Eur J Cardiothorac Surg 2009;35:553. doi:10.1016/j.ejcts.2008.11.011
Copyright © 2009, European Association for Cardio-thoracic Surgery. Published by Elsevier. All rights reserved.
Reply to Chen et al.
Erdal Okur*,
Volkan Baysungur,
Cagatay Tezel,
Semih Halezeroglu
Sureyyapasa Chest Diseases and Thoracic Surgery Teaching Hospital, Maltepe, 34758 Istanbul, Turkey
Received 10 November 2008;
accepted 12 November 2008.
* Corresponding author. Tel.: +90 5327961600; fax: +90 216 4214265. (Email: erdalokur{at}hotmail.com).
Key Words: Chest tubes • Randomized controlled trials • Pulmonary surgical procedures
We thank Dr Chen and his colleagues for their comments on our recent manuscript [1]. The method of clinical intervention and the parameters studied may oblige any randomized controlled trial (RCT) study to have some limitations. The limitations of our study were acknowledged in the manuscript [2]. However, we would like to respond the authors’ comments point by point.
- 1. In our study, patients were allocated into two groups in turn. We applied one chest tube in a standard lobectomy patient and two chest tubes to ‘the next’. Ideally in a RCT, participants should be assigned to comparison groups on the basis of a chance (random) process characterized by unpredictability [3]. Three surgeons performed the lobectomy operations in our study. Surgeons did not know if one or two chest tube(s) would be applied in any patient until completion of the operation. We think that our method may not be defined as quasi-random allocation. As the authors explained, the aim of randomization is to make a uniform distribution of preoperative variables which was achieved at the end of our study. Although the authors say that there may be some parameters which were not balanced between the groups, they do not give an example for this point. Indeed, all the important preoperative factors were similar in both groups in the study. In addition, we do not agree that 100 standard lung lobectomy patients is a small number for such a comparison.
- 2. Because it could be noticed easily in which group the patient was included during recording of the postoperative parameters, we acknowledged in the manuscript that the study could not be designed as totally ‘blind’.
- 3. The sample size for a trial needs to be planned at the beginning as suggested by Altman and colleagues [3], but, during the course of the study, interim analyses can be used to continue recruiting or to stop it. As explained in our manuscript, interim analyses were performed after every 50 cases and study was stopped when the number had reached 100, and statistically significant results were raised. We think that there was no need to continue the study and to apply one more ‘unwarranted’ chest tube to the patients after these results.
- 4. There was no sign of complication in this study that patients with one chest tube had inadequate pleural fluid drainage. Residual pleural effusion, in our practice, is very rare if chest tubes have been placed properly. We did not compare the arterial blood gasses measurements since we do not perform routine blood gas analyses before or after regular lobectomy operation. One would expect better postoperative pulmonary functions in patients with a single tube than double tube but not the opposite as authors say. Median values are preferred for the numeric data that have very large scale. Indeed, the results showed relatively narrow distribution in our study and we used the mean values instead of the median values.
References
- Chen H-L, You Q-S, Wen X-D. Experimental design should be regulated for the study. Eur J Cardiothorac Surg 2009;35:552-553.[Free Full Text]
- Okur E, Baysungur V, Tezel C, Sevilgen G, Ergene G, Gokce M, Halezeroglu S. Comparison of the single or double chest tube applications after pulmonary lobectomies. Eur J Cardiothorac Surg 2009;35(1):32-36.[Abstract/Free Full Text]
- Altman DG, Schulz KF, Moher D, Egger M, Davidoff F, Elbourne D, Gøtzsche PC, Lang T, CONSORT GROUP (Consolidated Standards of Reporting Trials) The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Inter Med 2001;134:663-694.[Abstract/Free Full Text]
