| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2006;132:1042-1046
© 2006 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease |
a Department of Cardiac Surgery, Northern General Hospital, Sheffield, United Kingdom
b Department of Cardiac Surgery, Glenfield Hospital, Leicester, United Kingdom.
Received for publication February 23, 2006; revisions received June 7, 2006; accepted for publication June 15, 2006. * Address for reprints: Mr Neil Roberts, Research Registrar Cardiac Surgery, St Georges Hospital, Blackshaw Rd, London SW17 0QT, United Kingdom. (Email: neil.roberts{at}stgeorges.nhs.uk).
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
METHODS: A prospective randomized controlled trial was conducted at two university hospitals. A total of 199 patients undergoing first-time cardiac procedures were randomized at surgery to receive either small-bore flexible Blake drains (19F) or larger Portex drains (28F). Drains were removed according to the study protocol and patients underwent echocardiography 3 to 5 days after drain removal to measure residual pericardial effusion.
RESULTS: Ninety patients received Blake drains and 109 patients received Portex drains. There were no statistically significant differences in preoperative variables between the groups. There was no difference in the number of drains inserted per patient between groups (Blake 2.1 ± 0.4 vs Portex 2.0 ± 0.5). Mean difference in size of pericardial effusion between groups was 1.96 mm (95% confidence interval –0.02, 3.95 mm), which did not exceed the predefined non-inferiority margin of 10 mm. There was no significant difference in the mean maximal hourly drainage rate between groups (Blake 94.7 mL vs Portex 123.1 mL; P = .070) or in the total drainage rates (Blake 541 mL vs Portex 679 mL; P = .066). Although the Blake group had a higher percentage of patients with detectable effusion (46.3% vs 27.4%; P = .011) than the Portex group, there was no difference in need for late drainage of pericardial effusions (1.1% vs 1.9%) or insertion of further chest drains (8.8% vs 7.2%).
CONCLUSION: The performance of small-bore Blake drains is not inferior to that of standard Portex drains after cardiac surgery.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
The usual method of drainage is to place large-bore (28F-36F) rigid Portex tubes (Portex, Inc, Hythe, Kent, United Kingdom) in the chest before wound closure. These drains are effective but cause pain,1
may interfere with bypass grafts,2,3 and may provoke cardiac arrhythmias. Despite encouraging reports of the use of smaller silicone rubber Blake drains4 (Ethicon, Inc, Somerville, NJ) and then convincing retrospective study data,5 the routine use of such drains after cardiac surgery has not been generally accepted.
Blake drains are small (10-19F) and have a fluted design, which allows a large surface area for drainage despite the small size. The drainage channels running along the side of the Blake drains reduce the likelihood of tissue invagination into the drains and hence may be less likely to interfere with coronary grafts.
A prospective randomized trial confirmed that Blake drains are less painful than traditional large-bore drains,6 but no efficacy data are available based on a larger prospective randomized controlled trial.
Our study was designed to investigate the efficacy of the two drain types in a randomized controlled trial with primary outcome measure being depth of pericardial effusion postoperatively.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
Local research ethics committees from the two participating centers approved the study and written consent was obtained from all patients.
Patients undergoing second-time procedures, emergency or in-house urgent surgery, triple valve surgery, concomitant carotid endarterectomy, and those with preoperative creatinine values greater than 180 mmol/L were excluded.
Anticoagulants were discontinued 5 days before the operation. Antiplatelet agents were administered 6 hours after the operation unless there was significant postoperative bleeding. Antiplatelet agents were then commenced on day 1 after the operation. Recipients of mechanical valves received warfarin on postoperative day 1. Recipients of tissue valves received high-dose aspirin or warfarin depending on individual consultant protocol.
Patients were randomly assigned to receive either Blake (19F) or Portex (28F) drains in the theater, immediately before drain insertion. The protocol of the study did not specify the site of drain placement; individual consultants followed their standard surgical practice, and insertion sites, number, and location of drains were recorded. All consultants in the study inserted drains via the mediastinum; there were no insertions via the lateral chest wall.
All drains were connected to a chest drain bottle with an underwater seal and were subject to 2.5- to 5-kPa suction from the time of chest closure.
Drains were removed the day after the operation if no air leak was present and if total drainage was less than 20 mL/h for at least 2 hours.
Echocardiographic Assessment
Transthoracic echocardiography was carried out between days 3 and 5 after drain removal to evaluate the presence of pericardial effusion. Echocardiography was performed by dedicated sonographers in each center who were blinded to drain type used. The echocardiographer used a standard form that detailed eight sites for measurements of pericardial effusion depth. At each site three separate measurements were made, which were then averaged. The largest of these averages was recorded and used for analysis.
All chest x-ray films were reviewed by a radiologist in accordance with standard hospital practice, and the presence of pneumothoraces and pleural effusions was reported. Decisions to insert further chest drains or to drain pericardial effusions were made depending on the need of the individual patient concerned.
Adverse events were recorded and monitored throughout the duration of the trial, with regular communication between the two centers.
Statistical Analysis
Statistical analysis was conducted with SAS (Statistical Analysis Software) version 8.02 software (SAS Institute, Inc, Cary, NC). Normally distributed continuous variables were analyzed by a Student t test. Nonnormally distributed continuous data were analyzed with a Wilcoxon-Mann-Whitney test. A logistic procedure with stepwise inclusion of parameters was used to identify key factors that predicted postoperative atrial fibrillation.
Primary Outcome Measure
The study was designed as a non-inferiority test. Non-inferiority was predefined as a difference of less than 10-mm mean pericardial effusion between the two groups as determined by a calculated 95% confidence interval around the mean difference in effusions between the groups. The calculation of the nonparametric 95% confidence interval for effusion used CIA (Confidence Interval Analysis, v2.0.0, T. Bryant, University of Southampton). This statistical analysis was predetermined on the basis of examination of data from a previous study by our group.6 This illustrated a predominance of zero effusion in postoperative patients, meaning the data were nonparametric but the median value was invariably zero. Hence median is a poor comparison measure of primary end point and so is mean. Our statistical analysis used a calculated 95% confidence interval around the mean difference between groups, which, given the distribution of the data, was considered the most appropriate method of comparison.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
|
Despite a trend toward higher maximal hourly drainage rate and total drainage in the Portex group, these differences did not reach statistical significance (Table 2).
|
Table 3 describes the adverse events in the study and demonstrates no differences between the two groups. The incidence of pleural effusion is high in both groups, but no distinction was made between tiny, small, and moderate effusions. The majority of effusions noted were tiny in both groups, as reflected by the low incidence of repeat chest drainage procedures required.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
A recent article addressed the issue of pain during drain removal, concluding that analgesia during drain removal is a challenge.8 Correspondence published in response suggested that the only way to reduce patient discomfort is to use smaller and softer drains. Blake drains fulfill this criterion and have been available for a number of years. Fears over suboptimal performance seem to have prevented their wider use. Our study illustrates that Blake drains performed in a non-inferior manner to larger Portex drains, with the primary outcome measure being depth of postoperative pericardial effusion. The study set clear limits for the definition of non-inferiority, which were developed on the basis of detecting a performance difference that would change clinical practice. The 95% confidence interval of the mean difference between groups, our predetermined primary outcome measure, fell well short of the 10-mm non-inferiority limit. It also crossed zero with a maximal value of 3.95 mm, which is too small a difference to affect clinical practice. Hence we clearly demonstrate non-inferiority between the drains.
This study was not designed to detect a difference in reopening for bleeding or tamponade between the treatments. To detect such a difference, a much larger number of patients would have been required. However, our results of equal safety profile agree with other previously published case-control studies of small and large size. In 2000 Obney and associates4 reported the use Blake drains for mediastinal drainage after cardiac surgery with no increased risk of tamponade in a case series of 100 patients; they also reported less pain with the smaller drains. Since then, a further retrospective case series has reported adequate performance of Blake drains,5 and a retrospective nonrandomized case-control study of 554 patients showed equal efficacy of Blake drains as compared with conventional drains, but with greater patient mobility in the Blake group.9
Akowauh and colleagues6 performed a prospective randomized controlled trial of Blake drains versus Portex drains in 2002, which demonstrated less pain in the Blake group along with equivalent efficacy, including echocardiography of all patients. This, however, was a study of only 70 patients with 35 in each group.
Recently, Ege and associates10 have reported a larger randomized controlled trial investigating the use of Blake drains after cardiac surgery. However, this study randomized patients to standard drains or standard drains plus infracardiac Blake drains and hence the safety of sole Blake drain usage was not evaluated. Interestingly, Ege's group defined an advantage of the addition of an infracardiac Blake drain to the standard drain configuration. We did not identify any advantage of Blake drains over standard Portex drains; however, our study did clearly define the non-inferiority of the smaller drains.
In conclusion, this randomized controlled trial confirms that the performance of small Blake drains was similar to that of larger Portex drains after cardiac surgery. This confirms other retrospective reports and smaller randomized controlled trials. As there is evidence that smaller, more flexible drains cause less pain than the larger alternative, on the basis of our results it is justified to use them routinely for most cardiac surgical patients.
|
Study Limitations |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
The other main study limitation is that consultants in charge of patient care made decisions such as when to insert further chest drains when they were not blinded to the drain type the patient received. Again, we would suggest that the primary end point is unlikely to have been affected by consultant knowledge. However, we concede the possible source of bias for secondary end points.
Finally, this study was designed as a test of non-inferiority against a predefined standard. Hence it is possible that there is a small performance difference between the two drain types in the region of 1 or 2 mm of postoperative pericardial effusion that was not detected in this study due to the study size. However, we would suggest that should such a difference exist, it is too small to have any clinical relevance. Hence we are satisfied with our conclusion of non-inferiority of Blake drains.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionStudy LimitationsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Nakamura, Y. Taniguchi, K. Miwa, Y. Adachi, S. Fujioka, and T. Haruki The use of Blake drains following general thoracic surgery: is it an acceptable option? Interactive CardioVascular and Thoracic Surgery, January 1, 2009; 8(1): 58 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Mair, I. Kaczmarek, and S. Daebritz Modern drainage techniques include not only smaller drains for pain reduction J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1124 - 1124. [Full Text] [PDF]
|
|
|
|
|
|
N. Roberts and T. J. Spyt Reply to the Editor J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1124 - 1125. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bjessmo, S. Hylander, J. Vedin, D. Mohlkert, and T. Ivert Comparison of three different chest drainages after coronary artery bypass surgery -- a randomised trial in 150 patients Eur. J. Cardiothorac. Surg., March 1, 2007; 31(3): 372 - 375. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
