Prognostic factors of extracorporeal membrane oxygenation as a rescue therapy for refractory septic shock in children

IntroductionOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

Severe septic shock remains important causes of death in children worldwide (1). The key pathophysiological mechanisms are impaired oxygen delivery and utilization due to circulatory failure and microcirculatory and mitochondrial distress syndrome (MMDS), which ultimately lead to multiple organ dysfunction syndrome (MODS) and even death. Once refractory septic shock (RSS), characterized by resistance to fluid resuscitation and vasoactive drugs, the body can undergo severe sepsis-induced myocardial depression and MMDS. Extracorporeal membrane oxygenation (ECMO) can support the cardiovascular system and provide sufficient oxygen delivery, thus potentially preventing or even reversing the progression of RSS and improving the prognosis of RSS patients. Since the 1990s, ECMO support for children and neonates with RSS has yielded successful results. Since 2002, ECMO has gradually been incorporated into the Surviving Sepsis Campaign International Guidelines and recommended as a rescue therapy for RSS. Whether adults or children, the timing and risk factors prognosis of ECMO intervention remain unclear because the researches are limited in the number of samples, and homogeneity and high-quality prospective studies are deficient. This multicenter retrospective study aims to explore the risk factors of prognosis of ECMO, which is recommended as the rescue therapy for pediatric RSS, and investigate the optimal timing of clinical intervention to provide a basis for future prospective studies and guide clinical practice. We present this article in accordance with the STROBE reporting checklist (available at https://pm.amegroups.com/article/view/10.21037/pm-24-21/rc).

MethodsOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

Study population

Children who were admitted consecutively to eight major pediatric ECMO centers in Chinese mainland and accepted ECMO support for RSS from January 1, 2018 to December 31, 2020, were enrolled.

The inclusion criteria were as follows (2): (I) children between 0–18 years old; (II) satisfying both: (i) ineffective fluid resuscitation despite the administration of 40–60 mL/kg of normal saline or balanced salt solution within the first hour of diagnosis of septic shock; (ii) resistance to vasoactive drugs (epinephrine at 0.05–0.3 µg/kg/min or dopamine at 5–9 µg/kg/min in cold shock, or norepinephrine at ≥0.05 µg/kg/min or dopamine at ≥10 µg/kg/min in warm shock); and (iii) meeting one of the following criteria: mean arterial pressure (MAP)-central venous pressure (CVP) < (55 + 1.5 × age), central venous oxygen saturation (ScvO₂) <70%, or cardiac index <3.3.

The exclusion criteria were as follows (3): absolute contraindications according to the 2017 Extracorporeal Life Support Organization (ELSO) guidelines: gestational age <30 weeks or weight <1 kg, lethal chromosomal abnormalities (such as trisomy 13 or 18), uncontrolled bleeding, irreversible brain injury, or end-stage chronic and malignant diseases. And those children whose parents refuse ECMO or give up ECMO treatment within 24 hours. The prognosis of those whose parents gave up midway was considered as death.

MODS in children was defined according to the 2005 International Pediatric Sepsis Consensus Conference (IPSCC) (4). MODS in this study is defined as organ dysfunction other than the cardiovascular system for all the patients in this study on ECMO support were due to cardiovascular system failure.

Methodology

This study was a multicenter retrospective observational study that did not involve clinical interventions.

Data collection

Method

A survey form was distributed to the directors of eight pediatric ECMO centers via email. The form required validation those directors.

Survey form design

The survey form was formulated by discussions among the ECMO team members of eight centers involved in the research and covered the following aspects: (I) demographic data of the patients; (II) infection characteristics and organ function, including site of infection, pathogenic microorganisms, MODS status, Pediatric Risk of Mortality (PRISM) III score, etc.; (III) baseline treatment prior to ECMO implementation, including fluid resuscitation, invasive mechanical ventilation (IMV), continuous blood purification (CBP), cardiopulmonary resuscitation (CPR), use of vasoactive drugs; (IV) laboratory parameters, including blood gas analysis, blood biochemical and key circulatory perfusion indicators such as ScvO₂, blood lactate value, and 6-hour lactate clearance rate after RSS diagnosis ( pre-ECMO lactate is defined as 6h lactate value); (V) ECMO-related indicators, including the time interval from RSS diagnosis to ECMO implementation (defined as the time from persistent hypotension under two high-dose vasopressors to the start of ECMO), ECMO blood flow rate, and ECMO complications; (VI) prognosis: survival to hospital discharge or death; (VII) grouping: survival group and non-survival group and (VIII) time points of data collection: except for ECMO related indicators, other data were collected immediately upon the point of diagnosis of RSS. To monitor the 6-hour lactate clearance rate, lactate value of 6 hours after RSS diagnosis was also collected.

RSS homogeneous treatment

Basic treatment plan

All children enrolled accepted bundle therapy according to the pediatric section of the Surviving Sepsis Campaign Guidelines 2012 (2), which includes fluid resuscitation (at least 40–60 mL/kg fluids within the first hour) of RSS, anti-infective therapy (broad-spectrum antibiotics and appropriate antiviral or antifungal therapy), actively identifying the pathogen of infection (including two site blood cultures), and vasoactive drugs, along with auxiliary therapies including IMV, and CBP, and etc.

ECMO treatment

All patients received peripheral venoarterial (VA)-ECMO. The initial blood flow rate was set according to the patient’s condition to maintain arterial partial pressure of oxygen (PaO₂) above 80–100 mmHg, ScvO₂ between 65–75%, and a gradual decrease in lactate value (5,6). Positive inotropic agents or vasopressors could be used as needed. Heparin was continuously infused for anticoagulation at a rate of 10–50 U/kg/hour to achieve the activated clotting time target range of 180–220 s (7). The hematocrit should be maintained at 35–40%, with a platelet count above 75×10⁹/L (8). Improvement of vital signs, blood gas parameters, and hemodynamic stability (9) are the indications to withdraw ECMO.

Statistical analysis

All statistical tests were two-tailed, and P<0.05 was considered statistically significant. For quantitative variables, the mean and standard deviation were used to describe normally distributed data, while the median and interquartile range were used for nonnormally distributed data. Categorical variables were described using counts and percentages. Paired t-tests or Wilcoxon signed-rank tests were used to compare group differences for continuous data, while Chi-squared tests or Fisher’s exact tests were used for categorical data. The study focused on RSS children treated with ECMO and analyzed the relationship between baseline parameters before ECMO cannulation and the clinical outcomes of these patients (survival or non-survival). After screening for statistically significant indicators through univariate analysis, multiple logistic regression analysis was conducted with non-survival as the dependent variable and relevant prognostic factors as independent variables. The impact of each variable on the outcome was expressed as an odds ratio (OR) and a 95% confidence interval (CI).

Ethical considerations

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional ethics board of Children’s Hospital of Fudan University [No. (2020) 3171] and individual consent for this retrospective analysis was waived.

ResultsOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

General characteristics

From January 2018 to December 2020, a total of 41 children with RSS were on ECMO support. No one dropped out during the study, and all data were collected completely. Among them, there were 25 males (61.0%) with a median age of 35.5 (IQR, 2, 169) months. The bloodstream (22, 53.7%), respiratory system (14, 34.1%), and central nervous system (4, 9.8%) were the top three infection sites. The main pathogens detected were Pseudomonas aeruginosa (16.3%), Acinetobacter baumannii (11.6%), and Klebsiella pneumoniae (9.3%). Thirty-eight children (92.7%) had MODS, and primary or secondary immunodeficiency was detected in 13 patients (31.7%), including 6 patients (14.6%) treated with immunosuppressive agents for connective tissue diseases, 4 patients (9.8%) with tumor radiotherapy and/or chemotherapy, and 3 (7.3%) with primary immune deficiency. The median vasoactive inotropic score (VIS) for all children was 100 (IQR, 80, 160), and the VIS of the survival and non-survival groups were 100.0 (IQR, 76.0, 150.0) and 110.0 (IQR, 87.5, 160.0), respectively. Before ECMO, all children received IMV, 26 patients (63.4%) received continuous venovenous hemodiafiltration (CVVHDF) treatment for acute kidney injury (AKI) and fluid overload, of which 5 ones (12.2%) also received therapeutic plasma exchange (TPE) treatment in combination for inflammatory mediators removing, and 12 patients (29.3%) received CPR.

Comparison of baseline information between the survival and non-survival groups

Among the 41 pediatric patients, 21 withdrew ECMO successfully and survived to discharge with the survival rate of 51.2%, while 20 died due to the inability to maintain circulation during ECMO treatment.

Baseline conditions before ECMO

There were no differences in baseline characteristics, including sex, age, body mass index, immune status, pathogens and MODS situation before ECMO between the two groups (Table 1).

The severity of illness and organ function

In the non-survival group, the 6-hour lactate clearance rate (P=0.002) and the MAP-CVP (P=0.02) were lower; the 6-hour lactate value was higher. In the survival group, the 6-hour lactate clearance rate was −20% (i.e., lactate increased by 20%), and the median 6-hour lactate value was 5.2 (IQR, 3.8, 7.5) mmol/L, while in the non-survival group, these values were −60% (i.e., lactate increased by 60%) and 11.2 (IQR, 5.7, 19.5) mmol/L, respectively. There were no differences in the immediate lactate value at the point of RSS diagnosis, PRISM III severity score, VIS, ScvO₂, left ventricle ejection fraction (EF), and other values (Table 2).

Treatment

The children with RSS who were on ECMO support earlier had a higher survival discharge rate (P=0.03). The median time interval from RSS diagnosis to ECMO initiation in the survival group was 6.0 (IQR, 5.0, 12.0) hours, compared to 13.8 (IQR, 9.5, 19.0) hours in the non-survival group. The children in the non-survival group accepted more CPR treatment before ECMO (45.0% vs. 9.5%, P=0.02). There was no significant statistical difference in terms of other ECMO-related parameters, including ECMO maximum blood flow and complications, and other conventional treatments (Table 3).

Multivariate analysis

Multifactorial analysis showed that lactate levels before ECMO initiation and the time interval from diagnosis of RSS to ECMO initiation were significant prognosis predictive factors of children with RSS on ECMO support (Table 4). Further quartile grouping of the time interval was performed to compare the difference in mortality. There was no difference in the survival discharge rate between the time intervals <6 and 6–13.8 hours. Beyond 13.8 hours, mortality increased 7.17-fold, and beyond 20 hours, mortality increased 10.96-fold. CPR treatment and the 6-hour lactate clearance rate were not statistically different (Table 4, Figure 1).

Figure 1 Cumulative mortality rates during different RSS to ECMO intervals. ECMO, extracorporeal membrane oxygenation; RSS, refractory septic shock.

DiscussionOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

RSS remains one of the leading causes of child mortality worldwide. All Surviving Sepsis Campaign Guidelines, including pediatric chapters since 2002, propose that the main cause of death in septic shock is MODS caused by circulatory failure, and death mainly occurs in the first 48–72 hours of treatment. The key points to blocking the progression of septic shock to MODS are early identification, dealing with circulatory dysfunction, including cardiac, microcirculation and mitochondrial function, restoration of oxygen and substance delivery, and the use of ECMO as a rescue therapy for RSS children with ineffective fluid resuscitation and vasoactive drug resistance. Although ECMO is a supportive treatment by providing sufficient oxygen delivery, it “buys time” in the patient with septic shock to ensure organ perfusion until anti-infection and other treatments work.

Currently, ECMO as a rescue therapy has been widely recognized. But the optimal timing of ECMO intervention to block MMDS is still unclear in clinical work whether for adults or children. The difference in pediatric survival discharge rates is also staggering, with the range from 27% to 70% (10-12). Existing retrospective adult and pediatric studies show that the main prognostic indicators are age, infectious pathogens, CPR before ECMO, VIS, lactate levels, ScvO₂, and the time interval from RSS diagnosis to ECMO initiation. Some authors have also suggested that the proficiency level of the ECMO teams and the running of a high ECMO flow rate (>150 mL/kg) are important factors. However, almost all of these studies are single-center retrospective studies, with relatively few patients enrolled, and no consistent quantitative indicators have been obtained (11,13). Research on the timing of ECMO intervention is almost nonexistent. Our study is the first retrospective multicenter study to identify the prognostic factors of ECMO in children with RSS, particularly in obtaining quantifiable evidence for the optimal timing of ECMO intervention for future prospective studies.

The results of our study showed that the survival discharge rate of children with RSS supported on ECMO was 51.2%, which was consistent with those of international reports. Furthermore, it was found that high 6-hour lactate levels (11.2 mmol/L) and 6-hour lactate levels increased by 60% strongly predicted high mortality rates, showing that lactate level and dynamic lactate monitoring may be a good indicator for microcirculatory perfusion and mitochondrial function and clinical ECMO decision-making in RSS children. Lee et al. (14) showed that both the 6-hour lactate level and initial lactate level were associated with the 30-day mortality rate in adult patients with septic shock and that the 6-hour lactate level was a better predictor than the initial lactate level. When predicting the 30-day mortality rate, a 6-hour lactate level of ≥3.5 mmol/L had a sensitivity of 60.8% and a specificity of 74.4%. Rambaud et al. (15) also stated that lactate value that did not decrease within 6 hours of maximum conventional treatment was an indication for septic shock patients to transfer to ECMO referral centers. In this study, only 3 patients whose time intervals of RSS to ECMO were less than 6 hours, 3, 4, and 5 hours, respectively. So, we can consider the 6-hour lactate value as the pre-ECMO value, which may guide pediatricians in making ECMO decisions.

More than half of the deaths of children with severe septic shock occur within the first 24 hours of pediatric intensive care unit (PICU) admission (16). Therefore, it is vital to study the optimal time interval from RSS diagnosis to ECMO initiation. But the results of existing studies vary greatly. Cheng et al. (17) found that the optimal timing interval of ECMO support on adult patients with septic shock was within 3.8 days of admission (sensitivity 77.5%, specificity 70.7%). Bréchot et al. (18) reported that the optimal interval was 24 (3–108 hours) after onset. Aubron et al. (19) proposed that a time interval of 96 hours or less from the diagnosis of septic shock to ECMO support resulted in better outcomes. Another article showed that the best time for ECMO support in pediatric patients was 19.5 (16.5–30.5 hours) after RSS diagnosis (13). However, these studies did not precisely define the “time interval from RSS to ECMO”. Park et al. (20) defined it as the time from the start of vasopressor application to ECMO initiation and found that among 32 adult patients with RSS who accepted VA-ECMO treatment, the survival rate was 21.9%. In the survival group, the time interval was 21.1 hours, while in the 25 non-survival patients, it was 24.9 hours, with no survivor with an interval beyond 30.5 hours. Other studies have suggested using the start time of mechanical ventilation as the time point RSS diagnosis. Han et al. (21) found that the median time interval was 29.5 (20–46 hours). Early ECMO support increased the survival discharge rate of children with RSS from 14.3% to 57.2%. The study also pointed out that patients with longer septic shock duration before ECMO had significantly decreased chances of survival. In our study, we defined the time interval from RSS to ECMO as the time from persistent hypotension under two high-dose vasopressors to the ECMO initiation. The results showed that the median time interval was 6.0 (IQR, 5.0, 12.0) hours in the survival group and 13.8 (IQR, 9.5, 19.0) hours in the non-survival group (P=0.03). If the time interval exceeded 13.8 hours, the mortality rate increased by 7.17 times, and when it exceeded 20 hours, the mortality rate increased by 10.96 times. Study on adults yielded similar results, with a median interval from RSS to ECMO of 5 hours in the survival group of 5 patients and 18 hours in the non-survival group of 18 patients (P=0.04) (22). We also studied the duration of mechanical ventilation before ECMO and found no difference between the survival group and the death group.

Septic shock is primarily distributive shock, but in older children and adults, it often presents as peripheral vasodilation, while infants are more prone to sepsis-related cardiomyopathy (23) and VA-ECMO should theoretically have a better therapeutic effect. Skinner et al. (24) reported that the survival discharge rate of neonatal, children aged 1 month to 12 years and those above 12 years was 73%, 40% and 32%, respectively. In our study, ECMO patients were grouped according to age (newborn, 1 month to 12 years, and over 12 years), but no significant difference in the survival discharge rates was found among the different age groups, which might be due to the heterogeneity of EF, VIS, and other indicators in children of different ages. Septic shock itself is a dynamic and complex process, often with multiple types of shock coexisting; thus, clinical prognosis cannot be simply determined by age. Dynamic monitoring of hemodynamics and continuous evaluation of myocardial damage are necessary for better prognosis prediction.

CPR before ECMO has been confirmed as an important predictor of non-survival outcomes in several adult studies (20,25), and some even suggest that ECMO should be prohibited in RSS-related cardiac arrest patients. However, there is still controversy over whether CPR before ECMO is a predictor for RSS children. A single-center study by Solé et al. (13) that lasted for 15 years and included 21 ECMO patients found that CPR was not a predictive factor for prognosis. However, a multicenter study by Schlapbach et al. (26) that included 80 ECMO patients found that the absence of cardiac arrest before ECMO was an important positive prognostic factor. Regarding pre-ECMO CPR and the mortality in our study, the univariate and multivariate analyses yielded opposite directions. This might be the confounding factors and small sample size cannot explain this paradoxical stat result. There are reports that venovenous (VV)-ECMO is effective in RSS support, especially with high blood flow. However, our study did not investigate VV-ECMO. VV-ECMO for RSS children may have a higher survival rate (79% vs. 64%) with lower neurological complications and more convenient operation than VA mode (24). Due to the lack of pediatric venous catheters in Chinese mainland, all patients in this study were treated with VA-ECMO. Further research can be conducted with a future update of consumables.

This study still has some limitations: (I) the sample size of this study is not optimal, which may have led to the inability to yield more prognostic results from the multifactorial analysis of surviving and non-surviving children; (II) the diagnosis of RSS and the initiation of ECMO are influenced by many factors, such as physician experience, institution resources, and parental cooperativeness, which may have caused some bias; (III) all patients had peripheral cannulation; thus, ECMO flow was limited, so the impact of flow on prognosis was not thoroughly discussed.

ConclusionsOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

The significant elevation of lactate levels at 6 hours after the diagnosis of RSS and high lactate level before ECMO suggest high mortality rate in RSS children who receive ECMO treatment. However, the survival rate is improved in children with an RSS diagnosis to ECMO implementation duration within 13.8 hours.

AcknowledgmentsOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

None.

FootnoteOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://pm.amegroups.com/article/view/10.21037/pm-24-21/rc

Data Sharing Statement: Available at https://pm.amegroups.com/article/view/10.21037/pm-24-21/dss

Peer Review File: Available at https://pm.amegroups.com/article/view/10.21037/pm-24-21/prf

Funding: This study was supported by the National Key R&D Program of China (Nos. 2021YFC2701800 and 2021YFC2701805), Science and Technology Innovation Program (No. 20Y11900600), and Shanghai Municipal Health System major supports discipline projects (No. 2023ZDFC0103).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://pm.amegroups.com/article/view/10.21037/pm-24-21/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by institutional ethics board of Children’s Hospital of Fudan University [No. (2020) 3171] and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

ReferencesOther Section

• Introduction
• Methods
• Results
• Discussion
• Conclusions
• Acknowledgments
• Footnote
• References

1. Weiss SL, Fitzgerald JC, Pappachan J, et al. Global epidemiology of pediatric severe sepsis: the sepsis prevalence, outcomes, and therapies study. Am J Respir Crit Care Med 2015;191:1147-57. [Crossref] [PubMed]
2. Davis AL, Carcillo JA, Aneja RK, et al. American College of Critical Care Medicine Clinical Practice Parameters for Hemodynamic Support of Pediatric and Neonatal Septic Shock. Crit Care Med 2017;45:1061-93. [Crossref] [PubMed]
3. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support. Extracorporeal Life Support Organization, Version 1.4. Ann Arbor, Michigan, USA: Extracorporeal Life Support Organization; 2017 [2021-06-06]. Available online: https://www.elso.org/ecmo-resources/elso-ecmo-guidelines.aspx
4. Goldstein B, Giroir B, Randolph A, et al. International pediatric sepsis consensus conference: definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit Care Med 2005;6:2-8. [Crossref] [PubMed]
5. Maratta C, Potera RM, van Leeuwen G, et al. Extracorporeal Life Support Organization (ELSO): 2020 Pediatric Respiratory ELSO Guideline. ASAIO J 2020;66:975-9. [Crossref] [PubMed]
6. Brown G, Moynihan KM, Deatrick KB, et al. Extracorporeal Life Support Organization (ELSO): Guidelines for Pediatric Cardiac Failure. ASAIO J 2021;67:463-75. [Crossref] [PubMed]
7. Bridges BC, Rannucci M, Lequie L. Anticoagulation and disorders of haemostasis. In: Brogan TV, Lequier L, Lorusso R, et al. editors. Extracorporeal Life Support: The ELSO Red Book. 1^st edition. Ann, Arbor, Michigan, USA: Extracorporeal Life Support Organization; 2017:93-103.
8. Lequier L, Annich G, Al-Ibrahim O, et al. ELSO anticoagulation guidelines. Ann Arbor, Michigan, USA: Extracorporeal Life Support Organization, 2014 [2021-08-10]. Available online: https://www.elso.org/ecmo-resources/elso-ecmo-guidelines.aspx
9. Fried JA, Masoumi A, Takeda K, et al. How I approach weaning from venoarterial ECMO. Crit Care 2020;24:307. [Crossref] [PubMed]
10. Maclaren G, Butt W, Best D, et al. Extracorporeal membrane oxygenation for refractory septic shock in children: one institution's experience. Pediatr Crit Care Med 2007;8:447-51. [Crossref] [PubMed]
11. MacLaren G, Butt W, Best D, et al. Central extracorporeal membrane oxygenation for refractory pediatric septic shock. Pediatr Crit Care Med 2011;12:133-6. [Crossref] [PubMed]
12. Tembo M, Harvey C, Duthie M, et al. Extracorporeal membrane oxygenation for refractory septic shock in children: One institution's experience. Pediatr Crit Care Med 2009;10:534-5. [Crossref] [PubMed]
13. Solé A, Jordan I, Bobillo S, et al. Venoarterial extracorporeal membrane oxygenation support for neonatal and pediatric refractory septic shock: more than 15 years of learning. Eur J Pediatr 2018;177:1191-200. [Crossref] [PubMed]
14. Lee SG, Song J, Park DW, et al. Prognostic value of lactate levels and lactate clearance in sepsis and septic shock with initial hyperlactatemia: A retrospective cohort study according to the Sepsis-3 definitions. Medicine (Baltimore) 2021;100:e24835. [Crossref] [PubMed]
15. Rambaud J, Guellec I, Léger PL, et al. Venoarterial extracorporeal membrane oxygenation support for neonatal and pediatric refractory septic shock. Indian J Crit Care Med 2015;19:600-5. [Crossref] [PubMed]
16. Cvetkovic M, Lutman D, Ramnarayan P, et al. Timing of death in children referred for intensive care with severe sepsis: implications for interventional studies. Pediatr Crit Care Med 2015;16:410-7. [Crossref] [PubMed]
17. Cheng A, Sun HY, Tsai MS, et al. Predictors of survival in adults undergoing extracorporeal membrane oxygenation with severe infections. J Thorac Cardiovasc Surg 2016;152:1526-1536.e1. [Crossref] [PubMed]
18. Bréchot N, Luyt CE, Schmidt M, et al. Venoarterial extracorporeal membrane oxygenation support for refractory cardiovascular dysfunction during severe bacterial septic shock. Crit Care Med 2013;41:1616-26. [Crossref] [PubMed]
19. Aubron C, Cheng AC, Pilcher D, et al. Factors associated with outcomes of patients on extracorporeal membrane oxygenation support: a 5-year cohort study. Crit Care 2013;17:R73. [Crossref] [PubMed]
20. Park TK, Yang JH, Jeon K, et al. Extracorporeal membrane oxygenation for refractory septic shock in adults. Eur J Cardiothorac Surg 2015;47:e68-74. [Crossref] [PubMed]
21. Han L, Zhang Y, Zhang Y, et al. Risk factors for refractory septic shock treated with VA ECMO. Ann Transl Med 2019;7:476. [Crossref] [PubMed]
22. Yeh YC, Lee CT, Wang CH, et al. Investigation of microcirculation in patients with venoarterial extracorporeal membrane oxygenation life support. Crit Care 2018;22:200. [Crossref] [PubMed]
23. Inwald DP, Tasker RC, Peters MJ, et al. Emergency management of children with severe sepsis in the United Kingdom: the results of the Paediatric Intensive Care Society sepsis audit. Arch Dis Child 2009;94:348-53. [Crossref] [PubMed]
24. Skinner SC, Iocono JA, Ballard HO, et al. Improved survival in venovenous vs venoarterial extracorporeal membrane oxygenation for pediatric noncardiac sepsis patients: a study of the Extracorporeal Life Support Organization registry. J Pediatr Surg 2012;47:63-7. [Crossref] [PubMed]
25. Cheng A, Sun HY, Lee CW, et al. Survival of septic adults compared with nonseptic adults receiving extracorporeal membrane oxygenation for cardiopulmonary failure: a propensity-matched analysis. J Crit Care 2013;28:532.e1-10. [Crossref] [PubMed]
26. Schlapbach LJ, Chiletti R, Straney L, et al. Defining benefit threshold for extracorporeal membrane oxygenation in children with sepsis-a binational multicenter cohort study. Crit Care 2019;23:429. [Crossref] [PubMed]