Current practice involves screening for low-lying placentas at the routine second trimester anatomy scan in order to predict the likelihood of placenta previa at delivery. If the placenta remains low lying in the third trimester, a transvaginal scan should be performed to confirm the diagnosis. Transvaginal sonography is deemed safe and more accurate in the diagnosis of placenta previa [15–18].

The majority of women in whom the placental edge reaches or overlaps the internal os in mid-pregnancy will not have placenta previa at term. ‘Placental migration’ is believed to occur in the second and third trimesters as the lower segment of the uterus develops. It is less likely to occur in cases where the placenta is posterior, in women who have had previous Cesarean section(s) and when the placenta completely or substantially overlaps the internal os [19–27].

Regardless of previous scan results, placenta previa should be suspected in any pregnant woman who presents with painless vaginal bleeding, particularly with a high presenting part or an abnormal fetal lie (because the placenta prevents the fetal presenting part, usually the head, from engaging into its mother’s pelvis).

Women who have had a previous CS and have an anterior placenta or placenta previa are at significantly increased risk of placenta accreta. Clinical suspicion of placenta accreta may be confirmed by ultrasonography; when ultrasonography is equivocal, magnetic resonance imaging (MRI) may clarify the situation. Two recent meta-analyses suggest that ultrasound has a sensitivity of 80–90 % in the diagnosis of placenta accreta, with specificity of 95–97 % [28, 29]. One of these meta-analyses [29] also looked at MRI diagnosis; sensitivity for placenta accreta was 82 % with a specificity of 88 %. Definitive diagnosis, however, can be made only at the time of CS and by histopathological examination.

Women with placenta previa and/or accreta may be asymptomatic (i.e. with no history of APH) or symptomatic (i.e. present with vaginal bleeding).

Elective CS is recommended for women with placenta previa persisting at term. CS is usually performed at 38 weeks’ gestation in cases with placenta previa and at 36–37 weeks in cases of placenta accreta, in order to minimise the risk of labour leading to bleeding prior to the booked CS date. Of course, if there is major APH prior to this, urgent delivery may be required, regardless of the gestation [20]. Evidence around the mode of delivery for women who have minor placenta previa is less clear. Women with a placental edge which lies less than 2 cm from the cervical os are likely to require delivery by CS. If the fetal head is engaged closer to term (as the lower segment continues to develop beyond 36 weeks), a transvaginal ultrasound scan is performed. If the leading placental edge is more than 2 cm from the internal os, the likelihood of achieving a vaginal delivery is at least 63 % [30]. The risk of postpartum hemorrhage (PPH) remains higher in this group of women, so should be anticipated and managed accordingly.

The management of women who present with bleeding depends on the extent of bleeding and fetal maturity. All women should have a full blood count (FBC) and ‘group and save’ on admission. Corticosteroids should be administered in women less than 35 weeks’ gestation to improve fetal lung maturity in the event of threatened preterm delivery [31]. All RhD (Rhesus) negative women should have estimation of fetal maternal hemorrhage by the Kleihauer-Betke test or flow cytometry, and receive prophylactic anti-D immunoglobulin intramuscularly (IM) [32]. Women who present with massive ongoing bleeding will require urgent surgical intervention following hemodynamic stabilisation. Women that are stable at presentation (i.e. no maternal or fetal compromise) with a self-limiting episode of painless unprovoked bleeding can be managed conservatively. However, they are thought to be at high risk of further sudden unprovoked bleeding, so are routinely admitted for 24–48 h following cessation of bleeding.

The use of tocolytics [33–37] or cervical cerclage [38, 39] to treat women who bleed with placenta previa has been investigated; their use remains controversial and is not recommended for routine clinical use.

In the past it was recommended that women with major previa who had previously bled should receive in-patient care from approximately 34 weeks of gestation [40]. There is, however, no clear evidence to suggest an association between the extent of placenta previa and the likelihood of bleeding or the requirement for urgent delivery [22, 25]; the risk of bleeding and need for in-patient management should be assessed on an individual basis. All women with a risk of major APH should be advised to remain close to the recommended hospital for delivery for the duration of the third trimester.

As a minimum requirement, a consultant obstetrician and consultant anaesthetist should be present on site when women with a placenta previa undergo CS. All women should have a baseline FBC and four units of red cells cross-matched. The Royal College of Obstetricians & Gynaecologists’ (RCOG) guideline [25] suggests that prophylactic steroids for fetal lung maturation should be given prior to any elective CS performed up to 38 + 6 weeks’ gestation. Prior to delivery a consultant led discussion regarding the delivery, risk of significant hemorrhage, indications for transfusion and the possibility of peripartum hysterectomy should be held and documented clearly.

In cases of confirmed or suspected placenta accreta a consultant obstetrician and anaesthetist should be present and directly supervising delivery. The procedure is ideally planned ahead, involving a multidisciplinary team (including haematologist, interventional radiologist, intensive care staff, and blood transfusion laboratory staff), with blood and blood components available on standby. Discussion regarding the significant risk of peripartum hemorrhage, anticipated skin and uterine incisions and possible interventions including conservative management, use of cell salvage, interventional radiology and the possibility of hysterectomy should be discussed with the woman and documented clearly in both the notes and consent form.

When placenta accreta is suspected antenatally, the surgical avoidance of the placental site and non-separation of the placenta is thought to be associated with reduced maternal mortality by reducing the risk of bleeding [25, 41]. Data from observational studies support avoiding the placenta and leaving it attached, or proceeding with peripartum hysterectomy [41–43] in cases of placenta percreta.

The risk of massive PPH is 12 times greater in women with placenta previa [40]. This is mainly because the lower segment of the uterus (where the placenta was implanted) contracts less efficiently than the upper segment (the usual site of placental implantation), leading to continued bleeding from the placental bed vessels. The rate of peripartum hysterectomy is also increased, particularly if the woman has had a previous CS.

Leaving the placenta adherent may reduce intra-operative bleeding but is associated with an increased risk of postpartum bleeding and infection. Prophylactic antibiotics may be considered postpartum. Prophylactic arterial embolisation or methotrexate have not been shown to reduce the risk of these complications and are therefore not routinely recommended [25].

If the placenta separates, it should be delivered. Subsequent hemorrhage must be anticipated and managed appropriately. Partial separation of the placenta is also possible and adherent portions may be left attached. Massive hemorrhage is highly likely in this situation and timely and appropriate management is essential [12, 44].

Placental abruption is defined as the partial or complete separation of the placenta prior to the delivery of the fetus [45].

Vasa previa is a condition in which the fetal blood vessels traverse the fetal membranes of the lower segment of the uterus below the presenting part, unsupported by the umbilical cord or placental tissue [67]. Vasa previa has an estimated incidence of 1 in 2,000–5,000 deliveries [67, 68]. Associated risk factors are summarised in Table 14.3 [67–74].

Postpartum hemorrhage is the commonest form of obstetric hemorrhage. There is currently no single globally agreed definition for PPH, hence various definitions exist [95]. Traditionally, PPH is defined as blood loss from the genital tract more than 500 mL occurring after delivery until 12 weeks after birth [96]. At CS, PPH has traditionally been defined as blood loss greater than 1,000 mL [97]. Primary PPH is that occurring within the first 24 h of delivery, though usually occurs shortly after birth. Secondary PPH is hemorrhage which occurs more than 24 h but before 12 weeks after delivery [95]. PPH can be minor (500–1,000 mL) or major (more than 1,000 mL). Major PPH may be further sub-classified into moderate (1,000–2,000 mL) or severe (more than 2,000 mL) [40].

Massive blood loss is normally defined as the loss of one blood volume in 24 h. An alternative definition which may be more useful in an acute situation is loss of 50 % of total blood volume in 3 h or >150 mL/min [40, 114, 115].

Estimated blood volume: pregnant women at term have an approx EBV of 100 mL/kg, when allowing for physiological increase. A pregnant woman weighing 70 kg at delivery would have an EBV of 7,000 mL; a loss of 20 % blood volume would be approximately 1,500 mL and a loss of 40 %, approximately 2,800 mL, and will require rapid access to red cell transfusion [40].

Many women will be able to tolerate an estimated blood loss between 500 and 1,500 mL without requiring blood component transfusion, provided they have had adequate fluid resuscitation. Consideration should be given to the patient’s underlying condition (e.g. respiratory/cardiac status), body weight & baseline hemoglobin. Reliable estimation of blood loss is extremely difficult and blood loss is frequently underestimated; if a woman is showing clinical signs of shock during massive obstetric haemorrhage (MOH), with ongoing blood loss, the MOH protocol should be activated to mobilise additional resources. Many institutions define the trigger for activation of an MOH protocol as an estimated blood loss of between 1,000 and 1,500 mL [113].

Successful management of PPH will require the involvement of clinicians, departments and personnel not present at the site of hemorrhage: