Initiation of abortion before there is definitive ultrasound evidence of intrauterine pregnancy: A systematic review with meta-analyses.

Abstract

Women are increasingly presenting for abortion at very early gestation. However, providers may be reluctant to conduct abortion at this stage as they may be concerned that they cannot exclude an ectopic pregnancy or that they may terminate a non-viable pregnancy, or may be concerned that both medical and surgical methods may be less effective at this stage of gestation. This provider concern may result in delays in the abortion as additional investigations may be required until an intrauterine pregnancy can be confirmed. Additional unnecessary visits may be distressing for women and waste health service resources. The objective of this systematic review was to determine whether it is safe and effective to initiate abortion before there is ultrasound evidence of an intrauterine pregnancy. We searched Embase Classic, Embase; Ovid MEDLINE® Epub Ahead-of-Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE® Daily, Ovid MEDLINE®; and Cochrane Library on 25 October 2019. Eligible studies were randomized and non-randomized comparative studies, published in English from 1985, comparing initiation of abortion before there is definitive evidence of an intrauterine pregnancy with initiation afterwards. We assessed risk-of-bias using the Newcastle-Ottowa scale. All outcomes were analyzed as risk ratios (RR) and meta-analyzed using the Mantel-Haenszel method. The quality of the evidence was assessed using GRADE. Two non-randomized studies (n=3785) showed no differences in "missed ectopic pregnancy" (RR=0.26, 95% CI 0.03-2.12), "ongoing pregnancy" (RR=1.06, 95% CI 0.34-3.34), or "complete abortion without surgical intervention" (RR=1, 95% CI 0.98-1.02) between initiation of medical abortion before or after ultrasound evidence of an intrauterine pregnancy. A third non-randomized study (n=1530) showed no differences between initiation of surgical abortion before or after ultrasound evidence of an intrauterine pregnancy in "missed ectopic pregnancy" (no events), "ongoing pregnancy" (RR=0.56, 95% CI 0.03-11.59) or "complete abortion without repeat surgical intervention" (RR=1, 95% CI 0.99-1.01). The quality of evidence was very low. Initiation of abortion before there is definitive ultrasound evidence of an intrauterine pregnancy in women without signs or symptoms of an ectopic pregnancy should be considered.