Role of the anaesthetist in obstetric critical care

Best Practice & Research Clinical Obstetrics and Gynaecology Vol. 22, No. 5, pp. 917–935, 2008 doi:10.1016/j.bpobgyn.2008.06.006 available online at http://www.sciencedirect.com 9 Role of the anaesthetist in obstetric critical care Felicity Plaat * BA, MBBS, FRCA Consultant Anaesthetist Department of Anaesthesia, 5th Floor, Hammersmith House, Hammersmith Hospital, Du Cane Road, London W12 0HS, UK Sarah Wray MBBS, FRCA Locum Consultant Anaesthetist Department of Anaesthesia, 2nd Floor, Alex Wing, Royal London Hospital, Whitechapel, London E1 1BB, UK The anaesthetist plays a key role in the management of high-risk pregnancies, and must be a member of the multidisciplinary team that is required to care for the critically ill obstetric patient. Anaesthetists are trained in advanced life support and resuscitation. They are experienced in the management of the critically ill, and provide anaesthesia, sedation and pain management. The obstetric anaesthetist should undertake education of medical and midwifery staff in the early recognition, monitoring and treatment of the sick mother, resuscitation training, running ‘skills drills’ for emergency simulations, risk management and audit of maternal morbidity on the labour ward. To date, there is little evidence to inform the anaesthetic management of the critically ill obstetric patient; most recommendations and guidelines are based on the management of non-obstetric, critically ill patients. Management must be adapted to encompass the physiological changes of pregnancy. Evidence-based guidelines on management of the critically ill woman with specific obstetric conditions are also lacking. Key words: pregnancy; critical care; obstetric anaesthesia; haemorrhage; pre-eclampsia; respiratory failure; resuscitation. In both the high-dependency and intensive care settings, the anaesthetist can contribute in terms of resuscitation skills, organ support, pain management and knowledge of maternal physiology. Anaesthetists are trained in airway management, the use of non-invasive and invasive ventilation, vascular access, advanced haemodynamic * Corresponding author. Tel.: þ44 (0) 208 383 3991; Fax: þ44 (0) 208 383 5373. E-mail address: felicity.plaat@imperial.nhs.uk (F. Plaat). 1521-6934/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved. 918 F. Plaat and S. Wray monitoring and cardiovascular support. The obstetric anaesthetist should participate in the planning and care of the high-risk pregnancy, and should be involved at an early stage if complications arise. OBSTETRIC CRITICAL CARE Although some of the physiological changes of pregnancy increase the ability to withstand physiological stresses, others can make the parturient more vulnerable to developing critical illness, such as the increase in oxygen consumption and enhanced coagulability. Women with pre-existing medical disease are at particular risk. For example, parturients with cardiac disease tolerate the normal alterations in cardiovascular physiology, such as the increase in blood volume and cardiac output, poorly. Those caring for the critically ill obstetric patient on the delivery suite itself face unique challenges. A multidisciplinary approach is essential, with input from obstetricians, anaesthetists, physicians, neonatologists and midwives. Providing such a level of care on the delivery suite, where traditionally, the emphasis has been on promoting delivery as a normal process with minimal medical intervention, can be difficult. The relative rarity of severe illness in the obstetric population increases the challenge of early detection should it arise. In the antenatal period, the consequences of any maternal medical interventions on the fetus must be taken into consideration. Critical care outside the intensive care unit Ideally, high-dependency care should be delivered on or near the delivery suite to minimize the need to transfer patients, ensure the proximity of appropriately trained staff, and enable mother and baby to remain together.1 If this is unachievable, women should be transferred to a general high-dependency unit in the same hospital. Adult intensive care facilities should be available on site.2 A survey of consultant-led obstetric units in the UK found that only 41% of units had specific obstetric high-dependency beds, although 92% of those units without dedicated obstetric beds reported that they could provide temporary high-dependency facilities.3 A more recent survey found very similar results.4 The admission of obstetric patients to intensive care facilities in the developed world occurs in approximately two to four per 1000 deliveries.5 Ryan et al reported a trend towards reduced intensive care admissions once a dedicated obstetric high-dependency unit had been established within their unit.1 ROLE OF THE ANAESTHETIST IN THE ANTENATAL PERIOD It is essential that the obstetric anaesthetist is involved in the planning of high-risk pregnancies, and is informed at an early stage should complications occur. There should be clear, comprehensive and explicit guidelines for all staff caring for pregnant women regarding which mothers require anaesthetic review antenatally (Table 1). Dedicated anaesthetic antenatal clinics allow assessment and optimization of the patient by an experienced anaesthetist well in advance of delivery. Early antenatal care provides an opportunity for expert consultation where unusual disorders are concerned. The advantages of such early anaesthetic intervention are well illustrated by the high-risk cardiac parturient. Accounts of the peripartum management of women with pulmonary hypertension undergoing caesarean section reveal the complexity of antenatal preparation. Patients receive antenatal input from obstetricians, Role of the anaesthetist in obstetric critical care 919 Table 1. Indications for antenatal anaesthetic assessment. System Cardiovascular Respiratory Renal Neurological Musculoskeletal Haematological Anaesthetic Miscellaneous Disorder Congenital heart disease Ischaemic heart disease Cardiomyopathy Cystic fibrosis Acute or chronicrenal failure Spinal cord injury Spina bifida Multiple sclerosis Previous neurosurgery Scoliosis Previous spinal surgery Connective tissue disorders Thrombocytopenia Previous anaesthetic problem Suxamethonium apnoea Needle phobia Jehovah’s Witness Valvular heart disease Pulmonary hypertension Pacemaker Severe asthma Severe epilepsy Myasthenia gravis Cerebrovascular disease Ankylosing spondylitis Severe rheumatoid arthritis Coagulopathy Allergy to anaesthetic agents Severe obesity anaesthetists, intensivists, cardiologists and obstetric physicians, with optimization on the intensive care unit prior to delivery in some cases.6,7 Currently in the UK, only a minority of units run such clinics. This makes it all the more necessary for robust referral systems to be in place, allowing staff caring for the high-risk parturient to access anaesthetic services reliably. In centres without dedicated antenatal anaesthetic clinics, obstetric anaesthetists may attend the obstetric antenatal clinic regularly, or arrangements may be in place for high-risk women to be referred to the senior on-call anaesthetist covering the delivery suite. In conjunction with the obstetric team, management plans for both elective and emergency delivery should be discussed and, most importantly, disseminated, along with appropriate contingency plans. A written plan of care for anaesthesia or labour analgesia should be in the patient’s notes, and a copy should be available on the delivery suite should the patient present as an emergency, especially out of hours, when access to hospital-held notes can be unreliable. ANAESTHESIA AND ANALGESIA FOR DELIVERY Labour pain and the stress response Labour pain initiates a neuro-endocrine response involving the secretion of adrenocorticotrophic hormone, endorphins and vasopressin. Stimulation of the sympathetic nervous system increases plasma catecholamines. The magnitude of this response is related to pain intensity.8 This stress response leads to an increase in metabolic rate and oxygen consumption. This increases the likelihood of hyperventilation. The accompanying respiratory alkalosis will cause a left shift of the maternal 920 F. Plaat and S. Wray oxyhaemoglobin dissociation curve, impairing oxygen delivery to the fetus. At the same time, the rise in circulating maternal catecholamines stimulates an increase in systemic vascular resistance, which may compromise placental blood flow. In late labour, the mother develops metabolic acidosis, worsened by hyperventilation, which is passed on to the fetus thus reducing its ability to withstand the detrimental effects of intra-uterine hypoxia. The critically ill mother and her fetus are especially vulnerable to these detrimental consequences of labour pain, particularly the parturient with pulmonary hypertension or a cardiac condition associated with impaired ventricular function or a fixed cardiac output. Labour analgesia for the critically ill parturient Only neuraxial blockade can abolish labour pain (Table 2). In the absence of contra-indications, regional analgesia should be offered early in labour and kept under close review to ensure that analgesia is effective throughout. Effective regional analgesia minimizes the cardiovascular effects of labour pain, but a rapid onset of sympathetic blockade must be avoided. Sympathetic blockade causes vasodilatation and reduced venous return, and may reduce cardiac output, especially in women whose compensatory mechanisms are obtunded by critical illness. These effects will be compounded by aortocaval compression; the supine position should be scrupulously avoided and the parturient should be nursed in the sitting or lateral position. The successful use of low-dose epidural analgesia has been described in parturients with fixed cardiac output states.9 In order to minimize sympathetic blockade, the amount of local anaesthetic is reduced through the use of boluses rather than infusion to maintain analgesia, and the addition of an opioid to the epidural solution. The use of a mixture of bupivacaine 0.1% and fentanyl 2 mg/mL has been described frequently. It should be remembered that neuraxial opioids are also associated with side-effects that are potentially hazardous to the high-risk parturient. A meta-analysis of the use of epidural analgesia compared with systemic opioid analgesia in labour has shown an improved neonatal acid-base status at delivery with epidural analgesia, suggesting that placental exchange can be well preserved in association with (minimal) maternal sympathetic blockade and good analgesia.10 The contra-indications to regional analgesia are shown in Table 3. Although there is little supporting data, most anaesthetists would site an epidural with a platelet count >100  109/L or >80  109/L with a normal coagulation screen.11 Some parturients with cardiac disease require anticoagulation throughout pregnancy, most commonly with low-molecular-weight heparins (LMWHs). Neuraxial anaesthesia can be safely administered to the patient receiving LMWHs provided that certain precautions are met. The American Society of Regional Anaesthesia recommends that in patients receiving prophylactic doses of LMWH, neuraxial blockade should occur at least 10–12 h after the last dose of LMWH, and in patients receiving a treatment dose of LMWH, e.g. enoxaparin 1 mg/kg twice a day, neuraxial block should be delayed for 24 h after the last dose.12 A 10-year review of reported clinical practice suggested the incidence of spinal haematoma to be one in 190 000 epidurals, and the use of therapeutic but not prophylactic LMWHs was found to be a predominant risk factor.13 The decision to perform a regional technique for labour or delivery in a patient receiving LMWH must be made on an individual basis, Role of the anaesthetist in obstetric critical care 921 Table 2. Advantages and disadvantages of different methods of labour analgesia in the critically ill. Method Physical Massage Advantages Disadvantages Harmless Perceived as effective by women Does not attenuate stress response May interfere with required level of monitoring Hydrotherapy (immersion) Harmless Popular Cannot use with other methods Does not attenuate stress response May prevent required level of monitoring Transcutaneous electrical nerve stimulation Non-invasive Gives parturient a sense of control May interfere with monitoring Minimal evidence for efficacy Does not attenuate stress response Acupuncture Some weak evidence for efficacy Drug-free Invasive Needs trained operator Psychological techniques/hypnosis Non-invasive Popular Does not attenuate stress response Gives parturient a sense of control Some evidence for analgesic efficacy Widespread availability Maternal hyperventilation Does not abolish stress response Possible bone marrow suppression Inhalational anaesthetics Improved analgesia compared with Entonox Maternal sedation Continuous maternal monitoring required Possible cardiovascular consequences Does not attenuate stress response Not widely available Parenteral opioids: Pethidine Diamorphine Fentanyl Readily available Alternative when regional technique contra-indicated Sedative not analgesic Does not attenuate stress response Respiratory depression Delayed gastric emptying, (increasing hazards of general anaesthesia) Risk of accumulation with repeat dosing Suppression of stress response Ensures anaesthetic involvement May be topped-up if operative delivery required Sympathetic blockade e hypotension Risk of severe cardiovascular instability if inadvertent intrathecal or intravascular local anaesthetic Not always fully effective Pharmacological Entonox Regional Epidural/combined spinal-epidural 922 F. Plaat and S. Wray Table 3. Contra-indications to regional anaesthesia more likely to occur in the critically ill parturient. Absolute Relative Thrombocytopenia/coagulopathy (pathological or iatrogenic) Increased incidence in the critically ill Severe systemic infection Raised intracranial pressure Hypovolaemia with cardiovascular instability Maternal pyrexia e may be performed after systemic antibiotic therapy commenced Fixed cardiac output states: severe aortic/mitral stenosis, hypertrophic obstructive cardiomyopathy weighing the risk of spinal haematoma against the benefits of regional anaesthesia in each individual case. Anaesthesia for the critically ill parturient Single-shot spinal, combined spinal-epidural (CSE), ‘topped-up’ epidural or general anaesthesia have been used to provide anaesthesia for the ill obstetric patient. The choice of technique depends on the urgency of delivery and maternal state of health. There is a trend towards the wider use of regional anaesthesia in the ill parturient as techniques ensuring greater haemodynamic stability have been developed. The singleshot spinal is the most commonly used technique for caesarean section. This necessitates the use of a relatively high dose of local anaesthetic to ensure that the duration of anaesthesia is adequate (up to 15 mg bupivicaine), which is associated with the rapid onset of a dense sympathetic block, vasodilation and hypotension. It is therefore unsuitable for the parturient with impaired cardiovascular function. CSE has become increasingly popular as the preferred technique in the sick mother, as it has the advantage of a small initial spinal dose with later extension of the block by incremental epidural top-ups. This achieves greater cardiovascular stability, and the epidural catheter allows anaesthesia to be prolonged as necessary and can be used for postoperative analgesia. Cardiac disease Cardiac disease remains the most common cause of maternal mortality in the UK14, and the anaesthetist should be made aware of any patient with significant congenital or acquired cardiac disease early in pregnancy. Estimation of risk in pregnancy and delivery in the parturient with cardiac disease is based upon the New York Heart Association (NYHA) functional classification, shown in Table 4. It should be noted that this classification was developed for the non-obstetric patient. Table 4. New York Heart Association functional classification. Class I II III IV Definition Uncompromised, no functional limitation Slightly compromised, some functional limitation but asymptomatic at rest Moderately compromised, limitation of most activity but asymptomatic at rest Severely compromised, inability to perform any physical activity and symptomatic at rest Role of the anaesthetist in obstetric critical care 923 Women with NYHA Class I or II disease usually present little problem in pregnancy, and the method of delivery is based purely on obstetric indications. However, parturients with NYHA Class III and IV disease have a high risk of deterioration in pregnancy and need detailed, multidisciplinary management. There should be detailed plans for each stage of pregnancy and delivery, which should cover a variable amount of the postpartum period. There should be agreed contingency plans if conditions change or added complications arise, and the mode, time and place of delivery should be considered. The anaesthetic management of NYHA Class III and IV can be very challenging, as alterations in heart rate, preload and afterload may be catastrophic. It is difficult to justify and extremely difficult to perform randomized controlled trials in such a high-risk population, and therefore evidence for management options is based on case series and reports. Women with moderate to severe cardiac disease require careful monitoring that may include both invasive arterial blood pressure monitoring, central venous pressure (CVP) monitoring and serial echocardiography throughout labour and the puerperium. Postpartum monitoring should continue for at least 48 h in a high-dependency or intensive care unit, and patients with pulmonary hypertension should ideally be monitored closely for up to 9 days postnatally, as late deaths in this group are not uncommon.15 In labour, epidural analgesia should be instituted early to minimize the stress response to the pain. A low-dose solution, as described previously, is well tolerated if titrated slowly against analgesic requirements. Labour should be conducted in the full left lateral position to minimize the effects of aortocaval compression. As epidural analgesia is associated with a fall in systemic vascular resistance and afterload, it is particularly beneficial in ischaemic heart conditions and regurgitant valvular lesions, as this improves forward flow and reduces ventricular work. However, a fall in afterload is particularly hazardous in pulmonary hypertension, fixed cardiac output states (e.g. severe aortic and mitral stenosis) and right-to-left cyanotic shunts (e.g. Eisenmenger’s syndrome). It is rare even in those with severe disease to require delivery by caesarean section for non-obstetric reasons, except when there is a significant deterioration in maternal condition (e.g. Marfan’s syndrome with aortic root dilatation). However, prolonged or difficult labour should be avoided. A shortened second stage with an assisted vaginal delivery will reduce or avoid the cardiovascular effects of the Valsalva manoeuvre when pushing. Reduced venous return, resulting from increased intrathoracic pressure during straining, is particularly poorly tolerated by those with a fixed cardiac output. The risk from a potentially severe drop in systemic vascular resistance induced by regional anaesthesia should be balanced against the negatively inotropic effects of general anaesthesia.16 Whichever technique is chosen, close monitoring, meticulous fluid balance and vigilance when changing position is necessary (e.g. turning the patient from full left lateral to wedged supine position or positioning head down for central venous catheterization). The UK registry of high-risk obstetric anaesthesia17, a database set up in 1996, shows that slow incremental epidural anaesthesia and incremental CSE are both popular techniques for parturients with cardiac disease, combining the advantages of reduction of afterload with relative haemodynamic stability.18 The successful use of continuous spinal anaesthesia, with a 28-gauge subarachnoid microcatheter and incremental boluses of 0.5% heavy bupivacaine, was first reported in a parturient with aortic stenosis undergoing caesarean section by Pittard and 924 F. Plaat and S. Wray Vucevic.19 This technique allows close control of block height and provides cardiovascular stability, but remains an unfamiliar technique to many anaesthetists in the UK and CSE is more widely used.20 Incremental epidural anaesthesia provides a gradual onset of anaesthesia with associated cardiovascular stability, and has also been used successfully in patients with cardiac disease.21 Some authorities argue that general anaesthesia remains the anaesthetic of choice in the severely compromised patient, as it bypasses the complication of sympathetic blockade.22 When general anaesthesia is used, the cardiac patient may not tolerate the cardiovascular instability associated with the traditional obstetric rapid sequence induction of anaesthesia. The rapidly administered induction agent causes hypotension, and laryngoscopy and intubation hypertension. A modified ‘cardiac’ anaesthetic utilizing high-dose opiates (e.g. fentanyl, alfentanil, remifentanil) and a small induction dose of thiopentone or etomidate should be performed. As the mother remains vulnerable to acid aspiration, a rapidly acting muscle relaxant is required. Although suxamethonium remains the most widely used agent, the non-depolarizing relaxant rocuronium has been advocated for obstetric patients and may afford greater haemodynamic stability.23 Short-acting opioids have been reported to attenuate the pressor response to intubation successfully in these patients and provide cardiovascular stability.24,25 High-dose opioids may cause significant neonatal respiratory depression, and the paediatrician should be made aware that systemic opiates have been given. Uterotonics. Uterotonics can have deleterious side-effects in the cardiovascularly compromised parturient. Oxytocin is the most haemodynamically upsetting and may be associated with profound tachycardia and hypotension when administered as a 5 U bolus.26,27 This effect is exaggerated by hypovolaemia and in those with compromised cardiac function. If syntocinon is necessary, bolus dosing should be avoided and the dose reduced, although some authorities argue that it should not be used at all in such patients. Ergometrine may cause hypertension and pulmonary vasoconstriction, and must be avoided in pulmonary hypertension. Carboprost (prostaglandin F2a), a potent smooth muscle constrictor, may cause bronchospasm, hypertension and (rarely) pulmonary oedema. The prophylactic placement of a uterine compression suture has been used to reduce the need for oxytocics. Cardiovascular emergencies Massive obstetric haemorrhage Massive obstetric haemorrhage is one of the most common reasons for obstetric patients to require critical care. The recognition and management of haemorrhage and hypovolaemia is fundamental to anaesthetic practice. The classical signs of haemorrhage in a young fit patient are shown in Table 5. However, in the obstetric patient, the clinical manifestations of bleeding develop later due to the altered physiology of pregnancy. Reports of the Confidential Enquiry into Maternal Deaths (CEMACH) repeatedly highlight the need for greater awareness that tachycardia, tachypnoea and signs of peripheral shutdown can signal concealed intra-abdominal bleeding, with hypotension a late sign of imminent catastrophe.14 In some respects, the challenges of the bleeding obstetric patient are more akin to those of haemorrhage in a trauma patient, rather than the patient bleeding in theatre. Traumatic and obstetric haemorrhage often occur outside the operating theatre, in an Role of the anaesthetist in obstetric critical care 925 Table 5. Classification of severity of haemorrhage in a previously fit young patient. Parameter Class I Class II Class III Class IV Blood loss (% blood volume) Blood loss (mL) Pulse rate (beats/min) <15% 15e30% 30e40% >40% <750 (in a 70-kg patient) <100 750e1500 1500e2000 >2000 >100 >120 Blood pressure (mm Hg) Normal Normal Decreased Respiratory rate (beats/min) Urine output (mL/h) Central nervous system/mental status 14e20 20e30 30e40 >140 (thready), may be bradycardic Decreased, may be unrecordable by non-invasive methods >35 >30 Slightly anxious 20e30 Sweating, restless 5e15 Agitated, confused, sweating Negligible Drowsy, confused environment not conducive to management (the delivery room). This necessitates removal of an unstable patient to a different environment. Initially, the woman may not have any monitoring, may have had no or few baseline investigations, and may be surrounded by relatives. The trauma patient may have multiple life-threatening injuries, whereas the obstetric patient may have concurrent obstetric problems that will complicate the management of haemorrhage (e.g. pre-eclampsia). In both the obstetric and trauma patient, haemorrhage may be concealed, torrential and very difficult to estimate. All departments should plan for obstetric haemorrhage in advance, have management protocols in place (Table 6), and practice multidisciplinary major haemorrhage drills regularly. It is helpful to have a designated ‘haemorrhage trolley’ on the delivery suite, where all the required equipment can be accessed rapidly in an emergency. The initial aim of anaesthetic management is rapid resuscitation and restoration of oxygen supplies to the tissues. The anaesthetic options for operative management are determined by maternal condition. If haemorrhage is severe and there is hypovolaemia and possible coagulopathy, regional anaesthesia is contra-indicated. In a patient who is already cardiovascularly compromised, general anaesthesia offers greater stability than regional techniques as sympathetic blockade, which may compound hypotension, is avoided. If the patient has an epidural catheter in situ and is stable with no signs of ongoing bleeding, the epidural may be ‘topped-up’ using cautious, incremental boluses of local anaesthetic and opioid. For elective cases where heavy bleeding is anticipated (e.g. caesarean section for placenta praevia), a regional technique such as CSE may be employed. Evidence suggests that estimated blood loss and fall in haematocrit are reduced when regional anaesthesia is used compared with general anaesthesia.28 A retrospective study by Parekh et al found a significantly reduced blood loss and need for transfusion in women with placenta praevia undergoing caesarean section under regional anaesthesia.29 It is essential that the patient is made aware that conversion to general anaesthesia may be 926 F. Plaat and S. Wray Table 6. Management of massive obstetric haemorrhage. Activate staff - ‘Massive obstetric haemorrhage’ call to hospital switchboard summons: obstetric team, labour ward anaesthetist, labour ward operating department assistant, senior midwife, porters, blood transfusion service staff - Low threshold for calling consultant staff Monitoring Basic non-invasive monitoring should be applied rapidly: - electrocardiogram, NIBP, pulse oximetry - Urinary catheter with hourly urine output monitoring Invasive monitoring as required: - Arterial line if cardiovascularly unstable - Central venous pressure monitoring if ongoing blood loss - Consider cardiac output monitoring, e.g. Doppler Use of near patient testing to guide ongoing management: - Haemacue - Arterial blood gases - Thromboelastography Initial resuscitation - Left lateral position if antenatal - Secure airway if required, otherwise high-flow oxygen 10 L/min via face mask - Insert two large bore (14-gauge) intravenous cannulae, send urgent samples for full blood count and coagulation screen, cross-match at least 6 U blood Fluid management: warm all fluids - Crystalloids: up to 2 L blood loss Colloids (starches or gelatines): if EBL >1500 mL Blood: O negative or group-specific blood while awaiting fully cross-matched blood Clotting products: FFP, cryoprecipitate and platelets, need determined by clinical situation, laboratory results and haematology advice - Use active fluid warming and high-pressure infusing devices - Consider the use of cell salvage and recombinant factor VIIa Stop the bleeding Medical therapies: - Uterotonics (oxytocin, ergometrine, prostaglandins) Surgical therapies: - Ensure no retained products or genital tract trauma - Bimanual compression/B-lynch suture - Uterine packing, e.g. Rusch balloon - Arterial ligation/interventional radiology procedures - Hysterectomy Continuing care - Continuing multidisciplinary input on high-dependency or intensive care unit - Serial haemoglobin, coagulation arterial blood gas studies until corrected NIBP, Non invasive blood pressure measurement; EBL, Estimated blood loss; FFP, Fresh frozen plasma. Role of the anaesthetist in obstetric critical care 927 required. The indications for converting a regional to a general anaesthetic include severe cardiovascular instability, inadequate anaesthesia and maternal anxiety. With the increasingly reported use of emergency interventional radiology procedures for uncontrolled obstetric haemorrhage30,31, the anaesthetist will increasingly face the challenge of safely transferring the cardiovascularly unstable, bleeding patient to the interventional radiology suite. In severe obstetric haemorrhage, invasive monitoring is indicated to allow close observation of haemodynamic parameters and also to facilitate sampling for frequent haemoglobin and coagulation studies, until haematological variables normalize. Adequacy of resuscitation can be assessed by improving cardiovascular variables and urine output, and correcting metabolic acidosis, anaemia and coagulopathy. It is vital that measures are actively taken to prevent and correct hypothermia (warmed fluids, warming blankets), as left untreated this will exacerbate coagulopathy. If the patient remains unstable, requires inotrope or ventilatory support, or is acidotic or coagulopathic with ongoing blood loss, admission to the intensive care unit may become necessary. If the pregnant patient survives the initial insult and is resuscitated, the prognosis is very good although secondary haemorrhage and infection can be late complications. Special attention should be paid to the parturient who refuses blood and blood products. The majority are Jehovah’s Witnesses who are generally well informed and carry a clear ‘advance directive’ with them. Many hospitals have consent forms specifically designed for Jehovah’s Witnesses that include a section for detailing specific exclusions. Pre-operatively, the precise details of what is acceptable or unacceptable must be documented. Although whole blood, packed red cells and plasma are generally unacceptable, most Jehovah’s Witnesses will accept recombinant products (erythropoietin, factor VIIa), antifibrinolytics, acute hypervolaemic haemodilution and intra-operative red cell salvage. Pre-eclampsia The role of the anaesthetist in severe pre-eclampsia and eclampsia is to provide safe, effective analgesia and anaesthesia for delivery, to establish appropriate monitoring, and to manage fluid balance. Severe pre-eclampsia remains an important cause of maternal mortality, particularly from intracerebral haemorrhage.14 There has been a reduction in deaths resulting from pulmonary oedema over the last 20 years in the UK, as successive triennial reports have emphasized the importance of fluid balance and avoiding aggressive fluid therapy to treat oliguria. This is in contrast to the aggressive fluid administration widely used to maximize oxygen delivery in the general intensive care setting. As cerebral haemorrhage is now the leading cause of death in pre-eclampsia, improved (systolic) blood pressure control is still needed. There should be guidelines in place for the management of fluid balance, antihypertensive regimen, seizure prophylaxis and anaesthesia in the pre-eclamptic patient. The patient should be cared for in a high-dependency environment. She requires an appropriate level of monitoring with one-to-one care by an experienced and appropriately trained midwife and input from senior members of a multidisciplinary team. Women with severe pre-eclampsia should have their blood pressure measured every 15 min until it stabilizes and subsequently every 30 min. Readings from automated blood pressure cuffs should be treated with caution in pre-eclampsia as they can under-read. If in doubt, the readings should be verified with a manual mercury sphygmomanometer.32 928 F. Plaat and S. Wray Magnesium sulphate has become the drug of choice for seizure prophylaxis and treatment.33 Labetalol, hydralazine or nifedipine may be used for the acute treatment of severe hypertension.34 Close attention must be paid to fluid balance with a strict fluid input–output chart, and total fluid input restricted to 1 mL/kg/h. Studies have shown poor correlation between CVP and left atrial pressure in severe pre-eclampsia35, particularly at pressures >6 mm Hg. Pulmonary artery flotation catheters may be needed in some cases, although these are associated with maternal morbidity and mortality and careful consideration should be given to their use. Optimum fluid therapy is difficult to achieve since the combination of low colloid oncotic pressure and left ventricular dysfunction in severe pre-eclampsia makes pulmonary oedema a great risk. It is essential that a pre-eclamptic patient is stabilized prior to delivery. The appropriate level of monitoring should be instituted, the results of baseline investigations known, and blood pressure should be brought under control. Low-dose epidural analgesia is advantageous for optimal blood pressure control during labour, and may be administered safely provided that there is no coagulopathy or thrombocytopenia. Coagulation tests may be normal in the presence of abnormal platelet function or low platelet counts. Two monitors that have recently become available as bedside tests to assess platelet and coagulation function are thromboelastography (TEG) and the platelet function analyser (PFA-100 platelet function analyser). TEG is a monitor of whole blood coagulation and the interaction between the coagulation cascade, fibrinogen and platelets, providing information on both the process of clot initiation/formation and clot strength/stability. Studies have found that at platelet counts >54–75  103/L, the thromboelastogram suggests normal function.36,37 The PFA-100 is an in vitro analyser of platelet function. It measures the time it takes for a sample of whole blood to occlude a hole in a membrane coated with platelet agonists at 37  C. One study has suggested that the TEG may be a better tool with which to monitor platelet function in pregnancy.38 It is now recognized that pre-eclamptic patients demonstrate greater haemodynamic stability during regional blockade compared with their healthy counterparts. Spinal anaesthesia has been shown to be safe in the context of preeclampsia.39,40 Regional anaesthesia is the preferred technique as general anaesthesia may be complicated by difficult airway management due to laryngeal oedema, or large fluctuations in blood pressure during induction of anaesthesia and endotracheal intubation. Regional anaesthesia is contra-indicated in the presence of coagulopathy or severe pulmonary oedema. Prior to general anaesthesia, a detailed assessment of the airway is indicated. Signs of upper body oedema, especially facial, are particularly worrying. Awake intubation may be considered to be the safest approach, although nasal intubation could precipitate significant bleeding due to the combination of venous engorgement of the upper airway and disordered coagulation, and should therefore be avoided. The hypertensive pressor response to laryngoscopy and intubation must be obtunded. Failure to do so in the pre-eclamptic or eclamptic patient has been identified as substandard care with fatal consequences.14 Pharmacological therapy includes short-acting opiates (e.g. alfentanil 10 mg/kg), beta-blockade (e.g. esmolol 0.5 mg/kg or labetalol 10–20 mg) and lidocaine (1.5 mg/kg given 5 min prior to intubation). Extubation may also cause an exaggerated cardiovascular response which should be attenuated. Magnesium may prolong the effects of depolarizing and non-depolarizing muscle relaxants. Following intubation, reduced bolus doses of non-depolarizing muscle relaxant should be given and neuromuscular blockade monitored. Role of the anaesthetist in obstetric critical care 929 Following delivery, the parturient should be nursed in a high-dependency setting for up to 48 h. If an epidural catheter had been sited for labour or delivery, it may be left in situ to be used for postoperative analgesia. An alternative analgesic regimen would be regular paracetamol with an opiate. Non-steroidal anti-inflammatory drugs are contraindicated in pre-eclampsia as they may exacerbate renal impairment and platelet dysfunction. Non-haemorrhagic shock: maternal collapse Cardiac arrest is a very uncommon event in pregnancy. The causes of nonhaemorrhagic shock in the obstetric patient (amniotic fluid embolization, pulmonary thromboembolism, sepsis, acute uterine inversion) are rare but a significant cause of maternal death in the developed world.41 Immediate management of the collapsed woman requires a multidisciplinary team. Standard cardiopulmonary resuscitation protocols should be followed, with important modifications for the pregnancy. Aortocaval compression must be minimized by applying left lateral tilt or displacing the uterus manually. Expert obstetric and neonatal assistance should be summoned urgently. Chest compressions should be performed with the hands higher on the chest, to make allowance for the elevated diaphragm in pregnancy, and early intubation is needed due to the risk of pulmonary aspiration. If spontaneous cardiac output does not return, delivery within 5 min by caesarean section may improve maternal outcome by relieving aortocaval compression.42 Respiratory disease Patients with limited respiratory reserve may not be able to cope with the increased physiological demands of pregnancy. The most common respiratory problems seen in pregnant women are asthma and cystic fibrosis. In the second trimester of pregnancy, airway closure may occur within normal tidal ventilation in a significant number of women in the supine position, and towards the end of pregnancy, the enlarging uterus displaces the diaphragm and reduces functional residual capacity by approximately 18% at term.43 At the same time, oxygen consumption increases. The effects of these changes may severely compromise the woman with respiratory disease. Lung function tests should be carried out in early pregnancy to give baseline values, and should be repeated at regular intervals if clinically indicated. Asthma is the most common respiratory condition seen in pregnancy. Asthmatics should be encouraged to deliver vaginally with effective regional analgesia to minimize pain and hyperventilation. During labour, the mother should maintain adequate hydration and receive supplemental oxygen as needed. Regional anaesthesia is the technique of choice in asthmatics if there are no contraindications and the patient is able to lie supine. General anaesthesia is often poorly tolerated, as intubation and mechanical ventilation may exacerbate bronchospasm and cause basal atelectasis and subsequent chest infection. Postoperatively, the asthmatic patient should be nursed in a high-dependency area with humidified oxygen, inhaled bronchodilators, regular monitoring of respiratory function and adequate analgesia to allow chest physiotherapy and early mobilization. The number of parturients with cystic fibrosis is increasing as, due to advances in treatment, life expectancy now extends well into the reproductive years, and the rates of conception have increased dramatically. The main aim of anaesthetic management in these patients is to maintain optimum respiratory function. 930 F. Plaat and S. Wray For vaginal delivery, a carefully controlled low-dose epidural block should be established early in labour. It is important to encourage mobility in these women, as many of them need regular postural drainage. Howell et al described successful vaginal delivery with regional analgesia in women with moderate to severe respiratory failure.44 If operative delivery is indicated, regional anaesthesia is preferred. Effective regional anaesthesia for caesarean section requires blockade to the higher thoracic dermatomes, which may temporarily impair the ability of these patients to cough and clear secretions. In cases where maternal respiratory function deteriorates, general anaesthesia followed by transfer to intensive care may be indicated. Cameron and Skinner described the anaesthetic management of a parturient, delivered by caesarean section, with severe respiratory failure (FEV1 17% of predicted value) and pulmonary hypertension secondary to cystic fibrosis. Invasive arterial monitoring and non-invasive bi-level positive airway pressure (BiPAP) ventilation were established in the operating theatre, before sequential CSE was performed. Postoperatively, the parturient was transferred to intensive care for monitoring and ongoing non-invasive respiratory support. Low-dose epidural infusion was used successfully for postoperative analgesia, enabling compliance with chest physiotherapy.45 In the authors’ opinion, patients with chronic severe respiratory disease are highly experienced at managing their condition themselves, and may bring their own non-invasive ventilator [continuous positive airway pressure (CPAP)/BiPAP] into hospital, to be used as needed. GENERAL PRINCIPLES OF CARDIOVASCULAR AND RESPIRATORY SUPPORT Limits of critical care within the delivery suite The early provision of intensive care management to the sick parturient can minimize progression of the condition to multi-organ failure, and reduce morbidity and mortality. One of the important roles of the anaesthetist on a delivery suite is to identify, as early as possible, when the need to step up to intensive care occurs. The limits to what care can be safely provided on the delivery suite will, to some extent, be determined by local resources. Indications for transfer to intensive care in the obstetric patient are shown in Table 7, and include the need for mechanical ventilation, renal replacement therapy, the presence of multi-organ failure and the need for intensive cardiovascular Table 7. Indications for transfer to the intensive care unit. System Cardiovascular Respiratory Renal Neurological Miscellaneous Indication Inotrope support Pulmonary oedema Mechanical ventilation Airway protection Tracheal toilet Renal replacement therapy Significantly depressed conscious level Multi-organ failure Uncorrected acidosis Hypothermia Role of the anaesthetist in obstetric critical care 931 support. The effects of intensive care management on the fetus should also be considered, with attention being paid to optimizing maternal oxygen carrying capacity, considering the effects of pharmacological agents on uteroplacental blood flow, ensuring adequate maternal nutrition and minimizing radiological investigations. Cardiovascular monitoring on the delivery suite All critically ill patients should have continuous electrocardiogram monitoring, with the CMV5 electrode positions that are best placed to detect arrhythmias and left ventricular ischaemia. The CMV5 lead configuration is obtained by attaching the right arm electrode over the suprasternal notch, the left arm electrode over the apex of the heart (V5 position), and the left leg electrode on the left shoulder to act as the indifferent lead. When using non-invasive blood pressure monitoring, care should be taken to ensure that the correct sized cuff for the patient is used. If the patient is unstable or repeated arterial blood samples are required, an arterial line should be sited to allow invasive arterial pressure monitoring. The radial artery is most commonly cannulated, but the brachial, femoral and dorsalis pedis arteries offer alternative sites. A CVP monitoring line may be placed in the internal jugular, subclavian or femoral veins, and used as a guide to right ventricular filling pressures. Central venous cannulation using a long line via the antecubital fossa avoids some of the risks associated with the central sites, but correct positioning of the catheter is more difficult and many anaesthetists are less experienced in this method. Titrating fluid challenges against CVP should be performed cautiously, due to poor correlation between right- and left-sided filling pressures. Femoral lines are less reliable as a monitoring aid in the obstetric patient due to increased intra-abdominal pressure from the gravid uterus. The use of invasive cardiac output monitoring with pulmonary artery floatation catheters (PAFC) has declined, as concerns grow about the well-documented associated hazards compared with unproven clinical benefits. The risks of insertion and use include pneumothorax, arrhythmias, valve damage, pulmonary artery rupture and pulmonary infarction.46 The PAC-Man study, a randomized controlled trial of 1041 critical care patients in the UK, found no difference in mortality between patients treated with or without a PAFC.47 Similarly, a Cochrane Database Systematic Review on the use of PAFCs for adult patients in intensive care could not demonstrate any difference in mortality or hospital length of stay associated with their use.48 Less-invasive monitoring techniques, including Doppler, intermittent transpulmonary thermodilution (PiCCO) and lithium chloride dilution (LiDCO) which allow trends in cardiac output, filling status and systemic vascular resistance to be monitored, have gained popularity. Doppler provides a means of real-time continuous cardiac output monitoring by measuring the velocity of blood flow in the aorta. The oesophageal Doppler incorporates a Doppler transducer at the tip of a flexible probe, which is inserted into the oesophagus via the oral route. Until recently, insertion of the probe required the patient to be sedated or anaesthetized, but newer models have been designed for use in the awake patient, and non-invasive suprasternal probes have also been developed recently. The suprasternal Doppler measures cardiac output via a handheld stationary probe which directs a continuous ultrasound beam at the ascending or aortic arch. The measurement of cardiac output by Doppler has been validated against thermodilution methods, and has been used to monitor cardiac function successfully in a range of physiological states including pregnancy.49,50 932 F. Plaat and S. Wray Table 8. Causes of respiratory failure in the obstetric patient. Amniotic fluid embolism Pre-eclampsia Pulmonary embolism Aspiration pneumonitis Massive haemorrhage/transfusion Sepsis Pulmonary oedema Severe trauma Peripartum cardiomyopathy Pneumonia Drugs and toxins Asthma Pulse contour cardiac output monitoring is also less invasive than PAFC as it utilizes information from central venous (usually femoral) and arterial lines. This method involves both pulse contour analysis of the arterial trace, and PiCCO or LiDCO to provide a continuous display of cardiac output.51 Respiratory support on the delivery suite The causes and management of respiratory failure in the obstetric patient are similar to those of the non-pregnant patient, with the addition of some specific obstetric causes of acute lung injury (shown in Table 8). The need for mechanical ventilation is very rare in pregnancy, and indications include airway protection, pulmonary toilet, hypoxia and ventilatory failure.52 The aim of ventilatory support is to improve gas exchange, allowing optimal oxygen delivery to the tissues whilst maintaining normocarbia. The choice between noninvasive and invasive ventilation with tracheal intubation depends on the level of consciousness of the patient and the severity of respiratory impairment. Although invasive ventilation can only be managed on the intensive care unit, non-invasive ventilation with CPAP or BiPAP may be managed in the obstetric high-dependency setting, depending on the familiarity of midwifery and nursing staff with the technique. Non-invasive ventilation delivered through a tight-fitting face or nasal mask is advantageous compared with invasive ventilation, as sedation is not required and it carries a reduced risk of developing ventilator-acquired pneumonia. However, there is an increased risk of vomiting and subsequent pulmonary aspiration with the use of non-invasive ventilation in late pregnancy.53 CPAP provides positive airway pressure throughout all phases of spontaneous ventilation. It is an effective treatment for pulmonary oedema and increases functional residual capacity, opening collapsed alveoli. CPAP pressures are usually limited to 5–10 cmH2O. BiPAP administers both positive end expiratory pressure and pressure support during the inspiratory phase. Cycling between inspiratory and expiratory modes may be triggered by patient breaths or can be preset. SUMMARY The anaesthetist is involved in some way in the care of >50% of women who enter the delivery suite, and the role has expanded considerably over the last 20 years. In addition to providing analgesia and anaesthesia for labour and delivery, the anaesthetist plays an integral role in the obstetric team and in managing mothers who become seriously ill. In order to minimize maternal morbidity and mortality, high-risk mothers need to be highlighted early in the antenatal period, and followed through their Role of the anaesthetist in obstetric critical care 933 pregnancy by a multidisciplinary team. The most recent CEMACH report has emphasized the need for more timely recognition, treatment and referral of women who are developing a critical illness.14 Anaesthetists can use their knowledge and skills to educate medical and midwifery staff on how to recognize the sick patient. High-dependency obstetric care should be delivered on or near the delivery suite where possible, to allow mother and baby to stay together, and to avoid the unnecessary transfer of a sick patient around the hospital. Hospitals offering obstetric services should have an intensive care unit on site. Low-dose epidural or CSE offers optimal analgesia for labour, and is safe in parturients with cardiac or respiratory compromise. Regional anaesthesia remains the technique of choice for caesarean section, except in the presence of cardiovascular instability or coagulopathy. Unless specifically indicated, general anaesthesia is avoided due to the risk of difficult intubation and pulmonary aspiration in the obstetric patient. Practice points  obstetric critical care requires a multidisciplinary approach, with communication and co-operation between all specialties involved  the anaesthetist must be involved as early as possible in the planning and management of high-risk pregnancies  due to the specific physiological changes of pregnancy, the care of the sick obstetric patient should involve an obstetric anaesthetist as well as a general intensivist  standard analgesic and anaesthetic techniques for delivery need to be modified in the face of maternal disease Research agenda  further investigation of the haemodynamic consequences of different regional anaesthetic techniques in the parturient with cardiac compromise  establish the safety of regional blockade in parturients with thrombocytopenia or receiving LMWHs REFERENCES 1. Ryan M, Hamilton V, Bowen M & McKenna P. The role of a high-dependency unit in a regional obstetric hospital. Anaesthesia 2000; 55: 1155–1158. 2. Obstetric Anaesthetists Association, Association of Anaesthetists of Great Britain and Ireland. Guidelines for obstetric anaesthetic services. Revised edn. London: Association of Anaesthetists of Great Britain and Ireland, 2005. 3. Cordingley JJ & Rubin AP. A survey of facilities for high risk women in consultant obstetric units. Int J Obstet Anesth 1997; 6: 156–160. 4. Srinivas K, Yadthore S & Collis RE. Obstetric high dependency care: a survey of current practice. Int J Obstet Anesth 2007; 16(S1): 563. 934 F. Plaat and S. Wray *5. Zeeman GG. Obstetric critical care: a blueprint for improved outcomes. Crit Care Med 2006; 34(Suppl. 9): S208–S214. 6. Duggan AB & Katz SG. Combined spinal and epidural anaesthesia for caesarean section in a parturient with severe primary pulmonary hypertension. Anaesth Intensive Care 2003; 31: 565–569. 7. Price LC, Forrest P, Sodhi V et al. Use of vasopressin following caesarean section in idiopathic pulmonary arterial hypertension. Br J Anaesth 2007; 99: 552–555. 8. Bacigalupo G, Riese S, Rosendahl H et al. Quantitative relationships between labor and beta-endorphin and cortisol concentrations in plasma. Decline of the hormone concentrations in the early postpartum period. J Perinat Med 1990; 18: 289–296. 9. Suntharalingam G, Dob D & Yentis SM. Obstetric epidural analgesia in aortic stenosis: a low-dose technique for labour analgesia and instrumental delivery. Int J Obstet Anesth 2001; 10: 129–134. *10. Reynolds F, Sharma SK & Seed PT. Analgesia in labour and fetal acid-base balance: a meta-analysis comparing epidural with systemic opioid analgesia. Br J Obstet Gynaecol 2002; 109: 1344–1353. 11. Kam PCA, Thompson SA & Liew ACS. Thrombocytopenia in the parturient. Anaesthesia 2004; 59: 255–264. 12. Horlocker TT, Wedel DJ, Benzon H et al. Regional anaesthesia in the anticoagulated patient: defining the risks (the second ASRA consensus conference on neuraxial anesthesia and anticoagulation). Reg Anesth Pain Med 2003; 28: 172–197. 13. Wulf H. Epidural anaesthesia and spinal haematoma. Can J Anaesth 1996; 43: 1260–1271. 14. CEMACH. Saving mothers’ lives: reviewing maternal deaths to make motherhood safer – 2003–2005. the seventh report on confidential enquiries into maternal deaths in the United Kingdom. London: CEMACH, 2007. 15. Khan MJ, Bhatt SB & Kryc JJ. Anesthetic considerations for parturients with primary pulmonary hypertension: review of the literature and clinical presentation. Int J Obstet Anesth 1996; 5: 36–42. 16. Dob DP & Yentis SM. Practical management of the parturient with congenital heart disease. Int J Obstet Anesth 2006; 15: 137–144. *17. Lewis NL, Dob DP & Yentis SM. UK registry of high-risk obstetric anaesthesia: arrhythmias, cardiomyopathy, aortic stenosis, transposition of the great arteries and Marfan’s syndrome. Int J Obstet Anesth 2003; 12: 28–34. *18. Gomar C & Errando CL. Neuroaxial anaesthesia in obstetrical patients with cardiac disease. Curr Opin Anaesthesiol 2005; 18: 507–512. 19. Pittard A & Vucevic M. Regional anaesthesia with a subarachnoid microcatheter for caesarean section in a parturient with aortic stenosis. Anaesthesia 1998; 53: 169–173. 20. Hamlyn EL, Douglass CA, Plaat F et al. Low-dose sequential combined spinal-epidural: an anaesthetic technique for caesarean section in parturients with significant cardiac disease. Int J Obstet Anesth 2005; 14: 355–361. 21. Brighouse D. Anaesthesia for caesarean section in patients with aortic stenosis: the case for regional anaesthesia. Anaesthesia 1998; 53: 107–109. 22. Whitfield A & Holdcroft A. Anaesthesia for caesarean section in patients with aortic stenosis: the case for general anaesthesia. Anaesthesia 1998; 53: 109–112. 23. Levy DM. Non-depolarising neuromuscular blockers can be used routinely instead of suxamethonium at induction of general anaesthesia for caesarean section. Int J Obstet Anesth 1999; 8: 266–272. 24. Scott H, Bateman C & Price M. The use of remifentanil in general anaesthesia for caesarean section in a patient with mitral valve disease. Anaesthesia 1998; 53: 691–701. 25. Redfern N, Bower S, Bullock RE et al. Alfentanil for caesarean section complicated by severe aortic stenosis. Br J Anaesth 1987; 59: 1309–1312. 26. Thomas JS, Koh SH & Cooper GM. Haemodynamic effects of oxytocin given as i.v. bolus or infusion on women undergoing caesarean section. Br J Anaesth 2007; 98: 116–119. *27. Pinder AJ, Dresner M, Calow C et al. Haemodynamic changes caused by oxytocin during caesarean section under spinal anaesthesia. Int J Obstet Anesth 2002; 11: 156–159. *28. Afolabi BB, Lesi FE & Merah NA. Regional versus general anaesthesia for caesarean section. Cochrane Database Syst Rev 2006; 18. CD004350. *29. Parekh N, Husaini SW & Russell IF. Caesarean section for placenta praevia: a retrospective study of anaesthetic management. Br J Anaesth 2000; 84: 725–730. Role of the anaesthetist in obstetric critical care 935 30. Sundaram R, Brown AG, Koteeswaran SK et al. Anaesthetic implications of uterine artery embolisation in management of obstetric haemorrhage. Anaesthesia 2006; 61: 248–252. 31. Royal College of Obstetricians and Gynaecologists, Royal College of Radiologists, British Society of Interventional Radiology. The Role of Emergency and Elective Interventional Radiology in Postpartum Haemorrhage. Guideline no. 6. London: Royal College of Obstetricians and Gynaecologists, 2007. 32. Natarajan P, Shennan AH, Penny J et al. Comparison of auscultatory and oscillometric automated blood pressure monitors in the setting of preeclampsia. Am J Obstet Gynecol 1999; 181: 1203–1210. 33. Altman D, Carroli G, Duley L et al. Do women with pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial: a randomised placebo-controlled trial. Lancet 2002; 359: 1877–1890. 34. Royal College of Obstetricians and Gynaecologists. The management of severe pre-eclampsia/eclampsia. Guideline no. 10 (A). London: Royal College of Obstetricians and Gynaecologists, 2006. 35. Young P & Johanson R. Haemodynamic, invasive and echocardiogenic monitoring in the hypertensive parturient. Best Pract Res Clin Obstet Gynaecol 2001; 15: 605–622. *36. Orlikowski CEP, Rocke DA, Murray WB et al. Thromboelastography changes in pre-eclampsia and eclampsia. Br J Anaesth 1996; 77: 157. 37. Sharma SK, Phillip J, Whitten CW et al. Assessment of changes in coagulation in parturients with preeclampsia using thromboelastography. Anesthesiology 1990; 90: 385. 38. Beilin Y, Arnold I & Hossain S. Evaluation of the platelet function analyser (PFA-100) vs. the thromboelastogram (TEG) in the parturient. Int J Obstet Anesth 2006; 15: 7–12. 39. Santos AC & Birnbach DJ. Spinal anaesthesia in the parturient with severe preeclampsia: time for reconsideration. Anesth Analg 2003; 97: 621–622. *40. Aya AG, Mangin R, Vialles N et al. Patients with severe preeclampsia experience less hypotension during spinal anesthesia for elective caesarean delivery than healthy parturients: a prospective cohort comparison. Anesth Analg 2003; 97: 867–872. 41. Thomson AJ & Greer IA. Non-haemorrhagic shock. Baillieres. Best Pract Res Clin Obstet Gynaecol 2000; 14: 19–41. 42. Resuscitation Council (UK). Cardiac arrest in special circumstances. In: Advanced life support. 5th edn. London: Resuscitation Council (UK), 2006, pp. 135–137. 43. Rizk NW, Kalassian KG, Gillian T et al. Obstetric complications in pulmonary and critical care medicine. Chest 1996; 110: 791–809. 44. Howell PR, Kent N & Douglas MJ. Anaesthesia for the parturient with cystic fibrosis. Int J Obstet Anesth 1993; 2: 152–158. 45. Cameron AJD & Skinner TAJ. Management of a parturient with respiratory failure secondary to cystic fibrosis. Anaesthesia 2005; 60: 77–80. 46. Fujitani S & Baldisseri MR. Hemodynamic assessment in a pregnant and peripartum patient. Crit Care Med 2005; 33(Suppl.): S354–S361. 47. Harvey S, Harrison DA, Singer M et al. Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet 2005; 366: 472–477. 48. Harvey S, Young D, Brampton W et al. Pulmonary artery catheters for adult patients in intensive care. Cochrane Database Syst Rev 2006; 3. CD003408. 49. Bray JK, Fernando R, Patel NP et al. Suprasternal Doppler estimation of cardiac output: standard versus sequential combined spinal epidural anesthesia for cesarean delivery. Anesth Analg 2006; 103: 959–964. 50. Gan T. The esophageal Doppler as an alternative to the pulmonary artery catheter. Curr Opin Crit Care 2000; 6: 214–221. 51. Berton C & Cholley B. Equipment review: new techniques for cardiac output measurement – oesophageal Doppler, Fick principle using carbon dioxide, and pulse contour analysis. Crit Care 2002; 6: 216– 221. 52. Jenkins TM, Troiano NH, Graves CR et al. Mechanical ventilation in an obstetric population: characteristics and delivery rates. Am J Obstet Gynecol 2003; 188: 549–552. *53. Lapinsky SE. Cardiopulmonary complications of pregnancy. Crit Care Med 2005; 33: 1616–1622.