Placenta previa: Management

INTRODUCTION — Placenta previa refers to the presence of placental tissue that extends over the internal cervical os. Because this can lead to severe antepartum, intrapartum, and/or postpartum bleeding, placenta previa is associated with high risks for preterm birth and maternal and fetal/neonatal morbidity.

The management of pregnancies complicated by placenta previa is best addressed in terms of the patient's clinical setting:

●Asymptomatic

●Active antepartum bleeding

●Stable after one or more episodes of active antepartum bleeding

●Postpartum hemorrhage

This topic will discuss the management of these patients. The epidemiology, clinical features, diagnosis, morbidity, and mortality of placenta previa are reviewed separately. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality".)

ASYMPTOMATIC PATIENTS

Goals — The main goals during management of asymptomatic patients are to:

●Determine whether the previa resolves with increasing gestational age

●Determine whether the placenta is also morbidly adherent (placenta accreta spectrum)

●Reduce the risk of bleeding

●Determine the optimal time for planned cesarean birth if the previa persists

Monitoring placental position — Follow-up transvaginal ultrasound examination of placental position is indicated when the placenta covers or is <2 cm from the internal os on a second-trimester ultrasound examination. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality", section on 'Asymptomatic finding on midtrimester ultrasound examination'.)

We agree with the consensus approach of an expert group for monitoring placental position of these pregnancies (algorithm 1) [1]:

●If the placental edge covers or is <2 cm from the internal os in the second trimester, follow-up transvaginal ultrasonography for placental position is indicated at 32 weeks of gestation.

●At the 32-week follow-up examination:

•If the placental edge is ≥2 cm from the internal os, the placental position is reported as normal and additional follow-up ultrasound examinations for placental position are not indicated. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality", section on 'Asymptomatic finding on midtrimester ultrasound examination'.)

Color and pulsed Doppler examinations are useful to confirm the position of the placental edge and rule out vasa previa, which may result from resolution of a low-lying placenta. (See "Velamentous umbilical cord insertion and vasa previa".)

•If the placental edge covers or is <2 cm from the internal os, a follow-up transvaginal ultrasound for placental position is indicated at 36 weeks.

●At the 36-week follow-up examination:

•If the placental edge covers the internal os, cesarean birth is scheduled, as the previa is likely to persist until the time of birth. (See 'Timing of delivery' below.)

•If the placental edge does not cover but is <2 cm from the internal os, the risks and benefits of a trial of labor should be discussed with the patient. The risk of bleeding increases as the distance between the placental edge and internal os decreases and if vasa previa is present. (See 'Timing of delivery' below.)

Excluding placenta accreta spectrum — The possibility of placenta accreta spectrum (PAS) should be excluded in all patients with placenta previa, given the strong association between the two disorders. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality", section on 'Placenta previa-accreta spectrum' and "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences", section on 'Prenatal diagnosis'.)

If present, antepartum management of PAS is the same as for placenta previa, but delivery risks are substantially greater. Cesarean birth is scheduled earlier in gestation (34+0 to 35+6 weeks of gestation) than for previa alone, and preoperative preparation includes planning for cesarean-hysterectomy (which is usually required) and interventions that will reduce the risk of massive hemorrhage (which is more common than with previa alone). (See "Placenta accreta spectrum: Management".)

Reducing the risk of bleeding — For an individual patient, it is not possible to accurately predict whether spontaneous bleeding will occur, the gestational age when it will start, the volume, or the duration and frequency of bleeding events. Sonographic features reported to be associated with a higher likelihood of antepartum bleeding are described separately. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality", section on 'Bleeding'.)

Interventions have been proposed to reduce the risk of bleeding:

●There is consensus that digital cervical examination should be avoided. It is clear from anecdotal experience that palpation of a placenta previa through a partially dilated cervix can result in severe hemorrhage.

●We advise patients with placenta previa after 20 weeks of gestation (earlier if they have experienced vaginal bleeding) to avoid any sexual activity that may lead to orgasm. The rationale is that this activity, especially if orgasm occurs, may be associated with transient uterine contractions, which, in turn, may provoke bleeding. Additionally, there is concern that vaginal intercourse (or putting any object deep into the vagina) might cause direct trauma to the previa, resulting in bleeding. There are no published studies that either support or refute this advice. It is extrapolated from observations that palpation of the previa through a partially dilated cervix can result in severe hemorrhage.

●We also advise patients with placenta previa after 20 weeks to avoid moderate and strenuous exercise, heavy lifting (eg, more than approximately 20 pounds), or standing for prolonged periods of time (eg, >4 hours). In the general obstetric population, this degree of activity was linked to small but statistically significant increases in preterm birth in a meta-analysis of observational studies (median odds ratios 1.10 to 1.20) [2].

In addition, we advise patients to seek immediate medical attention if contractions or vaginal bleeding occur, given the potential for severe bleeding and need for emergency cesarean birth.

Inpatient versus outpatient monitoring — Whether asymptomatic patients benefit from hospitalization is unclear. Findings from observational studies suggest that patients who have not experienced any antepartum bleeding are at low risk of sudden hemorrhage requiring an emergency cesarean birth for control of bleeding [3-6]. Therefore, it is likely that most of these patients can be managed on an outpatient basis until vaginal bleeding occurs or until admission for scheduled cesarean birth.

However, patient-specific risk factors need to be taken into account. Such factors may include [7-9]:

●Short cervical length on transvaginal ultrasound examination (eg, there is an increased risk for preterm labor if cervical length is ≤25 mm, and we consider hospitalization if ≤15 mm)

●Rapid cervical shortening (eg, >10 mm over a one- to two-week period on transvaginal ultrasound)

●Inability to get to the hospital promptly (within approximately 20 minutes), and lack of home support in case of an emergency

These parameters, which are based on personal clinical experience/expert opinion and data from studies on risk of preterm birth across the spectrum of cervical length, represent our approach and should not be considered a standard of care. The Society for Maternal-Fetal Medicine recommends not routinely performing cervical length screening in the late preterm period in patients with placenta previa due to a lack of data on an appropriate management strategy [10].

Antenatal corticosteroids — For patients who have not received a course of antenatal corticosteroids for standard obstetric indications at some point during the pregnancy, we administer a course 48 hours before a planned cesarean birth scheduled for <37+0 weeks of gestation. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery", section on 'Candidates for a first ACS course by gestational age'.)

Timing of delivery — We perform a cesarean birth at 36+0 to 37+6 weeks of gestation in pregnancies with uncomplicated placenta previa, in agreement with recommendations of the American College of Obstetricians and Gynecologists and the Society for Maternal-Fetal Medicine [10,11]. Uncomplicated placenta previa is defined as without fetal growth restriction, superimposed preeclampsia, or other issues that take precedent for delivery decision-making.

Review of available evidence suggests that the maternal-fetal risks associated with continuing the pregnancy (severe bleeding, emergency unscheduled delivery) are greater than the neonatal risks associated with preterm birth at this gestational age range. (See 'Cesarean birth' below.)

ACUTE CARE OF BLEEDING PATIENTS — An actively bleeding placenta previa is a potential obstetric emergency. These patients should be admitted to the Labor and Delivery Unit for maternal and fetal monitoring, and the anesthesia team should be notified.

Goals — The major goals in managing a patient with an acutely bleeding placenta previa are to:

●Achieve and/or maintain maternal hemodynamic stability

●Determine if emergency cesarean birth is indicated

Maternal and fetal assessment

●Maternal blood pressure, heart rate, respiratory rate, peripheral oxygen saturation, and urine output are closely monitored. Tachypnea, tachycardia, hypotension, low oxygen saturation, and air hunger are signs of hypovolemia (table 1).

●The fetal heart rate is monitored continuously for patterns suggestive of hypoxemia or anemia. (See "Intrapartum category I, II, and III fetal heart rate tracings: Management".)

●Blood loss is quantified. The accuracy of blood loss assessment is improved with quantitative measurement techniques [12]. Accurate estimation of vaginal blood loss is difficult to determine visually, particularly when blood is partially saturating or soaking towels, maternity pads, or gauze sponges or dripping onto the floor [13,14]. One or more of the following techniques can be used to quantify blood loss [14,15]:

•Collect blood in graduated volumetric containers.

•Measure the total weight of bloody materials and subtract the known weight of the same materials when dry. The difference in weight between wet and dry in grams approximates the volume of blood in milliliters.

•Attempt to account for fluids other than blood (eg, amniotic fluid, irrigation fluid, urine) that are collected or absorbed.

•Use visual aids that correlate the size and appearance of blood on specific surfaces (eg, maternity pad, emesis basin, bed sheet, lap sponge) with the volume of blood absorbed by that surface (picture 1). Regularly scheduling standardized training in the use of these charts can be helpful for this assessment.

Although the data do not support any one method of quantifying blood loss as superior to another, quantification of blood loss, such as using graduated drapes or weighing, provides a more accurate assessment of actual blood loss than visual estimation; however, the effectiveness of quantitative blood loss measurement on clinical outcomes has not been demonstrated [16-18]. Successful implementation of an obstetric hemorrhage bundle has been associated with improved outcome measures related to obstetric hemorrhage [19].

Laboratory testing and notifying the blood bank — There is no consensus about the components of routine laboratory assessment of patients with bleeding placenta previa [20,21]. We do the following:

●Obtain a complete blood count

●Send blood sample for type and antibody screen and notify the blood bank that a patient with bleeding placenta previa has been admitted. We cross-match two to four units of packed red blood cells when bleeding is heavy or increasing, delivery is likely for any reason, or difficulty in procuring compatible blood is anticipated. If massive transfusion might be needed, the blood bank is notified of that as well and a massive transfusion protocol is initiated, if available.

●Evaluate for coagulopathy (fibrinogen level, activated partial thromboplastin time, prothrombin time) in patients suspected of coexistent abruption or with severe blood loss resulting in hemodynamic instability. If available, thromboelastography of whole blood at the point of care synthesizes information obtained from multiple coagulation tests into a single readout providing information regarding clot initiation, clot strength, and fibrinolysis simultaneously. (See "Etiology and diagnosis of coagulopathy in trauma patients", section on 'Viscoelastic hemostatic assays'.)

Prolonged oozing from needle puncture sites also suggests coagulopathy. A crude test for identifying patients with impaired clotting function can be performed at the bedside using blood in a red top tube (clot observation test) (see "Disseminated intravascular coagulation (DIC) during pregnancy: Clinical findings, etiology, and diagnosis", section on 'Diagnostic evaluation').

Fetal bleeding can occur if disruption of fetal vessels in placental villi, vasa previa, or a velamentous cord occurs, and can be detected by performing a Kleihauer-Betke or flow cytometry test on a specimen of vaginal blood. However, fetal bleeding from disruption of one of these sources is rare and typically results in fetal demise or a nonreassuring fetal heart rate tracing necessitating emergency delivery. Therefore, it is unlikely that results of the test will impact management. (See "Spontaneous massive fetomaternal hemorrhage", section on 'Kleihauer-Betke assay'.)

Stabilization

Intravenous access and crystalloid — One or two large bore intravenous lines are inserted and crystalloid (Lactated Ringer solution or normal saline) is infused to achieve/maintain hemodynamic stability and adequate urine output (at least 30 mL/hour). (See "Treatment of severe hypovolemia or hypovolemic shock in adults".)

Transfusion — Transfusion of blood products in patients with active bleeding should be guided by the volume of blood loss over time, changes in hemodynamic parameters (eg, blood pressure, maternal and fetal heart rates, peripheral perfusion, and urine output), coagulation studies, and hemoglobin level. Not getting behind in replacement of blood products is essential. Acute hemorrhage may not be associated with an immediate reduction in either blood pressure or hematocrit in an otherwise healthy young patient. Thus, we have a low threshold for transfusing symptomatic patients. We believe a failure to rapidly correct tachycardia or hypotension with a normal saline bolus or a hemoglobin value <10 g/dL should prompt immediate transfusion. Types and actions of blood replacement products are shown in the table (table 2).

Our general approach is as follows. We do not administer tocolytic drugs to actively bleeding patients.

●Initially, we suggest transfusing two to four units of typed and crossed packed RBC, without fresh frozen plasma or platelets as long as the fibrinogen level is >250 mg/dL and the platelet count is >100,000/microL. Our goal is a final hemoglobin value >10 g/dL.

●If the patient fails to stabilize, we initiate a massive transfusion protocol (algorithm 2). If a massive transfusion protocol is not available, type O RhD-negative blood should be administered until type-specific or typed and cross-matched blood is available.

●If the patient continues to bleed, we suggest using the same blood product transfusion ratios used for patients with severe hemorrhage of other etiologies: a 1:1:1 ratio of packed RBC:fresh frozen plasma:platelets. If available, thromboelastography (TEG) or rotational thromboelastometry (ROTEM)-based transfusion algorithms may be helpful. (See "Intraoperative transfusion and administration of clotting factors".)

●If there is no imminent need for delivery, we continue transfusion until the patient has stabilized, bleeding is decreased, and hemoglobin is at least 10 g/dL. We choose this hemoglobin threshold to provide a margin of safety since the patient is at increased risk for another, more severe bleeding event. However, if there is an imminent need for delivery, a preoperative or intraoperative target hemoglobin of about 8 g/dL is reasonable.

●Tranexamic acid is generally not administered before delivery because it freely crosses the placenta. However, it has been recommended for treatment of antepartum and intrapartum bleeding related to several inherited bleeding disorders [22] and one UpToDate contributor sometimes administers it prophylactically at the start of cesarean delivery. Fetal/neonatal harm has not been reported, but data are limited. It is commonly used as a component of treatment of postpartum hemorrhage but prophylactic use is less common. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer tranexamic acid' and "Management of the third stage of labor: Prophylactic pharmacotherapy to minimize hemorrhage", section on 'Tranexamic acid'.)

In low resource settings, non-pneumatic anti-shock garments have been used to restore adequate blood pressure in pregnant/postpartum patients who are hemodynamically unstable due to severe bleeding [23-25]. These devices have not been used when the fetus was viable and there is no information on their effect on uteroplacental blood flow and the fetus. An unfavorable effect is likely since in postpartum patients these garments force blood to the essential organs, such as the heart, lungs, and brain, and significantly increase the resistive index of the internal iliac artery, which is responsible for supplying the majority of uterine blood flow [26].

Outcome — Many patients who initially present with symptomatic placenta previa respond to supportive therapy, as described above, and do not require immediate delivery [27-31]. In observational series, 50 percent of patients with a symptomatic previa (any amount of bleeding) were not delivered for at least four weeks [28,30,31]. Even a large bleed does not preclude conservative management. In one series, 50 percent of patients whose initial hemorrhagic episode exceeded 500 mL were successfully managed with aggressive use of antepartum transfusions and had a mean prolongation of pregnancy of 17 days [27].

Indications for delivery — Cesarean birth is indicated for:

●Active labor.

●A category III fetal heart rate tracing unresponsive to resuscitative measures. (See "Intrapartum category I, II, and III fetal heart rate tracings: Management", section on 'General approach'.)

●Severe and persistent vaginal bleeding such that maternal hemodynamic stability cannot be achieved or maintained.

●Significant vaginal bleeding after 34+0 weeks of gestation – When the clinician believes the volume of vaginal bleeding is significant, we believe the balance of fetal benefit versus maternal risk favors delivery in pregnancies ≥34+0 weeks because the neonatal benefits from avoiding preterm birth decrease with advancing gestational age and the maternal risks from persistent or recurrent bleeding probably increase. However, the exact gestational age threshold and amount of bleeding considered significant are matters of clinician judgment. The decision to deliver these pregnancies is made on a case-by-case basis while observing the patient's course on the labor unit. Delivery should not be delayed to administer antenatal corticosteroids [10].

Magnesium sulfate — Magnesium sulfate therapy for neuroprotection is a common practice for pregnancies <32 weeks in which the family opts for neonatal interventions and a decision has been made to deliver within 24 hours but not emergently. Emergency delivery because of maternal or fetal status should not be delayed to administer magnesium sulfate. (See "Neuroprotective effects of in utero exposure to magnesium sulfate".)

Anesthesia — For scheduled cesarean births, neuraxial anesthesia is generally the preferred approach. For patients who are actively bleeding or require an emergency cesarean birth, general anesthesia is usually preferred. (See "Anesthesia for the patient with peripartum hemorrhage", section on 'Placenta previa'.)

EXPECTANT MANAGEMENT OF STABLE PATIENTS AFTER A BLEED

Candidates and goals — We believe that symptomatic patients <34+0 weeks of gestation who are hemodynamically stable or quickly stabilized and have a normal fetal heart rate pattern are candidates for expectant management. The goal is to prolong pregnancy to enable further fetal growth and maturation, without placing the mother at excessive risk from persistent or recurrent bleeding.

Management of placenta previa after acute bleeding is based upon findings from observational studies and clinical experience. A systematic review that attempted to assess the impact of clinical interventions in these pregnancies concluded there were insufficient data upon which to make evidence-based recommendations for clinical practice; only three randomized trials involving a total of 114 participants were identified [32].

After the patient has been stabilized, we take the approach described in the following sections.

Antenatal corticosteroids — A course of antenatal corticosteroid therapy is administered to patients who experience bleeding. Administration can be considered as early as 22+0 to 22+6 weeks of gestation for a patient in whom delivery in the next seven days is anticipated and if the patient is requesting aggressive neonatal intervention. We would give an initial course of steroids to patients whose first bleed is as late as 34+0 to 36+6 weeks of gestation, but the benefits at this gestational age are less clear and use is controversial. Use of antenatal corticosteroids, including rescue (salvage, booster) dosing, is reviewed separately. (See "Antenatal corticosteroid therapy for reduction of neonatal respiratory morbidity and mortality from preterm delivery", section on '34+0 or more weeks'.)

Correction of anemia — Oral or parenteral iron supplementation may be needed for optimal correction of anemia. Parenteral iron therapy has the advantages of a faster rise in hemoglobin levels and less gastric upset compared with oral therapy. If oral iron is prescribed, stool softeners and a high-fiber diet help to minimize constipation and avoid excess straining that might precipitate bleeding. (See "Anemia in pregnancy", section on 'Treatment of iron deficiency'.)

Anti-D immune globulin — Theoretically, disruption of the fetomaternal placental interface can result in fetomaternal bleeding. For this reason, guidelines for prevention of RhD alloimmunization suggest administering anti-D-immune globulin to RhD-negative patients who have bleeding from placenta previa [33,34].

Readministration is not necessary if rebleeding occurs within three weeks of administration. If repeated episodes of bleeding occur, the anti-D antibody titer can be checked: a low titer suggests that the anti-D level is sufficient to provide ongoing protection against alloimmunization.

Readministration is also not necessary if delivery occurs within three weeks of administration, unless routine postpartum Kleihauer-Betke or flow cytometry detects a large fetomaternal bleed that would not be covered by standard doses of anti-D immune globulin. Use of anti-D immune globulin is reviewed in detail separately. (See "RhD alloimmunization: Prevention in pregnant and postpartum patients".)

Fetal assessment — As discussed above, active vaginal bleeding is an indication for continuous fetal monitoring. (See 'Maternal and fetal assessment' above.)

There is no proven value of nonstress testing or performing a biophysical profile in a patient with an asymptomatic placenta previa who has no evidence of uteroplacental insufficiency (eg, preeclampsia, fetal growth restriction, oligohydramnios) or another standard indication for antepartum fetal assessment.

In pregnancies with coexistent vasa previa, we perform nonstress testing to look for any evidence of cord compression. (See "Velamentous umbilical cord insertion and vasa previa", section on 'Management'.)

Autologous blood donation — Some patients may consider autologous blood donation, given the high frequency of bleeding requiring blood transfusion (12 percent in one series [35] and 22 percent in another [36]). However, most pregnant patients, especially those who have bled from a placenta previa, will not meet standard criteria for autologous donation [37,38]. Furthermore, autologous donation is not cost-effective, given the increasing safety of allogeneic blood and the wastage of unused autologous blood. These and other issues related to preoperative autologous blood donation are discussed in detail separately. (See "Surgical blood conservation: Preoperative autologous blood donation".)

Other interventions

Tocolysis — We do not routinely use tocolytic drugs in the management of placenta previa, given the lack of proven benefits and the known possible harms (see "Inhibition of acute preterm labor"). In patients with contractions, we may use tocolytics (usually nifedipine) while administering a course of antenatal corticosteroids if bleeding is diminishing or has ceased and delivery is not otherwise mandated by the maternal or fetal condition. In such patients, uterine contractions may mechanically promote placental separation and bleeding, and thus, tocolysis may reduce or eliminate a potential contributing factor. Indomethacin is not used due to its inhibitory effect on platelet function and potential adverse fetal/neonatal effects. (See "Inhibition of acute preterm labor", section on 'Fetal side effects'.)

Some observational studies in patients with symptomatic placenta previa suggest tocolysis may prolong pregnancy and result in an increase in birth weight without causing adverse effects on the mother or fetus [39,40]. However, it is likely that underlying differences in the treated and untreated (control) patients accounted for this benefit. Furthermore, these studies have generally not shown a decrease in the number of episodes of hemorrhage after admission, the total amount of blood loss, or the number of blood transfusions. Other studies have reported that maintenance tocolysis in patients with placenta previa is not beneficial [41].

Cerclage — In the absence of high-quality evidence of efficacy and safety, we advise not performing cerclage to improve pregnancy outcome in placenta previa. However, the presence of a stable placenta previa is not a contraindication to cerclage placement when indicated for cervical insufficiency. (See "Cervical insufficiency".)

Cervical cerclage has been placed in an attempt to minimize early widening of the isthmus, which is thought to promote placental separation [42]. The efficacy of this approach is unproven. Although a meta-analysis of two small randomized trials that evaluated cerclage for improving pregnancy outcome in primarily symptomatic placenta previa [43,44] reported that cervical cerclage reduced the risk of delivery before 34 weeks (relative risk [RR] 0.45, 95% CI 0.23-0.87) and the birth of an infant weighing less than 2000 g (RR 0.34, 95% CI 0.14-0.83), the lack of consistency between trials and methodologic issues prevent making a clear conclusion of benefit [32].

Pessary and progesterone — Use of pessary and progesterone showed promising results (less bleeding and preterm birth <34 weeks) in a randomized trial of patients with placenta previa and high risk of preterm birth [45]. A smaller trial using a pessary alone showed promising results but was stopped early due to budgetary issues [46]. More data are needed before we would consider managing patients with this investigational approach.

Inpatient versus outpatient management

●Patients with one or two bleeding episodes – If the patient's bleeding stops after the first or second bleeding event, we discharge them to home if they meet the criteria described below. (See 'Criteria for discharge' below.)

●Patients with three or more bleeding episodes – If a third bleeding episode occurs, we generally hospitalize the patient until delivery [30]. Since the frequency and severity of recurrent bleeding episodes are unpredictable, maintaining close proximity to the labor and delivery unit may minimize the risk of serious maternal or fetal complications by enabling prompt access to transfusion therapy and emergency cesarean birth when needed.

One study found that the risk for emergency cesarean birth progressively increased with one, two, and three or more episodes of antepartum bleeding (odds ratio [OR] 7.5, 14, and 27, respectively) and in patients who had had a blood transfusion (OR 6.4) compared with patients with no antepartum bleeding [47]. In this study, 57 percent of patients with bleeding placenta previa had an emergency delivery, and the frequency of emergency delivery was 42 percent after three or more bleeding episodes versus 25 to 30 percent after one or two bleeds.

The only randomized trial of outpatient versus inpatient management of patients with placenta previa after resolution of the initial bleeding episode found that outpatient care was not associated with greater morbidity than inpatient management [30]. Patients randomly assigned to the outpatient arm who had a recurrent bleed were treated initially as inpatients and then discharged home after a minimum of 48 to 72 hours if they remained stable with minimal ambulation. Outpatients were instructed to maintain bedrest. However, if these patients had a third episode of bleeding, they were hospitalized until delivery. Significant differences in outcome may not have been appreciated given the small number of participants (53) who participated in this trial.

Criteria for discharge — We discharge selected patients with placenta previa in whom bleeding has stopped for a minimum of 24 hours and who have no other pregnancy complications, although the safety and efficacy of this approach has not been established [30,48-50]. In our practice, candidates for outpatient care should:

●Be able to return to the hospital within 20 minutes [51].

●Be reliable (ie, will comply with instructions about sexual activity, etc). (See 'Reducing the risk of bleeding' above.)

●Be able to maintain modified bed rest at home.

●Understand the risks entailed by outpatient management.

●Have an adult companion available 24 hours/day who can immediately transport them to the hospital if there is light bleeding or call an ambulance for severe bleeding.

Management of coexistent PPROM — Antepartum decidual hemorrhage is a major risk factor for preterm prelabor rupture of membranes (PPROM), which can occur despite the presence of a complete placenta previa. In these cases, each condition is managed independently. (See "Preterm prelabor rupture of membranes: Clinical manifestations and diagnosis".)

Timing of delivery — As discussed above, planned cesarean birth of patients with stable (no bleeding or minimal bleeding) placenta previa should be accomplished at 36+0 to 37+6 weeks. (See 'Timing of delivery' above.)

Expectant management is terminated, and emergency cesarean birth is indicated if any of the following occur:

●Labor

●Any vaginal bleeding with a category III fetal heart rate tracing unresponsive to resuscitative measures

●Severe and persistent vaginal bleeding such that maternal hemodynamic stability cannot be achieved or maintained

●Significant vaginal bleeding at ≥34+0 weeks of gestation (see 'Indications for delivery' above)

CESAREAN BIRTH — A cesarean birth is always indicated when there is sonographic evidence of placenta previa and the fetus has a reasonable chance of extrauterine survival. As long as the mother remains hemodynamically stable, vaginal birth may be considered in rare circumstances, such as in the presence of a fetal demise or fetus without a reasonable chance of extrauterine survival.

Preparation — Planning for the possibility of postpartum hemorrhage is critical for reducing morbidity. (See "Overview of postpartum hemorrhage", section on 'Institutional planning and preparation' and "Overview of postpartum hemorrhage", section on 'Early recognition, assessment, and intervention'.)

Two to four units of packed red blood cells should be available for the cesarean birth. Appropriate surgical expertise and instruments for performance of a cesarean hysterectomy should also be available since these patients are at increased risk of placenta accreta spectrum, even in the absence of a prior cesarean birth. Evaluation for placenta previa-accreta spectrum should have been performed antenatally, with appropriate preparations for management, if present. (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences" and "Placenta accreta spectrum: Management" and "Peripartum hysterectomy for management of hemorrhage" and "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality", section on 'Placenta previa-accreta spectrum'.)

Management of the placenta — The surgeon should try to avoid disrupting the placenta when entering the uterus. If the placenta is incised, hemorrhage from fetal vessels can result in significant neonatal anemia; therefore, the fetus should be delivered rapidly, and the cord promptly clamped.

Preoperative or intraoperative sonographic placental localization is helpful in determining the position for the hysterotomy incision. For intraoperative imaging, the transducer is placed in a sterile plastic bag and sleeve, and sterile gel is applied to the uterine serosa; the best incision site for avoiding the placenta can then be mapped sonographically.

If the placenta is in an anterolateral location, a vertical incision can be made in the lower uterine segment on the opposite anterolateral side from the placenta. If the placenta wraps around the cervix from the anterior to posterior lower uterine segment in the midline, a transverse or vertical incision above it may be possible [52,53], although a vertical incision often results in extension into the upper uterine segment. Anterior previas are associated with a larger volume of bleeding at cesarean than posterior previas [54-57].

Management of postpartum hemorrhage — Oxytocin is given routinely to reduce the risk of postpartum hemorrhage. The addition of second uterotonic drug or tranexamic acid can be considered as well (see "Management of the third stage of labor: Prophylactic pharmacotherapy to minimize hemorrhage", section on 'Active management'). Nevertheless, severe bleeding may occur from the placental bed after placental expulsion: in a systematic review, 16 to 29 percent of patients with a placenta previa had a postpartum hemorrhage [58]. The reason for the increased risk of postpartum hemorrhage is thought to be that the myometrium of the lower uterine segment does not contract as effectively as the upper uterine segment, and thus may impede physiologic hemostasis from a lower segment placental bed. In some cases, hemorrhage is due to focal placenta accreta. Even in the absence of the accreta spectrum, patients with a previous cesarean may lack sufficient lower uterine segment contractility in the area of the scar to stop bleeding.

Postpartum hemorrhage is approached in the following ways.

Step one: Administer uterotonic drugs and tranexamic acid; consider temporizing maneuvers

●Pharmacotherapy – Oxytocin is administered and titrated, as needed, to control heavy bleeding. If oxytocin alone does not rapidly control hemorrhage, we administer tranexamic acid concomitantly with additional uterotonic drugs (carboprost tromethamine [prostaglandin F₂ alpha] or methylergonovine). Some providers administer tranexamic acid prophylactically rather than therapeutically. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer tranexamic acid' and "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Manage atony' and "Management of the third stage of labor: Prophylactic pharmacotherapy to minimize hemorrhage", section on 'Tranexamic acid'.)

●Tourniquets – Tourniquets may be useful as a temporizing measure while pharmacotherapy is administered and taking effect or while surgical interventions are being considered [59,60]. A bladder catheter or Penrose drain is tied tightly around the uterus as low as possible to occlude the uterine vessels in the broad ligaments and then secured with a clamp (figure 1). A second or third tourniquet can also be applied, as needed. (See "Techniques to reduce blood loss during abdominal or laparoscopic myomectomy", section on 'Tourniquets'.)

Step two: Treat focal bleeding, if present — The following options may control bleeding from a small focal area:

●Hemostatic square sutures

●Placement of fibrin glue or patch over area of oozing to promote clotting [61]

●Application of ferric subsulfate (Monsel's solution) to oozing area

●Placental site injection of vasopressin

●Excision, if the area is small and easily accessible, particularly in cases of focal placenta accreta with persistent bleeding

Hemostatic square sutures — Ligation of myometrial vessels at the placental site will control focal bleeding in patients who have a focal placenta accreta and some patients who responded poorly to intravenous uterotonic therapy.

The Affronti endouterine hemostatic square suture technique involves making four to six 2 by 2 cm squares in the area of placental bed bleeding [62]. The 1.0 polyglactin 910 sutures should penetrate the decidua and extend into the myometrium but not beyond the uterine serosa (figure 2). The ends of the sutures are tied down tightly to compress the enclosed vessels.

In one study, multiple square sutures stopped postpartum hemorrhage in 28 of 30 cases (93 percent) [63]. Twenty patients subsequently underwent hysteroscopy: eight (40 percent) had no intrauterine adhesions, nine (45 percent) had thin adhesions that were removed easily by the tip of the hysteroscope, two (10 percent) had moderate intrauterine adhesions that were resected, and one (10 percent) had severe intrauterine adhesions.

Placental site injection of vasopressin — The authors have not used this technique in patients with placenta previa. Placental site injection of vasopressin to reduce bleeding, while biologically plausible and potentially clinically relevant, is based on two small case series. After removal of the placenta, subendometrial injection of vasopressin at the placental implantation site may reduce bleeding [64,65]. The favorable effect has been attributed to binding to the vasopressin V1α receptor, which is highly expressed in smooth muscle cells in the lower segment of the uterus. Intravascular injection should be avoided, as it can cause severe adverse cardiovascular effects (bradycardia, cardiac arrhythmia, ischemia, right heart failure, shock, cardiac arrest, limb ischemia).

In one study, local injection of 4 units of vasopressin in 20 mL of saline into the placental implantation site significantly reduced blood loss without increasing morbidity (blood loss: with vasopressin 1149 mL, without vasopressin 1634 mL) [65]. In a case report, 5 units of vasopressin in 20 mL saline injected in 1 to 2 mL amounts into the area of placental implantation stopped bleeding within 90 seconds [64].

Step three: Ligation of the uterine and utero-ovarian arteries (O'Leary stitch) — Ligation of the uterine and utero-ovarian arteries (O'Leary stitch) can decrease diffuse uterine bleeding by reducing perfusion pressure in the myometrium. The technique, which can be applied rapidly, is described separately. (See "Postpartum hemorrhage: Management approaches requiring laparotomy", section on 'Laceration of the uterine artery or utero-ovarian artery branches'.)

Step four: Intrauterine balloon tamponade and/or uterine compression sutures — If bleeding persists, the next step is either intrauterine balloon tamponade or placement of uterine compression sutures. As no trials have compared the two approaches directly in this setting, the choice depends on the clinical judgment/preference of the surgeon.

Uterine compression sutures may be more effective for atony and fundal bleeding, whereas the balloon may be more effective for lower segment bleeding. However, an advantage of placing the balloon first is that it is a quick and easy procedure and if it does not work, the balloon can be deflated, compression sutures can be placed, and then the balloon can be reinflated, if needed. If the compression sutures are placed first, then they will have to be removed in order to place a balloon.

Intrauterine balloon tamponade — Intrauterine balloon tamponade is a fast and often effective procedure for control of hemorrhage in patients who have not responded to drug therapy or focal placental site suture ligation. A meta-analysis found that, after placement of an intrauterine balloon, 87 percent of patients with previa had arrest of bleeding without maternal death and additional surgical or radiologic interventions [66]. In a small trial in which 13 patients with intractable bleeding from placenta previa were randomly assigned to insertion of a balloon or hemostatic square sutures, the balloon group had greater reductions in operative time and intraoperative blood loss (time: 63 versus 78 minutes; intraoperative blood loss: 1520 versus 1946 mL) [67]. Routine prophylactic use of a balloon has also been shown to be effective in controlling blood loss during and after cesarean birth in patients with placenta previa [68].

The stem of a balloon catheter can be passes through the uterine incision, through the cervical os, and then into the vagina; an assistant then pulls the end of the catheter out of the introitus. If it is difficult to pass the catheter through the cervical os, an assistant can pass the tip of a small forceps from the vagina through the cervical os to the uterine incision [69]. The surgeon then places the stem of the catheter into the open tip so the assistant can pull it into the vagina while the surgeon places the balloon end in an appropriate place within the uterine cavity. The balloon can be left deflated or inflated with 50 to 100 mL sterile saline to reduce the risk of puncture when the hysterotomy incision is closed. After the uterine incision is closed, the assistant inflates the balloon to its maximum volume with sterile fluid (table 3) while the surgeon inspects the uterus from above. (See "Postpartum hemorrhage: Use of an intrauterine hemorrhage-control device".)

An intrauterine vacuum-induced hemorrhage-control device has been used to manage postpartum hemorrhage caused by atony, with minimal experience in patients with bleeding related to abnormal placentation. (See "Postpartum hemorrhage: Use of an intrauterine hemorrhage-control device", section on 'Intrauterine vacuum'.)

Uterine compression sutures — If balloon tamponade is ineffective, the balloon can be deflated and a B-Lynch uterine compression suture is applied (figure 3). The hysterotomy site is then closed, the uterus replaced in the pelvis (if previously exteriorized), and the balloon is reinflated, thereby applying pressure to both the outer and inner surfaces of the myometrium. [69,70]. It is important to observe the myometrium and stop instilling fluid before undue blanching occurs at the compression suture sites, which might lead to rupture or uterine necrosis. It is also important to observe for vaginal bleeding to see if hemorrhage has been controlled. (See "Postpartum hemorrhage: Use of an intrauterine hemorrhage-control device".)

Refractory patients

Consider arterial embolization — When the above measures have failed, uterine or hypogastric artery embolization in an operating room with the full surgical team in attendance is an option if the facility has a hybrid operating room, or an operating room that allows simultaneous surgery and embolization (an appropriately sensitive portable C-arm and carbon fiber table). We believe embolization is preferable to surgical internal iliac artery ligation. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Consider uterine or hypogastric artery embolization'.)

Hysterectomy — Hysterectomy is a definitive treatment of uterine bleeding when fertility preserving procedures have not reduced the bleeding to a manageable level. Ideally, it should be performed before severe hypovolemia, tissue hypoxia, hypothermia, electrolyte abnormalities, and acidosis have developed, which further compromise the patient's status. (See "Peripartum hysterectomy for management of hemorrhage".)

DELIVERY OF PATIENTS WITH A LOW-LYING PLACENTA — The optimal route for delivery of pregnancies where the distance between the placental edge and internal os is 0 to 20 mm is debatable. In these cases, the fetal head may tamponade the adjacent placenta, thus preventing hemorrhage.

Although a variety of factors influence the decision to perform cesarean birth, the following data from a meta-analysis support allowing a trial of labor in pregnancies in which the placenta is more than 10 mm from the internal os. If this distance is ≤10 mm, these pregnancies are at high risk of intrapartum hemorrhage necessitating emergency cesarean birth; therefore, we suggest scheduled cesarean birth at 39 weeks to minimize this risk. However, this is a shared decision, and some patients may choose to undertake a trial of labor since a substantial proportion will be able to successfully deliver vaginally. If a trial of labor with a low-lying placenta is attempted, the facility should have the ability to perform an emergency cesarean birth.

A systematic review of 10 studies of nearly 600 patients with a low-lying placenta undergoing a trial of labor reported outcomes by the distance between the placental edge to the internal os [71]:

●0 to 10 mm – vaginal delivery: 43 percent (95% CI 28-59), emergency cesarean: 45 percent (95% CI 22–69)

●11 to 20 mm – vaginal delivery: 85 percent (95% CI 70-96), emergency cesarean: 14 percent (95% CI 4.2–29)

●>20 mm – vaginal delivery: 82 percent (95% CI 58-97), emergency cesarean: 10 percent (95% CI 2.2–22.3)

There was no difference in blood transfusion or postpartum hemorrhage among groups.

PREGNANCY TERMINATION IN WOMEN WITH PLACENTA PREVIA — The presence of a placenta previa does not preclude second-trimester pregnancy termination by standard techniques, although data are limited to a few studies [72-77]. The approach to termination in this clinical setting is described separately. (See "Overview of second-trimester pregnancy termination", section on 'Placental abnormalities'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Obstetric hemorrhage".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Placenta previa (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Digital cervical examination should be avoided. It can result in severe hemorrhage. (See 'Reducing the risk of bleeding' above.)

●Acute management of bleeding previa

•An actively bleeding placenta previa is a potential obstetric emergency. Patients with active bleeding are hospitalized for close maternal and fetal monitoring, including (see 'Maternal and fetal assessment' above and 'Laboratory testing and notifying the blood bank' above):

-Monitoring maternal blood loss and hemodynamic status.

-Obtaining a complete blood count and sending blood for type and antibody screen or cross-match, depending on the likelihood of transfusion.

-Evaluating the fibrinogen level, activated partial thromboplastin time, and prothrombin time in patients suspected of coexistent abruption or with heavy blood loss resulting in hemodynamic instability.

-Monitoring the fetal heart rate

•Not getting behind in replacement of blood products is essential. Acute hemorrhage may not be associated with an immediate reduction in either blood pressure or hematocrit in an otherwise healthy young patient (table 1). Thus, we have a low threshold for transfusing symptomatic patients. (See 'Transfusion' above.)

Initially, we transfuse two to four units of typed and crossed packed RBCs, without fresh frozen plasma or platelets as long as the fibrinogen level is >250 mg/dL and the platelet count is >100,000/microL. Our goal is a final hemoglobin value >10 g/dL.

If the patient continues to bleed, we use the same blood product transfusion ratios as for patients with severe hemorrhage of other etiologies: a 1:1:1 ratio of packed RBC:fresh frozen plasma:platelets. If available, thromboelastography (TEG) or rotational thromboelastometry (ROTEM)-based transfusion algorithms can be helpful.

If the patient fails to stabilize, we initiate a massive transfusion protocol (algorithm 2).

•Most patients who initially present with symptomatic placenta previa respond to supportive therapy. Indications for emergency cesarean birth include active labor, refractory life-threatening maternal hemorrhage, a category III fetal heart rate tracing, and significant vaginal bleeding at ≥34+0 weeks of gestation. (See 'Acute care of bleeding patients' above.)

●Expectant management after an acute bleed

•After a bleeding episode has resolved, outpatient management of select patients is reasonable. We counsel these patients to avoid excess physical activity, including sexual intercourse, and to call their provider promptly if bleeding or labor occurs. They should be able to return to the hospital quickly if rebleeding occurs and should not have additional pregnancy complications. (See 'Expectant management of stable patients after a bleed' above.)

•A course of antenatal corticosteroid therapy is administered. Anti-D immune globulin is administered to RhD-negative patients. (See 'Antenatal corticosteroids' above and 'Anti-D immune globulin' above.)

•We schedule cesarean birth at 36+0 to 37+6 weeks (see 'Timing of delivery' above). Incision of the placenta should be avoided, as this increases the risk of fetal hemorrhage. (See 'Cesarean birth' above.)

●Asymptomatic previa

•The main goals during management of asymptomatic patients with placenta previa are to:

-Determine whether the previa resolves with increasing gestational age. Follow-up transvaginal ultrasonography is performed at 32 weeks of gestation. If the placenta is morbidly adherent (placenta accreta spectrum), cesarean birth is planned for 34+0 to 35+6 weeks of gestation. If the placenta is over or <2 cm from the internal os but not morbidly adherent, transvaginal ultrasound is repeated at 36 weeks (algorithm 1). (See 'Monitoring placental position' above and 'Excluding placenta accreta spectrum' above.)

-Counsel the patient on measures to reduce the risk of bleeding. (See 'Reducing the risk of bleeding' above.)

•Cesarean birth is performed at 36+0 to 37+6 weeks of gestation. (See 'Timing of delivery' above.)

•We suggest a course of antenatal corticosteroids 48 hours before a cesarean birth scheduled before 37+0 weeks of gestation, if not previously given (Grade 2C). (See 'Antenatal corticosteroids' above.)

●Cesarean birth

•Planning for the possibility of postpartum hemorrhage, which occurs in up to an approximate 30 percent of cases, is critical for reducing morbidity. Disruption of the placenta at hysterotomy should be avoided, if possible. (See 'Preparation' above and 'Management of the placenta' above.)

•The management of postpartum hemorrhage involves a stepwise approach, initially with uterotonic drugs and tranexamic acid, followed by standard surgical interventions (eg, suturing focal bleeding, ligation of the uterine and utero-ovarian arteries [O'Leary stitch], placement of uterine compression sutures and/or intrauterine balloon tamponade), and ultimately arterial embolization or hysterectomy. Uterine tourniquets may be useful as a temporizing measure while pharmacotherapy is administered and taking effect or while surgical interventions are being considered (figure 1). (See 'Management of postpartum hemorrhage' above.)

●Low-lying placenta

•A low-lying placenta is one in which the distance between the placental edge and internal cervical os is 0 to 20 mm. For women with a low-lying placenta where the placental edge is 0 to 10 mm from the edge of the internal os, we suggest planned cesarean birth (Grade 2C). These pregnancies are at high (>50 percent) risk for intrapartum hemorrhage necessitating emergency cesarean delivery. We suggest a trial of labor for those with placental edge to internal os distance 11 to 20 mm (Grade 2C). These patients have a >80 percent chance of successful vaginal birth. However, these are shared decisions between clinician and patient. (See 'Delivery of patients with a low-lying placenta' above.)