Eur J Cardiothorac Surg 2007;31:372-375. doi:10.1016/j.ejcts.2006.12.027
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Comparison of three different chest drainages after coronary artery bypass surgery — a randomised trial in 150 patients
Staffan Bjessmoa,*,
Susanne Hylandera,
Jenny Vedina,
Dag Mohlkertb,
Torbjörn Iverta
a Department of Cardiothoracic Surgery and Anaesthesiology, Karolinska University Hospital, 171 76 Stockholm, Sweden
b Department of Thoracic Radiology, Karolinska University Hospital, Stockholm, Sweden
Received 28 November 2006;
accepted 15 December 2006.
* Corresponding author. Tel.: +46 851770822; fax: +46 8322701. (Email: staffan.bjessmo{at}karolinska.se).
 |
Abstract
|
|---|
Objective: Drainage of blood from the mediastinum and pleura following open cardiac procedures is usually carried out using one or more large-bore plastic chest tubes. Recently small diameter siliastic drains have been reported to evacuate blood with a better patient comfort. The efficacy and safety of different chest tubes have not yet been fully evaluated. Methods: One hundred fifty patients undergoing coronary artery bypass surgery were randomised to have either Blake® 24F (Ethicon, Inc, Somerville, NJ), Argyle® 32F plastic (Tyco Healthcare, Tullamore, UK) or Jostra® 32F silastic (Maquet Cardiopulmonary AG, Hirrlingen, Germany) drains inserted for evacuation of postoperative bleeding. Bleeding rate per hour, total blood loss, patient discomfort during drain removal, residual pleural fluid at chest X-ray 3 days and 3 weeks after the operation were recorded. Results: Bleeding pattern and total bleeding did not differ significantly in the three groups. Median blood loss was 615 ml (quartile range 390–820 ml) in the Blake-group, 750 ml (quartile range 430–870 ml) in the Jostra-group and 580 ml (quartile range 450–750 ml) in the Argyle-group, respectively (p
= 0.17). Pain at removal the day after the operation was similar in the three groups. Residual fluid in the left pleura did not differ significantly at 3 days (p
= 0.41) or at 3 weeks postoperatively (p
= 0.42). Conclusions: None of the three chest tubes was superior to drain postoperative bleeding or considering pain at removal. Local clinical routines and cost aspects should be the guide in choosing drainage system for open cardiac operations.
Key Words: Coronary bypass surgery • Chest tube • Bleeding • Randomised
|
1. Introduction
|
|---|
Drainage of blood from mediastinum and pleural space following open cardiac surgery is carried out using one or more large-bore plastic chest tubes with large diameter to establish adequate evacuation of blood to prevent early tamponade. There are concerns that large stiff tubes because of pain might cause hypoventilation, atelectasis, and increased use of sedatives and analgesic agents. The drains are usually removed on the first postoperative day when bleeding has ceased and haemodynamic and respiratory functions are stable. There are reports supporting that drains with small diameter are as effective as larger drains after open heart operations and cause less discomfort on removal and allow better patient mobility [1–3]. The results are however conflicting and only two of the studies were randomised [3–5].
This prospective randomised trial was designed to compare three different commercially available chest drains after coronary artery bypass grafting (CABG). The amount of blood drained, residual pleural fluid and pain at drain removal were evaluated.
|
2. Material and methods
|
|---|
One hundred fifty patients undergoing a first time CABG were randomised to use of either Blake® 24F silastic drain (Ethicon, Inc, Somerville, NJ), Argyle® 32F plastic drain (Tyco Healthcare, Tullamore, UK) or Jostra® 32F silastic drain (Maquet Cardiopulmonary AG, Hirrlingen, Germany). The Blake Silastic drain has four grooves instead of holes running along the intra-thoracic length of the tube (Fig. 1
).
An inclusion criterion was opening of the left pleura, usually occurring during dissection of the internal mammary artery graft. Randomisation was done with closed envelope method before closing the chest. Fifty patients were randomised in each group after approval from the hospital's Ethics Committee and obtaining informed consent. There were no differences of baseline characteristics in the patient groups (Table 1
).
At the completion of the operation one drain was placed behind sternum and one in the left pleura through separate stab incisions near the inferior aspect of the sternotomy. The drains were connected to a Pleura-Vac® suction device (Genzyme Corporation; Cambridge, MA) at 15 cm H2O negative pressure. Bleeding was continuously recorded from the time of sternal closure in the operating room until the drains were removed the following day. All patients were put in upright position early the first postoperative day before drain removal in order to evacuate any residual fluid in the pericardial or pleural cavity. Pain at drain removal was estimated using a visual analogous scale (VAS) where 0 was regarded as no discomfort at all and 10 a very significant grade of pain [6]. A bleeding rate exceeding 300 ml/h for two consecutive ours or clinical signs on cardiac tamponade such as hypotension, increased pulmonary pressure and low urinary output resulted in acute re-exploration.
Chest X-ray was performed to evaluate any residual fluid in the pleural cavity on the third postoperative day and at 3 weeks postoperatively. One radiologist (D.M.) examined all chest X-rays without knowledge of the type of chest tube that had been used. The patients were examined in upright frontal position, with a side-projection and lying on the left side in order to approximate level of residual fluid. Any sign of free pleural fluid on chest X-ray of less than 1 cm in lying position was classified as minimal. One to three centimetre of pleural fluid was classified as moderate and more than three cm as significant amount.
2.1 Statistical methods
Sample size calculation was performed before the study applying total amount of postoperative bleeding as a primary endpoint. A power of 0.8 in the primary endpoint, alpha-level of 0.05 was considered clinical significant. Based on these assumptions, 126 patients were to be included in order to confirm any differences in the primary endpoint. Thus, inclusion of 150 patients seemed a reasonable sample size. Secondary endpoints were patient pain at drain removal and residual fluid on chest X-ray. In case of re-operation for bleeding or tamponade total amount of blood loss until removal of the chest tubes was recorded. Because of the skew distribution of postoperative bleeding, median values were used when differences between groups were analysed. Patient characteristics in the three groups were compared using non-parametric statistical methods. Kruskall–Wallis’ one-way analysis of variance was applied to analyse differences in postoperative bleeding and Chi-square was used to analyse proportions for categorical variables. A p value of less than 0. 05 was considered statistical significant.
|
3. Results
|
|---|
Bleeding from the chest tubes did not differ significantly in the patient groups. The median blood loss was 615 ml (quartile range 390–820 ml) in the Blake-group, 750 ml (quartile range 430–870 ml) in the Jostra-group and 580 ml (quartile range 450–750 ml) in the Argyle-group, respectively (p
= 0.17). Bleeding rate varied as expected with the highest bleeding rate during the first postoperative hours followed by a similar pattern of diminishing bleeding per hour with in all drain types (Fig. 2
). Use of antithrombotic and haemostatic drugs was similar in the three groups (Table 2
). Three patients (6%) in the Blake-group, two (4%) in the Jostra-group and none in the Argyle-group were re-explored for bleeding during the first 24 h postoperatively (p
= 0.21).
Need for postoperative erythrocyte, plasma or platelet transfusions did not differ significantly, nor did the recorded drop in haemoglobin concentration during the first three postoperative days (Table 3
). One patient in the Blake-group, one in the Jostra-group and two patients in the Argyle-group were re-explored due to deep sternal wound infection (p
= 0.77).
Estimated VAS pain at chest tube removal the day after the operation did not differ in the patient groups. One experience was that the 32F silicon tube (Jostra®) was somewhat more difficult to remove because of larger holes with at tendency to get stuck in the stab incision during removal.
The amount of free residual fluid in the left pleural cavity was examined at chest X-ray after 3 days and 3 weeks postoperatively. There were no statistical differences on the third postoperative day (p
= 0.41) or at 3 weeks (p
= 0.42) regarding both moderate and significant amounts of free pleural fluid.
|
4. Comment
|
|---|
Commercially available effective chest tubes are compelling for successful outcome after open heart operations. The small diameter and flexible design of the BlakeTM drain is promoted to be less painful at removal. There is however no clear randomised data supporting these claims. In our experience, the BlakeTM chest tube was pliable and worked excellent except in case of excessive bleeding.
All chest tubes were removed the first day after the operation causing similar pain regardless of type of drain used. The nursing routine in all patients was to give 3–5 mg of intravenous morphine as analgesia before the tubes were pulled.
Concerns, not only of patient discomfort, but also more severe complications as compression and unsettling of bypass grafts have been reported [7–8]. Care was taken not to position holes of the chest tubes towards grafts so that the tubes could not interfere with position of grafts when removed.
Our method of subxiphoidal insertion of the drainages caused minimal postoperative reduction in pulmonary function and less pain compared to insertion at the intercostal position [1,9]. Hypothetically, the more flexible silastic drainages (Blake® and Jostra®) should be easier to place in a preferred position, deep posterior in the opened left pleura to accomplish more complete drainage in a lying patient. However, there were no differences in total postoperative blood loss or residual fluid on chest X-ray in this study supporting this. In fact the flexible Blake® drain was sometimes difficult to direct deep into the posterior pleura. We experienced none of the complications previously described using larger chest tubes and the patients did not report any significant difference in discomfort during removal. Use of a more flexible, small calibre drain could be preferred when maintaining the chest tube for several days can be anticipated, but this is usually not necessary after routine open heart operations [10].
A limitation of the study is the possibility of a Type II error due to small patient groups, thus leading to false negative results. However, in our sample size analysis, 150 patients were to be sufficient to confirm differences in the primary endpoint. Further limitations were the lack of ‘hard’ endpoints such as tamponade or re-exploration for bleeding. Well aware of this, the study is based on the strong co-variation between high postoperative bleeding volumes and risk of tamponade/re-exploration. A randomised study based on tamponade and/or re-exploration would need very large sample sizes, to a high study cost, in order to avoid Type II errors.
In summary, the three different drainages functioned equally well after open coronary bypass operations. No chest tube was significantly superior to the others regarding efficacy to drain initial bleeding, completeness of fluid evacuation or patient discomfort when removed. Local clinical routines and cost aspects should be the main guide in choosing drainage system for open cardiac operations.
|
References
|
|---|
- Frankel TL, Hill PC, Stamou SC, Lowery RC, Pfister AJ, Jain A, Corso PJ. Silastic drains vs. conventional chest tubes after coronary artery bypass. Chest 2003;124:108-113.[Abstract/Free Full Text]
- Obney JA, Barnes MJ, Lisagor PG, Cohen DJ. A method for mediastinal drainage after cardiac procedures using small silastic drains. Ann Thorac Surg 2000;70:1109-1110.[Abstract/Free Full Text]
- Kolsi K, Frikha I, Kolsi M, Khannous M, Masmoudi S, Salah Kechaou M, Sahnoun Y, Karoui A. The effects of drainage with a Redon versus a conventional drain on postoperative pain and blood loss after valve replacements. J Cardiovasc Surg 2004;45:565-568.[Medline]
- Roberts N, Boehm M, Bates M, Braidley PC, Cooper GJ, Spyt TJ. Two-center prospective randomized controlled trial of Blake versus Portex drains after cardiac surgery. J Thorac Cardiovasc Surg 2006;132:1042-1046.[Abstract/Free Full Text]
- Ikishura H, Kimura S. The use of flexible silastic drains after chest surgery: novel thoracic drainage. Ann Thorac Surg 2006;81:331-333.[Abstract/Free Full Text]
- Dowling R, Thielmeier K, Ghaly A, Barber D, Boice T, Dine A. Improved pain control after cardiac surgery: results of a randomized, double-blind, clinical trial. J Thorac Cardiovasc Surg 2003;126:1271-1278.[Abstract/Free Full Text]
- Svedjeholm R, Håkansson E. Postoperative myocardial ischemia caused by chest tube compression of vein graft. Ann Thorac Surg 1997;64:1806-1808.[Abstract/Free Full Text]
- Kollef MH, Dothager DW. Reversible cardiogenic shock due to chest tube compression of the right ventricle. Chest 1991;99:976-980.[Abstract/Free Full Text]
- Hagl C, Harringer W, Gohrbandt B, Haverich A. Site of pleural drain insertion and early postoperative pulmonary function following coronary artery bypass grafting with internal mammary artery. Chest 1999;115:757-761.[Abstract/Free Full Text]
- Smulders YM, Wiepking ME, Moulijn AC, Koolen JJ, van Wezel HB, Visser CA. How soon should drainage tubes be removed after cardiac operations?. Ann Thorac Surg 1989;48:540-543.[Abstract]
Top
Abstract
1. Introduction
