CLINICAL OBSTETRICS AND GYNECOLOGY

Volume 57, Number 4, 791–796

r 2014, Lippincott Williams & Wilkins

Obstetric
Hemorrhage:
Recent Advances
ANTONIO SAAD, MD, and MAGED M. COSTANTINE, MD
Department of Obstetrics and Gynecology, Division of Maternal
Fetal Medicine, The University of Texas Medical Branch,
Galveston, Texas

Abstract: Hemorrhage is the most common cause of maternal mortality and the most common
maternal mortality worldwide, and represents the form of shock in obstetrical practice.1
third most common obstetrical cause of maternal
death in the United States. Although uterine atony Traditionally, uterine atony had been
was previously a major cause of peripartum hemor- the major cause for peripartum hysterec-
rhage, more recently, it appears that abnormal pla- tomy; however, recently abnormal pla-
centation is the leading etiology and the main centation has surpassed it in the
indication of peripartum hysterectomy. Early identi- developed world.2,3 This may be because
fication and aggressive management of obstetrical
hemorrhage is of utmost importance to prevent ma- of the current trends in obstetrical pa-
ternal morbidities and mortality. tients and practice including advanced
Key words: obstetrical hemorrhage, massive trans- maternal age at time of conception, higher
fusion, placenta accreta obesity rates, and increasing cesarean de-
livery rates. These may have resulted in
increased prevalence of placental abnor-
malities including placenta previa and
Introduction placental accretism (accreta, increta, or
According to the Centers for Disease percreta). It is estimated that placental
Control, about 650 women die each year accretism complicates 1 in 533 pregnan-
in the United States as a result of preg- cies; a rate that represents a sharp increase
nancy-related complications, with obstet- from 1 in 4027 pregnancies in the 1970s.4
ric hemorrhage being a top tier leading In recent years, novel pathways
etiology of maternal mortality.1 World- thought to play crucial role in the patho-
wide, hemorrhage remains the #1 cause of genesis of hemorrhage have been identi-
fied. In addition, the standard transfusion
Correspondence: Maged M. Costantine, MD, Depart- guidelines currently in place have been
ment of Obstetrics and Gynecology, Division of Mater- questioned, with a push to make hemo-
nal Fetal Medicine, The University of Texas Medical
Branch, 301 University Blvd, Galveston, TX. E-mail: static resuscitation (early aggressive
[email protected] blood product replacement) and massive
The authors declare that they have nothing to disclose. transfusion protocols essential elements

CLINICAL OBSTETRICS AND GYNECOLOGY / VOLUME 57 / NUMBER 4 / DECEMBER 2014

www.clinicalobgyn.com | 791
792 Saad and Costantine

in these guidelines.3 In this paper, we will and its effects on the fetus remains un-
review the diagnosis and management of certain. Therefore, use of paramagnetic
placental accretism, as well as, discuss contrast media during pregnancy may be
advances in the medical management of undertaken only after evaluating the risk
obstetrical hemorrhage. These topics benefit ratio of such approach, and if the
have been recently the subject of many improved accuracy of the study is ex-
extensive reviews by Pacheco and pected to significantly improve patient
colleagues.5–7 care and guide essential therapeutic
interventions.
DIAGNOSIS OF PLACENTA ACCRETA
In the majority of cases, an ultrasound ANTEPARTUM CARE AND TIMING OF
performed by an experienced and trained DELIVERY
sonographer has good sensitivity and spe- To ensure optimal care, it is recom-
cificity to diagnose placental accretism mended that patients with suspected pla-
(sensitivity of 77% to 87%, specificity of cental accretism or at risk for peripartum
96% to 98%, and negative predictive hemorrhage should be referred to a ter-
value of 98%).8,9 Traditional sonographic tiary center and managed by a multidisci-
clues suggestive of placental accretism plinary team including specialists in
include loss of the normal retroplacental maternal fetal medicine, urology, vascu-
hypoechoic space, increased vascularity lar surgery, interventional radiology,
with uterine wall vessel invasion on color blood bank, and neonatology. Such an
Doppler, and the presence of placental approach has been shown to decrease
lacunae giving a ‘‘moth-eaten or Swiss maternal morbidity.5,13
cheese’’ appearance, which is the most The use of serial ultrasound examina-
predictive sonographic sign with sensitiv- tions to monitor fetal growth is not rec-
ity of 79% and positive predictive value of ommended in patients with isolated
92%. placenta previa without accretism, as the
Although an ultrasound may be latter was not found to be associated with
enough in the majority of patients, mag- fetal growth restriction.14 However, when
netic resonance imaging (MRI) may be placental accretism is suspected or
useful in patients suspected of having present, serial sonographic follow-up
placental accretism to evaluate the depth every 3 to 4 weeks is recommended to
of invasion, especially for posterior pla- assess fetal growth, and the depth of
centas, as well as to better discriminate the placental invasion.
involvement of adjacent organs when a Patients with suspected placental ac-
placenta percreta is suspected.10 Bulging cretism or at risk for peripartum hemor-
of the lower uterine segment, the presence rhage need to have their hemoglobin
of a heterogeneous placenta, and dark levels optimized before delivery. It is com-
intraplacental linear bands on T2- mon to start these patients on iron and/or
weighed MRI images are suggestive of folic acid supplements, and occasionally,
placental accretism.11 In general, the ac- intravenous iron and/or recombinant er-
curacy of ultrasound and MRI in diag- ythropoietin may be needed as adjuvant
nosing placenta accretism is similar.5,12 Of therapy. Preoperative autologous blood
note, although the use of paramagnetic donation may be considered in select pa-
contrast media (gadolinium) would likely tients, such as those with rare blood types
improve the diagnostic performance of and/or alloimmunization to rare antibod-
MRI by enhancing the outer placental ies. Patients [with hemoglobin >10 g/dL,
surface relative to the myometrium; how- and within 42 d of surgery, the maximum
ever, the agent readily crosses the placenta accepted time to maintain packed red

www.clinicalobgyn.com
 Systolic Blood Pressure 793

blood cells (RBC) in the blood bank] may cardiovascular disease), there is minimal
donate their own blood units during preg- evidence that it alone reduces significantly
nancy, and these will be later used intra- the need for transfusion. This technique
operatively if needed. Disadvantages of involves collecting 500 to 1000 mL of
such approach is that it has not been blood from the patient and replacing it
shown to be cost-effective, is not well with either colloid (1:1 ratio) or crystal-
studied in pregnancy, and may not be loid (3:1 ratio) fluids to maintain hemo-
acceptable by Jehovah’s Witnesses dynamics.15 After surgical bleeding is
patients.15 controlled, the patient is transfused the
Regarding timing of delivery, the Soci- previously collected blood. This techni-
ety of Maternal Fetal Medicine recom- que is acceptable for some Jehovah’s Wit-
mends that patients with placenta previa nesses as long as the collection bag is
should be delivered at 36 to 37 weeks, and connected to the central venous line at
those with suspected placental accretism all times, without any ‘‘circuit’’
at 34 to 35 weeks.16 Testing for fetal lung disconnections.15
maturity before delivery is not thought to Another safe and effective technique
not provide any additional benefit, and that may be accepted by Jehovah’s Wit-
thus is not recommended.17 The decision nesses patients is intraoperative cell sal-
to deliver a pregnant woman with sus- vage.5,6 After collecting and filtering
pected accretism between 34 and 37 weeks aspirated blood from the surgical field,
should be individualized depending on sterile filters will separate tissue factor, a-
the patient obstetrical history, pregnancy fetoprotein, platelets, and circulating pro-
complications, and risk of preterm labor coagulants.21 The collected blood is then
and/or vaginal bleeding.5,18 reinfused to the patient after its reaches a
hematocrit concentration >55%. A Klei-
PERIOPERATIVE INTERVENTIONS hauer-Betke stain and appropriate ad-
Although a complete sympathectomy (eg, ministration of anti-D immunoglobulin
spinal anesthesia) may impair the pa- is required for prevention of alloimmuni-
tient’s ability to vasoconstrict and in- zation in Rh-negative blood type pa-
crease systemic vascular resistances and tients.15 Moreover, there is no evidence
thus may put the patient at risk from the of increased risk of amniotic fluid embo-
hypovolemia associated with acute blood lism with the use of this technique in the
loss; a continuous epidural technique is 400 plus cases described in the obstetric
safe and may be appropriate for patients literature.22
with placental accretism.19,20 However, in Another therapy that is being more
emergent situations of massive bleeding, frequently used in obstetric hemorrhage
general anesthesia may be more appro- is recombinant factor VII a (rFVIIa).
priate. In contrast, permissive hypoten- Traditionally, this therapy has been ap-
sion (with systolic blood pressure of 80 to proved for patients with hemophilia and
100 mm Hg) may be discussed with the inhibitory alloantibodies.23 However, it
anesthesia team and considered in pa- has been used off-label in various clinical
tients with postpartum hemorrhage. No scenarios including trauma, cardiovascu-
data are available on its safety before lar surgery, and obstetrical hemorrhage;
delivery of the fetus as it may compromise and is being added to many massive trans-
uterine perfusion.5,6 fusion protocols. The agent has a short
Although acute normovolemic hemo- half-life (2 to 6 h), requires multiple ad-
dilution may be considered in certain ministrations, and its use is associated
patients (those with hemoglobin level with increased risk of arterial throm-
>10 g/dL and no history of boembolism. rFVIIa is not the first line

www.clinicalobgyn.com
794 Saad and Costantine

FIGURE 1. Massive transfusion protocol in obstetrics. Once activated, blood bank personnel
will continue preparing blood products until the protocol is inactivated by the surgical team.
After 2 sequences of rounds 1 to 3, if not inactivated, the protocol will start again from round 1.
*The utilized dose of recombinant activated factor VII is based on local expert opinion. ^If
bleeding is not controlled the sequence is repeated. FFP indicates fresh frozen plasma; pRBC,
packed red blood cells. Adapted from Pacheco et al.7

of treatment for massive hemorrhage and based on massive administration of crys-

should only be used after surgical bleed- talloids, colloids and packed RBC, with
ing has been controlled, with a combina- the use of fresh frozen plasma (FFP),
tion of various blood products, and cryoprecipitates, and platelets limited to
only after meeting multiple prerequisites low fibrinogen (<100 mg/dL), prolonged
including platelet count >50,000/mm3, prothrombin or activated partial throm-
fibrinogen >50 to 100 mg/dL, tempera- boplastin times (usually >1.5  normal),
ture >321C, pH>7.2, and normal ionized and thrombocytopenia (platelet count
calcium.24 Multiple, nonobstetric, rando- <50,000/mm3).26 However, it is been
mized controlled trials have been pub- shown that massive early crystalloid re-
lished where rFVIIa has been used to suscitation may worsen bleeding.27 This is
control bleeding, and none showed that thought to be due to the increased intra-
its use improves survival. However, 4 of 17 vascular hydrostatic pressure, disruption
studies concluded that it reduced trans- of fresh clot formation, and enhancement
fusion requirements or blood loss.25 The of the fibrinolytic pathway.27,28 On the
optimal dose of rFVIIa, especially in preg- basis of these mechanisms of early coa-
nancy, is unknown, but the majority of gulopathy, it is suggested that early
reports in obstetrical hemorrhage have replacement of clotting factors, and con-
used a dose of 40 to 90 mg/kg.6,7 trol of the fibrinolytic state may be
Acute coagulopathy remains a major beneficial.6,7
complication in the setting of massive In addition to restriction of early ex-
hemorrhage, most commonly secondary cessive crystalloid and packed RBC ad-
to ‘‘dilutional coagulopathy.’’ Tradition- ministration, recent data from the
al resuscitative maneuvers are heavily nonobstetric literature also suggest

www.clinicalobgyn.com
 Systolic Blood Pressure 795

transfusion of packed RBC concomi- cardiac output and blood pressure.32

tantly with platelets and FFP at a ratio Clinically, ACS frequently presents as
as close to 1:1:1 as possible. This has been oliguria and/or hypotension postopera-
shown in several retrospective studies in tively which does not respond, and may
military and civilian trauma settings to worsen, with administration of intrave-
reduce mortality by 15% to 62% com- nous fluids. If there is suspicion of in-
pared with higher ratios of FFP:PRBC.29 creased abdominal pressures, clinicians
However, these studies are limited by should consider bladder pressure meas-
survival bias30,31 and they did not include urement, which reflects the pressures in
pregnant women.6 the abdomen.32 The diagnosis of ACS
In an attempt to improve the response includes a bladder pressure of >20 mm
to massive obstetrical hemorrhage, many Hg and dysfunction of at least 1 organ,
centers in the United States including our and these patients are managed with fluid
labor and delivery unit have implemented restriction, enteral feeding, but most will
massive transfusion protocols that in- need surgical decompression.32
clude early administration of high plate-
let:FFP:PRBC ratios and early use of
rFVIIa. An example of such protocol, Summary
adapted from Pacheco et al7 is depicted Obstetrical hemorrhage remains among
in Figure 1. This protocol is activated if the major causes of maternal morbidity
hemorrhage is expected to be massive and mortality, with placental accretism
(anticipated need to replace Z50% of representing the major cause and the most
blood volume within 2 h) or in patients common indication for a cesarean
with ongoing bleeding after the transfu- hysterectomy in the developed world.
sion of 4 U of packed RBCs within a short Early identification of placental accretism
period of time (1 to 2 h).7 and patient transfer to a tertiary care
center with multidisciplinary team ap-
POSTOPERATIVE CARE proach, improves maternal and fetal out-
Patient recovering from a massive obstet- comes. Various strategies have been
rical hemorrhage are at increased risk for identified that may assist in management
thromboembolic events during the post- of these patients including, normovolemic
partum period. This is secondary to endo- hemodilution, intraoperatively use of cell-
thelial injury, inflammation, prolonged saver, hemostatic resuscitation, and the
surgery, immobilization, as well as the use of massive transfusion protocols.
thrombotic state of the postpartum peri-
od. Mechanical deep venous thrombosis
prophylaxis should be applied intraoper- References
atively and continued thereafter; and 1. Kung HC, Hoy DL, Murphy SL. Deaths: Final
pharmacological prophylaxis should be Data for 2005. Volume 56, No. 10. Hyattsville,
started as soon as the clinician feels it is MD: National Center for Health Statistics; 2008.
safe and the bleeding risk is minimal.6 In 2. Kwee A, Bots ML, Visser GH, et al. Emergency
peripartum hysterectomy: a prospective study in
addition, abdominal compartment syn- The Netherlands. Eur J Obstet Gynecol Reprod
drome (ACS) may occur secondary to Biol. 2006;124:187–192.
fluid third spacing after crystalloid and 3. Smith J, Mousa HA. Peripartum hysterectomy
colloid administration and endothelial in- for primary postpartum haemorrhage: incidence
jury. This leads to increase intra-abdomi- and maternal morbidity. J Obstet Gynaecol. 2007;
27:44–47.
nal pressures, collapse of abdominal and 4. Wu S, Kocherginsky M, Hibbard JU. Abnormal
retroperitoneal vessels, leading to de- placentation: twenty-year analysis. Am J Obstet
crease cardiac preload, and ultimately Gynecol. 2005;192:1458–1461.

www.clinicalobgyn.com
796 Saad and Costantine

5. Pacheco LD, Gei AF. Controversies in the man- 19. Parekh N, Husaini SW, Russell IF. Caesarean section
agement of placenta accrete. Obstet Gynecol Clin for placenta previa: a retrospective study of anaes-
North Am. 2011;38:313–322. thetic management. Br J Anaesth. 2000;84:725–730.
6. Pacheco LD, Saade GR, Gei AF, et al. Cutting- 20. Ioscovich A, Mirochnitchenko E, Halpern S, et al.
edge advances in the medical management of Perioperative anaesthetic management of high
obstetrical hemorrhage. Am J Obstet Gynecol. order repeat caesarean section: audit of practice
2011;205:526–532. in a university affiliated medical center. Int J
7. Pacheco LD, Saade GR, Costantine MM, et al. Obstet Anesth. 2009;18:314–319.
The role of massive transfusion protocols in ob- 21. Tawes RL Jr. Clinical applications of autotrans-
stetrics. Am J Perinatol. 2013;30:1–4. fusion. Semin Vasc Surg. 1994;7:89–90.
8. Warshak CR, Eskander R, Hull AD, et al. Accu- 22. Rainaldi MP, Tazzari PL, Seagliarini G, et al.
racy of ultrasonography and magnetic resonance Blood salvage during cesarean section. Br J
imaging in the diagnosis of placenta accreta. Anaesth. 1998;80:196–198.
Obstet Gynecol. 2006;108:573–581. 23. Stanworth SJ, Birchall J, Doree CJ, et al. Re-
9. Comstock CH, Love JJ Jr, Bronsteen RA, et al. combinant factor VIIa for the prevention and
Sonographic detection of placenta accreta in the treatment of bleeding in patients without haemo-
second and third trimesters of pregnancy. Am J philia. Cochrane Database Syst Rev. 2007;18:
Obstet Gynecol. 2004;190:1135–1140. CD005011.
10. Baughman WC, Corteville JE, Shah RR. Placenta 24. Spahn DR, Cerny V, Coats TJ, et al. Management
accreta: spectrum of US and MR imaging find- of bleeding following major trauma: a European
ings. Radiographics. 2008;28:1905–1916. guideline. Crit Care. 2007;11:R17.
11. Teo TH, Law YM, Tay KH, et al. Use of magnetic 25. Johansson PI, Ostrowski SR, Secher NH. Man-
resonance imaging in evaluation of placental in- agement of major blood loss: an update. Acta
vasion. Clin Radiol. 2009;64:511–516. Anaesthesiol Scand. 2010;54:1039–1049.
12. Dwyer BK, Belogolovkin V, Tran L, et al. Pre- 26. Nascimento B, Callum J, Rubenfeld G, et al.
natal diagnosis of placenta accreta: sonography Clinical review: fresh frozen plasma in massive
or magnetic resonance imaging? J Ultrasound bleedings-more questions than answers. Crit
Med. 2008;27:1275–1281. Care. 2010;14:202.
13. Eller AG, Bennett MA, Sharshiner M, et al. Ma- 27. Ickx BE. Fluid and blood transfusion manage-
ternal morbidity in cases of placenta accreta man- ment in obstetrics. Eur J Anaesthesiol. 2010;27:
aged by a multidisciplinary care team compared 1031–1035.
with standard obstetric care. Obstet Gynecol. 28. Brohi K, Cohen MJ, Davenport RA. Acute
2011;117:331–337. coagulopathy of trauma: mechanism, identifica-
14. Harper LM, Odibo MO, Macones GA, et al. tion, and effect. Curr Opin Crit Care. 2007;13:
Effect of placenta previa on fetal growth. Am J 680–685.
Obstet Gynecol. 2010;203:330.e1–330.e5. 29. Cotton BA, Gunter OL, Isbell J, et al. Damage
15. Catling S. Blood conservation techniques in ob- control hematology: the impact of a trauma
stetrics: a UK perspective. Inter J Obstet Anesth. exsanguination protocol on survival and
2007;16:241–249. blood product utilization. J Trauma. 2008;64:
16. Spong CY, Mercer BM, D’Alton M, et al. Timing 1177–1182.
of indicated late-preterm and early-term birth. 30. Snyder CW, Weinberg JA, McGwin G, et al. The
Obstet Gynecol. 2011;118:323–333. relationship of blood product ratio to mortality:
17. Robinson BK, Grobman WA. Effectiveness of survival benefit or survival bias? J Trauma.
timing strategies for delivery of individuals with 2009;66:358–362.
placenta previa and accreta. Obstet Gynecol. 31. Scalea TM, Bochicchio KM, Lumpkins K, et al.
2010;116:835–842. Early aggressive use of fresh frozen plasma does
18. Stafford IA, Dashe JS, Shivvers SA, et al. Ultra- not improve outcome in critically injured trauma
sonographic cervical length and risk of hemorrhage patients. Ann Surg. 2008;248:578–584.
in pregnancies with placenta previa. Obstet Gyne- 32. Cheatham ML. Abdominal compartment syn-
col. 2010;116:595–600. drome. Curr Opin Crit Care. 2009;15:154–162.

www.clinicalobgyn.com