Pathophysiology of Amniotic Fluid Embolism: new considerations and remarks for syndrome diagnosis
Bouchè C. MD1; Casarotto M, MD2; Wiesenfeld U. MD1; Bussani R. MD3; Addobati R.4; Bogatti P. MD1
1 Dipartimento Ostetricia e Ginecologia Istituto per l’Infanzia Burlo Garofolo; Clinica Ostetrica e Ginecologica Università di Trieste.
2 Dipartimento Materno-Infantile Ospedale S. Maria degli Angeli Pordenone; Divisione di Ostetricia e Ginecologia.
3 Dipartimento di Patologia e Medicina Legale; Istituto di Anatomia ed Istologia Patologica Università di Trieste.
4 Dipartimento di Medicina di Laboratorio Istituto per l’Infanzia Burlo Garofolo Trieste.
Amniotic fluid embolism is a little known syndrome having a rather complex pathogenesis. This condition, even today, can be responsible for maternal death.
Dr Ricardo Meyer128 was the first to publish in 1926 some cases of pulmonary-caseous embolism in a Brazilian journal.
In 1941, Dr Paul Steiner and Dr Clarence Lushbaugh172, pathologists at the University of Chicago, found amniotic material in pulmonary vessels at the autopsy of women who had died due to peripartum or post-partum shock. They supposed that powerful uterine contractions, at times of tetanic character, can generate a mechanism whereby amniotic fluid is pushed into uterine veins during labor or delivery, probably due to uterine hyperkinesis.
Since then, many clinical cases and reviews have been published and several studies have been performed on animals in an attempt to reproduce the syndrome. Final results have not been promising and many key issues of the syndrome are still unclear.
After having carefully evaluated the majority of cases published and analyzed two cases from two different hospitals in our area, we have managed to give an actual pathophysiological interpretation of the syndrome by considering the sequence of clinical signs and symptoms and by interpreting the histological specimens.
This review is intended to clarify the pathophysiology mechanism that distinguishes the moderate embolic forms by the severe forms of AFE, which is still the subject of a controversial and open debate.
INCIDENCE, MATERNAL AND FETAL-NEONATAL MORTALITY AND MORBIDITY
The rate of maternal mortality ranges between 1/8000 and 1/80000 of all pregnancies.
Amniotic Fluid Embolism (AFE) is responsible for 10% of all maternal deaths in the USA, 13% in France and 16% in the United Kingdom. It is the first cause of maternal death during labor and in the immediate post-partum period.
Morgan133 attributed to AFE 86% of maternal deaths. Clark39 refers a 61% maternal mortality rate, with only 15% of patients that survived neurologically intact. More recently, Kramer106 has reported an incidence of 14.8 per 100,000 births and Abenhaim2 an incidence of 7.7 per 100,000 births, while maternal mortality has been reported to be 13% and 21.6% respectively.
The discrepancy of reported observations leads to one important consideration: this still enigmatic syndrome can manifest itself in sudden, subclinical, mild or in the classical severe forms reported in Literature. Therefore, disease incidence and even more so maternal mortality can vary widely if diagnosis is performed with accurate clinical criteria and the rigorous fulfillment of diagnostic procedures. Undoubtedly, there may be cases of false positives, but this report will clarify that the number may be lower than the the one accounted for until now.
In the following years, have been published some case reports whose diagnosis had been solely made by exclusion of other disorders which did not generally show any of the typical AFE syndrome signs or symptoms11,192,51,159. These are often less severe forms that manifest themselves only with a dysfunction affecting the hemocoagulative system.
Due to the difficulty in diagnosing these rarely fatal AFE forms with subtle symptoms, it is extremely hard to establish the real incidence of all cases, from severe to fatal to paucisymptomatic forms. Some believe that AFE ought to be classified as a pregnancy-related disorder, which includes only very severe cases, by setting highly strict inclusion criteria. Clark39 in the National American Registry and later Tuffnell185 in the United Kingdom Registry set out precise selection criteria for including a new case. These criteria classifies only the most severe
They are namely:
1. Acute hypotension or cardiac arrest.
2. Acute hypoxia defined as dyspnea, cyanosis or respiratory arrest.
3. Coagulopathy, defined as a laboratory evidence of intravascular consumption or fibrinolysis or severe clinical hemorrhage in the absence of other explanations.
4. Onset of the above symptoms during labor, cesarean section, or dilatation and evacuation or within 30 minutes post partum.
5. Absence of any other confounding significant condition or potential explanation for the signs or symptoms
The American Registry however does not include cases, even if mortal, of profuse hemorrhage secondary to acute coagulopathy in the absence of other specific listed symptoms.
Another factor that has led to the various reported incidence rates has been the different importance that was attributed to histopathological findings for diagnosis in the 1940s, 1950s and 1960s and the value these findings have today.
Additionally, some cases characterized by severe obstetric hemorrhage secondary to coagulopathy can favor the development of a hemorrhagic or hypovolemic shock, triggering in turn cardiac and respiratory symptoms, and some potentially life-threatening maternal diseases, such as placenta accreta, placental abruption and uterine rupture, have been found to be significantly associated with AFE.
AFE is a syndrome with a significant incidence of fetal or neonatal deaths or of permanent outcomes for them. When maternal symptoms manifest themselves during labor, they are almost often accompanied by acute fetal distress, which is often the onset symptom. In many cases, the fetus is already dead.
Clark39 reported a neonatal mortality of 21% and only half of the survivors were neurologically
The diagnosis of AFE is mainly based on clinical signs and symptoms, which develop during the evolution of the syndrome. Histology performed at autopsy or on samples of organs and tissues of survived women is useful to confirm the diagnosis.
The frequently reported symptoms of the classical forms of AFE are characterized by sudden cardiovascular collapse, in the course of labor or post-partum, which may be preceded by respiratory distress that rapidly leads to shock and coma. The clinical picture may be complicated by neurological manifestations, such as tonic-clonic seizures and rigor, which in some cases may precede the other signs or symptoms.
Half of the patients die within the first hour of onset of symptoms and those who survive the first phase usually develop significant coagulopathy. In at least 40% of cases, alteration of coagulation manifest itself through disseminated intravascular coagulation (DIC), generally of an acute type, with massive consumption of coagulation factors and genital hemorrhage. Development of coagulopathy is generally subsequent to cardiorespiratory symptoms. Nonetheless, aberration of one single district may prevail in the evolution of symptoms and their onset can be characterized by the presence of other signs and symptoms.
Peterson and Taylor stated that dyspnea and cyanosis may be the first symptom and sign of embolization in 50% of patients139. Morgan133 found that respiratory distress (51%) is the principal onset symptom of this disease, followed by hypotension (27%), coagulopathy (12%) and seizures (10%). In his presentation of data from the National US Registry, Clark39 reported on convulsive or convulsive-like conditions as being the initial symptom for 30% of patients, followed by dyspnea (27%), fetal distress (17%) and hypotension (13%).
Often, the acute phase of AFE syndrome is marked by elevated leukocytosis, supporting the proven anaphylactic reaction to fetal antigens in the amniotic
There are, however, some prodromic signs which can often precede the classical picture of this disease. They are:
1. restlessness or agitation122,154,123,3,133,146,85,191,59
2. change of behavior185,191
6. tachypnea or breathing alteration191
7. inability to answer questions52
8. mental confusion171,100
9. perception of bad taste29
These signs, or at least some of them, can certainly be related to a condition of maternal partial hypoxia and, since they are not specific, they can “a posteriori” aid in the clinical diagnosis of AFE. Some of these manifestations, such as shortness of breath, numbness and change of behavior, have been recently taken into account for AFE classification by the Confidential Enquiry into Maternal Death in the United Kingdom, but their appearance is not said to be relevant in early syndrome recognition.
These hypotheses confirm a possible prodromic phase of the syndrome and justify the fact that classical onset of symptoms occurs in the post-partum phase in a relevant number of cases, supporting the concept that the embolic event often does not coincide with the evident onset of symptoms.
The first important sign of the disease can be a condition of fetal distress. Hogberg and Joelsson85 remark that a condition of severe fetal distress is one of the first of a series of symptoms leading to the development of a frank AFE syndrome.
Barrows and Levy17,113 report of some cases where fetal asphyxia was the first sign.
Clark, Dashow, Martin and Perozzi34,47,122,138 remark this occurrence and underline the fetus’ particular sensitivity and response to a partial maternal hypoxic insult.
Johnson91 emphasizes the rapid changes of cardiotocographic readings in two AFE cases.
Fetal asphyxia is indeed a common sign of AFE cases during labor. This justifies the high incidence of operative deliveries described in case reports on this pathological condition that is typical of delivery labor and induction of a state of acute fetal hypoxia.
In relation to the disease onset, fetal death in utero precedes maternal frank symptoms in up to 40% of cases139,39, confirming that maternal hypoxia in some cases can develop gradually and the fetus can also be affected by slight alterations in the concentration of maternal oxygen. Alternatively, fetal death may precede embolization, which may expose the mother to a greater risk of contracting the syndrome153, although specific studies conducted on animal subjects have not been able to reproduce the syndrome.
In other cases, clinical onset is characterized by the presence of genital hemorrhage associated with alterations in coagulation, which almost always corresponds to acute disseminated intravascular coagulation (DIC). With the exception of acute DIC cases with catastrophic genital hemorrhage, prognosis depends on whether other signs and symptoms of the disease will develop.
However, DIC generally represents the second phase of severe AFE cases and usually appears from within half an hour to 4 hours after collapse196.
Acute DIC122,19, that is the form of coagulopathy leading to massive and rapid consumption of fibrinogen and platelets, occurs exclusively in cases of AFE, severe sepsis and abruptio.
Another typical sign associated with AFE is uterine atony, which, in the presence of coagulopathy, contributes to emphasize the hemorrhagic phenomenon typical of these patients. It has been supposed that biochemical mediators acting on the myocardium also interfere with the contracting functionality of the myometrium, preventing the uterus from contracting50. It has also been hypothesized that a high FDP (Fibrinogen Degradation Products) concentration deriving from a fibrinolytic process, that usually occurs in the course of DIC, can contribute to worsening this condition197,75.
Apart from the classical severe forms involving more apparatuses, we have already mentioned that other atypical forms or forms classified as" forme fruste”, have been described, in which the main characteristic is a state of coagulopathy associated with a more or less severe genital hemorrhage145,33,159,113,51,192,18,71,108. Ratten152 reported that coagulopathy was the only pathological aberration in 11% of Australian AFE patients and that patient death may be related to the effects of genital hemorrhage rather than the direct hemodynamic dysfunction.
HISTOPATHOLOGICAL AND IMMUNOHISTOCHEMICAL DIAGNOSIS
We have seen that AFE prevalence can vary if paucisymptomatic forms or cases with banal and transitory symptoms are classified with rapidly fatal cases. Clearly, diagnosis of severe cases is fairly easy, while it is not for forms of average severity. Nonetheless, the diagnosis of AFE is mainly clinical and laboratory or histological examinations are useful to establish suspected or confirmed clinical cases.
Clark has greatly influenced this concept, since he observed the presence of squamous cells in the blood taken from the pulmonary artery of patients who had not suffered from AFE39. In the wake of these studies, he may have felt that the search for fetal debris in the maternal circulation of living patients or at autopsy was not of relevant importance for diagnostic purposes. Autopsy is in fact carried out several days after death and little is known about the time when these cells disappear by autolysis. In other cases, specific immunohistochemical analyses are necessary to determine the presence of these cells, which however can only be easily detected by expert eyes. Since there are also cases of false negatives related to the histological examination of lung sections, Clark thus excluded the results of pathological examinations from the inclusion criteria to the National American Registry.
The detection of isolated squamous cells in maternal pulmonary artery blood samples of symptomatic patients has lost much of its value due to possible contamination41. Placement of a catheter in the pulmonary artery is very likely to introduce cells of the mother’s epidermis which, unfortunately, cannot be distinguished from the squamous cells of the fetus’ skin.
The risk of air contamination is also possible when the blood sample taken from the pulmonary arterial catheter is placed on histological slides in order to isolate other debris components, as mucin and lanugo.
Despite these premises, histopathological diagnosis is still necessary today, if not indispensable, to confirm the diagnosis of AFE. Moreover, debris detection in small pulmonary vessels at autopsy is still viewed as a “golden standard” to ascertain diagnosis.
Nonetheless, Obstetric Assessors in the United Kingdom believe that, in some dubious cases lacking autopsy, an AFE diagnosis can be based exclusively on clinical factors.
Clearly, a further analysis through histological and immunohistochemical examinations is believed to be of fundamental importance in complex cases.
Rushton and Dawson160 suggested that a special staining of Alcian Blue-Phloxine-Tartrazine should be employed on all lung sections during autopsy on all cases of maternal death. Such staining techniques should also be used on histological sections of the uterus of both deceased and living hysterectomized patients15. The authors also underlined the importance of an in-depth examination of the possible pathways through which amniotic fluid enters the birth canal in suspected cases.
Lau111 too points out the need to use special staining techniques, rather than the traditional hematoxylin-eosin technique. Indeed, he calls for an analysis of sections of lungs and other organs using Phloxine-Tartrazine (Lendrum staining) to detect squames, Alcian Blue to detect mucin and Sudan Black or Oil Red to detect vernix caseosa.
Despite all the reservations expressed by many authors, debris detection, particularly of some of its components such as mucin and lanugo hair112,157,50,75 is important to confirm the clinical diagnosis or provide further assurance in dubious cases. In the United Kingdom, some Assessors recommend that differential diagnosis in cases of maternal death be accurately excluded from a differential diagnosis of DIC, shock and hemorrhage at autopsy. Pathologists are also recommended not to perform simple histochemical examinations, but rather to carry out further in-depth immunohistochemical analyses on the pulmonary parenchyma of suspected cases. The soft birth canal should always be accurately examined at autopsy and veins adjacent to lacerated areas (even if there are only a few) should be searched for amniotic debris.
Cheung and Luk, through an extensive sampling of the cervix, observed squamous cells and fragments of hair respectively in five of the 14 blocks and in two of the 12 blocks of their reported suspected cases. Therefore, following their observations and since debris can be identified only locally, a thorough examination of the cervix through an extensive sampling is necessary. The detection of squamous cells in the cervical veins has a different meaning than the detection in the pulmonary vasculature, because the latter can hardly result from the contamination of maternal squames for clear anatomical reasons.
Today, a correct diagnosis of AFE can be made by investigating further elements through specific laboratory tests. Since the anaphylactic crisis is the final outcome of severe AFE forms, some researchers have investigated the concentration of histamine and triptase in maternal serum. The latter accounts for about 80% of the vacuolar content of mast-cells, that undergo rapid degranulation in this clinical situation. Farrar67 and Nishio135 published two fetal AFE cases presenting with high triptase serum levels, suggesting that this examination could be performed to confirm the clinical diagnosis of an anaphylactoid syndrome of pregnancy.
Similarly, Fineschi et al.68 observed, through immunohistochemical analyses on triptase receptors, the same high quantities of mast-cells in the lung parenchyma of deceased patients presenting with typical AFE signs and anaphylactic shock, compared to other patients who died due to other pregnancy-related complications. They emphasized the typical characteristics of severe forms, that is a violent immune reaction from the mother to fetal antigens.
Benson25 demonstrated a significant decrease in C3 levels in 87.5% of AFE patients and a decrease in C4 in 100% of these cases, while serum triptase levels were normal. He hypothesized extensive and direct complement activation due to the presence of biochemical mediators in the amniotic fluid.
The effects of a possible DIC should be researched at autopsy to be able to confirm the clinical diagnosis. Signs can already be present at the macroscopic examination of the corpse, such as fluid bleeding from cannulated venous sites, subcutaneous petechiae and collection of fluid blood in the viscera or the peritoneum and the pericardium. Fibrin thrombi need to be searched, particularly in the lung vessels, using specific staining techniques. Their detection, however, is infrequent160,161 and depends on the degree of reactive fibrinolysis, a process that continues after death146,129,134,165,79.
Histochemical and immunohistochemical examinations still play a fundamental role in confirming the diagnosis. Nonetheless, negative histological exams do not exclude the possibility that severe, clinically evident AFE cases are due to minimum quantities of amniotic fluid reaching the pulmonary vessels, which may trigger a severe anaphylactic or anaphylactoid syndrome. The lack of histological data due to poor debris traceability may thus hinder confirmation of AFE diagnosis.
M.B., at 40 weeks of gestation, with a history of cesarean section due to mechanical dystocia, was admitted to our delivery room in labor. After amniotomy, the amniotic fluid appeared to be stained with meconium (grade 2). CTG (cardiotocography) was normal for approximately one hour until the appearance of deep, repeated Fetal Heart Rate (FHT) decelerations. Dilatation was about 8 cm and the presenting part was not engaged, but adapting to the pelvic brim.
It was decided to perform an emergency cesarean section. A laparotomy was carried out within a few minutes. The lower segment appeared intact. Fetus extraction was successful and the operation ended with a blood loss of around 200 cc (cubic centimeters). After about one hour from the end of the operation, the uterus appeared to be very high and atonic. Abundant coagulated blood, approximately 1000 cc, was squeezed out. Uterotonics were immediately administered, another vein was cannulated, emergency blood and presence of the anesthesist were required, blood samples were drawn for complete blood count and coagulation. Approximately 1500 cc of plasma expanders were administered. Despite the intravenous administration of oxytocin and PGE2 and the continuous bimanual massage, the uterus remained strongly atonic. After about 10 minutes of squeezing of the uterus, this manoeuvre was repeated since the uterine bottom rose back to previous levels. Another 1000 cc of very liquid blood were expelled this time. Blood transfusion was started and fresh frozen plasma was required. After a further 5 minutes, it was decided to perform a total hysterectomy due to persistent uterine atony and hemorrhage. On reopening the peritoneum, the uterus presented with good suture retention and no intraperitoneal hemorrhage. The operation did not raise any particular problems, except for the tendency to diffuse pelvic microbleeding. The results of coagulation analysis suggested the presence of an acute disseminated intravascular coagulation. The patient was transferred to the Intensive Care Unit.
Despite an estimated total blood loss of about 3000 cc, the patient did not present with a condition of particular hemodynamic instability.
The post-operation period did not present any particular problems and the patient was discharged six days after admittance.
AFE was immediately suspected, due to absence of other clinical elements or other correlated disorders that could justify an acute post-partum DIC. In particular, an abundant outflow of fluid blood was observed at the second squeezing of the uterus, before blood transfusion was started.
The uterus underwent histological examination. Macroscopically, the cervix showed an area of hemorrhagic infarct with mild laceration of the deciduas. There were no areas of hemorrhagic suffusion in the hysterotomy incision performed for cesarean section.
The microscopic exam of the cervix showed the presence of consistent acid mucopolysaccharides and thus debris of meconial origin in a vein of medium caliber.
Moreover, the vessel was completely thrombosed, while inclusions on the hysterotomy incision did not show any element of fetal origin and the patency of venous vessels could be easily demonstrated.
A blood sample taken during the operation to determine the concentration of zinc-coproporphyrin - a meconium component - resulted negative, reporting values that were significantly similar to normal serum concentrations in healthy subjects.
|Case 1: Alcian Blue PAS magnification X 20: deposit of acid mucopolysaccharidic matter into a cervical vein
||Case 1: Azan-Mallory magnification X 60: endovascular occlusive clot of cervical vein
|Case 1: Alcian Blue PAS magnification X 20: deposit of acid mucopolysaccharidic matter in some cervical veins
||Case 1: Alcian Blue PAS magnification X 40: deposit of acid mucopolysaccharidic matter in a cervical vein
Zinc Coproporphyrin in Maternal Plasma
We measured the concentration of zinc coproporphyrin I (ZnCP-I), a characteristic component of meconium, in maternal plasma by fluorometry after HPLC.
Zinc coproporphyrin I (ZnCP-I), which is present in high concentration in meconium , can be detected in maternal plasma, and it was found to be high in the plasma of patients with amniotic fluid embolus. We found a particular case of AFE with non-detectable level of ZnCP-I.
CP-III dihydrochloride and ZnCP-I were purchased from Frontier Scientific (Logan, UT 84323).
HPLC-grade acetonitrile, reagent grade acetic acid, KH2PO4 and other reagents were all obtained from Sigma Aldrich.
High-performance liquid chromatography of ZnCP-I.
The HPLC system consisted of a series 200 lc Perkin Elmer biocompatible pump equipped with a Perkin Elmer autosampler, a fluorimeter LS 50 B Perkin Elmer and a pc station with Turbochrome as data processor software. Samples were injected into a Rheodyne injector with a 200μl loop.
For ZnCP-I analysis we used an 4.6 x 150 mm column SunFire C18 5μm (Waters Corporation, Milford, MA) with a security guard cartridge ODS C18 4 x 3 mm (distributed by Phenomenex Inc. USA) and a mobile phase of acetonitrile/potassium phosphate, 50 mmol/L (1/5 by vol), pH 6.8. The flow rate was 2.0 mL/min at room temperature. The excitation and emission wavelenghts were 405 and 580 nm, respectively.
Assay of ZnCP-I content of plasma.
ZnCP-I concentration was assayed by HPLC with synthetic ZnCP-I as the standard and Cp-III as the internal standard.
Internal standard was added to calibrator solutions and plasma solutions.
For analysis for ZnCP-I, we followed the spectrofluorometric method used by Gourley et al.  that has showed a sensibilty of about 0.6 pmol/100μL.
The linear regression of calibration curve was R2=0.997.
As Gourley et al. outlined, we did not even detect ZnCP-I in any samples unrelated to pregnancy.
Here below the unexpected non-detectable level of ZnCP-I of AFE diagnosed patient compared to one of the standard solutions of plasma doped with ZnCP-I, used for calibration.
Labor in a 35-year-old nullipara with normal pregnancy was induced via oxytocin administration at 41 weeks + 6 days of gestation. Her complete blood count was normal with hemoglobin level being 12.4 g/dl; coagulation values were also within standard parameters. After normal labor, when uterine cervix dilatation was completed, sudden persistent fetal bradycardia developed, leading the doctor on duty to perform an operative delivery by obstetrical vacuum. Amniotic fluid, which is of a light color during labor, turned to grade 3 due to the presence of meconium. A male newborn weighing 3720 g and in good condition was extracted.
Before the delivery of the placenta, about ten minutes after birth, blood loss was of about 700 cc.
Following the third stage, diffuse fluid blood oozing from small lacerations in the vagina and cervix was observed. They were sutured, but nonetheless genital hemorrhage persisted with outflow of fluid blood from the birth canal.
Coagulation tests suggested a disseminated intravascular coagulation (prothrombin time, 23.8 seconds, PTT 71.4 seconds, fibrinogen 64 mg/dl, ATIII 36%). Platelet count rapidly decreased and reached the amount of 139,000.
Total hysterectomy and bilateral ligation of the hypogastric arteries were performed due to the hemorrhage not resolving. Hemoglobin level was 6.7 g/dl and platelet count 69,000.
The patient was transferred to the ICU (Intensive Care Unit) maintaining good hemodynamic stability. She was administered 26 units of concentrated erythrocytes, 2 units of platelets, 8 units of fresh frozen plasma, 6000 IU (International Units) of antithrombin III and 2.4 mg of activated factor VII. After several hours, clotting too was under control.
In this case too, the clinical diagnosis of AFE was made by exclusion of the other possible causes of coagulopathy.
The patient was taken to the ward after two days and discharged without sequels after another 11 days of hospitalization.
During the patient’s hospitalization in the ICU, blood samples were taken from the pulmonary artery to determine the presence of mucin, vernix caseosa and fetal squames. Twenty-three cytological sections (12 with whole blood and 11 after lysis of erythrocytes) were variably fixed (Delaunoi, Citofix and air) and variably stained (Papanicolau, Emocolor and May-Grunwald-Giemsa). Fetal debris was not found on any of the sections.
The removed uterus microscopically showed vascular thrombotic formations. For the microscopic examination, inclusions on the body, decidua, lower uterine segment and cervix were carried out. A massive hemorrhagic necrosis of the endometrium was evident, with recent isolated endovasal thrombi. Thrombotic formations were observed in the cervical venous vessels, together with Alcian-PAS positive material compatible with fetal debris.
|Case 2: H.&E. (hematossilin-eosin) magnification X 60: venous myometrial vessel with an occlusive clot
||Case 2: H.&E. magnification X 40: fetal debris into venous cervical vessel
|Case 2: Alcian magnification X 60: mucin into a venous cervical vessel
||Case 2: PAS ( Periodic Acid Schiff ) magnification X 20: mucin into a venous cervical vessel
PATHOGENESIS AND PATHOPHYSIOLOGY OF AMNIOTIC FLUID EMBOLISM: NEW ISSUES AND SOME CONSIDERATIONS
Amniotic Fluid Embolism is a typical disorder of delivery labor or abortive labor.
The sites of entry of amniotic fluid into maternal circulation are well known and universally, and almost unanimously, shared by various researchers. For amniotic fluid to overcome the resistance of maternal venous pressure, the presence of a pressure gradient facilitating its flow into the maternal circulation is necessary. The pressure gradient favoring the entry of amniotic fluid into maternal circulation is uterine contraction.
However, we have reports of some cases where the presence of myometrial contraction could not be demonstrated3, even though a certain degree of hypertone or uterine contraction may have preceded the phase of embolization. Amniotic fluid may also enter maternal circulation in the course of either an elective or emergency cesarean section31,182,155. In any case, the pressure gradient pushing the amniotic fluid into the venous vessels opened by hysterectomy can be determined by the development of myometrial contractions appearing during fetal extraction or immediately after. Similarly, there are published reports on cases occurring after amnioinfusion118,49, amniocentesis57,137,167,82, insertion of intrauterine transducer12, removal of cervical suture80, presence of intrauterine device at term180, an attempt of external cephalic version86, uterine manipulation46, feticide or termination of pregnancy63,167 and after abdominal trauma64,150,93,141,136.
In the above cases, the professional’s invasive action or external trauma may cause a break in the blood-placental barrier and simultaneously bring changes to the uterine tone.
In AFE cases, during the labor of pregnant women at term, a large fetal part, mostly the cephalic extremity, may close the pelvic cavity like a sort of “plug” preventing amniotic fluid from outflowing from genitals and redirecting it towards open venous sinuses. This mechanism may be more understandable in the phase of comparison of cephalic extremity and pelvic bone, particularly during the phases of rotation, where the myometrial tissue of the lower segment and cervix are compressed between the presenting part and the maternal pelvic bone, and are thus particularly stretched during contraction. For the hematoplacental barrier to break, amniochorial membranes too need to be broken, although AFE cases with integral membranes have been reported.
A precise temporal relation between natural rupture of membranes, amniotomy and other invasive manipulations of the amniotic fluid, such as placement of intrauterine catheter, amniocentesis, amnioreduction and amnioinfusion, has been widely demonstrated by numerous reported cases.
In some case reports, symptoms either appear immediately after membrane rupture126,11 or several hours later121,124.
Rather than knowing the moment of rupture or the fact that membranes broke, it would have been interesting to know the moment when significant uterine contractions started in these cases and relate it to the onset of AFE symptoms.
In the absence of labor, the pressure gradient in AFE cases, which allows amniotic fluid to enter the maternal circulation, may be due to transitory alterations of the uterine tone or hypertone, and to the simultaneous opening of a uterine venous sinus, probably associated with premature separation of amniochorial membranes or partial placental abruption. Alternatively, in rare cases, anaphylactic or anaphylactoid reaction may occur due to antigen stimulation unrelated to amniotic fluid, such as in the case of feto-maternal transfusion, and the clinical picture may be related to a reaction to antigenically incompatible blood transfusion.
Various researchers have objected that uterine contraction activity, particularly hypertonic contractions, can carry amniotic fluid into the maternal venous system. These authors have in fact underlined that venous sinuses in the myometrium are mechanically closed during the peak of uterine contractions; they believe it is difficult to correlate an increase in intra-amniotic pressure with the passage of amniotic fluid into the venous vessels.
Nonetheless, this well-defined concept of obstetric pathophysiology can only be valid for the uterine venous system, but is less valid for the lower segment and is completely absent for the uterine cervix166. Indeed, the two latter anatomical structures have much less muscular tissue compared to the uterine myometrium and are subject to distension rather than shortening and consequent swelling during uterine contraction, thus potentially favoring the opening of lacerated venous sinuses161,14. This concept confirms what has been supposed by many authors: endocervix and lower segment veins are the most probable portals of entry of amniotic fluid into maternal circulation during labor. However, the concomitant membrane rupture and premature placental separation have been clearly illustrated by Peterson and Taylor139, who have observed this possible mechanism of amniotic fluid entry into the maternal circulation in a significant percentage of their 40 reported AFE cases. When myometrial veins are the portal of amniotic fluid entry, as it occurs in cases of abruptio or premature separation of the placenta, the accompanying, persistent uterine contraction may prevent the amniotic fluid from entering the uterine venous circulation. In these cases, embolization could occur either in the initial phase of contraction or, most likely, during the relaxation phases, when an endouterine depression occurs, favoring, in a way, aspiration of the amniotic fluid through open venous sinuses.
Other studies have suggested that a biochemical gradient may channel amniotic fluid into the maternal circulation95. This hypothesized mechanism, unconfirmed by experimental studies, may be valid for amniotic fluid transit into the capillary system, but not to the venous system where vessel calibre is greater and there is blood outflow. Moreover, flow induced by a biochemical gradient may be theoretically possible for pure amniotic fluid, but less for its debris.
In second trimester AFE cases, particularly in cases of pharmacologically induced abortion, amniotic fluid passage is likely to occur after total or partial placental abruption secondary to protracted uterine hypertonia or the organ’s relaxation phases.
In summary, entry portals are open uterine vessels that can be:
1) Lacerations of the cervix and the vagina or even simple abrasions which cause the opening of a cervical venous sinus or a varix and can also occur during labor in a physiological delivery.
2) Uterine venous sinuses which remain open after premature membrane separation or marginal or partial abruption of the placenta, a condition that is often associated with rupture of membranes, tetanic uterine contractions and premature separation of amniochorial membranes.
3) Uterine venous sinuses that are opened either artificially during cesarean section, or spontaneously through rupture or leakage of a uterine scar during labor, or traumatically after uterine perforation or car accident143,93, during invasive or other intrauterine diagnostic procedures.
It is very important to underline that the penetration of amniotic fluid into the maternal circulation is always described as a pathological condition in the Literature170,157, although some researchers have a different opinion and assume that small quantities of amniotic fluid can rather frequently enter the maternal circulation without causing the syndrome. These researchers have supposed that amniotic fluid embolism is due to the release of pathogenic, toxic or antigenically incompatible amniotic fluid into the maternal circulation34,36,42,31,69. However, if amniotic fluid were able to enter the maternal circulation with extreme ease, we should observe an ease of syndrome relapse in a successive pregnancy of surviving patients, since the fetal antigens causing the syndrome are very likely to be exposed again to maternal circulation. The case reports on new pregnancies after AFE are very few, but in no case has the syndrome reappeared117,37,53,173,62,42,1.
Many studies have been performed on animal models, by injecting the homologous or heterologous amniotic fluid, either filtered or not, through a peripheral vein. In contrast to the studies by Steiner and Lushbaugh, who have reproduced the syndrome mostly by injecting pure or diluited human meconium into monkeys and dogs, successive studies have found contrasting results and the signs and symptoms of the syndrome have been detected only in some cases.
Hankins et al.81 detected debris in the pulmonary vessels of 34.5% of the hyrcus goats to whom they had injected homologous amniotic fluid with meconium.
Petroianu140 reported in two articles the effects meconium-free amniotic fluid or amniochorial membranes have on the coagulation of mini pigs and the lack of systemic effects that often lead to death in humans.
Some doubts could have distorted the results:
1) By overlooking the uterine circulation, researchers have only considered eventual systemic effects of the amniotic fluid, overlooking the fact that some pathophysiological mechanism may have occurred in the uterine vessels, interfering with final results.
2) The amniotic fluid has been injected through a large calibre peripheral venous vessel, which is a condition that does not apply to the uterus where amniotic fluid is very likely to enter the maternal circulation through a small or medium venous sinus.
3) The filtering of amniotic fluid, the administration of human amniotic fluid, the non-pregnant state of some animals and the possible general anesthesia prior to euthanasia may have strongly affected results.
Although many studies have not been able to reproduce the syndrome on experimental animals, other numerous in vivo and in vitro studies have confirmed that amniotic fluid has a certain effect on maternal coagulation197,45.
A thorough analysis of reported cases in literature and the study of our two cases have allowed us to reconcile the various hypotheses and shed further light on the initial pathophysiological mechanism.
As underlined by Bick27, lipid content, cell count, fetal debris, procoagulant activity and amniotic fluid viscosity increase with evolution of pregnancy, reaching their peak on delivery. He emphasizes the effect of in vitro amniotic fluid, such as the acceleration of prothrombin time, activation of thromboplastin partial time, acceleration of coagulation of factor VII deficient plasma and thromboplastin effects. The author believes that not only does the amniotic fluid act as total thromboplastin, but even as activator of the tissue phase, with direct activation of factor X. Factor Xa is one of the most thrombogenic substances known today and, in the presence of factor V and additional phospholipids, rapidly converts prothrombin into thrombin.
Amniotic fluid contains coagulation activators, namely the tissue factor116, and many in vivo and in vitro studies have confirmed the precocious activation of the coagulation system.
Salem163 described the presence of a platelet activation factor in the amniotic fluid and a procoagulant activity independent of factor VII.
Liu et al.115 carried out a study on thromboelastogram and demonstrated that amniotic fluid accelerates in vitro coagulation and, even if not significantly, fibrinolysis. Based on the results of their tests, they believe that the amniotic fluid activates coagulation both through the extrinsic factor and the intrinsic factor, and this effect plays a specific role in primary uterus hemostasis after delivery. The strengthening of coagulation and activation of fibrinolysis is due to the presence of an activator of tissue plasminogen, an activator of a urokinase-like plasmin and thrombin-antithrombin complexes in the amniotic fluid.
Harnett et al.83 have recently confirmed an increase of in vitro platelet activation on fresh blood after the addition of some microlitres of amniotic fluid. These authors suggest that the amniotic fluid tissue factor promotes the conversion of factor VII into activated factor VII obtaining factor VIIa-tissue factor complexes that act on factors IX and X, leading to thrombin formation.
Some studies have emphasized the effects caused by the addition of amniotic fluid in freshly drawn blood, namely a relevant decrease in coagulation time, induction of platelet aggregation, release of platelet factor 3, complement activation and factor X activation.
Other in vitro studies have not been able to demonstrate that procoagulant factors present in significant quantities of amniotic fluid can cause a DIC142, and the genesis of coagulopathy could thus be multifactorial. These studies have clearly emphasized only the in vitro effects, although some local factors are believed to play a more incisive role in coagulopathy induction.
Uszynski at al.187 have correlated the possible procoagulant effect of amniotic fluid, with plasmatic alterations in the relation between tissue factor (TF) and tissue factor pathway inhibitor (TFPI) of the pregnant woman at term and in amniotic fluid factors. TF concentration in the amniotic fluid is on average 45 times higher than in plasma.
Let us thus consider a given fact: amniotic fluid has some effect on blood coagulation. Therefore, how do we translate the entry of the amniotic fluid into the circulation, that is of an element inducing the activation of coagulation?
The majority of studies have been performed to identify the DIC-causing biochemical mediator, while amniotic fluid is very likely to accelerate coagulation and possibly lead to the formation of a venous thrombus.
DIC, on the other hand, could be a secondary effect. This is the starting point for fully understanding the pathophysiological mechanism which begins with exposure of the amniotic fluid in uterine veins, before interpreting and justifying the reasons that trigger the syndrome!
In 1990, M.J. Kool105 reaffirmed that the procoagulant effect of amniotic fluid on stagnant blood, which can be observed in the pelvic vessels of a pregnant patient close to term, is greater and can thus cause a thrombus formation. These concepts had already emerged from other studies176,177,44.
Furthermore, Strickland177 also demonstrated that coagulation time of whole blood mixed with amniotic liquid extracted from women in labor was accelerated in comparison to a similar group whose whole blood was mixed with the amniotic fluid of women who were not in labor.
It can thus be hypothesized that in the majority of embolic events, amniotic fluid does not pass over the uterine venous circulation because a rapid activation of the local coagulant system occurs, leading to immediate thrombosis of venous vessels. Its effect would thus be limited to a blood compartment and could be mainly identified as a potential cofactor of a local coagulopathy. Platelet activation and the formation of a platelet thrombus would occur first, followed by the transformation into a fibrin thrombus.
This process could be accentuated by a possible vasoconstriction of the uterine venous network affected by embolization, due to the direct action of specific vasoconstrictors found in the amniotic fluid, such as endothelin.
Steiner and Lushbaugh172, during their experimental studies on animal models, observed the presence of vernix caseosa and thrombosis of the jugular vein in a monkey that had been administered 5 cc of amniotic fluid and subsequently another 20 cc of the same fluid.
The most relevant data is that the animal did not die after fluid administration, but it was killed after 72 hours from the first injection, while three other monkeys died within six hours from administration of one or three cc of meconium-stained amniotic fluid. While the deceased monkeys showed significant alterations of the pulmonary microcirculation, the lung sections of the monkey that was killed after 72 hours from the first injection showed the presence of a normal pulmonary condition, with only a sling alveolar edema and sporadic vascular thrombosis.
Hymes et al.87 managed to prevent and regress cell and platelet aggregation, and thus restore or maintain an adequate blood flow in monkeys that had been previously or successively administered a lethal quantity of human amniotic fluid. The authors injected 10 cc of surfactant drug, polyoxypropylene-polyoxyethylene glycol," Pluronic F-68 “, into the carotid or the aortic arch of the animals. This potentially prevented platelet aggregation and the formation of microthrombi, dissolving platelet thrombi and cellular aggregates with debris and re-establishing a normal micro-circulation.
Russell and Jones161 described a classical case of a deceased patient due to AFE syndrome showing poor debris in the pulmonary vessels at autopsy. Squames and amorphous material were detected in the vessels of the uterine wall and in the decidual sinuses, but platelet thrombi were not observed.
Cheung and Luk32 published two obstetric cases that developed a DIC post-partum. The histological examination of the cervix in both cases showed not only the presence of amniotic residue, but also confirmed the presence of recently thrombosed venous vessels.
Kern and Duff98 reported on a case where the only documented clinical manifestation was the isolated thrombosis of the right ovarian vein, which, once the adnexa was removed, a significant presence of debris was observed. The patient did not develop neither symptoms of cardio-respiratory failure, nor a severe coagulopathy with significant decrease of fibrinogen levels. The authors believe that the mechanism preventing entry into the pulmonary circulation may have been related to a venous vasospastic reaction to exposure of meconial elements with the interior surface of the vessels.
Gogola and Hankins77, during a cesarean section, observed the presence of bolus material mixed with vernix caseosa stagnating in the venous plexus of the left Couvelaire area, in proximity to the uterus-ovary ligament. Clamping was performed, first manually and then through double clamping of the infundibular-pelvic ligament and uterine veins of the area affected by the embolic process. The patient did not develop hemodynamic instability nor relevant coagulopathy.
It is thus possible that this feature was selected during the evolution of mammals and in some way protected the various species from this frightful disease. Indeed, the presence of procoagulant factors in the amniotic fluid of some species could be related to their greater reproductive capacity.
Uterine venous thrombosis induced by the mixture of amniotic fluid with the maternal coagulation system may be a form of protection preventing amniotic fluid components from reaching the general circulation and thus potentially triggering the syndrome.
The DIC that is associated with AFE is known to be often characterized by a massive fibrinolytic process21,22,61, which often continues after death134,165,79 to the extent that a paradoxically poor detection of fibrin thrombi in vessels occurs at the autopsy of some clinical cases with frank DIC160,21,22,3,27 and some cases show a form of primitive hyperfibrinolysis28.
Morgan133, too, stated that detection of fibrin thrombi is rare in some cases and, to support the concept of the fibrinolytic process continuing after death, he reported a case published by Bowman in 1955 where a patient collapsed during labor, a venous varix was cut on her ankle and a significant presence of thrombi was observed. At the subsequent autopsy, thrombosed veins were no longer present and blood was completely fluid.
A missing or late thrombotic-occlusive effect on uterine venous vessels can therefore also be identified in an early or predominant fibrinolytic effect. In other words, an early or rapid lysis of uterine venous thrombi may pave the way for amniotic fluid reaching the pulmonary circulation.
Amniotic fluid may be trapped in the uterine veins and then released after delivery, explaining the numerous cases where the temporal relation between embolic event and maternal signs and symptoms are not synchronized, in other words that symptoms often appear late in its most acute phase148,124.
Courtney43 first and Margarson121 later assumed, with different theories, that amniotic fluid could be trapped in uterine veins and then released successively. Courtney assumed that the amniotic fluid penetrating the circulation, while the fetus is still in utero, is then released gradually post-partum, with a decrease in uterine tone, but he could not explain the mechanism underlying this vascular aberration. Similar conclusions were drawn by Margarson, who hypothesized that amniotic fluid is trapped due to spinal blockade after administration of bupivacaina and later mobilized with restored sympathetic tone after local anesthesia.
Stolte174, on the other hand, believed that the amniotic fluid that penetrated in the uterine venous vessels was channelled towards the general venous circulation only during the pauses between contractions, since the venous flow stopped under the mechanical effect of myometrial fibres during the phase of increased uterine tone and until after the peak of contraction.
Spapen169 published the case of a gravida undergoing an emergency cesarean section, without complications, due to fetal distress. However, after surgery, the patient suffered from massive genital hemorrhage, mild arterious hypotension and dyspnea. A second laparotomy was performed, but the uterus was intact and atonic. Coagulation showed the presence of a DIC; a Swan-Ganz catheter placed in the pulmonary artery showed data that were not suggestive of AFE: increase in pressure in the pulmonary arteries, elevated PCWP (Pulmonary Capillary Wedge Pressure) and left hearth failure, namely a hemodynamic picture of anemia and hypovolemic shock.
These hypotheses state that the embolization process may occur a few hours before the onset of real cardiorespiratory and neurological symptoms. The presence of a prodromic phase of the syndrome, often characterized by light neurological symptoms, justifies the fact that classical onset of symptoms occur post-partum in a certain number of cases, but embolization is precocious.
The stagnation of amniotic fluid in uterine venous vessels may thus be due to a mechanical obstruction of the vessels undergoing partial or total thrombosis, rather than to the effects of a reduction in sympathetic tone.
The crossing of the uterine venous bed by the amniotic fluid may depend on several factors, such as a primary hyperfibrinolysis induced by biochemical mediators that are found in anomalous concentrations in the amniotic fluid21,28,197 or on a possible maternal inability, even partial, to activate platelet aggregation and the coagulation system.
In light of these hypotheses, the marginal effects detected in cases where the signs and symptoms are limited to an isolated organ aberration, such as in cases of acute fetal asphyxia associated with post partum atony and coagulation, may be justified.
These new considerations shed some light on a hitherto dark scenario, marked by an ongoing dispute among researchers on how and why amniotic fluid entered or not maternal circulation or induced a DIC.
Clark34,35 believes that amniotic fluid can easily penetrated the maternal venous circulation during labor without causing major consequences for the mother and fetus in the majority of cases. In fact, Clark presupposes that amniotic fluid reaches the pulmonary circulation unhindered, without taking into account that amniotic fluid may not reach the general circulation. On the contrary, the influential researcher states that the presence of squames in the pulmonary circulation is due to a possible maternal contamination.
It has always been difficult to place severe or fatal forms on the same level of cases showing an isolated aberration or the dysfunction of a single organ only. We have already stated that the diagnosis of some cases of amniotic fluid embolism reported in the Literature is performed solely by excluding other diseases. These cases have shown reduced maternal mortality (milder forms), because pathological effects are very likely to be confined to the uterus, where the embolic event affecting only the uterine venous circulation can be the cause of fetal asphyxia and/or coagulopathy, without having significant systemic effects.
Uterine or cervical venous vessels thrombosis secondary to amniotic embolization may in fact affect the uterine arterious vessels, causing a remarkable reduction of flow and a consequent more or less extensive local ischemia, which in turn is responsible for post-partum coagulopathy, fetal asphyxia and uterine atony.
Theoretically, this obstruction may allow the amniotic fluid to outflow from the site through which it entered, when the pressure gradient that favored its entry stops, and thus limit disease severity. Coagulopathy, which is particularly marked at local level, can only extend to the general circulation in a marginal way, through the arrival in the general circulation of blood containing a pathological activation of the coagulation system through collateral venous vessels of the uterine plexus that have not been affected by the embolic flow. A general phenomenon can only be observed with the alteration of the laboratory parameters of coagulopathy. Natural detachment of the placenta during the third stage, may also play a significant and well-defined role in the evolution of this complex pathological process in these cases. In fact, we may only detect a mild form of coagulopathy from a peripheral blood sample, while local effects may be a lot more severe.
Courtney43,45 supports this possibility by emphasizing the concepts that had already been discussed by Basu19. It was assumed that FDP concentrations in the uterine venous circulation were higher than in the general circulation, thus justifying the myorelaxing effects on the myometrium76. Courtney also assumed that amniotic fluid can have a two-fold effect on uterine musculature, the first being hyperstimulation to contract and the second the myometrial inhibition causing organ atony.
The state of uterine atony which often characterizes the syndrome can be a final consequence of all these pathogenic mechanisms, but in our opinion it mainly results from uterus ischemia, even if local, due to the consequent vasoconstriction of uterine arteries. The direct action of factors found in amniotic fluid or FDPs19,44,45,30,197,75, which reach the capillary circulation with retroactive flow after venous obstruction, may rather increase the duration and severity of this anomaly.
One may suppose that many post-partum uterine atonies, particularly those unresponsive to therapy based on uterotonics and not associated with risk factor, are a consequence of these local forms of amniotic embolism, mainly associated with coagulopathy.
Katcky97 published 58 cases of patients who underwent emergency post-partum hysterectomy. Several inclusions, comprising the cervix, were carried out on the removed uteri, and sections with Hematoxylin-Eosin were prepared. Where necessary, Alcian - Blue – Floxine staining and immunohistochemical examinations for cytokeratin and actin-smooth muscle were performed. The number of blocks was not always considered adequate, it was lower than 10. Debris was observed in seven patients. Out of the nine cases where uterine atony was an indication of hysterectomy, four were positive for presence of debris, but only one had clinical evidence of AFE. Out of the ten cases where the indication was post-partum hemorrhage, seven were negative for histological signs, even if in these cases too the number of blocks was lower than 10, and two cases showed evident amniotic residue. Only one case showed clinical and laboratory evidence of AFE, which was confirmed by autopsy findings in the lungs.
Liban and Raz114 performed the clinical and histological assessment of 14 cases of women that died due to AFE. They were associated with three control groups: the first included 13 deaths not associated with AFE, the second encompassed 22 uteri removed at peripartum for various reasons, the third included 12 uterine biopsies performed during cesarean section. The AFE cases were all confirmed by detection of squames and mucus in the blood, lung sections and in nearly all uterine sections. In group one, lung histological preparations were negative, a small quantity of mucus was detected in a venous vessel of one uterine block. In group 2, squames were observed in two cases and mucoid material was found in one case. In the third group, two cases showed a small quantity of squames and mucus in one single uterine vein. The cases of positive histology of the first and second control groups can be suggestive of a process of embolization affecting locally the uterine venous sinuses and confirm the data published by Katchy. The two histological positive cases of the third group may confirm that minimum quantities of amniotic fluid can reach the uterine circulation during a cesarean section, but the process stops in that site and the quantity of penetrated fluid may not be sufficient to induce the process of uterine venous thrombosis or may induce the process only in a reduced and asymptomatic way.
Thompson and Budd181 described six cases and emphasized that AFE may be wrongly diagnosed. One of the reported cases, referred to as case E, was of a woman who had naturally delivered a dead fetus four days after the premature rupture of membranes, with the possible cause of death being sepsis. Interestingly, the authors emphasized the presence of squames and vernix caseosa in the veins of the uterus sections, while ialin thrombi were present in lung arterioles at autopsy. Nonetheless, the venous vessels of the lower segment of the uterus were dilated and thrombosed and this process had affected the pelvic vessels. In this case, the presence of squames in uterine veins could not have been caused by maternal contamination, because the patient did not undergo a cesarean section, niether a hysterectomy. Indeed, in cases undergoing a cesarean section or hysterectomy, the blade of the scalpel or scissors used to incise the maternal skin and the myometrium or to cut the vascular plexus, may theoretically contribute to the entry of the mother’s squames into the maternal circulation. The underlying pathophysiological mechanism is, however, difficult to understand. Amniotic fluid embolization, confined to the uterine venous sinuses, may occur during fetal extraction, with a mechanism similar to that hypothesized for labor. After all, the detection of a vernix caseosa can only confirm the process of embolization, and the presence of dilated and thrombosed uterine veins may lead to a diagnosis of a minor AFE form, as we suggested, without excluding that the embolic fluid may have also channelled microorganisms of the amniotic fluid into the maternal venous circulation and may have therefore caused sepsis.
These concepts thus justify the fact that many cases diagnosed as post-partum hemorrhage, where coagulopathy is attributed to the often conspicuous blood loss, are rather the expression of an association between uterine atony and coagulopathy secondary to amniotic embolization that occurred during labor or delivery.
There are also reports on cases where clinical onset is represented by fetal asphyxia during labor, followed by fatal DIC, which caused substantial hemorrhage post-partum147 and led in some cases to patient death145. The death of these patients can be related to the hemorrhage only and we will never know if these women would have later developed an anaphylactic or anaphylactoid syndrome if they had survived. In other cases18 the anaphylactic reaction to amniotic fluid can have a late onset and coagulopathy can have a two-fold origin.
If, on the other hand, the amniotic fluid crosses the uterine venous circulation, the maternal coagulation system may not be activated significantly at the level of uterine venous vessels once exposed to amniotic fluid procoagulant agents. This anomaly may be due to the congenital absence in the mother of some coagulation factor or to the fact that the quantity of fluid entering the circulation is particularly elevated. Another possibility, underlined by several researchers, is that amniotic fluid may have atypical characteristics. In our hypothesis, it may be unable to activate coagulation locally or it may do so late or it may activate a simultaneous hyperfibrinolysis or it may not control it and thus amniotic fluid may reach the pulmonary circulation and cause cardio-respiratory symptoms. In these cases, however, severe respiratory distress or cardiac arrest cannot be due to the activation of coagulation in the pulmonary vessels and thus to the development of thrombi, because if no such response occurs precociously in the uterine veins, it is less likely to occur in other areas. This hypothesis is in agreement with the fact that coagulopathy or DIC in a significant number of severe AFE cases is an epiphenomenon successive to the first phase of disease (second AFE phase).
Quance148 published a case where, after fetal extraction with standard cesarean section, the woman presented with sudden O2 desaturation and dyspnea, which persisted for the first 45 minutes following operation. After an initial and temporary improvement, jugular vein distension and pulmonary edema were observed. The patient was transferred to the ICU, a pulmonary catheter was inserted and conditions were stable. Thirty minutes after pulmonary catheterism, three hours after delivery and during uterine massage, the patient developed sudden malaise, seizures and suffered from a cardiac arrest after 1 minute and died.
There could thus be two factors causing the classical syndrome: the first interacting with the coagulation system, and the second, if present, being responsible for the anaphylactic or anaphylactoid syndrome characterized by cardio-respiratory insufficiency.
The following possibilities can thus be considered:
1) Amniotic fluid rapidly reaches the pulmonary circulation for the aforementioned reasons. If this amniotic fluid is antigenically incompatible, it is generally responsible for the most severe forms of the classical syndrome, particularly in those cases with acute onset of cardio-respiratory symptoms. The exitus of these patients is related to dysfunction of the right ventricle or cardiac arrest caused by intense vasoconstriction of pulmonary arterioles after exposure of their internal surface to amniotic fluid. The cases surviving this phase (the blue phase) generally develop a cardiogenic shock induced by left hearth failure due to reduction of coronary flow that may be a secondary effect of anaphylaxis. These manifestations may also be compatible with the entry of small quantities of amniotic fluid in the pulmonary circulation.
Viceversa, if the amniotic fluid rapidly reaching the pulmonary circulation is antigenically compatible, it may be responsible for a mild or moderate maternal respiratory distress of an obstructive nature149 or it may immediately cause no symptom at all. Some AFE cases with non-cardiogenic pulmonary edema may be associated with this pathophysiological condition104,55,194.
A coagulopathy or DIC may thus appear after the anaphylactic or anaphylactoid reaction and thus after the cardio-respiratory symptoms.
2) Amniotic fluid is trapped in the uterine veins, but is then gradually released into the circulation during labor and particularly post-partum, due to precocious interruption of the protective barrier created by uterine venous thrombosis, as it occurrs with hyperfibrinolysis or conspicuous embolism.
The passage may initially be of a low quantity or gradual in time. If the amniotic fluid is antigenically incompatible, once it has reached the pulmonary vessels, even in small quantities, it can initially generate prodromic neurological signs and symptoms of mild cerebral hypoxia, (restlessness, behavioral change, feeling of cold, shortness of breath) and then, when the embolic flow becomes consistent, often gradually lead to classical signs and symptoms with varying onset, such as respiratory distress, seizures, etc. In these cases, the DIC can be either triggered by an initial, local uterine disorder and successively can be aggravated by activation of the complement, secondary to anaphylaxis. Clearly, the effects of an antigenically compatible amniotic fluid on maternal coagulation will depend on the duration of occlusion of the uterine venous vessels, the concomitant arterious vasoconstriction and the degree of local or total ischemia of the uterus. A contextual and consistent activation of fibrinolysis or the lack of an antifibrinolytic reaction may clearly contribute to the flow of amniotic fluid that is partially trapped in the uterine venous system, towards the pulmonary circulation and thus self- generate and worsen the syndrome.
The amniotic fluid in these cases may move towards the lungs after a slow, but at times even rapid, process of fibrinolysis on the obstructed uterine venous vessels and the anomaly may depend on the fact that the fluid is not trapped by the thrombotic process or on some other mechanism which prevents the fluid from outflowing once the pumping phase has terminated.
Even maternal response to fetal antigens may be of a varying degree and can thus influence syndrome evolution and maternal prognosis.
If the amniotic fluid that slowly reaches the pulmonary circulation is antigenically compatible, the cardio-respiratory effects are generally minimal or absent and there is rather a prevalence of symptoms caused by coagulopathy localized in the uterus, secondary to thrombosis of a portion of the embolized uterine venous tree. In these cases, the signs of uterine venous thrombosis may be present, but would not be generally widespread and would depend on the degree and time of development of coagulopathy.
3) The amniotic fluid, either antigenically compatible or incompatible, is confined to the uterine venous system by the barrier of the thrombosed veins, cannot reach the pulmonary circulation and can even outflow from the site through which it entered. These are the cases of incomplete clinical picture of AFE that are related to alterations induced on the uterus and the fetus still in the cavity.
Generally, disease onset is characterized by post-partum hemorrhage secondary to coagulopathy or DIC, even severe, which is often associated with or followed by uterine atony contributing to worsening the hemorrhage. In the cases where the process starts during labor, the placental delivery may also play a precise role in the genesis of coagulopathy.
Onset can also be represented by a condition of fetal hypoxia, not secondary to maternal hypoxia, but successive to a sudden reduction of arterious flow in the uterine arteries due to a vasoconstriction caused by local venous thrombosis. These are the cases where a coagulopathy or DIC are evident, but there are no signs of hemodynamic instability, despite conspicuous blood loss.
Clearly, the genesis of a DIC form in these cases cannot be related to complement activation.
If these signs are present as a consequence of a massive hemorrhage, the appropriate and timely therapy can rapidly help restore a condition of stability.
The histological examination of the uterus, if removed, may nearly always be positive due to the presence of venous thrombi and debris in the uterine venous plexus.
Some of these cases may be fatal, but the death-causing hemodynamic insult may be more easily related to the hypovolemic shock consequent to acute DIC hemorrhage.
It can be theoretically assumed that if a certain amount of antigenically incompatible amniotic fluid enters the uterine venous circulation, the vascular and thrombotic response would prevent the fluid from reaching the pulmonary circulation and from triggering the syndrome.
Similarly, small quantities of amniotic fluid in the uterine venous system can have a thrombotic effect on small venous vessels, but the affected area may be too small and thus insufficient to generate a local coagulopathy or a consequent effect on uterine arterious vessels.
Similarly, small quantities of antigenically incompatible amniotic fluid which, for the above reasons, can rapidly pass over the uterine venous system, may trigger AFE and severity of symptoms will depend on maternal immune response.
The mechanism whereby the presence of antigenically incompatible amniotic fluid does not trigger an anaphylactic response when the fluid stagnates in the uterine venous plexus remains to be clarified.
The rapid formation of platelet thrombus may isolate the antigens responsible for maternal anaphylactic response; therefore, the amniotic fluid and its components, even if antigenically incompatible, is trapped by the thrombus induced by amniotic fluid mediators. The pregnant uterus may also be unresponsive to the immune response and thus be tolerant towards external antigens as it tolerates the placenta23.
Much information has been published on cases where maternal symptoms secondary to anaphylactic response is delayed in relation to the time of embolization. Therefore this type of response is very likely to occur only when the amniotic fluid reaches the pulmonary circulation and not when the fluid is confined to the uterus venous plexus, showing its inability to trigger an anaphylactic syndrome when it is confined to that area.
It is thus very likely that an anaphylactic or anaphylactoid response occurs only when the amniotic fluid manages to cross the uterine venous vascular plexus and reaches the pulmonary arterioles.
The DIC consequent to amniotic embolization may have a two-fold genesis. In cases where the embolic process stops in the uterine venous circulation, the DIC may develop from the combination of thrombotic effect on uterine venous vessels, venous vasoconstriction, insult on endothelium and capillary bed, myometrial richness of the tissue factor, insufficient action of myometrial and placental TFPI (tissue factor pathway inhibitor)107, hischemic reaction of uterus induced by the consequent arterious vasoconstriction. In these cases, the third stage, even when it is physiological, may be the moment when DIC symptoms appear, that is when uterine hemorrhage is observed.
Viceversa, in cases where amniotic fluid immediately reaches the pulmonary circulation, the mechanism may be primarily related to anaphylactic response and complement activation.
It may be assumed that the uterine pathogenetic mechanism prevails in cases of frank DIC onset, causing rapid evolution of the hemocoagulative disorder; greater effects are determined by genital hemorrhage which is nearly always worsened by uterine atony. There can often be a discrepancy between coagulation laboratory parameters, extent of hemorrhage and poor therapeutic efficacy.
On the other hand, DIC in classical forms occurs from half an hour to several hours after collapse and may be directly related to the synergic effect between amniotic fluid thrombo-platelet activity and the effects secondary to anaphylaxis, mast cell activation and complement activation23.
In the forms where the antigenically incompatible amniotic fluid reaches late the pulmonary circulation, these two mechanisms may overlap and greater effects may be determined by fluid quantity localized in the uterine veins, and the patient’s degree of anaphylactic response.
In the 1940s, 1950s and 1960s many published studies emphasized the role of the uterine factor in the syndrome’s pathogenesis.
P.E. Steiner and C.C. Lushbaugh172, in their renowned work, underlined the presence of debris in uterine veins in one of their reported cases, giving great importance to this observation, although many histological exams on the deceased patients were not extended to the uterus and uterine cervix.
Liban and Raz114 reported on the first control group of cases of maternal death not clinically associated with AFE, and observed in one case the presence of mucous material in the interstitium of cervical vessels, but not in the pulmonary vessels.
Many other published cases have described the presence of coagulopathy or DIC as the main or sole dysfunction associated with AFE.
As less diagnostic importance has been gradually attributed to the detection of squames in the pulmonary vasculature, major works have mainly focused on understanding the hemodynamic aberrations caused by the syndrome and the description of case reports. Studies on animal models have also followed this direction.
In fact, over the past decades studies have no longer focused on understanding the underlying pathophysiological process of the syndrome that is the possible hemodynamic and functional changes that occur when the amniotic fluid enters the mother’s uterine venous circulation.
We have managed to demonstrate that amniotic fluid stops in the uterine venous circulation, where it exercises its pathogenic effect. Any systemic effects, such as hemorrhagic shock or DIC, result from this initial aberration. At the same time, we have emphasized the fact that no maternal sensitivity to fetal antigens is developed in case of incompatibility, if the amniotic fluid remains confined to the uterus.
Our first described case developed a condition of fetal distress during labor, making it necessary to resort to an emergency cesarean section. Soon after the operation, the patient developed coagulopathy and uterine atony with subsequent genital hemorrhage non responsive to therapy. A hysterectomy was thus necessarily performed. The coagulopathy could not have been induced by other disorders, which were all accurately excluded. The only remaining hypothesis was that of a meconium-amniotic fluid embolic process. The histological examination confirmed this hypothesis since it resulted positive for the presence of acid mucopolysaccharides in a diffusely thrombosed venous branch of the cervix. A blood sample was taken during the mother’s critical stage to measure zinc- coproporphyrin, which was within the normal range. The normal levels of serum zinc- coproporphyrin, which is found in meconium and amniotic fluid, showed that the embolic process stopped in the uterine cervix and the uterus, inducing local effects (fetal distress, uterine atony and genital hemorrhage linked to coagulopathy), without leading to evident hemodynamic, systemic effects, despite massive genital blood loss (3000 c.c.).
Similarly, in the second case, the amniotic fluid that entered the cervical venous vessels contributed to the formation of local thrombi, which prevented the fluid from flowing towards the pulmonary circulation. The absence of debris in the numerous cytological sections of pulmonary arterious blood could confirm this analysis. The negative cytological examination of pulmonary arterious blood plays a negative predictive role, relevant for diagnostic purposes in suspected cases71.
We have thus assumed that the amniotic fluid entering and stagnating in the uterine venous vessels during labor can induce a local thrombosis, which in turn can be responsible for a process of local coagulopathy. This condition may result in local disseminated intravascular coagulation with evident maternal genital hemorrhage, with specific systemic symptoms being related only to the consequences of the hemorrhage. Should this disorder occur during labor, uterine venous thrombosis can cause an arterious vasospasm, generally and rapidly leading to fetal hypoxia with acute fetal distress. In many of these cases, immediate uterus removal could help to resolve every problem, since this operation would also remove any amniotic fluid stagnating in the uterine vessels and favoring local coagulopathy. In other cases, coagulation affects the areas adjacent to the uterus, and hemorrhage extends to the visceral peritoneum and utero-ovaric ligaments. In these cases, histology or immunoistochemistry could confirm the presence of debris in the vessels where amniotic fluid has entered or where it has stagnated.
A sign often associated with AFE cases is pelvic pain. Symptoms in women in labor with a risk of AFE can be characterized by intense contractile activity of the uterus. However, the pain is often gravative-ischemic and can be easily associated with thrombosis of uterine veins.
Uterus atony is another event secondary to the effects of local coagulopathy and myometrial hypoxia. We believe that more or less severe AFE forms localized in the genital area may have a far greater prevalence, since uterine atony and consequent genital hemorrhage is the first or second most frequent indication for emergency post-partum hysterectomy70,14,96. In AFE forms with anaphylactoid or anaphylactic syndrome, uterine atony may result from shock, generalized maternal hypoxia, which means it can be related to the reduced quantity of blood, mainly desaturated, reaching the uterus172,137. The development of DIC or a late fibrinolytic process in these cases may worsen this condition due to the direct inhibition of fibrin degradation products on the myometrial fibre.
The entry of amniotic fluid in the uterine venous circulation may thus accelerate the coagulation process leading to the formation of venous thrombi. However, the coagulopathy or DIC may have a more complex genesis due to the interaction of several factors, even if the process is started by the entry of amniotic fluid into the maternal venous circulation.
In AFE forms where amniotic fluid rapidly crosses the uterine venous circulation, without locally activating the coagulation process, the effects on the coagulative state may appear late and after complement activation.
Therefore, in agreement with Benson and Ratten24,26,152, the severity of coagulopathy in AFE cases may be directly proportional to the quantity of amniotic fluid entering the circulation. Other factors such as the patient’s gestational age and labor can strongly condition the procoagulant effect of amniotic fluid.
Our hypotheses suggest that the embolic flow of amniotic fluid may be solely responsible for modest alterations in coagulation parameters, but may not directly cause a DIC, which would instead indirectly develop with the described two-fold mechanism. However, the severity of coagulopathy can be conditioned by the factors listed in Table 4.
FACTORS CONDITIONING SEVERITY OF COAGULOPATHY
1) Patient’s gestational age.
2) Quantity of penetrated amniotic fluid.
3) Presence of meconium.
4) Intensity of localized effect due to persistent presence of amniotic fluid in uterine venous vessels in the case of thrombotic obstruction.
5) Antigenically incompatible amniotic fluid.
6) Intensity of anaphylactic reaction, if any, in the case of rapid access of the amniotic fluid in the pulmonary circulation.
7) Degree of partial obstruction of uterine veins and temporary or partial entrapment of amniotic fluid in that area.
8) Onset of coagulopathy clinical signs in the first or second stage of delivery.
Amniotic fluid can thus cross the uterine venous tree and reach the maternal circulation. If the fluid is antigenically compatible, it does not cause any significant systemic symptom other than a possible, paucisymptomatic or transitory obstructive activity on the pulmonary circulation and a significant coagulative disorder that is secondary only to a possible initial thrombotic-obstructive involvement of the uterine venous circulation. In some cases, respiratory distress due to obstruction and consequent right heart disorder, which may require intensive care, may occur after a moderate embolic flow of amniotic fluid55,104.
The antigenically compatible amniotic fluid may thus enter the circulation and trigger a uterine venous thrombosis, and therefore severe localized coagulopathy, subsequent massive hemorrhage and be thus potentially fatal. Therefore, the cause of death in these cases ought to be attributed more to the consequences of rapid and conspicuous hypovolemia than to the alterations secondary to an anaphylactic process152.
In light of our new theories, we suggest that a severity scale for AFE diagnosis could be introduced, encompassing the various cases according to observed symptoms and severity of clinical picture. Alternatively, the term “Anaphylactic or Anaphylactoid Pregnancy Syndrome” could be added in the diagnosis of severe cases actually affected by the syndrome.
The anaphylactic response may also be roused to the maternal pulmonary vessels exposure to trophoblastic antigens, particularly if they are the expression of placental anomalies. Therefore, also in these cases coagulopathy is related to the effects of maternal immune response, rather than the direct thromboplastin-like action of the trophoblast tissue or of the amniotic fluid166, 34.
As a consequence of what has been stated so far, another possible cause of maternal death related to the penetration of amniotic fluid in the venous circulation, may be the mobilization of one or more uterine venous thrombi20,144, extended to the pelvic vessels , that developed due to the procoagulant effects of amniotic fluid. This could lead to severe or fatal thrombo-embolism post-partum or during puerperium66. Bauer20 in fact demonstrated the coexistence of AFE clinical signs and symptoms and thrombo-embolism, respectively, confirmed histologically, in a fatal AFE case.
Malvino119 later reported an AFE case presenting with tonic-clonic seizures and DIC and a large thrombus, which was detected during insertion of a catheter in the pulmonary artery and was in part rechannelled mechanically.
It is thus recommended that, in case of hysterectomy, histochemical and immnuhistochemical examinations are systemically performed on uterus inclusions from all the possible entry sites of amniotic fluid. This procedure is particularly important because not only does uterine histology show the portal of entry of amniotic fluid, but it also helps detect the presence of venous thrombosis, which is often still present since the fibrinolytic process that generally develops after DIC and genital hemorrhage has not yet been able to dissolve the previously formed thrombi. Paradoxically, thrombotic formations in the uterine venous network may be occasional or even absent in cases where histological examination is performed at autopsy. This is due to the fact that the fibrinolytic process may have dissolved the thrombi during the patients’ hospitalization and even after death.
Similarly, in cases of maternal death due to thrombo-embolism, debris ought to be searched in pulmonary thrombotic formations, particularly in cases of negative medical history or with no thrombophilic risk, in subjects who die under heparin prophylaxis and in cases that simultaneously present with a coagulation disorder or other complications during delivery.
In many ways, little is known about amniotic fluid embolism.
There are severe, often fatal forms characterized by respiratory distress, cardiovascular collapse, neurological manifestations and disseminated intravascular coagulation.
There are also less severe forms that are more difficult to diagnose, mainly because symptoms are often mild, less intense or only limited to a form of coagulopathy.
The effects of amniotic embolization can be very different and depend on whether amniotic fluid reaches the pulmonary circulation and whether it is antigenically incompatible or not.
We have assumed that in these" milder forms" or “forme fruste”, amniotic fluid remains confined to the uterine venous vessels, causing thrombosis of vessels and preventing entry into the general venous circulation. Our data are supported by numerous AFE cases reported in the Literature where debris and thrombosis have been observed in uterine veins. The consequent effects would be limited to the uterus and/or the fetus, if it is still in the cavity. In some of these cases, coagulopathy can generate a massive genital hemorrhage that can also lead to exitus.
Many forms of uterine atony non responsive to therapy or forms of acute fetal distress can be cases of amniotic fluid embolism, where significant reduction of arterious blood flow has resulted from the thrombosis of uterine venous vessels. The failure of this protective mechanism is likely to push amniotic fluid beyond the uterine veins and reach the pulmonary circulation.
It should not be excluded, however, that the incomplete formation of a hemostatic block depends on the early activation of the fibrinolytic system; this could justify those cases where cardio-respiratory failure appears later than the presumed embolic event.
Small quantities of amniotic fluid entering the uterine circulation generally have a modest effect on the local vascular system and can thus be asymptomatic. If these small quantities of amniotic fluid reach the pulmonary circulation, for the reasons described above, the effects can be modest or absent if the amniotic fluid is antigenically compatible. However, if the amniotic fluid is incompatible, the same small quantities may cause AFE and, in these cases, the severity of the clinical picture is related more to the mother’s immune response than to the quantity of penetrated fluid.
Coagulopathy, often associated with uterine atony, contributes to the development of this disorder through the pathological presence of fibrin degradation products or biochemical mediators found in the amniotic fluid, which interact with the contractile system of the uterine myofibril. Many post-partum hemorrhages attributed to uterine atony are clearly cases of primitive coagulopathy and successive atony, secondary to non-systemic amniotic embolism.
The pathogenesis of DIC may be two-fold, resulting from the effects of a uterine venous thrombosis and the consequences of an anaphylactic crisis. Both mechanisms may complement one another in some cases, particularly in cases presenting with initial uterine venous thrombosis and successive flow of antigenically incompatible amniotic fluid towards the pulmonary circulation.
Therefore, in agreement with Davies50, when a mother is exposed to amniotic fluid elements, the maternal and neonatal outcome depends on the degree of her pathophysiological response and the fluid’s antigenic degree.
In light of our observations, AFE may be classified as a “minor form” in cases where embolization is confined to the uterine or pelvic venous vessels and as a “major form"in cases presenting with signs and symptoms of an anaphylactoid or anaphylactic response.
In AFE “minor forms”, a coagulopathy induced by amniotic fluid embolization of uterine veins is an unforeseeable event. The often consequent hemorrhage is frequently characterized by sudden onset and massive genital blood flow, often non responsive to medical treatment, even if administered promptly. In such cases, it is fundamental, in the event of medical-legal disputes, to carefully establish the cause of hemorrhage and also consider aspects of above, particularly in cases where the uterus was to be removed or the hypogastric arteries were to be ligated, or particularly in fatal cases.
We believe that the role of the uterine factor needs to be further assessed to better understand the pathophysiology of the syndrome.
In light of this analysis, rapidly fatal AFE cases may thus be justified by the total or partial lack of thrombotic effects of the amniotic fluid on maternal blood. To confirm our hypotheses, future studies on animal models are needed to assess uterine vascular modifications after injection of homologous amniotic fluid in a medium caliber of the uterine cervix or myometrium of a chosen mammal at term or during labor, and study local and general effects on coagulation.
At the same time, the effects of the in vitro mixture of amniotic fluid and human blood extracted from women in labor could be studied on a large scale, in order to identify the cases that do not develop the classical coagulation changes and deepen our knowledge on this deficit, particularly on the fibrinolytic system.
Some forms of hereditary platelet disorders, such as von Willebrand’s disease, may in fact play a precise role in the development of severe forms. Platelet inactivation after the arrival of amniotic fluid procoagulants and the consequent lack of platelet thrombi formation in the small uterine venous vessels may pave the way for amniotic fluid towards the pulmonary circulation, with all the possible aforementioned consequences. Retrospective studies examining these issues in light of these new theories are thus possible.
Moreover, thorough pathological and immunohistochemical studies on uteri removed post partum in suspected cases may further clarify the initial pathophysiology of the syndrome.
These further studies may help clarify the main unclear and worrying issues regarding AFE, and could help shed light on the syndrome in the near future in order to identify cases at risk with more precision.
Signs and symptoms of AFE minor
1) Coagulopathy or DIC with rapid onset with more or less severe genital hemorrhage not preceded by respiratory distress, cardiovascular collapse or neurological manifestations.
2) Fetal asphyxia or cardiotographic signs of potential asphyxia.
3) Early uterine atony, generally with scant response to therapy.
4) Possible hemorrhagic shock, but patients treated promptly maintain a general hemodynamic stability.
Signs and symptoms of AFE major
1) Cardiocirculatory collapse with sudden onset and cardiac arrest.
2) Maternal hypoxia due to acute or subacute respiratory distress, with typical neurological symptoms, such as seizures, rigor and coma. In some cases, the most severe symptoms can be preceded by a prodromic phase (restlessness, change of behavior, tachypnea, feeling of cold, etc.)
3) Cardiogenic shock due to failure of left ventricle
4) Significant coagulopathy or late DIC, genital hemorrhage, uterine atony.
5) Fetal asphyxia secondary to maternal hypoxia.
6) Pulmonary edema and ARDS (Adult Respiratory Disease Syndrome).
In some cases presenting with a simultaneous action on uterine vessels and release of incompatible amniotic fluid towards the pulmonary circulation, the signs and symptoms of both forms can be observed.
The authors describe two clinical cases presenting with an atypical form of amniotic fluid embolism characterized by rapid onset of DIC without apparent severe hemodynamic instability. The histological examination of the removed uterus showed the presence of debris and thrombotic formations in some uterine veins. At the same time, the authors observed the absence of amniotic fluid components outside the uterine venous plexus.
The authors thus supposed that, in different cases, the embolic process does not reach the general circulation, but remains confined to the uterus and, if this phenomenon is of a considerable quantity, it can be responsible for local pathogenic effects such as fetal distress, coagulopathy and uterine atony.
Many cases of post-partum hemorrhage and/or atony can be classified as moderate AFE forms, especially in cases of fetal asphyxia developed during labor and of primitive coagulopathy.
Classical AFE appears to develop in cases where amniotic fluid rapidly reaches the pulmonary circulation unhindered or when it manages, even partially, to cross the obstacle of uterine thrombotic formations induced by amniotic fluid procoagulant factors and is at the same time antigenically incompatible. Symptoms are believed to vary according to timing, quantity of amniotic fluid reaching the pulmonary vessels and maternal immunological response.
Only if antigenically incompatible amniotic fluid reached the pulmonary circulation, a maternal anaphylactoid or anaphylactic response may be triggered.
The DIC developing with amniotic embolization can have a two-fold genesis. An early thrombosis of uterine veins could be responsible for the organ’s ischemia, even partial, due to consequent arterious vasoconstriction, endothelial damage, acidosis and fetal asphyxia, if the fetus is still in utero.
At the same time, it would prevent amniotic fluid from reaching the pulmonary circulation. AFE cases with isolated coagulopathy may be thus justified.
The unhindered arrival of antigenically incompatible amniotic fluid into the pulmonary circulation may be responsible for the immediate vasoconstriction of pulmonary arterioles and anaphylaxis. In these cases, the generally late DIC may be the result of complement activation.
The entry of amniotic fluid into the central venous circulation may be justified by the maternal inability to develop a platelet thrombus or a fibrin thrombus in the presence of procoagulant amniotic fluid factors or it may be due to the congenital lack of these factors in the amniotic fluid.
Alternatively, a condition of hypefibrinolysis activated on uterine venous thrombi and thrombotic induction would in some cases pave the way for the amniotic fluid trapped in the uterine venous plexus to gradually enter the pulmonary circulation. Successive symptoms may depend on the degree of maternal response and fetal antigenic compatibility.
Similarly, fetal asphyxia may have a different pathogenetic mechanism. It may originate from a state of hypoxia induced by consequent arterious vasoconstriction of uterine arteries in forms localized in the uterus. In generalized forms, on the other hand, this condition may be generated by maternal hypoxia successive to spasm of pulmonary arterioles or to cardiogenic shock.
The authors thus explain the possible, effective pathophysiological mechanism of AFE which can help classify milder and severe forms, and suggest further studies to identify cases at risk.
- Abecassis P.; Benhamou D.: "Is amniotic fluid embolism likely to recur in a subsequent pregnancy?" Int. J. Obstet. Anesth. 15 (1): 90, 2006.
- Abenhaim H.A.; Azoulay L.; Kramer M.S.; 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. 199. 49e1-49e8, 2008.
- Anderson D.: "Amniotic fluid embolism: A re-evaluation" Am. J. Obstet. Gynecol. 98 (3): 336-348, 1967.
- Anonymous: "Amniotic fluid Embolism" Lancet 25: 398-400, 1979.
- Attwood H.D.: "Amniotic fluid embolism" Pathol. Ann. 7: 145-172, 1972.
- Attwood H.D.: "Fatal pulmonary embolism by amniotic fluid" J. Clin. Pathol. 9: 38-46, 1956.
- Attwood H.D.: "The histological diagnosis of amniotic fluid embolism" J. Pathol.
Bacteriol. 76: 211-215, 1958.
- Attwood H.D.; Delprado W.J.: "Amniotic fluid embolism confirmed at autopsy five weeks after
delivery" Pathology 20: 381-382, 1988.
- Attwood H.D.; Downing S.E.: "Experimental amniotic fluid and meconium
embolism" Surg. Gynecol. Obstet. 120: 255-262, 1965.
- Attwood H.D.; Rome R.M.: "Amniotic fluid embolism and uterine
rupture" Austr. N.Z. J. Obstet. Gynaecol. 3: 73-77, 1963.
- Awad IT; Shorten GD: "Amniotic fluid embolism and isolated coagulopathy: atypical presentation of amniotic fluid
embolism" Eur. J. Anaesth 18: 410-413, 2001.
- Ayoub C.M.; Zreik T.G.; Dabbous A.S.; Baraka A.S.: "Amniotic fluid embolus: can we affect the outcome?” Curr. Opin. Anesthesiol. 16: 257-261, 2003.
- Azegami M.; Mori N.: "Amniotic fluid embolism and leukotrienes" Am. J. Obstet. Gynecol. 155: 1119-1124, 1986.
- Bakshi S.; Meyer B.A.: "Indications foe and Outcomes of Emergency Peripartum
Hysterectomy" J. Reprod. Med. 45 (9): 733-737, 2000.
- Bannister M.: "Amniotic fluid embolism" Can. J. Anesth. 47: 381, 2000.
- Barno A.; Freeman D.W.: "Amniotic Fluid Embolism" Am. J. Obstet. Gynecol. 77 (6): 1959.
- Barrows J.J.: "A documented case of amniotic fluid embolism presenting as acute fetal
distress" Am. J. Obstet. Gynecol. 143: 599-600, 1982.
- Bastien J.L.; Graves J.R.; Bailey S.: "Atypical Presentation of Amniotic Fluid
Embolism" Anesth. Analg. 87: 124-126, 1998.
- Basu H.: "Fibrinolysis and abruption placentae" J. Obstet. Gynaec. Brit. Cwith. 76: 481, 1969.
- Bauer P.; Lelarge P.; Hennequin L.; Lambert H.; Larcan A.: “ Thrombo-embolism during amniotic fluid
embolism" Intensive Care Med. 21: 384, 1995.
- Beller F.K.; Douglas G.W. Debrovner C.H. et al: "The fibrinolytic system in amniotic fluid
embolism" Am. J. Obstet. Gynecol 87: 48-55, 1963.
- Beller F.K.; Uszynski M.: "Disseminated intravascular coagulation in
pregnancy" Clin. Obstet. Gynecol. 17 (4): 250-278, 1974.
- Benson M.D.: "A hypothesis regarding complement activation and amniotic fluid
embolism" Med. Hypoth. 68 (5): 1019-1025, 2007.
- Benson M.D.: "Nonfatal amniotic fluid embolism: three possible cases and new clinical
definition" Arch. Far. Med. 2: 989-994, 1993.
- Benson M.D.; Kobayashi H.; Silver R.K.; Ohi H.; Greenberger P.A.; Terao T:
"Immunologic studies in presumed amniotic fluid embolism" Obstet. Gynecol. 97 (4): 510-514, 2001.
- Benson M.D.; Lindberg R.E.: "Amniotic fluid embolism, anaphylaxis and
tryptase" Am. J. Obstet. Gynecol. 175: 737, 1996.
- Bick R.L.; Frenkel E.P.; Baker W.F.; Sarode R.: "Hematological Complications in Obstetrics, Pregnancy, and
Gynecology" Cambridge University Press, 2006.
- Biron-Andreani C.; Morau E.; Schved J.F.; Hédon B.; Dechaud H.: "Amniotic Fluid Embolism with Haemostasis Complications: Primary Fibrinogenolysis or Disseminated Intravascular Coagulation?" Pathophysiol. Of Haemostasis and Thrombosis 33: 170-171, 2003.
- Blanc P.L.; Bethenod J.M.; Muguet D.; Gerard M.; Robert D.: "Un signe prodromique d’embolie amniotique?” La Presse Médicale 14 (41): 2107, 1985.
- Boulton F.E.; Letsky E.: « Obstetric haemorrhage : causes and management « 14 (3) : 683-728, 1985.
- Burrows A.; Khoo S.K.: "The amniotic fluid embolism syndrome: 10 years’ experience at a Major Teaching
hospital" Austr. N.Z. Obstet. Gynecol. 35 (3): 245-250, 1995.
- Cheung A.N.; Luk S.C.: ”The importance of extensive sampling and examination of cervix in suspected cases of amniotic fluid
embolus" Arch. Gynecol. Obstet. 255 (2): 101-105, 1994.
- Choi D.M.A.; Duffy B.L.: ”Amniotic Fluid Embolism” Anaesth.
Intens. Care 23 (6): 741-743, 1995.
- Clark S.L.: "Amniotic fluid embolism" Crit. Care Clin. 7: 877-882, 1991.
- Clark S.L.: "Amniotic Fluid Embolism" Clin. Perinatol. 13: 801-811, 1986.
- Clark S.L.: "New concepts of amniotic fluid embolism: A review" Obstet Gynecol. Surv. 45: 360-368, 1990.
- Clark S.L.: "Successful pregnancy outcomes after amniotic fluid
embolism" Am. J. Obstet. Gynecol. 167 (2): 511-512, 1992.
- Clark S.L.; Cotton D.B.; Gonik B. et al: "Central hemodynamic alterations in amniotic fluid
embolism" Am. J. Obstet. Gynecol. 158: 1124-1126, 1988.
- Clark S.L.; Hankins G.D.V.; Dudley D.; Dildy G.A.; Porter T.F.: "Amniotic fluid embolism: analysis of a national
registry" Am. J. Obstet. Gynecol. 172: 1158-1167, 1995.
- Clark S.L.; Montz F.J.; Phelan J.P.: "Emodynamic alterations associated with amniotic fluid embolism: a
reappraisal" Am. J. Obstet. Gynecol. 151: 617-621, 1985.
- Clark S.L.; Pavlova Z.; Greenspoon J.;Horestein J.; Phelan J.P. : "Squamous cells in the maternal pulmonary
circulation" Am. J. obstet. Gynecol. 154: 104-106, 1986.
- Collier C.: "Recurring amniotic fluid embolism" Anaesth. Int. Care 26 (5): 599-600, 1998.
- Courtney L.D.: "Amniotic fluid embolism" Obstet Gynecol. Surv. 29: 169-177, 1974.
- Courtney L.D.: "Coagulation failure in pregnancy" Br. Med. J. 1: 691, 1970.
- Courtney L.; Allington M.: "Effects of amniotic fluid on blood
coagulation" Br. J. Haematol. 22: 353-355, 1972.
- Cragun J.M.; Ellestad S.C.; Panni M.K.: "Amniotic fluid embolism presenting after uterine manipulation under general anesthesia: A case with a positive
outcome" Resuscitation 77 (1): 145-146, 2008.
- Dashow E.E.; Cotterill R.; Benedetti T.J.: "Amniotic fluid embolus: A report of two cases resulting in maternal
survival" J. Reprod. Med. 34: 660-666, 1989.
- Daver J.; Nicolas F.; Colinot M.; Nicolas G.: "Apropos of 1 case of blood incoagulability in the course of obstretrical shock after uterine rupture and injection of amniotic liquid into the maternal circulation. Contribution to the study of the action of amniotic liquid on
hemostasis" Anesth. Analg. 25 (1): 93-111, 1968.
- Davey A.M.; Becker J.D.; Davis J.M.: "Possible amniotic fluid embolism associated with
amnioinfusion" J. Mat. Fetal Med. 146: 1287-1293, 1992.
- Davies S.: "Amniotic fluid embolus: a review of the literature" Can. J. Anaesth. 48: 88-98, 2001.
- Davies S.: "Amniotic fluid embolism and isolated disseminated intravascular
coagulation" Can J. Anaesth. 46 (5): pt 1: 456-459, 1999.
- Davis D.: "Amniotic Fluid Embolism: Exploring This Rare but Typically Fatal
Condition" AWHONN Lifelines 7 (2): 127-131, 2003.
- Demianczuk C.E.; Corbett T.F.: "Successful pregnancy after amniotic fluid embolism: a case
report" J. Obstet. Gynaecol. Can. 27 (7): 699-701, 2005.
- De Rooij G.M.N.; Gelissen H.P.M.M.; Wester J.P.J.; Spijkstra J.J.; Go A.T.J.J.; Girbes A.R.J.:
"Severe maternal respiratory distress due to the amniotic fluid embolism syndrome in a twin
pregnancy" Neth. J. Med. 61 (10): 337-339, 2003.
- Devriendt J.; Machayekhi S.; Staroukine M.: ”Amniotic fluid embolism: another case with non-cardiogenic pulmonary edema” Int. Care Med. 21 (8): 698-699, 1995.
- Dibble L.A.; Elliot J.P.: "Possible amniotic fluid embolism associated with
amnioinfusion" J. Matern. Fetal Med. 1: 263-266, 1992.
- Dodgson J.; Martin J.; Boswell J.; Goodall H.B.; Smith R.: "Probable amniotic fluid embolus precipitated by amniocentesis and treated by exchange
transfusion" Br. Med. J. 294: 1322-1333, 1987.
- Dorairajan G.; Soundararaghavan S.: "Case report: Maternal death after intrapartum saline amnioinfusion – report of two
cases" Br. J. Obstet Gynaecol. 112: 1331-1333, 2005.
- Dorne R.; Pommier C.; Emery J.C.; Dieudonnè F.; Bongiovanni J.P.: "Embolie de liquide amniotique: evolution favourable après embolisation thérapeutique des artères utérines" ANN. Fr. Anesth. Réanim. 21: 431-435, 2002.
- Drife J.: "Amniotic fluid embolism" in Why mothes die?" 1997-1999, capt. 5.
- Ducloy-Bouthors A.S.; Wantellet A.; Tournoys A.; Depret S.; Krivosic-Horber R.:
"Amniotic fluid embolism suspected in a case of seizure and mild uterine haemorrhage with activation of coagulation and
fibrinolysis" Ann. Fr. Anesth. Reanim.: 23: 149-152, 2004.
- Duffy B.L.: "Does amniotic fluid embolism recur?" Anaesth Int. Care 26 (3): 333, 1998.
- Edwards G.J.; Davies N.J.: "Amniotic fluid embolism following feticide –a cautionary
tale" J. Obstet. Gynaecol. 20 (2): 191, 2000.
- Ellingsen C.L.; Eggebo T.M.; Lexow K.: "Amniotic fluid embolism after blunt abdominal
trauma" Resuscitation 75 (1): 180-183, 2007.
- El Maradny E.; Kanayama N.; Halim A.; Maehara K.; Terao T.: "Endothelin has a role in early pathogenesis of amniotic fluid
embolism" Gynecol Obstet. Invest. 40: 14-18, 1995.
- Esposito R.A.; Grossi E.A.; Coppa G.; Giangola G.; Ferri D.P.; Angelides E.M.; Andriakos P.:
"Successful treatment of post partum shock caused by amniotic fluid embolism with cardiopulmonary bypass and pulmonary artery
thromboembolectomy" Am. J. Obstet. Gynecol. 163: 572-574, 1990.
- Farrar S.C.; Gherman R.B.: ”Serum tryptase in a woman with amniotic fluid embolism” J. Repr. Med. 46 (10): 925-928, 2001.
- Fineschi V.; Gambassi M.; Gherardi M.; Turillazzi E.: "The diagnosis of amniotic fluid embolism: an immunohistochemical study for quantification of pulmonary mast cell
tryptase" Int. J. Legal Med. 111 (5): 238-243, 1998.
- Fletcher S.J.; Parr M.J.A.: "Amniotic fluid embolism: a case report and
review" Resuscitation 43: 141-146, 2000.
- Forna F.; Miles A.; Jamieson D.J.: "Emergency peripartum hysterectomy: A comparison of caesarean and postpartum
hysterectomy" Am. J. Obstet. Gynecol. 190: 1440-1444, 2004.
- Gamerre L.; Tramoni G.; Lhuillier F.; Boisson C.; Clement H.J.; Viale J.P.: "Embolie de liquide amniotique: évolution favorable d'une CIVD isolée et diagnostic biologique précoce" Ann. Fr. Anesth. Reanim. 25: 633-637, 2006.
- Garland I.W.C.; Thompson W.D.: "Diagnosis of amniotic fluid embolism using an antiserum to human
keratin" J. Clin. Pathol. 36: 625-627, 1983.
- Giampaolo C.; Schneider V.; Kowalski B.H. et al: "The cytologic diagnosis of amniotic fluid embolism: A critical
reappraisal" Diagn. Cytopathol. 3: 126-128, 1987.
- Gilbert V.M.; Danielsen B.: "Amniotic fluid embolism: decreased mortality in a population-based
study" Obstet. Gynecol. 93 (6): 973-977, 1999.
- Gillie M.H.; Hughes S.C.: "Amniotic Fluid Embolism" Anesth. Clin. North Amer. 11 (1): 55- 78, 1993.
- Gilmore D.; Wakim J.; Secrest J.; Rawson R.: "Anaphylactoid syndrome of pregnancy: A review of the literature with latest management and outcome
data" AANA J. 71 (2): 120-126, 2003.
- Gogola J.; Hankins G.D.V.: "Amniotic Fluid Embolism in Progress: A Management Dilemma!"
Am. J. Perinat. 15 (8): 491-493, 1998.
- Greene M.F.; Roberts D.J.; Mark E.J.: "Case 9-1998 - Cardiovascular Collapse after Vaginal Delivery in a Patient with a History of Cesarean
Section" N. Eng. J. Med. 338 (12): 821-826, 1998.
- Haba K.; Kumano O.; Mohri M.; Takemaru H.; Kawanishi K.; Tobe S.; Kamada T.; Ueno S.:
"Post-mortem activation of human blood fibrinolytic enzyme in sudden and natural
deaths" Acta Med. Okayama 17: 279-288, 1963.
- Haines J.; Wilkes R.G.: "Non-fatal amniotic fluid embolism after cervical suture
removal" Br. J. Anaesth. 90 (2): 244-247, 2003.
- Hankins G.D.V.; Snyder R.R.; Clark S.L. et al: “Acute hemodynamic and respiratory effects of amniotic fluid embolism in the pregnant goat
model" Am. J. Obstet. Gynecol. 168: 1113-1130, 1993.
- Hasaart T.H.; Essed G.G.: "Amniotic fluid embolism after transabdominal
amniocentesis" Eur. J. Obstet. Gynecol. Reprod. Biol. 16: 25-30, 1983.
- Harnett M.J.P.; Hepner D.L.; Datta S.; Kodali B.S.: "Effect of amniotic fluid on coagulation and platelet function in pregnancy: an evaluation using
thromboelastography" Anaesth. 60: 1068-1072, 2005.
- Hiderazu O.I.; Kobayashi H.; Hirashima Y.; Yamazaki T.; Kobayashi T.; Terao T.:
"Serological and Immunohistochemical Diagnosis of Amniotic Fluid
Embolism" Sem. Thromb. Hemost. 24 (5): 479-484, 1998.
- Hogberg U.; Joelsson I.: "Amniotic Fluid Embolism in Sweden,
1951-1980" Gynecol. Obstet. Invest. 20: 130-137, 1985.
- Hussain S.A.; Sondhi D.S.; Munir A.; Rosner F.: "Amniotic Fluid Embolism with Late Respioratory
Failure" Hospital Physician, 40-43, 2001.
- Hymes A.C.; Robb H.J.; Margulis R.R.: "Influence of an industrial surfactant ( pluronic-F68 ) on human amniotic fluid
embolism" J. Obstet. Gynaec. Brit. Cmnwlth 78: 423-429, 1971.
- Ikarashi T.; Takeuchi S.; Ohnishi Y.: "Trophoblastic embolism in sudden maternal
death" Nippon Sanka Fuji. Gak. Zasshi 40 (6): 793-795, 1988.
- Ishiyama I.; Mukaida M.; Komuro E.; Keil W.: "Analisys of a Case of Generalized Amniotic Fluid Embolism by Demostrating the Fetal Isoantigen (A Blood Type) in Maternal Tissued of B Bllod Type, Using Immunoperoxidase
Staining" Am. J. Clin. Pathol. 85: 239-241, 1986.
- James C.F.; Feinglass D.M.; Menke D.M.; Grinton S.F.; Papadimos T.J.: "Massive amniotic fluid embolism: diagnosis aided by emergency transesophageal
echocardiography" Int. J. Obstet. Anesth. 13: 279-283, 2004.
- Johnson T.R.B.; Abbasi I.A.; Urso P.J.: "Fetal heart rate patterns associated with amniotic fluid
embolism" Am. J. Perinat. 4 (3): 187-190, 1987.
- Josey W.E.: "Hypofibrinogenemia complicating uterine rupture; relationship to amniotic fluid
embolism" Am. J. Obstet. Gynecol. 94 (1): 1966.
- Judich A.; Kuriansky J.; Engelberg I.; Haik J.; Shabtai M.; Czerniak A.:
"Amniotic fluid embolism following blunt abdominal trauma in pregnancy" Iniury 29 (6): 475-477, 1998.
- Kanayama N.; Yamazaki T.; Naruse H. et al: "Determining zinc-coproporphyrin in maternal plasma: A new method for diagnosing amniotic fluid
embolism" Clin. Chem. 38: 526-529, 1992.
- Karetzky M.; Ramirez M.: "Acute respiratory failure in pregnancy: an analysis of 19
cases" Medicine 77: 41-49, 1998.
- Kastner E.S.; Figueroa R.; Garry D.; Maulik D.: "Emergency Peripartum Hysterectomy: Experience at a Community Teaching
Hospital" Obstet. Gynecol. 99 (6): 971-975, 2002.
- Katchy K.C.; Ziad F.; Al Nashmi N.; Diejomaoh M.F.E.: "Emergency obstetric hysterectomy in Kuwait: a clinico pathological
analysis" Arch. Gynecol. Obstet. 273: 360-365, 2006.
- Kern S.B.; Duff P.: "Localized Amniotic Fluid Embolism Presenting as Ovarian Vein Thrombosis and Refractory Postoperative
Fever" Am. J. Clin. Path. 76 (4): 476-480, 1981.
- Khong T.Y.: "Expression of endothelin-1 in amniotic fluid embolism and possible pathophysiological
mechanism" J. Obstet. Gynecol. 105 ( 7): 802-804, 1998.
- Killam A.: ”Amniotic Fluid Embolism" Clin. Obstet. Gynecol. 28: 32-36, 1985.
- Knight M.: "Amniotic fluid embolism: active surveillance versus retrospective database
review" Am. J. Obstet. Gynecol. Letters to the Editors, 2008.
- Kobayashi H.; Ohi H.; Terao T.: "A Simple, Noninvasive, Sensitive Method for Diagnosis of Amniotic Fluid Embolism by Monoclonal Antibody TKH-2 That Recognizes
NeuAcalpha2-6GalNAc" Am. J. Obstet. Gynecol. 168: 848-853, 1993.
- Kobayashi H.; Ooi H.; Hayakawa H.; Arai T.; Matsuda Y.; Gotoh K.; Tarao T.:
"Histological Diagnosis of Amniotic Fluid Embolism by Monoclonal antibody TKH-2 That Recognizes NeuAcalpha2-6GalNAc
Epitope" Human Path. 28 (4): 428-433, 1997.
- Koegler A.; Sauder P.; Marolf A.; Jaeger A.: "Amniotic fluid embolism: a case with non-cardiogenic pulmonary
edema" Int. Care Med. 20 (1): 45-46, 1994.
- Kool M.J.: "Letters To the Editors". Am. J. Obstet. Gynecol. 164 ( 2): 701-702, 1991.
- Kramer M.; Rouleau J., Baskett T.F.; Joseph K.S.: "Amniotic-fluid embolism and medical induction of labour: a retrospective, population-based cohort
study" Lancet 368: 2006.
- Kuczynski J.; Uszynski W.; Zekanowska E.; Soszka T.; Uszynski M.: "Tissue factor ( TF ) and tissue factor pathway inhibitor ( TFPI ) in the placenta and
myometrium" Eur. J. Obstet. Gynecol. Repr. Biol. 105: 15-19, 2002.
- Laforga J.B.: "Amniotic fluid embolism. Report of two cases with coagulation
disorder" Acta Obstet. Gynecol. Scand. 76 (8): 805-806, 1997.
- Landing B.H.: "The pathogenesis of amniotic fluid embolism: uterine
factors" New. Eng. J. Med. 19: 590-596, 1950.
- Lau G.; Chui P.P. : "Amniotic Fluid Embolism: a review of 10 fatal
cases" Singapore Med. J. 35 (2): 180-183, 1994.
- Lau G.: "Amniotic Fluid Embolism as a Cause of Sudden Maternal
Death" Med. Sci. 34 (2): 1994.
- Lee W.; Ginsburg K.A.; Cotton D.B.; Kaufman R.H. : "Squamous and trophoblastic cells in the maternal pulmonary circulation identified by invasive hemodynamic monitoring during the peripartum
period" Am. J. Obstet. Gynecol. 155: 999-1001, 1986.
- Levy R.; Furman B.; Hagay Z.J.: "Fetal bradycardia and disseminated coagulopathy: atypical presentation of amniotic fluid
embolism" Acta Anaesth. Scand. 48: 1214-1215, 2004.
- Liban E.; Raz S.: "A clinicopathologic study of forteen cases of amniotic fluid
embolism" Am. J. Clin. Pathol. 51: 477-486, 1969.
- Liu E.H.C.; Shailaja S.; Koh S.C.L.; Lee T.L.: "An assessment of the effects on coagulation of mid-trimester and final-trimester amniotic fluid on blood by Thromboelastograph
analisys" Anesth. Analg. 50: 333-336, 2000.
- Lookwood C.J.; Bach R.; Guha A.; Zhou X. ; Miller W.A.; Nemerson Y.: "Amniotic fluid contains tissue factor, a potent initiator of
coagulation" Am. J. Obstet. Gynecol. 165: 1335-1341, 1991.
- Lumley J.; Owen R.; Morgan M.: "Amniotic fluid embolism: a report of three
cases" Anaesthesia, 34: 33-36, 1979.
- Maher J.E., Wenstrom K.D.; Hauth J.C.; Meis P.J.: "Amniotic fluid embolism after saline amnioinfusion: two cases and review of the
literature" Obstet. Gynecol. 83: 851-854, 1994.
- Malvino E.; Korin J.; Bruno M.; Medrano J.; Miano J.: "Angiografìa pulmonar en un caso de embolia de lìquido amniòtico y revisiòn de la
literatura" Medicina Intensiva, 18 (2): 52-57, 2002.
- Manchanda R.; Sriemevan A.: "Anaphylactoid syndrome caused by amniotic fluid embolism following manual removal of
placenta" Am. J. obstet. Gynaecol. 25 (2): 201-202, 2005.
- Margarson M.P.: "Delayed amniotic fluid embolism following Caesarean section under spinal
anaesthesia" Anaesthesia 50: 804-806, 1995.
- Martin R.W.: “Amniotic Fluid Embolism" Clin. Obstet. Gynecol. 39 (1): 101-106, 1996.
- Masson R.G.: "Amniotic fluid embolism" Clin. Chest Med. 13 (4): 657-665, 1992.
- Masson R.G.; Ruggieri J.; Siddiqui M.M.: "Pulmonary microvascular cytology. A new diagnosis application of the pulmonary artery catheter
“. Chest 88: 908-914, 1985.
- Mathelier A.C.; Karachorlu K.: "Placenta previa and accrete complicated by amniotic fluid
embolism" Int. J. Fertil. Womens Med. 51 (1): 28-32, 2006.
- Mato J.: "Suspected Amniotic Fluid Embolism Following Amniotomy: A Case
Report" A.A.N.A. J. 76 (1): 53-59, 2008.
- McDougall R.J.; Duke G.J.: "Amniotic Fluid Embolism Syndrome : Case Report and
Review" Anaesth. Intens. Care 23: 735-740, 1995.
- Meyer J.R."Embolis pulmonary caseosa" Braz. J. Med. Biol. Res. 1: 301-303, 1926.
- Mole R.H.: "Fibrinolysis and the fluidity of the blood post mortem" J. Pathol. 60: 413, 1948.
- Moore J.: "Amniotic fluid embolism : on the trail of an elusive
diagnosis" Lancet 368: 1399-1401, 2006.
- Moore J.; Baldisseri M.R.: "Amniotic fluid embolism" Crit. Care Med. 33 (10) Suppl: S279-S285, 2005.
- Moore L.E.; Ware D.: "Amniotic Fluid Embolism" emedicine 2004. www.emedicine.com/med/topic122.htm.
- Morgan M.: "Amniotic.Fluid Embolism" Anaesthesia 34: 20-32, 1979.
- Morioka H.; Haba K.; Tanaka M.; Takemaru H.; Kawanishi K.; Tobe S.; Kamada T.; Ueno S.:
"Studies on fluidity of blood on the basis of fibrinolysis. II. Fibrinolytic activity of post-mortem blood in death from
disease" Nihon Hoig. Zasshi 19: 58-67, 1965.
- Nishio H.; Matsui K.; Miyazaki T.; Tamura A.; Iwata M.; Suzuki K.: "A fatal case of amniotic fluid embolism with elevation of serum mast cell
tryptase" Forensic Sc. Int. 126 (1): 53-56, 2002.
- Olcott C.O.; Robinson A.J.; Maxwell T.M et al. : "Amniotic fluid embolism and disseminated intravascular coagulation after blunt maternal
trauma" J. Trauma 13: 737-740, 1973.
- Paterson W.G.; Grant K.A.; Grant J.M.; McLean N.: "The pathogenesis of amniotic fluid embolism with particular reference to transabdominal
amniocentesis" Eur. J. Obstet. Gynecol. Repr. Biol. 7: 319-324, 1977.
- Perozzi K.J.; Englert N.C.: "Amniotic Fluid Embolism An Obstetric
Emergency" Crit. Care Nurse 24 (4): 2004.
- Peterson E.P.; Taylor H.B.: "Amniotic Fluid Embolism: an analysis of 40
cases" Obstet. Gynaecol., 35: 787-793, 1970.
- Petroianu G.; Altmannsberger S.H.; Maleck W.; Assmus H.P.; Friedberg C., Bergler W.; Rufer R.:
"Meconium and amniotic fluid embolism: Effects on coagulation in pregnant
mini-pigs" Critical Care Med. 27 (2): 348-355, 1999.
- Petrone P.; Asensio J.A.: "Trauma in pregnancy: assessment and
treatment" Scand. J. Surg. 95 (1): 4-10, 2006.
- Philips L.L.; Davidson E.C.: "Procoagulant properties of amniotic
fluid" Am.J. Obstet. Gynecol. 113: 911-919, 1972.
- Pluymarkers C.; De Weerdt A.; Jacquemyn Y.; Colpaert C.; Van de Poel E.; Jorens P.G.:
"Amniotic fluid embolism after surgical trauma: Two case reports and review of the
literature" Resuscitation 72 (2): 324-332, 2007.
- Porat S.; Leibowwitz D.; Milwidsky A.; Valsky D.V.; Yagel S.; Anteby E.Y.:
"Transient intracardiac thrombi in amniotic fluid embolism" Br. J. Obstet. Gynaecol. 111: 506-510, 2004.
- Porter TF; Clark SL; Dildy GA; Hankins GDV: "Isolated disseminated intravascular coagulation and amniotic fluid
embolism" Am. J. Obstet. Gynec. 174: 486, 1997.
- Price T.M.; Baker V.V.; Cefalo R.C.: "Amniotic Fluid Embolism. Three Case Reports with a Review of the
Literature" Obstet. Gynecol. Surv. 40 (7): 462-465, 1985.
- Prosper S.C.; Goudge C.S.; Lupo V.R.: "Recombinant Factor VIIa to Successfully Manage Disseminated Intravascular Coagulation From Amniotic Fluid
Embolism" Obstet. Gynecol. 109 (2): part 2; 524-525, 2007.
- Quance D.: "Amniotic Fluid Embolism: Detection by Pulse Oximetry" Anesthesiology 68: 951-952, 1988.
- Quinn A.; Barrett T.: "Delayed onset of coagulopathy following amniotic fluid embolism : two case
reports" Int. J. Obstet. Anaesth. 2: 177-180, 1993.
- Raijnio J.; Penttila A.: "Amniotic fluid embolism as cause of death in a car accident- A case
report" Forensic Sci. Int. 137: 231-234, 2003.
- Rashid A.M.; Moir C.L.; Butt J.C.: "Sudden death following caesarean section for placenta previa and
accreta" Am. J. Forens. Med. Pathol. 15 (1): 32-35, 1994.
- Ratten G.J.."Amniotic Fluid Embolism - 2 Case Reports and a Review of Maternal Deaths From This Cause in
Australia" Austr. N.Z. J. Obstet. Gynecol. 28: 33-35, 1988.
- Ray B.K.; Vallejo M.C.; Creinin M.D.; Shannon K.T.; Mandell G.L.; Kaul B.; Ramanathan S.:
"Amniotic fluid embolism with second trimester pregnancy termination : a case
report" Can. J. Anesth. 51(29): 139-144, 2004.
- Resnik R.: "Amniotic fluid embolism" Contemporary OB/GYN Archive, 1996.
- Ricou B.; Reper P.; Suter P.M.: "Rapid diagnosis of amniotic fluid embolism causing severe pulmonary
failure" Intens. Care Med. 15: 129-131, 1989.
- Robillard J.; Gauvin F.; Molinaro G.; Leduc L.; Adam A.; Rivard G.E.:
"The syndrome of amniotic fluid embolism: A potential contribution of
bradykinin" Am. J. Obstet. Gynecol. 193: 1508-1512, 2005.
- Roche W.D.; Norris H.J.: "Detection and significance of maternal pulmonary amniotic fluid
embolism" Obstet. Gynecol. 43: 729-731, 1974.
- Romero R.; Emamian M.; Wan M. et al: "Increased concentrations of arachidonic acid lipoxygenase metabolites in amniotic fluid during
parturition" Obstet. Gynecol. 70: 849-851, 1987.
- Rufforny-Doudenko I.; Sipp C.; Shehata B.M.: "Pathologic Quiz Case: A 30-Year-Old Woman With Severe Disseminated Intravascular Coagulation During
Delivery" Arch. Pathol. Lab. Med. 126: 869-870, 2002.
- Rushton D.I.; Dawson I.M.P.: "The maternal autopsy" J. Clin. Pathol. 35: 909-921, 1982.
- Russell W.; Jones W.N.: "Amniotic Fluid Embolism: a review of the Syndrome with a Report of 4
Cases" Obstet. Gynecol. 26 (4): 476-485, 1965.
- Sagava N.; Hasegawa M.; Itoh H.: "The role of amniotic endothelin in human
pregnancy" Placenta 15: 562-575, 1994.
- Salem H.H.; Walter W.A.; Perkin J.L.: "Aggregation of human platelets by amniotic
fluid" Br. J. Obstet. Gynaecol. 89: 733, 1982.
- Scott M.: "Cardiopulmonary Considerations in Nonfatal Amniotic Fluid
embolism" J.A.M.A. 183 (12): 989-993, 1963.
- Senianskaia N.L.: "Electron microscopic study of post-mortem
fibrinolysis" Arkh. Patol. 47 (2): 58-65, 1985.
- Shnider S.L.; Moya F.: "Amniotic Fluid Embolism" Aneshesiology 22 (1): 108-119, 1961.
- Shojai R.; Chau C.; Boubli L.; d’Ercole C. : "Amniotic fluid embolism during late term termination of pregnancy"
Pren. Diagn. 23: 944-951, 2003.
- Sistac Ballarin J.M.; Gomez Agraz J.L.; Sanz Martin P.J.; Martinez Salavera
T.: "Amniotic fluid embolism during the birth process. Apropos of an atypical
presentation" Re. Esp. Anestesiol. Reanim. 47 (9): 424-425, 2000.
- Spapen H.: "Use of the Swan-Ganz catheter in amniotic fluid
embolism" Int. Care Med. 14 (6): 678, 1988.
- Sparr R.A.; Pritchard J.A.: "Studies to detect the escape of amniotic fluid into the maternal circulation during
parturition" Surg. Gynecol. Obstet. 107: 560-564, 1958.
- Stafford I.; Sheffield J.: "Amniotic Fluid Embolism" Obstet. Gynecol. Clin. North Am. 34: 545-553, 2007.
- Steiner P.E.; Lushbaugh C.C.: "Maternal pulmonary embolism by amniotic fluid as a cause of obstetric shock and unexpected death in
obstetrics" J.A.M.A. 117: 1245-1254, 1341-1345, 1941.
- Stiller R.J.; Siddiqui D.; Laifer S.A.; Tiakowski R.L.; Whetham J.C.:
"Successful pregnancy after suspected anaphylactoid syndrome of pregnancy (amniotic fluid embolus). A case
report" J. Reprod. Med. 45 (12): 1007-1009, 2000.
- Stolte L.A.M.; Lim H.T.; Van Arkel C.; Eskes T.K.A.B.; van Kessel H.: “Activation of coagulation, stasis and amniotic fluid embolism
syndrome" Eur. J. Obstet Ginaecol. 219: 1971.
- Stolte L.A.M.; Lim H.T.; Van Arkel C.; Eskes T.K.A.B.; van Kessel H.I.A.M.; De Leeuw J.H.A.:
"Amniotic Fluid Embolism" in Amniotic Fluid “2nd revised edition. Excerpta Medica, Amsterdam 435-452, 1978.
- Strickland M.A.; Johnson M.P.; Whitworth N.S. et al: "Procoagulant properties of amniotic fluid from labouring and non-laboring
women" ACOG Annual Clinical Meeting, 1982.
- Strickland M.A.; Bates G.W.; Whitworth N.S.; Martin J.N.: "Amniotic Fluid Embolism: Prophylaxis With Heparin and
Aspirin" South. Med. J. 78 (4): 377-379, 1985.
- Styron A.G.; George R.B.; Allen T.K.; Peterson-Layne C.; Muir H.A.: "Multidisciplinary management of placenta percreta complicated by embolic
phenomena" Int. J. Obstet. Anesth. 17: 262-266, 2008.
- Tews G.; Yaman C.; Ebner T.: "Fatal trophoblastic embolism during cesarean
section" Int. J. Gynecol. Obstet. 76: 179-180, 2002.
- Thomsen R.J. : "Pregnancy in the presence of an intrauterine contraceptive device and amniotic fluid embolism: a possible
association" Am. J. Obstet. Gynecol. 15: 121 (2): 279-280, 1975.
- Thompson W.B.; Budd J.W.: "Erroneous diagnosis of amniotic fluid
embolism" Am. J. Obstet. Gynecol. 91: 606…, 1963.?????????
- Tramoni G.; Valentin S.; Robert M.O.; Sergeant M.V..; Branche P.; Duperret S..; Clement H.J.; Lopez F.; Boisson C.; Audra P..; Rudigoz R.C.; Viale J.P.:
"Amniotic fluid embolism during caesarean section" Int. J. Obstet. Anaesth. 13 ( 4 ): 271-274, 2004.
- Tuffnell D.J.: "Amniotic fluid embolism" Curr. Opin. Obstet. Gynaecol. 15: 119-122, 2003.
- Tuffnell D.J.: "Managing amniotic fluid embolism" OBGManagement.com 2005.
- Tuffnell D.J.:” United Kingdom Amniotic Fluid Embolism Register" Br. J. Obstet. Gynaec. 112: 1625-1629, 2005.
- Tuffnell D.J.; Johnson H.: "Amniotic fluid embolism: the UK
register" Hosp. Med. 61 (8): 532-534, 2000.
- Uszynski M.; Zekanowska E.; Uszynski W.; Kuczynski J.: "Tissue factor ( TF ) and tissue factor pathway inhibitor (TFPI) in amniotic fluid and blood plasma: implications for the mechanism of amniotic fluid
embolism" Eur. J. Obstet Gynaecol. Repr. Biol. 95: 163-166, 2001.
- Vanmaele L.; Noppen M.; Vincken L.; De Catte L.; Huyghens L.: "Transient left hearth failure in amniotic fluid embolism"
Intens. Care Med. 16: 269-271, 1990.
- Viscomi C.M.; Clark S.L.: "Amniotic Fluid Embolism" in Obstetric Anestesia" 2nd Edition, edited by M.C. Norris, Lippincott Williams & Wilkins, Philadelphia, 1999.
- Vlies R.: "Amniotic fluid embolism" Why Mothers die ? 2000-2002, cap 5.
- Vlies R.: "Amniotic fluid embolism" Savig Mothers 2003-2005, 86-90.
- Yang J.I.; Kim H.S.; Chang K.H.; Ryu H.S.; Joo H.J.: "Amniotic fluid embolism with isolated coagulopathy: a case
report" J. Reprod. Med. 51 (1): 64-66, 2006.
- Yentis S.: "Amniotic fluid embolism" Can. J. Anesth. 48: 829-830, 2001.
- Wasser W.G.; Tessler S.; Kamath C.P.; Sackin A.J.: "Nonfatal amniotic fluid embolism: a case report of post-partum respiratory distress with histopathologic
studies" Mt. Sinai J. Med. 46 (4): 388-391, 1979.
- Wegelius G.; Bergstrom M.; Ahlbom L.; Thomassen P.: "A case of life-threatening pulmonary edema associated with amnioinfusion during
labor" Eur. J. Obstet. Gynaecol. Reprod. Biol. 65 (2): 237-239, 1996.
- Weiner C.P.: "The obstetric patient and disseminated intravascular
coagulation" Clin. Perinat. 13 (4): 705-717, 1986.
- Weiner A.E.; Reid D.E.: "The pathogenesis of Amniotic Fluid Embolism. III. Coagulant activity of amniotic
fluid" New Engl. J. Med. 243: 597-598, 1950.