Ann Thorac Surg 1996;61:204-205
© 1996 The Society of Thoracic Surgeons
Invited Commentary
Invited Commentary
Hani Shennib, MD
Division of Cardiothoracic Surgery Montreal General Hospital 1650 Cedar Ave Room L9-120 Montreal, Que H3G 1A4 Canada
See also page 202.
There is sweeping enthusiasm for the establishment of registries to collect various types of surgical information. The usefulness of any registry will depend entirely on the objectives it wishes to achieve and the type of data that it accumulates. Currently, most registries are voluntary and lack the mechanisms that render provision of data obligatory. As a result, a voluntary registry that receives information from a large number of medical centers risks the serious disadvantage of being incomplete. The implications of this are the following: (1) Many centers with results that are perceived not to be at par with what is generally accepted will not report their data. This will usually occur despite assurances that information will remain confidential (in today's world of rapid telecommunication, words go around quickly). (2) Reporting of good results will falsely raise the standard of expected outcome for various surgical procedures. This will make it even more difficult for centers with average results to report their current experience. (3) The release of registry reports with a higher standard of outcome would also have medicolegal implications in that patients, hospitals, and other parties will consider this to be the norm, putting more pressure on centers with new initiatives and new surgical techniques. (4) Finally, a falsely elevated standard of practice often discourages centers with more realistic, lower outcomes from submitting their experience for publication. The result is that an overzealous effort to collect scientific data through registries may do a disservice to academic surgery and clinical practice.
This video-assisted thoracic surgery (VATS) registry report must be interpreted with caution. Beyond the description of demographic details, I find it extremely difficult to make conclusions based on any of its data. Discrepancies between one set of information and another are most likely due to incomplete compilation of data. For example, the authors of this article state that resection of pulmonary nodules was performed on 388 patients (171 benign and 217 malignant). As to the type of resection, wedge resection was performed in 300 cases and lobectomy in 21 cases. Whatever happened to the other 57 cases? Furthermore, the complication rate reported (8%) is unusually low. This could not be attributed to the technical skill of the surgeons, as there was a wide variation in surgical experience from one group to another, with a difference in the conversion rate to thoracotomy among centers that was proportional to the number of procedures performed in each. To make things more complicated, Bernard and associates report their experience mostly in percentages. For example, the rate of malignancy in nodules of 1 cm or less was 3.5%, whereas the rate of benignancy was 24%. What about the other 62.5%? Was it not reported? If so, does it not have a serious impact on the interpretation of data? Does this report imply that smaller lesions have a higher chance of being malignant? Bernard and associates report that 171 nodules resected were benign and that approximately two thirds of the VATS procedures performed were for diagnostic purposes (226 cases). In my opinion, this is a prohibitively costly way of managing benign lung nodules. One wonders whether this reflects the early enthusiasm of surgeons to apply the new technology of VATS and whether this trend will decline with the use of sound clinical judgment and the application of basic diagnostic tests such as repeat interval chest radiographs, transthoracic needle aspiration, and transbronchial biopsy. Overall, the data presented are no different from those previously reported by the Video-Assisted Thoracic Surgery Study Group in 1993. They also have the same limitations.
The effort by the French surgical centers to put their data together in one registry should be commended and encouraged. What needs to be done, however, is to tidy up the registry and upgrade its data-collection mechanisms so that a better quality of data can be accumulated, which would then be worth reporting. Finally, I would suggest that a registry without teeth to implement near-complete data collection may better be served by restricting its membership to a smaller number of seriously committed centers.
Related Article
-
Resection of Pulmonary Nodules Using Video-Assisted Thoracic Surgery
- Alain Bernard
Ann. Thorac. Surg. 1996 61: 202-204.
[Abstract]
[Full Text]
