Amniotic fluid embolism is a rare and potentially catastrophic, but poorly understood condition that is unique to pregnancy. It may range from a relatively minor subclinical episode through to one which is rapidly fatal.
Despite being first reported over 70 years ago, the underlying pathophysiology of amniotic fluid embolism (AFE) is still not well understood. Even the name AFE has been criticised as not being representative of the underlying condition, with suggested alternatives being ‘anaphylactoid syndrome of pregnancy’ or ‘sudden obstetric collapse syndrome’.
The incidence of AFE in Australia has recently been reported as 6.1 per 100 000 deliveries (95 per cent CI 5.2–6.9) with a case fatality rate of 14 per cent.1 Despite its rarity, AFE is a major contributor to maternal mortality in developed countries. It is currently the leading cause of direct maternal death in both Australia and New Zealand, while in the UK triennial reports it is consistently in the top four.
Traditionally, AFE was associated with a very high morbidity and mortality rate. Early published data from Clark2 and Morgan3 documented a mortality of between 61 and 85 per cent, with poor outcomes in those women who did survive. However, more recent data would suggest the mortality is much lower than this, ranging from 11–43 per cent (see Table 1). In addition, the outcomes in women who do survive the initial insult also appear encouraging. This improvement in morbidity and mortality is likely to be secondary to a number of factors, including greater awareness of the condition (such that less severe cases are now reported), developments in resuscitation and intensive care, as well as multidisciplinary training for the management of the collapsed obstetric patient.
The outcomes for the neonate, if the AFE occurs in utero, are unfortunately not as positive compared to the mother. Fetal distress may be an initial presenting sign of AFE and, if an in utero reaction does occur, the neonatal mortality may be up to 40 per cent.1
The pathophysiology of AFE is poorly understood and no reliable animal model exists for AFE. For this reason, most of the theories surrounding the pathophysiology of AFE are derived from clinical cases. The term ‘embolism’ itself is a common source of confusion. While some of the typical features of an AFE may be similar to the mechanical obstruction of pulmonary blood flow seen with a traditional pulmonary embolism, a number of other features do not fit with this model. For this reason possible immune-mediated mechanisms have also been suggested.4
Despite the lack of understanding as to why amniotic fluid can trigger a reaction, what is well accepted is that amniotic fluid must first enter the maternal circulation. Amniotic fluid is normally separated from the maternal circulation by the intact fetal membranes and AFE is thought to occur when there is a breach in this barrier. A breach may occur at a number of sites, including the endocervical veins, the area of placentation and sites of uterine trauma. Conditions that increase intrauterine pressure may contribute to the passage of amniotic fluid into the maternal circulation (for example, augmentation of labour with oxytocin) and go some way to explaining some of the identified risk factors for AFE.
However, potentially confusing the current understanding of AFE is the observation that amniotic fluid may enter the maternal circulation as a relatively normal aspect of childbirth and not trigger any adverse reaction. Evidence supporting this includes studies whereby components of amniotic fluid have been found in the maternal circulation without any other evidence of an AFE reaction4, in addition to the failure of AFE to be reliably reproduced by the injection of amniotic fluid into the circulation in animal models. This would suggest that amniotic fluid only triggers an AFE in a small proportion of women.
It is unclear what components of amniotic fluid or meconium lead to the clinical syndrome that is seen with AFE. Amniotic fluid contains a number of potentially vasoactive substances as well as substances that may interfere with coagulation.5 In addition, there may be immune-mediated processes that contribute. This is supported by a number of observational findings, such as AFE being more common in women carrying a male fetus, as well as decreased complement levels (suggesting complement activation). An elevated mast cell tryptase, usually a finding in cases of anaphylaxis, is not routinely seen with AFE, suggesting that anaphylaxis does not play a significant role. However, it has been suggested that an anaphylactoid process (which is not an IgE-linked reaction) may contribute to AFE and hence one of the suggested alternative names.2
AFE is currently a leading cause of maternal mortality in the developed world and it cannot be either predicted or prevented. However, there would appear to be a significant improvement in the both the morbidity and mortality of AFE. Severe cases of AFE are likely to present with a combination of cardio-respiratory compromise and coagulopathy, although the exact mechanism for the presenting signs and symptoms is unclear. The management is essentially supportive following the principles of care for unwell obstetric patients. Coagulation abnormalities may be significant and expert assistance may be required. The greater availability of echocardiography may help guide haemodynamic therapy in severely unwell women. In Australia and New Zealand, AFE is the leading cause of direct maternal death, it is currently unclear why this is the case and further research, such as that being conducted by Australasian Maternity Outcomes Surveillance System, is required.
A large number of risk factors for AFE have been identified, including advanced maternal age, placenta praevia, placental abruption, operative delivery and induction of labour.6 However, given the rare and unpredictable nature of AFE, these risk factors are only useful for retrospective analysis. Identified risk factors should not currently be used to alter the clinical management of individual woman (for instance, avoidance of induction of labour or caesarean delivery) as the baseline risk is still very low.
Table 1. Incidence of AFE and case fatality rates in published series.
|Author||Year Published||Incidence (per 100 000 maternities)||Case fatality rate (%)|
Table 2. Signs and symptoms of AFE.
|Signs or symptoms||Frequency (%)|
|Pulmonary oedema or ARDS||93|
This data, while dated and reflecting the presentation of more severe cases of AFE, demonstrates the wide range of signs and symptoms of AFE. Adapted from Clark.2
The majority of episodes of AFE are reported to occur in the intrapartum and immediate postpartum period, although cases have been described during amniocentesis, termination of pregnancies, abdominal trauma and post-caesarean delivery. The classic description is of a sudden and severe deterioration that is rapidly fatal. With greater recognition of the syndrome, less-severe cases may present with more subtle signs and symptoms. Table 2 shows the wide range of features that AFE may present with and, while these data are comparatively dated and reflect more serious cases, it does highlight the significant cardio-respiratory involvement in the condition as well as the likelihood of a compromised fetus.
Hypotension is a common sign in severe episodes of AFE, although the exact mechanism is unclear and may vary between patients. Case reports have documented a variety of contributing factors, including severe pulmonary vasospasm and pulmonary hypertension; impaired left ventricular filling secondary to severe right ventricular failure; and myocardial ischaemia. It has also been suggested that either amniotic fluid or meconium may contain substances that have a direct myocardial depressant effect, or that a substance with direct myocardial depression is released as part of the episode.
Respiratory signs and symptoms may range from shortness of breath to hypoxia or respiratory arrest. The underlying mechanism for the hypoxia is likely to be multifactorial. Initially, severe ventilation and perfusion mismatching may occur; then, as the episode progresses, cardiogenic pulmonary oedema may then develop secondary to left ventricular failure. In addition, non-cardiogenic pulmonary oedema, secondary to capillary damage from amniotic fluid, may also develop.
Coagulation disturbances can occur rapidly in AFE and, in some cases, may be the initial presenting sign. Amniotic fluid contains both tissue factor that acts as a pro-coagulant as well as plasminogen activation inhibitor-1, which is involved in fibrinolysis. Thus, both a consumptive coagulopathy as well as massive fibrinolysis may occur.7
The diagnosis of AFE is a clinical one, based on the presence of cardiovascular and respiratory compromise and coagulopathy after the exclusion of other causes.4 There is currently no widely available diagnostic test that can be used in survivors. The diagnosis can be made at autopsy by examining for the presence of fetal material in the maternal pulmonary circulation. However, given that fetal material may be present in otherwise normal women, the sensitivity of this finding may not be as high as first thought.
The initial management of a suspected episode of AFE should be tailored to the severity of the presentation. In severe cases, management is essentially supportive, following the principles of basic and advanced life support, with modifications made for the pregnant or recently pregnant state (uterine displacement and early securing of the maternal airway).8 If the presentation occurs prior to delivery then consideration should be given to urgent delivery of the fetus. While this may limit hypoxic damage to the fetus, its main benefit is for the mother, with potentially improved cardiac output by the limitation of aortocaval compression as well decreasing oxygen consumption.
Maternal oxygenation is likely to be significantly compromised in severe cases such that intubation and positive pressure ventilation will be required. Given the potential difficulties associated with intubation in the pregnant patient as well as the compromised maternal state, expert assistance should be obtained where available. Haemodynamic support is likely to be required and echocardiography can be a useful modality to help guide appropriate therapy.7
Coagulation abnormalities can develop rapidly; the potential for major haemorrhage should be anticipated and coagulation studies performed as soon as possible. Point-of-care coagulation monitoring (for example with thromboelastography [TEG] or rotational thromboelastometry [ROTEM]) may allow more specific targeting of coagulation factors, compared with traditional tests such as the international normalised ratio (INR) and activated partial thromboplastin time (aPTT). In addition to conventional blood products, antifibrinolytics, such as tranexamic acid, may also be of benefit, while the use of recombinant activated factor VII has been potentially linked to poorer outcomes.9
A number of novel therapies have also been described in case reports for AFE. These include the use of cardio-pulmonary bypass, extracorporeal membrane oxygenation, inhaled nitric oxide and prostacycline and haemofiltration or plasma exchange. The consideration of such therapies will be dependant on the local expertise available as well as the clinical condition of the patient.
Future pregnancies in survivors of AFE
A common issue faced by survivors of AFE is whether it will recur in future pregnancies. This may be a significant source of anxiety when the next pregnancy does occur. Currently, there is no evidence to suggest these women are at a higher risk of recurrence, with no documented cases of women having suffered a subsequent AFE.
- Knight M, Berg C, Brocklehurst P, Kramer M, Lewis G, Oats J, et al. Amniotic fluid embolism incidence, risk factors and outcomes: a review and recommendations. BMC Pregnancy Childbirth. 2012;12:7.
- Clark SL, Hankins GD, Dudley DA, Dildy GA, Porter TF. Amniotic fluid embolism: analysis of the national registry. Am J Obstet Gynecol. 1995 Apr;172(4 Pt 1):1158-67.
- Morgan M. Amniotic fluid embolism. Anaesthesia. 1979
- Benson MD. Current concepts of immunology and diagnosis in amniotic fluid embolism. Clin Dev Immunol. 2012:946576.
- O’Shea A, Eappen S. Amniotic fluid embolism. Int Anesthesiol Clin. 2007;45(1):17-28.
- Abenhaim HA, Azoulay L, Kramer MS, Leduc L. Incidence and risk factors of amniotic fluid embolisms: a population-based study on
3 million births in the United States. Am J Obstet Gynecol. 2008
- Gist RS, Stafford IP, Leibowitz AB, Beilin Y. Amniotic fluid embolism. Anesth Analg. [Review]. 2009 May;108(5):1599-602.
- Soar J, Perkins GD, Abbas G, Alfonzo A, Barelli A, Bierens JJ, et al. European Resuscitation Council Guidelines for Resuscitation 2010 Section Cardiac arrest in special circumstances: Electrolyte abnormalities, poisoning, drowning, accidental hypothermia, hyperthermia, asthma, anaphylaxis, cardiac surgery, trauma,pregnancy, electrocution
Resuscitation. 2010 Oct;81(10):1400-33.
- Leighton BL, Wall MH, Lockhart EM, Phillips LE, Zatta AJ. Use of recombinant factor VIIa in patients with amniotic fluid embolism: a systematic review of case reports. Anesthesiology. 2011 Dec;115(6):1201-8.
- Knight M, Tuffnell D, Brocklehurst P, Spark P, Kurinczuk JJ. Incidence and risk factors for amniotic-fluid embolism. Obstet Gynecol. 2010 May;115(5):910-7.
- Tuffnell DJ. United Kingdom amniotic fluid embolism register. BJOG. 2005 Dec;112(12):1625-9.
- Gilbert WM, Danielsen B. Amniotic fluid embolism: decreased mortality in a population-based study. Obstet Gynecol. 1999 Jun;93(6):973-7.
- Burrows A, Khoo SK. The amniotic fluid embolism syndrome: 10 years’ experience at a major teaching hospital. Aust N Z J Obstet Gynaecol. 1995 Aug;35(3):245-50.