Conference Coverage
Review shows D&C raises preterm birth risk
Dilation and curettage increases the risk for prematurity in subsequent pregnancy, according to a systematic review and meta-analysis of nearly 2...
LISBON – Expectant management is a safe and effective treatment for incomplete miscarriage after misoprostol, results of the MisoREST trial show.
“Because it is non-invasive, we think it should be considered as first-line treatment and to reserve curettage only for specific patients,” Dr. Marianne Verschoor said at the annual meeting of the European Society of Human Reproduction and Embryology.
Patrice Wendling/Frontline Medical News
Dr. Marianne Verschoor
Misoprostol (Cytotec) is an inexpensive and effective treatment for miscarriage, but about 30% of women are left with an incomplete evacuation of the uterus. Curettage is often performed in these typically asymptomatic women, but carries both short- and long-term risks, said Dr. Verschoor of the Academic Medical Center, Amsterdam.
No study had directly compared the effectiveness of curettage and expectant management, she said. To fill this void, the researchers performed a randomized controlled trial (RCT) and parallel observational cohort study at 27 community and academic hospitals. Women who had sonographic findings of intrauterine remnants and/or a thick endometrium (> 10 mm) 1-2 weeks after misoprostol treatment for first-trimester miscarriage were randomly assigned to curettage within 3 days of diagnosis or expectant management. Women who declined randomization were prospectively followed.
The primary outcome was sonographic evidence of an empty uterine cavity at 6 weeks or an uneventful clinical course at 3 months in the absence of sonography.
Because of strong preferences for expectant management (EM), 59 women were randomized in the RCT (30 curettage, 29 EM) and 197 women were followed prospectively (65 curettage, 132 EM). All groups were comparable at baseline with regard to age, gestational age, and previous miscarriage or curettage.
In the RCT, 93% of women had an empty cavity or uneventful clinical course with curettage vs. 76% with EM. This difference was not statistically significant (relative risk, 1.23; 95% confidence interval, 0.098-1.54), Dr. Verschoor said.
In the observational cohort, 95% of patients choosing curettage had a successful outcome vs. 83% choosing EM. This difference was statistically significant (RR, 1.14; CI, 1.04-1.26).
When the two populations were pooled, 95% of patients (90/95) treated with curettage reached the primary end point vs. 82% of patients (132/161) with EM, which was statistically different (RR, 1.16; CI, 1.06-1.26), she said.
There were no significant differences in secondary outcomes in the RCT between the curettage and EM groups, including intrauterine infection (each 1 patient), excessive blood loss defined as > 500 mL (1 patient vs. 2 patients), or duration of bleeding (mean 5.4 days vs. 5.5 days).
However, one patient had already developed Asherman syndrome in the curettage group within 3 months of follow-up, Dr. Verschoor said. Two patients in each arm underwent reintervention with hysteroscopy and 5 EM patients underwent curettage.
Secondary outcomes in the observational cohort were also comparable between the curettage and EM groups for intrauterine infection (3 vs. 2), excessive blood loss (0 vs. 1), and duration of bleeding (mean 5.7 days vs. 7.1 days). One patient who underwent curettage developed Asherman syndrome vs. none in the EM group, she said. There were 11 reinterventions in the curettage group (7 hysteroscopy, 4 recurettage), compared with 30 in the EM group (12 hysteroscopy, 18 curettage).
A closer look at the EM patients who underwent reintervention revealed that “the interventions were primarily elective,” Dr. Vershcoor said. “Only in 26% of cases was there a medical reason for performing the intervention like bleeding, fever, or excessive abdominal pain.” Histology performed in these patients found no pregnancy tissue in 46%.
A separate quality of life analysis of the MisoREST participants found no statistical differences at baseline, 2, 4, and 12 weeks for the Short Form 36 Health Survey mental and physical score in all four treatment arms. Anxiety and depression measured with the Hospital Anxiety and Depression Scale were also not statistically different at any time, study coauthor Dr. Marike Lemmers, also of the Academic Medical Center, Amsterdam, reported in a poster at the meeting.
ZonMw, a Dutch organization for health research and development, funded the study. Dr. Verschoor reported having no financial conflicts.
On Twitter @pwendl