Lung malignancies are the leading cause of cancer death in United States today in both men and women. A heightened sense of awareness of this problem by primary care physicians and improved imaging techniques have yielded a low rate of unresectable tumors at thoracotomy and a subset of patients with small primary site neoplasms. Ongoing research in the field of oncologic agents and regimens has resulted in tumor downsizing in greater than half of patients. Newer modalities such as positron emission tomography have shown promise in terms of predicting the sites of metastatic disease.
Despite advances in pharmacotherapy, complete tumor resection remains the goal of thoracic oncologic surgery. As a prelude to lung resection, complete staging is crucial. This usually entails computed tomography of the chest and upper abdomen, to include views of the adrenal glands. Bone scanning and cranial imaging is performed pre-operatively if clinically indicated. Obtaining tissue for a histologic diagnosis is achieved by flexible bronchoscopy with brushings, washings, and/or biopsies. In addition to providing tissue for diagnosis, valuable information regarding tumor location and line of bronchial resection is garnered from this simple procedure.
Debate continues over the need for mediastinoscopy in light of radiographically negative (i.e., < 1 cm diameter) mediastinal lymph nodes. False negative rates of as much as 18% have been referenced in patients with these nodal properties. As a prelude to thoracoscopic lobectomy, almost all investigators have mandated pre-operative evaluation with mediastinoscopy. The finding of ipsilateral mediastinal disease is an indication in most cases for neoadjuvant chemotherapy and radiation followed by restaging and resection. Patients with contralateral mediastinal nodal disease, given the current data available, are best treated with chemotherapy and radiation only, although clinical trials that add surgical resection are under review.
Standard approaches to the thorax for lobectomy/pneumonectomy are via an incision that begins between the posterior border of the scapula and the vertebral bodies and follows a path one to two fingerbreadths below the tips of the scapula and extends anterior to the mid-axillary line. Extension of this incision anteriorly is sometimes necessary for better exposure to the hilum of the lung. A variety of muscle-sparing variants of the standard approach have been devised, however, frequently a portion of the latissimus dorsi and serratus anterior muscle are sacrificed to improve exposure. This is followed by division of the intercostal muscles, generally in the fifth or sixth interspace, with placement of a rib spreader.
Given the invasive nature of this approach it is not uncommon for patients to have a protracted course of recovery after undergoing lung resection. Even after this initial six to eight week recovery course, post-thoracotomy pain syndromes with paresthesias over the anterior chest and wound are common.
The introduction of videothoracoscopy to treat pleural effusions and provide an access portal for lung and pleural biopsy has revolutionized how surgeons treat intrathoracic malignancy. In addition to lung biopsy, VATS techniques have been employed to assist in nodal sampling and evaluate visually depth and location of tumor invasion as a screen test for potential resectability. The natural progression of thoracoscopic techniques has brought to the forefront the feasibility of thoracoscopic, anatomic lung resection.
As with any “minimally invasive” approach, the guiding principle of the surgeon must be the provision of an oncologically sound procedure and providing patients with as good or better an operation than secured by open techniques. To this end thoracoscopic lobectomy remain true to anatomic dissection and nodal retrieval that has become the gold standard of oncologic lung surgery.
Understanding the changes in physiology that occur with lung resection, candidates for thoracoscopy resection need a clinical assessment of their existing lung function. This can be achieved clinically by a detailed examination of a patient's daily activities. Formal pulmonary function testing is frequently used with analysis of arterial blood gases to provide a quantitative assessment of function. After screening for pulmonary function, cardiac function, and radiographic evidence of resectability the patient is considered for surgery.
Given the technical constraints of the operation, thoracoscopic lobectomy has been limited to patients with T1 (i.e., < 3 cm diameter) lesions that are located away from hilar structures. Three to four port sites are employed with the main working site being in the anterior axillary line at the fourth or fifth interspace. The utility of this site is its ready access to the superior and inferior pulmonary vessels for endoscopic stapling. After exploration of the thoracic cavity to rule out extrapulmonary spread of malignancy, pleural reflection around the hilum is divided to expose the pulmonary vessels.
For upper lobectomies, arterial branches to the superior and anterior segments of the upper lobe tend be adjacent to one another and after careful dissection with either endoscopic and/or conventional instruments, an endoscopic stapler can be used to divide these branches. Access to the posterior segmental branch most often requires division of the major fissure to locate the vessel. Similarly, identification of the main lower lobe branch of the pulmonary artery is made possible by definition of the major fissure.
Optimal positioning of the lung with cranial traction on the upper lobe and caudal traction on the lower lobe allows the surgeon to place the endoscopic stapler through a port site adjacent to the posterior border of the scapula at the fourth interspace. Care must be taken to define the plane of dissection by first identifying the bronchial branches to the upper, middle/lingula, and lower lobes. Remembering that the middle lobe and lingula belong anatomically to the upper lobe, dissection anterior to the vessels in this region will allow a clean staple line across lung parenchyma only. Failure to define this plane can result in an inadvertent division of either the middle lobe arterial branch, or the arterial branch to the superior segment of the lower lobe.
Having completed division of the arterial branches, either the superior or inferior pulmonary veins are then dissected circumferentially. The vessels are divided as well using a linear endoscopic stapler with a vascular load (usual 2.5 mm staples). Once again, the operator must be aware of the location of the venous drainage of the middle lobe which although is generally to the superior pulmonary vein, can also drain via collaterals to the inferior vein.
Finally, with the lobe pedicled on its bronchus, a linear stapler with a thicker load (i.e., 4.8 mm) is used to complete the lobectomy. The specimen is placed in an endoscopic bag and delivered through the anterior-most port site by enlarging it to its smallest possible diameter.
Nodal sampling along the segmental bronchi (level 11) and mainstem bronchi (level 10) is performed, along with a thorough examination of the remainder of the mediastinal nodal stations. Specimens obtained from these areas provide critical staging information and should as well be delivered through the chest wall via an endoscopic bag to prevent chest wall implantation of tumor.
As manipulation of the lobe is limited during thoracoscopy, size of the tumor is a critical factor to consider prior to deciding to attempt thoracoscopic lobectomy. Adhesions from post-obstructive pneumonia or neo-adjuvant radiotherapy can also present a formidable challenge to the surgeon prior to beginning work on the lobectomy itself. Failure to mobilize the lung completely will limit the surgeon in attempts to divide the fissure. This step is key as it not only anatomically defines the extent of resection, but also provides the required exposure to avoid injury to adjacent structures. An incomplete or poorly defined fissure is a technical challenge as one needs to use the bronchial branches posteriorly and the pulmonary veins anteriorly as the anatomic landmarks along which to place the staple line. Excessively thick pulmonary parenchyma makes even fastidious stapling more prone to lung injury and small air leaks.
Remembering that the pulmonary vessels (especially arteries) are especially fragile, undue traction and torsion can result in bleeding or intimal injury with endoluminal thrombus formation. A careful, clean dissection with smooth passage of the stapling device is an essential part of good surgical technique. Should bleeding be encountered during dissection, control of the vessel should be achieved prior to continuation of the operation. Injudicious application of a staple line can result in stenosis or occlusion of the main pulmonary artery or a segmental branch to a region of lung to be preserved. Conversion to an open procedure should be contemplated early in the course of bleeding to prevent excessive hemorrhage and adjacent vascular injury. Adherence to this principle will prevent the need for unexpected pneumonectomy or extensive vascular repair procedures.
Dissection in the region of the bronchus should seek to sweep all nodal and associated tissue distally toward the lobe to be resected. This not only fulfills the oncologic mandate for a thorough node dissection, but also will provide a clean, well-defined structure across which a linear stapler can be fired. Excessive tissue, or lack of a perpendicular cut across the bronchus, can result in poor tissue approximation with post-operative air leak and/or luminal impingement on a remaining lobe’s bronchus. As in open procedures, the surgeon should ask the anesthetist to inflate the remaining lung segments once having applied the stapler across the bronchus and after closing the jaws. Any question of poor lung aeration should be addressed via flexible bronchoscopy or by reapplication of the stapler prior to firing.
“Thoracoscopic” lobectomy is a loosely applied term for anatomic lung resection performed via a procedure other than a standard muscle splitting/sparing thoracotomy. This minimally invasive approach almost uniformly requires a limited anterior thoracotomy through which the lung specimen is removed from the thorax. Surgeons often utilize this access portal for all, or a portion of the hilar vessel and bronchus dissection. An anatomic dissection of the vessels with individual suture ligation or stapling of the vessels is employed in all instances. A 1998 survey by Yim, et al. found that of the 33 of 45 thoracic surgeons contacted who replied, the majority had greater than 5 years operative experience and/or were working in an academic institution.1 One-third of the respondents used VATS approach in > 40% of their lobectomies, whereas one-third used it in less than 10%.
One of the earliest series of thoracoscopic lobectomies reported was by Walker, et al. in 1993.2 Eleven patients underwent pulmonary resections. Ten patients had peripheral pulmonary opacities and one a lobar bronchietasis. Mean operative time decreased from 3.3 hours overall to 2.3 hours in the final five cases. Blood loss was estimated to be less than 100 cc. There was no conversion to an open procedure and no major complications noted. That same year, Kirby and colleagues reviewed their experience with 44 patients with primary bronchogenic carcinoma accepted as candidates for potential VATS lobectomy.3 At the time of operation, three patients were found to have N2 disease and were excluded from the study group. In the remaining group of 41 patients, 35 underwent a successful thoracoscopic lobectomy (success rate = 85%). There were no major intraoperative complications and the mean hospital stay for the group was 5.7 ± 1.6 days.
Weber et al. examined results of thoracoscopic versus open lobectomy for benign disease in 117 patients between 1992 and 1999.4 Surgical indications included bronchietasis, chronic infections, tuberculosis, emphysema, arteriovenous malformations, and severe hemoptysis. Sixty-four of 117 operations were completed thoracoscopically. Duration of chest tube drainage, blood loss, and hospital stay while lower in the thoracoscopic group and did not differ significantly from that of the open group.
Nomori and colleagues sought to compare Stage I lung cancer patients undergoing VATS (n = 33) versus anterior limited thoracotomy (ALT) (n = 33) lobectomies.5 A total of 38 VATS lobectomies were attempted, with 5 conversions to open procedures (13.2% conversion rate). Compared to the ALT group, VATS patients had less post-operative pain and lower analgesic requirements between POD 1 and 7. No significant differences were noted between the groups however after POD 14. Similarly, there were no significant differences in post-operative impairment in vital capacity, respiratory muscle strength, or 6 minute walk testing.
Demmy and Curtis reviewed their series of 22 high-risk (FEV1 < 1.5 L or < 50% of predicted) patients undergoing an attempted or successful VATS lobectomy for lung cancer.6 Three of the 22(13.6%) required conversion to an open procedure due to adhesions, unfavorable anatomy, or bleeding. They had three operative mortalities in the VATS group (3/19 = 15.8%), one due to severe pneumonia, and two due to gastrointestinal complications associated with steroid use.
The first published series to focus on complications of VATS lobectomy was reported by Daniels, et al.7 Exclusion criteria for thoracoscopic lobectomy included tumors > 5 cm in diameter, T3 tumors, cancers with visible endobronchial tumor at bronchoscopy, extensive N1 disease on CT scan, and N2 disease at mediastinoscopy. Thoracoscopic lobectomy was performed successfully in 108/110 patients (98.1%) with 2 patients requiring conversion to an open thoracotomy due to hilar bleeding. Four post-operative deaths occurred, 3 from ARDS and 1 from a stroke (operative mortality = 3.6%). Major complications included pneumonia (5), stroke (1), and bronchopleural fistula requiring return to the operating room for closure (1). Minor complications included prolonged air leak (6), new onset atrial fibrillation (4), blood transfusion (2), and ileus (1). Median length of hospital stay was 3 days.
The data of Daniels reflects that of our experience with this technique.8 In our series of 21 patients who underwent successful thoracoscopic lobectomy from September 2002 to April 2003 after flexible bronchoscopy and staging mediastinoscopy, one death occurred in the post-operative period secondary to mesenteric ischemia. Eighteen complications occurred in nine of the twenty patients. Two patients suffered major complications (mesenteric ischemia, colonic volvulus, empyema, and atrial fibrillation). Four patients had prolonged air leaks and two of these patients had clinical subcutaneous emphysema. Eighteen patients were discharged in good condition. Length of stay (LOS) analysis showed a median of 4 days and a mean of 7.3 ± 5.8 days. LOS was £ 4 days in 55% of study participants. Malignant tumors were found in 80% of cases with a tumor size was £ 3 cm. Non-malignant findings included: focal pulmonary vascular congestion, lymphoid infiltrate (2), and emphysema with focal fibrosis. The mortality rate of our study group was 5%, and some type of complication occurred in 45% of patients. In patients with cancer, 69% had stage 1a and 31% had stage 1b disease.
With the growth in popularity of video thoracoscopic lobectomy, Ferguson and Walker from the Royal Infirmary of Edinburgh examined approached to teaching the VATS technique to various trainees.9 A total of 276 consecutive VATS lobectomies were divided into cohorts of 46 patients with the various trainees. The authors found that training resulted in an increase of 22 minutes of operating time with no increase in blood loss, morbidity, mortality, or length of stay. The trainees' operating time was similar to the early cases performed by the consultant staff.
Sawada and colleagues reviewed their experience in patients with early stage NSCLC undergoing VATS versus open anatomic lobectomies to determine if a bias existed in selecting patients for a minimally invasive procedure.10 One hundred and sixty-five patients underwent VATS lobectomy and 123 patients underwent an open lobectomy. The authors found a 94.5% five-year survival in the VATS group and an 81.5% survival in the open group. Analysis of survival revealed that male sex, CEA > 5, and tumor size > 3 cm were negative prognostic factors in terms of survival. The authors concluded that the apparent improved patient survival after VATS lobectomy was most likely due to a patient selection bias to receive a less invasive procedure.
Other authors such as Congregado, et al. reviewed their long-term experience with VATS lobectomy.11 Between 1992 and 2008, 237 major pulmonary resections were performed using the VATS technique. The majority of procedures (204/237) were done for NSCLC. An overall conversion rate to open thoracotomy of 14% was observed with a post-operative stay of 4.2 days, a 30-day operative mortality of 3.7%, and an actuarial five year survival rate of 77.7%. The authors conclude that VATS is ideal for resection of small, node-negative, lung cancers.
Kawachi and colleagues reviewed their series of 249 consecutive anatomic lung resections between 1999 and 2003.12 Within the group, 73 patients underwent VATS lobectomy and 176 patients underwent open procedures. The mean operative time in the VATS group was 291 minutes and in the open group 215 minutes. The mean blood loss was equivalent between the two procedures and estimated at approximately 175 mL. Operative mortality was 1.4% in the VATS group and 2.3% in the open group. The most common morbidity in both groups was post-operative air leak. Incidence of pulmonary vessel injury was 8.2% in the VATS group and 1.7% in the thoracotomy group. The incidence of vascular injury in the VATS group was noted to decrease as experience with the technique increased.
Flores et al. from the Sloan Kettering Cancer Center in New York reviewed their series of clinical stage 1a lung cancer patients undergoing VATS lobectomy (n = 398) versus open lobectomy (n = 343) between May 2002 and August 2007.13 Survival in both groups was equivalent, with increasing age, large tumor size, and higher nodal stage being associated with worse survival. There were fewer complications in the VATS lobectomy group (p = NS), while advanced age (p < .001) and tumor size (p < .02) were associated with a higher risk of operative complications. VATS lobectomy patients had a 2-day shorter length of hospital stay than their open thoracotomy counterparts.
VATS lobectomy in properly selected patients has been shown to provide a safe and oncologically sound surgical alternative to formal thoracotomy. Early investigators (Daniels, Kirby, Walker, Demmy) have shown their success with the VATS alternative in a variety of subclasses of patients with peripheral, limited stage cancers. 2,3,6,7 Ferguson and Walker have reiterated the need for formal training in minimally invasive techniques to improve dexterity and ensure safe and quality patient care.9 A standard time for VATS lobectomy varies greatly between investigators, but most groups can demonstrate a two-day shorter hospital stay with a minimally invasive lobectomy. Most authors continue to avoid VATS lobectomies in patients with significant mediastinal and hilar nodal disease. VATS lobectomy in the hands of an experienced thoracoscopic surgeon should be endorsed as the treatment of choice for early stage pulmonary malignancies.
1. Yim, AP, Landreneau, RJ, Izzat, MB, Fung, AL, Wan, S. Is video-assisted thoracoscopic lobectomy a unified approach? Ann Thorac Surg. 1998;66:1155-8.
2. Walker, WS, Carnochan, FM, Pugh, GC. Thoracoscopic pulmonary lobectomy. Early operative experience and preliminary clinical results. J Thorac Cardiovasc Surg. 1993;106:1111-7.
3. Kirby, TJ, Mack, MJ, Landreneau, RJ, Rice, TW. Initial experience with video-assisted thoracoscopic lobectomy. Ann Thorac Surg. 1993;56:1248-52.
4. Weber, A, Satmmberger, U, Inci, I, Schmid, RA, Dutly, A, Weder, W. Thoracoscopic lobectomy for benign disease–a single center study on 64 cases. Eur J Cardiothorac Surg. 2001;20(3):443-8.
5. Nomori, H, Horio, H, Naruke, T, Suemasu, K. What is the advantage of a thoracoscopic lobectomy over a limited thoracotomy procedure for lung cancer surgery? Ann Thorac Surg. 2001;72:879-84.
6. Demmy, TL, Curtis, JJ. Minimally invasive lobectomy directed toward frail and high-risk patients: A case-control study. Ann Thorac Surg. 1999;68:194-200.
7. Daniels, LJ, Balderson, SS, Onaitis, MW, D’Amico, TA. Thoracoscopic lobectomy: a safe and effective strategy for patients with stage I lung cancer. Ann Thorac Surg. 2002;74:860-4.
8. Podbielski, FJ, Connolly, AE, McEnaney, PM, McNamee, CJ, Conlan, AA. Video-assisted thoracoscopic lobectomy: a single institution study. Chest. 2003;124(4):234S.
9. Ferguson, J, Walker, W. Developing a VATS lobectomy programme -- can VATS lobectomy be taught? Eur J Cardiothorac Surg. 2006;29(5):806-9.
10. Sawada, S, Komori, E, Yamashita, M, Nakata, M, Nishimura, R, Teramoto, N, Segawa, Y, Shinkai, T. Comparison in prognosis after VATS lobectomy and open lobectomy for Stage 1 lung cancer: retrospective analysis focused on a histological subgroup. Surg Endosc. 2007;21(9):1607-11.
11. Congregado, M, Merchan, RJ, Gallardo, G, Ayarra, J, Loscertales, J. Surg Endosc. 2008;22(8):1852-7.
12. Kawachi, R, Tsukada, H, Nakazato, Y, Takei, H, Koshi-ishi, Y, Goya, T. Morbidity in video-assisted thoracoscopic lobectomy for clinical stage 1 non-small cell lung cancer: is VATS lobectomy really safe? Thorac Cardiovasc Surg. 2009;57(3):156-9.
13. Flores, RM, Park, BJ, Dycoco, J, Aronova, A, Hirth, Y, Rizk, NP, Bains, M, Downey, RJ, Rusch, VW. Lobectomy by video-assisted thoracic surgery (VATS) versus thoracotomy for lung cancer. J Thorac Cardiovasc Surg. 2009;138(1):11-18.