If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Cardio-Thoracic Surgery, Heart and Vascular Centre, Maastricht University Medical Centre (MUMC), Cardiovascular Research Institute Maastricht (CARIM), Maastricht, The Netherlands
Yong Loo Lin School of Medicine, National University of Singapore, SingaporeCardiothoracic ICU, Department of Cardiac Thoracic and Vascular Surgery, National University Heart Centre, National University Hospital, Singapore
Yong Loo Lin School of Medicine, National University of Singapore, SingaporeCardiothoracic ICU, Department of Cardiac Thoracic and Vascular Surgery, National University Heart Centre, National University Hospital, Singapore
The use of extracorporeal membrane oxygenation (ECMO) in the peripartum period is described infrequently.
•
There were more cardiovascular than respiratory indications for ECMO in our review.
•
Respiratory indications were associated with longer ECMO runs and higher mortality.
•
91% of mothers and 78% of fetuses survived, albeit with possible publication bias.
•
ECMO is a viable option in peripartum patients with severe cardiopulmonary failure.
Abstract
The use of extracorporeal membrane oxygenation (ECMO) as salvage therapy for patients with severe cardiopulmonary failure has increased significantly in the past decade. However, the use of ECMO in pregnant and peripartum patients has received scant attention. We performed a systematic review of case reports in the literature, documenting indications and outcomes of ECMO in pregnancy and postpartum patients. Case reports on ECMO use in pregnant and postpartum patients were retrieved from MEDLINE, EMBASE and SCOPUS databases up to December 2018. Ninety publications reporting on 97 patients met our inclusion criteria. The majority of publications reported peripartum or postpartum ECMO use for cardiovascular failure (60.8%), while the remainder had respiratory failure. Adult Respiratory Distress Syndrome (91.9%) was the most common respiratory indication while pulmonary embolism (23.7%) and peripartum cardiomyopathy (16.9%) accounted for the two most common cardiovascular indications. Hemorrhage was the most common complication of ECMO reported (31.9%). Of 96 documented neonatal outcomes, 80 neonates (83.3%) survived while 88 of 97 (90.7%) mothers survived. Extracorporeal membrane oxygenation appears to be a viable life support modality in pregnant and postpartum women with severe cardiopulmonary failure, but publication bias in our study cohort should be considered.
Extracorporeal membrane oxygenation (ECMO) has become an important tool in the care of adults and children with severe cardiac and pulmonary dysfunction which is resistant to conventional management. Extracorporeal membrane oxygenation is a form of modified cardiopulmonary bypass that provides artificial ventilatory and/or mechanical circulatory support to critically- ill patients in intensive care units (ICU). Venous blood is removed from the patient and pumped through an artificial membrane lung (the oxygenator), where carbon dioxide is removed and oxygen is added. The oxygenated blood is then returned to the patient. In veno-venous (V-V) ECMO configuration, the oxygenated blood is returned to the venous circulation, providing support of gas exchange only (Fig. 1). In comparison, the veno-arterial (V-A) ECMO configuration allows the oxygenated blood to be returned to the arterial circulation (Fig. 2), thereby providing circulatory support in addition to gas exchange.
Fig. 1Schematic representation of venovenous (V-V) ECMO. Blood is drained out of the femoral vein and goes through the centrifugal pump and oxygenator before it is returned to the jugular vein in patients with severe respiratory failure. Alternatively, blood can be returned to the contralateral femoral vein with femoro-femoral V-V ECMO. Dual-lumen single cannulas in the jugular vein can also be used for V-V ECMO
Fig. 2Schematic representation of veno-arterial (V-A) ECMO. Blood is drained out of the femoral vein, goes through the centrifugal pump and oxygenator before it is returned to the femoral artery in patients with severe cardiac failure
The rate of ICU admissions in peripartum patients ranges from 0.7 to 13.5 per 1000 deliveries, with 3.3% mortality in high-income countries and 14% in low- and middle-income countries.
Maternal mortality in pregnant or postpartum women who are admitted to ICU with severe Adult Respiratory Distress Syndrome (ARDS) or those who suffer cardiac arrest has been reported to be approximately 40%.
Heart failure rates among antepartum hospitalizations have been increasing by 5% annually in the United States, with an increased risk of maternal mortality and morbidity.
The use of ECMO in pregnancy and the postpartum period has been reported infrequently, since it was thought to increase the risk of maternal or fetal hemorrhage. The literature about ECMO during pregnancy is limited to case reports, case series and selective systematic reviews on pregnant patients with influenza A(H1N1) infections. It is estimated that up to 40% of maternal deaths are potentially preventable, so clinicians should be aware of salvage therapies available for severe cardiopulmonary failure during the peripartum period.
The aim of this study was to review peer-reviewed case reports documenting the indications and outcomes of ECMO in pregnant and postpartum patients, with analyses of maternal and fetal outcomes.
Materials and methods
This study was performed following waiver of institutional review board approval and was conducted in adherence to the ethical guidelines stated in the Declaration of Helsinki. A comprehensive literature search was obtained from three databases: MEDLINE, EMBASE and SCOPUS. These databases were searched from 1972 until November 2017, using an advanced strategy including the following MeSH terms and text words: Extracorporeal Membrane Oxygenation, ECMO, extracorporeal life support, extracorporeal membrane oxygenator support, extracorporeal membrane oxygenation device, extra-corporeal oxygenation, extra-corporeal membrane oxygenation, extra-corporeal life support, postpartum, peripartum, prenatal, postnatal, gestation, gestational and pregnancy (Supplementary Table ST1).
We followed Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines in our systematic review of the literature (Fig. 3). Our literature search produced 1901 unique articles for review. The articles were screened based on title and abstract, and the final selection was performed after full-text evaluation. The eligibility of studies was independently assessed by two review authors (JO and JZ) and disagreements were resolved by consensus or appeal to a third author (KR). Citations from selected articles were also screened and were included as hand-picked articles. We included all pregnant patients or patients up to a maximum of 15 days’ postpartum who received ECMO. Case reports where ECMO was on standby but not used were excluded. Larger case series (>3 peripartum patients) which had minimal details about individual patients were not included.
Fig. 3Preferred Reporting Items for Systematic Reviews and Meta-analyses flow chart
Data on maternal demographics and presenting indications as well as adjuvant therapies employed, type and duration of ECMO used, and fetal and maternal outcomes, were extracted. Outcomes were quantified based on survival, the presence of complications and, when possible, duration of hospital stay. This information was collated and analyzed accordingly.
The characteristics of the patients at baseline were reported as percentages for categorical variables and as mean with standard deviation for discrete variables. Categorical variables (ECMO indications, ECMO configuration) were compared with Chi-square or Fisher’s exact tests, as appropriate.
Univariate analysis was used to detect any statistically significant difference in the observed maternal survival associated with pre-ECMO adjuvant therapies. All the analyses were conducted at a two-sided alpha level of 5% using SPSS software, version 25, except for the univariate analysis of pre-ECMO adjuvant therapies, for which we used R software, version 3.3.3 (R Foundation).
Results
A total of 97 patients treated with ECMO during pregnancy or the early postpartum period were reported in 90 publications (Supplementary Table ST2) and were included in this review. The mean maternal age was 28.5 ± 5.8 years. We observed more case reports on cardiovascular indications (n=59, 61%) for ECMO compared with case reports on respiratory indications (n=37, 38%). The most common underlying etiology for respiratory ECMO was Adult Respiratory Distress Syndrome (ARDS) (n=34/37, 92%), mainly secondary to influenza A(H1N1). Pulmonary embolism (14/59, 24%) and peripartum cardiomyopathy (10/59, 17%) were the two main cardiovascular aetiologies requiring ECMO (e-Supplement, Table 1).
Table 1Summary of underlying etiologies
Underlying etiology
Pregnant (n=50)
Postpartum (n=47)
Total (n=97)
Respiratory insufficiency
25
12
37
Adult Respiratory Distress Syndrome (ARDS) with pneumonia
Analysis of the various pre-ECMO adjuvant therapies (Table 2) showed that several pregnant or post-partum patients received inhaled nitric oxide (iNO), prostaglandin infusions, vasopressors, high frequency oscillation ventilation, corticosteroids or cardiopulmonary resuscitation (CPR) prior to initiation of ECMO. All patients who received CPR prior to ECMO survived. Corticosteroid use prior to ECMO initiation was associated with higher maternal mortality on univariate analysis (odds ratio (OR) 5.71, 95% confidence interval (95% CI) 1.17 to 27.9, P=0.03)
Table 2Univariate analysis of pre-ECMO adjuvant therapies and maternal survival.
Adjuvant therapy
Category
Survival/Mortality (n)
Non-survivors (%)
P-value
Prostaglandin infusion
Yes No
8/1 79/8
11.1 9.1
0.851
Nitric Oxide
Yes No
10/2 77/7
16.7 8.3
0.364
Vasopressors
Yes No
30/1 56/8
3.2 12.5
0.179
Corticosteroids
Yes No
7/3 80/6
30.0 7.0
0.031
HFOV
Yes No
5/0 82/9
0.0 9.9
0.999
CPR
Yes No
16/0 71/9
0.0 11.3
0.998
Proning
Yes No
4/0 83/9
0.0 9.8
0.999
Pacing/Anti-arrhythmic
Yes No
2/0 85/9
0.0 9.6
0.999
CPR: cardiopulmonary resuscitation. ECMO: extracorporeal membrane oxygenation. HFOV: High frequency Oscillation ventilation.
Hemorrhagic complications were the most frequently reported ECMO-related complication (n=31, 32%). Intracranial and multiple site bleeding were associated with mortality, while vaginal or upper gastro-intestinal bleeding were common causes of non-fatal bleeding. Other reported bleeding problems included disseminated intravascular coagulation, hemothorax, intracerebral hemorrhage, rectus sheath hematoma, renal/bladder hemorrhage and bleeding from the cannulation sites. Other ECMO-related complications such as nosocomial infections, atrial fibrillation, generalized myopathy and renal impairment were also reported.
Extracorporeal membrane oxygenation during the peripartum period was associated with survival rates of 91% (88/97) of mothers and 78% (39/50) of babies born to mothers who received ECMO (Table 3). The causes of maternal deaths included multi-organ failure,
Extracorporeal membrane oxygenation in severe acute respiratory failure in postpartum woman with rheumatic mitral valve disease: benefit, factors furthering the success of this procedure, and review of the literature.
Of the 16 neonatal deaths, 11 fetuses were exposed to ECMO in utero and five were not. Fetal mortality appeared to be slightly higher in the pregnant group compared with the postpartum group. Of the 50 case reports about pregnant patients that reported fetal outcomes, 11 fetal deaths occurred. Among 47 postpartum patients, there were five infant deaths.
Of the 11 fetuses on ECMO, five were electively terminated,
ECMO devant une hypoxémie réfractaire compliquant une grippe maligne et fulminante H1N1 chez une femme enceinte: poursuite ou interruption de la grossesse? Extracorporeal membrane oxygenation for refractory hypoxia secondary to a severe viral pneumonia due to influenza A (H1N1) in a pregnant woman: continuation or termination of pregnancy?.
Of the 11 fetal deaths associated with ECMO exposure, seven occurred among the 26 patients placed on V-A ECMO and three occurred among the 22 patients initiated on V-V ECMO during pregnancy. The ECMO configuration employed was not specified in one fetal death.
Our comparison of the indications and outcomes of ECMO in pregnant and postpartum patients revealed that 50 patients (52%) received ECMO antepartum and intrapartum, while the remaining 47 patients (48%) received ECMO up to 15 days postpartum. The mean age of patients in the pregnant group was 29.7 ± 5.45 years while it was 27.4 ± 5.89 years in the postpartum group. Mean gestational age at time of ECMO cannulation was 27.1 ± 6.4 weeks in the pregnant subgroup and 1.7 ± 4.1 days in the postpartum subgroup.
The number of pregnant patients who received ECMO therapy for respiratory (n=25, 50%) and cardiovascular (n=25, 50%) indications were comparable, while indications for ECMO therapy were predominantly cardiovascular in the postpartum group (n=35, 74%). Within the pregnant group, ARDS was the most common ECMO indication, making up half the underlying aetiologies. The four most common ECMO indications in the postpartum group were ARDS (n=9, 19%), pulmonary embolism (n=8, 17%), postpartum cardiomyopathy (n=6, 13%) and amniotic fluid embolism (n=6, 13%). Maternal mortality within our study population was similar between pregnant and postpartum ECMO patients. Of 50 pregnant patients who received ECMO, four died, whereas of 47 postpartum patients, five died (8% vs 11%, respectively).
The number of pregnant patients who received V-A or V-V ECMO was similar (Table 5). However, more postpartum patients received V-A than V-V ECMO. The mean ECMO duration was slightly longer in postpartum patients than pregnant patients. Patients with respiratory indications had longer ECMO duration than patients with cardiovascular indications in both groups (431 h vs 131 h) and a higher rate of complications (44/91 reported complications).
Table 5ECMO Variables in pregnant and postpartum ECMO patients
ECMO variables
Pregnant
Postpartum
Overall
V-A
26
32
58
V-V
22
15
37
Unknown ECMO configuration
2
0
0
Mean ECMO run time (h)
219.9 ± 197
300.0 ± 518
256.6 ± 390
Mean ECMO run time in patients with respiratory indication (h)
292.1 ± 203
742.1 ± 854
431.2 ± 540
Mean ECMO run time in patients with cardiovascular indication (h)
Our review of case reports of peripartum patients demonstrated that they were supported on ECMO for respiratory as well as cardiovascular indications. Adult respiratory distress syndrome was the main etiology among respiratory indications, while pulmonary embolism and peripartum cardiomyopathy were more common etiologies among the cardiovascular indications. Overall, cardiovascular indications for ECMO made up the majority of indications, which contrasts with the 2019 Extracorporeal Life Support Organization (ELSO) registry data on adult ECMO patients, in which ECMO was initiated more commonly for respiratory indications (respiratory vs cardiac indications, 59% vs 42%, respectively).
We observed that peripartum patients needing V-V ECMO had lower mortality than those receiving V-A ECMO, while those who needed ECMO for respiratory indications had higher mortality than those with cardiac indications. Patients with cardiac indications received V-V ECMO and vice versa in our cohort. We also observed that patients who had ECMO for respiratory indications had a longer mean duration of ECMO and a higher rate of complications, accounting for the higher mortality compared with those with cardiac indications. The Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries (MBRRACE-UK) Maternal Mortality report
published in 2016 identified the 2009–2010 influenza A(H1N1) pandemic as the single most significant issue in maternal critical care in the United Kingdom. While the majority of ICU admissions in this cohort were during the postpartum period, the report identified pneumonia as the most common cause of ICU admission in pregnant women. Current guidance for patients with ARDS includes lung protective ventilation, consideration of infusion of neuromuscular blockade, prone positioning (difficult in pregnancy and the peripartum period) and early consideration of ECMO.
The MBRRACE-UK report also recommended early ECMO referral for pregnant women and new mothers in severe respiratory failure who had failed conventional ventilation.
reported cases of ECMO used exclusively during pregnancy. The authors conducted a comprehensive literature review of ECMO use during pregnancy between 1991 and 2007, and observed that overall maternal survival was 78% (35/45) and fetal survival was 65% (28/43). In contrast, our study consisted of a comprehensive literature search of all of the reported case reports of ECMO during pregnancy and the postpartum period, and demonstrated maternal survival of 91% and fetal survival of 79%. A single-center observational study,
including both pregnant and postpartum patients, also reported maternal survival of 80% and fetal survival of 70%.
With regards to adjuvant therapies employed before ECMO cannulation, corticosteroid use was associated with higher maternal mortality. Although only 10 patients in our cohort received steroids, their use as an adjuvant therapy has long been controversial, particularly with respect to the dosage and timing of administration. Improvements in oxygenation have been reported with the use of low- to moderate-dose methylprednisolone (<2.5 mg/kg/day) but other studies have shown no benefit or possible harm.
Recent studies on steroids for community-acquired pneumonia document a potential benefit, by reducing the median time to clinical stability and to treatment failure.
The prevalence of obstetric bleeding raises a relevant concern with regards to anticoagulation practices and bleeding diatheses in ECMO patients. The level of anticoagulation must be judicially adjusted considering all complex factors that may affect it, such as the severity of illness, underlying pathology and circuit components. While using anticoagulants is standard practice in ECMO patients, they can be discontinued for patients with significant bleeding diatheses. There are guidelines issued by ELSO to help guide physicians manage the use of anticoagulants during ECMO.
Our study is limited by publication bias. It is likely that published case reports are biased toward better outcomes, given that centers are less inclined to publish poorer outcomes. In addition, the small cohort of patients and study heterogeneity (pregnancy vs postpartum) across different case reports and case series are such that comprehensive statistical analysis was not feasible. Another limitation is the lack of a control group to compare maternal and fetal outcomes in the presence and absence of ECMO. However, the review throws insight into the indications and complications of ECMO in pregnant and postpartum females, which may be valuable to further investigation.
In summary, this literature review of case reports confirms the feasibility of ECMO during late pregnancy and the early postpartum period, and demonstrates the potential for good maternal and fetal outcomes, even in mothers with advanced respiratory or circulatory failure. The risk of peripartum bleeding during ECMO therapy should not be considered a contraindication to this strategy. Patients with respiratory indications in this study had longer duration on ECMO and poorer outcomes compared to patients with cardiovascular indications. There was no difference in outcomes between pregnant and postpartum patients. The expanded use of ECMO in critically-ill obstetric populations, such as those with ARDS and pulmonary hypertension, as well as early ECMO intervention, may result in improved maternal and fetal outcomes and warrants further investigation.
Declarations of interests and funding
The authors declare that there are no conflicts of interests. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
Extracorporeal membrane oxygenation in severe acute respiratory failure in postpartum woman with rheumatic mitral valve disease: benefit, factors furthering the success of this procedure, and review of the literature.
ECMO devant une hypoxémie réfractaire compliquant une grippe maligne et fulminante H1N1 chez une femme enceinte: poursuite ou interruption de la grossesse? Extracorporeal membrane oxygenation for refractory hypoxia secondary to a severe viral pneumonia due to influenza A (H1N1) in a pregnant woman: continuation or termination of pregnancy?.
30053-4&id=gr1.jpg){kind=link}
30053-4&id=gr2.jpg){kind=link}
30053-4&id=gr3.jpg){kind=link}