Ann Thorac Surg 2007;83:1831-1836
© 2007 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Pneumonectomy for Bronchogenic Carcinoma: Analysis of Factors Predicting Survival
Nithya Ramnath, MDa,*,
Todd L. Demmy, MDb,
Ana Antun, MDd,
Nachimuthu Natarajan, MSc, MSc,
Chukwumere E. Nwogu, MDa,
Gregory M. Loewen, DOa,
Mary E. Reid, PhDc
a Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York
b Department of Thoracic Surgery, Roswell Park Cancer Institute, Buffalo, New York
c Division of Cancer Prevention and Population Sciences, Roswell Park Cancer Institute, Buffalo, New York
d Department of Medicine, Jacobi Medical Center, South Bronx, New York
Accepted for publication December 18, 2006.
* Address correspondence to Dr Ramnath, Roswell Park Cancer Institute, Elm and Carlton St, Buffalo, NY14263 (Email: nithya.ramnath{at}roswellpark.org).
 |
Abstract
|
|---|
Background: The aim of this study was to identify risk factors associated with survival after pneumonectomy for non-small cell lung cancer.
Methods: This was a retrospective study of 155 patients who underwent a pneumonectomy for non-small cell lung cancer at Roswell Park Cancer Institute between 1986 and 2002. Medical record review ascertained information on preoperative assessment including pulmonary function tests and clinical characteristics, postoperative complications, and overall survival. Multivariate Cox proportional hazards models to calculate the hazard ratios and 95% confidence intervals were used. Kaplan-Meier cumulative survival curves (with log-rank p values) were generated for selected variables.
Results: The median age was 58 years at the time of surgery; 65% of patients were males. Squamous cell carcinoma (54%) and adenocarcinoma (33%) were the predominant histologic types. The median time to relapse was 11 months, and the overall median survival was 15.6 months. An American Society of Anesthesiology score of less than 3, squamous histology, and lower pathologic stage were significant independent predictors of improved survival. Current smoking status (hazard ratio = 1.87, 95% confidence interval: 1.30 to 2.70) and left tumor location (hazard ratio = 1.40, 95% confidence interval: 0.97 to 2.03) were associated with a trend toward poorer survival. Sixty-four patients (41%) had postoperative complications. The operative mortality from pneumonectomy was 9 of 155 (5.8%).
Conclusions: American Society of Anesthesiology score, histology, pathologic stage, smoking status, and location of the tumor were important predictors of survival in this patient sample. Pneumonectomy for non-small cell lung cancer carries an acceptable operative mortality and provides an important survival benefit.
|
Introduction
|
|---|
Non-small cell lung cancer (NSCLC) accounts for 80% of all lung cancers. Of all NSCLCs, only 30% are eligible for surgery, the only curative modality. For these surgically operable patients, anatomic lobectomy or pneumonectomy with mediastinal lymph node dissection is the accepted standard of care. More often, the location of the primary tumor (for example, hilar) dictates the need for a pneumonectomy. We specifically looked at outcomes after pneumonectomy for NSCLC at our institution. The aim of this study was to identify factors that may predict outcome.
|
Material and Methods
|
|---|
This is a single-institution retrospective study of patients who underwent a pneumonectomy for NSCLC at Roswell Park Cancer Institute between 1986 and 2002. The Institutional Review Board/Ethics Committee approved this study. The research was done under an approved existing data review protocol that was expedited, in accordance with the federal regulations. In accordance to Roswell Park Cancer Institute procedures, chart review met the criteria for a waiver of informed consent. In addition, a Health Insurance Portability and Accountability Act waiver was granted so that the investigator could obtain and use protected health information.
Of 220 consecutive patients identified, 160 charts were available with complete information. Of the remaining 60, we had information on date of pneumonectomy and date of death. There was information on side (right/left) for 52 of 60 patients; there were no significant differences in either survival or side between patients included in the final analysis and the total population (data not shown). In addition, patients who had a diagnosis of carcinoid were excluded, resulting in a final sample size of 155 patients.
Preoperative Evaluation
Patients were evaluated by history and physical examinations, radiographic tests (computed tomography [CT] scans of the chest and upper abdomen), pulmonary function tests, electrocardiograms, and a complete blood count and serum chemistry studies for organ function. Smoking status was recorded on all patients. Performance status was assessed according to the Eastern Cooperative Oncology Group system (score from 0 to 5). We used the American Society of Anesthesiology (ASA) scoring system in abstracting the comorbidity variables from the charts. The ASA classification system is based on the presence or absence of systemic disturbances: absent (class 1), mild (class 2), moderate (class 3), severe (class 4), or almost certain to cause death (class 5) [1]. Brain and bone scans were performed only if symptoms warranted. Most patients underwent preoperative bronchoscopy. Cervical mediastinoscopy or anterior mediastinotomy (Chamberlain procedure) was undertaken in patients with larger tumors (>3 cm) or if the mediastinal lymph nodes were larger than 1 cm on CT images. Split lung function tests (by ventilation/perfusion scintigraphy) or exercise tests were performed if clinically indicated. Patients without distant metastasis were considered for pneumonectomy if they had adequate cardiac and pulmonary reserve (predicted postoperative forced expiratory volume in first second of at least 0.8 L or mVO2 greater than 10 mL · kg–1 · min–1). Data regarding preoperative chemotherapy or adjuvant chemotherapy or radiation were collected. The ASA score and data extraction were verified in 90 patients by a second reader and found to be concordant.
Surgical Procedure
The operative approach included a thoracotomy and pneumonectomy with mediastinal lymph node dissection as primary therapy. Mediastinal lymph nodes were dissected according to the standard protocol and labeled for pathology review. Stage of disease was determined based on clinical assessment before the surgery and based on pathologic findings after the surgery. All postoperative complications were recorded. These are detailed in the results section. Operative mortality was defined as death occurring within 30 days after surgery (1 patient died on day 32 and another on day 35; they are included as postoperative deaths).
Patient Follow-Up
Patients were followed every 3 months for the first year, every 6 months for the second year, and then yearly with chest roentgenography. A CT scan was performed if the chest roentgenogram or symptoms warranted. Relapse was determined radiographically. Date of death was obtained from the tumor registry and the medical records. The mean follow-up time for the cohort was 982 days.
Statistics
Frequencies, proportions, means, and standard deviations were calculated for baseline characteristics and reported complications. Kaplan-Meier cumulative survival curves were generated, and the resulting curves were compared with log-rank tests. Potential risk factors for an adverse event were identified by Cox proportional hazard models. Adjusted hazard ratios (HR) and 95% confidence intervals (CI) were calculated. Any p value of 0.05 or less were considered significant. For all statistical analyses, STATA software, version 9.0 (STATA Corp, College Station, Texas), was used.
|
Results
|
|---|
Table 1
summarizes the characteristics of the 155 patients who underwent a pneumonectomy. The mean age was 58.0 years, with men slightly older than women. The majority of the patients were men (65%), former or current smokers (93.5%), and at pathologic stage III (46.4%). Squamous cell carcinoma (54%) and adenocarcinoma (33%) were the predominant histologic types. The median months to relapse was 10.8, and the median months to death was 15.6. Table 2
summarizes the reported complications from pneumonectomy. Of the 64 patients with any reported complication, cardiac arrhythmias and respiratory failure were the most common. Of the 7 patients who had a postoperative bronchopleural fistula, 4 had a left and 3 had a right pneumonectomy. None of these patients had neoadjuvant radiation, but 3 of 7 had preoperative chemotherapy.
Nine patients died within 35 days of surgery. Of the 9 patients, 5 were women and 4 were men. The mean age of these patients ranged from 32 to 81 years, with a median of 57 years. All 9 patients had ASA scores of III or IV. Six had a diagnosis of adenocarcinoma, and 3 had squamous cell carcinoma. The stages of cancer were stages I through IV (2 had lymphangitic carcinomatosis detected at surgery). One patient died during surgery of uncontrolled bleeding due to injury to the descending thoracic aorta. The other deaths are summarized in Table 3.
Figures 1 through 4
represent Kaplan-Meier survival curves for years after pneumonectomy by several variables. Figure 1 shows the overall unstratified survival of these patients in years. Approximately 50% survived just over 1 year. Figure 2 presents the Kaplan-Meier survival curve by ASA scores: low (I, II) versus high (III, IV). Both categories of ASA scores showed comparable survival until 12 months after surgery. At 1 year, the curves separated, and the lower scores had a significantly improved overall survival (log rank, p = 0.007). In a similar analysis, survival for patients with squamous cell carcinoma versus adenocarcinoma was compared (Fig 3). Overall, patients with squamous cell carcinoma survived significantly longer than patients with adenocarcinoma (log rank, p = 0.003). Finally, Figure 4 compares the survival of these patients by the pathologic stage of the tumor. As expected, stage I patients survived longer than did stage II, III, and IV patients combined. This trend was significant (log rank, p = 0.0003).
View larger version (9K):
[in this window]
[in a new window]
|
Fig 2. Survival curves for groups with American Society of Anesthesiology (ASA) scores I and II versus III and IV.
|
|
Figure 5
presents the Cox proportional hazard ratios and 95% confidence intervals for lung cancer survival by individual covariates included in a multiple regression model. The overall model was adjusted for age (continuous) and sex. Current smoking (versus former and never smokers), pathologic stages III and IV (versus stage I), and an ASA score of III or IV (versus I or II), significantly increased the risk of death. Left location of the tumor and pathologic stage II also increased the odds of a lung cancer death, but they were not significant (p = 0.05). In 56% of patients, there was a documented site of relapse. The more common sites of relapse were the contralateral lung (n = 37) followed by brain (n = 21), and the remaining relapses were in the spine, bones, and other sites (data not shown). Some patients had relapses in more than one anatomical site.
View larger version (9K):
[in this window]
[in a new window]
|
Fig 5. Cox proportional hazard ratios for lung cancer survival of selected covariates. 1Model adjusted for age and sex. 2Hazard ratios for comparisons: current smokers versus never and former smokers, squamous cell carcinoma versus adenocarcinoma, left versus right tumor location, stage II versus stage I, stage III/IV versus stage I, American Society of Anesthesiology (ASA) scores III and IV versus I and II.
|
|
|
Comment
|
|---|
Lobectomy and pneumonectomy with mediastinal lymph node dissection are accepted curative operations for resectable NSCLC. Because video-assisted thoracoscopic surgery leads to shorter patient stays and equivalent outcomes for patients with smaller tumors (T1/T2), there has been debate on the need for extensive parenchymal lung resections. There has been a decline in the number of pneumonectomies performed for NSCLC since 1980. In our institution, they currently represent less than 10% of all lung cancer operations (lobectomies and bilobectomies). In fact, our rates for pneumonectomy have ranged between 2.5% and 7.8% from 1985 to 2003. Some of the reasons include better staging with positron emission tomography (PET) scans and increased ability to perform sleeve vascular and bronchial resections to save pulmonary parenchyma.
Our study describes the outcomes after a pneumonectomy for lung cancer. These were consecutive patients in the time period noted. Diagnostic bronchoscopies were done for 108 patients. Preoperative bronchoscopies may have been performed as part of obtaining airway control in collaboration with anesthesia and not documented as separate procedure. Advances in CT scanning have made surprises during bronchoscopy less common. In fact, coronal reconstruction of the airways is possible with available software that allows better visualization of airways. With regard to left-sided pneumonectomies, we agree that it is somewhat unusual, both in terms of incidence and in that they seemed to be associated with a trend toward poorer survival, contrary to what is reported in the literature. That is indeed a peculiarity of our dataset. We analyzed other factors in this group, particularly stage, to see if there was an imbalance of poor prognostic factors in this group, but found none.
The overall operative mortality in our institution was 5.8%. This rate is acceptable and similar to other reported series on pneumonectomies for NSCLC [2–6]. The overall median survival was 15.6 months for the entire group. The overall 5-year survival was 21%. In contrast to other series, older patient age and female sex was not associated with poorer survival [7–10]. Patients who had adenocarcinoma fared worse than patients who had squamous cell carcinoma, in keeping with some of the other reports [11]. Predictably, patients with higher pathologic stages did worse compared with stage I patients. Patients who received neoadjuvant chemotherapy or radiation, or both, or adjuvant radiation (most of these patients had mediastinal lymph node involvement) did not have a significant survival advantage compared with patients who did not. The patients with stage IIIB cancer (n = 21) either had mediastinal invasion (diagnosed intraoperatively) or intralobar satellite nodules. Among the patients with stage IV cancer (n = 24), there were 8 patients with intrapulmonary metastasis and 6 patients with a solitary brain lesion. These patients would not be considered for surgical resection in modern practice, with the availability of PET scans. The median survival for stage III and IV patietns in this study was 12 months.
We found that the ASA score predicted survival at 12 months. Patients with a lower ASA score lived longer than patients with a higher ASA score, and that was statistically significant. The ASA scoring system was developed by anesthesiologists for preoperative risk assessments to predict perioperative adverse events, including mortality. Typically, this measure was not designed for use as a predictor of survival beyond the perioperative period. Most surgical series have shown a correlation between a high ASA score and 7-, 14-, and 30-day mortality [12–15]. There has been one report of patients undergoing surgery for supraglottic tumors in which the ASA score was predictive of crude 5-year survival [16]. Also reported in head and neck cancer patients, the ASA score was a better predictor of 4-year survival after surgery compared with other indices [17]. We found the ASA score to be a predictor of long-term survival. It is likely that, since the ASA score takes into account chronic conditions that affect the patient, it is more likely to be reflective of long-term morbidity as well as mortality. That is particularly true for patients with lung cancer who almost always have a chronic condition of chronic obstructive pulmonary disease.
It is true that this population had a disproportionate number of patients with advanced disease. We submit that preoperative staging might not have been adequate in these patients. Neither PET nor mediastinoscopy was routinely performed in all cases. There were no specific protocols; however, patients with locally advanced disease were treated with neoadjuvant chemotherapy or radiation, or both.
Positron emission tomography scanning, which has become "de rigor" in the past 5 years, has considerably changed the nature of patients undergoing thoracotomy. Routine use of preoperative PET scanning reduced futile thoracotomies to 25% for stages I and II (compared with 46% for conventional methods) and to 11% for stage III (from 29% using conventional methods) [18]. With respect to neoadjuvant chemotherapy, the intergroup study reported by Albain and coworkers [19] did not show an overall survival advantage for neoadjuvant chemoradiotherapy for stage III patients. It has been suggested that the study was underpowered to detect a small difference in overall survival with adjuvant surgery. The authors of the study cautioned against this approach, particularly with patients for whom pneumonectomy is a consideration, because of the higher morbidity and mortality noted in this group. However, studies of neoadjuvant chemotherapy (alone) do not bear this out, and it continues to be explored. Neoadjuvant chemotherapy has been shown to be of particular benefit for stage IIIA disease, especially when N2 nodes are converted to N0. The Bimodality Lung Oncology Team studies reported by Pisters and colleagues [20] showed that this approach was feasible. The results of the phase III trial are awaited. If downstaging can be accomplished and proven pathologically, lesser resections such as sleeve resections could be undertaken with equivalent results; the small number of patients eligible for this approach would likely make a statistically significant inference difficult.
Although we realize the limitations of a retrospective study at a single institution, we have long-term follow-up on more than 150 patients who underwent a pneumonectomy for NSCLC. In addition to the usual clinical variables of higher pathologic stage and current smoking status, a high ASA score at the time of surgery predicted poor long-term survival. We believe that with modern staging techniques for the right patient with early stage lung cancer, anatomical pneumonectomy will continue to provide a vital therapeutic option, with a potential for long-term survival.
|
References
|
|---|
- Dripps RD, Lamont A, Eckenhoff JE. The role of anesthesia in surgical mortality JAMA 1961;178:261-266.[Medline]
- Kadri MA, Dussek JE. Survival and prognosis following resection of primary non small cell bronchogenic carcinoma Eur J Cardiothorac Surg 1991;5:132-136.[Abstract]
- Licker M, de Perrot M, Hohn L, et al. Peri-operative mortality and major cardio-pulmonary complications after lung surgery for non-small cell carcinoma Eur J Cardiothorac Surg 1999;15:314-319.[Abstract/Free Full Text]
- Mizushima Y, Noto H, Sugiyama S, et al. Survival and prognosis after pneumonectomy for lung cancer in the elderly Ann Thorac Surg 1997;64:193-198.[Abstract/Free Full Text]
- Patel RL, Townsend ER, Fountain SW. Elective pneumonectomy: factors associated with morbidity and operative mortality Ann Thorac Surg 1992;54:84-88.[Abstract]
- Roxburgh JC, Thompson J, Goldstraw P. Hospital mortality and long-term survival after pulmonary resection in the elderly Ann Thorac Surg 1991;51:800-803.[Abstract]
- Alexiou C, Beggs D, Rogers ML, et al. Pneumonectomy for non-small cell lung cancer: predictors of operative mortality and survival Eur J Cardiothorac Surg 2001;20:476-480.[Abstract/Free Full Text]
- Bouchardy C, Fioretta G, De Perrot M, Obradovic M, Spiliopoulos A. Determinants of long term survival after surgery for cancer of the lung: a population-based study Cancer 1999;86:2229-2237.[Medline]
- Ouellette D, Desbiens G, Emond C, Beauchamp G. Lung cancer in women compared with men: stage, treatment, and survival Ann Thorac Surg 1998;66:1140-1144.[Abstract/Free Full Text]
- Ferguson MK, Wang J, Hoffman PC, et al. Sex-associated differences in survival of patients undergoing resection for lung cancer Ann Thorac Surg 2000;69:245-250.[Abstract/Free Full Text]
- Deneffe G, Lacquet LM, Verbeken E, Vermaut G. Surgical treatment of bronchogenic carcinoma: a retrospective study of 720 thoracotomies Ann Thorac Surg 1988;45:380-383.[Abstract]
- Marx GF, Mateo CV, Orkin LR. Computer analysis of postanesthetic deaths Anesthesiology 1973;39:54-58.[Medline]
- Prause G, Offner A, Ratzenhofer-Comenda B, et al. Comparison of two preoperative indices to predict perioperative mortality in non-cardiac thoracic surgery Eur J Cardiothorac Surg 1997;11:670-675.[Abstract]
- Prause G, Ratzenhofer-Comenda B, Pierer G, et al. Can ASA grade or Goldman’s cardiac risk index predict perioperative mortality?A study of 16,227 patients. Anaesthesia 1997;52:203-206.[Medline]
- Vacanti CJ, VanHouten RJ, Hill RC. A statistical analysis of the relationship of physical status to postoperative mortality in 68,388 cases Anesth Analg 1970;49:564-566.[Free Full Text]
- Nicolai P, Redaelli de Zinis LO, Tomenzoli D, et al. Prognostic determinants in supraglottic carcinoma: univariate and Cox regression analysis Head Neck 1997;19:323-334.[Medline]
- Reid BC, Alberg AJ, Klassen AC, Koch WM, Samet JM. The American Society of Anesthesiologists’ class as a comorbidity index in a cohort of head and neck cancer surgical patients Head Neck 2001;23:985-994.[Medline]
- van Tinteren H, Hoekstra OS, Smit EF, et al. Effectiveness of positron emission tomography in the preoperative assessment of patients with suspected non-small-cell lung cancer: the PLUS multicentre randomised trial Lancet 2002;359:1388-1393.[Medline]
- Albain K, Swann RS, Rusch VR, et al. Phase III study of concurrent chemotherapy and radiotherapy (CT/RT) vs CT/RT followed by surgical resection for stage IIIA(pN2) non-small cell lung cancer (NSCLC): outcomes update of North American Intergroup 0139 (RTOG 9309); 2005 ASCO annual meeting proceedings, part I of II J Clin Oncol 2006;23(Suppl):2005(abstract 7014).
- Pisters KM, Ginsberg RJ, Giroux DJ, et al. Bimodality Lung Oncology Team Induction chemotherapy before surgery for early-stage lung cancer: a novel approach J Thorac Cardiovasc Surg 2000;119:429-439.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
T. Gudbjartsson, E. Gyllstedt, A. Pikwer, and P. Jonsson
Early surgical results after pneumonectomy for non-small cell lung cancer are not affected by preoperative radiotherapy and chemotherapy.
Ann. Thorac. Surg.,
August 1, 2008;
86(2):
376 - 382.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
