IA patients but slightly worse among the small number of stage IB patients with sublobar resection. Overall Survival With a median follow-up for surviving patients of 31 months, the cohort had a 5-year probability of survival of 51%. Sublobar resection was a strong predictor of overall survival with a decreased 5-year survival of 40% versus 54% for the lobectomy group (log-rank test, p ; Fig 5A). Age and number of lymph nodes also predicted overall survival. Hazard ratios for individual predictors of overall survival are shown in Table 5. Owing to the imbalance in lymph node removal between cohorts (fewer nodes with sublobar resection), we estimated the hazard ratios for resection type adjusting for age and number of nodes, which were strongly associated with overall survival, and stage, which was only weakly associated. As shown on Figure 5B and 5C, when simultaneously adjusting for age, stage, and number of lymph nodes, overall survival for lobectomy and sublobar resection was equivalent. Limited or no lymph node sampling appears to be a much stronger predictor than sublobar resection, such that when adjusting for the number of resampled lymph nodes, patients with sublobar resection had equivalent overall survival as lobectomy patients. In summary, patients selected for sublobar resection tended to be slightly older and to have limited node sampling, which is a proxy variable for undocumented comorbidities that adversely affect overall health. When adjusting for the limited node sampling and for age, sublobar GENERAL THORACIC

5 412 EL-SHERIF ET AL Ann Thorac Surg SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC 2006;82: Fig 4. Disease-free survival by stage with 95% confidence intervals at 2-year intervals. Patients with sublobar resection (dashed lines) had similar disease-free survival compared with patients receiving lobectomy (solid lines) if they were stage IA (A; log-rank test, p 0.308), but worse disease-free survival if they were stage IB (B; log-rank test, p ). resection did not confer worse overall survival than lobectomy. Comment These data demonstrate improved overall survival for patients with stage I NSCLC undergoing lobectomy compared with sublobar resection. One can look at these results and conclude that lobectomy must theore be the superior management option compared with sublobar resection for stage IA NSCLC. Important confounding issues questioning this conclusion identified in this work relate to the similar disease-free survival despite differences in local and regional recurrence (Table 3, Fig 4A). As the disease-free survival was similar and not affected by the type of resection, we may assume that the overall mortality differences are affected by other patientrelated variables. The fact that sublobar resection was Fig 5. Marginal and conditional estimates of overall survival by type of resection with 95% confidence intervals at 2-year intervals. When considered alone, sublobar resection conveys a worse prognosis for overall survival (A). The marginal effect of sublobar resection lowers 5-year survival from 54% in the lobectomy group to 40% in the sublobar group (log-rank test, p ). However, when adjusted for other significant predictors, sublobar resection has no effect on overall survival. This is equally true for patients with limited (B) or extensive (C) node sampling.

6 Ann Thorac Surg EL-SHERIF ET AL 2006;82: SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC Table 5. Individual Predictors of Overall Survival Covariate Hazard Ratio 95% CI p Value Age Stage IA Stage IB Lobectomy Sublobar resection VPI no a VPI yes a lymph nodes 0 1 lymph nodes lymph nodes a VPI analysis limited to 335 stage IB patients. CI confidence interval; VPI vascular plueral invasion. exclusively used by the operating surgeon to manage stage I NSCLC patients thought to be poor candidates for lobectomy because of cardiopulmonary physiologic concerns can explain a possible greater risk of death from comorbid diseases during follow-up. Although we could not study this difference in comorbid conditions between the two groups given the retrospective analysis and the limitations of the lung registry, we can only speculate based on our standard approach of lobectomy for stage I disease in patients with good risk. This effect of comorbid disease in evaluating survival from coronary artery disease has been recently analyzed and emphasized by a group from Duke [13]. In this analysis of coronary artery disease survival, the survival influence of comorbidities such as chronic obstructive pulmonary disease or peripheral vascular disease in 1,471 patients was equivalent to poor left ventricular function in affecting patient survival. The effect of comorbidity on older lung cancer patients survival after lung resection has also been documented in a report by Birim and coworkers [14]. The authors assessed the risk factors of survival in 126 consecutive patients with NSCLC who were older than 70 years of age and underwent resection. During a period of 12 years, the patient survival was compared with the yearly expected survival rates of the general population. The study showed a 5- and 10-year survival rate of 37% and 15%, respectively. Smoking (odds ratio, 2.3) and chronic obstructive pulmonary disease (odds ratio, 2.1) were identified as risk factors for long-term survival. The observed survival was lower than the expected survival related to the natural history of their comorbid disease and the patient age. The recent subanalysis of the medical treatment arm of the National Emphysema Treatment Trial further emphasized the importance of the natural history of moderate to severe chronic obstructive pulmonary disease [7]. The mortality among patients with predominant upper lobe emphysema and poor exercise tolerance was 34% at 2 years. Even among patients with the least chronic obstructive pulmonary disease limitations (lower lobe predominance and good exercise tolerance), the mortality was 13% at 2 years. 413 The possible effect of erroneous understaging of a significant number of the patients undergoing sublobar resection related to the thoracic lymph node sampling differences observed in this study could have also led to disparate survival results between lobectomy and sublobar resection patients. The Lung Cancer Study Group conducted the only randomized study comparing sublobar resection with lobectomy for stage IA NSCLC patients [15]. Importantly, all patients considered for the study had to be considered functional candidates for lobectomy if necessary. In this study, the intraoperative nodal staging requirements were limited to sampling a single mediastinal, hilar and interlobar node before randomization to the mode of surgical resection. Distant recurrence rate was unaffected by the method of resection used. There was no difference in hospital mortality between the two groups of patients; however, patients who underwent lobectomies did have significantly more postoperative pulmonary complications requiring mechanical ventilatory support. A significant threefold increase in local recurrence was seen with sublobar resection; however, there was no difference in overall survival or cancer-related survival between sublobar resection and lobectomy. Although all patients in this report were staged as stage I on the basis of pathologic staging, we think there may be an artificial tumor understaging of patients undergoing sublobar resection as a consequence of disproportionate dissection between lobectomy and sublobar resection techniques. Our analysis showed a 2% decrease in the risk of dying for every node that was resected (p ); this resulted in a bias in survival favoring lobectomy. Our results are consistent with the recent data from the ACOSOG (American College of Surgeons Oncology Group) Z0030 trial presented at the Annual Meeting of The Society of Thoracic Surgeons in Tampa, Florida, in January 2005, and published in 2006 [16]. This study included 1,111 randomized patients undergoing anatomic resection of stage I NSCLC. The report showed a 4% increase in number of positive lymph nodes when lymph node sampling was compared with lymph node dissection. Interestingly, an 18% stage shift from stage I to stage II and III was seen when one compares the preoperative clinical staging to the intraoperative or postoperative pathologic staging. Recent investigations from Japan comparing extended segmentectomy with thorough thoracic lymph node staging for small peripheral stage I lesions have reported equivalent survival to that of lobectomy [17]. These reports may be considered an expected consequence to the lung cancer computed tomographic screening programs begun in that country more than a decade ago. Many thoracic surgeons began to selectively use sublobar resection as the primary therapy for small peripheral NSCLC in lieu of lobectomy. An interesting report of sublobar management of small peripheral NSCLC came from Swanson and coworkers [18] and Mery and associates [19]. These investigators reported their experience among patients undergoing wedge resection as primary therapy for subcentimeter peripheral lung cancers. GENERAL THORACIC

7 414 EL-SHERIF ET AL Ann Thorac Surg SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC 2006;82: Table 6. Summary of Studies Comparing Sublobar With Lobar Resection for Stage I Non Small Cell Lung Cancer Study Sublobar Resection Lobar Resection N 5-year Survival (%) Local Recurrence (%) N 5-year Survival (%) Local Recurrence (%) Pro lobar resection LCSG [15] a a 6.4 Warren [22] a a 4.9 Miller [23] a 11 Martini [24] a a 24 Pro sublobar resection Errett [25] NA NA Pastorino [26] Read [27] Landreneau [28] a Okada [17] NA NA Kodama [29] Koike [30] a Statistically significant. LCSG Lung Cancer Study Group; NA not available. Kondo and colleagues [20] reviewed the outcome of 104 patients with peripheral stage IA adenocarcinomas of the lung. They specifically analyzed the survival among 57 of these patients with lesions less than 1 cm in diameter. These patients with subcentimeter lesions underwent the following procedures: 23 lobectomies, 13 segmentectomies, and 21 wedge resections. Interestingly, the authors found that the mode of resection used for these small peripheral lung cancers did not affect postoperative cancer-related survival. The 5-year survival for the group was 97%. These investigators also advocated the use of the Noguchi s histologic classification for these small peripheral tumors to aid in predicting the risk of nodal metastasis [21]. This system identifies small ground-glass opacities with bronchioalveolar histology as having a very minimal chance of nodal metastasis and also an extremely favorable prognosis with sublobar wedge resection alone. In contrast, lesions with more characteristic glandular adenocarcinoma histologic appearance had greater than 10% chance of nodal metastasis. Because of this occult lymph node metastasis risk, Kondo and associates [20] advocated anatomic resection, perably lobectomy, for such lesions. An interesting recent report by Okada and colleagues [6] looked at the effect of tumor size on prognosis after resection in 1,272 patients. There was no statistical difference in the 5-year cancer-specific survival between lobectomy and segmentectomy in patients with tumors 20 mm or less in diameter (92.4% versus 96.7%, respectively). In addition, tumors 20 to 30 mm in diameter had no significant difference between lobectomy and segmentectomy in the 5-year cancer-specific survival (87.4% versus 84.6%, respectively). The authors suggested segmentectomy was an acceptable operation for stage I disease 2 cm or less in diameter. A summary of the studies comparing sublobar with lobar resection for stage I NSCLC is shown in Table 6. In conclusion, there are compelling results of several recent and past studies suggesting that sublobar resection may have primary therapeutic utility for selected patients with stage IA NSCLC. This may be particularly the case for tumors located in the periphery of the lung, within anatomic segmental boundaries, with no endobronchial component, and less than 2 cm in diameter. Retrospective analyses of the results of lobectomy compared with sublobar resection will inherently be affected by confounding patient-related issues and selection biases of the thoracic surgeons performing the surgical management. These data emphasize the need for future prospective randomized trials comparing segmentectomy with lobectomy for selected small peripheral stage IA NSCLC. References 1. Patel AN, Santos RS, De Hoyos A, Luketich JD, Landreneau RJ. Clinical trials of peripheral stage I (T1N0M0) non-small cell lung cancer. Semin Thorac Cardiovasc Surg 2003;15: Bach PB, Cramer LD, Warren JL, Begg CB. Racial differences in the treatment of early-stage non-small cell lung cancer. N Engl J Med 1999;341: Keenan RJ, Landreneau RJ, Maley RH, et al. Segmental resection spares pulmonary function in patients with stage I lung cancer. Ann Thorac Surg 2004;78: Bando T, Yamagihara K, Ohtake Y, et al. A new method of segmental resection for primary lung cancer: intermediate results. Eur J Cardiothorac Surg 2002;21: Lewis RJ. The role of video-assisted thoracic surgery for carcinoma of the lung: wedge resection to lobectomy by simultaneous individual stapling. Ann Thorac Surg 1993;56: Okada M, Nishio W, Sakamoto T, et al. Effect of tumor size on prognosis in patients with non-small cell lung cancer: the role of segmentectomy as a type of lesser resection. J Thorac Cardiovasc Surg 2005;129: National Emphysema Treatment Trial Research Group. A randomized trial comparing lung-volume-reduction surgery

8 Ann Thorac Surg EL-SHERIF ET AL 2006;82: SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC with medical therapy for severe emphysema. N Engl J Med 2003;348: Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997;111: Daniel WW. Applied nonparametric statistics. Second Edition. Boston: PWS-KENT, 1990: Mehta CR, Patel NR. A network algorithm for the exact treatment of Fisher s exact test in RxC contingency tables. J Am Stat Assoc 1983;78: Kaplan E, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958;53: Peto R, Peto J. Asymptotically efficient rank invariant procedures (with discussion). J R Stat Soc Series A 1972;135: Sachdev M, Sun JL, Tsiatis AA, Nelson CL, Mark DB, Jollis JG. The prognostic importance of co-morbidity for mortality in patients with stable coronary artery disease. J Am Coll Cardiol 2004;43: Birim O, Zuydendorp HM, Maat AP, Kappetein AP, Eijkemans MJ, Bogers AJ. Lung resection for non-small-cell lung cancer in patients older than 70: mortality, morbidity, and late survival compared with the general population. Ann Thorac Surg 2003;76: Ginsberg RJ, Rubinstein LV, for the Lung Cancer Study Group. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Ann Thorac Surg 1995;60: Allen MS, Darling GE, Pechet TT, et al. Morbidity and mortality of major pulmonary resections in patients with early-stage lung cancer: initial results of the randomized, prospective ACOSOG Z0030 trial. Ann Thorac Surg 2006;81: Okada M, Yoshikawa K, Hatta T, Tsubota N. Is segmentectomy with lymph node assessment an alternative to lobectomy for non-small cell lung cancer of 2 cm or smaller? Ann Thorac Surg 2001;71: Swanson SJ, Jaklitsch MT, Mentzer SJ, Bueno R, Lukanich JM, Sugarbaker DJ. Management of the solitary pulmonary nodule: role of thoracoscopy in diagnosis and therapy. Chest 1999;116(Suppl):523S 4S Mery CM, Pappas AN, Burt BM, et al. Diameter of non-small cell lung cancer correlates with long-term survival: implications for T stage. Chest 2005;128: Kondo D, Yamada K, Kitayama Y, Hoshi S. Peripheral lung adenocarcinomas: 10 mm or less in diameter. Ann Thorac Surg 2003;76: Noguchi M, Morikawa A, Kawasaki M, et al. Small adenocarcinoma of the lung. Cancer 1995;75: Warren WH, Faber LP. Segmentectomy versus lobectomy in patients with stage I pulmonary carcinoma. J Thorac Cardiovasc Surg 1994;107: Miller DL, Rowland CM, Deschamps C, et al. Surgical treatment of non-small cell lung cancer 1 cm or less in diameter. Ann Thorac Surg 2002;73: Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109: Errett LE, Wilson J, Chiu RC-J, Munro DD. Wedge resection as an alternative procedure for peripheral bronchogenic carcinomas in poor-risk patients. J Thorac Cardiovasc Surg 1985;90: Pastorino U, Valente M, Bedini V, Infante M, Tavecchio L, Ravasi G. Limited resection of stage I lung cancer. Eur J Surg Oncol 1991;17: Read RC, Yoder G, Schaeffer RC. Survival after conservative resection for T1N0M0 non-small cell lung cancer. Ann Thorac Surg 1990;49: Landreneau RJ, Sugarbaker DJ, Mack MJ, Hazelrigg SR, et al. Wedge resection versus lobectomy for stage I (T1 N0 M0) non-small-cell lung cancer. J Thorac Cardiovasc Surg 1997; 113: Kodama K, Doi O, Higashiyama M, Yokouchi H. Intentional limited resection for selected patients with T1 N0 M0 nonsmall cell lung cancer: a single-institution study. J Thorac Cardiovasc Surg 1997;114: Koike T, Yamato Y, Yoshiya K, Shimoyama T, Suzuki R. Intentional limited pulmonary resection for peripheral T1 N0 M0 small-sized lung cancer. J Thorac Cardiovasc Surg 2003;125: GENERAL THORACIC DISCUSSION DR SCOTT J. SWANSON (New York, NY): That was a very interesting paper. I wonder if you can clarify a point for me. Did you show that the local recurrence was similar until 10 years and then it diverged, and if that s true, can you explain that? Also, if you understaged the patients, why wouldn t you expect to see a greater local recurrence in the sublobar resection group? DR EL-SHERIF: Thank you, Dr Swanson, we analyzed the 10-year follow-up period as a means of evaluating overall recurrence. Accordingly, we did not find a difference between the sublobar resection and lobectomy groups with regard to overall tumor recurrence. Most local recurrences will be identified within the first 2 years after resection. The relative understaging seen between lobectomy and sublobar resection patients related to the difference in lymph node yield may confound the accuracy in comparing tumor recurrence. This would potentially bias against the sublobar resection patients who had significantly fewer lymph nodes sampled as a group. These sublobar resection patients also were physiologically more impaired than the lobectomy patients, which would also potentially affect the overall survival time owing to death from unrelated disease. DR FRANK C. DETTERBECK (Chapel Hill, NC): I was surprised by the number of sublobar resections you had. It was about a third as many as the lobectomy patients, and I m wondering what your criteria for poor cardiopulmonary reserve are that preclude lobectomy. They seem to be fairly liberal, I think. Secondly, I think that in comparing the patients, what would probably be best is if you could compare the same patients by clinical stage and tumor size, in other words, clinical stage I, peripheral 2-cm tumor, in both groups as opposed to what we have here, which are probably discrepant stages and, of course, the whole issue of the accuracy of your postoperative pathologic staging, which is obviously different in the two groups. So if you look at clinical staging, can you give us a subset of your lobe resection patients and compare those two groups? DR EL-SHERIF: Thank you, Dr Detterbeck, our standard of surgical care for the good risk stage I non small cell lung cancer patient was lobectomy during the entire period reviewed for this investigation. However, we do see many patients with poor preoperative functional reserve for whom we do not recommend lobectomy but who could potentially obtain an important survival advantage with a sublobar resection that they could tolerate. The criteria to utilize sublobar resection was variable among surgeons in our group. There were multiple factors affecting the decision to recommend sublobar resection. The patient s pulmonary function was a primary determinant for many patients; however, other im-

9 416 EL-SHERIF ET AL Ann Thorac Surg SUBLOBAR RESECTION VS LOBECTOMY: STAGE I NSCLC 2006;82: portant medical comorbidities and advanced patient age were also important variables leading to the decision to perform less than lobectomy for these stage I lung cancer patients. We did not look at the patterns of recurrence between resection groups among patients with small, less than 2-cm tumors. We will be investigating this subset of patients in the future. DR SUDISH C. MURTHY (Cleveland, OH): This is just to get back to Dr Swanson s question. You must be caul when you re comparing a group of very high-risk patients from a medical standpoint to a group of more fit patients. Because of this type of analysis, you often have to normalize the data for the competing risk of non-cancer death (which is likely much higher in the patients undergoing limited resection). By not accounting for this, you may be overestimating the impact of your intervention. This is just a suggestion for your additional analyses. I enjoyed your presentation. DR EL-SHERIF: You are absolutely correct. DR DAVID RICE (Houston, TX): You alluded to the Japanese literature, and I think a point that I would make is that a significant number of patients in these trials have very earlystage bronchoalveolar cancer (BAC) or ground-glass opacities. What was the percentage of BAC in both groups in your study? DR EL-SHERIF: Twelve percent, without a differential between resection groups. DR RICE: Thank you. Online Discussion Forum Each month, we select an article from the The Annals of Thoracic Surgery for discussion within the Surgeon s Forum of the CTSNet Discussion Forum Section. The articles chosen rotate among the six dilemma topics covered under the Surgeon s Forum, which include: General Thoracic Surgery, Adult Cardiac Surgery, Pediatric Cardiac Surgery, Cardiac Transplantation, Lung Transplantation, and Aortic and Vascular Surgery. Once the article selected for discussion is published in the online version of The Annals, we will post a notice on the CTSNet home page ( with a FREE LINK to the full-text article. Readers wishing to comment can post their own commentary in the discussion forum for that article, which will be informally moderated by The Annals Internet Editor. We encourage all surgeons to participate in this interesting exchange and to avail themselves of the other valuable features of the CTSNet Discussion Forum and Web site. For August, the article chosen for discussion under the Adult Cardiac Dilemma Section of the Discussion forum is: Reduction Ascending Aortoplasty: Midterm Follow-Up and Predictors of Redilatation Gianluca Polvani, MD, Fabio Barili, MD, Luca Dainese, MD, Veli K. Topkara, MD, Faisal H. Cheema, MD, Eleonora Penza, MD, Sandro Ferrarese, MD, Alessandro Parolari, MD, PhD, Francesco Alamanni, MD, and Paolo Biglioli, MD Tom R. Karl, MD The Annals Internet Editor UCSF Children s Hospital Pediatric Cardiac Surgical Unit 505 Parnassus Ave, Room S-549 San Francisco, CA Phone: (415) Fax: (212) karlt@surgery.ucsf.edu 2006 by The Society of Thoracic Surgeons Ann Thorac Surg 2006;82: /06/$32.00 Published by Elsevier Inc