|
|
||||||||
J Thorac Cardiovasc Surg 2007;134:594-600
© 2007 The American Association for Thoracic Surgery
Cardiopulmonary Support and Physiology |
a Department of Cardiac Surgery, Innsbruck Medical University, Tyrol, Austria
b Department of Medical Statistics, Informatics and Health Economy, Innsbruck Medical University, Tyrol, Austria
c Department of Forensic Medicine, Innsbruck Medical University, Tyrol, Austria
d Department of Anesthesiology and Intensive Care Medicine, Innsbruck Medical University, Tyrol, Austria.
Presented at the World Congress of Cardiology, September 2006, Barcelona, Spain.
Received for publication January 14, 2007; revisions received March 12, 2007; accepted for publication March 22, 2007. * Address for reprints: Elfriede Ruttmann, MD, Department of Cardiac Surgery, Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Austria. (Email: elfriede.ruttmann{at}i-med.ac.at).
| Abstract |
|---|
|
|
|---|
Methods: A consecutive series of 59 patients with accidental hypothermia in cardiocirculatory arrest between 1987 and 2006 were included. Thirty-four patients (57.6%) were resuscitated by standard extracorporeal circulation, and 25 patients (42.4%) were resuscitated by extracorporeal membrane oxygenation. Accidental hypothermia was caused by avalanche in 22 patients (37.3%), drowning in 22 patients (37.3%), exposure to cold in 8 patients (13.5%), and falling into a crevasse in 7 patients (11.9%). Multivariate logistic regression analysis was used to compare extracorporeal membrane oxygenation with extracorporeal circulation resuscitation, with adjustment for relevant parameters.
Results: Restoration of spontaneous circulation was achieved in 32 patients (54.2%). A total of 12 patients (20.3%) survived hypothermia. In the extracorporeal circulation group, 64% of the nonsurviving patients who underwent restoration of spontaneous circulation died of severe pulmonary edema, but none died in the extracorporeal membrane oxygenation group. In multivariate analysis, extracorporeal membrane oxygenation-assisted resuscitation showed a 6.6-fold higher chance for survival (relative risk: 6.6, 95% confidence interval: 1.2–49.3, P = .042). Asphyxia-related hypothermia (avalanche or drowning) was the most predictive adverse factor for survival (relative risk: 0.09, 95% confidence interval: 0.01–0.60, P = .013). Potassium and pH failed to show statistical significance in the multivariate analysis.
Conclusions: Extracorporeal rewarming with an extracorporeal membrane oxygenation system allows prolonged cardiorespiratory support after initial resuscitation. Our data indicate that prolonged extracorporeal membrane oxygenation support reduces the risk of intractable cardiorespiratory failure commonly observed after rewarming.
| Introduction |
|---|
|
|
|---|
|
Because of the protective effects of deep hypothermia, prolonged resuscitation in patients with severe accidental hypothermia is recommended.1
Support with extracorporeal circulation (ECC) is considered the gold standard in the treatment of hypothermic patients with cardiocirculatory arrest; however, survival of these patients is still poor.2,3
Several years ago, portable extracorporeal membrane oxygenation (ECMO) became commercially available and was introduced as an alternative treatment option to ECC for emergency mechanical circulatory support in cardiothoracic surgery.
The successful use of ECMO has been reported in normothermic patients undergoing noncardiac surgery who have cardiogenic shock4,5
and in patients undergoing open surgery who cannot be weaned from cardiopulmonary bypass.6,7
ECMO has several advantages compared with standard ECC equipment: (1) Cannulation can be performed percutaneously (even outside of the operating room during external chest compression) and can be used for single pulmonary support when used as a venovenous system.8
(2) ECMO requires lower levels of anticoagulation and can be used without systemic anticoagulation in case of major bleeding. (3) Prolonged extracorporeal support for hours and days is possible.9
(4) The system can be transferred within hospitals. (5) Negative pressure on the venous cannulas allows higher extracorporeal flow rates.
In contrast with ECC, ECMO support cannot decompress a severely impaired left ventricle because it is a closed system.
In hypothermic patients with cardiocirculatory arrest, ECC is used during cardiopulmonary resuscitation in several centers.10-13
The limited support time of ECC, because of the inflammatory activation and hemolysis, might not be sufficient to adequately treat reperfusion injury.14
Reperfusion edema is a well-known phenomenon in lung transplantation, frequently requiring prolonged ECMO support to provide sufficient oxygenation.15,16
Although previous case series of hypothermia have reported successful rewarming with the use of ECC, there has never been a study published investigating the effect of ECMO-assisted rewarming in hypothermic patients with cardiocirculatory arrest. Most previous studies could not evaluate such an effect because of the limited sample size or unavailability of ECMO support.17,18
During the past few years, we have observed a marked improvement in survival in hypothermic patients resuscitated by ECMO support that could not be solely explained by the more restrictive use of extracorporeal rewarming in patients with concomitant asphyxia (eg, in avalanche accidents). Considering these advantages, we hypothesized that prolonged ECMO support is superior in the treatment of accidental hypothermia with cardiocirculatory arrest.
| Materials and Methods |
|---|
|
|
|---|
Thirty-four patients (57.6%) were resuscitated by ECC (ECC group, 1987–1996), and 25 patients (42.4%) were resuscitated by venoarterial ECMO (ECMO group, 1996-2006).
At admission, 31 patients (52.5%) had asystole, 25 patients (42.4%) had ventricular fibrillation, and 3 patients (5.1%) had pulseless electrical activity. Hypothermia was caused by avalanche in 22 patients (37.3%), drowning in 22 patients (37.3%), exposure to cold in 8 patients (13.5%), and falling into a crevasse in 7 patients (11.9%).
Patient Management and Cannulation Technique
Before patients were admitted to the emergency department, anesthesiologists, cardiac surgeons, and perfusionists were frequently instructed to organize a rescue team for extracorporeal rewarming.
In the ECMO group, patients with return of spontaneous circulation (ROSC) remained on the extracorporeal assist device up to several days until cardiopulmonary function was stabilized (Figure 1).
|
After initiation of cardiopulmonary bypass, the patient was kept at the admission temperature for the first 5 minutes, and then active rewarming was started with a warming rate of approximately 7°C per hour. Body core temperature was measured by tympanum temperature during the rewarming period. The active rewarming period took approximately 1.5 hours until normothermia was reached; the maximum gradient between core temperature and water bath of the heat exchanger was kept at less than 8°C. At a core temperature of approximately 27°C, active (and if necessary recurrent) defibrillation and buffering with sodium bicarbonate were started.
The ECC system consisted of a roller pump (Stöckert, Bonn, Germany) and an oxygenator with an integrated heat exchanger (from 1987 to 1992, a bubble oxygenator, Polystan AS, Denmark; from 1993 to the present, a membrane oxygenator, Jostra AG, Verloese, Denmark). Polyvinyl cannulas (Stöckert, Bonn, Germany) were used for peripheral femoral cannulation.
The ECMO system consisted of a centrifugal pump (Biopump, Bio-Medicus, Eden, Prairie, Minn), a hollow-fiber oxygenator (Maxima, Medtronic Bloodsystem, Anaheim, Calif), polyvinyl chloride tubing, percutaneous arterial and venous cannulas (Medtronic Inc, Minneapolis, Minn), and an integrated heat exchanger (Stöckert, Bonn, Germany).
Statistical Analysis
Possible associations between successful resuscitation and potentially predictive clinical features were assessed by means of univariate and multivariate analyses. To test for univariate differences in categoric variables, the Pearson chi-square test or Fisher exact test (when appropriate) was applied. Continuous variables were tested with the Student t test or Mann–Whitney U test (if assumption of a Gaussian distribution was not fulfilled). Multivariate logistic regression analysis was performed to determine the relative risk (RR) and 95% confidence intervals (CIs) for potential predictors of surviving extracorporeal resuscitation. Thereby, the selection of variables was based on univariate comparisons (entry criteria P < .05) and clinical relevance.
| Results |
|---|
|
|
|---|
The characteristics of the patients according to the applied extracorporeal technique are summarized in Table 1.
|
|
The ages of nonsurvivors were not significantly different from those of survivors (29.5 ± 16.7 years vs 38.3 ± 20.1 years, P = .20). In addition, male gender was not associated with successful resuscitation (87.2% vs 75.0%, P = .37). Asphyxia-related cause of hypothermia, either avalanche or drowning as previously described by Walpoth and colleagues,13
was significantly associated with death from hypothermia (85.1% vs 33.3%, P = .001). The heart rhythm at admission (asystole vs ventricular fibrillation/pulseless electrical activity) did not have a significant influence on survival (57.4% vs 33.3%, P = .20). In comparison with ECC, ECMO-assisted rewarming was significantly associated with improved survival (34.0% vs 75.0%, P = .02). At admission, nonsurvivors had significantly higher potassium levels (8.1 ± 4.3 mmol/L vs 4.9 ± 1.9 mmol/L, P = .014) and lower pH (6.64 ± 0.24 vs 6.80 ± 0.32, P = .026).
The mean burial/submersion time in persons in avalanche and near-drowning accidents was not significantly different with regard to survival (94.7 ± 86 minutes vs 92.2 ± 134.6 minutes, P = .44). In addition, there was no statistical difference in core body temperature (24.5°C ± 0.68°C vs 24.2°C ± 0.35°C, P = .77), mean lactate levels (136.9 ± 64.2 mg/dL vs 118.1 ± 59.0 mg/dL, P = .57), or spontaneous activated clotting time (201.7 ± 45.5 seconds vs 204.7 ± 32.0 seconds, P = .92).
The median ECMO support time was 42 hours (range 18-216 hours) in nonsurvivors and 27 hours in survivors (range 2.5-48 hours).
In the ECC group, the median time from extracorporeal weaning to death was 23 hours (range 0.5-110 hours) in nonsurvivors.
Causes of death in nonsurvivors with a reestablished sinus rhythm are summarized in Table 3.
|
The results of the multivariate analysis are summarized in Table 4.
|
According to Locher and colleagues,14
who described avalanche victims due to the inferior prognosis of concomitant asphyxia separately, we performed a second subanalysis excluding all avalanche victims (n = 22). Furthermore, we wanted to prove the stability of the parameters adjusted in the multivariate regression analysis. In this subanalysis, ECMO-assisted resuscitation was again confirmed to be superior compared with ECC-assisted rewarming regarding survival (RR: 7.6, 95% CI: 1.14–50.8, P = .036). Again, serum potassium (RR: 0.75, 95% CI: 0.49–1.13, P = .16) and pH (RR: 2.53, 95% CI: 0.055–115.8, P = .63) did not indicate statistical significance.
| Discussion |
|---|
|
|
|---|
Furthermore, because both treatment groups (ECC and ECMO) showed a similar restoration of spontaneous circulation but significantly different survivals, our data suggest that the improved outcome of patients who received ECMO was attributable to prolonged cardiopulmonary support. In fact, 9 of 14 patients with ROSC in the ECC group (64%) died of intractable pulmonary edema within a few hours after rewarming, but none in the ECMO group died (however, severe pulmonary edema was frequently observed during the rewarming period in ECMO-assisted patients too) (Figure 1). We therefore conclude that the improved outcome of ECMO-assisted patients was most likely attributable to a reduction of early mortality after initial rewarming.
Extracorporeal support is typically terminated immediately after rewarming in patients on standard ECC support because of bleeding, hemolysis, and inflammatory response. This is often possible only with high-dose catecholamine support and invasive ventilatory support. In contrast, by using ECMO, prolonged extracorporeal support is possible with minimal side effects until the cardiac and pulmonary functions of the patient have been stabilized.9,19
Furthermore, several studies have demonstrated that vasopressors are ineffective in hypothermia and therefore not recommended by international task forces.2,20
A high incidence of severe cardiopulmonary dysfunction is reported in most publications on extracorporeal rewarming of accidental hypothermia among both survivors and nonsurvivors.13,14
Vretenar and colleagues21
reported 60% survival among 68 hypothermic patients in a collective literature review; however, the high survival was explained to be the result of the publication bias of these types of studies and low rate of asphyxia-related cause of hypothermia. In addition, they reported that adult respiratory distress syndrome and severe pulmonary edema were the most frequent complications after successful weaning from ECC. However, in concordance with Vretenar and coworkers, Walpoth and colleagues13
and Locher and colleagues14
reported similar pulmonary complication rates among survivors and nonsurvivors of hypothermia. Thalmann and colleagues17
and Gilbert and colleagues22
reported on 2 hypothermic infants in cardiocirculatory arrest who were resuscitated by ECC, followed by ECMO support for several days because of weaning failure.
Severe pulmonary edema was frequently observed during the rewarming period of our patients. This phenomenon can be partly explained as ischemia-reperfusion injury but may be aggravated by severe diastolic dysfunction. Several studies on therapeutically induced hypothermia have reported severe stiffness of the left ventricle and acute volume overload during rewarming.23-25
If patients have to be weaned from ECC during refractory diastolic dysfunction with elevated left ventricular filling pressures, acute aggravation of reperfusion pulmonary edema occurs. Our study therefore underlines that prolonged ECMO support can prevent this complication if patients are weaned from ECMO after successful stabilization of cardiopulmonary function.
The high mortality rate among our patients can be explained by the high rate of avalanche accidents, which is typical in a high alpine region such as ours. This factor, however, was most predictive for surviving hypothermia.
Locher and colleagues14
described an 84.6% death rate among avalanche victims with cardiocirculatory arrest. In another study by Walpoth and colleagues,11
only 1 of 16 resuscitated hypothermic patients was an avalanche victim. Farstad and colleagues10
reported a survival of 26.9% after ECC, despite only a 3.8% rate of avalanche victims.
Despite the reported inferior outcome of avalanche victims, the results of our analysis confirmed that ECMO support in hypothermic patients had a significant impact on later survival even when excluding these patients.
In 1996, Brugger and colleagues,26
from the Mountain Rescue Service of our region, were the first to propose guidelines for the triage of avalanche victims showing asystole. Their on-site differentiation between asphyxia and hypothermia was based on the burial time, core temperature, and existence of an air pocket.
Hyperkalemia is a well-known marker of poor prognosis in patients with severe accidental hypothermia, indicating cardiac arrest from asphyxia before cooling.27,28
Potassium levels exceeding 10 mmol/L have been considered to justify termination of resuscitation in asphyxia-related accidents; however, Mair and colleagues27
argued that a decision about further resuscitation efforts cannot be based on laboratory parameters. However, these parameters can help avoid using limited ECC facilities for patients with no hope of survival. Therefore, as we have demonstrated that the technique of extracorporeal rewarming had a much higher impact on later survival than obtained laboratory parameters, our data are consistent with the previous observations.28
Although this is the largest consecutive study on extracorporeal-assisted resuscitation in hypothermia, we cannot rule out additional effects of missing covariates. In addition, the lack of randomization and the fact that the histories of the ECC group extend from 1987 to 1996, rather than during the present time, are further limitations on interpreting the findings. However, the clear difference in early mortality secondary to severe cardiorespiratory failure provides an obvious and logical explanation for the findings of our study clearly supporting our hypothesis.
| Conclusions |
|---|
|
|
|---|
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
L. Gordon, G. J. Peek, and J. A. Ellerton Is recommended for severe accidental hypothermia BMJ, December 30, 2010; 341(dec30_2): c7411 - c7411. [Full Text] |
||||
![]() |
T. L. Vanden Hoek, L. J. Morrison, M. Shuster, M. Donnino, E. Sinz, E. J. Lavonas, F. M. Jeejeebhoy, and A. Gabrielli Part 12: Cardiac Arrest in Special Situations: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Circulation, November 2, 2010; 122(18_suppl_3): S829 - S861. [Full Text] [PDF] |
||||
![]() |
L. J. Morrison, C. D. Deakin, P. T. Morley, C. W. Callaway, R. E. Kerber, S. L. Kronick, E. J. Lavonas, M. S. Link, R. W. Neumar, C. W. Otto, et al. Part 8: Advanced Life Support: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations Circulation, October 19, 2010; 122(16_suppl_2): S345 - S421. [Full Text] [PDF] |
||||
![]() |
R. R. Thiagarajan, T. V. Brogan, M. A. Scheurer, P. C. Laussen, P. T. Rycus, and S. L. Bratton Extracorporeal Membrane Oxygenation to Support Cardiopulmonary Resuscitation in Adults Ann. Thorac. Surg., March 1, 2009; 87(3): 778 - 785. [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 |