Chapter 56 General obstetric emergencies in the ICU
The intensive care unit (ICU) will receive obstetric patients with the usual range of medical and surgical emergencies, and also provide supportive care for patients who suffer specific obstetric complications. The pattern of admission varies widely among countries with different standards of obstetric care, but complications of hypertensive disorders and haemorrhage usually make up a large proportion of cases,1 while respiratory failure and sepsis are also common.2 The proportion of obstetric patients in most ICUs is still low, and this may lead to relative inexperience in management and teamwork between the intensivist and obstetrician. The latest report of the UK Confidential Enquiries into Maternal Deaths noted that one-third of mothers who died had some intensive care involvement.3 Unfortunately, major problems identified on the ward were:
Maternal outcome is usually favourable because patients are often young and healthy. Scoring systems appear to be valid when the primary problem is medical, but overestimate mortality when the problem is obstetric.1 This is partly because normal pregnant physiological variables are scored as abnormal.
PATHOPHYSIOLOGY
Two important points to recognise in treating obstetric patients are:
Table 56.1 Changes in physiological variables during late pregnancy
| Systolic arterial pressure | -5 mmHg |
| Mean arterial pressure | -5 mmHg |
| Diastolic arterial pressure | -10 mmHg |
| Central venous pressure | No change |
| Pulmonary capillary wedge pressure | No change |
| Heart rate | +15% |
| Stroke volume | +30% |
| Cardiac output | +45% |
| Systemic vascular resistance | -15% |
| Pulmonary vascular resistance | -30% |
| Tidal volume | +40% |
| Respiratory rate | +10% |
| Minute volume | +50% |
| Oxygen consumption | +20% |
| pH | No change |
| PaO2 | +10 mmHg |
| PaCO2 | -10 mmHg |
| HCO3– | -4 mmol/l |
| Total blood volume | +40% |
| Haematocrit | -0.06 |
| Plasma albumin | -5 g/l |
| Oncotic pressure | -3 mmHg |
AIRWAY AND VENTILATION
Several factors may complicate tracheal intubation in pregnancy:
Intensivists must be familiar with the difficult airway algorithm and the use of the laryngeal mask airway.36,37 Avoidance of intubation and the use of non-invasive ventilation may be a good option in selected cases.
Some causes of respiratory failure are modified by pregnancy (e.g. aspiration of gastric contents, viral pneumonia) and some are unique to pregnancy (e.g. amniotic fluid embolism (AFE), pre-eclampsia).5 Pregnant patients are more susceptible to pulmonary oedema because of the increased blood volume and lower oncotic pressure. Mechanical ventilation can be more problematic in the pregnant patient.6 Respiratory alkalosis is normal during pregnancy and fetal gas exchange must be considered when titrating respiratory support. Although the changes in anatomy and lung compliance present no difficulty to the mechanics of ventilation, strategies for managing adult respiratory distress syndrome (ARDS) such as permissive hypercapnia may be more difficult to implement5,7 (see below).
CIRCULATION
ALTERED SIGNS
Haemodynamic support should generally start with good hydration, and assessment should take into account the altered cardiovascular variables in pregnancy. Non-invasive cardiac output monitoring is inaccurate and invasive monitoring with pulmonary artery catheters may be helpful in severe pre-eclampsia, pulmonary oedema and cardiac disease.8 The uterine vascular bed is considered maximally dilated but still responsive to stimuli that cause vasoconstriction, such as circulating catecholamines. Ephedrine has traditionally been the vasoconstrictor used in obstetrics because it was thought to preserve uterine blood flow better than pure alpha-agonists. However, alpha-agonists such as phenylephrine are more effective and not associated with fetal acidosis when used to manage hypotension during caesarean delivery.9 There is no evidence favouring any particular inotrope.
CARDIOPULMONARY RESUSCITATION10
Cardiac arrest is rare in pregnancy and estimated to occur once in every 30 000 deliveries.
Normally, external cardiac massage produces only 30% of cardiac output and this is reduced further if there is vena caval compression. After about 20 weeks’ gestation it becomes increasingly necessary to relieve aortocaval compression during basic life support (BLS). Left lateral tilt decreases the efficiency of closed-chest compression, but a wedge providing an angle of 27° gives significant relief of vena caval obstruction while allowing 80% of the maximal force for chest compression.11 During BLS, aortocaval compression may be minimised by manually displacing the uterus using a wedge, or positioning the pregnant patient’s back on the rescuer’s thighs. Chest compression should be performed with a slightly higher hand position (slightly above centre of sternum). Early tracheal intubation after cricoid pressure will facilitate ventilation and decrease the risk of acid aspiration. During advanced life support (ALS), drugs are given and defibrillation performed according to the normal protocols. Apical placement of the paddle may be difficult because of position and breast enlargement, and adhesive defibrillation pads are preferred.
Fetal or uterine monitors should be removed before defibrillation.
Case reports at advanced gestation indicate that both maternal and fetal survival from cardiac arrest may depend on prompt caesarean delivery to relieve the effects of aortocaval compression. The European Resuscitation Council Guidelines12 and International Liaison Committee on Resuscitation (ILCOR) advisory statement13 suggest that, if there is no immediate response to ALS, perimortem caesarean delivery should be considered. The decision must be made quickly because the surgery should start within 4 minutes of the arrest, with the aim of delivering the infant within 5 minutes of the arrest. There is no need to perform the delivery if the fetus is less than 20 weeks because aortal caval compression is not problematic then.
There is now some experience with somatic support after brain death to allow the fetus to mature.10
TRAUMA
Trauma occurs in approximately 6–7% of all pregnancies but only requires hospital admission in 0.3–0.4% of pregnancies. Trauma is the leading non-obstetric cause of maternal mortality and survivors have a high rate of fetal loss.14 Head injuries and haemorrhagic shock account for most maternal deaths, while placental abruption and maternal death are the most frequent causes of fetal death.15 Most injuries occur as the result of motor vehicle accidents, but other common causes are suicide (usually postpartum), falls and assaults.
Initial resuscitation should follow the normal plan of attention to airway, breathing and circulation.16 Oxygen 100% should be given and cricoid pressure applied during tracheal intubation. Blood volume is increased during pregnancy and hypotension may not be evident until 35% or more of total blood volume is lost. Uterine blood flow is not autoregulated and may be decreased despite normal maternal haemodynamics, so that slight overhydration may be preferred to underhydration. Excessive resuscitation with crystalloids or non-blood colloids may increase the mortality from severe haemorrhage.
Necessary radiological investigations should be performed as indicated because radiation hazard to the fetus is very unlikely, except in the early first trimester, where exposure to more than 50–100 mGy is a cause for concern. A chest X-ray delivers less than 5 mGy to the lungs and very little to the shielded abdomen. The fetal radiation dose from abdominal examinations can range from 1 mGy for a plain film, up to 20–50 mGy for an abdominal pelvic computed tomography (CT) with fluoroscopy.17
Cardiotocographic monitoring is considered essential, but there is wide variation in practice and in the recommended duration of monitoring. Premature labour and placental abruption may not be diagnosed unless regular monitoring is continued for at least 6 hours and even 24 hours if indicated.18 Rh immune globulin 300 µg should be considered for all Rh D-negative within 72 hours of injury. The Kleihauer–Betke test can be used to detect fetal blood in the maternal circulation and give an estimate of the volume of transplacental haemorrhage.
