|
|
||||||||
J Thorac Cardiovasc Surg 2007;134:1322-1331
© 2007 The American Association for Thoracic Surgery
Cardiothoracic Transplantation |
a Institute of Social Medicine, Medical University of Schleswig-Holstein, Campus Luebeck, Luebeck, Germany
b Department of Cardiovascular Surgery, Medical University of Schleswig-Holstein, Campus Luebeck, Luebeck, Germany.
* Address for reprints: Maike Schnoor, Institute of Social Medicine, Medical University of Schleswig-Holstein, Campus Lübeck, Beckergrube 43-47, 23552 Lübeck, Germany. (Email: Maike.schnoor{at}uk-sh.de).
| Abstract |
|---|
|
|
|---|
Methods: A systematic review with meta-analysis was performed. As data sources, we used the electronic databases EMBASE and Medline (1966–August 2006), hand searching in 4 journals, expert consultation, and reference lists of reviews. Observational and randomized and prospective and retrospective controlled trials that reported outcomes on the 2 techniques of heart transplantation were considered.
Results: A total of 23 retrospective and 18 prospective studies were included. Meta-analyses of prospective trials including between 228 and 472 patients revealed significant superiority of the bicaval technique in comparison with the biatrial procedure for early atrial pressure (weighted mean difference, –3.95; 95% confidence interval, –6.50 to –1.40), perioperative mortality (odds ratio, 0.41; 95% confidence interval, 0.17 to 0.98), tricuspid valve regurgitation (odds ratio, 0.23; 95% confidence interval, 0.15 to 0.36), and sinus rhythm (odds ratio, 7.01; 95% confidence interval, 2.57 to 19.13). The latter also showed a significant difference in the analysis of retrospective studies (odds ratio, 2.69; 95% confidence interval, 1.55 to 4.66).
Conclusion: In summary, this systematic review and meta-analysis provides evidence of clinically relevant beneficial effects of the bicaval technique in comparison with those of the standard technique. Nevertheless, the longer-term beneficial effects of the bicaval technique remain to be evaluated.
| Introduction |
|---|
|
|
|---|
There are several studies comparing these 3 different techniques of orthotopic heart transplantation that have also been summarized in recent reviews.9-11
It is essential to summarize and appraise the available studies under the rigorous methods of evidence-based medicine to help in the decision making on what technique should be preferred. This has not been done thus far.
We therefore aimed to compare the more recent bicaval heart transplantation techniques (Figure 1),
6,7
both combined under the term "bicaval techniques," with the standard procedure (Figure 1) for clinically relevant outcomes and, by implementing the methods of a systemic review and meta-analysis, to achieve the best available level of evidence for that subject.
|
| Materials and Methods |
|---|
|
|
|---|
We conducted a literature search in PubMed and the database of the German Institute of Medical Documentation and Information, including EMBASE and Medline, from 1966 through August 2006 using the following search strategy: [("heart transplantation") OR ("cardiac transplantation") AND bicaval]. In addition, the Cochrane library of systematic reviews was visited.
For a hand search, the 4 journals (The Annals of Thoracic Surgery, The European Journal of Cardiothoracic Surgery, The Journal of Heart and Lung Transplantation, and The Journal of Thoracic and Cardiovascular Surgery) that provided the most studies on the use of the bicaval technique thus far were identified. A hand search was then performed for the period from 1999 through June 2005.
In addition, HHS was asked to give hints on further literature and research groups that were not covered by the achieved studies. Furthermore, the reference lists of recent reviews were checked for relevant literature. Studies had to evaluate orthotopic heart transplantation by using the bicaval technique to be eligible for inclusion.6,7
We included only studies in the English or German languages. The search was not limited to randomized controlled trials. Observational controlled and uncontrolled, prospective and retrospective studies were included. The search was not age restricted. We excluded letters, comments, case reports and series, and nonhuman studies.
A 2-stage filter process applying the eligibility criteria was implemented by screening titles and abstracts first and then full texts. Both stages of the filter process were performed independently by 2 investigators (DL, TS). Results were then compared and showed identical results.
Data Extraction and Critical Appraisal
The data extraction and critical methodologic appraisal of the included studies was undertaken by MS. Levels of evidence were assigned to each study according to the Oxford Centre of Evidence-based Medicine. For the detailed methodologic assessment, a standardized form, which was developed on the basis of the Scottish Intercollegiate Guideline Network checklist, and an evaluation sheet of the German Institute for Quality and Efficiency in Health Care were used. Both are standardized instruments for the quality assessment of randomized controlled trails. The extracted information included lead author, publication year, intervention and observation period, study group characteristics (number, sex, and age), indication, criteria of inclusion and exclusion, techniques of the operation, statistical methods, outcomes, and adverse events. Quantitative results, either as means and standard deviations or rates, were summarized in tables.
With respect to the methodologic quality of the study, we further appraised the randomization process (sequence generation and allocation concealment), blinding, statistical methods, baseline characteristics of both groups, and handling of losses to follow-up.
Outcome Measures
All reported outcomes were retrieved and could be divided into 3 groups. The first group included clinically relevant outcomes, which were assessed in more than 2 studies in a comparable way and on which a meta-analysis could be performed. This group included intraoperative ischemic time of transplantation, permanent pacemaker insertion, early mortality (30-day mortality), 1-year survival, 3-year survival, duration of hospital stay, and atrial pressure in the early postoperative period. Furthermore, tricuspid valve regurgitation and sinus rhythm after cardiopulmonary bypass, although measured at different time points in prospective studies (Table E1), were considered.
|
The last group included outcomes that were determined only in single studies, such as the hemodynamic vasomotor responses to lower body negative pressure. These results were not considered in our review.
Statistical Analyses
We reported and summarized descriptive results as either means or rates according to the original publications. Meta-analyses were performed for outcomes as mentioned above for prospective and retrospective controlled trials separately. The software RevMan 4.2 (Cochrane Collaboration, http://www.cc-ins.net/RevMan/) was used for these analyses. Odds ratios or weighted differences of the means and corresponding 95% confidence intervals (CIs) were reported as measures of association and stability. Heterogeneity among studies for every outcome was assessed by using the Cochrane Q test. Fixed-effect models were chosen in case no significant heterogeneity among studies was observed. Otherwise, random-effect models are presented. Sensitivity analyses were performed by comparing random and fixed-effect models, evaluating the effect of omitting influential studies, presenting funnel blots for parameters in which fixed-effect models were used with more than 3 included studies, and omitting studies with overlapping patient samples.
| Results |
|---|
|
|
|---|
|
|
|
Methodologic Quality
Prospective studies
Most of the studies had potential or obvious methodologic limitations. Two of the included 18 prospective studies were uncontrolled trails. In 9 studies the study population was randomized to the operation technique, in most of them (n = 8) on an alternate basis. The latter cannot be considered as adequate because the allocation was foreseeable. One study did not describe the randomization technique at all. The study population consisted of less than 20 patients per group in 8 studies. Two studies failed to give information about inclusion or exclusion criteria. The time between transplantation and observation was shorter in the bicaval group compared with the standard group in 4 studies. For 2 studies the observation time point is unknown. Another 2 studies did not give information on the indication for transplantation. In all but 2 controlled studies,E14,E19
the study groups (bicaval and standard) were comparable regarding age, sex and preoperative parameters. One studyE14
showed a significant age difference between study groups, and the other studyE19
reported a significant difference in the preoperative right atrial pressure.
Retrospective studies
We included 23 retrospective studies in our review. Four of the studies were not controlled by a patient group undergoing heart transplantation by means of the standard technique. In 9 studies the allocation to a study group occurred by pseudorandomization (time intervals). Some centers compared the time periods before and after the introduction of the bicaval technique. In 3 studies the observation period was shorter for the bicaval group than for the standard group. Four studies reported no exact observation period. In 4 studies the sample size was small (<20 patients per group). Inclusion and exclusion criteria were given in 15 studies, and the indication for heart transplantation was given in another 15 studies.
Outcomes
Meta-analyses
The results of the meta-analyses are displayed in Figure 3
for prospective and Figure E1
for retrospective studies.
|
|
The summary of 3 prospective studies each proved a significantly reduced early atrial pressure of 4.0 mm Hg (Figure 3, B) and a significantly reduced perioperative mortality of 59% (relative risk reduction; Figure 3, C) by means of the bicaval technique. The proportion of patients with tricuspid valve regurgitation was reduced significantly by 77% in the bicaval group according to the summarized results of 7 prospective studies (Figure 3, D). In case the assessment of tricuspid valve regurgitation was graded, only moderate and severe cases were considered, and in case multiple observation points were given, results of the latest observation were included. According to 2 prospective studies, a sinus rhythm was achieved significantly more frequently with the bicaval technique (Figure 3, D). It turned out that the meta-analyses of prospective studies only included those that implemented the bicaval transplantation technique according to Sievers and associates,7
with the exemption of the study of Beniaminowitz and coworkers,E19
in which both bicaval techniques were used.6,7
The summary of 4 retrospective studies showed a nonsignificant reduction in hospital stay of 1 day in the bicaval group (Figure E1, B). Patients undergoing the bicaval technique also had a nonsignificant 88% risk reduction for a permanent pacemaker implantation (Figure E1, C). Furthermore, 1-year and 3-year mortality was reduced in patients who underwent the bicaval technique by 40% and 36%, respectively (nonsignificant; Figure E1, D and E). The summary of retrospective studies also indicated a significantly higher proportion of patients achieving sinus rhythm with the bicaval technique (Figure E1, F).
Outcomes not included in the meta-analyses
The quantitative results of the outcome measurements not included in the meta-analysis are shown in Table E1. In summary, superiority of the bicaval technique introduced by Sievers and associates7
is indicated for the outcomes "need for temporary pacemaker,"E9-E11,E20
"right atrial pressure 12 months after transplantation,"E3,E7,E21
"pulmonary artery pressure after 1 year,"E7,E22
"cardiac index at first postoperative day,"E22-E24
"mitral valve regurgitation,"E7,E11
"tricuspid valve regurgitation,"E7,E20,E24-E26
and "left atrial thrombosis."E27,E28
No differences were seen for "pulmonary vascular resistance,"E3,E9
"systolic blood pressure,"E5,E29
"cardiac output,"E7,E9,E25
and intensive care unit stay.E5,E22
Sensitivity analysis
A comparison of the results of fixed- and random-effect models is displayed in Table E2. Overall, the effect estimates did not differ substantially by either model. As expected, the higher variability under the random-effect assumption led to loss of statistical significance in some cases (perioperative mortality in prospective studies and permanent pacemaker in retrospective studies).
|
|
|
| Discussion |
|---|
|
|
|---|
Heart transplantation in its most anatomic form would require 8 circumferential anastomoses of 4 pulmonary veins, 2 venae cavae, and 2 arteries. This technique, however, has several shortcomings, such as prolonged ischemic time, potentially difficult accessible suture lines in case of bleeding, anastomotic stenoses, and surgical complexity. Therefore, different transplantation techniques were developed experimentally to reduce the number of anastomoses. Berman and coworkers in 195714
sutured the donor left atrium to cuffs around the pulmonary veins of the recipient on each side, reducing the number of anastomoses to 6. Berman and associates14
left the posterior cuff of the recipient left atrium in place, reducing the number of anastomoses to 5. Cass and Brock,3
in 1959, and Lower and Shumway,4
in 1960, introduced the concept of 2 atrial and 2 arterial anastomoses, reducing the number of anastomoses to 4.
This latter technique has been the standard clinical procedure since 1967, when the first human heart transplantation was performed by Barnard.1
Major parts of the recipient right and left atria are left in situ, the donor left atrium is partially excised, and the donor right atrium is incised from the inferior vena cava to the right atrial appendage. Postoperatively, the atria are acutely enlarged, showing a figure-of-eight configuration.5
Theoretically, these morphologic alterations might interfere with hemodynamic, electrophysiologic, innervative, and valvular function of the donor heart. Therefore former experimental alternative surgical principles were introduced into clinical practice recently. In 1989, Banner and associates6
first used the total transplantation technique, leaving the left and right atria completely intact. In 1991, Sievers and colleagues7
introduced clinically the bicaval technique, preserving the right atrium intact, only using 2 vena caval anastomoses, and leaving a small bridge of recipient left atrial tissue in place for simple left atrial anastomosis similar to those of the standard technique. During the last 15 years, the bicaval technique has become the most commonly used procedure for orthotopic heart transplantation.15
Different refinements of the principal technique have been used, such as interrupted sutures for the caval anastomosis with absorbable or unabsorbable material, and performed during the ischemic state or even leaving some right atrial bridge between the superior and inferior vena cava.16,17
This meta-analysis provides evidence that the expected theoretic advantages of bicaval transplantation in comparison with the standard technique have come true in clinical practice. In prospective trials, a reduction in right atrial pressure was found. The absolute difference in right atrial pressure is probably of no clinical relevance at rest. Conclusions with respect to clinical relevance under exercise with increased tricuspid regurgitation cannot be drawn.E17
The higher rate of sinus rhythm after transplantation, the significant reduced rate of tricuspid valve regurgitation, the prevention of contraction abnormalities by the acute atrial enlargement with the standard technique, and the asynchrony of recipient and donor atrial innervation probably have contributed to the beneficial hemodynamic effects after bicaval transplantation.18-20
The enlargement and distension of the atria after the standard technique might not only induce impairment of electrical impulse initiation and conduction, as well as trigger arrhythmias,21,22
but also promote atrial thrombus formation most likely avoided by the bicaval technique.23,E27
There was a trend toward reduced permanent pacemaker requirement for the bicaval groups.E3,E7,E10,E11,E23
However, also with the standard technique, the incidence of permanent pacemaker implantation can be kept comparably low when sinus node area is protected.E30,E33,E34
Furthermore, Bernardi24
found that the bicaval technique leads to an increased parasympathetic reinnervation compared with the standard technique, which might be of clinical relevance because an increase in control of blood pressure by larger reflex changes in heart rate might improve adaptation to various stimuli and to physical exercise.
Potential shortcomings of the bicaval technique include the marginally prolonged ischemic transplantation time of some minutes, which is likely of no clinical relevance, as well as some kind of stenosis at the level of the venous anastomoses. Both problems, however, can be neutralized by refined surgical techniques, such as performing anastomoses with the unclamped aorta and partially interrupted caval sutures. Furthermore, it can be discussed whether the longer hospital stay (when omitting an influential study) in the bicaval group is related to the particular surgical technique or the patients clinical conditions.
This review has limitations. Although we implemented several strategies to obtain all relevant studies, including searches in electronic databases, hand searching, contact with experts, and screening of actual reference lists, it cannot be excluded that we have missed some information.
The quality of the review is reflected by the quality of the included single studies. Well-performed single studies can provide valid information and bear the potential advantage of a low variability in important parameters, such as myocardial protection, reperfusion techniques, or immunosuppression. The majority of studies were retrospective in nature. These rather observational retrospective cohort studies have not implemented a rigorous randomization principle, which makes these studies prone to several biases, including selection and information bias (Level of evidence 2b). Furthermore, in 2 studies it is not always ascertained that the outcomes were obtained in a standardized and comparable way, which downgraded these studies to level of evidence 4 (poor-quality cohort studies). We identified 7 prospective randomized trials. All studies did not describe the randomization technique in detail or the randomization was on alternate basis, which downgraded these studies to level of evidence 2b (low-quality randomized controlled trial). Overall, the study quality was limited mainly because of a small sample size, lack of randomization, and blinding. Blinding should be possible for the patient, and an independent outcome assessment should be introduced.
Some reports were based on overlapping patient samples. We therefore performed meta-analyses with results from different centers only. The results changed only marginally. The estimate on early right atrial pressure in prospective studies lost its statistical significance.
Another potential limitation of meta-analysis is that different studies assessed outcomes in different ways and at different time points. We included in the meta-analysis studies assessing tricuspid valve regurgitation at different time points. This could bias our results and should be considered in the interpretation. The different time points could influence the pacemaker implantation, as well as the tricuspid valve insufficiency. Especially the latter might be impaired by biopsy and rejection. In addition, differences in the protocols of the practical performances of the outcome assessments cannot be ruled out, which might have affected comparability. There might also be a time trend in survival because of changing conditions in heart transplantation, which is not related to a particular transplantation technique.
A consequence for future studies should be the use of a minimum set of clinically relevant outcomes, which must be measured in a standardized and comparable way.
In summary, this systematic review and meta-analysis proves evidence of clinically relevant beneficial effects of the bicaval technique in comparison with the standard technique of orthotopic heart transplantation, which warrants careful consideration for further decision making. As a result of this analysis, the perioperative mortality seems to be reduced significantly in subjects after bicaval transplantation. Nevertheless, the longer-term beneficial effects of the bicaval technique remain to be evaluated, especially with regard to exercise capacityE16
and patient-oriented outcomes, such as health-related quality of life (Short Form–36).
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
| E-References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. M. Dell'Aquila, S. Mastrobuoni, G. Bastarrika, B. L. Praschker, P. A. Aguero, S. Castano, J. Herreros, and G. Rabago Bicaval versus standard technique in orthotopic heart transplant: assessment of atrial performance at magnetic resonance and transthoracic echocardiography Interact CardioVasc Thorac Surg, January 4, 2012; (2012) ivr084v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. R. Banner, A. R. Simon, and M. M. Burke 49 Heart transplantation Oxford Textbook of Heart Failure, January 1, 2011; 1(1): med-9780199577729-chapter - med-9780199577729-chapter. [Abstract] [Full Text] |
||||
![]() |
R. R. Davies, M. J. Russo, J. A. Morgan, R. A. Sorabella, Y. Naka, and J. M. Chen Standard versus bicaval techniques for orthotopic heart transplantation: An analysis of the United Network for Organ Sharing database J. Thorac. Cardiovasc. Surg., September 1, 2010; 140(3): 700 - 708. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Fernandez, D. Babadjanov, and R. S. Farivar Mitral and tricuspid valve repair 21 years after cardiac transplantation J. Thorac. Cardiovasc. Surg., July 1, 2010; 140(1): e3 - e4. [Full Text] [PDF] |
||||
![]() |
J. H. Shuhaiber, J. Moore, and D. B. Dyke The effect of transplant center volume on survival after heart transplantation: A multicenter study J. Thorac. Cardiovasc. Surg., April 1, 2010; 139(4): 1064 - 1069. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Jacob and F. Sellke Is bicaval orthotopic heart transplantation superior to the biatrial technique? Interact CardioVasc Thorac Surg, August 1, 2009; 9(2): 333 - 342. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. A. Hunt and F. Haddad The Changing Face of Heart Transplantation J. Am. Coll. Cardiol., August 19, 2008; 52(8): 587 - 598. [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 |