The overall operative mortality for lung cancer surgery in Japan has been reported to be satisfactorily low [1], with a 30-day death of 1.3%. In addition, the rate of pneumonectomy (PN) is also relatively low because surgeons generally try a bronchoplasty to avoid pneumonectomy. A bronchoplastic resection achieves local control and preserves pulmonary function without missing the complete removal of the central located tumors. Since Price-Thomas [2] first applied this procedure for therapeutic option, it has become a basic therapeutic strategy in the general hospitals provided that oncologic radicality and reconstructive aspects are evident during the past 20 years [3,4]. In contrast, pneumonectomy for lung cancer has been reported to be associated with significant morbidity and mortality [5–7], including postpneumonectomy lung edema, adult respiratory distress syndrome (ARDS), bronchopleural fistula, and postpneumonectomy syndrome [7]. Recent studies have compared pneumonectomy and sleeve lobectomy (SL) procedures in terms of late outcome and morbidity using matched patients [8,9]. A meta-analysis [6] of comparisons between a sleeve lobectomy and pneumonectomy in stages I and II non-small cell lung cancer (NSCLC) revealed an advantage of sleeve lobectomy for mortality (4.1% vs 6.0%), while there was no significant difference in 5-year survival rate (51% vs 49%). We always employ a sleeve lobectomy when technically possible, however, a pneumonectomy is inevitable in certain situations, when a complete resection could not otherwise be achieved and provided that pulmonary function permits it. Recently induction therapy followed by surgery has emerged as a promising treatment for advance staged NSCLC [10,11], for which an extensive resection is also required by using a sleeve resection or pneumonectomy [12]. In the present study, we retrospectively analyzed our experiences of the two procedures without adding selection bias. The objective of the present retrospective analysis was to characterize the indications, patient demographics, morbidity, mortality, and late outcomes over time in patients who underwent sleeve lobectomy and pneumonectomy. Between January 1984 and December 2003, a pulmonary resection was performed for 1211 consecutive patients with primary NSCLC (920 males, 291 females; average age 62.5 years) at Toneyama National Hospital in Japan. Ten patients who underwent a carinal resection, including five that received a sleeve pneumonectomy were excluded from the analysis. In our institutional experience, tracheo-bronchoplastic procedures using an end-to-end anastomosis were performed [3] in the early era (1984–1993), and then telescoping techniques have been achieved in the recent era (1994–2003) [4], including those who received induction chemoradiotherapy [12]. After induction therapy, re-staging of local tumor extension and nodal status were assessed to determine the operative procedures. Survival, perioperative morbidity, and sites of recurrence were analyzed and compared in 62 consecutive patients who underwent a sleeve lobectomy, and 110 consecutive patients who underwent a pneumonectomy. The sleeve lobectomy procedures included 35 right upper lobectomies, 1 middle lobectomy, 2 upper and middle lobectomies, 6 middle and lower lobectomies, 12 left upper lobectomies, and 6 left lower lobectomies. A concomitant pulmonary angioplasty was performed in 12 patients (19.3%) in the SL group. As for the PN group, we performed 38 right and 72 left pneumonectomies (Table 1). Comparison of patients’ characteristics of sleeve lobectomy and pneumonectomy All patients were evaluated for predicted postoperative function, as we have previously reported [13]. However, patients with borderline pulmonary functional reserve were posed a special preoperative pulmonary rehabilitation if necessary. Furthermore, an aggressive pulmonary toilet, including a bronchoscopy, was performed during anesthesia and prior to extubation. Other postoperative management included, early ambulation, bronchial toilet including bronchoscopy, as necessary and prolonged low-flow nasal oxygen supplementation, as necessary. Postoperative complications were divided into minor complications, which included pneumonia based on chest radiographic findings that required antibiotics, hypoxemia, atelectasis, a persistent air leak for more than 7 days, chyle leak, and cardiac arrhythmia that could be treated medically, and major complications, which included myocardial infarction, respiratory failure requiring reintubation with mechanical ventilation, and bronchopleural fistula, empyema, and severe chylothorax requiring reoperation. The hospital mortality included 30-day mortality and operation-related death during the same hospitalization for up to 3 months. Most sites of relapses were documented through hospital re-admission. A locoregional recurrence being defined as a recurrence that developed within the ipsilateral hemithorax including the mediastinum. Data are reported as mean ± standard deviation or as a proportion. Survival rate was estimated by the Kaplan–Meier method, and the log-rank test was used to compare survival rates between the two groups. Other comparisons were made using an unpaired t-test or χ2-test. Significance was accepted as a p value of less than 0.05. Comparisons of the patients’ clinical characteristics and operative results for the SL group (n = 62) and PN group (n = 110) are shown in Table 1. There were no significant differences in age, gender, or prevalence of histological subtypes including adenocarcinomas, squamous cell carcinomas, and others (Table 1). A sleeve lobectomy on the right side and left pneumonectomy were more common than those on the respective contralateral sides (p = 0.0001). Pathological staging for the SL and PN groups were as follows: stage IA, B (SL: n = 29/62, 46.8%, PN: n = 24/110, 21.8%), stage IIA, B (SL: n = 16/62, 25.8%, PN: n = 14/110, 12.7%), stage IIIA, B (SL: n = 17/62, 27.4%, PN: n = 70/110, 63.6%), with significant differences found between the groups (p = 0.0069) (Table 1). Namely, 72.6% of the SL group was stage I or II disease, whereas those comprised 34.5% in the PN group. Tumor size was significantly greater in the PN group than that of the SL group (p = 0.0026). There were no significant differences in the ratio of the combined resection including the chest walls, pericardium, diaphragm, or great vessels (SL, 30.6% vs PN, 44.5%, p = 0.0645) or the rate of induction therapy (SL, 25.8% vs PN, 17.2%, p = 0.236) between the groups. Three SL group patients, who received induction therapy, had serious complications, and finally died of empyema or respiratory failure, respectively, while one PN group patient who underwent a right pneumonectomy following induction chemoradiotherapy died of adult respiratory distress syndrome 6 days postoperatively. Three cases were considered to be technically hazardous, which consisted of one patient who had a history of tuberculosis pleuritis and underwent a pneumonectomy, and two patients with T4 tumors that had severely invaded to the aortic arch and left atrium, respectively. The 30-day postoperative mortality was 1.6% (1/62) in the SL group and 1.8% (2/110) in the PN group (Table 2). As for the hospital deaths, the rate was 4.8% (3/62) in the SL group and 3.6% (4/110) in the PN group. Comparisons of the early (1984–1993) and recent (1994–2003) eras did not show significant differences in morbidity and mortality, though recent pneumonectomy cases showed a higher rate of mortality than those in the early era (statistically not significant) (Table 2). Regarding respiratory and cardiac complications, there were no significant differences in respective complications or overall morbidity between the groups (SL, 45.2% and PN, 40.9%), as shown in Table 3. Local and distant recurrence rates were 9.7% and 29.0%, respectively, in the SL group. They were 10.9% and 42.7%, respectively, in the PN group, which were not significantly different between the groups (Table 3). Two patients in the SL group were later treated with a completion pneumonectomy without later recurrence. Further four patients in the PN group who were free of relapse required home oxygen therapy (HOT) in the late postoperative period, whereas none in the SL group required HOT throughout the 5-year follow-up period. The overall 5-year survival rates in the SL and PN groups were 54.3% and 32.9%, respectively (p = 0.0028) (Fig. 1). However, there was no difference in 5-year survival rate in pathologically stage I and II patients (SL, n = 45 and PN, n = 38) between the groups (Fig. 1). Moreover, as for the patients who received induction therapy with clinical stage III patients, the PN group (n = 19) showed a marginally better survival rate compared to the SL group (n = 16) (Fig. 3). Time trends of surgical results and outcome: sleeve lobectomy versus pneumonectomy Details of description and classification of complications Overall survival rates for 62 sleeve lobectomy patients (SL group) and 110 pneumonectomy patients (PN group). A significant difference was found between the SL and PN groups (p = 0.0029). Overall survival rates for 16 sleeve lobectomy patients (SL) and 19 pneumonectomy patients (PN group) who received induction therapy. No significant difference was found between the SL and PN groups (p = 0.061). Resectability for locally advanced lung cancer is improving with recent advances in perioperative care, surgical techniques [4,14,15], and induction therapy [10–12], which downstages the tumors to render them resectable. Thus, avoidance of pneumonectomy can be achieved in particular patients at an early disease stage, and the ratio of sleeve lobectomy to pneumonectomy (SL/PN) has nearly reached 1.0 in the recent reports [8,14–17]. It has been shown that a sleeve resection can offer a better long-term survival and quality of life (QOL) in patients with relatively early stage disease [6,13,15]. In the current retrospective analysis, overall survival in p-stage I/II was similar in both groups. Over the time, the rate of operative mortality decreased in patients who underwent standard lobectomy at our institution, while there was a slight increase in mortality over the time in both the SL and PN groups. An explanation may be due to differences in the severity of disease in the recent era, while the use of extended resection and/or induction therapy may also be related. All three patients with hospital death in the SL group received induction chemoradiotherapy. In the light of this trend, 35 (40.2%) of the 87 patients in the recent era received induction therapy, which may adversely affect the surgical morbidity and mortality [10–12]. In addition, our series indicate that the recent sleeve lobectomy and pneumonectomy procedures encompassed a variety of additional procedures, including combined resection of adjacent organs, mainly because of locally advanced cancer [12,14]. There have been many retrospective analyses of the operative mortality and morbidity of sleeve lobectomy and pneumonectomy for NSCLC [5–7,9,14–22] (Table 4). Recent reports have shown that a sleeve lobectomy can be performed with a much lower rate of operative mortality (1.2–7.5%) as compared to a pneumonectomy (4.9–12.0%) [6,14,20–22]. In some reports [4,15], sleeve lobectomy is safer than pneumonectomy even in patients who received induction therapy. Literature on operative mortality of lobectomy, sleeve lobectomy, and pneumonectomy for non-small cell lung cancer Considering the high percentage of patients who underwent an extended resection and/or induction therapy, the present surgical results for both procedures seemed to be acceptable in terms of morbidity and mortality [12]. In the current series, no significant differences were found for local relapse rates between the SL and PN groups, which were similar to those of the previous reports [16,19,23]. We consider that a pneumonectomy following induction therapy for complete resection of locally advanced NSCLC is a last option and a difficult challenge to the thoracic surgeons. Regarding the late outcome (Fig. 1), it was considered that the two groups did not have the same biologic disease and stage for an adequate comparison of survival for the total patients. It is well documented that TNM stage and nodal status are important to obtain the best benefits of a sleeve lobectomy or pneumonectomy [5,8,21–25]. A 5-year survival was similar in patients of the stage-matched (p-stage I/II) was also demonstrated (Fig. 2). On the other hand, pneumonectomy following induction therapy (Fig. 3) resulted in a marginally better survival rate than that of sleeve lobectomy in the clinical stage III patients. Fair 5-year survival of patients with pneumonectomy after induction therapy is an encouraging result in the current study. Furthermore, we believe that an optimal preoperative functional evaluation of cardiorespiratory risks [12], meticulous surgical techniques, and perioperative care, led to the acceptable immediate results in the present study. Overall survival rates for 45 sleeve lobectomy patients (SL group) and 38 pneumonectomy patients (PN group) with pathological stages I and II. No significant difference was found between the SL and PN groups (p = 0.567). Based on our results, we did not think that there are any drawbacks to a pneumonectomy in terms of morbidity, mortality, and overall survival as compared to a sleeve resection, despite our finding that pneumonectomy procedures were associated with 49 combined resections of the adjacent organs and 19 resections following induction therapy. Just one opponent was that four pneumotectomized patients required HOT and eventually died without cancer recurrence, implying a potential of late cardiorespiratory failure in terms of QOL. In retrospect, the sleeve lobectomy and pneumonectomy procedures performed for the central tumors, presumably with different stages, were done efficiently, with acceptable morbidity, mortality, and 5-year survival. The present indication of pneumonectomy is aimed to perform a curative resection of locally advanced lung cancer in particular following induction therapy as a final surgical option otherwise unresectable, and the properly selected patients will likely benefit by a complete resection with a pneumonectomy. The authors wish to thank Dr Osamu Kuwahara, Dr Hideki Dohi, Dr Yasuhiko Tani, Dr Kiyohiro Fujiwara, Dr Hiroki Kishima, Dr Keiji Inada, Dr Yukiyasu Takeuchi, and Dr Masanobu Hayakawa, former attending thoracic surgeons, for contribution to this project, and thank Miss Yukari Hirai and Miss Machiko Inoue for their secretary assistance.
No takes yet. Share an insight, caveat, or question.
Takeda et al. (2006) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: