Best Practice & Research Clinical Obstetrics and Gynaecology

Vol. 22, No. 5, pp. 775799, 2008

doi:10.1016/j.bpobgyn.2008.06.001
available online at http://www.sciencedirect.com

2
Aims of obstetric critical care management
Laura Claire Price *

BSc, MBChB, MRCA, MRCP

SpR Respiratory and Intensive Care Medicine

Adult ICU, St Georges Hospital, London, UK

Andrew Slack

MBBS, MRCP

SpR Renal and Intensive Care Medicine

Adult ICU, Kings College Hospital, London, UK

Catherine Nelson-Piercy

MA, FRCP, FRCOG

Consultant Obstetric Physician

Directorate Office, North Wing, St Thomas Hospital, London, UK

The aims of critical care management are broad. Critical illness in pregnancy is especially pertinent as the patient is usually young and previously fit, and management decisions must also
consider the fetus. Assessment must consider the normal physiological changes of pregnancy,
which may complicate diagnosis of disease and scoring levels of severity. Pregnant women
may present with any medical or surgical problem, as well as specific pathologies unique to pregnancy that may be life threatening, including pre-eclampsia and hypertension, thromboembolic
disease and massive obstetric haemorrhage. There are also increasing numbers of pregnancies
in those with high-risk medical conditions such as cardiac disease. As numbers are small and
clinical trials in pregnancy are not practical, management in most cases relies on general intensive care principles extrapolated from the non-pregnant population. This chapter will outline the
aims of management in an organ-system-based approach, focusing on important general principles of critical care management with considerations for the pregnant and puerperal patient.
Key words: pregnancy; high risk; complications; pre-eclampsia; postpartum haemorrhage;
thromboembolism; intensive care; critical care.

GENERAL OVERVIEW
Maternal mortality is fortunately rare in the UK, with 13.95 maternal deaths per
100 000 deliveries in 20032005.1 However, despite modern-day advances in care,
* Corresponding author. Tel.: 44 2074311250; Fax: +44 2084584505.
E-mail address: [email protected] (L.C. Price).
1521-6934/$ - see front matter 2008 Elsevier Ltd. All rights reserved.

776 L. C. Price et al

this figure has remained static over the last few decades, and this may relate to a higher
number of high-risk pregnancies progressing to term.2 Worldwide, maternal mortality
is greater with 55920 maternal deaths per 100 000 live births3, with the highest rates
in sub-Saharan Africa. The most common reasons for intensive care unit (ICU) admissions in the UK are pre-eclampsia, sepsis and haemorrhage.4 Overall, 0.9% of pregnant
women require ICU admission in the UK, comparable to US figures.5 The most
common cause of maternal death on ICU is acute respiratory distress syndrome
(ARDS). UK maternal mortality could be improved with prompt recognition of critical
illness, earlier use of critical care facilities and earlier senior involvement. Perinatal
mortality rates are up to 2025% depending on the underlying maternal diagnosis.
The assessment and management of obstetric admissions to critical care can be challenging, with unique disease states and physiological changes seen. These changes occur
in all major systems and persist for up to 6 weeks post partum. In 5080% of cases, pregnant women require ICU admission due to a direct obstetric cause; the remainder relate
to medical causes. There may be diseases specific to pregnancy (massive obstetric haemorrhage, amniotic fluid embolism, pre-eclampsia, peripartum cardiomyopathy); an increased susceptibility to certain diseases due to pregnancy (venous thromboembolism,
urinary tract infection, varicella pneumonia); pre-existing disease exacerbation (asthma,
cardiac disease); or incidental diseases during pregnancy (e.g. diabetic ketoacidosis). The
requirement for critical care support for one or more organ failures usually results from
the development of a multisystem disorder such as shock, ARDS or sepsis.
The aim of critical care management in any population is to ensure adequate oxygen
delivery and tissue perfusion. There are specific conditions requiring attention in pregnancy, and this review will consider general ICU principles and these with reference to
obstetric physiology. The altered maternal physiology should be considered during
each stage of assessment, resuscitation, monitoring, use of pharmacological therapies,
and single or multiple organ support. Young, previously healthy patients often show
relative compensation in critical illness. The signs of sepsis and haemorrhage are often
masked initially and abnormal signs may overlap normal signs of pregnancy. Some signs
always indicate abnormality, including tachypnoea and metabolic acidosis. Obstetric
and medical staff should be trained in the recognition of critical illness in this population as the disease processes can follow a rapid and fulminant course.
The effects on fetal perfusion are dependent on placental perfusion and oxygen
delivery, both reflecting maternal wellbeing. The fetus is adapted to living in a relatively
hypoxic environment with the oxygen dissociation curve shifted to the left, high
haematocrit and high cardiac output (CO). Despite this, small changes in maternal
homeostasis may have an adverse effect on the fetus. If the mother does become critically ill, premature delivery may be indicated with the associated neonatal complications of respiratory distress syndrome, jaundice, intracranial haemorrhage and
necrotizing enterocolitis.
AIMS OF ORGAN SUPPORT IN CRITICAL CARE
General supportive care
Immediate resuscitation is the primary aim in the management of any critically ill patient, with systematic assessment of deranged physiology using an organ-based
systematic approach. This is generic, irrespective of the underlying pathology, and is
followed by more specific diagnostic consideration when the patient is more stable.

Aims of obstetric critical care management 777

Appropriate investigations and procedures are then carried out, depending on the
clinical history and examination. It is important not to withhold or delay useful investigations because the patient is pregnant.
The gestational age and condition of the fetus must always be considered, as well
as the effects of any drugs or procedures on the fetus. The obstetric team should
perform a fetal scan to assess viability. Often if the mother is or has been critically
ill, the fetus may have succumbed. Once this is established, liaison with obstetrics
to deliver the baby will usually improve maternal physiology, although delivery itself
is associated with significant temporary haemodynamic demands and changes. If the
fetus is viable, the advisability of delivery needs a multidisciplinary discussion balancing
the fetal risks of prematurity versus any maternal benefits from delivery. The mothers health should be the priority. The fetus is at significant risk if the mother is seriously ill, particularly if she is acidotic, and regular fetal monitoring is appropriate.
Fetal management primarily involves maternal resuscitation, maintaining adequate placental oxygenation and perfusion. Delivery may be indicated in some settings of severe maternal illness such as cardiac arrest, severe asthma, acute fatty liver of
pregnancy or pre-eclampsia, or with fetal distress. When considering the timing of
delivery, the effects of drugs on fetal physiology and the requirement for neonatal resuscitation should be considered. Drugs are classed according to the risk of fetal effects into Category A (no fetal risk) to Category D (fetal risk), and Category X
(contra-indicated). The maternal volume of distribution is increased and glomerular
filtration rate (GFR) is elevated in pregnancy, so higher doses are needed for drugs
with renal elimination. Sedative effects on fetal respiratory function should be anticipated, although polarized agents such as neuromuscular blocking agents do not cross
the placenta.
Critical care involves intensive monitoring and physiological support for patients
with life-threatening but potentially reversible conditions. Patients should be managed
in either an ICU (Level 3) or high-dependency (Level 2) setting. Level 3 usually involves
patients with multi-organ failure and/or requiring mechanical ventilation, while Level 2
involves non-invasive ventilation (NIV), renal replacement therapy or intensive
monitoring.
Early involvement and planning should involve all members of the multidisciplinary
team. This includes obstetricians, midwives, physicians (obstetric physicians if available), intensivists, anaesthetists, haematologists, paediatricians and neonatologists.
Elective ICU beds should be booked for certain patients such as those with significant
cardiac disease undergoing elective caesarean section.
Early recognition of critical illness is essential. There is good evidence that early
intervention in the first 6 h of severe sepsis influences survival. Ideally, early warning
scoring systems adapted to pregnancy physiology should be implemented on obstetric
wards to identify patients at an early stage. ICU care can often be commenced in the
operating theatre or emergency department (ED) prior to transfer to the ICU to
avoid delays, and stabilization and elective intubation may be necessary prior to
transfer.
Aims should also include adherence to recent developments in critical care practice including the Surviving Sepsis Campaign guidelines6, considered use of activated
protein C7, insulin therapy to maintain normoglycaemia8, corticosteroids in septic
shock9, lung protective ventilatory strategies in ARDS management, and early
goal-directed therapy for severe sepsis and septic shock.10 Although not written
with the obstetric patient in mind, most aspects of management should not differ
dramatically.

778 L. C. Price et al

Haemodynamic management
The aims of cardiovascular support in any setting are to maintain adequate cardiac output (CO) and blood pressure (BP). There is a 20% increase in maternal oxygen demand in pregnancy; CO increases by 3050% due to an increase in stroke volume
and heart rate, and BP is reduced by 510 mm Hg by mid-pregnancy. There is no
change in central venous pressure (CVP) or pulmonary artery wedge pressure
(PAWP) as systemic and pulmonary vascular resistance are both reduced.11 These haemodynamic changes begin as early as 8 weeks, and gradually return to normal 212
weeks post partum.12
The supine hypotension syndrome is an important consideration after 24 weeks. It
relates to compression of the inferior vena cava by the gravid uterus, and can lead to
a dramatic drop in preload, leading to a 2530% drop in CO, severe hypotension and
bradycardia. It can be reduced by positioning in a lateral tilt although a full left lateral
position is sometimes required.
The most significant cardiovascular challenge occurs at birth. Up to 500 mL of
blood is autotransfused back into the circulation following relief of aortocaval
compression by the fetoplacental unit aortocaval compression by the fetoplacental
unit and contraction of the uterus, and CO may increase by up to 80% of preterm
baseline. These fluid shifts are especially challenging in parturients with cardiac disease,
who should be monitored for 72 h post partum.13
Shock
Cardiovascular support may be required in states of circulatory shock. In shock,
reduced tissue oxygenation leads to anaerobic metabolism, reflected by an increased
serum lactate and reduced maternal central venous oxygen saturations (SvO2). This
will affect fetal oxygenation despite adaptation to a relatively hypoxic environment.
Lactate levels >2 mmol/L indicate tissue hypoxia except in settings with poor
lactate clearance such as liver failure. Serial measurements are useful, for example,
in septic shock (Table 1).14
The most common causes of shock in the obstetric population are major haemorrhage and sepsis. When reversible causes have been controlled, support of the

Table 1. Types of shock.

Type of shock

Pathophysiology

Hypovolaemia Loss of circulating

volume
Distributive
Pathological
vasodilatation
Cardiogenic
Severe pump failure
Obstructive
Obstruction to
cardiac output
Neurogenic
Loss of sympathetic
outflow if lesion
above T6

Cardiac
output

Systematic
vascular
resistance

Examples of causes

Low

High

High/low

Low

Low
Low

High
High

Low

Low

Massive haemorrhage, diabetic

ketoacidosis, burns
Sepsis, liver failure, pancreatitis,
anaphylaxis
Cardiomyopathy, valvular dysfunction
Pulmonary embolism, amniotic fluid
embolus, tamponade
Spinal trauma, intracranial haemorrhage

Aims of obstetric critical care management 779

cardiovascular system may involve fluid administration, correction of heart rate and
rhythm disorders, the use of vasoactive agents, thrombolysis, pacemakers, ventilatory
support and mechanical devices. Invasive or non-invasive haemodynamic monitoring is
an important tool. Techniques may be invasive (lithium dilution pulsed contour analysis
and the pulmonary artery catheter), semi-invasive (oesophageal Doppler) or non-invasive (echocardiography).

Fluids
Fluid administration is an important part of most resuscitation scenarios, aiming to
improve microvascular blood flow by increasing plasma volume, and improve CO by
the Frank-Starling mechanism.15 However, the use of fluids in obstetrics should be
more cautious than in the non-obstetric setting. Even in a normal pregnancy, colloid
oncotic pressure (COP) is reduced by 14% (from 20.8 to 18 mm Hg).16 This reduction
in COP, when combined with any condition predisposing to either increased capillary
leak (e.g. in pre-eclampsia and many critical illness states) or raised left atrial pressure,
can relatively easily lead to pulmonary oedema.17 It is not thought to be as common as
this suggests, because pulmonary lymphatic drainage is increased in pregnancy.
The choice of fluid used will depend on the setting. In major haemorrhage, replacement is needed with blood products. Otherwise, the choice between crystalloid and
colloid remains under debate. Excessive use of normal saline leads to hyperchloraemic
acidosis18, and Hartmanns solution is currently the most physiological crystalloid in
use. Colloids remain in the intravascular compartment for longer. Overall, the choice
will depend on the underlying diagnosis, local availability, the evaluation of the content
of each fluid type, and the benefits and complications of each (see Table 2).19
The concept of a fluid challenge requires haemodynamic monitoring to assess
volume responsiveness to a set volume of fluid.20 Following this, a 20% rise in stroke
volume suggests intravascular depletion, and sufficient intravascular volume is suggested when the rise is sustained. Small rises or a fall in the stroke volume suggest fluid
overload as the myocardium is stretched over the Starling curve. Obstetric patients
are best managed with a neutral or even negative fluid balance, as the development of
acute prerenal failure is usually reversible if recognised and treated early, compared
with the potential risks of non-cardiogenic pulmonary oedema or ARDS.
Table 2. Fluid composition.
Fluid

pH

Complications

Normal saline
(NaCl 0.9%)
5% dextrose
Hartmanns
(compound Na lactate)

Na 154, Cl 154

5.0

Hyperchloraemic acidosis

Nil
Na 131, K 5, Cl 111,
Ca2 2, HCO-3 29

4.0
6.5

Gelofusin

Na 154, K 0.4, Cl 154,

Ca2 0.4
Na 150, K 2, Cl 120

7.4

Hyponatraemia, oedema
Fluid overload
(with all fluids), K may
accumulate
Allergy (<0.1%),
hyperchloraemia
Transfusion reactions,
TRALI

Albumin 4.5%

Ionic content (mmol/L)

TRALI, transfusion-related lung injury.

7.4

780 L. C. Price et al

Furthermore, the appropriate management of oliguria in these patients is not necessarily a fluid challenge, as discussed further in the renal section.
Cardiac failure
Cardiac failure may be primarily left sided with pulmonary oedema, right sided with evidence of hepatic and renal congestion, or biventricular. Reversible causes such as coronary ischaemia or pulmonary embolism should be managed as appropriate. Non-invasive
continuous positive airways pressure (CPAP) is a very effective treatment for cardiogenic
pulmonary oedema21,22, and will increase CO, reduce left ventricular afterload, increase
functional residual capacity (FRC) and respiratory mechanics, and reduce work of
breathing. Invasive ventilation may be needed if the woman is severely acidotic, hypercapnic, hypotensive or has very poor left ventricular function.23 Vasodilator therapy is important, and both glyceryl trinitrate (GTN) and hydralazine are safe in pregnancy.
Loop diuretics are used to reduce pulmonary congestion, although they may reduce
CO and uteroplacental perfusion if hypovolaemic. If inotropes are needed, dobutamine
is safe in pregnancy. Newer inotropes such as levosimendan may improve arterioventricular coupling24, although there are only case reports of use in pregnancy in peripartum
cardiomyopathy.25 Mechanical devices to augment CO such as ventricular assist devices
and intra-aortic balloon pumps may be required.
Vasoactive agents
Heart rate and rhythm disorders are corrected in the usual way according to resuscitation council guidelines. Most supraventricular tachycardias and ventricular ectopics
require no drug therapy in pregnancy, and vagal manouevres such as carotid sinus
massage should be tried first in paroxysmal supraventricular tachycardia. Atrial fibrillation should be restored to sinus rhythm in pregnancy with electrical cardioversion,
or rate controlled with digoxin or cardioselective beta-blockers. Replacement of
potassium and magnesium may be required, then chemical or electrical cardioversion
as appropriate, or use of specific anti-arrhythmic agents. Cardioselective betablockers, diltiazem, verapamil and adenosine are safe; however, amiodarone should
be avoided as it can inhibit fetal thyroid function.26
Vasopressors are often used for hypotension in the setting of obstetric regional anaesthesia. Ephedrine may lead to maternal tachycardia due to its chronotropic effects on the
cardiac beta-1 receptor and reduced fetal pH, although it is less likely to cause uteroplacental vasoconstriction. Phenylephrine, an alpha-1 adrenergic agonist and powerful arteriolar constrictor, causes less fetal acidosis although it may reduce maternal heart rate
and therefore CO.27 If inotropes or more powerful vasopressors are required in an
ICU setting, the effects on uteroplacental perfusion should be considered, and CO monitoring is useful. The choice of agent will depend on the mean arterial pressure (MAP),
CO and systematic vascular resistance (SVR), and should follow appropriate fluid resuscitation. In the setting of vasodilatory shock with low SVR and high CO (sepsis, liver failure), sympathomimetic vasopressors such as norepinephrine are used. Vasopressin has
been used in some settings such as resistant septic shock and cardiac arrest28, but reduced blood flow to some organs may occur29 so it should be reserved for salvage therapy.30 In cardiogenic shock, increased inotropy from beta-1 agonism may be useful using
dobutamine (and epinephrine with added vasoconstrictive effects). Inodilators are useful
when low CO occurs with vasoconstriction (high SVR), with agents including milrinone
and the calcium-sensitizer levosimendan (Table 3).24

Aims of obstetric critical care management 781

Table 3. Vasoactive agents.

Use if
Vasopressors
Less potent
Ephedrine
Phenylephrine
Metaraminol
More potent
Noradrenaline
Vasopressin
Inotropes
Inoconstrictor
Adrenaline
Inodilators
Milrinone
Levosimendan
Variable
Dobutamine
Dopexamine

Clinical example

Important side effects

Low SVR

Spinal/epidural/general
anaesthetic-induced
vasodilatation

All have effects on

UP perfusion

Low SVR with

shock unresponsive
to fluid challenge

Septic shock

Lowers UP perfusion;
digital necrosis

Low CO
low SVR

Septic shock

Lactic acidosis

Low CO
high SVR

Cardiogenic shock

Hypotension, dose prolonged

in renal failure (milrinone)

Low CO
Low CO with poor
splanchnic
perfusion

Sometimes combined
with a vasoconstrictor
in some shock settings,
or used alone in
cardiogenic shock

Variable effects on blood

pressure (dobutamine)
Tachycardia (dopexamine)

CO, cardiac output; SVR, systematic vascular resistance; UP, uteroplacental.

Hypertension
Hypertension complicates 12% of pregnancies31 and may be classified into pre-eclampsia,
gestational hypertension, chronic essential hypertension and malignant hypertension
presenting in pregnancy. The overall management strategy of pre-eclampsia with delivery as a prime goal is discussed in detail elsewhere. There is no single ideal anti-hypertensive regime used for acute severe hypertension in pregnancy, and most units
have their own protocol. Intravenous hydralazine and labetalol are equally effective
but labetalol is preferred as it has fewer side effects.32 Sodium nitroprusside is an effective vasodilator, and is therefore a good choice if pulmonary oedema is present,
and its short half-life is useful when titrating the dose. However, toxicity may occur
with cyanide or thiocyanate accumulation, usually in those with renal failure and if
treated for more than 24 h. GTN is useful and safe for short-term use, although
mainly as a venodilator as it has little effectiveness in hypertensive emergencies
and so is not usually a first-line agent, and problems result with tachyphylaxis. Oral
nifedipine, doxasocin and alpha-methyldopa are oral options. Angiotensin-converting
enzyme inhibitors are teratogenic33 and are therefore contra-indicated, although they
may be life-saving in some cases of malignant hypertension including scleroderma.
Respiratory management
The overall aim of respiratory management is to maintain gas exchange. This encompasses airway management, administering oxygen therapy and ensuring adequate

782 L. C. Price et al

ventilation (CO2 clearance). These general principles are as for the non-pregnant
population, with added considerations of the physiological changes, including the left
lateral tilt in positioning, and the need in later pregnancy to consider a delivery
plan. It is important to remember that oxygen delivery to the tissues and effective
CO2 removal involves the heart and circulation as well as the lungs. Oxygen delivery
to the tissues involves a cascade of processes depending on alveolar oxygen concentration, oxygen transfer, haemoglobin and the oxygen dissociation curve, then diffusion
from capillary blood into mitochondria along a concentration gradient. Defects can
occur at any of these levels.
The differential diagnosis of primary respiratory failure in pregnancy is broad, and
the changes in respiratory physiology should be remembered in assessment and
management. Erect partial pressure of oxygen (PaO2) increases by the end of the first
trimester, and falls during each of the following trimesters. This is due to an increased
arteriovenous oxygen difference as oxygen consumption increases above CO with the
advancement of pregnancy. After mid-gestation, PaO2 is <13.1 kPa in supine patients
due to airway closing capacity being above FRC, and also from aortocaval compression. The reduced FRC and greater oxygen consumption make episodes of desaturation more rapid. Adequate fetal oxygenation requires a maternal paO2 > 9.2 kPa,
corresponding to a maternal arterial oxygen saturations (SaO2) > 95%34; much higher
than the saturations usually tolerated outside pregnancy.
It is important to consider the normal progressive respiratory alkalosis when considering respiratory management, as ventilation requirements may change depending
on the stage of pregnancy. In severe asthma, a normal partial pressure of CO2
(pCO2) of 45.5 kPa is usually a warning of impending respiratory failure. In pregnancy,
this is, in fact, relative hypercapnia as the usual upper limit in pregnancy is 3.5 kPa. The
effects of pCO2 on fetal wellbeing are not clear, although they are likely to be detrimental with acidosis leading to reduced ability of fetal haemoglobin to bind oxygen.
A diffusion gradient of approximately 1.3 kPa is needed for placental pCO2 clearance35, and high maternal levels may interfere with this. A maternal pCO2 target of
<5.9 kPa or pH >7.30 has therefore been suggested36, avoiding hypercapnia (Table 4).
Upper airway
Airway management may be challenging in obstetric practice. There are changes in maternal anatomy with increased upper airway oedema, especially in pre-eclampsia where
fluid retention enlarges the tongue and makes identification of airway landmarks more

Table 4. Causes of respiratory failure in pregnancy.

Pulmonary causes
Asthma, severe pneumonia, pleural effusion, pneumothorax, pulmonary haemorrhage, interstitial
lung disease, exacerbation of underlying respiratory disease (e.g. cystic fibrosis, chronic obstructive
pulmonary disease, bronchiectasis, pulmonary hypertension), atypical infection (including human
immunodeficiency virus), respiratory muscle myopathies (hypercapnic respiratory failure)
Acute respiratory distress syndrome (ARDS) and acute lung injury (see later)
Cardiac causes
Cardiogenic pulmonary oedema, e.g. peripartum cardiomyopathy, mitral stenosis
Iatrogenic fluid overload
Tocolytic pulmonary oedema (rare now with alternatives to beta-sympathomimetics)

Aims of obstetric critical care management 783

difficult.37 Any neck or face oedema in a woman with pre-eclampsia should forewarn
of a likely difficult intubation.38 These patients may even develop stridor and may have
difficult extubations.39 Airway oedema in pre-eclampsia can even lead to obstructive
sleep apnoea.37 Sleep-disordered breathing is probably underdiagnosed in pregnancy40
and may have adverse fetal effects.41 Obesity is more common and correlates with difficult intubations42, instrumental deliveries, a higher incidence of postpartum haemorrhage43, and an even greater incidence of gastric acid aspiration during general
anaesthesia.44 The upper airway in pregnancy is prone to contact bleeding with any
airway manipulation. This mucosal swelling leads to a smaller laryngeal inlet. Any airway intervention should involve a skilled anaesthetist with a fluent difficult/failed intubation drill.
Ventilation
The aims of ventilation are to oxygenate, ventilate and relieve the work of breathing. In
pregnancy, ventilation can be problematic. As in the non-pregnant patient, ventilation
may be invasive, through an endotracheal tube or tracheotomy, or non-invasive,
through a tightly fitting facemask. NIV may exert CPAP or bi-level positive airway pressure (BiPAP). CPAP is the application of a constant positive end-expiratory pressure,
called positive end-expiratory pressure in the invasively ventilated patient. Non-invasive CPAP differs from BiPAP in several ways. BiPAP enables spontaneous breathing
over two pressure levels, and is therefore more effective at clearing CO2, hence its
use in hypercapnic respiratory failure. CPAP circuits can deliver a higher fractional inspired oxygen concentration (FiO2) and can be humidified, and the FiO2 can be measured accurately. With BiPAP, FiO2 is not measured accurately, oxygen is entrained in
L/min, the circuit tolerates leaks, and the delivered FiO2 is dependent on the inspiratory flow rate.
Non-invasive ventilation. The evidence for acute NIV is well established in the treatment
of hypercapnic respiratory failure in chronic obstructive pulmonary disease, where it
reduces the need for intubation in mildly acidotic patients.45 There is also good evidence for NIV in cardiogenic pulmonary oedema and in immunocompromised states.
It has physiological benefit in other causes of respiratory failure, but should be used in
a controlled, monitored setting as a trial of therapy. Patients should be awake with
good respiratory drive, haemodynamically stable and without excessive respiratory secretions. Those with more marked acidosis (pH <7.25) should be considered for earlier invasive ventilation. There is less evidence for its use in other causes of
hypoxaemic respiratory failure, but it has been used in asthma, pneumonia, ARDS
and others including postoperative cases.46 There is little evidence for its use in the
obstetric population, although it has been used in sleep-disordered breathing during
pregnancy.40 NIV cannot therefore be currently recommended except as a closely
monitored trial of therapy in selected obstetric patients, in an ICU setting. Furthermore, there is a theoretical even greater risk of gastric acid aspiration with the gastric
distention that occurs with NIV.
Invasive ventilation. Invasive ventilatory support may be necessary in primary respiratory failure, when a trial of NIV fails, in severe acidosis, and with reduced levels of consciousness and respiratory drive. The overall management of a ventilated obstetric
patient is similar to the non-pregnant population. This includes adequate thromboprophylaxis, enteral feeding, sedation breaks and weaning techniques. Spontaneous

784 L. C. Price et al

breathing modes are preferred to maintain respiratory muscle strength, but increased
sedation and paralysis may be needed to optimize ventilation.47 Gastric acid suppression therapy should be routine, especially with the increased risk of gastric acid aspiration in these patients.
It is important to prevent ventilator-associated pneumonia (VAP), thought to follow
bacterial colonization of the upper respiratory tract. Diagnosis based on clinical criteria has poor specificity, and microbiological samples are used including direct bronchoalveolar lavage48, protected specimen brush49 or non-invasive endotracheal
aspiration. Biomarkers such as C-reactive protein and procalcitonin50 can be useful,
although they are probably less specific in the obstetric population. Methods to prevent VAP include nursing ventilated patients head-up, avoiding re-intubation and minimizing changes in ventilator circuits.51
Strategies for weaning from mechanical ventilation, including the use of weaning
protocols52, are no different to the non-obstetric population. Percutaneous tracheostomy insertion may be indicated early in patients requiring prolonged ventilation53,
bearing in mind that the upper airway in any parturient is more oedematous and prone
to contact bleeding.
Acute respiratory distress syndrome
Acute lung injury and ARDS are a disease spectrum with many underlying causes. The
diagnosis is based on three factors: the presence of pulmonary oedema; evidence of
poor oxygenation; and clinical, echo or direct evidence of normal left atrial pressure
to exclude a cardiogenic cause of pulmonary oedema.54 The extent of poor oxygenation is based on a P:F ratio of PaO2 to FiO2. For example, breathing air (FiO2 0.21),
the normal range of PaO2 would be 1317 kPa, giving a P:F ratio of 72 kPa. With
ARDS, the FiO2 may be 0.8 and the PaO2 only 15 kPa, giving a P:F ratio of 18.7 kPa
(or 142 mm Hg), hence indicating poor oxygenation. It is always important to note
the FiO2 when measuring blood gases for this reason (Table 5).
Any pathology causing inflammation-induced injury to the alveolarcapillary interface can lead to ARDS. There is no suggestion that it is more common in pregnancy,
but it may be the end result in several obstetric diseases. Maternal mortality may be
3560% with higher mortality post partum, and morbidity often persisting after recovery in survivors (Table 6).36
Ventilatory strategies in ARDS are similar to the non-pregnant patient with consideration of the normal progressive respiratory alkalosis and the effects of hypoxia, hypercapnia and acidosis on the fetus. Invasive ventilation is often necessary, and is
thought to contribute to a syndrome of lung injury itself. There has recently been
a drive towards a lung protective ventilation strategy to reduce the need for invasive
ventilation. The components of iatrogenic ventilator-associated lung injury include
those due to excess airway pressure (barotrauma55), too high tidal volumes (volutrauma56), airway collapse (atelectrauma9) and local inflammation (biotrauma57). The

Table 5. Definitions of acute respiratory distress syndrome (ARDS) and acute lung injury (ALI).54
Bilateral chest X-ray infiltrates
P/F ratio < 26 kPa (ARDS) or <39 kPa (ALI)
Pulmonary artery wedge pressure <18 mm Hg (or 2.4 kPa)

Aims of obstetric critical care management 785

Table 6. Causes of acute respiratory distress syndrome in pregnancy.

Causes specific to pregnancy
Massive haemorrhage, pre-eclampsia (especially with fluid overload), sepsis (due to chorioamnionitis,
endometritis, pyelonephritis), amniotic fluid embolism, trophoblastic embolism, gastric acid
aspiration
Other causes
Sepsis (other causes), pneumonia, transfusion-related acute lung injury, trauma, inhalational injury,
burns, near-drowning, acute pancreatitis etc.

principles of a protective ventilation strategy in ARDS are to minimize airway pressures and to tolerate a higher pCO2. There are, however, fetal effects of hypercapnia
and acidosis, so this permissive hypercapnia is not recommended in the obstetric patient. Furthermore, the low tidal volumes desired in ARDS in a non-obstetric patient
may not be sufficient for the higher ventilatory requirements needed in pregnancy. In
fact, higher tidal volumes and plateau pressures than usual may actually be necessary
for this reason, and higher SaO2 and pO2 targets are recommended. As in any setting,
positive end-expiratory pressure is useful to prevent end-expiratory lung collapse, and
can improve recruitment and oxygenation of alveolar units. These effects may be offset
by negative effects on cardiac filling with reduced systemic BP, exaggerated with intravascular fluid depletion.
Renal support
The incidence of acute renal failure (ARF) in pregnancy has reduced over time in both
developed (12.8%) and developing (1015%) countries with improvements in obstetric care, especially a reduction in septic abortions.58 The World Health Organization
has estimated that one in eight pregnancy-related deaths in the developing world are
the result of unsafe abortions and about 8% of these deaths are due to ARF.59
The diagnosis of ARF is complex with over 30 definitions of ARF in the literature.
The RIFLE criteria were developed to provide a uniform means of classifying ARF;
these have been modified recently and the term acute kidney injury (AKI) has
been introduced to encompass all causes of ARF. The classification is based on changes
in serum creatinine and/or urine output over a 48-h period.60 It is easy to apply clinically, and importantly highlights the requirement of repeated testing of serum creatinine and the close monitoring of urine output in any patient with suspected or at risk
of ARF.
Obstructive uropathy must always be excluded, and this can be difficult in pregnancy due to physiological hydronephrosis, therefore an early ultrasound can provide
an invaluable baseline reference for the future when obstructive uropathy is suspected.
Progesterone-induced ureteric smooth muscle relaxation plus compression by the
gravid uterus leads to ureteric and renopelviceal dilatation. There is greater hydronephrosis on the right side due to physiological engorgement of the right ovarian
vein and dextrorotation of the uterus.
Early recognition and treatment of AKI saves nephrons and prevents further decline
in glomerular filtration rate (GFR). Serum creatinine concentration is a poor marker of
GFR because it does not rise appreciably until GFR has fallen significantly. From the
first trimester of pregnancy until term, there are significant increases in renal blood

786 L. C. Price et al

flow (5085%) and GFR (4065%); consequently, there is a parallel decline in the serum creatinine concentration. Therefore, small changes in the serum creatinine concentration can represent a significant deterioration in GFR. The calculation of
estimated GFR using the modification of diet in renal disease study (MDRD) equation
is not recommended in pregnancy as it overestimates GFR.
Aetiology
Conventionally, the major causes of AKI are grouped into three general categories:
prerenal; intrinsic renal; and obstructive causes. This holds true in pregnancy, but
most importantly includes the placenta-driven diseases including pre-eclampsia and
HELLP (haemolysiselevated liver enzymeslow platelets) syndrome (Table 7).
Urinary electrolyte measurements are rarely helpful clinically as they are unreliable
in distinguishing between prerenal failure and acute tubular necrosis, and they do not
affect management. It is useful to quantify the degree and change in proteinuria, which
requires repeated random urine samples for protein:creatinine ratio measurement.
This method has been validated in hypertensive and non-hypertensive pregnant
patients and is comparable to 24-h collections.61
Management of acute kidney injury
The mainstay of treatment in ARF is aimed at minimizing damage to surviving nephrons
whilst providing support until the kidney recovers. This includes the removal of tubular toxins, specific treatment of glomerular diseases and restoration of the circulation.
Haemodynamic monitoring using clinical and invasive methods guides volume resuscitation and the use of vasopressors with the goal of improving perfusion pressure and
urine output (>0.51 mL/kg/h). Failure to achieve this goal will usually become apparent at an early stage, and persistent oliguria will eventually lead to volume derangements. Oliguria can be converted to a non-oliguric state by using diuretics and lowdose dopamine, liberating valuable time prior to commencing renal replacement therapy. There is, however, little evidence to suggest that these two agents improve renal
recovery times; ultimately, this depends on the number of remaining functional nephrons increasing their filtration to maintain GFR. In the postpartum patient, oliguria may
Table 7. Aetiology of acute kidney injury.
Pre renal
Hypovolaemia
(e.g. sepsis, haemorrhage)
Low cardiac output states
(e.g. heart failure)
Pre-eclampsia and
HELLP syndrome

Intrinsic renal
Inflammatory
(eg anti-GBM (glomerular basement membrane)
disease, ANCA-associated GN, SLE)
Thrombotic
[e.g. DIC, thrombotic micro-angiopathy
(TTP, HUS), antiphospholipid syndrome]
Interstitial nephritis (e.g. infections, drugs),
pre-eclampsia and HELLP syndrome

Post renal
Urolithiasis

Blood clot

Urethral stricture

ANCA, antineutrophil cytoplasmic antibody; GBM, glomerular basement membrane; GN, glomerulonephritis; SLE, systemic lupus erythematosus; DIC, disseminated intravascular coagulation; TTP, thrombocytopenic purpura; HUS, haemolytic uraemia syndrome; HELLP, haemolysiselevated liver enzymeslow
platelets.

Aims of obstetric critical care management 787

be normal, and, as described above, inappropriate intravenous fluids can precipitate

iatrogenic pulmonary oedema. Lower targets for urine output are therefore
appropriate.
Metabolic derangements are an important aspect of AKI management. Hyperkalaemia is best managed by limiting potassium input, augmenting elimination with diuretics,
and promoting cellular uptake with insulin and glucose boluses. Severe acidosis (pH
<7.2) may require treatment with sodium bicarbonate, although avoiding hypernatraemia and excessive volume expansion. The intracellular acidosis that follows should be
remembered, although it is not a problem if the spontaneously breathing patient is able
to blow off the CO2 or if the patient is being adequately mechanically ventilated.
The indications for renal replacement therapy are the same in pregnancy as for the
general population, but it is rarely required (<1 in 10 00015 000 pregnancies). There
is no clear evidence guiding dialysis dose prescription in AKI. Pregnant chronic dialysis
patients achieve better fetal outcomes at higher dialysis doses (usually >20 h/week)
compared with the standard regimen of 4 h thrice weekly.62 GFR is higher in pregnancy so the goal should be to avoid under-dialysing these patients. The aim is for
more frequent and longer dialysis sessions if using haemodialysis, and in the ICU
setting, continuous haemofiltration should be >35 mL/kg/h.
Hypothalamic pituitary adrenal function in critical illness
Hypothalamic pituitary adrenal function during critical illness is important and has
been shown to affect outcome in septic patients.63 Adrenergic receptor desensitization has been demonstrated with a reduction in alpha- and beta-adrenergic receptors.
Steroids have been shown to help resensitize these receptors.
The evaluation of adrenal function during critical illness is controversial. Adrenal function should still be tested, but commencement of steroid therapy should not be withheld
until results are available.64 Currently, testing relies on random and low-dose Synacthenstimulated serum total cortisol levels. In sepsis, adrenal insufficiency is likely when baseline cortisol levels are <10 mg/dL or the change in cortisol is <9 mg/dL, and unlikely when
the Synacthen-stimulated cortisol level is 44 mg/dL or the change in cortisol is
16.8 mg/dL.9 Free cortisol or salivary cortisol when they become available will be
more useful. Plasma aldosterone and renin level are useful when primary adrenal disease
is suspected, but pregnancy-specific normal ranges must be used.
When suspected, glucocorticoid doses should be physiological with either a continuous infusion of hydrocortisone or 46-hourly boluses with a daily dose 200 mg.
This regimen should be discontinued once there has been clinical improvement.
Such patients may include those requiring >24 h or an increasing dose of vasopressors, or those with diseases such as meningococcaemia.
Neurological management
Neurological support aims to relieve pain and anxiety. Sedation is usually required for
mechanical ventilation and invasive procedures. Agents include analgesics (paracetamol, opioids) and sedative-anxiolytics (benzodiazepines, propofol, haloperidol and clonidine). These are all used as needed, with consideration given to the implications on
the fetus. Intubation and some modes of ventilation may also require neuromuscular
blockade. Daily sedation breaks are recommended to allow a circadian sleepwake cycle and to re-assess neurology in these patients. Very agitated patients may need

788 L. C. Price et al

sedation if intensive monitoring and treatment is required. ICU delirium is common

and difficult to manage.65
Investigations for coma would be as in the non-pregnant population with computed
tomography / magnetic resonance imaging, blood tests including cultures, lumbar
puncture if considering meningitis, encephalitis or demyelination, and urine for toxicological screening (Table 8). Electroencephalography can be useful to assess metabolic
coma and non-convulsive status, and hypoxic brain injury. Antidotes such as naloxone
and flumazenil may be needed but with consideration of the risk of precipitating seizures and fetal effects. Cases of poisoning or overdose should be discussed with
the local poisons unit.
For those with known epilepsy, the risk of seizures increases at the time of delivery. For women at high risk for intrapartum seizures, intravenous phenytoin or rectal
carbamazepine can be given pre-emptively. In pre-eclampsia, the development of seizures defines eclampsia, and this is prevented and treated with magnesium sulphate.
There are many other causes of seizures in pregnancy (including conditions listed in
Table 8) as well as others including thrombotic thrombocytopaenic purpura) and initial management focuses on resuscitation with control of fitting. Metabolic abnormalities including hypoglycaemia and hypocalcaemia should be corrected and intravenous
magnesium sulphate given, even without known pre-eclampsia; eclampsia may present
de novo. Further seizure management is with lorazepam or diazepam, then loading
with intravenous phenytoin (20 mg/kg). Status epilepticus (seizures lasting >30 min
or multiple seizures without regaining consciousness) is managed as outside pregnancy, remembering the teratogenic effects of anticonvulsants during the first trimester, and fetal respiratory effects if peripartum.
The use of mild therapeutic hypothermia following cardiac arrest is a fairly recent
treatment concept that followed two large multicentre randomized
controlled trials

published in 2002.66,67 It involves active cooling to 3234 C when return of spontaneous circulation (ROSC) follows a ventricular fibrillation or ventricular tachycardia
out-of-hospital arrest. The proposed mechanism is a reduction in cerebral metabolic
rate which, with less free radical production and intracellular acidosis, is neuroprotective.68 However, adverse events including arrhythmias, coagulopathy, immunoparesis,
and abnormal blood glucose and acidbase control may occur, and its use in pregnancy
has not been reported to date.

Table 8. Causes of coma or collapse in pregnancy.

Intracerebral haemorrhage: arteriovenous malformation (AVM), Berry aneurysm, pre-eclampsia,
hypertension
Subarachnoid haemorrhage
Trauma: extradural or subdural haematoma
Prolonged hypotension or hypoxia
Embolic or ischaemic stroke, venous sinus thrombosis
Pituitary apoplexy
Brain structural abnormality, e.g. tumour
Infection: abscess, meningitis, encephalitis, septic encephalopathy
Demyelination
Metabolic: hypoglycaemia, hyponatraemia, acid-base disorders, rare causes (e.g. carbamyltransferase
deficiency)
Severe depression

Aims of obstetric critical care management 789

Neurological intensive care

The principles of neurological intensive care management are similar to the non-pregnant setting, with prevention of secondary brain injury following an initial insult. The
aetiology of most brain insults can be subdivided into those causing coma and those
leading to traumatic brain injury (TBI); the leading cause of mortality in young people.
Severe cases of TBI are best managed in major hospitals with neurosurgical and
neuro-intensive care facilities.69 Management is as for the non-pregnant woman using
the airways, breathing, circulation (ABC) approach (and cervical spine control) and
consideration for the fetus; immediate delivery may be indicated. Intubation and
ventilation is indicated when the Glasgow Coma Scale score is <8/15, or drops by
2 points, or earlier depending on the pathology and management plan. In trauma,
the secondary survey is essential in order not to miss other life-threatening
injuries.70
Prevention of secondary brain injury entails maintaining cerebral perfusion pressure (CPP) and normoglycaemia, and preventing complications such as infection.
CPP depends on the pressure difference between MAP and intracranial pressure
(ICP) within the rigid box that is the cranium (CPP MAP-ICP). ICP is usually
<10 mm Hg; an ICP >20 mm Hg may distort cerebral architecture, reduce cerebral
blood flow and lead to ischaemia and oedema. It is important to maintain MAP >80
90 mm Hg to maintain CPP in the non-obstetric population, and this target may be
lower in pregnancy.
Cerebral oedema may occur in TBI or other metabolic causes of coma due to disruption of the bloodbrain barrier, impaired mitochondrial ability to maintain normal
ionic cellular gradients, and ischaemia. ICP monitoring may be indicated, as the complication if left untreated is increasing pressure leading to coning and brain death. ICP
monitoring may be indicated and depends on local practice. Initial management for
raised ICP is sedation, mannitol, moderate hyperventilation and cerebrospinal fluid
drainage. Further therapies include profound hyperventilation, barbiturate coma and
surgical decompression. In the obstetric patient, the effects of further hypocapnia
are not clear. The diagnosis of brain death is no different to the non-pregnant patient,
with considerations for perimortem caesarean section for fetal survival.71 Somatic
support of a pregnant women following brain death is rare72, and organ support to
preserve the fetoplacental unit must consider the predictable physiological changes
that occur following brain death. These include haemodynamic instability from the excessive sympathetic discharge, panhypopituitarism, hypothermia, poor nutrition and
infectious complications.71
Haematological management
Haematological disorders are relatively common in pregnancy. Pregnancy is a thrombophilic state with increased levels of most procoagulants that prevent excessive maternal blood loss at the time of delivery. This and the reduction in venous flow due to the
gravid uterus leads to a five-fold increased risk of venous thromboembolism during
and immediately following pregnancy73, and is even higher in the puepurium.74 Prevention of venous thromboembolic events is therefore especially relevant in obstetrics.
Prophylaxis with low-molecular-weight heparin (LMWH) or low-dose unfractionated
heparin, and the use of a mechanical intermittent compression device or compression
stockings is essential. The anticoagulant dose of LMWH is increased as glomerular filtration is elevated; for example, the dose of enoxaparin is 1 mg/kg bd, reverting to the

790 L. C. Price et al

usual 1.5 mg/kg od dose following delivery.75 The prophylactic subcutaneous dose is
40 mg/day in a normal-sized woman.
Anaemia is common in critical illness for many reasons. Blood transfusion targets in
the non-pregnant population are >7 g/dL, except in those with cardiac disease who
are kept >8 g/dL.76 Pregnancy targets should be similar, bearing in mind normal physiological anaemia.
Disseminated intravascular coagulation (DIC) is an acquired coagulopathy that
follows a trigger of generalized coagulation activity. Other causes of coagulopathy
are outlined in Table 9. In DIC, further consumption of platelets, clotting factors
and fibrin occurs, leading to a cycle of continuing bleeding and consumption of clotting components. In pregnancy, it should be anticipated from some of the associated conditions such as massive obstetric haemorrhage (especially placental
abruption with direct exposure to fetal material), pre-eclampsia, retained fetal products and amniotic fluid embolism. Sepsis is the most common overall cause in any
ICU population, and other causes include transfusion reactions, trauma and drugs.
Diagnosis is based on deranged clotting in an appropriate clinical setting with low
fibrinogen and elevated fibrinogen breakdown products (FDPs) or D dimer levels.
Management is similar to that of major obstetric haemorrhage with prompt resuscitation and fluid replacement, with identification and treatment of the underlying
cause. Blood products need to be given as soon as available including packed red
cells, fresh frozen plasma, cryoprecipitate and platelets. Recent developments include recombinant activated factor VIIa (NovoSeven), originally used for haemophilia and factor VII deficiency. It has been used for intractable blood loss and
fulminant disseminated intravascular coagulation as it induces short-term local haemostasis. It has been used (off licence) with life-saving results in cases of massive
obstetric haemorrhage.77
NUTRITION AND GASTROINTESTINAL MANAGEMENT
Nutrition in pregnancy is important for the development of a healthy baby. The
gastrointestinal system is the portal through which macro- and micronutrients enter

Table 9. Causes of coagulopathy in pregnancy.

Obstetric causes
DIC due to placental abruption,
placenta praevia, massive obstetric
haemorrhage, amniotic fluid embolus
Thrombocytopenia, e.g. HELLP syndrome,
TTP
Liver failure due to acute
fatty liver of pregnancy
Bone marrow failure due to effects of
shock (many causes)
Iatrogenic

Non-obstetric causes
DIC due to any other causes including
sepsis, trauma, incompatible transfusion reaction
Thrombocytopenia, e.g. ITP, sepsis, alcohol
Liver failure due to alcohol, paracetamol overdose etc.
Bone marrow failure or infiltration (many causes)
Pre-existing, e.g. haemophilia
Hypothermia

DIC, disseminated intravascular coagulation; TTP, thrombocytopenic purpura; HELLP, haemolysis

elevated liver enzymeslow platelets; ITP idiopathic thrombocytopaenic purpura.

Aims of obstetric critical care management 791

the body. Proteins, fats and complex carbohydrates are digested, and vitamins, minerals and water are absorbed across the gut mucosa. In pregnancy, there is an increased requirement for zinc, folate and vitamin B12 in the first trimester, and
calories in the second and third trimester. Nutritional support is required to meet
the increased metabolic demand and limited nutritional reserve during critical illness.
However, despite extensive research, many areas in this field of critical care medicine
remain controversial, due to a lack of well-controlled randomized trials.
The availability of functioning small bowel is vital and should prompt early feeding,
preferably via the enteral route. Enteral nutrition is cheap and safe, but can be complicated by aspiration and poor gastric emptying, which results in poor absorption and
under-feeding. Increased progesterone levels in pregnancy cause smooth muscle relaxation. This results in an increased likelihood of aspiration and constipation due to the
reduced oesophageal
sphincter tone and bowel peristalsis. Aspiration can be pre
vented by a 45 head-tilt position and confirming the nasogastric tube placement radiologically. Good absorption of feed often requires the use of prokinetic agents to
promote gastric emptying when residual gastric volumes are high, and laxatives and
enemas to prevent constipation. If poor absorption is a problem, drugs given via the
nasogastric tube should also be considered to be poorly absorbed, and prompt conversion to the intravenous route should be instituted. Some drugs require enteral feed
to be discontinued to ensure good uptake and effective absorption, such as the anticonvulsive drug phenytoin.
Protocols have been developed and are implemented by nurses to ensure that adequate enteral nutrition is delivered early and effectively. The failure to deliver 25% of
a patients calculated calorific requirement has been shown to result in increased mortality and infection rates.78 Protocol-directed regimens have been shown to reduce the
incidence of under-feeding; however, it remains a significant problem for many patients
due to frequent and often avoidable interruptions.
Total caloric requirements can be either estimated or calculated, and are often estimated at 2535 kcal/kg/day. Calculations are required at extremes of weight and performed by dieticians. The recommended nutritional requirements of a septic patient
are 25 total kcal/kg/day, given in three forms: protein 1.32.0 g/kg/day; glucose
3070% of non-protein calories; and lipids 1530% of non-protein calories.
The H2 receptor blocker ranitidine is used as prophylaxis against stress ulceration;
there is no evidence to suggest that proton pump inhibitors are superior.79 The National Institute for Health and Clinical Excellence recommends that parenteral nutrition should be limited to 50% of the calculated calorific requirement, and it has
been shown to be less harmful than first thought. It is useful when the small bowel
is being rested or not functioning for prolonged periods. Glycaemic control has also
been shown to affect outcome in sepsis, and the Surviving Sepsis Campaign guidelines
advise controlling blood glucose <8.3 mmol/L. This should be achieved using insulin
infusions and close monitoring of blood glucose.8
Micronutrients and electrolytes, e.g. magnesium, iron, copper, zinc and selenium,
are necessary in small amounts. Phosphate is important since it is required for normal
metabolic processes resulting in the formation of ATP. Hypophosphataemia results in
reduced contractility of skeletal muscles including respiratory muscles, which can lead
to difficulties weaning from ventilatory support. Other micronutrients include fat-soluble (vitamins A, carotene) and water-soluble vitamins (vitamins B, C, D and E). The
precise requirements for specific vitamins remain unclear, although several studies
have shown extremely low circulating concentrations. In pregnancy, the supplementation of folate should be continued.

792 L. C. Price et al

Early goal-directed therapy

Haemodynamic goal-directed therapy (GDT) was first used in sepsis and has since
been introduced in the peri-operative care setting, targeting oxygen delivery as the
goal. Original studies of sepsis in the late 1980s noted that survivors of sepsis had
supranormal levels of oxygen delivery (DO2) compared with controls.80 However, targeting high DO2 levels did not initially appear to improve outcome.81 It became apparent from a meta-analysis of all the optimization trials that the timing of such
interventions was crucial82 where the earlier interventions did in fact reduce mortality.
There are evidently two phases to sepsis. The early phase is characterized by low oxygen delivery, high lactate and low SvO2 levels. This phase is associated with high morbidity and mortality if left untreated, but may be reversible if targeted early.83 In
comparison, the later phase of established sepsis is more hyperdynamic with a higher
SvO2, CO and DO2 (Table 10).84
This early work led to a landmark study of sepsis before admission to ICU. Rivers
et al studied patients with severe sepsis and septic shock over 6 h following presentation to the emergency department (ED).10 Patients with systemic inflammatory response syndrome criteria and BP <90 mm Hg or serum lactate >4 mmol/L were
randomized to standard ED care or the early GDT protocol with continuous central
venous oxygen saturations (ScvO2) monitoring with goals as follows: CVP 8
12 mm Hg, MAP > 65 mm Hg, urine output (UO) > 0.5 mL/kg/h, ScvO2 > 70% (or
SvO2 > 65%). If the ScvO2 was below target, protocolized manouevres were undertaken to increase DO2. These included fluid boluses, giving packed cells to increase
haematocrit >30% and using vasopressors or inotropes. The other important feature
in the trial was the use of antibiotics within 1 h. The overall results from the early
GDT protocol indicted a significant mortality benefit (30.5% for early GDT vs
46.5% for standard care). The important message from this is that survival in sepsis
does appear to be influenced by the care in the first few hours, and that sepsis appears
to have an early reversible phase. It has revolutionized the importance of early diagnosis and accessibility to critical care facilities for patients admitted with septic shock
and severe sepsis. This principle has been extrapolated to the peri-operative care of
patients.85
Cardiopulmonary resuscitation in pregnancy
Cardiac arrests in pregnancy are fortunately rare, occurring in approximately one in
30 000 late pregnancies. They are less likely to have a primary cardiac cause compared
Table 10. Definitions of sepsis syndromes.83
Infection

The inflammatory response to invasion of micro-organisms into normally sterile

host tissue

SIRS
Response to a variety of clinical insults with more than two of: temperature <36 C

or >38 C; pulse >90 beats/min, respiratory rate >20/min; white cell count >12 or <4
Sepsis
The systemic response to infection with more than two of the above SIRS criteria
Severe
Sepsis with organ dysfunction, hypoperfusion or hypotension
sepsis
(systolic blood pressure 40 mm Hg below baseline)
Septic shock Sepsis with hypotension despite adequate fluid resuscitation
SIRS, systemic inflammatory response syndrome.

Aims of obstetric critical care management 793

Table 11. Causes of cardiac arrest in pregnancy.

Obstetric causes
Pulmonary embolism
Amniotic fluid embolism
Massive blood loss
Complications of pre-eclampsia
(and magnesium toxicity)
Uterine inversion
Aortic dissection

Non-obstetric causes
Trauma
Septic shock
Drug overdose
Acute myocardial infarction
Severe asthma
Local anaesthetic toxicity
Failed intubation, high spinal block

with cardiac arrests outside pregnancy. The most common reason for syncope is the
supine hypotension syndrome, and management of patients in the left lateral position
is essential. Other common causes of cardiac arrest in pregnancy are shown in Table 11.
The concept of a chain of survival with basic and advanced life support is as important in pregnancy as in the general population. It includes early recognition and call for
help, early cardiopulmonary resuscitation (CPR), early defibrillation and postresuscitation care. The structured ABC, airway-breathing-circulation approach in cardiac arrest
underpins the principles of basic and advanced life support with the goal of establishing
a spontaneous return of circulation.
CPR in pregnancy is complicated by both physiological and physical changes that are
well established late in pregnancy. They affect the quality of CPR that can be delivered,
and increase the possibility of complications. Each component is considered below, including the important differences in pregnancy.
Airway
Simple airway manoeuvres remain the same as for the general population. Difficult intubations are more common and effective pre-oxygenation is essential. Insertion of an advanced airway at an earlier stage and with a 0.51 mm smaller endotracheal tube is
recommended. The short obese neck and full breasts make insertion of the laryngoscope

more problematic. Short-handled laryngoscopes, blades mounted greater than 90 and
dismountable blades first inserted into the mouth can all help to overcome these difficulties. A laryngeal mask airway may be needed in cases of failed intubation, although cricoid
pressure should be released and re-applied after insertion and inflation of the mask.
Breathing
Mouth-to-mouth and bag and mask ventilation should be performed without a pillow
and the head fully extended. Ventilation can be hampered by flared ribs, raised diaphragm, reduced chest compliance and difficulties seeing the chest rise. These along
with greater oxygen demand and reduced FRC lead to rapid desaturation with inadequate ventilation. The problems of ventilation can be compounded by aspiration,
which is more likely in late pregnancy due to the incompetent gastro-oesophageal
sphincter and raised intragastric pressure.
Circulation
Femoral intravenous access for resuscitation drugs is not advised, as central maternal
delivery will not occur until the fetus is delivered. Chest compressions should be

794 L. C. Price et al

performed at the standard rate of 100/min and a ratio of 30:2. Hands should be positioned further up the sternum because abdominal contents are displaced upwards
by the gravid uterus.
The patient must be rolled into the left lateral position. This

should be up to 30 or there is a tendency to roll into the full lateral position. This
angle is ideal for effective chest compression, and aids displacing the uterus off the inferior vena cava. This manouevre, along with raising the patients legs, will improve venous return. One technique to achieve this is the human wedge, whereby the patient
is tilted on to a rescuers knees to provide a stable position for CPR.
Defibrillation and drug administration is similar in pregnancy as for the general population. Defibrillation will not deliver significant current to the fetus, but any fetal monitoring electrodes must be removed. Adhesive defibrillation pads have removed the
difficulties of applying the paddles in the correct position, normally affected by the
full breasts and left lateral position adopted. Excessive magnesium sulphate used to
treat and prevent eclampsia can contribute to cardiac arrest. Empiric calcium should
be given and can be life saving.
Perimortem caesarean section
Failure to achieve ROSC after 5 min of CPR should prompt consideration of perimortem delivery. This should be considered by the team leader at the onset of cardiac arrest. The greatest chance of infant survival is in pregnancies >24 weeks of gestation as
neonatal resuscitation can commence. It is obviously an aggressive procedure, but infant and maternal survival may depend on it. In pregnancies >20 weeks, emptying the
uterus after this will improve CO by 6080% of prepregnancy levels, allowing venous
return and improving chances of ROSC, improving maternal and fetal outcome. Cardiac compressions are also more effective after delivery.86 In pregnancies <20 weeks,
the gravid uterus is unlikely to have detrimental compressive effects. In pregnancies
<23 weeks, infant survival is unlikely; if done, it should be for maternal survival.
Between 1975 and 2000, there were 56 postmortem caesarean deliveries with six
neurologically normal surviving infants (survival rate 10.7%). Eight infants were delivered alive but died in the neonatal period. Of the 40 perimortem deliveries, 25 survived neurologically intact (survival rate 62.5%).87
Post-resuscitation care
ROSC is an important step in resuscitation, but this must be followed by meticulous
postresuscitation care to increase the chance of a return to normal cerebral function.
It focuses on the re-assessment of ABCDE to ensure good oxygenation, stable cardiac
rhythm and BP. Once established, the patient can be transferred to a critical care area
for ongoing postresuscitation care.
SUMMARY
It is important to understand the aims of managing critically ill obstetric patients in
many areas of clinical medicine, surgery and anaesthesia. The most common reasons
for UK obstetric ICU admissions are pre-eclampsia, sepsis and haemorrhage. The maternal physiological differences make a significant impact on the assessment and management, and must be considered at all times, especially the supine hypotension
syndrome in positioning and resuscitation scenarios. Additional considerations are
for fetal wellbeing and appropriateness for delivery, which may be life saving for the

Aims of obstetric critical care management 795

mother. The developments and trials in critical care medicine over the last 50 years
have not included obstetric patients specifically. However, with the changes in maternal physiology borne in mind, there are no reasons why the principles should differ
greatly. When considering haemodynamic management, the requirement for a high
maternal CO should be remembered. The low colloid oncotic pressure leads to an
increased tendency to develop iatrogenic pulmonary oedema, and several obstetric pathologies tend to increase alveolar capillary permeability, compounding the problem.
Oliguria post delivery, especially in pre-eclampsia, should not usually be managed
with a fluid challenge. Airway management may be challenging with altered anatomy

Practice points
General supportive care
 immediate resuscitation of the mother is paramount for maternal and fetal
survival
 consider fetal age at an early stage; monitoring/assessment and delivery if
indicated
 early identification of warning signs of critical illness
 always plan in advance with the multidisciplinary team, where possible, such as
booking an ICU bed for expected high-risk deliveries
Haemodynamic management
 remember the supine hypotension syndrome
 beware of iatrogenic pulmonary oedema with fluid challenges
 labour is a high-risk time for cardiac patients, with fluid shifts for up to 72 h
post partum
Cardiopulmonary resuscitation
 airway: pregnant women may be very difficult to intubate (need skilled anaesthetist). Need smaller endotracheal tube, short-handled laryngoscope blade
and difficult intubation equipment
 breathing: may be difficult to ventilate. Increased risk of gastric acid aspiration
 circulation: always use lateral position to avoid the supine hypotension syndrome. CPR cycles and drugs given as outside pregnancy, remembering that
MgSO4 toxicity can lead to cardiac arrest (give empiric calcium intravenously)
 perimortem caesarean section may be necessary and should be considered at
the start of CPR

Research agenda
 more on the mechanism of ARDS in pre-eclampsia, and how to avoid iatrogenic pulmonary oedema
 ideal ventilatory targets throughout pregnancy
 short-term effects of fetal hypoxia, hypercapnia and acidosis

796 L. C. Price et al

and high oxygen consumption, and rapid oxygen desaturation may occur. Ventilatory
management may be difficult for this reason and higher pressures may be needed.
Some of the recommended ARDS ventilatory strategies are probably not appropriate
as fetal haemoglobin binds oxygen less well when acidotic. Overall, critical care management in obstetrics is mostly similar to a non-obstetric setting, applying the changes
in maternal and fetal physiology in each case. High-risk patients should be identified
early, with appropriate planning and involvement of the multidisciplinary team.
ACKNOWLEDGEMENTS
The authors would like to thank Dr John Dick, consultant anaesthetist, for his helpful
comments.
REFERENCES
1. Lewis G (ed.). The Confidential Enquiry into Maternal and Child Health (CEMACH). Saving Mothers Lives:
Reviewing Maternal Deaths to Make Motherhood Safer, 20032005. The Seventh Report on Confidential Enquiries into Maternal Deaths in the United Kingdom. London: CEMACH, 2007.
2. Butwick AJ & Carvalho B. Can we improve maternal outcome for high-risk obstetric patients? Int J Obstet Anesth 2007; 16: 311313.
3. Maine D & Chavkin W. Maternal mortality: global similarities and differences. J Am Med Womens Assoc
2002; 57: 127130.
4. Baskett TF & Sternadel J. Maternal intensive care and near-miss mortality in obstetrics. Br J Obstet Gynaecol 1998; 105: 981984.
5. Kilpatrick SJ & Matthay MA. Obstetric patients requiring critical care. A five-year review. Chest 1992;
101: 14071412.
*6. Dellinger RP, Carlet JM, Masur H et al. Surviving Sepsis Campaign guidelines for management of severe
sepsis and septic shock. Crit Care Med 2004; 32: 858873.
7. Bernard GR, Vincent JL, Laterre PF et al. Efficacy and safety of recombinant human activated protein C
for severe sepsis. N Engl J Med 2001; 344: 699709.
8. Mebis L, Gunst J, Langouche L et al. Indication and practical use of intensive insulin therapy in the critically ill. Curr Opin Crit Care 2007; 13: 392398.
9. Annane D, Maxime V, Ibrahim F et al. Diagnosis of adrenal insufficiency in severe sepsis and septic
shock. Am J Respir Crit Care Med 2006; 174: 13191326.
*10. Rivers E, Nguyen B, Havstad S et al. Early goal-directed therapy in the treatment of severe sepsis and
septic shock. N Engl J Med 2001; 345: 13681377.
11. Robson SC, Boys RJ, Hunter S et al. Maternal hemodynamics after normal delivery and delivery complicated by postpartum hemorrhage. Obstet Gynecol 1989; 74: 234239.
12. Capeless EL & Clapp JF. When do cardiovascular parameters return to their preconception values? Am J
Obstet Gynecol 1991; 165: 883886.
13. Uebing A, Steer PJ, Yentis SM et al. Pregnancy and congenital heart disease. BMJ 2006; 332:
401406.
14. Bakker J, Gris P, Coffernils M et al. Serial blood lactate levels can predict the development of multiple
organ failure following septic shock. Am J Surg 1996; 171: 221226.
15. Spodick DH. Frank-Starling effect. Circulation 1979; 60: 718719.
16. Oian P, Maltau JM, Noddeland H et al. Oedema-preventing mechanisms in subcutaneous tissue of normal pregnant women. Br J Obstet Gynaecol 1985; 92: 11131119.
17. Clark SL, Cotton DB, Lee W et al. Central hemodynamic assessment of normal term pregnancy. Am J
Obstet Gynecol 1989; 161: 14391442.
18. Reid F, Lobo DN, Williams RN et al. (Ab)normal saline and physiological Hartmanns solution: a randomized double-blind crossover study. Clin Sci (Lond) 2003; 104: 1724.

Aims of obstetric critical care management 797

19. Barrett NA & Kam PC. Transfusion-related acute lung injury: a literature review. Anaesthesia 2006; 61:
777785.
20. Vincent JL & Weil MH. Fluid challenge revisited. Crit Care Med 2006; 34: 13331337.
*21. Masip J, Roque M, Sanchez B et al. Noninvasive ventilation in acute cardiogenic pulmonary edema: systematic review and meta-analysis. JAMA 2005; 294: 31243130.
22. Peter JV, Moran JL, Phillips-Hughes J et al. Effect of non-invasive positive pressure ventilation (NIPPV) on
mortality in patients with acute cardiogenic pulmonary oedema: a meta-analysis. Lancet 2006; 367:
11551163.
23. Masip J, Paez J, Merino M et al. Risk factors for intubation as a guide for noninvasive ventilation in
patients with severe acute cardiogenic pulmonary edema. Intensive Care Med 2003; 29: 1921
1928.
24. Follath F, Cleland JG, Just H et al. Efficacy and safety of intravenous levosimendan compared with dobutamine in severe low-output heart failure (the LIDO study): a randomised double-blind trial. Lancet
2002; 360: 196202.
25. Benlolo S, Lefoll C, Katchatouryan V et al. Successful use of levosimendan in a patient with peripartum
cardiomyopathy. Anesth Analg 2004; 98: 822824.
26. Gowda RM, Khan IA, Mehta NJ et al. Cardiac arrhythmias in pregnancy: clinical and therapeutic considerations. Int J Cardiol 2003; 88: 129133.
27. Mercier FJ, Bonnet MP, De la Dorie A et al. Spinal anaesthesia for caesarean section: fluid loading, vasopressors and hypotension. Ann Fr Anesth Reanim 2007; 26: 688693.
28. Dunser MW, Mayr AJ, Ulmer H et al. The effects of vasopressin on systemic hemodynamics in catecholamine-resistant septic and postcardiotomy shock: a retrospective analysis. Anesthesia and Analgesia
2001; 93: 713.
29. Krejci V, Hiltebrand LB, Jakob SM et al. Vasopressin in septic shock: effects on pancreatic, renal and
hepatic blood flow. Crit Care 2007; 11: R129.
*30. Beale RJ, Hollenberg SM, Vincent JL et al. Vasopressor and inotropic support in septic shock: an evidence-based review. Crit Care Med 2004; 32(Suppl.): S455S465.
31. Koonin LM, MacKay AP, Berg CJ et al. Pregnancy-related mortality surveillance United States, 1987
1990. MMWR CDC Surveill Summ 1997; 46: 1736.
32. Magee LA, Cham C, Waterman EJ et al. Hydralazine for treatment of severe hypertension in pregnancy:
meta-analysis. BMJ 2003; 327: 955960.
33. Cooper WO, Hernandez-Diaz S, Arbogast PG et al. Major congenital malformations after first-trimester exposure to ACE inhibitors. N Engl J Med 2006; 354: 24432451.
34. Catanzarite V, Willms D, Wong D et al. Acute respiratory distress syndrome in pregnancy and the puerperium: causes, courses, and outcomes. Obstet Gynecol 2001; 97: 760764.
35. Campbell LA & Klocke RA. Implications for the pregnant patient. Am J Respir Crit Care Med 2001; 163:
10511054.
*36. Cole DE, Taylor TL, McCullough DM et al. Acute respiratory distress syndrome in pregnancy. Critical
Care Medicine 2005; 33(Suppl.): S269S278.
37. Izci B, Riha RL, Martin SE et al. The upper airway in pregnancy and pre-eclampsia. Am J Respir Crit Care
Med 2003; 167: 137140.
38. Ezri T, Szmuk P, Evron S et al. Difficult airway in obstetric anesthesia: a review. Obstet Gynecol Surv 2001;
56: 631641.
39. Rocke DA & Scoones GP. Rapidly progressive laryngeal oedema associated with pregnancy-aggravated
hypertension. Anaesthesia 1992; 47: 141143.
40. Izci B, Vennelle M, Liston WA et al. Sleep-disordered breathing and upper airway size in pregnancy and
post-partum. Eur Respir J 2006; 27: 321327.
41. Sahin FK, Koken G, Cosar E et al. Obstructive sleep apnea in pregnancy and fetal outcome. Int J Gynaecol Obstet 2008; 100(2): 141146.
42. Hood DD & Dewan DM. Anesthetic and obstetric outcome in morbidly obese parturients. Anesthesiology 1993; 79: 12101218.
43. Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet Gynecol
2004; 103: 219224.
44. Adams JP & Murphy PG. Obesity in anaesthesia and intensive care. Br J Anaesth 2000; 85:
91108.

798 L. C. Price et al
45. Plant PK, Owen JL & Elliott MW. Early use of non-invasive ventilation for acute exacerbations of chronic
obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial.
Lancet 2000; 355: 19311935.
46. Garpestad E, Brennan J & Hill NS. Noninvasive ventilation for critical care. Chest 2007; 132:
711720.
47. Tobin MJ. Advances in mechanical ventilation. N Engl J Med 2001; 344: 19861996.
48. Chastre J, Viau F, Brun P et al. Prospective evaluation of the protected specimen brush for the diagnosis
of pulmonary infections in ventilated patients. Am Rev Respir Dis 1984; 130: 924929.
49. Fagon JY, Chastre J, Wolff M et al. Invasive and noninvasive strategies for management of suspected
ventilator-associated pneumonia. A randomized trial. Ann Intern Med 2000; 132: 621630.
50. Seligman R, Meisner M, Lisboa TC et al. Decreases in procalcitonin and C-reactive protein are strong
predictors of survival in ventilator-associated pneumonia. Crit Care 2006; 10: R125.
51. Dodek P, Keenan S, Cook D et al. Evidence-based clinical practice guideline for the prevention of ventilator-associated pneumonia. Ann Intern Med 2004; 141: 305313.
52. Ely EW, Baker AM, Dunagan DP et al. Effect on the duration of mechanical ventilation of identifying
patients capable of breathing spontaneously. N Engl J Med 1996; 335: 18641869.
53. Griffiths J, Barber VS, Morgan L et al. Systematic review and meta-analysis of studies of the timing of
tracheostomy in adult patients undergoing artificial ventilation. BMJ 2005; 330: 1243.
54. Ware LB & Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342: 13341349.
55. Slutsky AS. Lung injury caused by mechanical ventilation. Chest 1999; 116(Suppl): 9S15S.
56. Dreyfuss D, Soler P, Basset G et al. High inflation pressure pulmonary edema. Respective effects of high
airway pressure, high tidal volume, and positive end-expiratory pressure. Am Rev Respir Dis 1988; 137:
11591164.
57. Tremblay LN & Slutsky AS. Ventilator-induced injury: from barotrauma to biotrauma. Proc Assoc Am
Physicians 1998; 110: 482488.
58. Prakash J, Kumar H, Sinha DK et al. Acute renal failure in pregnancy in a developing country: twenty
years of experience. Ren Fail 2006; 28: 309313.
*59. Gammill HS & Jeyabalan A. Acute renal failure in pregnancy. Crit Care Med 2005; 33(Suppl.): S372S384.
60. Mehta RL, Kellum JA, Shah SV et al. Acute Kidney Injury Network: report of an initiative to improve
outcomes in acute kidney injury. Crit Care 2007; 11: R31.
61. Leanos-Miranda A, Marquez-Acosta J, Romero-Arauz F et al. Protein:creatinine ratio in random urine
samples is a reliable marker of increased 24-hour protein excretion in hospitalized women with hypertensive disorders of pregnancy. Clin Chem 2007; 53: 16231628.
62. Barraclough K, Leone E & Chiu A. Renal replacement therapy for acute kidney injury in pregnancy.
Nephrol Dial Transplant 2007; 22: 23952397.
63. Arafah BM. Hypothalamic pituitary adrenal function during critical illness: limitations of current assessment methods. J Clin Endocrinol Metab 2006; 91: 37253745.
64. Goodman S & Sprung CL. The International Sepsis Forums controversies in sepsis: corticosteroids
should be used to treat septic shock. Crit Care 2002; 6: 381383.
65. Pun BT & Ely EW. The importance of diagnosing and managing ICU delirium. Chest 2007; 132: 624636.
66. Safar PJ & Kochanek PM. Therapeutic hypothermia after cardiac arrest. N Engl J Med 2002; 346:
612613.
67. Bernard SA, Gray TW, Buist MD et al. Treatment of comatose survivors of out-of-hospital cardiac
arrest with induced hypothermia. N Engl J Med 2002; 346: 557563.
68. Busto R, Globus MY, Dietrich WD et al. Effect of mild hypothermia on ischemia-induced release of
neurotransmitters and free fatty acids in rat brain. Stroke 1989; 20: 904910.
69. Patel HC, Bouamra O, Woodford M et al. Trends in head injury outcome from 1989 to 2003 and the
effect of neurosurgical care: an observational study. Lancet 2005; 366: 15381544.
70. Chesnut RM. Avoidance of hypotension: conditio sine qua non of successful severe head-injury management. J Trauma 1997; 42(Suppl.): S4S9.
*71. Mallampalli A & Guy E. Cardiac arrest in pregnancy and somatic support after brain death. Critical Care
Medicine 2005; 33(Suppl.): S325S331.
72. Powner DJ & Bernstein IM. Extended somatic support for pregnant women after brain death. Critical
Care Medicine 2003; 31: 12411249.
73. Toglia MR & Weg JG. Venous thromboembolism during pregnancy. N Engl J Med 1996; 335: 108114.

Aims of obstetric critical care management 799

74. Refuerzo JS, Hechtman JL, Redman ME et al. Venous thromboembolism during pregnancy. Clinical
suspicion warrants evaluation. J Reprod Med 2003; 48: 767770.
*75. Greer IA & Nelson-Piercy C. Low-molecular-weight heparins for thromboprophylaxis and treatment of
venous thromboembolism in pregnancy: a systematic review of safety and efficacy. Blood 2005; 106:
401407.
76. Hebert PC, Wells G, Blajchman MA et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian
Critical Care Trials Group. N Engl J Med 1999; 340: 409417.
77. Haynes J, Laffan M & Plaat F. Use of recombinant activated factor VII in massive obstetric haemorrhage.
Int J Obstet Anesth 2007; 16: 4049.
78. OLeary-Kelley CM, Puntillo KA, Barr J et al. Nutritional adequacy in patients receiving mechanical
ventilation who are fed enterally. Am J Crit Care 2005; 14: 222231.
79. Brett S. Science review: the use of proton pump inhibitors for gastric acid suppression in critical illness.
Crit Care 2005; 9: 4550.
80. Shoemaker WC, Appel PL, Kram HB et al. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest 1988; 94: 11761186.
81. Gattinoni L, Brazzi L, Pelosi P et al. A trial of goal-oriented hemodynamic therapy in critically ill patients.
SvO2 Collaborative Group. N Engl J Med 1995; 333: 10251032.
82. Kern JW & Shoemaker WC. Meta-analysis of hemodynamic optimization in high-risk patients. Crit Care
Med 2002; 30: 16861692.
83. Parrillo JE, Parker MM, Natanson C et al. Septic shock in humans. Advances in the understanding of
pathogenesis, cardiovascular dysfunction, and therapy. Ann Intern Med 1990; 113: 227242.
84. Silance PG & Vincent JL. Oxygen extraction in patients with sepsis and heart failure: another look at
clinical studies. Clin Intensive Care 1994; 5: 414.
85. Pearse R, Dawson D, Fawcett J et al. Early goal-directed therapy after major surgery reduces complications and duration of hospital stay. A randomised, controlled trial ISRCTN38797445. Crit Care 2005;
9: R687R693.
86. Morris S & Stacey M. Resuscitation in pregnancy. BMJ 2003; 327: 12771279.
*87. Whitten M & Irvine LM. Postmortem and perimortem caesarean section: what are the indications?
J R Soc Med 2000; 93: 69.