Early chest tube removal after video-assisted thoracic surgery lobectomy with serous fluid production up to 500 ml/day

Transcription

1 European Journal of Cardio-Thoracic Surgery 45 (2014) doi: /ejcts/ezt376 Advance Access publication 19 July 2013 ORIGINAL ARTICLE Early chest tube removal after video-assisted thoracic surgery lobectomy with serous fluid production up to 500 ml/day Lars S. Bjerregaard a, *, Katrine Jensen b, Rene Horsleben Petersen b and Henrik Jessen Hansen b a b Department of Cardiothoracic Anaesthesia and Section for Surgical pathophysiology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark Department of Cardiothoracic Surgery, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark * Corresponding author. Department of Cardiothoracic Anaesthesia and Section for Surgical pathophysiology, Rigshospitalet, section 7621, Blegdamsvej 9, DK-2100 Copenhagen. Tel: ; fax: ; lars.stryhn.bjerregaard@regionh.dk (L.S. Bjerregaard). Received 13 February 2013; received in revised form 26 March 2013; accepted 31 May 2013 Abstract OBJECTIVES: In fast-track pulmonary resections, we removed chest tubes after video-assisted thoracic surgery (VATS) lobectomy with serous fluid production up to 500 ml/day. Subsequently, we evaluated the frequency of recurrent pleural effusions requiring reintervention. METHODS: Data from 622 consecutive patients undergoing VATS lobectomy from January 2009 to December 2011 were registered prospectively in an institutional database. Data included age, gender, lobe(s) resected, bleeding and duration of surgery. Follow-up was 30 days from discharge. All complications requiring pleurocentesis or reinsertion of a chest tube, and all readmissions were registered. Twenty-three patients were excluded due to missing data, in-hospital mortality and loss to follow-up, leaving 599 for final analysis. Our primary outcome was the number of patients requiring reintervention due to recurrent pleural effusion. Secondary outcomes included time of chest tube removal and time to discharge. The incidence of recurrent pleural effusions requiring reintervention was compared between three groups according to the postoperative day (POD) of chest tube removal (Day 0 1, 2 3 and 4, respectively) using Fisher s exact test. RESULTS: Pleural effusion after chest tube removal required reintervention in 17 patients (2.8%). Of these, 7 needed readmission. Median time from surgery to chest tube removal was 2 days, and median time from surgery to discharge was 4 days. No statistically significant association was found between the incidence of reinterventions due to recurrent pleural effusion and the POD of chest tube removal (P = 0.50). The median time from chest tube removal to discharge was 1 day in all groups. Of the patients who needed reintervention, none had complications regarding this, except one who developed pneumothorax after pleurocentesis. CONCLUSIONS: Our findings suggest that chest tube removal after VATS lobectomy is safe despite volumes of serous fluid production up to 500 ml/day. The proportion of patients who developed pleural effusion necessitating reintervention was low (2.8%), and a complication of the reintervention was seen in only 1 patient. Keywords: Video-assisted thoracic surgery Lobectomy Chest tube Fast-track surgery Pleural effusion THORACIC INTRODUCTION In fast-track surgery, the idea is to enhance postoperative recovery by optimizing all factors involved in the treatment of surgical patients. The concept was first described in relation to colorectal surgery in the 1990s, and is still an area of on-going improvement [1]. During the last two decades, the fast-track surgery approach has been applied to different specialties including thoracic surgery [2 4]. An important step in fast-tracking pulmonary resections is the early removal of chest tubes, which cause pain, impair mobilization and impose a risk of infection. A consensus on when to remove the chest tube after pulmonary resections on the basis of the amount of pleural drainage has Presented at the 21st European Conference on General Thoracic Surgery, Birmingham, UK, May not yet been reached [5], and recently proposed standardized definitions and nomenclature on chest tube related topics, require studies of more standardized patients with clinically relevant outcomes [6]. In fast-track pulmonary resections, we have removed chest tubes after video-assisted thoracic surgery (VATS) lobectomy despite volumes of serous pleural drainage up to 500 ml/day. In this study, we evaluated the frequency of recurrent pleural effusions requiring reintervention as a result of this strategy. Although we recognize the relevancy of air leak in relation to early chest tube removal, this was not the objective of the present study and therefore the discussion focusses on early chest tube removal, recurrent pleural effusion and the consequences of this complication. The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 242 L.S. Bjerregaard et al. / European Journal of Cardio-Thoracic Surgery MATERIALS AND METHODS Patient inclusion and approval According to Danish law, this was considered a quality insurance process by The Regional Ethics Committee, due to its observational, non-interventional design. The study was approved by the Danish Data Protection Agency (Capital region, Denmark). Data were registered on 622 consecutive patients undergoing elective VATS lobectomy between January 2009 and December Nine patients were excluded due to missing data and in-hospital mortality, and 14 patients were lost to follow-up (Fig. 1). Patients characteristics, operative data and follow-up Data on age, gender, type of resection (uni- or bilobectomy, ± wedge resection), intraoperative bleeding (volume in floor suction), duration of surgery, postoperative day (POD) of chest tube removal, POD of discharge and complications during the admission were collected prospectively in an institutional database. Follow-up was 30 days from discharge, where records on all discharged patients were retrieved from E-journal (A Danish national system recording all admissions and reasons for admission). All complications requiring pleurocentesis or reinsertion of a chest tube were registered, as were all readmissions and their causes. If a patient were transferred to another hospital, the day of transmission were registered as POD of discharge, but follow-up was 30 days from final discharge. Outcomes Our primary outcome was the number of patients developing recurrent pleural effusion requiring pleurocentesis or reinsertion of a chest tube. Secondary outcomes were POD of chest tube removal and its association with the incidence of recurrent pleural effusions, POD of discharge, time from chest tube removal to discharge and the cumulative time of readmission due to recurrent pleural effusion. Surgery All operations were performed by one of two highly experienced surgeons (H.J.H. and R.H.P.) via a standardized anterior approach, with three ports identically placed independently of the lobe to be resected. The patient was placed in the lateral position, and the intercostal spaces were opened by arching the operating table. The procedure was performed in a bimanual fashion, the surgeon using two instruments through two of the insertions, while the camera Figure 1: Flow chart of study profile.

3 L.S. Bjerregaard et al. / European Journal of Cardio-Thoracic Surgery 243 was controlled by the assistant through the third insertion placed anterior and in the middle. Vessels, fissures and bronchus were divided one-by-one with endostablers (Endo GIA, Covidien, CT, USA) or in few cases with an electrothermal bipolar tissue-sealing system (LigaSure,Covidien,CT,USA). A no-touch the fissure technique was used, meaning no dissection or use of electrocautery in the fissure, but instead performing a stabling of the fissure with the visceral pleura intact. All specimens were removed in an endobag and a systematic lymph node dissection was performed if malignancy were suspected. The surgical technique used in our centre has previously been described in more detail by Hansen et al.[7]. Chest tube management A size 28 French, Portex chest tube (Smiths Medical, Ashford, UK) was placed at the end of operation. A few patients had two tubes placed if significant postoperative air leak was suspected, or in the case of bilobectomy. Chest tubes were removed when output (not blood, not chylos) was <500 ml in 24 h, and no air leakage was present. From January 2009 until April 2010, air leakage was objectively evaluated as no bubbles were seen in the water seal of the Redax Drentech chest drainage system (Redax, Mirandola, Italy) when the patient coughed. From April 2010 to December 2011, the Thopaz thoracic drainage system (Medela, Baar, Switzerland) was used and the acceptable air leakage for chest tube removal was defined as <50 ml/min in 12 h or <20 ml/min in 8 h. If chylous output was suspected (milky appearance), levels of triglycerides were compared between bloods and chest tube effusion, and if the concentration of triglycerides in the effusion were the same or higher than in blood, the patient had only intravenous alimentation, and the chest tubes were kept in place. The chest tubes were removed by nurses trained in the procedure, and a control chest X-ray was taken 2 4 h after removal. No patient was discharged or transferred with a chest tube. Data analysis Continuous data are presented as means with standard deviations and compared between groups using one way analysis of variance, if approximately normally distributed. If not approximately normally distributed, these data are presented as medians with interquartile ranges [IQR], and compared between groups using the Kruskal-Wallis Test. Categorical data are presented as numbers and percentages, and compared using Fisher s exact test. IBM SPSS statistics version 20 (IBM, New York, USA) was used for data analysis, and P < 0.05 was considered significant. RESULTS Baseline characteristics and operative data are presented in Table 1 for the 613 patients included, and for the 599 patients available for follow-up. The latter have been divided into three groups, according to POD of chest tube removal, and variables are compared between groups. The 14 patients lost to follow-up were comparable on all variables. Data on selected complications are given in Table 2. A total of 17 patients (2.8%) required reintervention because of recurrent pleural effusion. Of these, 5 patients received a new chest tube while 12 had pleurocentesis. The median time from surgery to chest tube removal was 2 days (IQR 1 4) and median time from surgery to discharge was 4 days (IQR 3 6) (Fig. 2). Postoperative lengths of stays ranged from 1 to 39 days. Of note, 15.3% of the patients were discharged the same day as they had their chest tube removed, 51.1% were discharged the day after chest tube removal and 17.3% were discharged 2 days after chest tube removal. 17 patients (2.8%) were transferred to another hospital or another department before discharge. In the group of patients having their chest tubes removed on POD 0 1, 9 patients (4.0%) developed pleural effusion requiring reintervention. In each of the groups POD 2 3 and POD > 4 4 patients required reintervention (2.2 and 2.1%, respectively) (Table 3) (between groups, P = 0.50). The median time from removal of chest tube to discharge was 1 day in all groups (IQR s: 1 2, 1 2 and 0 1, respectively). Of the 16 patients requiring reintervention due to recurrent pleural effusion after discharge, only 7 were readmitted. The remaining 9 patients had pleurocentesis in the outpatient clinic after which they returned home. Two patients were readmitted because of pleural effusion, but did not need reintervention. When including these 2 patients, the cumulative time of THORACIC Table 1: Basic characteristics and operative data for all patients included (n = 613), and for all patients available for the follow-up (n = 599) divided in groups according to POD of chest tube removal All patients included, n = 613 POD 0 1, n = 227 POD 2 3, n = 182 POD 4, n = 190 Age (years) a1 65 (11.4) 64 (11.0) 64 (11.9) 65 (10.8) Gender b2 Male 279 (45.5) 107 (47) 72 (40) 94 (49) Female 334 (54.5) 120 (53) 110 (60) 96 (51) Type of resection b3 Lobectomy 569 (92.8) 205 (90.3) 167 (92.8) 183 (96.3) Lobectomy + wedge resection 36 (5.9) 19 (8.4) 11 (6.0) 6 (3.2) Bilobectomy 8 (1.3) 3 (1.3) 4 (2.2) 1 (0.5) Duration of surgery (min) a4 116 (37.8) 101 (29.5) 120 (35.8) 129 (42.2) Bleeding (ml) c5 100 (50 150) 50 (25 100) 100 (50 200) 100 (50 250) Comparison between groups of POD for chest tube removal: 1 P = 0.498, 2 P = 0.1, 3 P = 0.118, 4 P < , 5 P < POD: postoperative day. a Mean (standard deviation). b Number of patients (group percentage). c Median (interquartile range).

4 244 L.S. Bjerregaard et al. / European Journal of Cardio-Thoracic Surgery Table 2: Complications leading to readmission and/or reintervention (pleurocentesis or reinsertion of chest tube) during admission and during the 30-day follow-up Table 3: The number of patients developing pleural effusion requiring reintervention divided into comparable groups according to POD of chest tube removal Figure 2: POD of chest tube removal and POD of discharge. readmission because of pleural effusion was 40 days. Eight patients were excluded because they died before discharge. Of these, 5 died within 30 days from surgery, yielding a 30-day mortality of 0.8% and an overall in-hospital mortality of 1.3%. DISCUSSION During admission, n = 613 During the follow-up, n = 599 Pleural effusion Pleurocentesis 0 12 (2.0) Reinsertion of chest tube 1 (0.2) 4 (0.7) Readmitted but no 0 2 (0.3) intervention Pneumothorax and/or SCE Insertion of additional chest 18 (2.9) a tube(s) Reinsertion of chest tube 22 (3.6) 29 (4.8) No intervention 28 (4.6) b Pleural empyema Chest tube (±Streptokinase) and antibiotics 2 (0.3) 4 (0.7) All values are given as the number of patients (group percentage). SCE: subcutaneous emphysema. a Only relevant during admission (before removal of primary chest tube). b Data not obtainable. A chest tube may cause pain and impair mobility. Furthermore, it has been suggested that chest tube removal improves forced expiratory volume in 1 s [8], and early chest tube removal may favour recovery of vital capacity and physical performance after lobectomy [9]. All of this makes chest tube removal crucial to recovery after pulmonary resection. Also, early chest tube removal has potential economic benefits, since patients who have their chest tubes removed early have shorter in-hospital stay [10]. Many Chest tube removed Reinsertion of chest tube due to pleural effusion Pleurocentesis due to pleural effusion POD: postoperative day. Number of patients (%) POD 0 1, n = 227 POD 2 3, n = 182 POD 4, n = 190 Total, n = (0.4) 2 (1.1) 2 (1.1) 5 (0.8) 8 (3.5) 2 (1.1) 2 (1.1) 12 (2.0) thoracic surgeons use a threshold of ml/day in deciding when to remove chest tubes after pulmonary resections, but these thresholds are mainly based on tradition and gut feeling more than evidence [11], and therefore might result in unnecessarily prolonged postoperative hospitalization. In recent years, several studies have been published, challenging the threshold of pleural drainage at which it is safe to remove the chest tube after pulmonary resection; McKenna et al. [3] reported minimal complications to removal at 300 ml in a 24-h period after VATS resections, Cerfolio and Bryant [11] reported it safe to remove chest tubes despite drainage of 450 ml/day after resections by open chest surgery and Göttgens et al. [12] found that chest tubes could be safely removed in the first h after VATS (bi-)lobectomy despite volumes of output up to 400 ml/day. Nakanishi et al. removed chest tubes on the first POD after VATS lobectomy, despite the volume of output. They found no association between drainage volume before chest tube removal and the risk of complications [13]. However, less encouraging results of high-output chest tube removal have also been reported [14]. To our knowledge, no study to date has included an equally large number of patients who all had major pulmonary resections by a standardized VATS-technique and allowed chest tube removal despite a volume of output up to 500 ml/day. Despite the allowed high volume output, we found that only 2.8% of the patients developed pleural effusion necessitating reintervention, that most of these could be treated without readmission, and that reinterventions were without complications except in 1 patient, who developed pneumothorax and needed a chest tube after pleurocentesis. This suggests that removal of chest tubes after VATS lobectomy despite output volumes up to 500 ml/day inflicts minimal risk of recurrent pleural effusion. While reintervention is rarely needed, this can in most cases be performed with minimal interference of the patient s everyday life, since hospitalization is not needed. In the study of Göttgens et al. [12], no patients were reported to have developed pleural effusion after VATS lobectomy and chest tube removal. In their study, the outcomes did not include the number of patients developing pleural effusion; however, the authors did report on complications in general, which we would expect to include symptomatic effusions, particularly if these necessitated reintervention. In the study of Nakanishi et al. [13] 4 of 100 (4%) had pleurocentesis due to pleural effusion while

5 L.S. Bjerregaard et al. / European Journal of Cardio-Thoracic Surgery 245 McKenna et al. [3] stated that no patients underwent reinsertion of a chest tube for pneumothorax or pleural effusion. However, with regard to the discussion following the article, one could ask if some patients might have had pleural effusions after discharge, but had pleurocentesis without the knowledge of the authors. The study by Cerfolio and Bryant [11] reported symptomatic pleural effusion in 11 of 1988 patients (0.55%), after pulmonary resections performed by thoracotomy. It is a matter of discussion whether it makes any difference in relation to early chest tube removal after pulmonary resections, if the procedure is performed by VATS or thoracotomy [3, 15]. In a recent study, chest tubes were omitted after major pulmonary resections, and no patients were reported to have received reintervention because of recurrent pleural effusion, regardless of whether a chest tube was placed during surgery or not [16]. However, no follow-up was carried out after discharge. One could hypothesize that allowing high-volume output before chest tube removal would leave a considerable amount of fluid in the pleural space, creating ideal conditions for bacterial proliferation, but the low incidence of pleural empyema in this study does not seem to support this assumption (Table 2). In this study, more than half of the patients discharged (56.2%), had their chest tube removed within 2 days from surgery, and 77.1% had it removed within 4 days from surgery (Fig. 2). Most of the patients were discharged the same day, or the day after chest tube removal, suggesting that early chest tube removal facilitates early discharge after major pulmonary resections, without increasing the risk of reintervention due to pleural effusion. In the group of patients having their chest tubes removed on POD 0 1 the frequency of reinterventions due to pleural effusions was higher than in the groups of patients having their chest tubes removed later (Table 3). Although these results are statistically insignificant (P = 0.5), they could raise the question whether it is justified to remove chest tubes already on the first POD after major pulmonary resections. In this series, chest tubes were removed on the first POD in more than one-third of the patients, making it possible to discharge almost 25% of the patients within 2 days from surgery. In our centre, the current chest tube duration is 2 days and the median length of in-hospital stay is 4 days after VATS lobectomy. We believe that this is advantageous for the patients, and definitely justifies removing chest tubes as soon as possible. Table 1 shows significant disparities in mean duration of surgery and median intraoperative bleeding between groups; however, these differences do not seem to explain the different incidences of reinterventions, and the possible associations between operative data and the POD of chest tube removal, is considered beyond the scope of this article. The cumulative time of readmission due to recurrent pleural effusion was 40 days, which equals 1 day of readmission for every 15 patients operated, or in other words: even if we assume that the frequency of readmissions due to recurrent pleural effusion was solely a result of our aggressive chest tube removal, it would still be advantageous from an economical point of view, as long as just 1 of 15 patients is discharged 1 day earlier. However, the expenses of utensils, personnel, etc. used in treating the patients with recurrent pleural effusions who did not require readmission, should of course also be included the equation. This study has limitations. First, despite the prospective set-up of the institutional database, exclusion of a few patients (2.4%) was necessary because of missing data. Secondly, the relatively small number of patients developing recurrent pleural effusion overall might prevent us from finding a significant association between the incidence of pleural effusions and the POD of chest tube removal. Finally, it would have been desirable to report data on actual volumes of daily pleural outputs; however, these were not registered except for the >500 ml/day limit. Future studies on this subject should undoubtedly include the registration and reporting of actual volumes of pleural output. In conclusion, the results of this study suggest that chest tube removal after VATS lobectomy is safe despite volumes of serous fluid production up to 500 ml/day. The proportion of patients developing pleural effusion necessitating reintervention was low (2.8%), and most could be treated without readmission and without further complications. In our centre, this has led to current median chest tube duration of 2 days and a median length of in-hospital stay of 4 days after VATS lobectomy. Funding This work was supported by the Doctor Fritz Karner and wife Edith Karner s foundation and the Doctor Edgar Schnohr and wife Gilberte Schnohr s foundation. Conflicts of interest: Rene Horsleben Petersen and Henrik Jessen Hansen both declare that they are consultants for Covidien, and that they have received travel-grants from Takeda. Lars S. Bjerregaard and Katrine Jensen declare no conflict of interest. REFERENCES [1] Kehlet H, Slim K. The future of fast-track surgery. Br J Surg 2012;99: [2] Cerfolio RJ, Pickens A, Bass C, Katholi C. Fast-tracking pulmonary resections. J Thorac Cardiovasc Surg 2001;122: [3] McKenna RJ Jr, Mahtabifard A, Pickens A, Kusuanco D, Fuller CB. Fast-tracking after video-assisted thoracoscopic surgery lobectomy, segmentectomy, and pneumonectomy. Ann Thorac Surg 2007;84: [4] Padilla AJ, Penalver Cuesta JC. Experience with lung resection in a fast-track surgery program. Arch Bronconeumol 2013;49: [5] Cerfolio RJ, Bryant AS. The management of chest tubes after pulmonary resection. Thorac Surg Clin 2010;20: [6] Brunelli A, Beretta E, Cassivi SD, Cerfolio RJ, Detterbeck F, Kiefer T et al. Consensus definitions to promote an evidence-based approach to management of the pleural space. A collaborative proposal by ESTS, AATS, STS, and GTSC. Eur J Cardiothorac Surg 2011;40: [7] Hansen HJ, Petersen RH, Christensen M. Video-assisted thoracoscopic surgery (VATS) lobectomy using a standardized anterior approach. Surg Endosc 2011;25: [8] Refai M, Brunelli A, Salati M, Xiume F, Pompili C, Sabbatini A. The impact of chest tube removal on pain and pulmonary function after pulmonary resection. Eur J Cardiothorac Surg 2012;41: [9] Nomori H, Horio H, Suemasu K. Early removal of chest drainage tubes and oxygen support after a lobectomy for lung cancer facilitates earlier recovery of the 6-minute walking distance. Surg Today 2001;31: [10] Bardell T, Petsikas D. What keeps postpulmonary resection patients in hospital? Can Respir J 2003;10:86 9. [11] Cerfolio RJ, Bryant AS. Results of a prospective algorithm to remove chest tubes after pulmonary resection with high output. J Thorac Cardiovasc Surg 2008;135: [12] Göttgens KW, Siebenga J, Belgers EH, van Huijstee PJ, Bollen EC. Early removal of the chest tube after complete video-assisted thoracoscopic lobectomies. Eur J Cardiothorac Surg 2011;39: [13] Nakanishi R, Fujino Y, Yamashita T, Oka S. A prospective study of the association between drainage volume within 24 hours after thoracoscopic lobectomy and postoperative morbidity. J Thorac Cardiovasc Surg 2009;137: [14] Grodzki T. Prospective algorithm to remove chest tubes after pulmonary resection with high output is it valid everywhere? J Thorac Cardiovasc Surg 2008;136: THORACIC

6 246 L.S. Bjerregaard et al. / European Journal of Cardio-Thoracic Surgery [15] Jiwnani S, Mehta M, Karimundackal G, Pramesh CS. Early removal of chest tubes after lung resection VATS the reason? Eur J Cardiothorac Surg 2012; 41:464. [16] Ueda K, Hayashi M, Tanaka T, Hamano K. Omitting chest tube drainage after thoracoscopic major lung resection. Eur J Cardiothorac Surg 2013;44:225 9; discussion 229. APPENDIX. CONFERENCE DISCUSSION Dr H. V. Kara (Istanbul, Turkey): Sorry if I missed the data. I have a question about your method. After you remove the chest tube, what is the next time point for undertaking a chest X-ray in the patient? You send the patient home and the day after you call them again, because of the criteria to reposition a chest tube. So when do you take the chest X-ray again? Dr Jensen: We take the chest tube out when the serous fluid production is under 500 ml, no air leakage and chylous, and then two hours later we take an X-ray to see if there is pneumothorax or fluid, and then we send the patient home. Dr Kara: So for the patients in whom you had to place a chest tube again, was the problem in that two hours chest X-ray or in the follow-up period? Dr Jensen: Sorry? Dr Kara: There were some patients who had to have a chest tube again, so was the problem in the chest X-ray in the two-hour period or in the early follow-up? Dr Jensen: It was in the follow-up group. When the patients were admitted, they were seen 14 days later in our outpatient clinic, and then we always do a chest X-ray. But the other patients (it was half of them) were admitted to a local hospital due to pneumonia, and they had a thoracocentesis there. But the amount of fluid they took out was 150 ml, and if we had seen the patient in the first place, we would not have done a thoracocentesis. So the numbers needing reintervention may seem higher than they actually are. Dr L. Molins (Barcelona, Spain): What have been the criteria to keep in mind about the amount of fluid to perform a thoracocentesis? I don t know if it was only considered on the amount of fluid or also on the results of the X-ray or even echo. Dr Jensen: We didn t have the exact amount of fluid production because we wanted to test a strategy and see how it went, not based on an exact amount. So this was a strategy we looked at, below 500. Dr H. Eid (Dubai, UAE): I just want to ask you, have you applied negative suction to the chest drain, because applying negative suction can increase the drainage? Personally I don t feel in favour of discharging a patient one day or two days earlier, then seeing them come back with a complication and being readmitted for one week when you are economizing or trying to study cost benefit for the patient. You had patients with pneumothorax referred back by peripheral clinics. I wonder if these patients went to a peripheral clinic and they placed a tube in them based on the criteria for reinsertion of the tube, or for drainage, and how long these patients with complications stayed in the hospital after that. Dr Jensen: So your first question was? Dr Eid: Negative suction. Dr Jensen: We applied -10 on all patients. Dr Eid: And you have no air leak in these patients, because there will be a percentage of air leak postoperatively after lobectomy, especially in such numbers of patients. So I want you to clarify in your study the percentage of air leak, application of negative suction, and the complicated patients, especially pneumothorax, that developed after early removal of the chest tube drain. Dr Jensen: I am not sure I understood your question. Dr H. Hansen (Copenhagen, Denmark): I might comment on this. This study is addressing the problem of pleural effusion afterwards. It is not addressing the problem with air leakage or the problem with pneumothorax. We had patients who had prolonged tube time due to pneumothorax or due to air leakage, but they are not addressed here. Dr Eid: So you excluded these patients with air leak? Dr Hansen: This is a study about whether tube removal has an impact concerning fluid production, and it shows that only 2.8% had any reintervention rate. That conclusion means that you don t have to follow-up concerning fluid production. You can look at the air leakage. That is important, not discussed here, but fluid production does not mean anything at all. We had very, very few insertions of thoracocentesis, only one tube afterwards. So you can aggressively remove the tube as long as there is no air leakage. And that is a major point in getting the fast-track patient out of the clinic, and also points to avoidance of having tubes in for a prolonged time, with the risk of empyema and other possibilities concerning tube pain and so on. Dr Jensen: We only had four patients with pleural empyema out of the 599 patients, and they were not in the group who had intervention. Dr S. Rathinam (Leicester, UK): Could you talk us through your postoperative drain management? You said you used digital suction and negative suction initially. Would you change them to gravity mode, get an X-ray? Just tell us what your protocol is before you remove the tubes. Dr Jensen: We remove it when there is no air leakage, that means below 20 ml/min in 12 h on the Topaz drainage system, and, of course, no chylous and fluid production under 500 ml/day. And if there is prolonged air leak, of course we will take a control X-ray. Dr G. Varela (Salamanca, Spain): When I investigated some points on postoperative management of pleural drainages, I was really surprised with the high rate of variation in clinical practice. In my institution, I don t remember the data, but probably the agreement between two different surgeons with comparable experience on the day to withdraw chest tubes was around 40%. In your study, there are several points needing some clarification. For example, did you agree on the discharge criteria, because when you are analysing an outcome such as time to discharge, you have to be sure that all surgeons have the same discharge criteria, otherwise the results are not valid. Dr Jensen: We all use the same criteria for discharge and removal of the chest tubes in our centre. Dr Varela: That s great. After the presentation, please give me some counselling on how you can do it, because it is really difficult in clinical practice. Second question: you are analysing how many patients required reinsertion of a chest tube due to pleural effusion, what were the criteria? Was it low po2, dyspnoea, a chest X-ray or what? Dr Jensen: We looked at a chest X-ray to see if there was pleural effusion or not, if they came in with dyspnoea, and if we could see a large amount we did an ultrasound to estimate how much there was, and then we did a thoracocentesis in the patient in whom we believed the lung could expand, but that is an individual decision. Dr Varela: So it was based mainly on radiological images, not on clinical criteria? Dr Jensen: Both. Dr Hansen: We are fast-tracking the patients, so they are discharged before we have the final pathological answer. So we saw all our patients except those 14 excluded because they went to Greenland or the Faroe Islands. All of our patients are seen in our outpatient clinic 10 to 14 days after surgery to have an X-ray control and have a final pathological answer and planning of adjuvant chemotherapy. So we actually did see all 599 patients in our outpatient clinic within two weeks after discharge. And this is one part of the data collected. The other part of the data collected is that we are looking to see whether any patients were admitted to any unit (not only people having a reintervention in our own unit), so we could count them also.