Ann Thorac Surg 2001;71:961
© 2001 The Society of Thoracic Surgeons
Invited commentary
Robert J. Cerfolio, MDa
a University of Alabama at Birmingham, 712 Tinsley Harrison Tower, 1900 University Blvd, Birmingham, AL 35294, USA
e-mail: robert.cerfolio{at}ccc.uab.edu
This intriguing report is essentially two articles in one. One is a retrospective review of 1,453 patients over 14 years, the other is a prospective trial over an 8-year period of 70 patients who underwent segmentectomy instead of lobectomy for lesions 2.0 cm or less. Because of this duality the report can be confusing at times. The basic idea Okada and colleagues support is that segmentectomy for 2.0-cm lesions offers the same survival as lobectomy and preserves more lung.
I have several concerns and questions. It is disturbing that 6% of the retrospectively reviewed specimens contained a nonradiologic imaged (and presumably a nonpalpable) T4 satellite nodule. If this incidence is accurate (in the prospective trial it was only 1 in 70, or 1.4%) then a significant number of patients who undergo segmentectomy may have satellite nodules (T4 lesions) left in the remainder of the unresected lobe.
Okada and colleagues suggest that 12 patients in the prospective trial "had to receive" other procedures because of nodal involvement. This raises a crucial concept. The authors imply that if sentinel lymph nodes outside the intended resected segment are positive, segmentectomy should not be performed because the lung with the involved positive lymph node sould be removed. I agree. But do the authors recommend pneumonectomy for a patient with a 1.5-cm peripheral lesion in the right upper lobe just because a subcarinal (#7) lymph node is involved with cancer? The answer seems an obvious no. If their answer is no, why then do the authors state that "when nodal involvement [is detected] ... a limited resection was not indicated"? Patients with N2 disease usually die from systemic disease. Why then recommend an operation that has higher risk but only provides better local control?
Finally, and most importantly, if a pulmonary resection smaller than lobectomy (which has minimal morbidity and mortality) is going to be recommended, then that smaller resection must afford a distinct advantage without compromising survival and local control. I do not believe, even with this excellent report, that the latter has been shown. There is little difference in postoperative pulmonary function in patients who undergo segmentectomy versus lobectomy, especially with certain types of segmentectomies. Segmentectomy should be reserved for patients with compromised pulmonary function. Some segmentectomies can be technically challenging, and because most lung operations worldwide are not performed by general thoracic surgeons, I am concerned others will not be able to duplicate Okada’s excellent results. Although the authors unfortunately did not list their "noncritical" complications, one might imagine that the incidence of air leaks or atelectasis may be higher. This article should spur further investigation.
