EJCTS Click here for details of sales representative
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Eur J Cardiothorac Surg 2008;33:660-665. doi:10.1016/j.ejcts.2008.01.001
Copyright © 2008, European Association for Cardio-thoracic Surgery. Published by Elsevier. All rights reserved.
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Torsten Doenst
Michael A. Borger
Richard D. Weisel
Terrence M. Yau
Vivek Rao
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Doenst, T.
Right arrow Articles by Rao, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Doenst, T.
Right arrow Articles by Rao, V.
Related Collections
Right arrow Cardiac - other
Right arrow Extracorporeal circulation
Right arrow Myocardial protection

Relation between aortic cross-clamp time and mortality — not as straightforward as expected

Torsten Doensta,b, Michael A. Borgera,b, Richard D. Weisela, Terrence M. Yaua, Manjula Magantia, Vivek Raoa,*

a Division of Cardiovascular Surgery, Toronto General Hospital and University of Toronto, Toronto, Ontario, Canada
b Department of Cardiac Surgery, University of Leipzig Heart Center, Leipzig, Germany

Received 21 September 2007; received in revised form 3 January 2008; accepted 3 January 2008.

* Corresponding author. Address: Toronto General Hospital, 200 Elizabeth Street, 4N-464, Toronto, Ontario, Canada M5G 2C4. Tel.: +1 416 340 3562; fax: +1 416 340 3337. (Email: vivek.rao{at}uhn.on.ca).


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A
 References
 
Objective: Due to modern techniques of cardio protection, less attention has been paid to aortic cross-clamp (XCL) times. However, patients with impaired cardiac contractile function are still at increased perioperative risk, which may be partially due to an increased susceptibility to myocardial ischemia. We tested whether XCL times are associated with perioperative mortality in patients with preserved versus poor left ventricular function. Methods: We determined predictors of operative mortality on all patients undergoing cardiac surgery with aortic cross-clamping in our institution between 1990 and 2003. We excluded patients with markedly prolonged XCL times (>120 min, n = 1426) in order to limit the effect of intraoperative technical difficulties and their known association with poor outcomes. Of the included patients (n = 27,215), 99.8% received antegrade, retrograde, or combined blood cardioplegia. Results: Overall mortality was 2.2%. Multivariable analysis revealed that XCL time was an independent predictor of mortality for patients with LVEF >40% (odds ratio 1.014 per min of XCL, CI 1.01–1.02). However, XCL time was not a predictor in patients with LVEF <40%, mainly due to high mortality in patients with short XCL times. Mortality of patients with an LVEF <40% was the same or higher at cross-clamp times of 1–30 min than at 91–120 min. Conclusions: Despite modern techniques of cardio protection, XCL time remains an independent predictor of mortality in patients with preserved preoperative contractile function. The unexpected lack of risk prediction by aortic cross-clamp time in patients with low ejection fraction appear to be due to a high mortality rate when XCL times were short.

Key Words: Adult cardiac surgery • Mortality • Cross-clamp time • Risk factors


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Appendix A
 References
 
Myocardial protection with cardioplegia is intended to extend the amount of time that hearts can remain arrested, while ascertaining adequate return of post-arrest function. Such an extension of ischemia tolerance is achieved by decreasing the heart's energy demand by applying hypothermia and/or arrest of the contractile apparatus [1–3]. Previous advances in this field have been so great that current techniques provide excellent means to extend the heart's ischemia tolerance, affording the surgical team more time to focus on other aspects of the procedure [3,4].

The detrimental impact of ischemia on myocardial contractile function, however, is not an ‘on’ or ‘off’ phenomenon. Unprotected ischemia causes a ‘wave-front’-picture of detrimental effects over time [5]. These effects do not disappear simply by giving cardioplegia, but rather are delayed, potentially beyond the point of irreversible myocardial damage.

Patients with impaired left ventricular function are at increased perioperative risk, as evident by greater mortality when compared to similar patients with normal contractile function [6–9]. It may be hypothesized that the impact of protected ischemia on contractile function, i.e. aortic cross-clamp time with cardioplegia, is responsible for at least a portion of this difference. A small ischemia-induced decrease in contractile function may not be harmful to a patient with normal contractile function, while it may greatly affect a patient with borderline contractile function.

We therefore assessed the impact of aortic cross-clamp time on mortality in a large cohort of patients according to whether they had preserved or impaired preoperative left ventricular function.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Appendix A
 References
 
We reviewed data on all patients undergoing cardiac surgery at our institution from January 1990 to June 2003 (n = 28,684). We excluded patients with markedly prolonged XCL times (>120 min, n = 1426) because of the likelihood of intraoperative technical difficulties that would disproportionately affect outcome. Patients undergoing heart transplant and extra-cardiac procedures were also excluded. The vast majority of patients (88%) received cold antegrade blood cardioplegia, while the remainder received either retrograde or combined antegrade/retrograde cardioplegia. The ratio of blood to crystalloid mixture was changed from 4:1 to 8:1 in 1996.

Perioperative management was the same for all patients. Cardiopulmonary bypass (CPB) flow rates were maintained between 2.0 and 2.5 l/min m2 and mean arterial pressure was kept above 70 mmHg in patients with renal insufficiency or peripheral vascular disease. Hematocrit was kept above 20% and mild hypothermia (34 °C) was employed during CPB. Moderate or severe hypothermia were used only if patients required circulatory arrest. Cardioplegia consisted of oxygenated blood mixed with crystalloid in a 4:1 ratio until 1996 and an 8:1 ratio thereafter. Cardioplegia was delivered cold and antegrade in the majority of patients, as previously described [8,10].

2.1 Data source
Data were gathered on all patients including pre-, intra-, and postoperative variables, and were entered into our computerized database by a full-time research nurse. We have maintained our institutional database for approximately 15 years, and quality assurance checks have consistently revealed a missing data rate of <2% and an error rate of <2%. All adverse outcomes were recorded by our research nurse, using clinical criteria that have been previously described [10]. Mortality was defined as in-hospital death.

2.2 Analysis
Categorical data are expressed as percentages and continuous data as mean ± standard deviation throughout the manuscript. SAS version 8.2 (SAS Institute; Cary, NC) was used for all statistical analyses. Categorical data were analyzed univariately by Chi square or Fisher's exact test, and continuous data were analyzed by Student's unpaired t-tests or Wilcoxon rank-sum tests where appropriate. Stepwise multivariate logistic regression analysis was used to calculate risk-adjusted odds ratios and to determine the independent predictors of each outcome of interest. For a complete list of variables assessed in the logistic regression models, please see Appendix A. All variables suggested by the univariate analysis (p < 0.25) or those judged to be clinically important were entered into the logistic regression models. Model discrimination was evaluated by the area under the receiver operating characteristic (ROC) curve and model precision was evaluated by the Hosmer–Lemeshow goodness-of-fit statistic, as previously described [10].


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Appendix A
 References
 
Table 1 shows demographic data of all patients undergoing cardiac surgery at the Toronto General Hospital between 1990 and 2003. There were 3.5 times more patients with preserved left ventricular function (ejection fraction ≥40%) than patients with poor left ventricular function (ejection fraction <40%). With the exception of atrial fibrillation, the incidence of all preoperative risk factors was significantly higher in the group with poor ventricular function.


View this table:
[in this window]
[in a new window]

  		Table 1 Demographic parameters of patients undergoing cardiac surgery at the Toronto General Hospital between 1990 and 2003
 
Fig. 1 shows the distribution of patients with preserved or poor ventricular function into the three surgical categories used in our database. There is a similar distribution of cases in both groups with a majority of patients undergoing isolated coronary artery bypass grafting (CABG), followed by valve ± CABG surgery (valve) and all other procedures (other, see Table 2 ). Other procedures included replacement of the ascending aorta, adult congenital procedures, left ventricular aneurysmectomy, and mapping procedures.


Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 1. Number of patients undergoing coronary artery bypass grafting (CABG), valve surgery with or without CABG (valve) or other procedures (other) with a left ventricular ejection fraction equal to or greater than 40% (filled bars) versus less than 40% (hollow bars). * p < 0.05.

 

View this table:
[in this window]
[in a new window]

  		Table 2 Perioperative parameters of patients undergoing cardiac surgery at the Toronto General Hospital between 1990 and 2003
 
Table 2 shows the perioperative parameters for the two patient groups. More patients with poor ventricular function were operated on under urgent/emergent conditions. Although CPB times were longer in the group with poor ejection fraction, there was little difference (2 min) in aortic cross-clamp times. As with the preoperative risk factors, all adverse outcomes were significantly higher in the low ejection fraction group including low output syndrome, renal failure, stroke, and pulmonary complications.

Fig. 2 illustrates the distribution of patients into four cross-clamp time intervals according to their preoperative left ventricular function. As in Fig. 1, the distribution of patients was similar. In the majority of patients, 30–90 min of cross-clamp time was required. Less than 2% of patients required cross-clamp times greater than 120 min and these patients were excluded for the reasons noted above.


Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 2. Distribution of cases as a function of aortic cross-clamp (XCL) time in patients with a left ventricular ejection fraction equal to or greater than 40% (filled bars) or less than 40% (hollow bars). Patients with XCL times above 120 min were excluded in order to limit the effect that intraoperative technical difficulties would have on outcomes.

 
Fig. 3 shows mortality of the two groups as a function of aortic cross-clamp time. Patients with preserved left ventricular function had the lowest mortality during short cross-clamp times, with a steady increase in mortality as cross-clamp times were extended. While patients with poor left ventricular function displayed higher mortality rates than patients with normal function at all time intervals, their operative mortality rates followed a ‘U-shape’ distribution with increased risk at both ends of the spectrum. Mortality in these patients was greatest when cross-clamp time was less than 30 min (8.3%, 13/157). In the other three time brackets, mortality increased with growing XCL times, similar to the group with normal ventricular function. We examined the effects of cardioplegia type (4:1 blood to crystalloid ratio vs 8:1 ratio) and method of delivery (antegrade, retrograde, or combined) and did not find a significant effect on mortality (data not shown).


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		Fig. 3. Relation between aortic cross-clamp times and mortality in patients with preserved (ejection fraction equal or greater than 40%, filled bars) versus poor left ventricular function (ejection fraction less than 40%, hollow bars).

 
Tables 3 and 4 list the different risk factors identified as independent predictors of mortality in patients with ejection fraction ≥than 40% (Table 3) and <40% (Table 4). While aortic cross-clamp time was an independent predictor of mortality in patients with normal ventricular function, it was not identified as a significant risk factor in patients with poor ventricular function. The discriminative ability of both models was very good with an area under the ROC curve of 0.830 and 0.805 for patients with normal LV function and impaired LV function, respectively. In addition, both models displayed good precision with a Hosmer–Lemeshow statistic of 0.41 and 0.86, respectively.


View this table:
[in this window]
[in a new window]

  		Table 3 Independent predictors of mortality in patients with EF more than 40%
 

View this table:
[in this window]
[in a new window]

  		Table 4 Independent predictors of mortality in patients with EF less than 40%
 
We also subdivided patients into those receiving CABG ± other procedures versus those receiving no CABG, to determine if coronary artery disease modified the effect of prolonged ischemia. The results were unchanged, i.e. there was a significant effect of prolonged cross-clamp time on mortality in patients with preserved LV function, but not in patients with impaired LV function (data not shown).


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Appendix A
 References
 
In the current study, we have demonstrated that aortic cross-clamp time is an independent predictor of mortality in patients with preserved preoperative contractile function. However, we failed to demonstrate a significant correlation between ischemic time and operative mortality in patients with impaired left ventricular function, provided ischemic time is kept below a certain limit (120 min). To the best of our knowledge, this is a novel observation in such a large group of patients.

Our results are somewhat counterintuitive in patients with left ventricular dysfunction. One would expect that a postoperative decrease in contractile function, which is more likely to occur with increasing ischemic times, would have a greater impact on outcome in patients with poor ventricular function than those with preserved function. However, our results may be used to argue to the contrary. It might be speculated that hearts with poor ventricular function have greater ischemia tolerance than hearts with normal ventricular function. While this conclusion appears to be novel, it is analogous to observations made in the field of ischemic preconditioning. Kloner and co-workers [11,12] have repeatedly demonstrated that patients with coronary artery disease are less likely to die from a myocardial infarction if previous angina has been documented. The investigators attribute this effect to previous repeated ischemic episodes in the patients with angina, providing the heart with preconditioning stimuli and therefore offering protection during the index ischemia. It is interesting to note that in our patients with poor ventricular function, the incidence of triple vessel disease, preoperative myocardial infarction and angina pectoris was significantly greater than in patients with preserved ventricular function (Table 1), potentially providing such ‘preconditioning-like’ stimuli. Based on this reasoning, it may be speculated that hearts with poor ventricular function may have increased ischemia tolerance and may therefore experience greater protection by cardioplegia than hearts with normal function. However, it has to be emphasized that these conclusions are hypothetical.

Despite the conclusion of potentially improved ischemia tolerance in hearts with impaired contractile function, overall mortality is still higher in these patients compared to those with normal ventricular function. It is clear from Table 1 that patients with poor ventricular function have a significantly higher prevalence of other risk factors that negatively impact on outcome [8,9,13,14]. Our multivariate analysis underscores these findings by identifying several other variables as independent predictors of mortality. It is interesting to note that the relative risks for emergency or redo coronary bypass grafting differed greatly between the two patient populations. However, attributing the differences in total mortality to any of these statistical variables is not appropriate, since our analysis did not identify exact causes for these differences. We are only able to state that the impact of extending aortic cross-clamp time on mortality is less than expected in patients with poor left ventricular function. Thus, it is still possible (and probably likely) that the ischemic event contributed to mortality. However, the duration of ischemia may be less important in these patients. This conclusion finds indirect support by a recent study from Rastan et al. [15]. The authors demonstrated that avoiding cardioplegic arrest through beating heart strategies reduced mortality and improved other outcomes in patients undergoing emergency coronary surgery.

Our analysis also revealed that mortality is high in patients with poor ventricular function and short cross-clamp times. The reason for this observation is not entirely clear, as we were unable to identify any major differences in this subset of patients compared to others (data not shown). It is conceivable that the majority of patients with short ischemic times and poor ventricular function received incomplete revascularization, a known risk factor for mortality [16]. Presumably, the target vessels were inappropriate and fewer grafts than planned could be performed. It is also possible that in the interest of time, distal target vessels were judged too small and incomplete revascularization was accepted. We attempted to include this information into our statistical analysis, but were unable to satisfactorily define incomplete revascularization using those variables, which are present in our database. Another possible explanation is that some high-risk cases required long cardiopulmonary bypass times with only a short period of cross-clamping, e.g. an adult congenital procedure focus on the right heart.

It may also be argued that cross-clamp time was not identified as an independent risk factor in patients with poor ventricular function due to the increased mortality rate in the group with cross-clamp times below 30 min. There is a similar increase in mortality with time in the other three time intervals compared to the group with normal function (see Fig. 3). Based on this interpretation, a word of caution should be noted from a practical standpoint of view. One should not arrive at the conclusion to deliberately extend cross-clamp times in patients with poor ventricular function in order to reduce mortality. However, it may not be necessary to ‘rush’ through a procedure in patients with ventricular dysfunction. The impact of cross-clamp time on mortality is either less or at least not greater than in patients with normal function.

In the current study, we excluded patients with cross-clamp times greater than 120 min. We reasoned that in such patients, there was a high likelihood that technical difficulties may have caused the increased cross-clamp times and that these difficulties would disproportionately affect outcome. Having limited the analysis to a maximum of 2 h of clamp time increases the likelihood of achieving the desired completeness of surgery with the first attempt and therefore should enhance the precision of the statistical analysis.

A potential limitation of our study is the smaller sample size in the group with poor ventricular function, which may have decreased our ability to detect long cross-clamp time as an independent predictor. However, there was still a large number of these patients (n = 6000) available for analysis with a significant number of outcomes of interest (i.e. mortality, n = 255). We therefore feel our conclusions are justified. Another possible limitation is our analysis of mortality, which we defined as in-hospital death, i.e. death that occurred any time within hospitalization. It is possible that this resulted in an underestimation of mortality rates when compared to 30-day mortality, but this is unlikely given that we do not transfer patients to other hospitals who have suffered serious postoperative complications and since sudden death at home shortly after hospital discharge is a very uncommon event. The final limitation of the current study is that we were unable to provide information on the mode of death. This information is not recorded in our database for several reasons, the most important being that only a small proportion of patients undergo autopsy at our institution and therefore we do not feel confident in assigning a cause of death in each case. However, it is likely that the majority of perioperative deaths were cardiac-related, as would be the norm for most cardiac surgery centers.

In summary, we demonstrate that despite modern techniques of cardio protection, aortic cross-clamp time is an independent predictor of mortality in patients with preserved preoperative contractile function. In contrast, this correlation was not found for operative mortality in patients with impaired left ventricular function. Further research should be performed to investigate the possible explanation for our novel observation.


   Appendix A
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A
References
 
Variables assessed in predictive models:


View this table:
[in this window]
[in a new window]

 
 


   Acknowledgments
 
We wish to thank Carole Cürten for editorial assistance.


   Footnotes
 
{star} TD was supported by grants from the Deutsche Forschungsgemeinschaft (Do602/3-1, 3-2, 8-1).


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A
 References
 

  1. Bretschneider HJ. Myocardial protection. Thorac Cardiovasc Surg 1980;28:295-302.[Medline]
  2. Bigelow WG, Mustard WT, Evans JW. Some physiologic concepts of hypothermia and their applications to cardiac surgery. J Thorac Surg 1954;28:463-480.[Medline]
  3. Buckberg GD. Update on current techniques of myocardial protection. Ann Thorac Surg 1995;60:805-814.[Abstract/Free Full Text]
  4. Guru V, Omura J, Alghamdi AA, Weisel RD, Fremes SE. Is blood superior to crystalloid cardioplegia? A meta-analysis of randomized clinical trials. Circulation 2006;114(1 Suppl.):I331-I338.[Medline]
  5. Jennings RB, Reimer KA. The cell biology of acute myocardial ischemia. Annu Rev Med 1991;42:225-246.[CrossRef][Medline]
  6. Hannan EL, Racz MJ, Walford G, Jones RH, Ryan TJBE, Culliford AT, Isom OW, Gold JP, Rose EA. Long-term outcomes of coronary-artery bypass grafting versus stent implantation. N Engl J Med 2005;352:2174-2183.[Abstract/Free Full Text]
  7. Doenst T, Wijeysundera D, Karkouti K, Zechner C, Maganti M, Rao V, Borger MA. Hyperglycemia is an independent predictor of morbidity and mortality in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2005;130:1144-1150.[Abstract/Free Full Text]
  8. Yau TM, Fedak PW, Weisel RD, Teng C, Ivanov J. Predictors of operative risk for coronary bypass operations in patients with left ventricular dysfunction. J Thorac Cardiovasc Surg 1999;118:1006-1013.[Abstract/Free Full Text]
  9. Christakis GT, Weisel RD, Fremes SE, Ivanov J, David TE, Goldman BS, Salerno TA. Coronary artery bypass grafting in patients with poor ventricular function. J Thorac Cardiovasc Surg 1992;103:1083-1092.[Abstract]
  10. Borger MA, Ivanov J, Weisel RD, Peniston CM, Mickleborough LL, Rambaldini G, Cohen G, Rao V, Feindel CM, David TE. Decreasing incidence of stroke during valvular surgery. Circulation 1998;98:II137-II143.[Medline]
  11. Kloner RA, Przyklenk K, Whittaker P, Hale S. Preconditioning stimuli and inadvertent preconditioning. J Mol Cell Cardiol 1995;27:743-747.[CrossRef][Medline]
  12. Kloner RA, Przyklenk K, Shook T, Matthews RV, Burstein S, Cannom DS. Clinical aspects of preconditioning and implications for the cardiac surgeon. J Card Surg 1995;10(4 Suppl.):369-375.[CrossRef][Medline]
  13. Koch CG, Weng YS, Zhou SX, Savino JS, Mathew JP, Hsu PH, Saidman LJ, Mangano DT, Group IRaEFMSoPIR Prevalence of risk factors, and not gender per se, determines short- and long-term survival after coronary artery bypass surgery. J Cardiothorac Vasc Anesth 2003;17:585-593.[CrossRef][Medline]
  14. Robinson K, Conroy RM, Mulcahy R. Risk factors and in-hospital course of first myocardial infarction in the elderly. Clin Cardiol 1988;11:519-523.[Medline]
  15. Rastan AJEJI, Hentschel B, Funkat AK, Gummert JF, Doll N, Walther T, Falk V, Mohr FW. Emergency coronary artery bypass graft surgery for acute coronary syndrome: beating heart versus conventional cardioplegic cardiac arrest strategies. Circulation 2006;114(1 Suppl.):I477-I485.[Medline]
  16. Hannan EL, Racz M, Holmes DR, King SBR, Walford G, Ambrose JA, Sharma S, Katz S, Clark LT, Jones RH. Impact of completeness of percutaneous coronary intervention revascularization on long-term outcomes in the stent era. Circulation 2006;113:2380-2382.[Free Full Text]



This article has been cited by other articles:


Home page
 Eur. J. Cardiothorac. Surg.Home page
T. Doenst, H. Bugger, M. Schwarzer, G. Faerber, M. A. Borger, and F. W. Mohr
Three good reasons for heart surgeons to understand cardiac metabolism
Eur. J. Cardiothorac. Surg., May 1, 2008; 33(5): 862 - 871.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Torsten Doenst
Michael A. Borger
Richard D. Weisel
Terrence M. Yau
Vivek Rao
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Doenst, T.
Right arrow Articles by Rao, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Doenst, T.
Right arrow Articles by Rao, V.
Related Collections
Right arrow Cardiac - other
Right arrow Extracorporeal circulation
Right arrow Myocardial protection

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