|
|
||||||||
Ann Thorac Surg 2007;83:2009-2016
© 2007 The Society of Thoracic Surgeons
a Department of Cardiac Surgery, Centro Cardiologico Monzino IRCCS, Milan, Italy
b Department of Pharmacological Sciences, University of Milan, Milan, Italy
Accepted for publication January 22, 2007.
* Address correspondence to Dr Parolari, Department of Cardiac Surgery, University of Milan, Centro Cardiologico-Fondazione Monzino IRCCS, Via Parea 4, Milan, 20138, Italy (Email: alessandro.parolari{at}cardiologicomonzino.it).
| Adult cardiac surgery:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
| Abstract |
|---|
|
|
|---|
Methods: A search of computerized databases supplemented with manual bibliographic review was performed for all peer-reviewed English language publications concerning randomized and nonrandomized studies reporting the results of left ventricular reconstruction after both linear and geometric reconstruction techniques. Meta-analyses of several short-term outcomes were performed.
Results: No randomized trial was identified. Eighteen nonrandomized trials were found with a total of 1,814 and 803 patients who underwent linear and geometric reconstruction, respectively. Meta-analysis of all studies (n = 18) revealed an increased risk of in-hospital death for patients undergoing linear reconstruction (relative risk = 1.59, 95% confidence interval: 1.12 to 2.26, p = 0.01). The subanalysis of studies in which linear reconstruction was adopted mainly in the first period of time, and geometric reconstruction was adopted in a later phase, still showed a significant advantage in terms of in-hospital mortality for patients undergoing geometric reconstruction (n = 11 studies, relative risk = 1.89, 95% confidence interval: 1.22 to 2.93, p = 0.004). By contrast, when the two surgical approaches were carried out in the same time lag, there was no difference between linear and geometric reconstruction techniques (n = 7 studies, relative risk = 1.04, 95% confidence interval: 0.57 to 1.92, p = 0.89). No differences in the other outcomes of interest were observed.
Conclusions: The advantage for geometric reconstruction techniques in terms of in-hospital mortality shown in some studies can be an effect of learning curve or of improvement over time in management of these difficult patients. Further studies are required to clarify this issue.
The surgical treatment of left ventricle (LV) aneurysms has been performed for nearly 50 years, but controversy still remains regarding the impact of different surgical techniques on postoperative results. Indeed, various surgical techniques were developed over the years with the aim of restoring LV function and of improving postoperative results. These techniques can be grossly classified into two different categories: linear reconstruction (plication and linear repairs) or geometric reconstruction (circular patch/endoventricular patch closure, direct LV reconstruction using multiple concentric purse string sutures) techniques [1, 2]. Geometric reconstructions have the theoretical advantage of maintaining LV shape and geometry, thus possibly improving postoperative LV performance. That, however, did not consistently translate into improved outcomes, as no effect of surgical technique on operative or on long-term mortality was demonstrated [3, 4], or lower in-hospital and late mortality rates [5] and a possible functional improvement [6] were shown after LV reconstruction with geometric techniques.
There are no systematic reviews or meta-analyses comparing the results of linear versus geometric reconstruction techniques in LV aneurysm surgery; also, meta-analysis may provide additional statistical power that overcomes the limited sample size of most studies together with the low incidence of the major endpoints, for example, in-hospital mortality. Thus, this study is designed to assess whether any difference between the two different LV reconstruction techniques occurs in the early postoperative outcomes of patients undergoing LV aneurysm surgery.
| Material and Methods |
|---|
|
|
|---|
Search Strategy
Two reviewers searched Medline (1966 to May 2006), Embase (1980 to May 2006), and PubMed (up to May 31, 2006), including electronic links to related articles. All peer-reviewed studies published in the English language that dealt with trials comparing different techniques of LV reconstruction for LV aneurysms (both prospective randomized and retrospective observational studies were searched) were identified and reviewed. The text string employed (formatted for PubMed) was: ((left ventricular aneurysm OR left ventricle aneurysm OR left ventricular reconstruction OR left ventricle reconstruction OR left ventricular remodeling OR left ventricle remodelling OR left ventricular aneurysmectomy OR left ventricle aneurysmectomy) AND (plication OR linear closure OR endoaneurysmorrhaphy OR patch repair OR patch closure OR circular patch OR circular patch plasty OR endoventricular patch).
The outcomes searched were the following: in-hospital mortality, stroke, myocardial infarction, acute renal failure, reoperation for bleeding, low output syndrome/postoperative inotropes requirement, and postoperative intra-aortic balloon pump need. The outcome definitions used by the original researchers were accepted. Bibliographies of included articles were also searched.
Several strategies were employed to avoid duplicate publications. If the same institution produced multiple studies, only studies reporting recruitment time periods were considered. If there was sample overlap between studies, only the largest study was included. To minimize temporal bias, as well as interinstitutional variability, studies were included in the meta-analysis only if they contained both linear and geometric reconstruction patient cohorts, with a minimum of 10 patients treated with either technique; also, separate analyses were performed for studies reporting the results of simultaneous use of both techniques and for studies where a temporal trend (eg, linear reconstruction techniques were used mainly or totally in the early years whereas geometric reconstruction techniques were used mainly or totally later on) in the use of linear and geometric reconstruction techniques was clearly identified. Data abstraction and analysis of temporal trends in the adoption of either technique of each study was performed by two reviewers (A.P. and P.D.), and disagreements were solved by consensus.
Analyses
Data abstracted were analyzed by means of RevMan 4.2.8 (Cochrane Collaboration, Oxford, United Kingdom). Effects on dichotomous outcomes were expressed as relative risk (RR) with 95% confidence intervals (CI). Heterogeneity was assessed with the Q statistic. In the absence of significant heterogeneity, treatment effects were pooled with the fixed-effects model. If there was significant heterogeneity (p
0.1), the random-effects model was used; in addition, each outcome was assessed first on all studies selected for meta-analysis, then separately on studies where both techniques were used simultaneously, and on studies where a temporal trend was clearly detectable (techniques used in sequence). For each endpoint, meta-analysis was carried out only when a minimum of 200 patients could be pooled in each treatment arm.
Sensitivity analyses on meta-analyses were performed by removing studies in which the largest (or smallest) effect was found; the study that enrolled the highest number of patients; and studies not reporting any event. Additionally, we performed random-effects meta-analysis on the outcomes of interest.
Publication bias was explored through visual inspection of funnel plots in which the inverse of the estimated variance of the natural logarithm of the adjusted relative risk was plotted against the natural logarithm of the adjusted relative risk for each outcome [8]. Statistical significance was defined by p value of 0.05 or less.
| Results |
|---|
|
|
|---|
Meta-analysis of all the studies (n = 18) showed a significantly increased risk of in-hospital death for patients undergoing linear reconstruction (RR = 1.59, 95% CI: 1.12 to 2.26, p = 0.01).
The temporal sequence of surgical procedure was found of particular relevance, when separate analyses were performed on studies where both techniques were used in temporal sequence (see Table 1 for clinical features) and on studies where both techniques were used simultaneously (see Table 2 for clinical features). The subanalysis of studies in which linear reconstruction was adopted mainly in the first period of time, and geometric reconstruction was adopted in a later phase still showed a significant advantage in terms of in-hospital mortality for patients undergoing geometric reconstruction (n = 11 studies, RR = 1.89, 95% CI: 1.22 to 2.93, p = 0.004; Table 3 and Fig 1). By contrast, when the two surgical approaches were carried out in the same time lag, there was no difference between linear and geometric reconstruction techniques in terms of in-hospital mortality (n = 7 studies, RR = 1.04, 95% CI: 0.57 to 1.92, p = 0.89; Table 3 and Fig 1).
|
|
|
|
Neither our sensitivity analyses nor the test for publication bias modified data, and meta-analyses conclusions remained robust to methodologic changes; also, no publication bias was detected, as tested using the Egger method [8]. As an example, Figure 2 shows the analysis of publication bias for in-hospital mortality of all studies: the shape of the funnel plot was symmetrical.
|
| Comment |
|---|
|
|
|---|
From a theoretical standpoint, and also from an intuitive one, geometric LV reconstruction techniques aimed at restoring and maintaining a more physiologic elliptical LV cavity offer several advantages with respect to linear reconstruction: among them, the chance to exclude the akinesis of the septum, the possible tension decrease in the transitional area, an easier revascularization of the left anterior descending artery, andespeciallyan improved muscle fiber alignment that might result in a more efficient contraction and, as a consequence, better postoperative LV performance that might warrant improved early outcomes.
For these reasons, the concept of restoring the elliptical LV shape and of excluding the noncontractile areas of the LV linked to the geometric reconstruction has been widely accepted by the CT surgical community, and in recent years, this approach has found a potentially much wider target population of patients who are affected by heart failure due to ischemic cardiomyopathy that does not show clearly dyskinetic but only akinetic LV areas [35], patients who may theoretically benefit from geometric left ventricular reconstruction/remodeling. As a consequence, the question of the results that can be achieved with different LV reconstruction strategies has become of great interest.
The results of this meta-analysis highlight some issues of the current surgical therapy of the LV aneurysms. First of all, the possible advantage in terms of in-hospital mortality for patients undergoing surgery following geometric principles of LV reconstruction is likely an effect of the improvement over time of management of these patients and of the learning curve, and this possible advantage needs further evidence from well-designed prospective randomized multicenter studies.
In fact, although the meta-analysis of all the papers pooled together shows that there is a potential protective effect of geometric reconstruction techniques, this evidence is dramatically weakened by the majority of the studies included in the analysis reporting the results of surgical series covering several years of practice, when linear reconstruction techniques were totally or mainly adopted earlier and geometric reconstruction strategies were implemented later, when both management strategies and technical skills of the surgeons might have substantially improved over time. In fact, when examining the effect of the two surgical strategies on the studies that reported the results of surgical series in which both techniques were used at the same time, the effect on mortality was no more evident at all (RR 1.04, 95% CI: 0.57 to 1.92), and there was a substantial equivalence of these two techniques. The same equivalence was also documented for some of the outcomes of interest, namely, reoperation for bleeding, the postoperative need of inotropes or the occurrence of low output syndrome, and the need of aortic counterpulsation; in all these three cases, there was no clear advantage of either approach.
Interestingly, our analysis showed that, with current literature evidence, it is not possible to document differences between these two strategies for three of the major complications occurring after cardiac surgical procedures: perioperative myocardial infarction, stroke, and renal failure. The number of patients who could be pooled for the analysis was in all these three cases quite low and did not reach the criteria that were settled a priori for meta-analysis.
Taken together, these data underscore the scantiness of evidence concerning the efficacy of different surgical approaches aimed at treating LV aneurysms, and highlight the compelling need for additional evidence to support the extension of the geometric reconstruction technique concept to ischemic cardiomyopathy patients who show akinetic areas who may theoretically benefit from LV reconstruction surgery. Even if the vast majority of surgeons (and we are among them) strongly believe that a geometric LV reconstruction strategy is the most appropriate approach even in these cases, our study shows that, unfortunately, definitive evidence about the superiority of this approach is still lacking.
Limitations of the Study
The findings of this meta-analysis must be interpreted with some caution. First, the design of the study may lack the experimental element of a random allocation to the linear or to the geometric reconstruction techniques, and very few studies included in the meta-analysis reported the criteria considered by the individual surgeons to allocate patients to either group. Second, the two groups were not comparable for all the factors that can alter the outcome of interest, and confounding factors cannot be excluded.
It is worth mentioning that most of the studies included in our meta-analysis showed higher mean values of cardiopulmonary bypass time in case of geometric reconstruction technique, whereas the number of grafts performed in both groups was similar. It was not possible in our study to evaluate the clinical importance of this observation in the early outcome, and we may only hypothesize that patients who underwent geometric LV reconstruction might be affected by a relatively more severe disease or disarrangement of the left ventricle requiring a more complex repair, or that geometric reconstruction is a more time-consuming procedure.
Finally, as already stated, it is well known that meta-analysis is most effective when analyzing randomized studies [36], but in this case only observational studies were available. It is obvious that this meta-analysis does not substitute for a randomized trial, and perhaps one is called for. Based on the results of this meta-analysis, it is not clear if such a trial should be designed as a superiority trial or an equivalence trial. In either event, the trial size might be very large; and regardless of its mathematical attractiveness, such a trial might not be clinically realistic.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
P. Klein, J. J. Bax, L. J. Shaw, H. H.H. Feringa, M. I.M. Versteegh, R. A.E. Dion, and R. J.M. Klautz Early and late outcome of left ventricular reconstruction surgery in ischemic heart disease Eur. J. Cardiothorac. Surg., December 1, 2008; 34(6): 1149 - 1157. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Anderson Invited commentary Ann. Thorac. Surg., June 1, 2007; 83(6): 2016 - 2016. [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 |