Critical Care Obstetrics part 58 potx

Immediate postpartum hemorrhage PPH, defi ned as excessive blood loss within 24 hours after childbirth, is the single most important cause of maternal death worldwide, accounting for alm

Critical Care Obstetrics, 5th edition Edited by M Belfort, G Saade,

M Foley, J Phelan and G Dildy © 2010 Blackwell Publishing Ltd.

Scott Roberts

Department of Obstetrics and Gynecology, The University of Texas Southwestern Medical Center (UTSMC) at Dallas, TX, USA

Introduction

Hemorrhage is one of the leading causes of pregnancy - related

mortality in the United States (2.0/100 000 live births) second

only to embolism (2.3/100 000 live births) (Table 40.1 ) [1]

Almost 99% of maternal deaths occur in developing countries

Immediate postpartum hemorrhage (PPH), defi ned as excessive

blood loss within 24 hours after childbirth, is the single most

important cause of maternal death worldwide, accounting for

almost half of all postpartum maternal deaths in developing

countries [2,3]

In the United States, hemorrhage was the leading cause of

death after stillbirth (from abruptions and uterine rupture), and

accounted for 93% of deaths associated with ectopic pregnancies

Hemorrhage was also prominent as a cause of death in

pregnan-cies ending in induced or spontaneous abortion (21.8%) [1]

These deaths are mediated through hypovolemic shock which is

also responsible for a number of other serious non - fatal

compli-cations, including acute renal failure, acute respiratory distress

syndrome (ARDS), and more rarely, postpartum pituitary

necro-sis The parturient undergoes several important physiologic

adaptations during pregnancy to protect her from the bleeding

expected at the time of delivery Peripartum complications can

occur quickly and since the uterus receives a blood fl ow of 450 –

650 mL/min quick, decisive, and coordinated action on the part

of the practitioner and supporting staff can be life - saving [4]

Shock is perhaps best defi ned as reduced tissue oxygenation

resulting from poor perfusion [5] Low fl ow or unevenly

distrib-uted fl ow from hypovolemia and disproportionate

vasoconstric-tion are major causes of inadequate tissue perfusion in the acutely

ill patient with circulatory dysfunction or shock In hemorrhagic

shock, the disparity is a result of blood loss that leads to both

compensatory neurohormonal activation as well as the release of

various endogenous mediators, which may aggravate the primary physiologic effects of hypovolemia [6 – 8] Because the purpose of the circulation is to provide oxygen and oxidative substrates for metabolic requirements, insuffi cient tissue perfusion and oxy-genation to support body metabolism is the common circulatory problem of acute critical illness This inadequate perfusion leads

to local tissue hypoxia, organ dysfunction, multiple organ failure, and death

Blood fl ow to the capillary beds of various organs is controlled

by arterioles, which are resistance vessels that in turn are con-trolled by the CNS On the other hand, 70% of the total blood volume is contained in venules, capacitance vessels controlled by humoral factors Hypovolemic shock evolves through several pathophysiologic stages as body mechanisms combat acute blood volume loss (Table 40.2 ) The diagnosis of shock is most often made by the presence of hypotension, oliguria, acidosis, and col-lapse in the late stage, when therapy is frequently ineffective Early

in the course of massive hemorrhage, there are decreases in mean arterial pressure (MAP), cardiac output (CO), central venous pressure (CVP), pulmonary capillary wedge pressure (PCWP), stroke volume and work, mixed venous oxygen saturation, and oxygen consumption Increases are seen in systemic vascular resistance (SVR) and arteriovenous oxygen content differences These latter changes serve to improve tissue oxygenation when blood fl ow is reduced [9] Catecholamine release also causes a generalized increase in venular tone, resulting in an autotransfu-sion effect from the capacitance reservoir These changes are accompanied by compensatory increases in heart rate, SVR and pulmonary vascular resistance, and myocardial contractility Redistribution of CO and blood volume occurs via selective arte-riolar constriction mediated by the CNS This results in dimin-ished perfusion to the kidneys, gut, skin, and uterus, with relative maintenance of blood fl ow to the heart, brain, and adrenal glands

In the pregnant patient, such redistribution may result in fetal hypoxia and distress, even before the mother becomes overtly hypotensive In such situations, the uterus is, from a teleologic viewpoint, relatively less important than the essential life - saving organs systems Regardless of the absolute maternal BP,

signifi cant maternal shock is highly unlikely in the absence of fetal distress [10] Peripheral vasoconstriction caused by the adreno-medullary stress response is an initial reaction to blood loss that maintains pressure in the presence of decreasing fl ow This vaso-constriction, however, is disparate and leads to unevenly distrib-uted microcirculatory fl ow These early changes precede the development of organ failure In the presence of continued hypo-volemia, the stress response may result in poor tissue perfusion, tissue hypoxia, covert clinical shock, organ dysfunction, ARDS, and other organ failure [11,12]

As the blood volume defi cit approaches 25%, such compensa-tory mechanisms become inadequate to maintain CO and arterial pressure At this point, small additional losses of blood result in rapid clinical deterioration, producing a vicious cycle of cellular death and vasoconstriction, organ ischemia, loss of capillary membrane integrity, and additional loss of intravascular fl uid volume into the extravascular space [13,14]

Increased platelet aggregation is also found in hypovolemic shock Aggregated platelets release vasoactive substances, which cause small vessel occlusion and impaired microcirculatory per-fusion These platelet aggregates can embolize to the lungs and

be a factor contributing to respiratory failure, which is often seen following prolonged shock

Table 40.2 Clinical classifi cation of maternal hemorrhage

Class Blood loss

(mL)

Volume defi cit (%)

Signs and symptoms

I ≤ 1000 15 Orthostatic tachycardia ( ↑ 20 bpm)

II 1001 – 1500 15 – 25 ↑ HR 100 – 120 bpm

Orthostatic changes ( ↓ 15 mmHg) Cap refi ll > 2 sec

Mental changes III 1501 – 2500 25 – 40 ↑ HR (120 – 160 bpm)

Supine ↓ BP

↑ RR (30 – 50 rpm) Oliguria

IV > 2500 > 40 Obtundation

Oliguria - anuria

CV collapse

BP, blood pressure; bpm, beats per minute; CV, cardiovascular; rpm, respirations

per minute; RR, respiratory

From Eisenberg M, Copass MK, eds Emergency Medical Therapy Philadelphia:

WB Saunders, 1982: 40

Table 40.1 Causes of pregnancy - related death, by outcome of pregnancy and pregnancy - related mortality ratios (PRMR * ) – United States, 1991 – 1999

Cause of death Outcome of pregnancy

(% distribution)

All outcomes (n = 4200) Livebirth Stillbirth Ectopic Abortion † Molar Undelivered Unknown Percent PRMR (n = 2519) (n = 275) (n = 237) (n = 165) (n = 14) (n = 438) (n = 552)

Total † † 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

* Pregnancy - related deaths per 100 000 livebirths

† Includes spontaneous and induced abortions

§ Pregnancy - induced hypertension

¶ Cerebrovascular accident

* * The majority of the other medical conditions were cardiovascular, pulmonary, and neurologic problems

† † Percentages might not add to 100.00 because of rounding

From Centers for Disease Control and Prevention Pregnancy - related mortality surveillance – United States, 1991 – 1999 MMWR 2003; 52: 55 – 62

gland may undergo ischemic necrosis Sheehan and Murdoch

fi rst described the syndrome of hypopituitarism secondary to postpartum hypotension as result of hemorrhage [19] This con-dition is now a rare complication in modern obstetrics The clini-cal presentation can vary, but secondary amenorrhea resulting from loss of pituitary gonadotrophs is usually present In severe cases, thyrotropic and adrenotropic pituitary hormones also may

be defi cient A typical or partial defi ciency syndrome of both anterior and posterior pituitary hormones has been reported Hypovolemia from any cause leads to reduced renal perfusion, which can result in acute tubular necrosis In one series, hemor-rhage and hypovolemia were precipitating factors in 75% of obstetric patients with acute renal failure [20] Prompt blood and

fl uid replacement is essential in order to avoid such sequelae Lung injury may result from hypovolemic shock [21] In the non pregnant state, a critical cardiac output exists below which oxygen extraction becomes impaired, and this critical oxygen delivery has been implicated in the pathogenesis of ARDS in humans The question of a critical oxygen delivery point in human pregnancy

is unclear although it has been suggested as a component of the pathology of severe pre - eclampsia [22] Evans and colleagues pre-sented evidence that in the pregnant sheep model, such a critical cardiac output does not exist [23]

Causes of o bstetric h emorrhage

Any disruption in the integrity of the maternal vascular system during pregnancy has the potential for devastating blood loss

As an overview, ectopic pregnancy is the leading cause of life threatening obstetric hemorrhage in the fi rst half of gestation (see Table 40.1 ) Beyond the fi rst trimester, antepartum obstetric hemorrhage usually results from a disruption of the placental attachment site (involving either a normally implanted placenta

or placenta previa) or uterine rupture (spontaneous or trauma related) During the intrapartum period, the likelihood of clini-cal shock is enhanced in patients with pre - eclampsia Because

of the intravascular volume depletion associated with this condition, even the usual blood loss associated with delivery may result in clinical instability Another pathophysiologic change often associated with pre - eclampsia is thrombocytope-nia, which when severe, may contribute to postpartum blood loss [24]

Most serious obstetric hemorrhage occurs in the postpartum period The most common cause is uterine atony following pla-cental separation Under normal conditions, shortening myome-trial fi bers act as physiologic ligatures around the arterioles of the placental bed Thus, uterine atony with failure of myometrial contraction results in arterial hemorrhage Factors that predis-pose a patient to uterine atony include precipitous or prolonged labor, oxytocin augmentation, magnesium sulfate infusion, cho-rioamnionitis, enlarged uterus resulting from increased intra-uterine contents, and operative deliveries [10,25]

Physiologic c hanges in p reparation for

p regnancy b lood l oss

The pregnant woman undergoes profound physiologic changes

to prepare for the blood loss that will occur at the time of

parturi-tion By the end of the second trimester of pregnancy, the

mater-nal blood volume has increased by 1000 – 2000 mL [15] The

maternal CO increases by 40 – 45% while total peripheral

resis-tance decreases [16] This decreased peripheral resisresis-tance results

from hormonal factors (progesterone, and prostaglandin

metab-olites such as prostacyclin) that reduce overall vasomotor tone

and from the development of a low - resistance arteriovenous

shunt through the placenta The decreased peripheral resistance

is maximal in the second trimester About 20 – 25% of the

mater-nal CO goes to the placental shunt to yield a blood fl ow of

approximately 500 mL/min Placental blood fl ow is directly

portional to the uterine perfusion pressure, which in turn is

pro-portional to systemic BP Any decrease in maternal CO results in

a proportionate decrease in placental perfusion The uterine

arte-rioles are very sensitive to exogenous vasopressor substances but

because of an incompletely understood pregnancy - related

stimu-lus of the renin – angiotensin system, the vasopressor effect of

angiotensin appears to be blunted during pregnancy [17] Thus,

during her pregnancy, the mother has been prepared for a blood

volume loss of up to 1000 mL Following a normal spontaneous

vaginal delivery, a fi rst - day postpartum hematocrit usually is not

altered signifi cantly from the admission hematocrit In practice,

blood loss at delivery is often underestimated Actual

measure-ments show that the average blood loss after normal spontaneous

vaginal delivery is over 600 mL [18] With a postpartum blood

loss of less than 1000 mL, the parturient ’ s vital signs may refl ect

acute blood loss (i.e., hypotension and tachycardia)

During the antepartum period, the obstetrician must be

con-cerned with both patients Fetal oxygenation decreases in

propor-tion to the decrease in maternal CO The catecholamine output

from the mother ’ s adrenal medulla may preferentially increase

arteriolar resistance of the spiral arterioles in the placental bed,

thus further decreasing oxygenation Under such circumstances,

the fetus may be in jeopardy, even though compensatory

mecha-nisms maintain stable maternal vital signs Thus, even in the

absence of overt hypotension, the healthcare team must act

quickly to preserve fetal well - being by expanding the

intravascu-lar volume of an antepartum patient who has lost a signifi cant

amount of blood

Although all vital organs receive increased blood fl ow during

pregnancy, three organs (other than the placenta) are particularly

susceptible to damage when perfusion pressure decreases as a

result of hemorrhagic shock These organs are the anterior

pitu-itary gland, the kidneys, and the lungs During pregnancy, the

anterior pituitary enlarges and receives increased blood fl ow

Under the condition of shock, blood fl ow is shunted away from

the anterior pituitary gland As a result, the anterior pituitary

Oxygenation

The most frequent cause of maternal death from shock is inad-equate respiratory exchange leading to multiple organ failure [11] The duration of relative tissue hypoxia is important in the accumulation of byproducts of anaerobic metabolism Thus, increasing the partial pressure of oxygen across the pulmonary capillary membrane by giving 8 – 10 L of oxygen per minute by tight - fi tting mask may forestall the onset of tissue hypoxia and is

a logical fi rst priority Also, increasing the partial pressure of oxygen in maternal blood will increase the amount of oxygen carried to fetal tissue [27] If the airway is not patent, or the tidal volume is inadequate, the clinician should not hesitate to perform endotracheal intubation and institute positive - pressure ventila-tion to achieve adequate oxygenaventila-tion

Studies in adult critical care indicate that tissue oxygen debt resulting from reduced tissue perfusion is the primary underlying physiologic mechanism that subsequently leads to organ failure and death [28,29] It seems that early identifi cation and treatment

of hypovolemic shock and its inciting cause is imperative to improving outcome One approach commonly used to assist the clinician is to classify the degree of hemorrhage from I to IV based

on the patient ’ s signs and symptoms (Table 40.2 )

Volume r eplacement

Protracted shock appears to cause secondary changes in the microcirculation; and these changes affect circulating blood volume In early shock, there is a tendency to draw fl uid from the interstitial space into the capillary bed As the shock state progresses, damage to the capillary endothelium occurs and is manifested by an increase in capillary permeability Capillary permeability further accentuates the loss of intravascular volume This defi cit is refl ected clinically by the disproportionately large volume of fl uid necessary to resuscitate patients in severe shock Sometimes, the amount of fl uid required for resuscitation is twice

Obstetric trauma is another common cause of postpartum

hemorrhage Cervical and vaginal lacerations are more common

with midpelvic operative deliveries, and as a consequence of an

extension of a uterine incision for cesarean birth Other causes of

postpartum hemorrhage (Table 40.3 ) include uterine inversion,

morbidly adherent placenta (accreta/percreta), amniotic fl uid

embolism, retroperitoneal bleeding from either birth trauma or

episiotomy, and coagulopathies of various causes [10,25,26]

Management of h ypovolemic s hock

in p regnancy

Fundamentally the most important and prerequisite

manage-ment tool in approaching hypovolemic shock is a complete

understanding of maternal blood volume and how that volume

is affected by pregnancy In 1989, Clark et al presented

central hemodynamic parameters of normal - term pregnancy

and contrasted them with non - pregnant values (Table 40.4 )

The calculation and demonstration of a 50% increased blood

volume in term pregnancy was delineated by Pritchard et al in

1965 [15]

So we are to understand that the average pregnant woman has

4.5 – 5 L of total blood volume, not 3 – 3.5 L as in the non - pregnant

state Further, there is a rise in CO of 50% in the term patient,

a result of an increased heart rate and stroke volume There

is also a dramatic decrease in SVR and pulmonary vascular

resistance Clark et al were not able to document increases in

left ventricular contractility and we are left with the knowledge

that pregnancy is not a hyperdynamic state, but rather a fi nely

written (evolved) and adapted symphony of perfect resilience

and capacity to perpetuate the gestation Excesses have also been

built into the system to withstand the blood loss of labor and

delivery After delivery, the low resistance placental shunt is

turned off and a process of autotransfusion helps to replenish lost

volume from the delivery phase We are fortunate to have

repro-duction occur at the zenith of human health in the early adult

years

Table 40.3 Common causes of obstetric hemorrhage

Antepartum and intrapartum

Placental abruption

Uterine rupture

Placenta previa

Postpartum

Retained placenta

Uterine atony

Uterine rupture

Genital tract trauma

Coagulopathy

Table 40.4 Central hemodynamic changes

Cardiac output (L/min) 4.3 ± 0.9 6.2 ± 1.0 Heart rate (beats/min) 71 ± 10.0 83 ± 10.0 Systemic vascular resistance (dyne/cm/sec − 5

) 1530 ± 520 1210 ± 266 Pulmonary vascular resistance (dyne/cm/sec − 5 ) 119 ± 47.0 78 ± 22 Colloid oncotic pressure (mmHg) 20.8 ± 1.0 18.0 ± 1.5 Colloid oncotic pressure – pulmonary capillary

wedge pressure (mmHg)

14.5 ± 2.5 10.5 ± 2.7 Mean arterial pressure (mmHg) 86.4 ± 7.5 90.3 ± 5.8 Pulmonary capillary wedge pressure (mmHg) 6.3 ± 2.1 7.5 ± 1.8 Central venous pressure (mmHg) 3.7 ± 2.6 3.6 ± 2.5 Left ventricular stroke work index (g/m/m − 2 ) 41 ± 8 48 ± 6 Reproduced with permission from Clark S, Cotton D, Lee W, et al Central hemodynamic assessment of normal term pregnancy Am J Obstet Gynecol 1989; 161: 1439 – 1442

risk of FFP includes disease transmission, anaphylactoid reac-tions, alloimmunization, and excessive intravascular volume [34]

Massive blood replacement is defi ned as transfusion of one total blood volume within 24 hours The NIH consensus confer-ence report noted that pathologic hemorrhage in the patient receiving a massive transfusion is caused more frequently by thrombocytopenia than by depletion of coagulation factors This

fi nding was demonstrated in a prospective study of 27 massively transfused patients in whom levels of factors V, VII, and IX and

fi brinogen could not be correlated with the number of units of whole blood transfused [35] A study of combat casualties sug-gested the thrombocytopenia was more important than depletion

of coagulation factors as a cause of bleeding in massively trans-fused patients [36] In this report, restoration of the prothrombin times (PT) and partial thromboplastin times (PTT) to normal with FFP had little effect on abnormal bleeding; however, platelet transfusions were effective There is no evidence that routine administration of FFP per a given number of units of RBCs decreases transfusion requirements in patients who are receiving multiple transfusion and who do not have documented coagula-tion defects [37] Thus, during massive blood replacement, cor-rection of specifi c coagulation defects (fi brinogen levels < 150 mg/ dL) and thrombocytopenia (platelet count < 30 000/mL) will minimize further transfusion requirements With massive obstet-ric hemorrhage (the usual reason for hypovolemic shock) coagu-lation factors as well as red blood cells are lost Specifi c replacement with PRBC ’ s and crystalloid solution may lead to dilutional coag-ulopathy and subsequently more blood loss

In acute hemorrhagic shock, central venous pressure (CVP)

or pulmonary capillary wedge pressure (PCWP) refl ect intravas-cular volume status and may be useful in guiding fl uid therapy

In the critically ill patient, however, CVP may be a less reliable indicator of volume status due to compliance changes in the vein walls [38] The clinician must use resources on hand to correct the volume defi ciency Central hemodynamic monitoring equipment, and personnel to introduce and maintain it, are not commonly present in the obstetric delivery suite or labor area Again, we digress to the obstetrician ’ s fundamental knowledge of maternal volume in pregnancy, and how that may

be authenticated by her current condition (e.g pre - eclampsia or abruption)

In the absence of diuretic use (an unusual if not proscribed therapy during the conduct of labor and delivery, or preparations for elective cesarean delivery); urine output measured with indwelling Foley catheter and drainage will provide approximate and important information about maternal volume and intravas-cular status in real time The operative or treating physician ’ s time is best confi ned to eliminating the focus of hemorrhage and relying on simple and adequate techniques for assessing patient response to resuscitation measures Serial hematocrit, platelets,

fi brinogen, PT, and PTT can monitor the hemoglobin and coagu-lation integrity in the maternal vascular tree Urine fl ow should

be maintained between 30 and 60 cc/h

the amount indicated by calculation of blood loss volume

Prolonged hemorrhagic shock also alters active transport of ions

at the cellular level, and intracellular water decreases

As can be appreciated from Table 40.3 most instances of

hypo-volemic shock in obstetrics are hemorrhagic and immediate

Although optimal measurements of this process may certainly

document its severity, quick action and volume replacement is

essential to optimizing outcome of the patient The two most

common crystalloid fl uids used for resuscitation are 0.9% sodium

chloride and lactated Ringer ’ s solution Both have equal plasma

volume - expanding effects The large volumes of required

crystal-loids can markedly diminish the colloid osmotic pressure (COP)

Fluid resuscitation in young, previously healthy patients can be

accomplished safely with modest volumes of crystalloid fl uid and

with little risk of pulmonary edema The enormous volumes of

crystalloids necessary to adequately resuscitate profound

hypo-volemic shock, however, will reduce the gradient between the

COP and PCWP and may contribute to the pathogenesis of

pul-monary edema [30]

Unfortunately, only 20% of infused crystalloid solution

remains intravascular after 1 hour in the critically ill patient

Their use should be limited to immediate resuscitation and

perfu-sion as the clinician orders and awaits the arrival of blood

prod-ucts Crystalloid solutions such as lactated ringers and normal

saline also help to replenish intracellular water and electrolytes,

and help to correct metabolic derangement created by the

hemor-rhagic and resuscitative event [31] Recently, data supporting the

use of colloid solutions (e.g 5% albumin) in the active

resuscita-tion of patients have come under re - evaluaresuscita-tion No trial or

analy-sis has purported to show any benefi t for the use of colloids over

crystalloids and some have suggested increased mortality with the

use of colloids [32]

The most effective replacement therapy for lost blood volume

is its replacement with whole blood The immediacy of obstetric

hemorrhage may, at times, demand this

Modern blood transfusion practice emphasizes the use of cell

components or component hemotherapy rather than whole

blood Red blood cells are administered to improve oxygen

deliv-ery in patients with decreased red cell mass resulting from

hemor-rhage A National Institutes of Health (NIH) consensus conference

concluded that transfusion of fresh frozen plasma (FFP) was

inappropriate for volume replacement or as a nutritional

supple-ment [33] In the past, up to 90% of FFP use was for volume

replacement The other 10% was for the following conditions

approved by the NIH consensus conference: replacement of

iso-lated coagulation factor defi ciencies, reversal of coumarin effect,

antithrombin III defi ciency, immunodefi ciency syndromes, and

treatment of thrombotic thrombocytopenic purpura The current

concern for excessive use of FFP is at least threefold Firstly, the

high profi le of cost containment has caused blood banks to

reevaluate use of blood products and the time involved in their

preparation Second, the routine use of FFP compromises the

availability of raw material for preparation of factor VIII

concen-trates for hemophiliacs Third, with regard to recipient safety, the

and aroyl preparations delivered in the third stage of labor were more effective than rectal misoprostol in the prevention of post-partum hemorrhage [42]

In cases of persistent vaginal bleeding, careful exploration of the vagina, cervix, and uterus is performed The clinician looks for retained products of conception or lacerations For hemor-rhage resulting from uterine atony that has failed to respond to the previously described conservative measures, as well as in cases

of extensive placenta accreta or uterine rupture not amenable to simple closure, laparotomy and hysterectomy may be indicated

If the patient does desire future fertility and is clinically stable, uterine artery ligation or stepwise uterine devascularization has been favorably described [43,44]

The fundus compression suture as described by B - Lynch has also been reported to abate uterine hemorrhage in many cases [45] Rarely, hypogastric artery ligation is surgically necessary Balloon occlusion and embolization of the internal iliac arteries have also been described in cases of placenta percreta [46,47] A good discussion of many of these techniques, as well as a more comprehensive discussion of techniques for achieving medical and surgical hemostasis in patients with postpartum bleeding, have been described elsewhere [48,49]

It should be emphasized that preventable surgical death in obstetrics may, on occasion, represent an error in judgment and

a reluctance to proceed with laparotomy or hysterectomy, rather than defi ciencies in knowledge or surgical technique Proper management of serious hemorrhage requires timely medical and surgical decision - making as well as meticulous attention to the aforementioned principles of blood and volume replacement

Cardiogenic s hock

This type of shock is caused by failure of the heart as an effective pump In the obstetric patient this most often occurs in the patient with pre - existing myocardial disease, peripartum cardio-myopathy, congenital or acquired valvular heart disease, and certain cardiac arrhythmias It is important to remember that ischemic changes in the heart may be induced in the settling of hypovolemic and septic shock [50]

Common causes of cardiogenic pulmonary edema in preg-nancy are diastolic heart failure due to chronic hypertension and obesity, leading to left ventricular hypertrophy [51,52] Cyanotic congenital heart disease leads to ischemic changes with increasing right to left shunting due to normal decreases in systemic vascular resistance in pregnancy [53] Patients with Eisenmenger syn-drome can develop right heart failure and cardiogenic shock as pulmonary hypertension worsens temporarily [54]

Pathogenesis

Cardiogenic shock is characterized by systemic hypoperfusion in the setting of an adequate intravascular volume Hemodynamic criteria include sustained hypertension (i.e systolic blood pres-sure < 90 mmHg), reduced cardiac index ( < 2.2 L/min/m 2 ), and an

Initial type and screening of labor and delivery patients can

provide valuable information if the need for blood replacement

arises in hemorrhagic morbidity Type - specifi c blood to the

patient with Coombs negative blood screens is associated with an

acceptably low level of incompatibility of 0.01% [39]

Pharmacologic a gents

During the antepartum and intrapartum periods, only correction

of maternal hypovolemia will maintain placental perfusion and

prevent fetal compromise Although vasopressors may

temporar-ily correct hypotension, they do so at the expense of

uteroplacen-tal perfusion Thus, vasopressors are not used in the treatment of

obstetric hemorrhagic shock

Further e valuation

After the patient ’ s oxygenation and expansion of intravascular

volume have been accomplished and her condition is beginning

to stabilize, it is essential for the healthcare team to evaluate the

patient ’ s response to therapy, to diagnose the basic condition that

resulted in circulatory shock, and to consider the fetal condition

Serial evaluation of vital signs, urine output, acid – base status,

blood chemistry, and coagulation status aid in this assessment In

select cases, placement of a pulmonary artery catheter should be

considered to assist in the assessment of cardiac function and

oxygen transport variables In general, however, central

hemody-namic monitoring is not necessary in simple hypovolemic shock

Evaluation of the fetal cardiotocograph may indicate fetal

dis-tress during an acute hemorrhagic episode As a rule, maternal

health trumps fetal health This means that delivery, under these

circumstances, should not be considered until maternal

condi-tion has been stabilized Once the pregnant woman is stabilized

and the fetus continues to demonstrate persistent signs of fetal

distress, the clinician should then consider delivery It is

impor-tant to realize that as the maternal hypoxia, acidosis, and

under-perfusion of the uteroplacental unit are being corrected, the fetus

may recover Serial evaluation of the fetal status and in utero

resuscitation are preferable to emergency delivery of a depressed

infant from a hemodynamically unstable mother

Hemostasis

In certain situations, such as uterine rupture with intraperitoneal

bleeding, defi nitive surgical therapy may need to be instituted

before stabilization can be achieved With postpartum

hemor-rhage resulting from uterine atony that has not responded to the

conventional methods of uterine compression and dilute

intra-venous oxytocin, the physician should consider intramuscular

methergine or 15 methyl prostaglandin F 2 α The latter is

admin-istered as a 250 - µ g dose, which may be repeated as necessary for

up to a maximum of eight doses at 15 – 90 - min intervals

In a small series of patients, rectal administration of

misopro-stol, a PGE 1 analogue, has been found effective in the treatment

of uterine atony [40] Other data indicate that rectal misoprostol

is no more effective than intravenous oxytocin in preventing

postpartum hemorrhage [41] In a systematic review oxytocin

Pregnancy - associated spontaneous coronary artery dissection (P - SCAD) is the most common cause of myocardial infarction in the immediate postpartum period In one report 78% of women with peripartum P - SCAD had no risk factors for coronary artery disease and 84% of lesions involved the left anterior descending artery [68] Successful treatment includes coronary stenting and emergency bypass grafting [69,70] One review of P - SCAD man-agement concluded that approximately one - third of women could be managed medically with antiplatelet therapy and β blocker administration and achieve excellent clinical and angio-graphic results [71]

Women who have evidence of atherosclerotic or intracoronary thrombosis are candidates for coronary stenting or the adminis-tration of tissue plasminogen activator (TPA) This large molecu-lar weight molecule should not cross the placenta and has been used successfully in thrombolysis of intracoronary thrombosis during pregnancy [72] However, TPA is contraindicated in the early postpartum period because the risk of postpartum hemor-rhage is greater than the risk of angioplasty and coronary artery stenting The foregoing discussion should emphasize the useful-ness of early coronary angiography after acute myocardial infarc-tion in pregnancy

Peripartum c ardiomyopathy

Peripartum cardiomyopathy (PPCM) is cardiomyopathy that develops in the last gestational month of pregnancy or in the fi rst

5 months postpartum By defi nition, it requires that there be no identifi able cause for heart failure and no identifi able heart disease It is rare in the United States (1 in 3000 to 1 in 15 000) but more prevalent in Africa (1 in 3000), and Haiti (1 in 350) Risk factors include, but are not limited to, the following: older women, obesity, multiparous mothers with multifetal gestation [73] , and may also include pre - eclampsia and severe hyperten-sion during pregnancy [74] Prior to delivery PPCM presents as

a patient with NYHA class III or IV functional status [75] Patients who present after delivery often have dramatic signs of congestive heart failure Symptoms may include, but are not limited to, dyspnea, orthopnea, persistent weight retention or weight gain, peripheral edema, nocturnal cough, and profound fatigue post-partum Evaluation of left ventricular size and systolic function should be performed with echocardiography Every attempt should be made to rule out other causes of cardiomyopathy Cunningham et al evaluated 28 cases of peripartum heart failure

of obscure etiology and was able to ascribe underlying causes to all but 7 (25%) The majority of cases on this cohort were due to underlying chronic hypertension, obesity, and forme fruste mitral stenosis that were eliminated by thorough evaluation In this study, peripartum heart failure was usually precipitated by, or may even have been caused by, a constellation of complications common to pregnancy, namely pre - eclampsia, cesarean delivery, anemia, and infection [76]

elevated fi lling pressure (pulmonary capillary wedge pressure

> 18 mmHg)

Cardiogenic shock is characterized by a vicious cycle in which

decreased myocardial contractility, usually due to ischemia,

results in reduced cardiac output and arterial pressure The cycle

continues with hypoperfusion of the myocardium and further

depression of maternal cardiac output Systolic myocardial

function reduces stroke volume and, together with diastolic

dys-function, leads to elevated LV end - diastolic pressure and PCWP

as well as to pulmonary congestion Reduced coronary perfusion

leads to worsening ischemic and progressive myocardial

dysfunc-tion and a rapid downward spiral, which, if uninterrupted, is

often fatal [55]

Due to the unstable condition of these patients, supportive

therapy must be initiated simultaneously with diagnostic

evalua-tion In this circumstance, clinical evaluation of the patient is

important in helping to establish a diagnosis and to guide patient

management Blood work including baseline ABG, cardiac

tro-ponin, metabolic profi le, hematocrit, and live enzymes should be

sent to the lab ECG, chest X - ray, and echocardiogram should be

obtained There is a split of opinion with respect to the use of

pulmonary artery catheterization in patients with suspected

car-diogenic shock However, many clinicians believe that the use of

the pulmonary artery catheter provides diagnostic clarity and

guidance for clinical management [56 – 59]

Acute m yocardial i nfarction

Acute myocardial infarction in pregnancy is a rare event Recent

studies place the incidence at 2.8 – 6.2 per 100 000 deliveries

[60,61] The three strongest independent predictors of

myocar-dial infarction in one population were chronic hypertension,

diabetes, and advanced maternal age [60] Case fatality rates have

been estimated from 5.1% to 37%; and the most dangerous time

for the gravida is in the last trimester of gestation or puerperium

[60 – 62] Women who sustain an infarction less than 2 weeks

prior to labor are at especially high risk of death [64] Myocardial

infarction is more common during the third trimester or

puer-perium of the fi rst or second pregnancies [65]

Patients typically present with ischemic chest pain in the

pres-ence of an abnormal ECG and elevated specifi c cardiac troponin

I Initially the maternal condition should be stabilized by medical

management Nitroglycerin and morphine sulfate should be

administered Oxygen is also a potent vasodilator and should be

administered initially by nasal cannula Ventricular arrhythmia

may occur and if left unchecked, can lead to cardiogenic shock

or sudden cardiac death Ventricular defi brillation should be

effective in this case Lidocaine should be started to prevent

arrhythmias In the case where left ventricular dysfunction occurs

as a result of ischemic changes, intra - aortic balloon pump has

been used to improve left ventricular output and coronary artery

perfusion [66,67] Calcium channel blockers and β - blockers may

be used to help decrease afterload

are common in mitral stenosis and should be treated aggressively with intravenous verapamil (5 – 10 mg) or cardioversion, if neces-sary Because of the increased risk of systemic embolism in patients with mitral stenosis and atrial fi brillation, anticoagulant therapy is indicated

Intrapartum, carefully controlled epidural anesthesia with attention to minimizing preload will decrease anxiety, which causes tachycardia and pulmonary congestion Choice of route of delivery should be left to the patient as both vaginal delivery and elective cesarean are obstetrically acceptable options Assuming the patient does not elect a cesarean using a regional anesthetic, instrumental vaginal delivery should be considered to shorten the second stage Once delivered, the low resistance uterine shunt disappears and “ autotransfusion ” may bring about pulmonary edema (Figure 40.1 )

During pregnancy diuretic therapy may be used to decrease preload along with prophylactic β - blockers to slow the heart rate response to activity and anxiety [83] Acute pulmonary edema

in pregnancy may also occur with the use of tocolytics, pre eclampsia, and fl uid overload [84]

Arrhythmias

Management of cardiac arrest attributable to life - threatening ventricular tachyarrhythmias is essential to prevent sudden cardiac death in the mother and fetus This requires a correct diagnosis and is usually possible with a 12 - lead surface ECG Clinical errors can occur in circumstances where medical person-nel believe that a young healthy woman is unlikely to develop a life - threatening ventricular arrhythmia Ventricular tachycardia (VT) is rare in pregnant women but can originate from either the right or left ventricle in structurally normal hearts [85,86] If the

Treatment of peripartum cardiomyopathy is similar to the

treatment of acute and chronic heart failure due to other causes

of left ventricular systolic dysfunction Patients who are

con-gested but have adequate perfusion will require treatment with

intravenous diuretics alone or in combination with vasodilators

such as nitroglycerin or nitroprusside Patients with diminished

perfusion will require augmentation of their cardiac output with

inotropic agents such as dobutamine [77] Beta - blockers are

used, since high heart rate, arrhythmias, and sudden death often

occur with PPCM Digitalis, an inotropic agent, is also safe during

pregnancy and may help to maximize contractility and rate

control Because of the high incidence of thromboembolism in

these patients, the use of heparin is considered medically

neces-sary, followed by warfarin (when not pregnant) in those with left

ventricular ejection fractions less than 35% After pregnancy is

over, ACE inhibitors are used to reduce afterload by

vasodilata-tion Angiotensinogen receptor blockers may be substituted for

those patients intolerant of ACE inhibitors

The acute treatment aims to reduce preload and afterload,

and increase contractility In the setting of cardiogenic shock,

invasive hemodynamic monitoring may be helpful in making

decisions regarding maternal responses to therapy and the need

for additional therapeutics Since an immune pathogenesis has

been postulated, immune modulation with intravenous

immu-noglobulin has been proposed [77,78] However, no consistent

benefi t has been demonstrated If patients remain in cardiogenic

shock despite aggressive medical therapy, additional therapies

to maintain circulatory support and organ perfusion should be

considered Intra - aortic balloon counterpulsation can be used

acutely Where resources exist, left ventricular assist devices

and heart transplantation may be used if clinically indicated

[79,80]

Mitral s tenosis

Rheumatic mitral stenosis (MS) is the most common clinically

signifi cant valvular abnormality in pregnant women and may be

associated with pulmonary congestion, edema, and atrial

arrhyth-mias during pregnancy or soon after delivery The contracted

valve impedes blood fl ow from the left atrium to the ventricle

The left atrium is dilated, left atrial pressure is chronically

ele-vated, and signifi cant passive pulmonary hypertension may

develop Twenty - fi ve per cent of women have cardiac failure for

the fi rst time during pregnancy [81] This has been confused with

peripartum cardiomyopathy [76]

With signifi cantly tight MS (surface area < 2.5 cm 2 ) symptoms

usually develop [82] The most common complaint is dyspnea

and others may include, but are not limited to, fatigue,

palpita-tions, cough, and hemoptysis Tachycardia shortens diastolic

fi lling time and increases the mitral gradient, which raises left

atrial and pulmonary venous and capillary pressures and may

result in pulmonary edema Sinus tachycardia is often treated

with β - blockers Atrial tachyarrhythmias, including fi brillation,

A 10 20 30

8 4 6 5

2 BC Time

Figure 40.1 Intrapartum alterations in pulmonary capillary wedge pressure

(PCWP) in eight patients with mitral stenosis (a) First - stage labor;

(b) second - stage labor, 15 – 30 min before delivery; (c) 5 – 15 min post partum; (d) 4 – 6 h post partum; (e) 18 – 24 h post partum (Reproduced with permission from Clark SL, Phelan JP, Greenspoon J, et al Labor and delivery in the presence of mitral stenosis: central hemodynamic observations Am J Obstet Gynecol 1985; 152: 986.)

pressure) associated with labor The combined α - and β - blocker labetolol can be titrated (1 – 10 mg/kg/min) to allow rapid control

of mean arterial blood pressure during labor and delivery A split of opinion exists as to the optimal anesthetic for cesarean delivery in these patients General anesthesia may be indicated in anticoagulated patients, but the hypertensive response to intuba-tion and surgical stimulaintuba-tion may increase cardiovascular stress, promoting rupture or progression of a pre - existing dissection [65]

Summary

Cardiogenic shock is quite different from hypovolemic shock, the latter requiring increased preload and the former inotropic support and decreased afterload Cardiogenic shock occurs infre-quently and calls for consultation, timely decision - making, and diagnostic integration Pulmonary congestion is the usual pre-senting symptom and has multiple etiologies Although there is

a split of opinion as to the effectiveness of invasive hemodynamic monitoring, accurate diagnosis as to the etiology of heart failure

is essential and many experts feel comfortable with this diagnostic modality Others have demonstrated that non - invasive echocar-diographic modalities to estimate cardiac parameters are reason-ably accurate [97] Obstetricians should have a working knowledge

of maternal hemodynamics in pregnancy and should avail them-selves of frequent cardiology and maternal – fetal medicine con-sultation Responding to postpartum hemorrhage should be fairly straightforward in most but not all instances Cardiogenic shock (and its imitators) should be suspected in the setting of pulmonary congestion and myocardial infarction In many of these circumstances, timely consultation and a multidisciplinary approach can enhance maternal and fetal outcome

References

1 Center for Disease Control and Prevention Pregnancy - related

mortality surveillance – United States MMWR 2003 ; 52 (Suppl 2 ):

1991 – 1999

2 McCormick M , Sanghvi H , Kinzie B Preventing postpartum hemor-rhage in low - resource settings Int J Gynecol Obstet 2002 ; 77 :

267 – 275

3 Li X , Fortney J , Katelchuk M , et al The postpartum period: the key

to maternal mortality Int J Gynecol Obstet 1996 ; 54 : 1 – 10

4 Edman C , Toofanian A , MacDonald P , et al Placental clearance rate

of maternal plasma and rostenedione through placental estradiol

for-mation Am J Obstet Gynecol 1981 ; 141 : 1029 – 1037

5 Shoemaker W Diagnosis and treatment of shock and circulatory dysfunction In: Grenvik A , Ayers S , Holbrook P , Shoemaker W , eds

Textbook of Critical Care , 4th edn Philadelphia : WB Saunders : 2000

6 Pullicino E , Carli F , Poole S , et al The relationship between circulat-ing concentrations of interleukin - 6, tumor necrosis factor, and the acute phase response to elective surgery and accidental injury

Lymphokine Res 1990 ; 9 : 231 – 238

pregnant patient has structural heart disease VT carries a poor

prognosis [87]

Acutely, it is important to determine whether VT is

hemody-namically stable or unstable If at any time the patient becomes

unstable or there is evidence of fetal compromise direct - current

(DC) countershock (50 – 100 J) should be delivered immediately

[88] Conservative therapy is appropriate for sustained VT and

stable hemodynamics Acute therapy should be initiated with

procainamide (50 – 100 mg intravenously) Another potential

antiarrhythmic drug is lidocaine Neither is known to be

terato-genic Life - threatening ventricular fi brillation (VF) or ventricular

fl utter can occur in any stage of pregnancy and is associated with

a high risk of sudden cardiac death [89] In these patients DC

defi brillation is the method of choice (100 – 360 J) Prompt

car-diopulmonary resuscitation and early defi brillation by either DC

countershock or an automated external defi brillator signifi cantly

improves the likelihood of successful resuscitation from VF

[89,90] The implantable cardioverter - defi brillator (ICD) is an

excellent approach to terminal tachyarrhythmias and prevents

sudden death [91]

Ventricular premature beats (VPBs) in pregnant women with

structurally normal hearts are benign and do not usually require

therapy [87] Exacerbating chemical stimulants (e.g caffeine,

cocaine) should be eliminated, and cardiology and maternal – fetal

medicine consultation should be considered If patients remain

highly symptomatic a selective β - blocker such as metoprotol may

be used

Aortic d issection

Acute aortic dissection may occur in pregnancy in association

with severe hypertension due to pre - eclampsia, coarctation of the

aorta, or connective tissue disease such as Marfan ’ s syndrome

[92,93] Although rare, aortic dissection during pregnancy

accounts for 50% of all dissections in women under 40 years of

age [94] Classic symptoms include severe retrosternal and/or

abdominal pain The patient may present with cardiogenic shock

due to aortic incompetence or hemopericardium and

tampon-ade Diagnosis is made with computed tomography or

trans-esophageal echocardiography Maternal outcome is poor with

mortality as high as 25% [93,95] If the dissection is limited to

the descending aorta some patients can be managed successfully

to term and surgery performed postpartum

Before 28 weeks gestation, aortic repair with the fetus kept in

situ is warranted, given the high mortality (80%) of non -

opera-tive treatment in this setting [93,96] After 32 weeks gestation

(fetal viability) primary cesarean followed by repair is appropriate

[93,95] Controlled vaginal delivery with regional anesthesia has

been described [95] Management between 28 and 32 weeks is

controversial If there is hemodynamic instability or non -

reassur-ing fetal status, then cesarean followed by operative repair should

be affected Epidural anesthesia reduces increases in vessel shear

stress (cardiac output) and wall tension (mean arterial blood

27 Dildy G , Clark S , Loucks CA Intrapartum fetal pulse oximetry: the effects of maternal hyperoxia on fetal arterial oxygen saturation

Am J Obstet Gynecol 1994 ; 171 : 1120 – 1124

28 Thangathurai D , Charbonnet C , Wo C , Shoemaker WC , et al Intraoperative maintenance of tissue perfusion prevents ARDS

New Horiz 1996 ; 4 : 466 – 474

29 Taylor RW Pulmonary Artery Catheter Consensus Conference Participants: Pulmonary Artery Catheter Consensus Conference

consensus statement Crit Care Med 1997 ; 25 : 910

30 Harms B , Kramer GC , Bodai BI The effect of hypoproteinemia and

pulmonary and soft tissue edema formation Crit Care Med 1981 ; 9 :

503

31 Chiao J , Minei J , Shires G , et al In vivo myocyte sodium activity and

concentration during hemorrhagic shock Am J Physiol 1990 ; 258 :

R684 – R689

32 SAFE Study Investigators A comparison of albumin and saline for

fl uid resuscitation in the intensive care unit N Engl J Med 2004 ; 350 :

2247 – 2256

33 National Institutes of Health Consensus Development Conference Fresh frozen plasma Indications and risks JAMA 1985 ; 253 :

551 – 553

34 Oberman HA Uses and abuses of fresh frozen plasma In: Garrity A ,

ed Current Concepts in Transfusion Therapy Arlington, VA : American

Association of Blood Banks , 1985

35 Counts R , Haisch C , Simon TL , et al Hemostasis in massively

trans-fused trauma patients Ann Surg 1979 ; 190 : 91 – 99

36 Miller R , Robbins T , Tong MJ , et al Coagulation defects associated

with massive blood transfusions Ann Surg 1971 ; 174 : 794 – 801

37 Mannucci P , Federici A , Sirchia G Hemostasis testing during massive blood replacement: a study of 172 cases Vox Sang 1982 ; 42 :

113 – 123

38 Shippy C , Appel P , Shoemaker WC Reliability of clinical monitoring

to assess blood volume in critically ill patients Crit Care Med 1984 ;

12 : 107 – 112

39 Boral L , Hill S , Apollon CJ , et al The type and antibody screen,

revisited Am J Clin Pathol 1979 ; 71 : 578 – 581

40 O ’ Brien P , El - Refaey H , Gordon A , et al Rectally administered miso-prostol for the treatment of postpartum hemorrhage unresponsive to

oxytocin and ergotomine: a descriptive study Obstet Gynecol 1998 ;

92 : 212 – 214

41 Gerstenfeld T , Wing DA Rectal misoprostol versus intravenous oxy-tocin for the prevention of postpartum hemorrhage after vaginal

delivery Am J Obstet Gynecol 2001 ; 185 : 878 – 882

42 Villar J , G ü lmezoglu A , Hofmeyr GJ , et al Systematic review of randomized controlled trials of misoprostol to prevent postpartum

hemorrhage Obstet Gynecol 2002 ; 100 : 1301 – 1312

43 AbdRabbo SA Stepwise uterine devascularization: a novel technique for management of uncontrolled postpartum hemorrhage with

preservation of the uterus Am J Obstet Gynecol 1994 ; 171 : 694 – 700

44 O ’ Leary JA Uterine artery ligation in the control of postcesarean

hemorrhage J Reprod Med 1995 ; 40 : 189 – 193

45 B - Lynch C , Coker A , Lawal AH , et al The B - Lynch surgical technique for the control of massive postpartum hemorrhage: an alternative to

hysterectomy? Br J Obstet Gynaecol 1997 ; 104 : 372 – 375

46 Dubois J , Burel L , Brignon A , et al Placenta percreta: balloon occlu-sion and embolization of the internal iliac arteries to reduce

intraop-erative blood loss Am J Obstet Gynecol 1997 ; 176 : 723 – 726

47 Shih J , Liu K , Shyu MK Temporary balloon occlusion of the common iliac artery: new approach to bleeding control during cesarean

7 Abraham E Physiologic stress and cellular ischemia: relationship to

immunosuppression and susceptibility to sepsis Crit Care Med 1991 ;

19 : 613 – 618

8 Hoch R , Rodriguez R , Manning T , et al Effects of accidental trauma

on cytokine and endotoxin production Crit Care Med 1993 ; 21 :

839 – 845

9 Bassin R , Vladeck B , Kim S , et al Comparison of hemodynamic

responses of two experimental shock models with clinical

hemor-rhage Surgery 1971 ; 69 : 722 – 729

10 Clark SL Shock in the pregnant patient Semin Perinatol 1990 ; 14 :

52 – 58

11 Shoemaker W , Appel P , Kram HB Role of oxygen debt in the

devel-opment of organ failure, sepsis, and death in high risk surgical

patients Chest 1992 ; 102 : 208 – 215

12 Shoemaker W , Appel P , Kram HB Hemodynamic and oxygen

trans-port responses in survivors and nonsurvivors of high risk surgery Crit

Care Med 1993 ; 21 : 977 – 990

13 Shoemaker WC Pathophysiologic basis for therapy for shock and

trauma syndromes: use of sequential cardiorespiratory measurements

to describe natural histories and evaluate possible mechanisms Semin

Drug Treat 1973 ; 3 : 211 – 229

14 Slater G , Vladeck B , Bassin R , et al Sequential changes in the

distribu-tion of cardiac output in various stages of experimental hemorrhagic

shock Surgery 1973 ; 73 : 714 – 722

15 Pritchard JA Changes in the blood volume during pregnancy and

delivery Anesthesiology 1965 ; 26 : 393

16 Clark S , Cotton D , Lee W , et al Central hemodynamic assessment

of normal term pregnancy Am J Obstet Gynecol 1989 ; 161 :

1439 – 1442

17 Greiss FC Uterine vascular response to hemorrhage during

preg-nancy and delivery Anesthesiology 1965 ; 26 : 393

18 Pritchard J , Baldwin R , Dickey JC , et al Blood volume changes in

pregnancy and the puerperium, 2 Red blood cell loss and changes in

apparent blood volume during and following vaginal delivery,

cesar-ean section, and cesarcesar-ean section plus total hysterectomy Am J Obstet

Gynecol 1962 ; 84 : 1271

19 Sheehan H , Murdoch R Postpartum necrosis of the anterior

pitu-itary: pathologic and clinical aspects Br J Obstet Gynaecol 1938 ; 45 :

456

20 Smith K , Browne J , Shackman R , et al Renal failure of obstetric

origin Br Med Bull 1968 ; 24 : 49

21 Mabie WC , Barton JR , Sibai BM Adult respiratory syndrome in

preg-nancy Am J Obstet Gynecol 1992 Oct; 167 ( 4 pt 1 ): 950 – 957

22 Belfort M , Anthony J , Saade G , et al The oxygen consumption:

oxygen delivery curve in severe preeclampsia: evidence for a fi xed

oxygen extraction state Am J Obstet Gynecol 1993 ; 169 : 1448 –

1455

23 Evans W , Capelle S , Edelstone DI Lack of a critical cardiac output

and critical systemic oxygen delivery during low cardiac output in the

third trimester in the pregnant sheep Am J Obstet Gynecol 1996 ; 175 :

222 – 228

24 Haddad B , Barton JR , Livingston JC , et al Risk factors for adverse

maternal outcomes among women with HELLP (hemolysis, elevated

liver enzymes, and low platelet count) syndrome Am J Obstet Gynecol

2000 Aug; 183 ( 2 ): 444 – 448

25 Naef R , Chauhan S , Chevalier S , et al Prediction of hemorrhage at

cesarean delivery Obstet Gynecol 1994 ; 83 : 923 – 926

26 Zelop C , Harlow B , Frigoletto FD , et al Emergency peripartum

hysterectomy Am J Obstet Gynecol 1993 ; 168 : 1443 – 1448