A uterus stops its own bleeding by squeezing shut. When it does not, a woman can lose a litre before anyone is certain she is bleeding

Childbirth ends with a wound that most people never think about. When the placenta separates, it leaves behind a raw area on the inner wall of the uterus, and that area was until moments earlier supplied by arteries wide enough to carry a substantial share of the mother's cardiac output. An open wound of that kind, anywhere else in the body, would bleed catastrophically.

It does not, because of a mechanism that is unusual in human physiology. The uterus does not rely on clotting to close those vessels. It squeezes them shut. The muscle fibres of the uterine wall are arranged in interlacing bundles that run around and between the blood vessels, so when the whole muscle contracts down after delivery it physically pinches the arteries closed. Obstetric texts call them living ligatures, which is exactly right.

Everything about postpartum haemorrhage follows from that single fact. If the uterus contracts, bleeding stops almost regardless of clotting. If it does not contract, no amount of clotting will compensate, because the vessels are too large. This is why uterine atony — a uterus that stays soft — causes the large majority of cases, and why the first action any midwife takes is to feel the abdomen and rub the uterus to make it contract.

Four things to work through, in order

Clinicians remember the causes as four words, and the order reflects how common they are.

Tone is atony, the soft uterus, and it accounts for most haemorrhage. A uterus that has been overstretched by a large baby, twins or excess fluid, or exhausted by a very long labour, or relaxed by certain drugs, may simply fail to clamp down.

Trauma means tears — of the cervix, vagina or perineum — or, rarely, uterine rupture. Here the uterus is contracting properly and blood is coming from somewhere else, which is why a firm uterus with continued bleeding sends the examination straight to looking for a laceration.

Tissue means retained placenta or fragments of it. The uterus cannot contract properly around something still inside it, so the atony is secondary and the fix is removal.

Thrombin covers coagulation disorders — pre-existing, or acquired during a catastrophic bleed, since massive haemorrhage consumes clotting factors and produces a coagulopathy that then worsens the haemorrhage. That loop is why speed matters so much.

Working through those four in sequence is the whole of the initial management, and none of it requires equipment beyond hands, a uterotonic drug and a light.

Prevention is a routine, not a rescue

The single most effective measure is not a treatment at all. It is giving a drug that makes the uterus contract, to every woman, immediately after every birth, before any bleeding has occurred.

A uterotonic given routinely in the third stage of labour — oxytocin is the standard — along with controlled traction on the cord and uterine massage, substantially reduces the incidence of haemorrhage. This is prophylaxis for an event that has not happened and will not happen in most cases, and it is one of the highest-value routine interventions in all of obstetrics.

It has a practical weakness that matters enormously in hot countries: oxytocin degrades when it is not kept cold, and a warm ampoule may be a placebo. This is why the World Health Organization's recommendation of heat-stable carbetocin, which does not require refrigeration, was significant rather than merely incremental — it removes the cold chain from the critical path. Where neither is available, misoprostol tablets are a recognised fallback, less effective but stable and easy to store.

The real problem is knowing that it is happening

Here is where the subject becomes genuinely counter-intuitive. The drugs work. The procedures work. What fails most often is recognition.

Blood loss at delivery is, in most places, estimated by looking at it. That estimate is unreliable in a specific and dangerous direction: visual assessment systematically underestimates large volumes. Blood soaks into linen, mixes with amniotic fluid, pools under the woman and goes onto the floor. A loss that is called moderate can be a litre. And because a healthy woman compensates well at first — maintaining her blood pressure while quietly losing volume — the vital signs can look reassuring until they collapse suddenly.

The fix is almost embarrassingly simple: measure instead of estimating. A calibrated collection drape, a plastic sheet with a graduated pouch that hangs below the woman and collects blood, turns a guess into a number.

A large multi-country trial reported in 2023 tested exactly this — early detection with a calibrated drape, combined with a bundle of treatments delivered together rather than one after another the moment the threshold was crossed: uterine massage, a uterotonic drug, tranexamic acid, intravenous fluids, examination to find the source, and escalation. It reported a very large reduction in severe bleeding outcomes. The intervention contained no new drug and no new device beyond a plastic sheet. What it changed was the timing of the alarm and the habit of doing everything at once rather than sequentially.

A woman rarely dies of postpartum haemorrhage because the right treatment did not exist in the building. She dies because the bleeding was called moderate, and then treatment happened one step at a time.

The three-hour drug

Tranexamic acid deserves its own mention because it is cheap, widely available, and has a property most drugs do not: a hard time window.

It works by inhibiting the breakdown of clots rather than by promoting clotting, which makes it useful precisely when a woman is bleeding and consuming her clotting capacity. A large international trial found it reduced deaths due to bleeding when given within three hours of onset — and found no benefit when given later. That is a clinically unusual finding and an operationally demanding one, because it means the value of the drug is destroyed by exactly the delays that characterise emergency obstetric care: recognising late, referring late, arriving late.

Alongside it sit two interventions designed for the gap between a bleeding woman and an operating theatre. Uterine balloon tamponade inflates a balloon inside the uterine cavity to press on the bleeding surface from within; the low-cost version, a condom tied over a catheter and filled with saline, was developed in South Asia and is used across the region because it costs almost nothing and can be assembled from ward supplies. The non-pneumatic anti-shock garment, a set of compressive neoprene segments, redirects blood from the lower body to the vital organs and buys time during transport. Neither is a cure. Both convert a woman who would have died in transit into one who arrives alive.

Anaemia decides how much bleeding is survivable

There is a reason maternal haemorrhage is far deadlier in some countries than others even when the same volume of blood is lost, and it has nothing to do with obstetric skill.

A woman who begins labour with a healthy haemoglobin has reserve. A woman who begins severely anaemic does not: the same blood loss takes her much closer to the point where her tissues stop receiving enough oxygen, and she decompensates faster and recovers worse. In a country where anaemia in women of reproductive age is extremely common, every case of haemorrhage starts from a worse position.

This makes antenatal anaemia treatment a haemorrhage intervention, which is not how it is usually classified. It also means that the same clinical event — a litre of blood lost, managed competently — can be a difficult afternoon in one setting and a death in another.

What has changed, and what the frontier now is

India's maternal mortality ratio has fallen substantially over the past two decades, and the proportion of births taking place in health facilities is now very high. Those two facts are connected, and they mean the binding constraint has moved.

When most births happened at home, the problem was getting women to a facility. Now that most births happen in facilities, the problem is what happens inside them: whether a uterotonic was given routinely and was still potent, whether blood loss was measured or eyeballed, whether the bundle was delivered in minutes, whether blood is available, whether the referral pathway from a small facility to one with a theatre works at night. These are questions about systems and nursing practice rather than about the availability of medicines, and they are where the remaining deaths sit.

Why it matters for students and researchers

Postpartum haemorrhage is an unusually clear demonstration that outcome improvements often come from measurement and process rather than from new therapeutics. The drugs have been available for decades. The recent gains have come from noticing sooner and acting simultaneously.

That makes the research agenda accessible and unglamorous. Audit of what actually happens in the third stage on a given labour ward, observed rather than recorded, since the two frequently differ. Whether uterotonics in a given district have been stored within their temperature range, which is measurable and rarely measured. Whether calibrated drapes, introduced without other changes, alter recognition times in Indian facilities. How often tranexamic acid reaches women inside three hours, and what specifically consumes those hours. Whether low-cost tamponade is actually available and staff are actually trained in assembling it. And the drill question — whether obstetric emergency simulation training changes response times durably, or only for a few weeks.

That practice-facing focus on women's healthcare within nursing is the scope of the International Journal of Women's Health Nursing and Practices (ISSN 2584-170X), a peer-reviewed hybrid open-access journal launched in 2015. For nursing and midwifery students, this is a subject where the most valuable contribution is rarely a new idea — it is documenting, precisely, where the minutes go.

Frequently asked questions

Why does the uterus not bleed heavily after every birth?

Because the muscle fibres of the uterine wall are arranged around the blood vessels, so when the uterus contracts after delivery it physically compresses the arteries at the placental site closed. Bleeding is stopped mechanically rather than primarily by clotting.

What is uterine atony?

A uterus that fails to contract firmly after birth, leaving the placental bed's vessels open. It is the cause of the large majority of postpartum haemorrhage, which is why uterine massage and a drug to make the uterus contract are the first actions taken.

Why is blood loss at delivery usually underestimated?

Because it is normally judged by eye, and visual estimation systematically undercounts large volumes as blood soaks into linen, mixes with amniotic fluid and spills onto the floor. A calibrated collection drape replaces the estimate with a measurement.

What did the 2023 trial change?

It combined early detection using a calibrated drape with delivering a bundle of treatments simultaneously rather than one at a time, and reported a large reduction in severe bleeding outcomes. It introduced no new drug — it changed when the alarm was raised and how treatment was sequenced.

Why must tranexamic acid be given within three hours?

A large international trial found it reduced deaths from bleeding when given within three hours of onset and showed no benefit afterwards. The delays typical of emergency obstetric care — late recognition, late referral, late arrival — can therefore remove the drug's value entirely.

Why does anaemia make haemorrhage more dangerous?

Because an anaemic woman has less reserve. The same volume of blood lost brings her much closer to inadequate oxygen delivery to her tissues, so she deteriorates faster and recovers less well, which is why antenatal anaemia treatment is effectively a haemorrhage intervention.