 |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
In this stage of labour placental separation and expulsion take place; for the mother the main risks are haemorrhage during or after separation of the placenta, and retention of the placenta. Postpartum haemorrhage is one of the main causes of maternal mortality; the large majority of these cases occur in developing countries (Kwast 1991). The incidence of postpartum haemorrhage and retention of the placenta is increased if predisposing factors are present, such as multiple pregnancy or polyhydramnios, and complicated labour: augmentation of labour, obstructed labour, or vaginal operative delivery (Gilbert et al 1987). Postpartum haemorrhage and placental retention also occur more frequently if these complications were present in the obstetric history of the woman (Doran et al 1955, Hall et al 1987, WHO 1989). To a certain extent therefore it is possible to select during pregnancy and in the course of labour those women with an increased risk of complications in the third stage. But even in low-risk pregnancies and after an uneventful first and second stage of labour serious haemorrhage and/or placental retention may sometimes occur. The management of the third stage may influence the incidence of these complications, and the amount of blood lost. Several measures aiming at the prevention of complications have been proposed, have been tested in randomized trials and are discussed below.
Oxytocics may be given prophylactically at various moments during the third stage. Most often they are administered intramuscularly immediately with the delivery of the anterior shoulder, or after delivery of the infant. The drugs usually given, and investigated in trials, are oxytocin and ergot derivatives like ergometrine, or a combination of the two, syntometrine (Daley 1951, McGinty 1956, Friedman 1957, Newton et al 1961, Howard et al 1964, Hacker and Biggs 1979, Rooney et al 1985, Prendiville et al 1988, Thornton et al 1988, Begley 1990). Both oxytocin and ergot derivatives decrease the estimated postpartum blood loss, but the effect of ergot seems to be somewhat less than the effect of oxytocin. The effect on retention of the placenta is not yet quite clear, although there are some data suggesting that routine oxytocics may increase the risk of retained placenta.
Complications of oxytocics are nausea, vomiting, headache and hypertension postpartum. These complications occur more often with ergot derivatives. Moreover, rare but serious maternal morbidity has been associated with oxytocics, especially with ergometrine: cardiac arrest and intracerebral haemorrhage, myocardial infarction, postpartum eclampsia and pulmonary oedema. Because these events are so rare, randomized trials cannot give useful information about the extent to which they may be attributed to oxytocics. The available evidence suggests that oxytocin is a better choice than ergot derivatives. Moreover, in tropical countries oxytocin is more stable than ergometrine or methylergometrine (Hogerzeil et al 1992, 1994).
Because in many developing countries the administration of oral tablets would be much easier, and the tablets would be more stable than injections under tropical conditions, a randomized study was undertaken to investigate the influence of oral tablets of ergometrine immediately after birth. The outcome was disappointing: compared with a placebo the medication had little demonstrable effect on blood loss after childbirth (De Groot et al 1996).
Controlled cord traction involves traction on the cord, combined with counterpressure upwards on the uterine body by a hand placed immediately above the symphysis pubis. In two controlled trials this procedure has been compared with less active approaches, sometimes entailing fundal pressure (Bonham 1963, Kemp 1971). In the controlled traction groups a lower mean blood loss and shorter third stages were found, but the trials do not provide sufficient data to warrant definite conclusions about the occurrence of postpartum haemorrhage and manual removal of the placenta. In one trial patient discomfort was less if controlled traction was used. However, in 3% the cord was ruptured during controlled cord traction. A rare but serious complication associated with controlled cord traction is inversion of the uterus. Although the association might be with a wrong application of the method, the occurrence of inversion of the uterus still is a matter of concern. The above mentioned trials have apparently gathered data on women in a supine position. The impression of midwives attending deliveries with the woman in the upright position during the second and third stage is that the third stage is shorter and placental separation is easier, although the loss of blood is more than in the supine position. However, apart from blood loss, these aspects have not been investigated in randomized trials. Presumably controlled cord traction as described in the textbooks would be more difficult to perform in the upright position.
The combined effects of oxytocics and controlled cord traction are sometimes summarized by the term “active management of the third stage”, as opposed to expectant or physiological management. Sometimes early clamping of the cord is included too, especially because in controlled cord traction early clamping is mandatory. However, because the main effects of this procedure relate to the newborn we shall deal with that aspect separately.
In the literature active management of the third stage compares favourably with expectant management, mainly because postpartum haemorrhage occurs less often and haemoglobin levels postpartum are higher (Prendiville et al 1988, Harding et al 1989, Begley 1990, Thigalathan et al 1993). The results with respect to the frequency of blood transfusion and manual removal of the placenta are not identical in the two largest trials, in Bristol and Dublin (Prendiville et al 1988, Begley 1990). In both trials active management resulted in more nausea, vomiting and hypertension, probably caused by the use of ergometrine.
Some remarks on these findings may be justified. Postpartum haemorrhage is defined by WHO as blood loss >= 500 ml (WHO 1990). The diagnosis is made by a clinical estimate of blood loss; such an assessment of the amount of blood often causes a significant underestimation. Apparently the definition is influenced by the fact that in large parts of the world 500 ml of blood loss (or even less) is a real threat to the life of many women, mainly because of the high prevalence of severe anaemia. Nevertheless, if meticulously measured, the mean blood loss at vaginal delivery is around 500 ml, and about 5% of women delivering vaginally lose more than 1000 ml of blood (Pritchard et al 1962, Newton 1966, De Leeuw et al 1968, Letsky 1991). In the Bristol trial (Prendiville et al 1988) 18% of the group of women with a physiological management of the third stage had blood loss >= 500 ml, and only 3% lost > 1000 ml.
In a healthy population (as is the case in most developed countries) postpartum blood loss up to 1000 ml may be considered as physiological, and does not necessitate treatment other than oxytocics. However, in many developing countries other standards may be applied. The 500 ml limit as defined by WHO should be considered an alert line; the action line is then reached when vital functions of the woman are endangered. In healthy women this usually only occurs after blood loss >1000 ml. This distinction is crucial in the light of efforts to minimise unnecessary blood transfusion and its associated risks, including HIV infection.
Definite conclusions about the value of active management of the third stage in healthy low-risk populations cannot yet be drawn. The term “active management” is used for a combination of various interventions with different effects and side-effects. All trials of expectant versus active management were carried out in centres where active management was the normal practice. A trial is needed in a setting where both expectant and active management are normal procedures. The occurrence of serious but rare complications (cardiac complications, eclampsia, inversion of the uterus, etc.) cannot be studied in randomized trials, but might nevertheless be of major importance if and when active management is recommended for large populations. Serious doubts are justified about the routine prophylactic use of ergometrine or a combination of oxytocin and ergometrine, and also about controlled cord traction as a routine procedure.
In conclusion, oxytocin administration immediately after delivery of the anterior shoulder, or after delivery of the infant, seems advantageous, especially in women with increased risk of postpartum haemorrhage or in women endangered by even a small amount of blood loss, for instance women with severe anaemia. Doubts remain about the combination with controlled cord traction, and about the routine application in healthy low-risk women. Recommendation of such a policy would imply that the benefits of this management would offset and even exceed the risks, including potentially rare but serious risks that might become manifest in the future. In our opinion it is too early to recommend this form of active management of the third stage for all normal low-risk deliveries, although we note the earlier recommendations made by WHO (1990, 1994c). If for various reasons active management is employed, a number of questions remain unresolved, particularly regarding the optimal timing of prophylactic oxytocin injections.
The umbilical cord can be clamped immediately after birth or at a later moment, and this may have effects on the mother and the infant (Prendiville and Elbourne 1989). The effects on the mother have been studied in some trials (Dunn et al 1966, Botha 1968, Nelson et al 1980). There was no evidence of a significant effect of the timing of cord clamping on the incidence of postpartum haemorrhage or on feto-maternal transfusion. The effects on the newborn have been studied by observational studies and randomized trials.
There are a number of observations on the effects of the timing of cord clamping on the neonate (Buckels and Usher 1965, Spears et al 1966, Yao et al 1971, Nelson et al 1980). If after birth the infant is placed at the level of the vulva or below that level for three minutes before clamping the cord, this results in a shift of about 80 ml of blood from the placenta to the infant (Yao et al 1971, 1974, Dunn 1985). The erythrocytes in this volume of blood will soon be destroyed by haemolysis, but this provides about 50 mg of iron to the infant’s reserve and reduces the frequency of iron-deficiency anaemia later in infancy (Michaelsen et al 1995, Pisacane 1996). Theoretically this transfusion of blood from the placenta to the infant might cause hypervolaemia, polycythemia and hyperviscosity, and also hyperbilirubinaemia. These effects have been studied in a number of trials (Prendiville and Elbourne 1989). Babies born after early cord clamping have lower haemoglobin values and haematocrits. With respect to neonatal respiratory disturbances there were no significant differences between the two management practices. Neonatal bilirubin levels were lower after early cord clamping, but no clinically relevant differences between the two practices were noticed, and no differences in neonatal morbidity.
Late clamping (or not clamping at all) is the physiological way of treating the cord, and early clamping is an intervention that needs justification. The “transfusion” of blood from the placenta to the infant, if the cord is clamped late, is physiological, and adverse effects of this transfusion are improbable, at least in normal cases. After an abnormal pregnancy or labour, for instance in rhesus sensitization or preterm birth, late clamping may cause complications, but in normal birth there should be a valid reason to interfere with the natural procedure.
If controlled cord traction after oxytocin administration is practised, as is the case in many obstetric departments worldwide, early or relatively early clamping of the cord is mandatory. However, where late clamping is taught and practised, i.e. after the pulsations of the cord have ceased, usually after about 3-4 minutes, adverse effects have not been recorded. In addition, recent research supports late clamping, because it may prevent iron deficiency anaemia in childhood, which might be of special importance in developing countries (Michaelsen et al 1995, Pisacane 1996). Although at present there is insufficient evidence on which to decide between early and late clamping, this issue clearly deserves more attention.
Directly after birth there should be attention to the condition of the newborn. Such attention is an integral part of care in normal birth, and the World Health Organization stresses the importance of a unified approach to care of the mother and the baby (WHO 1994c). Immediate care involves ensuring that the airway is clear, taking measures to maintain body temperature, clamping and cutting the cord and putting the baby to the breast as early as possible. Each of these elements has been the object of considerable research and debate, but the present Technical Working Group for Normal Birth has the advantage of being able to refer to the work and recommendations of the Technical Working Group on Essential Care of the Newborn (WHO 1996). In the present report only a few aspects of the immediate care of the newborn will be briefly mentioned.
· Immediately after the birth the baby has to be dried with warm towels or cloths, while being placed on the mother’s abdomen or in her arms. The baby’s condition is assessed and the existence of a clear airway is ensured (if necessary) simultaneously. Maintaining the body temperature of the baby is important; newborn babies exposed to cold delivery rooms may experience marked drops in body temperature, and concurrent metabolic problems. A fall in infant temperature can be reduced by skin-to-skin contact between baby and mother.· Early skin-to skin contact between mother and baby is important for several other reasons. Psychologically it stimulates mother and baby to get acquainted with each other. After birth babies are colonized by bacteria; it is advantageous that they come into contact with their mothers’ skin bacteria, and that they are not colonized by bacteria from caregivers or from a hospital. All these advantages are difficult to prove, but nevertheless they seem plausible. Early suckling/breast-feeding should be encouraged, within the first hour after birth (WHO/UNICEF 1989). The influence of nipple stimulation by the baby on uterine contractions and postpartum blood loss should be investigated. One randomized study has been performed (Bullough et al 1989), but only with traditional birth attendants. The influence of early suckling on blood loss could not be established. However, a study with professional birth attendants is needed.
· Cutting the cord should take place with sterile instruments, either disposable, for instance from the clean delivery kit, or thoroughly decontaminated by sterilization. This is of utmost importance for the prevention of infections.
The placenta should be examined carefully to detect abnormalities (infarcts, haematomas, abnormal insertion of the umbilical cord), but above all to ensure that it is complete. If there is a suspicion that part of the placenta is missing, preparations should be made to explore the uterine cavity. If part of the membranes are missing exploration of the uterus is not necessary.
In some countries it is customary for birth attendants routinely to explore the uterine cavity after every delivery, “uterine revision”. There is not the slightest evidence that such policy is useful; on the contrary, it can cause infection or mechanical trauma or even shock. The same holds true for another practice, the “lavage of the uterus”, the rinsing out or douching of the uterine cavity after delivery.
The mother should be observed carefully during the first hour postpartum. The most important observations include the amount of blood lost, and uterine fundal height: if the uterus contracts insufficiently blood may accumulate in the uterine cavity. If the blood loss is abnormal and the uterus is contracting poorly, gentle abdominal massage of the uterus can be helpful. It is essential to ensure that uterine contraction is not inhibited by the presence of a full bladder. Abnormal blood loss, estimated more than 500 ml, should be treated with oxytocics: ergometrine or oxytocin intramuscularly. The condition of the mother is also important: blood pressure, pulse and temperature, and general well-being should be assessed.
 |
 |
