Perhaps you have just done a vacuum extraction, and are just taking your gloves off, when there is an ominous splashing of blood into the bucket. Or, a midwife calls you in the middle of the night to say that a patient has had a severe postpartum haemorrhage. What are you going to do? A PPH can often be prevented, and can almost always be treated. Here is its management in hospital, which supplements that in Primary Mother Care.
PPH is caused by: (1) Bleeding from the placental site because the uterus has failed to contract[md]much the most important cause. (2) Tears of the genital tract[md]rupture of the uterus, cervical tears, tears of the upper vagina, and vulval tears, especially near the urethra and clitoris. (3) Occasionally by a clotting defect, especially DIC (disseminated intravascular coagulation), which produces a fibrinogen deficiency.
Aim to resuscitate the patient, to stop the bleeding, and to monitor her carefully. Bleeding most often occurs from the placental site, so your first objective must be an empty well-contracted uterus with the placenta out.
Obstetricians differ in what they do for the few patients who continue to bleed from a contracted uterus with the placenta out, who have no obviously suturable tear or bleeding vessel to tie, and no clotting defect. Some pack the uterus, some stitch quite minor tears (the parturient cervix is normally ragged so they may be stitching the normal), and some do nothing except transfuse. Of those who pack, some explore, inspect, and suture the uterus first, and only pack if they find no tear worth suturing. Others pack, and only explore if packing fails to control bleeding. We side with those who pack when exploring and suturing have failed.
When you pack, do so on the correct indications, and after all proper steps have been taken. Packing is messy and time-consuming, and needs large quantities of sterile dressings. If there is a steady ooze, blood is scarce, and HIV common, packing may save a mother's life. In theory, packing is undesirable; in practice it is very useful as a near last resort, before tying her uterine or her iliac arteries, or removing her uterus (see ''Stop Press'). It is much less effective in controlling bleeding from her uterus, than from her cervix. Much the best way to do this is to give her oxytocics to make her uterus contract[md]if it will.
DIC is probably the commonest cause of a massive PPH, when the uterus is empty, and is satisfactorily contracted. It is the commonest clotting defect, and is an important and preventable cause of maternal death. It is uncommon after a normal delivery, and is more common after abruption (16.13), an obstructed labour (18.3), or an intrauterine death (16.4). Try to keep two bottles of fibrinogen (one gram) in the refrigerator of your maternity unit. This is the only clotting factor which it is practicable for you to stock. If you cannot get it, or any fresh frozen plasma, you will have to give her fresh blood. To do this, you will find it helpful if all your permanent medical and nursing staff know their blood groups, and can be called upon in an emergency.
Bergstrom Steffan. (1) Modrahalsovard I U-Land. (2) Forlossningsvard I U-Land. Reklam and Katalogtryck Uppsala 1988. IF SHE HAS LOST MORE THAN A LITRE OF BLOOD, OR SHOWS SIGNS OF HYPOVOLAEMIA, REQUEST 2 UNITS OF BLOOD URGENTLY (decide how much she needs when it comes) Fig. 19-9 CONTROLLED CORD TRACTION. As soon as her uterus is contracting firmly from the action of oxytocin or ergometrine, grasp her uterus, push it upwards towards her umbilicus and gently pull on the cord, first downwards and backwards, and then more anteriorly as the cord comes out.
POSTPARTUM HAEMORRHAGE (''PPH') PREVENTING PPH BEFORE LABOUR RISK FACTORS FOR PPH IDENTIFIABLE DURING PREGNANCY. If a mother has a history of any of these, she is more likely to have a PPH and should deliver in hospital: (1) Grand multiparity ([mt]5 children). (2) An antepartum haemorrhage in this pregnancy. (3) A postpartum haemorrhage, or a retained placenta, in a previous pregnancy. (4) Multiple pregnancy or other cause of polyhydramnios. (5) Hypotonic uterine action in a previous pregnancy.
RISK FACTORS FOR PPH IDENTIFIABLE DURING LABOUR. (1) Prolonged labour. (2) General anaesthesia, usually with ether or halothane. (3) A full bladder. (4) ''Fiddling with the uterus' during the third stage. (5) Placenta praevia. (6) Placental abruption, mainly because this causes a clotting defect. (7) A clotting defect, especially DIC. (8) Incomplete expulsion of the placenta.
CAUTION ! (1) A postpartum haemorrhage may occur without there being any risk factors. (2) When you ''rub up a uterus', use the flat of your hand on the fundus. ''Fiddling' is all kinds of pushing, pulling, and rubbing, which cause partial separation of the placenta before the uterus has contracted firmly.
PREVENTING PPH DURING LABOUR Give every mother, especially those with risk factors, an oxytocic drug: (1) Ergometrine with oxytocin (''Syntometrine') 1 ml intramuscularly. Or, (2) 5 units of oxytocin intramuscularly. Or, (3) ergometrine 0.5 mg intramuscularly (usually one ampoule). They will work quicker if you give them intravenously, but there may be nobody around who can do this routinely. Give a mother one of these, as soon as her baby is born[md]and you are sure there is no twin in her uterus. Then deliver her placenta by controlled cord traction. If supplies are short, you may only be able to give an oxytocic drug to ''at risk' mothers.
If she has a risk factor for PPH, and you have sufficient intravenous fluids and drip sets, set up a drip of dextrose in water before she reaches the second stage. When her baby and her placenta have been delivered, add 20 units of oxytocin to the drip (500 ml), and run this in at about 30 drops a minute for at least 3 hours. Also, give her ergometrine as usual. Unfortunately, this is an expensive routine, and you may have to wait until a mother has already lost 500 ml, before you can afford to put up a drip.
CONTROLLED CORD TRACTION. As soon as her uterus is contracting firmly from the action of oxytocin or ergometrine, put your left hand on her abdomen, above her pubic symphysis, and turn your palm towards her head. Grasp her uterus. As soon as it feels hard from the effect of the oxytocic, push it upwards towards her umbilicus (deliver the placenta more by pushing her uterus up than by pulling on the cord). Wind two or three loops of cord round your index finger and gently pull on the cord, first downwards and backwards, and then more anteriorly as the cord comes out.
As soon as the placenta is delivered check to make sure that: (1) it is complete and that no lobes of it have been left behind (see below) and, (2) that there are no obvious tears in her birth canal. Keep her in the labour ward, and monitor her for at least an hour, before returning her to the ward. Check that her uterus is well contracted and note any bleeding.
Opinions differ about the use of controlled cord traction, without the use of an oxytocic drug. Ideally, you should never apply controlled cord traction before the uterus has hardened under the effect of an oxytocic drug, and if you don't have one, you should not use it. In practice, little harm results provided there are signs of placental separation (lengthening of the cord, hardness and mobility of the uterus). Although it is a very valuable procedure, there is a risk, particularly if you do it incorrectly, that you may turn her uterus inside out (inversion of the uterus), see below.
CAUTION ! Don't squeeze her uterus to try to get the placenta out. This is so painful that it may cause shock.
ASSESSMENT AND RESUSCITATION [s7]FOR PPH As soon as you are called to a patient with a PPH, quickly call an assistant: at least 2 people are needed. Assess and, if necessary, resuscitate her vigorously, as you would any other hypovolaemic patient (53.2). What is the state of her peripheral circulation? How much blood has she lost? Is it clotting normally in the receiver used to collect it? It may clot to start with, and then stop clotting later. What has been done so far? Monitor the volume of blood she continues to lose, her pulse and blood pressure, and her urine output.
If she is still bleeding: Is her uterus still contracted? Is the placenta out and complete? Does she have any obvious lacerations of her vulva, vagina or perineum?
If she is not still bleeding, is her uterus well contracted?
CAUTION ! Put someone in charge of her, and make sure that she is that person's sole responsibility, until bleeding has stopped, and her condition is stable. Poor supervision is an important cause of death in PPH.
PPH WITH THE PLACENTA IN Try to make her uterus contract. (1) If you have not given her ergometrine, or, better, ''Syntometrine', give it now. (2) If this fails to stimulate a contraction, gently massage her uterus (''rub up' a contraction). (3) Remove her placenta by controlled cord traction (see above), as soon as her uterus is contracting firmly. It should deliver immediately.
If controlled cord traction fails to deliver her placenta, remove it manually. Before doing a formal manual removal, you may be wise to do a vaginal examination, and see if it has stuck in her cervix, from which you can remove it quite easily.
While preparing to do a manual removal concentrate on: (1) resuscitating her, and (2) keeping her uterus contracted by putting 20 or perhaps 40 units of oxytocin into the bottle (500 ml) of her intravenous drip (not the tubing, it is needed as a continuous infusion). (3) If the oxytocin does not work, gently rub up a contraction.
Fig. 19-10. POSTPARTUM HAEMORRHAGE (PPH). A, bimanual compression of a bleeding uterus between a fist in the patient's vagina and a hand on her abdominal wall. B, manual removal of the placenta. Gently separate it from the wall of her uterus with a [f10]slow [f11]sawing movement with the side of your hand. C, packing the uterus is only occasionally necessary. Its main use is to control bleeding from the cervix. It is much less effective in controlling bleeding from the uterus. Much the best best way to do this, is to give drugs to make it contract. C, kindly contributed by Robert Lange.
MANUAL REMOVAL OF THE PLACENTA can either be fairly easy, or rather difficult. It is usually best done in the the labour ward (which must be equipped for anaesthetic resuscitation, A 3.1) rather than the theatre, which will cause delay and require moving her. You will need stirrups to maintain the lithotomy position and a good light. Before you start, set up a drip of saline or Ringer's lactate, or if she is very collapsed, two drips. If she is already being given an oxytocin drip, stop this just before manual removal to allow her cervix to relax, so that you can get your fingers through it.
Scrub up and gown yourself, then put her into the lithotomy position, and clean her vulva and the protruding cord. Cover her with a lithotomy towel (a towel cut to expose the vulva).
Unless she is very collapsed, she needs analgesia. Give her pethidine 25 to 50 mg and diazepam 10 to 20 mg intravenously. Or, if this is difficult, give her intravenous ketamine (A 8.1). Give them intravenously slowly, into the tubing of the drip, or into a vein. Manual removal without analgesia or an anaesthetic is very uncomfortable, particularly if it turns out to be difficult. Inexpert general anaesthesia, which may be all there is, is unnecessary, and potentially dangerous (18.1a).
Hold the cord in your left hand. Put the tips of the fingers of your right hand together, and introduce it into the upper part of her vagina. If her placenta has stuck in her cervix, grasp it and slowly remove it. Now let go of the cord, and place your left hand on her fundus (over the towel). Prevent her fundus from being pushed up, as you gradually work your way into her uterus with your right hand. Feel for the part of the placenta which has already separated, and push your fingers between it and the wall of her uterus. Gently separate her placenta from the wall of her uterus with a slow sawing movement, with the side of your hand.
CAUTION ! All this time keep your left hand pressing on her fundus, so as to bring her uterus as close to your right hand, as you can. If you don't do this there is a danger you may tear it.
As soon as the placenta has separated, grasp it with your right hand, remove it, and ask your assistant to inspect it. While this is being done, and whether it looks complete or not, explore her uterus for any pieces that may have been left behind, and remove them. Only now remove your right hand from her uterus. Finally, give her a further dose of intravenous ''Syntometrine', or ergometrine (0.5 mg), and wait for her uterus to contract. As it begins to do this, remove your hand. As you do so, check that the lower segment is intact.
Before you finish make sure that there are no other sites of bleeding; so explore her uterus as described below.
Inspect her placenta to see if: (1) a piece of it has been left inside, or (2) a vessel is running off one edge of it. This may lead to an extra lobe which has been left inside. If either of these things have happened, the missing piece of placenta must be removed.
If she continues to bleed, apply BIMANUAL COMPRESSION (A, Fig. 19-10). Put your right hand into her upper vagina. Put your left hand on her abdomen, and use it to push her fundus down onto your right hand. Press for at least 5 minutes, and then review the situation. If she continues to bleed, you are now in the situation of ''PPH with the placenta out', so see below.
Continue the oxytocin drip. Add 20 units of oxytocin to the intravenous fluid (500 ml), and run it in at a rate that will keep her uterus contracted. Continue the drip for at least 12 hours, using more intravenous fluid and oxytocin as necessary. Monitor her carefully. Some obstetricians would also give her an antibiotic. Keep her in hospital for at least 5 days, because of the increased risk of puerperal sepsis, particularly endometritis. A few days later check her haemoglobin.
PPH WITH THE PLACENTA OUT Failure of the uterus to contract is the most important cause, so aim for an empty, well-contracted uterus.
Feel her fundus. It should be hard and round, and below her umbilicus. If it is soft and difficult to feel, it may be relaxing. Rub it up to make it contract. This may expel some blood and clots. If her bladder is full, catheterize it. Give her ergometrine 0.5 mg, or ''Syntometrine' 1 ml, intravenously or intramuscularly (if she has not already had it).
Resuscitate her. Ideally, put up two drips of saline or Ringer's lactate (in practice you may have to use a single drip). To the first add 20 units of oxytocin to the intravenous fluid (500 ml). Run it in fast, until her uterus contracts well. Then slow it to 40 drops a minute. Continue this drip for two hours afterwards.
Use the second drip to replace the blood she has lost. Give her a plasma substitute (dextran), or blood. If her blood pressure falls below 80 mm systolic (90 mm is the usual value, but you will probably be worried about HIV), run it in rapidly. As soon as her blood pressure reaches 90 mm systolic, slow it to 40 drops a minute.
Inspect her placenta for missing pieces with great care, if you have not already done so. If a piece is retained it will have to be removed. If there are any obvious perineal tears, suture them.
If bleeding stops, continue to monitor her, to resuscitate her if necessary, and to give her intravenous oxytocin.
If she continues to bleed with an empty uterus (5% chance), note the following and take the appropriate action:
(A) Is her uterus still poorly contracted, despite the oxytocin? If so, increase the rate of infusion. If this fails, she may have a piece of placenta remaining inside, or, much less commonly, a ruptured uterus. So explore her uterus (see below), if you have not already done so.
(B) Does the blood coming from her uterus clot normally? If it fails to clot, she probably has a clotting defect (see below).
(C) Does her uterus remain well contracted, but she bleeds in spite of it? If so, explore her genital tract for tears, from her fundus to her clitoris. If you find tears, suture them. If you don't find any tears (and her blood clots), pack her uterus and vagina. If it does not clot, see below.
Fig. 19-11 REPAIRING A CERVICAL TEAR. A, search all round the patient's cervix with ring forceps, until you find the tear. B, a longer tear being sutured. C, if midwives cannot control bleeding they are asked to apply ring forceps, tie the patient's legs together, and refer her to you like this.
EXPLORATION [s7]FOR PPH INDICATIONS. (1) As a normal part of any manual removal (see above). (2) A mother who continues to bleed with the placenta out. Also see below on the indications for packing.
METHOD. Scrub up and put on gloves. Towel her, as for manual removal. Catheterize her. Give her intravenous analgesia (see above). Put her into the lithotomy position, get a good light, and find a Sims' speculum, and an assistant to help hold it. Wipe out the blood in her vagina with cotton wool swabs. Look at its walls. Check that her vaginal wall, and her perineal and vulval skin are intact. To inspect her cervix, use two swab- holding forceps. Grasp the front lip of her cervix with one of them. Pull her cervix gently down, and look for lacerations on it. If there are no lacerations in that bit of cervix, use the second forceps to pull down the next bit of cervix, and look at that. Go right round her cervix in this way, looking at every part, as in Fig. 19-11. Then put your hand into her uterus and carefully feel its front, sides, back, and fundus. Feel for a rupture of her uterus (18.17), and for any pieces of adherent placenta.
If she has lacerations of her perineum, vagina, or cervix which are big enough to suture, suture them. Only suture a cervical tear, if it is causing arterial bleeding. A venous ooze is not a sufficient indication for suturing.
If she is bleeding from multiple small tears rather than one large one which you can easily stitch, or there is a steady ooze, pack her uterus and vagina as described below.
If a piece of placenta remains inside, scrape it off with your fingers. If you cannot get it all off, she has an abnormally adherent placenta, leave it.
If you find a rupture in her uterus, apply bimanual compression (if the bleeding is severe), until you can get the theatre organized for a laparotomy (18.17).
PACKING THE UTERUS AND VAGINA [s7]FOR PPH INDICATIONS. (1) Continued bleeding, when there is no clotting defect, and no tear in the upper vagina, cervix, or uterus, which is large enough to repair surgically, and when other methods to control bleeding, particularly the adequate use of oxytocic drugs, have failed. (2) Continued bleeding after a clotting defect has been corrected, or when you are unable to correct it. Note: one contributor packs before exploring and only explores when packing has failed (and the blood clots), see above.
METHOD. Scrub up and glove yourself. Put her into the lithotomy position. Pack her uterus and vagina with a wide roll of sterile gauze, or laparotomy pads, or failing these, maternity pads, which are less satisfactory, because they may get lost inside. Start by packing her fundus and work downwards. Use ring forceps to push lengths of gauze through her vagina into her uterus, until both are firmly packed down to her perineum. Pack tightly to press on her cervix from below. The pack should fill her uterus. However, if both her cervix and her uterus are well contracted, you may not be able to pack her uterus completely. If so, a well packed vagina may press adequately on a bleeding cervix.
CAUTION ! (1) Be sure to pack her whole genital tract, from her fundus to her introitus if you possibly can, for which you will need large quantities of gauze. (2) Don't only pack her vagina, because she will bleed above the pack and her uterus will fill with blood, the only sign of which may be increasing shock. (3) If you use maternity pads or separate pieces of gauze, you must tie them together, or they will get lost.
When you have packed her uterus, she will have difficulty in passing urine, so pass a Foley's catheter, and connect this to a bag.
If the pack controls bleeding, continue to monitor her and to give her intravenous fluid or blood as necessary. Remove the pack at 24 to 48 hours, preferably at 24 hours.
DIFFICULTIES [s7]WITH PPH If a patient is BLEEDING SEVERELY and there is going to be some delay before you can treat her, compress her aorta. Stand on her left and feel for her left femoral pulse with your left hand. Clench your right fist and with your index finger level with her umbilicus and your knuckles in the line of her spine, press gently and firmly through her abdominal wall so as to compress her aorta against her spine. You will feel it pulsating. Press so that you no longer feel any pulsations and obliterate her femoral pulse. If necessary, this method can be kept up for hours, while she is referred or while preparations for surgery are being made, changing hands and workers as required. If her legs become numb, allow a little blood to flow through them. A method described by Staffan Bergstr[um]om (see above).
If you CANNOT GET YOUR WHOLE HAND THROUGH HER CERVIX TO DO A MANUAL REMOVAL (not uncommon if she has been given a lot of ergometrine shortly before the manual removal is done, or there has been a long delay), you are in difficulty. Avoid this problem, if you can, by using intravenous oxytocin in the drip, rather than ergometrine, and by discontinuing the drip just before manual removal. Try to get one or two fingers through her cervix, and push her fundus well down with your other hand. Usually, her cervix relaxes gradually so that, if you are slow and gentle, you can put your whole hand into her uterus.
If her placenta seems abnormally adherent to her uterus (PLACENTA ACCRETA or increta), remove what you safely can piecemeal, without perforating her uterus, and leave the rest. If her uterus does not contract well, she will not bleed from these areas, but only from the separated ones. The placenta which you have to leave will be slowly absorbed. She is at serious risk from sepsis and secondary postpartum haemorrhage. Continue the oxytocin drip for 48 hours, then stop if she is satisfactory. Give her antibiotics (chloramphenicol and metronidazole, 2.9). Monitor her carefully, and keep her in hospital for 12 to 14 days.
If YOU PUT YOUR FINGERS THROUGH HER UTERINE WALL as you remove the placenta (easily done, but this should be rare if you do the procedure properly), do a laparotomy and inspect the tear. If it is a minor one, you may be able to repair it. If it is a large tear, repair it, and if bleeding is not controlled, tie her uterine (see ''Stop Press') or her internal iliac arteries. If you don't think it is safe for her to labour again, and her relatives agree, tie her tubes. A hysterectomy is seldom necessary.
If her blood FAILS TO CLOT in the receiver as it comes from her vagina, she probably has DIC (DISSEMINATED INTRAVASCULAR COAGULATION). If necessary, you can confirm this with a bedside clotting test (16.13), but don't let this delay you; control is urgent.
d Give her 2 g of fibrinogen by rapid intravenous infusion. Give her 2, 4, or 6 units of blood with 10 ml of 10% calcium gluconate after the third bottle. Give her another gram of fibrinogen 15 to 30 minutes later, if necessary. If her problem is DIC causing afibrinogenaemia, this should be enough. If you don't have fibrinogen, give her fresh whole blood. Her clotting defect will probably correct itself within 12 hours of delivery of the placenta, so if you can only keep her alive during this period, she will probably live.
If she CONTINUES TO BLEED FROM AN EMPTY UTERUS, DESPITE ALL THE ABOVE MEASURES, try oxytocin 40 units to 500 ml of fluid in a fast running drip and repeated doses of ergometrine 0.5 mg intravenously. Try prostaglandins if you have them. If this fails, tie her internal iliac arteries (3.5), and only if all these measures fail (rare) resort to hysterectomy (20.12). She may have a small rupture of her uterus, which you can only diagnose at laparotomy. Some contributors consider hysterectomy easier than tying the iliac arteries, particularly under inadequate anaesthesia.
If her uterus TURNS INSIDE OUT as her placenta is delivered (rare), she has INVERSION OF HER UTERUS. This may
happen spontaneously, or as a complication of controlled cord traction, particularly in elderly multips. Untreated, she can easily die. Immediately push it back. If you can return it immediately, it should go back easily. If there is any delay, she may become shocked, and replacing it will be much more difficult. Wash her prolapsed uterus with warm saline, give her an antibiotic, resuscitate her, give her a general anaesthetic, and put her into the lithotomy position. There are two methods. (1) Use an enema nozzle and a douche can of warm saline suspended a metre above her. Wash out her vagina with fluid, insert the nozzle, and close her vagina with you left forearm. The hydrostatic pressure of saline will slowly return her fundus over 15[nd]30 minutes. Replace it slowly and manipulate it as little as possible. Check that reduction is adequate. (2) Slowly and gently replace it manually, her fundus last. See also Primary Mother Care.
If she presents after several weeks with CHRONIC INVERSION (rare), do a laparotomy. You will probably find that, whereas her uterus is protruding a considerable distance from her vulva, internally it seems to be inverted from her lower segment, which is much congested. Her tubes may enter pits on either side of her evaginated uterus and be attached at their bottoms. Isolate her bladder from the lower part of her uterus and divide its rolled-over rim where it is inverted and constricted. Alternatively, pull it up with a volsellum and incise the posterior rim of the depression in her uterus through both thicknesses of its inverted wall (Haultain's operation, 19- 12). This should allow you to withdraw her fundus from inside, aided if necessary by a finger passed through the incision into her vagina. Repair the wound you have made in her uterus in 2 layers.
CAUTION ! (1) Inversion of the uterus is much less common than prolapse of the swollen cervix through the vulva. You can easily push this back and it seldom recurs. (2) See also prolapsed fibroids (20-4).
Fig. 19-12 INVERSION OF THE UTERUS. A, B, and C, increasing degrees of inversion. If this happens spontaneously (rare), or as a complication of controlled cord traction, immediately push it back. If there is any delay, replacing it will be much more difficult. D, Haultain's operation for chronic inversion. After ''Bonney's Gynaecological Surgery', Fig. 431. Bailli[gr]ere Tindall, permission requested.