39: Antepartum hemorrhage

Antepartum hemorrhage

Joanna Gibson1, Nada Sabir2, and James Neilson3

1Obstetrics and Gynecology, Yorkshire and Humber, UK

2Obstetrics and Gynecology/Maternal Medicine, Bradford Teaching Hospitals NHS Foundation Trust, Bradford, UK

3Obstetrics and Gynecology, University of Liverpool, Liverpool, UK


Ante‐partum hemorrhage (APH) is defined as bleeding from 24 weeks of pregnancy until birth arising from or in the genital tract [1]. 27.1% (CI 19.9–36.2) of maternal deaths between 2003 and 2009, are due to hemorrhage. Antepartum hemorrhage was the primary culprit in 6.5%, intrapartum in 0.9%, and postpartum hemorrhage in 19.7% [2].

Antepartum hemorrhage complicates 2–5% of all pregnancies and remains a cause of maternal and fetal morbidity and mortality [3]. Antepartum hemorrhage of unknown origin occurs in 50% cases [4] and carries an increased risk of induction of labor and preterm delivery [5]. These babies have increased risk of admission to neonatal intensive care unit (NICU), low birth weights and hyperbilirubinemia.

Placental insertion abnormalities including vasa previa, placenta previa, accreta, increta, and percreta are an important cause of ante‐partum hemorrhage and may contribute to up to 50%. Other possible causes of antepartum hemorrhage include cervical ectropion, infection, genital tract malignancy, vulvo‐vaginal varicosities, trauma, and hematological disorders. Antepartum hemorrhage is a known sequela from domestic violence.


As antepartum hemorrhage can cause significant mortality and morbidity to both mother and fetus, timely assessment of maternal and fetal condition is essential. It is vital to manage significant hemorrhage promptly with maternal resuscitation to manage the subsequent hypovolemic shock. This can be complicated by severe anemia, acute kidney injury, disseminated intravascular consumptive coagulopathy, sepsis and complications from blood transfusions such as transfusion reactions and transfusion related lung injury. Antepartum hemorrhage is often followed by significant postpartum hemorrhage; therefore active measures should be taken to reduce this.

In the presence of antepartum hemorrhage, fetal mortality and morbidity is related to prematurity, intra‐uterine growth restriction, anemia, and hypoxia.

A Kleihauer test can be carried out in Rhesus negative patient to quantify the feto‐maternal hemorrhage. It is not a sensitive test to diagnose placental abruption [1].

Clinical questions

  1. What is the prevalence of placenta previa?

Placenta previa is diagnosed when the placenta is implanted into the lower segment of the uterus, close to (placenta previa minor) or covering the internal os of the cervix (placenta previa major), diagnosed on ultrasound.

The overall prevalence rate of placenta previa was shown to be 4 per 1000 births [6]. Previous C‐section as a consistently described risk factor for placenta previa, accreta, and abruption [7].

A retrospective cohort study between 2000 and 2009 showed a twofold increase in rates of placenta previa in first pregnancies delivered by C‐section (8.7 per 1000 births) compared to women with vaginal first deliveries (4.4 per 1000 births). By adjusting for the effect of previous C‐section and placenta previa rate in England, this meta‐analysis showed that 359 primiparous deliveries by C‐section would result in one additional case of placenta previa in subsequent pregnancies [8].

A systematic review and random‐effects meta‐analysis showed an overall worldwide prevalence of placenta previa of 5.2 per 1000 pregnancies (95% CI 4.5–5.9). Increased prevalence was seen in Asian studies (12.2 per 1000 pregnancies 95% CI 9.5–15.2) compared to, North America (2.9 per 100 pregnancies, 95% CI 2.3–3.5), European studies (3.6 per 1000 pregnancies, 95% CI 2.8–4.6) and Sub‐Saharan Africa (2.7 per 1000 pregnancies; 95% CI: 0.3–11.0) suggesting some variation in prevalence by region but it is unclear if this is due to genetic or ethnic differences or other unknown factors [9].

  1. 2. What is the accuracy of ultrasound in the diagnosis of low lying placenta?

Traditionally placenta previa was diagnosed as a result of painless bleeding or fetal malpresentation in later pregnancy, however placental localization is now possible through routine ultrasound in the ante‐natal period [10].

Concerns regarding the accuracy of transabdominal ultrasound (TAS) have been raised in cases of maternal obesity, overfilling of maternal bladder, difficulty identifying the lower edge of the [11]. Posterior placentas, under‐filling the bladder and interference of the fetal head have also been cited as possible causes for inaccuracy on TAS [12].

Transvaginal ultrasound (TVS) has been shown to be superior to the transabdominal route [13, 14]. It is accurate for diagnosing a low‐lying placenta but also can be used to measure the distance between the placental edge and cervical os in the second and third trimester. Therefore TVS can be used to assess the persistence of placenta previa at term [15].

The location of the placenta is identified at the routine anomaly scan (TAS) in the second trimester. If the placenta is found to be low lying, a TVS should be performed, as this may reclassify 26–60% of cases reducing the need for third trimester scanning [16].

Magnetic resonance imaging (MRI) has been shown to be useful to provide further information in cases of uncertainty when placenta accreta is suspected [17]. MRI may provide more detailed information in the degree of placenta previa when compared to ultrasound but there is no difference in placental localization [18].

  1. 3. How safe is trans‐vaginal scan in the diagnosis of placenta previa?

Several studies have shown that trans‐vaginal scanning has no increased risk of hemorrhage in cases of placenta previa [13, 1922].

Several of these also showed trans‐vaginal scanning to be superior to trans‐abdominal scanning in correctly identifying placenta previa [13, 19] with one study showing the accuracy of diagnosis of 92.8% for the trans‐vaginal route compared with 75.7% for trans‐abdominal method [22].

Trans‐vaginal scanning in the second trimester is deemed safe and important to gain more information about a low lying placenta [16].

  1. 4. What is a rising placenta?

The phenomenon of a “rising placenta” is due to the formation of the lower segment from around 28th week of pregnancy which displaces the placenta upwards. Thus, a low lying placenta may be diagnosed in about 5% of women at 16–18 weeks, but placenta previa is found at delivery in only 10% of the 5% (0.05% overall) as the placenta “rises” with the formation of the lower segment and growth of upper segment [10]. Recent evidence suggests that 96% of cases of low‐lying placenta diagnosed between 16 and 24 weeks had resolved by 36 weeks [23]. However partial or total placenta previa is more likely to persist at term [24].

This phenomenon is more pronounced when the placenta is anterior compared to posterior due to the relative growth of the anterior lower segment. Placental migration has been reported to be less probable in cases of posterior placenta. A mean rate of migration with anterior placenta previa is 2.6 mm per week, compared to 1.6 mm per week in the posterior placenta previa [25].

Placental migration is also less probable in cases where it existed with a previous scar. A previous C‐section delivery was an independent risk factor for persistent previa in women diagnosed with previa in the second trimester, P < 0.05 [26].

  1. 5. What happens if the placenta is covering the internal os at anomaly scan (18–24 weeks)?

Evidence suggests that if the placenta is found to be covering the internal os more than 2 cm at the anomaly scan, then adequate “migration” away from the os to enable a vaginal delivery is unlikely. Several studies have shown lower migration rates in these cases [26]. Follow up scans for placental position is therefore recommended.

It has been suggested that the initial position of placental edge relative to the internal os at 26 weeks along with rate of migration can be used to predict mode of delivery, with cases where the placental edge overlaps the internal os by more than 20 mm resulting in C‐section [15]. However, it was also found that if the placenta covers the internal os by 10 mm or more when scanned between 15 and 24 weeks’ gestation, those patients were at risk of placenta previa at delivery with 100% sensitivity and 85% specificity (positive predictive value 38), [27]. As placental migration is less likely in the presence of a uterine scar, therefore risk factors for placental invasion such as previous uterine surgery/trauma should be considered. In these cases, further imaging could be considered to identify possible placenta accreta/increta/percreta which would alter management of delivery. A large systematic review found ultrasound to have high specificity and sensitivity to diagnose invasion (sensitivity, 90.72 (95% CI, 87.2–93.6)%; specificity, 96.94 (95% CI, 96.3–97.5)%; especially when color Doppler is used (sensitivity, 90.74 (95% CI, 85.2–94.7)%; specificity, 87.68 (95% CI, 84.6–90.4)% [28].

There is inadequate data on which to base the most appropriate dates for follow up scanning in cases of low lying placenta. The Royal College of Obstetricians and Gynaecologist (RCOG) in the UK recommend follow up scanning at 32 weeks if major placenta previa or accreta is suspected at the 20/40 anomaly scan and the women remains asymptomatic. In cases of symptomatic placenta previa minor, a follow‐up scan at 36 weeks will aid decisions on mode of delivery [16, 29]. However in cases of ante‐partum hemorrhage, appropriate management should be based on the clinical situation.

Placental localization scans in women who have had a previous C‐section of uterine surgery are recommended [30]. MRI can be used to further characterize placental implantation disorders, however sensitivity and specificity have been reported to be similar to ultrasound. MRI has been shown to have good predictive accuracy for diagnosis with high predictive accuracy for assessment of depth and topography, with no difference identified in diagnosis in sensitivity (P = 0.24) or the specificity (P = 0.91) between ultrasound and MRI was identified [31].

  1. 6. Is there a role of tocolysis in the management of bleeding placenta previa?

Tocolysis is when drugs are given in order to stop labor with the aim of prolonging pregnancy. There is a growing body of evidence for their use in bleeding placenta previa.

Significant uterine activity has been observed in cases with bleeding placenta previa. The process of lower segment growth and development is associated with potential placental separation and triggering uterine activity. It has been reported that 40% of cases of placenta previa were associated with spontaneous rupture of membranes, spontaneous labor or other problems leading to delivery before 37 weeks [32].

It is conventionally believed that uterine contractions may help containing the ongoing bleed, by compressing bleeding vessels at the placental bed. However, a number of reports suggested beneficial effect of tocolysis in management of bleeding placenta previa including prolongation of pregnancy.

The use of tocolysis in the management of bleeding placenta previa theoretically can be beneficial.

One of the potential benefits to prolong pregnancy is to allow time to give corticosteroids to enhance fetal lung maturity. Cochrane systematic review supports their use e to reduce potential problems related to prematurity including respiratory distress syndrome, intra‐ventricular hemorrhage, necrotizing enterocolitis, and systemic infection [33].

In the only small randomized controlled trial of 60 patients with symptomatic placenta previa between 28 and 34 weeks of gestation, tocolysis with ritodrine for 7 days was compared to no treatment. Treatment was associated with prolongation of pregnancy (25.33 ± 17.7 days compared with 14.47 ± 20.33 days, P < 0.05) and an increased birth weight (2.27 ± 0.59 kg compared with 1.95 ± 0.55 kg). No difference in maternal complications were shown, and particularly no increased incidence of bleeding, no difference in total blood loss or number of blood transfusions found [34]. However ritodrine has known potential cardiovascular risks including tachycardia and palpitation and both are non‐desirable with patients suffering from bleeding related hypotension.

Retrospective analysis shows that in preterm labor with placenta previa, tocolysis can reduce preterm delivery rates, improve birthweights, especially when long term maintenance drugs (terbutaline) compared to short term drugs (magnesium sulfate). There was no observed statistical difference in incidence of recurrent bleeding, interval from admission to first recurrent bleeding, and need for blood transfusion [35].

When tocolysis was used to prolong pregnancy in women with abruption and placenta previa, there were no adverse maternal or fetal effects reported, with no increase in fetal distress or need for transfusion [36].

Although there are no known added risks reported with the available studies there is no clear evidence on the benefit gained. Therefore, there is currently not enough good evidence to support the use of tocolytics to improve perinatal outcome in cases of bleeding placenta previa. A review of the evidence (one randomized control trial and two retrospective studies) in 2011 suggested that if tocolytics are used, their use should be restricted to 48 hours [37]. More studies and trials are needed to back up or refute the potential benefits of tocolytics in the management of placenta previa.

  1. 7. What is the place of home management in the care of women with major placenta previa?

Traditionally patients with major placenta previa were hospitalized until delivery [38] aiming to prolong pregnancy to 37 weeks and to manage complications in a timely fashion. More recently, long hospital admissions have been shown to have social, psychological, financial, and domestic implications on a woman and her family [3]. The notion of home care was brought to discussion following a retrospective analysis. The study involved 15 930 patients including 58 with placenta previa and concluded that “in the majority of cases with or without bleeding and irrespective of the degree of previa out‐patient management would appear safe and effective” [39].

It has been shown that in cases of symptomatic placenta previa major early delivery, by C‐section, of lower birthweight babies requiring neonatal unit admission were more likely outcomes compared to those who were asymptomatic. Therefore out‐patient management can be appropriate and the number of bleeds rather than the degree of placenta previa should be considered with more caution [40].

The Cochrane systematic review on interventions for suspected placenta previa included one randomized control trial showing reduction in hospital stay for those managed as outpatients with no difference in neonatal morbidity [41].

It may also be important to consider cost implications of inpatient treatment. A retrospective analysis showed no difference in maternal or fetal morbidity by reduced inpatient stay and therefore cost for mothers and mother–infant pairs [42].

In conclusion, home or hospital management can be appropriate and it may be best to consider each individual case. During outpatient management the patient should be able to directly access the hospital if they become symptomatic in any way. International opinion agrees on the recommendation that women at risk of major obstetric hemorrhage should remain close to a hospital able to manage this emergency for the third trimester of pregnancy [16].

  1. 8. At what gestation should delivery be planned, and at what gestation should it occur?

A large systematic review showed the preterm delivery rates for low‐lying/marginal placenta (26.9%), placenta previa (43.5%), and increased risk for preterm delivery in patients with placenta previa (risk ratio [RR], 5.32; 95% confidence interval [CI], 4.39–6.45). This prematurity can lead to significant morbidity and mortality. This review also showed increased risk of NICU admissions (RR, 4.09; 95% CI, 2.80–5.97), neonatal death (RR, 5.44; 95% CI, 3.03–9.78), and perinatal death (RR, 3.01; 95% CI, 1.41–6.43) with placenta previa [43].

To reduce risk of respiratory distress and need for prolonged neonatal unit support, corticosteroids are routinely administered to those electively delivered before 38 weeks [33].

The care bundle working group [29] looked at over 1000 cases of placenta previa over 10 years and found that it is rare for delivery to be required prior to 32 weeks but around 40% will be delivered as an emergency before 38 weeks’ gestation. This is in general agreement with previous data [39].

Individualized patient care is likely to be appropriate in those cases where prediction of onset of labor is difficult. In patients with uncomplicated placenta previa the general consensus is to aim for delivery between 38 and 39 weeks as a balance between neonatal respiratory morbidity and risks of heavy bleeding.

  1. 9. Can women with pregnancies complicated by placenta previa aim for a vaginal birth?

Currently there is not enough good quality data from large scale studies which consider mode of delivery with associated outcomes of maternal and neonatal morbidity where the placenta lies <20 mm away from the internal os, therefore this is the cut off which is generally used when counseling women regarding mode of delivery in placenta previa.

Women with placenta previa are at higher risk of post‐partum hemorrhage, blood transfusion, and hysterectomy [44].

A vaginal birth can be considered when the placental edge – os distance is >2 cm and the placenta described as low lying, but delivery should be by elective C‐section if the distance is <1 cm or less [45].

Various small studies looking at rate of vaginal delivery in cases where placenta‐os distance of 11–20 mm concluded that the rate of vaginal birth varied between 36% and 76.5% [4648].

It has been suggested that other factors such as the thickness of the lower edge of the placenta and the position (anterior or posterior) may also affect clinical outcome. A small study was the first to show that women with thicker leading edge (>1 cm) of placenta are more likely to have a cesarean section (P = 0.02), hysterectomy or severe hemorrhage than those with a thinner placental edge [49].

Therefore these factors may be considered, in addition to the clinical situation, maternal preferences and resources when deciding the optimal mode of delivery.

  1. 10. How would the management plan differ in cases of suspected placenta previa accreta, increta, or percreta?

A morbidly adherent placenta is a low lying placenta that invades into or through the myometrium. This is further defined by the depth of placental invasion: accreta invades the myometrium (80% of cases), increta invades deeply through the full thickness of the uterus (15%) and a percreta invades through the uterus to the serosal layer, sometimes with associated invasion into the bladder and other structures within the pelvis (5%).

It is most likely to occur when the decidua basalis is deficient, therefore predisposing factors include infection, previous surgery, including cesarean section.

It is the leading cause for intrapartum hysterectomy and often associated with significant morbidity. The risk of hysterectomy with placenta previa and uterine scar is 10% but with placenta previa accreta it is 66% [50]. There is an increased risk of preterm labor, and perinatal morbidity and mortality in those diagnosed with major placenta previa or accreta [51].

The management of placenta previa accreta incorporates pre‐delivery planning, intrapartum multidisciplinary management and post‐partum intensive care support. Establishing diagnosis early in pregnancy allows for adequate multidisciplinary based care and planning for delivery. There are no randomized controlled trials found on management of placenta previa accreta. Most studies are observational and of small numbers.

Raised index of suspicion with those at increased risk including advanced maternal age and previous scarring, may improve early detection rate. Antenatal diagnosis can reduce estimated blood loss and transfusion requirements [52].

Ultrasound was found to be highly accurate, especially when combined with color Doppler for diagnosis of invasive placentation in high‐risk women [28].

MRI has also been found to be very accurate in diagnosing placental invasion for both depth and topography [31]. Both ultrasound and MRI have high specificity and sensitivity indicating that either mode of imaging can be appropriate in further characterization of placental site and invasion [31, 53].

A large prospective observation cohort study of 30 132 women undergoing C‐section before labor (evidence level II) showed serious maternal morbidity (such as surgical complications, admission to intensive care unit (ICU) and hysterectomy) with increasing and subsequent deliveries by C‐section. The risk of placenta accreta in placenta previa was 3%, 11%, 40%, 61%, and 67% for first, second, third, fourth, and fifth or more repeat cesarean deliveries, respectively. Hysterectomy was required in 0.65% first, 0.42% second, 0.90% third, 2.41% fourth, 3.49% fifth, and 8.99% sixth or more cesarean deliveries. Therefore it is suggested that counseling before a primary elective C‐section includes discussion over future family planning/number of desired pregnancies [54].

In the UK, to optimize clinical management, a care bundle has been recommended and tested for placenta previa after previous cesarean section. This includes six main criteria: multidisciplinary involvement in preoperative planning; consultant obstetrician planning and directly supervising delivery; consultant anesthetist planning and directly supervising anesthetic at delivery; blood and blood products available on site; discussion and consent including possible intervention (hysterectomy, leaving the placenta in situ, cell salvage, and intervention radiology); and high dependency care availability [29].

This care bundle was tested in a pilot study and was found easy to use, whilst encouraging multi‐disciplinary planning and prompting senior input [55].

The American College of Obstetricians and Gynaecologist (ACOG) also recommends individualized delivery planning including decisions regarding most appropriate surgical management. It generally recommends cesarean hysterectomy with placenta left in situ, but in individual cases where fertility preservation is preferred other options may be considered such as leaving the placenta in situ or partial resection of the placenta. However, further management, including hysterectomy may be required after the C‐section, therefore this should be considered and discussed with the patient before delivery [56].


Clinical history and assessment suggests a diagnosis of placenta previa in the presence of ante‐partum hemorrhage and uterine activity at 26 weeks gestation. Hemodynamic stability of the mother is imperative, followed by close monitoring of fetal heartbeat on the cardiotocograph. Preparations for delivery by C‐section should be made including corticosteroids for fetal lung maturity and tocolysis could be considered after discussion with a senior obstetrician. However if bleeding and uterine activity settle and there is no fetal distress, a conservative approach can be considered with further imaging such as TVS or MRI at a later gestation.

Jul 20, 2020 | Posted by in GYNECOLOGY | Comments Off on 39: Antepartum hemorrhage
Premium Wordpress Themes by UFO Themes