Obstetrical Hemorrhage

Obstetrical hemorrhage is a serious complication of childbirth and is responsible for about one of every four maternal deaths. Many of those deaths are preventable.

The clinical management of obstetrical hemorrhage is fundamentally the same as for other types of hemorrhage: that is…stop the bleeding and restore normal circulation. But there are complexities to obstetrical hemorrhage that make it unique among other forms of hemorrhage. The two primary complexities are maternal cardiovascular physiology and the difficulty in accurately quantifying blood loss among these patients.

During pregnancy, the total maternal blood volume increases about 50% to approximately 5 liters. Most of this increase is in plasma, while only some of it is increased red cells. Consequently, there is a so-called physiologic anemia, or dilutional anemia of pregnancy. Because of this differential increase in blood volume, a woman with a hemoglobin of 12.5 at the beginning of pregnancy, might normally have a hemoglobin of 11.5 at the end of pregnancy, although she’ll have 5 liters of blood, instead of 3.5 liters.

Also during pregnancy, her peripheral vascular system becomes more compliant or stretchy, to accommodate this increased blood volume. Her cardiac output normally rises about 50%, maxing out at the beginning of the 3^rd trimester. She normally develops a mild tachycardia, relative to her prepregnancy heart rate. Her systolic blood pressure falls some during the second trimester before rising back to near normal levels during the 3^rd trimester. Her pulse pressure widens during pregnancy.

The point of this explanation of pregnancy physiology is that you understand that a third trimester pregnant woman’s cardiovascular system is different in some important respects than a non-pregnant cardiovascular system. For this reason, it usually doesn’t respond in the same way to hemorrhage that we would expect a non-pregnant person to respond.

Absent a pregnancy, when confronted with progressive hemorrhage, we expect an individual to respond initially with mild tachycardia, followed by increasing respiratory rate. Orthostatic hypotension gives way to a slight fall in systolic pressure, narrowing of the pulse pressure, and ultimately frank hypotension. The urine output falls progressively and the patient’s mental status, initially anxious, becomes confused, combative, disoriented, and ultimately loses consciousness.

In contrast, pregnant women at full term tend to show very few of those signs and symptoms of hemorrhage until they have lost more than 30% of their blood volume. Other than tachycardia, they may not demonstrate the other signs or symptoms of bleeding until they are already significantly hypovolemic.

The second major complexity of obstetrical hemorrhage is the difficulty we all have in accurately measuring blood loss.

If I show a group of experienced obstetricians five surgical sponges, each soaked with varying amounts of blood and ask them to visually estimate the blood loss, I’ll get a wide range of estimates that both overestimate and underestimate the actual amount of blood by very large amounts.

If I ask them to look at a blood spill on the sheets and floor and ask them to estimate the loss, I’ll get equally wide and inaccurate estimates.

Suction devices typically provide an accurate recording of fluid volume, but don’t distinguish between blood and amniotic fluid. Women who bleed into a toilet have blood mixed with water. Women who bleed at home have blood loss that is unseen by healthcare providers.

Despite this difficulty in visual estimation of blood loss, we have some tools and benchmarks that can be very helpful.

The average vaginal delivery involves the loss of about 500 cc of blood, and the average cesarean section loses about 1,000 cc of blood. While experienced obstetricians may have difficulty accurately identifying the exact quantity of blood loss, they will usually have little difficulty identifying the blood loss as average, greater than average, or less than average.

We have accurate scales that can weigh each bloody item, which added together, will give an accurate estimate of blood loss contained within those items. This won’t be helpful if the bloody items are also filled with amniotic fluid, but for any ongoing bleeding or non-contaminated surgical sponges, weighing the underpads and sponges can be very helpful.

We can obtain laboratory measures of blood hemoglobin or hematocrit. Knowing that the woman started with 5 liters of blood, a 50% drop in either hemoglobin or hematocrit suggests a 2500 cc blood loss. Of course, it takes some time for the hemoglobin or hematocrit to equilibrate and this could take a very long time if IV fluids are not being administered. Nonetheless, if you observe a large drop in hemoglobin or crit, you know there has been a substantial blood loss that will not, over the short term, improve without a blood transfusion.

The 5 liter volume can also be used to predict the impact of transfusion on hemoglobin concentration or hematocrit. 5 liters of blood has the same amount of red blood cells as 10 units of packed red blood cells, and normally gives a hemoglobin concentration of a little over 10 gms. So transfusing one unit of packed red blood cells would be expected to increase the hemoglobin concentration by about 1 gram. For those of us who think in hematocrit terms…one unit of packed red blood cells will increase the hematocrit by 3 to 4 points.

Fresh frozen plasma contains all the clotting factors and is usually needed in cases of massive blood loss, since not only are red blood cells lost, but also clotting factors. A general rule of thumb is that for hemorrhage requiring less than 4 units of RBCs, fresh frozen plasma is usually not necessary, but are usually needed for hemorrhage beyond that level.

Cryoprecipitate is an excellent source of fibrinogen which is lost in massive hemorrhage and also in the presence of consumptive coagulopathies, such as is seen in large placental abruptions. 10 units of cryoprecipitate (or 400 ml), will increase fibrinogen by 50 to 100 gm/dl and will usually raise the serum concentration to above 150.

Platelets are given typically whenever the platelet count falls below 50,000 mm3. One pack of platelets will raise the platelet count by 5,000 to 8,000.

5% Albumin can be administered as a temporary method of restoring volume to the vascular tree as its’ oncotic pressure tends to keep fluid inside the vessels. This is in contrast to crystalloid solutions which typically leak out into the surrounding tissues relatively soon after administration. Albumin doesn’t contain any red blood cells, however, so it doesn’t restore oxygen carrying capacity to the blood.

Another temporizing method that is always available is elevating the patient’s legs, either by placing the bed in trendelenburg position, or by lifting the legs up in the air. This maneuver can instantly transfuse 300 to 500 cc of blood that had been pooled in the legs back into the general circulation, improving flow and oxygenation to the vital organs.

None of these temporizing techniques replace the need for blood transfusion, but they can support the circulation long enough for blood to be obtained and transfused.

Transfusions should generally be continued until the urine output is maintained at or just above 30 ml/hour, the mental signs have disappeared, and the patient is able to sit up in bed or ambulate without getting lightheaded. At this point, all of the vital organs are being adequately perfused.

It is best to use type-specific crossmatched blood, but in the presence of massive, acute hemorrhage, type O negative blood can be given initially, while the type-specific blood is being crossmatched. If type O negative blood is not available, but the patient is in shock, type O positive blood can be given, although rH sensitization then becomes an issue for rH negative women. Of course, if the patient doesn’t survive the hemorrhagic shock, rH sensitization will be irrelevant.

During the hemorrhage, some blood tests can be useful, including hemoglobin, hematocrit, platelet count, fibrinogen, PT, PTT, and INR. Because the results of these tests may not be known for while after they are drawn, some experienced obstetricians have found it useful to have a red-top tube drawn, one without any anticoagulant, and taped to the wall. After about 10 minutes, a clot will form in the bottom and clear serum will be on top. This will provide basic information that the blood is able to clot, that there is at the moment a reasonable hematocrit. It also provides an additional native source for cross-matching blood, should there be many, many transfusions.

While the maternal circulation is being restored, it is also very important to stop the bleeding.

Obstetrical hemorrhage can be from any number of sources, but the common ones prior to delivery are placental abruption and placenta previa. In the setting of obstetrical hemorrhage, both will require immediate cesarean section. Should there be lesser amounts of bleeding, then a careful evaluation with ultrasound followed by digital exam can be undertaken, but in the presence of frank hemorrhage, it is usually better to save time and go directly to the operating room.

Following delivery, uterine atony or hypotonus is probably the single most common cause of excessive bleeding. In this case, the uterus is soft and flabby, unable to contract sufficiently to close off the uterine vessels that had been supplying the vascular bed of the placenta. Uterine massage can be effective, but medications, such as oxytocin, methergine, hemabate or cytotec are usually given to encourage uterine contractions. Bimanual compression can be a temporizing expedient.

If the placenta or a large piece of placenta is retained inside the uterus, the uterus may not be able to firmly contract to control the bleeding. For that reason, the expelled placenta should be inspected for missing pieces, and the uterus explored manually in the face of hemorrhage, looking specifically for any large retained pieces of placenta, and removing them. Curettage may be necessary.

Lacerations anywhere in the genital tract may also contribute to hemorrhage, including vaginal lacerations, cervical lacerations, or uterine rupture. A careful inspection of the birth canal and digital examination of the uterus can usually reveal these problems.

Once coagulation factors and fibrinogen have been depleted, the patient’s blood will lose its ability to clot. At this point, there will likely be oozing from the skin puncture sites (IV, blood draw sites), and possibly from the mouth and perineum. This acquired coagulation defect will need to be corrected if you are to meet with success in stopping the bleeding.

Return to the OBGYN Morning Rounds Home Page

This information is provided by The Brookside Associates, a private organization, not affiliated with any governmental agency. The opinions presented here are those of the author and do not necessarily represent the opinions of the Brookside Associates. The patients presented and discussed here are fictitious and are merely representative of clinical conditions. Any resemblence to real patients is purely coincidental and not intentional.

For any clinical condition, many alternative diagnostic and therapeutic efforts may give satisfactory or superior results. The clinical approaches presented here are not intended to reflect and do not reflect the only way to provide good care for these patients.

This information is provided solely for educational purposes. The practice of medicine and surgery is regulated by statute and restricted to licensed professionals and those in training under supervision. Performing medical procedures outside of that setting is a bad idea, is not recommended, and may be illegal.

The presence of any advertising on these pages does not constitute an endorsement of that product or service by the Brookside Associates.