ALSO Syllabus Update

Late Pregnancy Bleeding

Evidence review completed: December 14, 2005

By Ellen Sakornbut, MD

Update published:

Placenta Previa

The greatest morbidity and mortality associated with placenta previa is related to complications of prematurity. Delay of delivery is frequently desired after initial bleeding is stabilized. A randomized controlled trial (RCT) of 60 women between 28 and 34 weeks gestational age with symptomatic placenta previa evaluated the safety and efficacy of tocolysis with ritodrine with a prolongation of pregnancy averaging 25.33 vs.14.47 d (p<0.05) in the trial group¹. Prolongation greater than 2 days and greater than 7 days occurred more frequently in the study group than the control group. No increase in complications occurred in women receiving tocolysis. Infant birth weights averaged 2270 g vs. 1950 g (p<0.05).

Another randomized controlled trial comparing 54 women randomized to outpatient expectant versus inpatient management after an initial 72 hour observation found no increase in hemorrhage, need for transfusion, or poor maternal or neonatal outcomes in women treated as outpatients². Hospital days averaged 10 versus 28 days in the control group.

Abruption

A large, prospective RCT of magnesium sulfate for the treatment of preeclampsia³ found a 58% reduction in the incidence of eclampsia and a 27% reduction in placental abruption.

No high-quality RCTs are available for the treatment of abruption, but a retrospective, case-control study of women admitted with severe abruption found a relationship between decision to delivery interval and poor neonatal outcome with relative risk of 0.44 (95% CI 0.22-0.86) comparing a decision to delivery interval of 20 minutes versus 30 minutes⁴.

Vasa Previa

A large retrospective study of more than 93,000 women in the second and third trimester detected 18 vasa previa in the antepartum setting, with the average gestational age at diagnosis 26 weeks and the earliest detection occurring at 15.6 weeks. Vasa previa was suspected with two-dimensional ultrasound and confirmed with color-flow Doppler⁶. The neonatal mortality associated with vasa previa undetected before onset of labor is more than 40%⁵. It is unlikely that screening specifically for vasa previa can be done efficiently in a general population, but case finding may be warranted in women diagnosed with placenta previa in the second trimester. Vasa previa is highly associated with placenta previa diagnosed by ultrasound in the second trimester with an odds ratio of 56.3 (95% CI, 8.9-354.1)[vii] Two-dimensional imaging of the lower uterine segment is a routine part of the standard second and third trimester ultrasound exam. Patients who should be considered for more directed studies include women followed for resolving placenta previa, those with a placenta with a succenturiate lobe, and any patient with an apparent velamentous insertion of the cord.

NNT = number-needed-to-treat   NNH = number-needed to harm