|
|
||||||||
J Thorac Cardiovasc Surg 2006;131:1114-1121
© 2006 The American Association for Thoracic Surgery
Surgery for Congenital Heart Disease |
a Department of Critical Care Medicine, The Hospital for Sick Children and University of Toronto, Toronto, Canada
b Division of Perfusion Services, The Hospital for Sick Children and University of Toronto, Toronto, Canada
c Division of Cardiovascular Surgery, The Hospital for Sick Children and University of Toronto, Toronto, Canada
d Division of Cardiology, The Hospital for Sick Children and University of Toronto, Toronto, Canada
Received for publication August 1, 2005; revisions received November 2, 2005; accepted for publication November 28, 2005. * Address for reprints: Ian Adatia, MB, ChB, University of California San Francisco Children's Hospital, 505 Parnassus Ave, Room M-655, San Francisco, CA 94143-0106 (Email: ian.adatia{at}ucsf.edu).
| Abstract |
|---|
|
|
|---|
METHODS: We reviewed all patients with a functional single ventricle receiving postoperative extracorporeal life support between January 1997 and May 2003.
RESULTS: We supported 25 infants (age range, 2-139 days; median age, 15 days; weight range, 1.9-5.9 kg; median weight, 3.4 kg) with extracorporeal life support. Operative procedures were Norwood stage 1 procedure in 18 patients, modified Blalock-Taussig shunt in 4 patients, bidirectional superior cavopulmonary shunt in 2 patients, and pulmonary vein repair in 1 patient. Indications for extracorporeal life support included cardiac arrest (14/25) and low cardiac output state (11/25). Extracorporeal membrane oxygenation was initiated in 19 patients, with conversion to a ventricular assist device in 7 patients. Ventricular assist device alone was initiated in 6 patients. Survival to decannulation was 76%, with 5 late deaths from multiorgan failure and 56% intensive care unit survival. Survival to hospital discharge was 44%. On univariate analysis, the presence of arrhythmia before extracorporeal life support (P = .005), renal failure (P = .0007), Candida speciesinduced sepsis (P = .026), and multiorgan failure (P = .0009) were significant risk factors in the nonsurvivors. Median hospital stay was 43.5 days (range, 6-181 days) for the whole group and 93 days (range, 36-181 days) for survivors. Eight patients completed next stage palliation.
CONCLUSIONS: Twenty percent of patients were supported with a ventricular assist device alone, with 50% conversion to a ventricular assist device from extracorporeal membrane oxygenation. Survival to decannulation was encouraging. Multiorgan failure and risk of invasive infection in the postextracorporeal membrane oxygenation period mitigate against survival to hospital discharge. Use of extracorporeal life support before cardiac arrest might reduce attrition between decannulation and hospital discharge.
| Introduction |
|---|
|
|
|---|
We sought to review the outcomes of patients with a functional single ventricle supported with ECLS postoperatively.
| Methods |
|---|
|
|
|---|
Throughout the study period, we maintained a primed ECLS circuit with an oxygenator for rapid deployment, with the option to convert to a ventricular assist device (VAD) alone if suitable. In addition, an in-house team was assembled that was capable of starting a patient on ECLS at short notice. We used standard antibiotic coverage for the first 3 days after starting open-chest ECLS. We maintained strict surveillance for infection with daily blood cultures; we did not use regular mediastinal washouts or change vascular lines routinely. We performed alternate-day cranial ultrasonographic scans or sooner if clinical concern was present.
Data are described as frequencies, medians with ranges, and means with standard deviations. Comparison of categoric data was performed by using
2 or Fisher exact tests. Continuous data were compared by using t tests, paired t tests, and analysis of variance. The Kruskal-Wallis test was used for nonparametric comparisons. Kaplan-Meier "survival" curves were explored to compare "time-to-event" rates. Log-rank tests were used to assess the statistical significance. Statistics were analyzed by using a commercially available software package (SAS version 8.2).
The research and ethics board of the Toronto Hospital for Sick Children approved the study.
| Results |
|---|
|
|
|---|
|
|
Mode of ECLS support on initiation and subsequent course
We initiated ECLS (ie, support with an oxygenator in 19 patients and as a VAD without an oxygenator in 6 patients). In the former group the oxygenator was removed from the circuit in 7 patients after a median of 48 hours (range, 1-62 hours). Removal of the oxygenator from the ECLS circuit was not associated with mechanical complications. One patient who returned from the operating room on VAD support required addition of the oxygenator in the cardiac critical care unit. The Blalock-Taussig (BT) shunt was clipped in 2 patients because of poor systemic perfusion. The patients who received a BT shunt were managed with high flows (mean, 180 ± 40 mL · kg1
· min1) and afterload reduction with phenoxybenzamine, milrinone, or both. We aimed to maintain a normal arteriovenous oxygen saturation difference with systemic oxygenation levels appropriate for a shunt-dependant patient. The patients who received the Sano shunt were managed with mean flows of 150 ± 20 mL · kg1
· min1. Anticoagulation was maintained by means of continuous infusions of unfractionated heparin, which were monitored on the basis of activated clotting time (ACT) and anti-Xa levels. Hourly ACT levels and anti-Xa levels every 12 hours guided the management of anticoagulation on ECLS. The target ACT levels for ECLS were 180 to 200 seconds and for VAD were 140 to 160 seconds with anti-Xa levels between 0.35 and 0.7 U/mL. We found that monitoring of antithrombin 3 levels and the use of constant plasma infusions in small patients improved the effectiveness of our anticoagulation protocol. We re-explored the mediastinum for bleeding, especially when the ACTs and anti-Xa levels were within the target range or there was associated hemodynamic instability. Cardiac catheterization was performed in 13 patients supported with ECLS. We detected residual anatomic lesions in 3 patients (neoaortic obstruction, thrombosed BT shunt, and precoronary obstruction in the native ascending aorta) and significant hemodynamic lesions in 2 patients (one with severe atrioventricular valve regurgitation and one with both atrioventricular and neoaortic regurgitation). In 2 patients with RAI-TAPVC, cryoablation of abnormal conduction pathways was performed during ECLS.
Complications on ECLS
Significant bleeding requiring surgical exploration occurred in 20% of patients supported with a VAD compared with 45% of patients on full ECMO support. Renal failure (defined as increased serum creatinine value above the standard laboratory reference range for the relevant age group and urine output of less than 1 mL · kg1
· h1) was present in 16 (64%) of 25 patients. Peritoneal dialysis was performed in 9 patients, continuous venovenous hemofiltration in 2 patients, and slow continuous ultrafiltration in 5 patients. Ten had sepsis (10/25 [40%]) with positive blood cultures. Six (1 with Pseudomonas species and 5 with systemic Candida species infections) subsequently died. Adverse neurologic events occurred in 7 (28%) of 25 patients. Intracerebral hemorrhage identified by means of cranial ultrasonography occurred in 3 patients (left parietal, thalamic, and intraventricular), cerebral infarcts were seen in 2 patients, and 2 patients had seizures. Six of the 7 died subsequently. There were no neurologic complications in the survivors supported with a VAD alone. Respiratory complications occurred in 3 patients who failed conversion to VAD, but all were decannulated and survived ICU discharge.
The overall median duration of mechanical ventilation was 504 hours (range, 144-2496 hours), and that in survivors was 828 hours (range, 264-2496 hours). Three patients failed initial decannulation, 1 with LCOS and 2 with arrhythmia and subsequent LCOS; all 3 were resupported on ECMO, 1 was bridged to transplantation, and 2 died later, both of whom were unsuitable candidates for transplantation.
RAI-TAPVC Subgroup
Five infants with complex RAI (median age, 62 days; range, 2-120 days; median weight, 3.6 kg; range, 3.1-5.9 kg) were supported on ECLS. Three were supported electively on a VAD for a median duration of 168 hours (range, 96-192 hours). Two were supported on ECMO as rescue after cardiac arrest. Two underwent cryoablation during ECLS for refractory tachyarrhythmias. The 2 supported after cardiac arrest died, and those supported electively were discharged from the critical care unit. One was an expremature baby with complex medical issues who died before hospital discharge.
ICU, Hospital Survival, and Overall Survival Until the End of the Study
Data on ICU, hospital survival, and overall survival until the end of study are shown in Figure 1. Overall, 19 (75%) of 25 patients survived decannulation from mechanical support. In 6 patients ECLS support was withdrawn. None of these 6 patients recovered cardiac function, and their course was complicated by intractable arrhythmias in 2 patients, multiorgan failure in 2 patients, and systemic fungal infection in 2 patients (Tables 3 and 4).
The average duration of ECLS in this group was 181 hours. Five died after decannulation after a median of 9 days (range, 5-23 days) after decannulation, all with multiorgan failure complicated by systemic infection in 4 (Pseudomonas species in 1 and Candida species in 3) and impaired pulmonary blood flow related to a thrombosed shunt in 1 patient. Fourteen patients were discharged from the cardiac critical care unit (survival rate of 56%). Three died after discharge from the cardiac critical care unit but before discharge from the hospital, 1 with RAI who had sepsis after line-related thrombus, 1 with hypoplastic left heart syndrome with severe neurologic impairment, and 1 with hypoplastic left heart syndrome with central core myopathy, aqueduct stenosis, and increased intracranial pressure. The median duration of critical care unit stay of survivors was 50 days (range, 12-120 days), and that in nonsurvivors was 18 days (range, 6-32 days). The median duration of hospital stay in the group who survived to hospital discharge was 97 days (range, 36-181 days). Survival to hospital discharge was 44%.
|
|
|
| Discussion |
|---|
|
|
|---|
The outcome of patients with right isomerism with or without obstructed abnormal pulmonary venous drainage is uniformly poor.
10-12
The elective use of VAD support in this group of infants might have a role to play in the immediate postoperative period. However, improving early and midterm outcomes for this complex group of patients is challenging, and there continues to be high morbidity and mortality.
In the current cohort of patients, we found that the median time of acute deterioration leading to cardiac arrest and ECLS support was 181 hours, which is 7.5 days after surgical intervention. This highlights that the vulnerable critical period after surgical intervention is not limited to the first 24 to 48 hours, as reported previously. Wright and colleagues
13
reported that a dangerous increase in systemic vascular resistance could occur remote from the initial operation. In our experience afterload reduction is crucial in the early postoperative management after the traditional stage 1 Norwood procedure (BT shunt), and continued afterload reduction therapy might be required to prevent late cardiovascular collapse.
14
Among the pre-ECLS variables, severe refractory tachydysrhythmia as a reflection or cause of poor ventricular function was a significant predictor of poor survival. The combination of LCOS and negative inotropic drugs used to control the tachydysrhythmia are poorly tolerated after the Norwood stage 1 procedure. Early support of patients with a tachydysrhythmia with ECLS might permit rhythm control before cardiac arrest. The tendency for survival was higher if ECLS was initiated electively (55% vs 36%), but this was not statistically significant. Jaggers and associates
9
found almost 100% mortality if ECMO was instituted during cardiac arrest. Timely institution of support can limit end-organ damage. Morbidity, mainly caused by multiorgan failure and sepsis after decannulation, however, is a significant problem resulting in long intensive care and hospital stays and significant time-related attrition. In this cohort multiorgan failure and sepsis contributed to significant morbidity. More than half of the patients in this series were supported on ECLS after cardiac arrest, which can increase the risks of multiorgan failure. We found that multiple organ system failure (defined as the failure of at least 2 organ systems,
15
P < .0009) and renal failure (P < .0007) in particular were significant risk factors for death among the nonsurvivors. Many studies have shown that renal failure is an independent factor predisposing to increased mortality.
3,4,9,16
In this study, although the decannulation from ECLS was encouraging, particularly in the high-risk subgroup with RAI-TAPVC, the attrition from the time of decannulation to hospital discharge was a major issue. The median overall survival to hospital discharge for patients with a single ventricle supported on ECLS in the reported literature is 40% (range, 33%-64%).
2,4,6,9,17-21
Our results compare favorably at 44%. In our study the tendency for survival was higher when a VAD alone was used (60% vs 40%) and if the duration of ECLS was less than 120 hours (72% vs 50%), but this was not statistically significant. The relatively small number of patients and the retrospective design of the study preclude use of multivariate analysis.
In a multivariate analysis of risk factors for operative outcome for stage 1 operations in 158 patients, Gaynor and colleagues
18
identified that ECLS support was one of the predictors of operative mortality (P < .001). In a recent study from the same institution reviewing the outcome of all patients supported with ECLS over a 5-year period,
6
single-ventricle physiology and failure to separate from cardiopulmonary bypass were not associated with an increased risk of mortality. The factors identified that increased the probability of mortality in postoperative patients were age less than 1 month, male sex, longer duration of mechanical ventilation before ECLS, and development of renal or hepatic dysfunction during ECLS. Continuing multiorgan failure, infection, and thrombosis in the post-ECLS phase are risk factors for prolonged critical care unit stay, hospital stay, and late mortality. Montgomery and coworkers,
22
in their review of 59 children treated with ECMO after cardiopulmonary bypass over a 9-year period, reported that progressive multiple organ system dysfunction and development of a nosocomial infection were significantly associated with death. This warrants aggressive management of LCOS, anticoagulation, and treatment and prevention of systemic infections.
The Extracorporeal Life Support Organization Registry has reported data over a 4-year period from 1996 through 2000 documenting the substantial increase in neonatal ECLS. Survival rates have remained constant at 35%.
23
However, survival rates in the disadvantaged patients with a single ventricle was not significantly lower than in the biventricular group of cardiac patients.
There is no doubt that in a small subset of patients after functional single-ventricle palliation, use of ECLS positively affects survival. Thus ECLS will remain an important resource, albeit highly technical and expensive, when faced with adverse hemodynamics in the immediate postoperative period. With further advances in the surgical management of hypoplastic left and right heart syndrome, optimum ECLS support in the initial postoperative period will be an important treatment option until identification of residual correctable lesions and recovery of the stressed myocardium. The long-term neurologic morbidity is significant,
24,25
and survivors might have cognitive impairment, motor impairment, or both and require prolonged care, with major implications on health care resources.
26
A second run of ECLS carried a high mortality. However, it was difficult to predict which patients might benefit without recourse to transplantation.
We identified that longer periods of circulatory arrest are associated with reduced survival (Table 4). We suggest that techniques to limit prolonged periods of total circulatory arrest by providing regional perfusion might be important determinants of short-term outcome.
Early identification, recognition, and management of LCOS is crucial because early use of ECLS before cardiac arrest when LCOS is present, especially in the presence of a tachydysrhythmia, might reduce the morbidity related to multiorgan failure between decannulation and hospital discharge. Strategies to improve survival to hospital discharge are required to justify the continued use of ECLS in this patient group, especially as critical care unit and hospital stays are prolonged.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. C. Polimenakos, P. Wojtyla, P. J. Smith, V. Rizzo, M. Nater, C. F. El Zein, and M. N. Ilbawi Post-cardiotomy extracorporeal cardiopulmonary resuscitation in neonates with complex single ventricle: analysis of outcomes Eur J Cardiothorac Surg, December 1, 2011; 40(6): 1396 - 1405. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Petrucci, S. M. O'Brien, M. L. Jacobs, J. P. Jacobs, P. B. Manning, and P. Eghtesady Risk Factors for Mortality and Morbidity After the Neonatal Blalock-Taussig Shunt Procedure Ann. Thorac. Surg., August 1, 2011; 92(2): 642 - 652. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Hetzer, E. V. Potapov, V. Alexi-Meskishvili, Y. Weng, O. Miera, F. Berger, E. Hennig, and M. Hubler Single-center experience with treatment of cardiogenic shock in children by pediatric ventricular assist devices. J. Thorac. Cardiovasc. Surg., March 1, 2011; 141(3): 616 - 623.e1. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Kleinman, L. Chameides, S. M. Schexnayder, R. A. Samson, M. F. Hazinski, D. L. Atkins, M. D. Berg, A. R. de Caen, E. L. Fink, E. B. Freid, et al. Part 14: Pediatric Advanced Life Support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Circulation, November 2, 2010; 122(18_suppl_3): S876 - S908. [Full Text] [PDF] |
||||
![]() |
M. E. Kleinman, A. R. de Caen, L. Chameides, D. L. Atkins, R. A. Berg, M. D. Berg, F. Bhanji, D. Biarent, R. Bingham, A. H. Coovadia, et al. Pediatric Basic and Advanced Life Support: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations Pediatrics, November 1, 2010; 126(5): e1261 - e1318. [Full Text] [PDF] |
||||
![]() |
M. E. Kleinman, L. Chameides, S. M. Schexnayder, R. A. Samson, M. F. Hazinski, D. L. Atkins, M. D. Berg, A. R. de Caen, E. L. Fink, E. B. Freid, et al. Pediatric Advanced Life Support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Pediatrics, November 1, 2010; 126(5): e1361 - e1399. [Full Text] [PDF] |
||||
![]() |
M. E. Kleinman, A. R. de Caen, L. Chameides, D. L. Atkins, R. A. Berg, M. D. Berg, F. Bhanji, D. Biarent, R. Bingham, A. H. Coovadia, et al. Part 10: Pediatric Basic and 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): S466 - S515. [Full Text] [PDF] |
||||
![]() |
L. Shekerdemian Perioperative manipulation of the circulation in children with congenital heart disease Heart, August 1, 2009; 95(15): 1286 - 1296. [Full Text] [PDF] |
||||
![]() |
B. Alsoufi, O. O. Al-Radi, C. Gruenwald, L. Lean, W. G. Williams, B. W. McCrindle, C. A. Caldarone, and G. S. Van Arsdell Extra-corporeal life support following cardiac surgery in children: analysis of risk factors and survival in a single institution Eur J Cardiothorac Surg, June 1, 2009; 35(6): 1004 - 1011. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Chan, R. R. Thiagarajan, D. Frank, and S. L. Bratton Survival after extracorporeal cardiopulmonary resuscitation in infants and children with heart disease. J. Thorac. Cardiovasc. Surg., October 1, 2008; 136(4): 984 - 992. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Tajik and M. G. Cardarelli Extracorporeal membrane oxygenation after cardiac arrest in children: what do we know? Eur J Cardiothorac Surg, March 1, 2008; 33(3): 409 - 417. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Chikovani, J.-H. Hsu, R. Keller, T. R. Karl, A. Azakie, I. Adatia, P. Oishi, and J. R. Fineman B-type natriuretic peptide levels predict outcomes for children on extracorporeal life support after cardiac surgery. J. Thorac. Cardiovasc. Surg., November 1, 2007; 134(5): 1179 - 1187. [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 |