Bone cancer rib

Study objectives: The treatment of patients with non-small cell lung cancer (NSCLC) that is invading the chest wall is still debated. We aim to illustrate the improvements in treatment results that have occurred over last decade.

Design: Retrospective analysis of our experience and an overview of the literature.

Setting: Department of Surgery, San Giuseppe Hospital, University of Milan.

Patients: From January 1970 to December 1999, of 2,738 patients with NSCLC, we operated on 146 patients (5.4%) with chest wall invasion by NSCLC. Superior sulcus tumors and tumors invading the diaphragm or mediastinum were excluded. We reclassified all cases according to the current TNM classification.

Results: We registered one postoperative death (0.69%) and five major complications (3.4%). From 1970 to 1979, of 32 patients, 10 underwent an exploratory thoracotomy (ET) and 22 underwent a radical resection (stage IIB disease, 17 patients; stage IIIA disease, 5 patients). The 5-year survival rate was 22.7% (25% for stage IIB disease). From 1980 to 1989, of 67 patients, 11 underwent an ET and 56 underwent a radical resection (stage IIB disease, 34 patients; stage IIIA disease, 12 patients; stage IIIB disease, 5 patients; and stage IV disease, 5 patients). The survival rate following radical resection was 14.1%, ranging between 23.5% for patients with stage lib disease and 0% (3 years, 14%) for those with stage IIIA disease. From 1990 to 1999, of 47 patients, 2 underwent an ET, 2 underwent an exploratory thoracoscopy, and 43 underwent a radical resection (stage IIB disease, 23 patients; stage IIIA disease, 20 patients). The survival rate was 42.7% (stage IIB disease, 78.5%; stage IIIA disease, 7.2%).

Conclusions: Considering the low morbidity, mortality, and significant improvement in survival during the last decade, we advocate the performance of radical en bloc resection for the treatment of chest wall invasive NSCLC.

Key words: chest wall invasion; lung cancer; surgical treatment; survival

Abbreviations: ET = exploratory thoracotomy; NSCLC = non-small cell lung cancer

**********

Advances in surgery and anesthesia and the improvement of adjuvant and neoadjuvant treatment gradually have extended the indications for the surgical treatment of lung cancer. Patients with tumors that formerly were considered to be inoperable, such as lung cancer invading the chestwall, which accounts for 5 to 8% of lung cancers, are now candidates for resection. (1,2) Coleman (3) reported the first important series regarding the resection of tumors invading the chest wall in 1947. However, the validity of this approach was debated for years.

A personal experience of 146 patients who had been treated for lung cancer invading the chest wall over the past 30 years led us to analyze the international literature to examine unsolved problems and to focus on the aspects of treatment that are unquestionably accepted.

MATERIALS AND METHODS

From January 1970 to December 1999, 146 of 2,738 patients (5.4%) with non-small cell lung cancer (NSCLC) with chest wall involvement underwent surgery. We did not include in this analysis the patients with superior sulcus tumors invading the thoracic inlet or those with tumors infiltrating the diaphragm.

During the 1970s, the available diagnostics included chest radiography, tomography and bone scan. Total-body CT scanning became available during the 1980s, followed by high-resolution CT scanning and routine thoracoscopy during the 1990s. Mediastinoscopy was performed routinely during the 1970s. During the 1980s, we performed mediastinoscopy using the same criteria as we used for other lung cancers, that is, only in cases of suspected N3 disease or extracapsular N2 invasion (ie, nodes with clinical or CT scan signs suggesting the infiltration of surrounding tissues). Since the 1990s, we have employed mediastinoscopy only in cases of suspected N3 disease, as N2 disease is normally assessed during thoracoscopic exploration. We always carry out a thoracoscopic exploration as the first step of an operation for lung cancer.

Considering the above-mentioned progress in approach and diagnostics, we divided our 30-year long series into three decades (ie, 1970 to 1979, 1980 to 1989, and 1990 to 1999) [Tables 1 and 2]. We reviewed all cases at the time of this study and reclassified them according to the present TNM classification. (4) The chief author of this study took part in or personally carried out all the operations in this series.

1970 to 1979

Of the 32 patients who underwent surgery during this decade, 10 (31.3%) underwent an exploratory thoracotomy (ET), 6 (18.7%) underwent a pneumonectomy, and 16 underwent a lobectomy (50%). Sixteen patients underwent extrapleural resection, and 6 patients (pneumonectomies, 2 patients; lobectomies, 4 patients) underwent en bloc resection of the ribs that had been invaded by the tumor (range, two to four fibs). Rib reconstruction was carried out in three cases using Kirschner needles. A radical resection was achieved in all the 22 patients who were operated on. Histology revealed 18 squalnous cell carcinomas and 4 adenocarcinomas. There were 17 stage IIB and 5 stage IIIA tumors (stage T3N1, 3; stage T3N2, 2 patients) according to the TNM classification. Nolle of the patients underwent postoperative chemotherapy or radiotherapy.

1980 to 1989

Of the 67 patients operated on during this period, 11 (16.4%) underwent an ET, 13 (19.4%) underwent a pneumonectomy, and 41 (61.2%) a lobectomy. Two patients (3%) with poor respiratory function underwent wedge resections. The resection of one or more fibs (range, one to four fibs) was carried out in 27 patients. The chest wall defect was repaired in six patients using different materials (metal struts, two patients; Kirschner needles, one patient; Mersilene mesh, three patients). Resection was achieved in 56 patients.

Histology revealed squamous cell carcinoma in 30 patients and adenocarcinoma in 26. Infiltration of the resection margins exposed a nonradical resection in five patients. There were 36 stage IIB tumors and 20 stage IIIA tumors (stage T3N1, 7 patients; stage T3N2, 13 patients) according to the TNM classification. (4) Eighteen patients underwent postoperative radiotherapy, and 1 patient underwent postoperative chemotherapy.

1990 to 1999

Forty-seven patients were treated during this last decade. Routine thoracoscopy revealed pleural dissemination of the tumor in two patients. In two other patients, thoracoscopy could not be adequately performed because of pleural adhesions. Subsequent surgical exploration revealed mediastinal invasion, and the operation was limited to an ET. The rate of ETs in this series was therefore 4.25%. Resection was achieved in 43 patients and consisted of 4 pneumonectomies (8.5%), 31 lobectomies (66%), and 8 wedge resections (17%), the latter having been performed because of poor respiratory function.

Eighteen patients (pneumonectomies, 2 patients; lobectomies, 13 patients; wedge resections, 3 patients) underwent resection of one or more ribs (range, one to three ribs). Chest wall reconstruction was not needed because all the parietal defects were limited or posterior. Histology revealed microscopic invasion limited to the parietal pleura in 8 patients and confirmed extrapleural invasion in 10 patients. There were 21 squamous cell carcinomas and 22 adenocarcinomas that were found. There were 23 stage IIB tumors and 20 stage IIIA tumors (stage T3N1 tumors, 15; stage T3N2 tumors, 5). All patients underwent postoperative radiotherapy, which was associated with preoperative or postoperative chemotherapy in 28 patients.

No patient was lost at follow-up. Survival was calculated from the date of surgery until death, or until the last follow-up, and was estimated by applying the Kaplan-Meier product limit method. The SD was calculated using the Greenwood method, and the survival analysis was evaluated with the log rank test.

RESULTS

Of 146 patients, we registered 1 hospital death (0.69%) due to respiratory insufficiency in a patient who had undergone upper lobectomy and en bloc resection of three ribs. The overall morbidity rate reached 8% when minor complications (eg, protracted air leaks, wound infection, arrhythmias) also were taken into account but was 2.0% if only major complications (bronchopleural fistulas, two patients; pulmonary embolism, one patient) were considered. During the last decade, we registered only minor complications in 2.8% of patients. During the second decade, five patients developed metastatic disease only 2 months after undergoing surgery, and clearly their tumor had been understaged.