Dr. Krasna provides a thoughtful review of thoracoscopy as an emerging technique for the staging of patients with lung and esophageal cancers. In lung cancer, thoracoscopy can be used as a complement to cervical mediastinoscopy in the evaluation of mediastinal and hilar lymph nodes. This is especially true in patients who have left-sided neoplasms with enlarged lymph nodes in the aortico-pulmonary window--a region typically inaccessible to cervical mediastinoscopy.
Perhaps the most useful application of thoracoscopy is in the preoperative assessment of patients with pulmonary lesions abutting the mediastinum. Of nine patients with lung cancer in Dr. Krasna's review who underwent thoracoscopic staging, three were found to have mediastinal invasion and thus avoided an unnecessary thoracotomy. Similarly, thoracoscopy enables evaluation of the pleural space, such that the presence of pleural implants or invasion can be assessed preoperatively. As Dr. Krasna points out, however, routine thoracoscopic evaluation of the pleural cavity prior to all thoracotomies for lung cancer is not indicated currently, and this technique should be limited to a study setting at this time.
Potential Benefits in Staging of Esophageal Cancer
The potential benefits of thoracoscopy in patients with esophageal cancer include improved staging so that an adequate determination of prognosis with present therapies can be made, as well as selection of the appropriate treatment modality for each disease stage. Lymph node involvement occurs in the majority of patients with esophageal cancer, including, in most studies, those undergoing curative surgical resection. Although the accuracy of endoscopic ultrasound for T-staging is 85%, its accuracy for lymph node involvement is only 75%. By contrast, the results reported by Krasna suggest that 93% accuracy for lymph node involvement may be possible with thoracoscopic evaluation.
Information about lymph node status is essential, as the number of lymph nodes involved is predictive of prognosis in patients managed with esophagectomy. Abe et al  found no survivors if two or more nodes were involved.
Skinner et al  and Sieward  observed no survivors if five or more nodes were involved, whereas Ellis et al  found that patients with one to four positive nodes had a survival intermediate to that of patients with no involved nodes or more than four positive nodes.
Information regarding the extent of the esophageal tumor at thoracoscopy may also be important for patients initially managed with a nonsurgical approach, such as chemoradiation alone, chemoradiation followed by surgical resection, or chemotherapy followed by surgical resection. At present, patients initially managed with a nonsurgical approach must be staged using the 1983 American Joint Committee on Cancer (AJCC) clinical staging system, or one must rely on the accuracy of ultrasound to appropriately stage patients with the 1988 AJCC criteria. As the volume of tissue irradiated correlates with morbidity, accurate information regarding lymph node involvement provided by video-assisted thoracoscopy may permit the use of tailored radiation fields that conform more closely to the actual regions of involvement, potentially resulting in a reduction in morbidity. Nearly two-thirds of the patients treated with chemoradiation in the landmark study by Herscovic et al  developed grade 3 or 4 acute toxicity, including esophagitis and hematologic toxicity, which could be attributed, at least in part, to the use of relatively generous radiation fields, typically encompassing the nodes from the supraclavicular area to the gastroesophageal junction.