Complication Rate Is Lower For Surgical Than Medical Second-Trimester Abortion

Susan London

First published online:

Dilation and evacuation (D&E) is safer than medical abortion for second-trimester pregnancy termination, and among medical methods, misoprostol is safer than others.1 A retrospective cohort study of women who had a second-trimester abortion found that complications occurred in 29% of women who had a medical abortion but only 4% of women who had a D&E. The higher rate in women in the medical group was primarily due to a higher rate of incomplete abortion requiring surgery; this complication occurred in one in every five women in this group. Women who had a medical abortion had 80% lower odds of having complications if they were given misoprostol than if they were given other medications.

Analyses for the study included 139 women who had a surgical abortion and 158 women who had a medical abortion (i.e., induction by administration of misoprostol, high-dose pitocin or prostaglandin suppository) at 14-24 weeks of gestation. The researchers identified the women by reviewing medical records and diagnostic codes at two university hospitals in Milwaukee for the seven-year period from January 1994 through February 2001. For all of the women, they recorded a range of background characteristics, including information on their reproductive history, and details about the abortion.

Women were classified as having complications if they had any of the following events: failed medical abortion (i.e., D&E was needed to complete the procedure), bleeding requiring transfusion, infection requiring intravenous antibiotics, retained products of conception requiring dilation and curettage, organ injury requiring additional surgery, cervical laceration requiring repair and hospital readmission.

On average, women in both the surgical and the medical abortion groups were about 30 years old and had a body mass index of about 26 kg/m2. They had had similar numbers of pregnancies (2.6 and 3.0, respectively) and live births (1.0 and 1.2), and they were about equally likely already to have a uterine scar (14% and 13%) and to have had a D&E (0% and 1%). The average gestational age was significantly younger for women who had surgery than for those who had a medical abortion (18.4 vs. 20.3 weeks). Women who had a surgical abortion were significantly more likely to have any laminaria inserted (92% vs. 65%) and to have had the abortion for health reasons (4% vs. 1%). They also had significantly more laminaria inserted (4.5 vs. 3.3) and had a significantly shorter hospital stay (0.3 vs. 1.6 days).

The proportion of women experiencing complications was significantly higher in the medical abortion group than in the surgical group (29% vs. 4%). Women who had a medical abortion were significantly more likely than those who underwent surgery to have retained products of conception requiring dilation and curettage (21% vs. fewer than 1%). Seven percent of women in the medical abortion group required a D&E to complete the abortion.

Among women who had a medical abortion, 79% were given misoprostol. The likelihood of complications in these women (22%) was greater than that in women who had a D&E (4%), but considerably lower than the likelihood among women who had other types of medical abortion (55%).

In logistic regression analysis controlling for gestational age, number of pregnancies and length of hospital stay, women who had a D&E had a significant 90% reduction in the odds of complications relative to women who had a medical abortion (adjusted odds ratio, 0.1). In addition, as the number of laminaria inserted increased, a woman's odds of complications following a D&E declined by 10% (0.9). Women who lost more than 500 ml of blood had substantially increased odds of complications (6.4).

When the same factors plus the number of laminaria inserted were taken into account in a logistic regression analysis involving only the medical abortion group, women who were given misoprostol had a significant 80% reduction in the odds of complications relative to women who were given other medications first (adjusted odds ratio, 0.2). Women who had significant blood loss again had sharply increased odds of suffering complications (23.0).

The study's findings add important new information, the researchers note, because little research has compared the safety of surgical abortion with the safety of misoprostol induction during the second trimester. Because medical abortion might have been favored for women with more advanced pregnancies, it is noteworthy that the greater safety of D&E persisted after gestational age was taken into account, they add.

"When skilled operators are available, dilation and evacuation should be considered the preferred method for second trimester abortion," the researchers conclude, emphasizing that a woman's preference must also be considered. They recommend that laminaria be inserted before D&E to reduce the need for intraoperative cervical dilation and thereby reduce the risk of complications. Women should be advised that their odds of needing surgery are fairly high if they have a medical abortion, the researchers note, and if a woman and her physician select medical abortion, "misoprostol is the medication of choice."--S. London


1. Autry AM et al., A comparison of medical induction and D&E for second-trimester abortion, American Journal of Obstetrics & Gynecology, 2002, 187(2): 393-397.