Postpartum Haemorrhage

Introduction

‘The dangerous efflux is occasioned by everything that hinders the emptied uterus from contracting … in these cases such things must be used as will assist the contractile power of the uterus and hinder the blood from flowing so fast into it and the neighboring vessels.’ William Smellie

Treatise on the Theory and Practice of Midwifery. London: D. Wilson; 1752:402–404 ( Fig 34.1 ).

Although global maternal mortality rates are falling, every hour there are still over 30 maternal deaths worldwide, and of these 10 bleed to death. Of the estimated 80,000 global deaths from postpartum haemorrhage (PPH) in 2015, only 189 occurred in the richest one-fifth of countries, whereas over 40,000 occurred in the poorest fifth. In rich settings the commonest causes are placental abnormalities (praevia, accreta and retained placenta) and surgery, whereas deaths occur from all causes in poorer settings, with women who are isolated, anaemic or with underlying disease at particular risk and unable to cope with even small bleeds. A helpful model to understand the underlying systemic cause of deaths is the ‘Three Delays’ model: delay in seeking care, delay in reaching care, and delay in receiving care. Although these ‘delays’ are most common in the developing world they are not unknown in countries with developed health services. The United Kingdom Confidential Enquiry Into Maternal Deaths continues to emphasize that deaths due to PPH are often associated with treatment that is ‘too little, too late’ or ‘major substandard care’. The importance of care provision can be seen by tracking the UK maternal deaths since 1874, where the greatest falls in PPH deaths occurred between 1874 and 1926, before the arrival of oxytocics or blood transfusion, but during a time that there were improvements in public health and access to services. Concerningly, while PPH rates in low-income settings are slowly improving, those in high-income settings are slowly rising with increased intervention.

This chapter will outline the causes and medical management of PPH. Other aspects, including surgical management, are covered in separate chapters: retained placenta ( Chapter 35 ); uterine inversion ( Chapter 40 ); lower genital tract trauma ( Chapter 39 ); uterine tamponade, uterine compression sutures, pelvic vessel ligation and embolization, and obstetric hysterectomy ( Chapter 27, Chapter 36, Chapter 37, Chapter 38 ).

Definition

Primary PPH is classically defined as bleeding from the genital tract in excess of 500 mL in the first 24 hours after delivery. However, this definition has been criticised as it is neither an appropriate level at which to start treatment, nor is a level that causes maternal morbidity. Furthermore, the use of 500 mL as a clinical trial outcome has led to the use of treatments that reduce minor bleeds, but which may not reduce maternal morbidity or mortality. It is therefore suggested that clinicians continue their current practice of giving PPH treatment as soon as they are unhappy with the blood loss speed or volume. However, clinical trials should use either 1000 mL blood loss or maternal near-miss mortality as the main outcome.

Estimation of blood loss at birth is notoriously inaccurate, with clinicians tending to underestimate blood loss and patients overestimating it. Accuracy can be improved by the weighing of blood loss and swabs (the ‘gravimetric’ method) or collection of the blood in a plastic drape, placed under the mother’s buttocks immediately after birth (the ‘volumetric’ method). Clinical estimation can be improved by training. A useful practical method is to remember that the volume of a male adult hand/fist is approximately 350 mL, as is a heavily soaked ‘abdominal’ swab. A typical female does not develop signs or symptoms of hypovolaemia until she has lost 1500 mL. However, it is important to remember that blood volume is related to body weight. Thus, the small, low-weight woman, particularly if she is anaemic, may tolerate small blood losses quite poorly (see Chapter 29 ).

Secondary PPH is defined as excessive bleeding between 24 hours and 6 weeks postnatally.

The Normal Third Stage of Labour

Before discussing the causes and management of primary PPH, it is necessary to understand the physiology of the third stage of labour, which lasts from delivery of the infant to delivery of the placenta. Although this is the shortest of the three stages of labour, it carries the greatest risk to the mother.

During pregnancy, the myometrial fibres have been stretched considerably to accommodate the enlarged uterus and its contents. When the infant is born, the uterus continues to contract, leading to a dramatic shortening of these elongated fibres. The permanent shortening of the muscle fibres is achieved by retraction, a unique property of uterine muscle which, in contrast to contraction, requires no energy.

Placental separation is caused by the uterine contraction and retraction greatly reducing the size of the placental implantation site. The placenta is thus sheared from the uterine wall: analogous to a postage stamp stuck to the surface of an inflated balloon becoming detached when the balloon is deflated. When the placenta is completely separated from the implantation site, the contractions continue its descent to the lower uterine segment, through the cervix and into the vagina.

Clinical Signs of Placental Separation

The triad of clinical signs associated with placenta separation is as follows:

• 1.

  The uterus will be felt to contract and, as the placenta is sheared off the uterine wall and descends to the lower uterine segment, the uterine fundus changes from a broad and flat discoid shape to a more elevated, narrow and globular shape. This change from the discoid to globular configuration can be quite difficult to appreciate clinically except in the very thin. However, the uterus will be felt to harden as it contracts, rises in the abdomen and becomes ballotable.

• 2.

  A small gush of blood often accompanies separation of the placenta from the uterine wall. This can be unreliable as bleeding may occur with only partial separation of the placenta and, even with complete separation of the placenta from the uterine wall, the blood may be contained behind the membranes and not be clinically apparent.

• 3.

  When the placenta has separated and descends to the lower uterine segment and through the cervix there is cord lengthening (8–15 cm) at the introitus. This is the most reliable sign.

The mechanism of haemostasis at the placental site is one of the physiological and anatomical marvels of nature. The muscle fibres of the myometrium are arranged in a criss-cross pattern and through this lattice-work of muscle fibres the radial and arcuate blood vessels pass to supply the placental bed. When the uterine muscle contracts, this lattice-work of fibres effectively compresses the blood vessels ( Fig. 34.2 ). This myometrial architecture is sometimes appropriately referred to as the ‘living ligatures’ or ‘physiological sutures’ of the uterus. If the placenta is forcibly separated from the uncontracted myometrium, then blood will continue to flow through the radial arteries and out into the uterus. Hence the importance of waiting for signs of separation before cord traction.

Routine Management to Prevent PPH

After the infant has been delivered, delayed cord clamping (after 2 minutes) is advocated to allow maximum transfusion to the fetus. Then the cord is divided and the necessary cord blood samples taken. Put very light tension on the cord to ensure there are no loops free in the vagina and then place the clamp on the cord at the level of the introitus, ensuring that real cord lengthening becomes clinically apparent. One hand cradles and ‘guards the fundus’ so that the changes associated with placental separation can be appreciated or to detect an atonic enlarging uterus filling with blood. When the clinical signs of placental separation are evident, assist delivery of the placenta by controlled cord traction (CCT). The abdominal hand moves to the lower part of the uterus just above the pubic symphysis and gently applies counter-pressure, pushing the uterus upwards and backwards while the other hand exerts steady downward traction on the cord. The distance between the suprapubic hand and the sacral promontory should be such as to prevent the possibility of uterine inversion ( Fig. 34.3 ).

There are essentially two approaches to the routine management of the third stage of labour: physiological and active.

• Physiological management involves observation while awaiting the physiological changes that bring about placental separation. This usually takes 10–20 minutes and is favoured by those who prefer limited intervention in the management of labour. Depending on definition, 95% of women who give birth normally do not bleed excessively and with this method, intervention is reserved for those who do. Secondary prevention is the term given for a form of expectant management in which treatment is given as soon as bleeding is thought to be excessive (i.e. around 350 mL) rather than at 500 mL.

  • Some will encourage suckling immediately after delivery to stimulate physiological oxytocin release. Unfortunately, this physiologically attractive approach is not as effective at reducing PPH when compared with active pharmacological management.

• Active management entails giving an oxytocic drug during or just after delivery of the infant in order to consistently cause the uterine contractions that lead to placental separation and haemostasis. Several randomized controlled trials have shown that it effectively reduces blood loss, need for therapeutic doses of oxytocic drugs, PPH and blood transfusion by 50–70% when compared with expectant management. The evidence and experience with active management is such that this has become the standard of care.

  • Active management of the third stage of labour has evolved over the past half century. The original protocol combined the oxytocic with early cord clamping (ECC) and CCT. With time however, ECC has been shown to be harmful and CCT of no benefit. Both have therefore been dropped from the active management protocol.

The timing of cord clamping has been the subject of considerable interest. Although its use was criticised as far back as 1794, it had become part of the usual culture of modern birthing methods. However, the realization that 20–30% of the newborn’s blood volume was made up of blood that had been transferred from the placenta in the first few minutes after birth led academics to question the practice. Randomized trials now demonstrate that ECC is harmful to the neonate. In babies born at term, ECC leads to increased anaemia and iron deficiency in both high- and low-resource settings. Iron is critical for neurological development, and babies who have undergone ECC have worse fine motor and social functioning at 4 years of age. In premature babies the harm of ECC is clearer, with babies who have been subjected to ECC having higher rates of transfusion, cardiovascular instability and mortality. Just delaying cord clamping by 1 minute in premature babies reduces all-cause mortality by a staggering 30%.

CCT also appears to be of little benefit, except to shorten the third stage of labour slightly.

The choice of oxytocic for routine active management is usually between the cheaper injectable drugs, oxytocin and ergometrine, or a combination of both in the compound Syntometrine (containing oxytocin 5 IU and ergometrine 500 μg). Of these, oxytocin is the cheapest, has the fewest side effects and does not cause retained placenta. It is, however, shorter-acting (15–30 minutes). It is given as a dose of 5–10 units intravenously or by intramuscular injection with delivery of the anterior shoulder or as soon thereafter as feasible. It is more effective when given intravenously, especially in nulliparous or induced women but needs to be given as a slow bolus over 1–5 minutes as rapid administration causes a transient fall in blood pressure of around 20 mmHg. Ergometrine is effective but has more side effects (see below), has a longer duration of action (60–120 minutes) and a slightly higher risk of causing retained placenta. Carbetocin is longer acting, but randomized trials show it to be no more effective than oxytocin. Misoprostol can be taken orally, but is less effective than oxytocin and has side effects of raised temperature and shivering.

Therefore the optimal choice for women at high risk appears to be either Syntometrine or intravenous oxytocin, although both are associated with potentially dangerous side effects. Intramuscular oxytocin or carbetocin are weaker, but also have less side effects and so may be a better option for routine use in low-risk women.

More details about the oxytocics can be found in the appendix to this chapter.

‘The uniform operation of the ergot to restrain uterine haemorrhage … has frequently been prescribed, a little previous to the birth of the child, or immediately after, to the patients who have been accustomed to flow immoderately, at such times, and it has always proved an effectual preventive.’ Oliver Prescott.

A dissertation on the natural history and medical effects of secale cornutum or ergot. Andover: flagg & gould; 1813:14.

‘In patients liable to haemorrhage, immediately after delivery … ergot may be given as a preventive a few minutes before the termination of the labour.’ John Stearns.

Observations on the secale cornutum or ergot, with directions for its use in parturition. Med Rec. 1822;5:90.

Pathophysiology of Postpartum Haemorrhage