Postcardiotomy extracorporeal membrane oxygenation in patients with congenital heart disease;the effect of place of initiation

Abstract

Background

Postcardiotomy extracorporeal membrane oxygenation (ECMO) in pediatric patients can be affected by the place of initiation, either in the operating room (OR) or the pediatric cardiac intensive care unit (PCICU). This study aimed to characterize and compare patients who had postcardiotomy ECMO initiation in the OR or PCICU and evaluate risk factors for hospital mortality.

Methods

This retrospective study included 103 patients who required postcardiotomy ECMO support after the repair of congenital cardiac lesions from 2010 to 2022. Patients were grouped according to the place of ECMO insertion into two groups. Group 1 (n = 69) had ECMO insertion in the OR, and Group 2 (n = 34) had ECMO insertion in the PCICU.

Results

Cardiac arrest occurred significantly more often in patients with ECMO insertion in the PCICU (21 (61.76%) vs. 13 (18.84%); p < 0.001). Pre-ECMO lactate levels, pH, VIS, base deficit, and PaO₂ did not differ between the groups. Re-exploration for bleeding was significantly higher in Group 1 (32 (46.38%) vs. 8 (23.53%); p = 0.03). Cannula repositioning (4 (11.76%) v. 2 (2.90%); p = 0.09) and mechanical ventilation time were nonsignificantly higher in Group 2 (19.5 (10–31) vs. 11 (5–25) days; p = 0.07). No difference in mortality was found between groups (42 (60.87%) vs. 23 (67.65%), p = 0.50). By multivariable analysis, elevated lactate on ECMO and low pH before ECMO were associated with mortality.

Conclusions

ECMO insertion in the OR has a comparable mortality rate to PCICU insertion. Pre-ECMO low pH and high lactate during ECMO could predict mortality.

Keywords

extracorporeal membrane oxygenation congenital cardiac disease lactate mortality

Introduction

Extracorporeal membrane oxygenation (ECMO) is widely used to support patients with congenital heart disease (CHD) and has become a solution for cardiorespiratory insufficiency when conventional therapies fail.¹ In the postoperative period following the repair of CHD, ECMO may be required for many reasons, including failure to separate the patients from cardiopulmonary bypass (CPB) or progressive hemodynamic deterioration with cardiogenic shock or cardiac arrest.² Despite recent advances in ECMO technology, the mortality rate in pediatric patients is still high.³ The Extracorporeal Life Support Organization (ELSO) registry reported survival to hospital discharge in 42% of neonatal cardiac ECMO, 41% in neonatal ECMO for cardiopulmonary resuscitation (E-CPR), 51% in pediatric cardiac ECMO, and 41% in pediatric E-CPR.⁴

ECMO remains the most commonly used mechanical cardiopulmonary support in infants and young children in the immediate postoperative period. Factors affecting hospital mortality and morbidity in patients with CHD supported with ECMO vary widely in the literature. This variation could be attributed to the heterogeneity of the population, different ECMO indications and techniques, and the timing of ECMO initiation.⁵ It is controversial whether ECMO initiation in the operating room (OR) or pediatric cardiac intensive care unit (PCICU) will affect hospital mortality. ECMO initiation in the PCICU could occur in the context of cardiac arrest or cardiogenic shock and, consequently, a higher mortality risk. On the other hand, ECMO insertion in the operating room may indicate severe cardiac dysfunction.⁶ Elmahrouk and associates found that ECMO connection in the OR was associated with a nonsignificantly higher survival (49%) than ICU ECMO (40%).⁷ Other studies reported that early ECMO support for myocardial failure before the commencement of prolonged cardiogenic shock was crucial for better outcomes in patients with severe cardiac failure after congenital cardiac surgery.^8,9 Therefore, this study aimed to characterize patients with postcardiotomy ECMO initiation in the OR or PCICU, compare the outcomes between these two groups, and evaluate risk factors for hospital mortality.

Patients and methods

Design and Setting

This research is a retrospective study conducted in the PCICU of Prince Sultan Cardiac Centre-Riyadh. All congenital heart disease patients admitted to the PCICU who required ECMO support between January 2010 and January 2022 were enrolled in the study. The data were collected from electronic and paper medical records. The case notes of 122 patients were reviewed. Out of these 122 patients, we excluded 19 patients: 7 had incomplete medical records, and 12 were excluded because they had no prior cardiac surgery. One hundred and three patients were included in the study. All patients included required ECMO support postcardiac surgery for congenital lesions due to low cardiac output syndrome. The place of ECMO insertion was either in the PCICU or OR. Patients were grouped according to the place of ECMO insertion into two groups. Group 1 (n = 69) had ECMO insertion in the OR, and Group 2 (n = 34) had ECMO insertion in the PCICU.

The study was approved by the Scientific Research Center of the Prince Sultan Military Medical City (IRB approval number: 1665).

Study variables and outcomes

The study variables included patients’ demographics, associated syndromes, the Risk Adjustment for Congenital Heart Surgery (RACHS) category, cardiac arrest, highest lactate, lowest PH, base deficit, highest creatinine, lowest PaO₂, and highest vasoactive inotropic score (VIS). Operative variables included bypass and ischemic times and the number of bypass runs. The biochemical parameters were reported before and during ECMO support. The primary outcome was hospital mortality. Secondary outcomes were hospital complications, ECMO-related complications, duration of PCICU, and hospital stay.

ECMO technique

The indication for ECMO in these patients was failure to wean from cardiopulmonary bypass due to low cardiac output syndrome. All patients received central veno-arterial ECMO support. According to protocol procedures, the type, size, and site of the cannula depend on weight, age, and indication of ECMO. Pump flow was started initially at 100 mL/kg/min and adjusted to maintain adequate perfusion; a higher flow of up to 200 mL/kg/min may be needed for patients with single ventricle physiology. Inotropes were kept at a minimum and adjusted as needed. Afterload-reducing agents are often utilized to improve cardiac performance in patients with high blood pressure. Before cannulation, a dose of 50–100 units/kg heparin was given. Activated clotting time (ACT) was performed within 5 min of initiation of ECMO and every 60 min until target ACT was achieved and bleeding was controlled (chest tube drainage less than 5 mL/kg/h). Anticoagulation was achieved by continuous heparin infusion, maintaining an ACT between 180 and 200 s with heparin infusion according to the protocol. ACT levels were monitored two hourly for the first 12 h; then, the follow-up was with partial thromboplastin time (PTT) according to the protocol. Heparin reversal was performed to achieve the target ACT on ECMO for patients who had ECMO insertion because of failure of separation from cardiopulmonary bypass. Hemoglobin was kept at 12 g/dl, and the platelet count was over 100,000/mm3. The left heart was decompressed via a left atrial cannula inserted through one of the pulmonary veins in all patients.

Ventilation was adjusted at a low setting to maintain oxygenation of blood generated by native cardiac output and to prevent lung atelectasis. Head ultrasound was performed on alternate days for patients with open fontanels to exclude intracranial hemorrhage. The target temperature was between 35.5 and 37°C. Diuretics were given to promote diuresis when urine output was below 2–3 ml/kg/hr. In persistently oliguric or anuric patients, slow continuous ultrafiltration or peritoneal dialysis was used to remove extra volume. In patients with renal failure, dialysis was preferred. Nutrition was commenced 24–48 h after the institution of ECMO. The enteral route was preferred unless the food was poorly absorbed. In this case, total parenteral nutrition was started alternatively. Broad-spectrum prophylactic antibiotics, usually ceftazidime and vancomycin, were prescribed, and the dose was adjusted as necessary.

Statistical analysis

Qualitative data are presented as numbers and percentages and were compared with the chi-squared or Fisher exact test when appropriate. Quantitative data are presented as the mean and standard deviation or median (25th −75th percentiles), according to the normality distribution tested with the Shapiro‒Wilk test. Quantitative data were compared with the t test or Wilcoxon test when appropriate. Repeated measures were compared between the two groups with random effect regression. Survival was plotted using a Kaplan‒Meier curve and compared with the log-rank test. Stepwise logistic regression with forward selection was used to identify the risk factors for hospital mortality. Significant variables in the univariable analysis were introduced into the multivariable models, and a p value of 0.05 or less was required to retain the variables in the final model. Stata 17 was used for all analyses (Stata Corp- College Station, TX, USA).

Results

Baseline and operative data

The study included 17 (24.64%) neonates and 33 (47.83%) infants in Group 1 vs 8 (23.53%) and 15 (44.12%) in Group 2, respectively. There were no differences in age groups, sex, gestational age, weight, height, or syndromes between groups. RACHS score categories were significantly higher in Group 1. ECMO insertion in the operating room was due to failure of separation from cardiopulmonary bypass in 56 patients (81.16%) and sudden cardiac arrest after separation from cardiopulmonary bypass and heparin reversal in 13 patients (18.84%). ECMO insertion in the PCICU was indicated for low cardiac output in 13 patients (38.24%) and for cardiac arrest in 21 patients (61.76%). The median time to ECMO insertion in the PCICU was 24 h (25th- 75th percentiles: 6–123 h).

Cardiac arrest occurred significantly more often in patients with ECMO insertion in the PCICU (13 (18.84%) vs. 21 (61.76%), in Group 1 vs 2, respectively; p < 0.001). Pre-ECMO lactate levels, pH, VIS, base deficit, and PaO₂ did not differ between the groups. Comparing the ischemic and cardiopulmonary bypass times and the number of cardiopulmonary bypass runs showed no difference between groups. (Table 1)

Table 1.

Comparison of preoperative and operative data between patients who received ECMO in the operating room (Group 1) and those who received ECMO in the pediatric cardiac intensive care unit (Group 2).

	Group 1 (n = 69)	Group 2 (n = 34)	p Value
Age group			0.67
Neonate	17 (24.64%)	8 (23.53%)
Infant	33 (47.83%)	15 (44.12%)
Child (1–14 years)	15 (21.74%)	11 (32.35%)
Adolescent (14–18 years)	1 (1.45%)	0
GUCH (>18 years)	3 (4.35%)	0
Female	27 (39.13%)	17 (50%)	0.29
Gestational age			0.84
Full-term	31 (45.59%)	14 (41.18%)
Preterm	4 (5.88%)	3 (8.82%)
Post-term	1 (1.47%)	1 (2.94%)
Unknown	33 (47.82%)	16 (47.06%)
Weight (kg)	3.7 (3–9.1)	5.25 (3.3–9.5)	0.38
Height (cm)	53 (48–74)	66.5 (51–80)	0.13
RACHS	(n = 67)	(n = 34)	0.038
Category 1	1 (1.49%)	0
Category 2	5 (7.46%)	10 (29.41%)
Category 3	35 (52.24%)	13 (38.24%)
Category 4	22 (32.84%)	8 (23.53)
Category 5	1 (1.49%)	0
Category 6	3 (4.48%)	3 (3 (8.82%)
Single ventricle	5 (7.25%)	7 (20.59%)	0.058
Syndrome			0.28
None	59 (85.51%)	27 (79.41%)
Down	4 (5.80%)	4 (11.76%)
William	3 (4.35%)	0
Arterial tortuosity	2 (2.90%)	1 (2.94%)
Keutel	1 (1.45%)	0
Dysmorphic	0	1 (2.94%)
DiGeorge	0	1 (2.94%)
Cardiac arrest before ECMO	13 (18.84%)	21 (61.76%)	˂0.001
Lactate pre-ECMO (mmol/L)	(n = 63) 6.8 (3.4–11.8)	(n = 34) 9.5 (3.7–20)	0.14
pH before ECMO	(n = 67) 7.25 (7.17–7.31)	(n = 34) 7.20 (7.04–7.3)	0.10
VIS score before ECMO	(n = 25) 20.76±15.31	(n = 32) 29.17±12.17	0.02
Base deficit before ECMO	(n = 67) −6 (−10.7–−1.3)	(n = 34) −5.4 (−10.3–−1)	0.11
PO2 before ECMO (kPa)	(n = 67) 4 (3.1–6)	(n = 34) 3.51 (3.03–4.42)	0.09
Creatinine (mmol/L)	(n = 50) 45 (28–71)	(n = 30) 41 (29–56)	0.42
Cardiopulmonary bypass time (min)	215.5 (123.5–302.5) (n = 68)	174.5 (125–246) (n = 32)	0.30
Ischemic time (min)	94 (50–135) (n = 67)	84 (42–107) (n = 31)	0.18
Number of bypass cycles	(n = 68)	(n = 32)	0.53
One	50 (73.53%)	28 (87.50%)
Two	10 (14.71%)	2 (6.25%)
Three	3 (4.41%)	1 (3.13%)
Four	5 (7.35%)	1 (3.13%)

ECMO, extracorporeal membrane oxygenation; GUCH, Grown Up Congenital Heart, PCICU, pediatric cardiac intensive care unit; RACHS, Risk Adjustment for Congenital Heart Surgery; VIS, vasoactive inotropic score

Continuous data are presented as the mean and standard deviation or median (Q1–Q3), and categorical data are presented as numbers and percentages.

Outcomes

There were no differences in lactate levels, pH, base deficit, lowest hemoglobin, VIS, and creatinine levels on ECMO between the groups. Re-exploration for bleeding was significantly higher in Group 1 (32 (46.38%) vs. 8 (23.53%); p = 0.03). Cannula repositioning (4 (11.76%) v. 2 (2.90%); p = 0.09) and mechanical ventilation time were nonsignificantly higher in Group 2 (19.5 (10–31) vs. 11 (5–25) days; p = 0.07). Death occurred in 65 patients, and all patients died in the PCICU. Death occurred on ECMO in 52 patients (80%) and after ECMO decannulation in 13 patients (20%). No difference in mortality was found between groups (42 (60.87%) vs. 23 (67.65%), p = 0.50). The causes of death were neurological insult (n = 11), septic shock (n = 2), hemolysis (n = 1), bowel ischemia (n = 1), and multiorgan failure (n = 50). (Table 2)

Table 2.

Comparison of extracorporeal membrane oxygenation (ECMO) and postoperative data between patients who had ECMO in the operating room (Group 1) vs. pediatric cardiac intensive care unit (Group 2).

	Group 1 (n = 69)	Group 2 (n = 34)	p value
Highest lactate on ECMO (mmol/L)	(n = 68) 10.25 (5.95–19.5)	15.5 (9–23)	0.09
Lowest pH on ECMO	(n = 68) 7.26 (7.15–7.3)	(n = 34) 7.22 (7.01–7.29)	0.11
Lowest base deficit on ECMO	(n = 68) −6.3 (−11.15–−3.55)	(n = 34) −7.85 (−15.3–−2.6)	0.40
Lowest PaO2 on ECMO (pKa)	(n = 68) 3.64- (3.01–4.11)	(n = 34) 3.53–2.72–4.26)	0.59
Lowest hemoglobin on ECMO (g/dl)	9.9 (8.8–10.8)	10.3 (9–11.4)	0.19
Highest VIS on ECMO	(n = 66) 23 (15–37.5)	(n = 32) 27.5 (18–33)	0.55
Highest creatinine on ECMO (mmol/L)	65 (44–92)	74 (41–97)	0.97
Steroid for low cardiac output	27 (39.13%)	12 (35.29%)	0.71
Multiple ECMO insertion	6 (8.70%)	3 (8.82%)	˃ 0.99
Return to OR while on ECMO	31 (44.93%)	12 (35.29%)	0.35
Reintervention for residual defects	7 (10.14%))	3 (8.82%)	˃ 0.99
Re-exploration for bleeding	32 (46.38%)	8 (23.53%)	0.03
Cannula repositioning	2 (2.90%)	4 (11.76%)	0.09
Circuit thrombosis	3 (4.35%)	0	0.55
Oxygenator malfunction	1 (1.45%)	1 (2.94%)	˃ 0.99
Dialysis	25 (36.23%)	11 (32.35%)	0.70
Nitric oxide	19 (27.54%)	10 (30.30%)	0.77
High-frequency oscillating ventilation	3 (4.35%)	4 (11.76%)	0.22
Diaphragmatic paralysis	12 (17.65%)	2 (5.88%)	0.13
Days on ECMO	6 (4–10)	9 (5–14)	0.06
Mechanical ventilation (days)	11 (5–25)	19.5 (10–31)	0.07
PCICU stay (days)	19 (8–43)	24 (11–47)	0.44
Hospital stay (days)	20 (7–44) (n = 64)	29 (11–52) (n = 31)	0.35
Hospital mortality	42 (60.87%)	23 (67.65%)	0.50

ECMO, extracorporeal membrane oxygenation; OR, operating room; PCICU, pediatric cardiac intensive care unit; VIS, vasoactive inotropic score.

Continuous data are presented as the mean and standard deviation or median (Q1–Q3), and categorical data are presented as numbers and percentages.

There was no difference in the change in lactate levels between the groups (β: 0.88 (95% CI: −1.22–2.98); p = 0.413). (Figure 1) There was no significant change in VIS score between groups (β: 4.78 (95% CI: −0.21–9.76); p = 0.06) (Figure 2).

Figure 1.

Changes in lactate levels for patients who had extracorporeal membrane oxygenation in the operating room (Group 1) or pediatric cardiac intensive care unit (Group 2). (Values are presented as the median, and the bar is the interquartile range.)

Figure 2.

Changes in the vasoactive inotropic score (VIS) for patients who had extracorporeal membrane oxygenation in the operating room (Group 1) or intensive care unit (Group 2). (Values are presented as the median, and the bar is the interquartile range.)

Survival at 3 months was 38% in Group 1 vs 31% in Group 2 (log-rank p = 0.678) (Figure 3).

Figure 3.

Kaplan‒Meier survival estimates comparing patients who had extracorporeal membrane oxygenation (ECMO) insertion in the operating room (Group 1) vs. pediatric cardiac intensive care unit (Group 2).

Factors affecting mortality

Hospital mortality occurred in 65 patients (63.11%). The place of ECMO initiation (either in the OR or PCICU) was not associated with hospital mortality (p = 0.503). Univariable analysis revealed that mortality was associated with pH before ECMO and lactate levels (after 24 h and the highest levels on ECMO), pH, base deficit, VIS, and hemoglobin levels on ECMO. By multivariable analysis, elevated lactate on ECMO and low pH before ECMO were associated with mortality (Table 3).

Table 3.

Multivariable logistic regression for factors affecting hospital mortality.

Mortality	Univariable analysis		Multivariable analysis
Mortality	Or (95% CI)	p Value	Or (95% CI)	p Value
ECMO insertion in PCICU	1.34 (0.57–3.20)	0.503	-	-
Lactate after 24 h on ECMO	1.92 (1.08–3.41)	0.027	-	-
Highest lactate on ECMO	1.15 (1.08–1.24)	<0.001	1.15 (1.07–1.24)	<0.001
pH before ECMO	0.02 (0.0008–0.67)	0.028	0.14 (1.07–1.24)	0.038
Lowest base deficit on ECMO	0.84 (0.77–0.92)	<0.001	-	-
Lowest pH on ECMO	0.002 (0.0003–0.11)	0.002	-	-
Highest VIS on ECMO	1.03 (1.004–1.06)	0.020	-	-
Lowest hemoglobin on ECMO	0.73 (0.57–0.93)	0.012	-	-

CI, confidence interval; ECMO, extracorporeal membrane oxygenation; PCICU, pediatric cardiac intensive care unit; OR, odds ratio; VIS, vasoactive inotropic score.

Discussion

ECMO in patients with congenital heart disease is still associated with high morbidity and mortality rates. Several studies have demonstrated variable risk factors predicting mortality in those patients. Early ECMO insertion before end-organ damage could be associated with better outcomes.¹⁰ The effect of ECMO initiation in the OR compared to the PCICU on hospital mortality has not been thoroughly evaluated. This study characterized patients who had ECMO insertion in the OR or PCICU after the repair of congenital cardiac lesions. Additionally, we compared the outcomes in those patients and assessed factors affecting hospital mortality. ECMO insertion in the OR has the potential advantage of rapid insertion because of the availability of the surgical team and the equipment. Additionally, insertion in OR mostly occurs before clinical deterioration and cardiac arrest and is a continuation of the cardiopulmonary bypass. ECMO insertion in the OR before heparin reversal and hemostasis could increase the risk of bleeding. Pizarro and colleagues reported that ECMO initiation outside the OR was a risk factor for mortality after Norwood palliation.¹¹ We reported a significantly higher prevalence of cardiac arrest in patients who had ECMO insertion in the PCICU. Re-exploration for bleeding was higher in patients who had ECMO in the OR, with no difference in mortality between groups. The lack of difference in hospital mortality between the place of ECMO initiation in our study could be explained by early insertion before clinical deterioration and the availability of the equipment and trained personnel in the PCICU. Timely ECMO initiation can be demonstrated in both groups by the absence of differences in the biochemical parameters between groups.

ECMO support can be used to support children in all age groups.¹² This study’s most common age groups were neonates and infants, with no difference in age groups between patients who had ECMO insertion in the OR and PCICU. Additionally, age group was not a predictor of mortality in our cohort. The survival rate was lower in neonatal ECMO compared to other age groups, according to the ESLO report (42% vs. 52%, respectively).¹³ However, this report included patients with diagnoses other than postcardiotomy cardiogenic shock. In our series, the mortality rate among neonates was 72%, 56% in infants, and 62% in older children. Morris and associates found that neonates and male sex were risk factors for mortality.¹⁴

We did not report demographic, clinical, or disease-related factors associated with hospital mortality. This study included 13 patients with a single ventricle, with a nonsignificantly higher prevalence in patients with ECMO initiation in the PCICU (p = 0.07). Alsoufi et al. reported in their 100 postoperative pediatric ECMO series that survival was 32% in patients with single ventricles.¹⁵ Several studies reported worse outcomes in patients with single ventricles supported with ECMO.^5,16–19 Several other risk factors were reported to increase hospital mortality, including total anomalous venous connection,²⁰ residual lesions,²¹ long bypass times,^22,23 low birth weight,²⁴ and prematurity.²⁵

In this study, pre-ECMO low pH and high lactate levels during ECMO predicted mortality. Several biochemicals were found to be significant predictors of mortality in pediatric postcardiotomy ECMO, including pre-ECMO parameters.²⁶ Walter and associates reported elevated creatinine and blood urea nitrogen on ECMO as risk factors for mortality.²² In Kolovos et al.’s study on 74 pediatric patients who received postcardiotomy ECMO, high lactate 48 h after ECMO initiation predicted mortality,²⁶ and Polimenakos reported that high lactate 24 h after ECMO initiation was an independent risk factor for mortality.²⁷ Alsoufi and colleagues found that delayed lactate normalization predicted mortality in postcardiotomy pediatric ECMO.¹⁵ Similar to our study, Howard and associates found that pre-ECMO low pH (<7.17) was a risk factor for mortality.²⁵ In their study on 47 children with veno-arterial ECMO, Chen and colleagues found that a high lactate level 12–24 h after ECMO insertion was an independent predictor of mortality, and the cutoff point was 48.6 mg/dl. They also found that venous pH, HCO₃, CO₂, base excess, and O₂ saturation did not affect mortality.²⁸ Our study demonstrated that biochemical parameters before and during ECMO were more prevalent in predicting hospital mortality than clinical, pathological, and demographic factors. Moreover, timely ECMO initiation in a well-prepared PCICU yielded comparable results to ECMO initiation in the OR.

Study limitations

The study is limited by the retrospective design with its known biases, and the difference between groups could be related to patient-specific variables or even to unmeasured variables. However, ECMO is not a frequent procedure, and the study duration is over 10 years; therefore, prospective randomized studies on ECMO patients are difficult. The study is also limited by the single-center experience. The characteristics of our patients and the center facilities could make our results not generalizable to all centers.

Conclusions

Poscardiotomy ECMO support in congenital cardiac lesions is still associated with high mortality rates. Mortality in patients with ECMO initiation in the OR was comparable to those who had PCICU ECMO. Pre-ECMO low pH and high lactate during ECMO could predict hospital mortality.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Amr A Arafat

References

Barbaro

Paden

Guner

, et al. Pediatric extracorporeal life support organization registry international report 2016. ASAIO J 2017; 63: 456–463.

Cooper

Jacobs

Moore

, et al. Cardiac extracorporeal life support: State of the art in 2007. Cardiol Young 2007; 17(Suppl 2): 104–115.

Kang

Lee

Han

. Predictors of mortality in patients with VA-extracorporeal membrane oxygenation. Eur Heart J 2020; 41: 946–1237.

Lorusso

Alexander

Rycus

. The extracorporeal life support organization registry: Update and perspectives. Ann Cardiothorac Surg 2019; 8: 93–98.

Ergün

Yildiz

Güneş

, et al. Use of extracorporeal membrane oxygenation in postcardiotomy pediatric patients: Parameters affecting survival. Perfusion 2020; 35: 608–620.

Lorusso

Raffa

Kowalewski

, et al. Structured review of post-cardiotomy extracorporeal membrane oxygenation: Part 2-pediatric patients. J Hear ung Transplant 2019; 38: 1144–1161.

Elmahrouk

Ismail

Hamouda

, et al. Extracorporeal membrane oxygenation in postcardiotomy pediatric patients--15 years of experience outside Europe and North America. Thorac Cardiovasc Surg 2019; 67: 28–36.

Misfeldt

Kirsch

Goldberg

, et al. Outcomes of single-ventricle patients supported with extracorporeal membrane oxygenation. Pediatr Crit Care Med 2016; 17: 194–202.

Chaturvedi

Macrae

Brown

, et al. Cardiac ECMO for biventricular hearts after paediatric open heart surgery. Heart 2004; 90: 545–551.

10.

Alsoufi

Awan

Manlhiot

, et al. Results of rapid-response extracorporeal cardiopulmonary resuscitation in children with refractory cardiac arrest following cardiac surgery. Eur J Cardiothorac Surg 2014; 45: 268–275.

11.

Pizarro

Davis

Healy

, et al.

Is there a role for extracorporeal life support after stage I Norwood?

Eur J Cardiothorac Surg 2001; 19: 294–301.

12.

Mascio

Austin

3rd Jacobs

, et al. Perioperative mechanical circulatory support in children: an analysis of the society of thoracic surgeons congenital heart surgery database. J Thorac Cardiovasc Surg 2014; 147: 658–664.

13.

Extracorporeal Life Support Organization . ELSO registry. Ann Arbor, MI, USA: Extracorporeal Life Support Organization 2022 (Accessed 3 Feb).

14.

Morris

Ittenbach

Godinez

, et al. Risk factors for mortality in 137 pediatric cardiac intensive care unit patients managed with extracorporeal membrane oxygenation. Crit Care Med 2004; 32: 1061–1069.

15.

Alsoufi

Awan

Manlhiot

, et al.

Does single ventricle physiology affect survival of children requiring extracorporeal membrane oxygenation support following cardiac surgery?

World J Pediatr Congenit Heart Surg 2014; 5: 7–15.

16.

Alsoufi

Manlhiot

Mahle

, et al. Low-weight infants are at increased mortality risk after palliative or corrective cardiac surgery. J Thorac Cardiovasc Surg 2014; 148: 2508–2514.

17.

Booth

Roth

Thiagarajan

, et al. Extracorporeal membrane oxygenation support of the Fontan and bidirectional Glenn circulations. Ann Thorac Surg 2004; 77: 1341–1348.

18.

Ugaki

Kasahara

Kotani

, et al. Extracorporeal membrane oxygenation following Norwood stage 1 procedures at a single institution. Artif Organs 2010; 34: 898–903.

19.

Huang

S-C

E-T

Chen

Y-S

, et al. Experience with extracorporeal life support in pediatric patients after cardiac surgery. ASAIO J 2005; 51: 517–521.

20.

Sasson

Cohen

Tamir

, et al. Extracorporeal membrane oxygenation in pediatric patients: Our experience in the last 10 years. Isr Med Assoc J 2013; 15: 13–16.

21.

Black

Coles

Williams

, et al. Determinants of success in pediatric cardiac patients undergoing extracorporeal membrane oxygenation. Ann Thorac Surg 1995; 60: 133–138.

22.

Walters

3rd Hakimi

Rice

, et al. Pediatric cardiac surgical ECMO: Multivariate analysis of risk factors for hospital death. Ann Thorac Surg 1995; 60: 329–336.

23.

Dalton

Siewers

Fuhrman

, et al. Extracorporeal membrane oxygenation for cardiac rescue in children with severe myocardial dysfunction. Crit Care Med 1993; 21: 1020–1028.

24.

Mistry

Trucco

Maul

, et al. Predictors of poor outcomes in pediatric extracorporeal membrane oxygenation. World J Pediatr Congenit Heart Surg 2018; 9: 297–304.

25.

Howard

Kalish

Wigmore

, et al. Association of Extracorporeal membrane oxygenation support adequacy and residual lesions with outcomes in neonates supported after cardiac surgery. Pediatr Crit are Med 2016; 17: 1045–1054.

26.

Tantway

Arafat

Albabtain

, et al. Sepsis in postcardiotomy cardiogenic shock patients supported with veno- arterial extracorporeal membrane oxygenation. Int J Artif Organs 2023; 46: 153–161.

27.

Polimenakos

Wojtyla

Smith

, et al. Post-cardiotomy extracorporeal cardiopulmonary resuscitation in neonates with complex single ventricle: Analysis of outcomes. Eur J Cardiothorac Surg 2011; 40(6): 1396–1405.

28.

Chen

T-Y

Chang

C-H

Hsu

J-Y

, et al. Comparison of the predictive ability of lactate and central venous blood gas in pediatric mode extracorporeal membrane oxygenation outcome. Pediatr Neonatol 2022; 63: 474–483.