Objective
To compare immediate vs delayed medical treatment for first-trimester miscarriage.
Study Design
Randomized open-label trial in a university hospital gynecologic emergency department. Between April 2003 and April 2006, 182 women diagnosed with spontaneous abortion before 14 weeks’ gestation were assigned to immediate medical treatment (oral mifepristone, followed 48 hours later by vaginal misoprostol, n = 91) or sequential management (1 week of watchful waiting followed, if necessary, by the above-described medical treatment, n = 91). Vacuum aspiration was performed in case of treatment failure, hemorrhage, pain, infection, or patient request.
Results
Compared with immediate medical treatment, sequential management resulted in twice as many vacuum aspirations overall (43.5% vs 19.1%; P < .001), 4 times as many emergent vacuum aspirations (20% vs 4.5%; P = .001), and twice as many unplanned visits to the emergency department (34.1% vs 16.9%; P = .009).
Conclusion
Delaying medical treatment of first-trimester miscarriage increases the rate of unplanned surgical uterine evacuation.
First-trimester miscarriage occurs in about 11% of pregnancies and is associated with a considerable human and financial burden. For decades, surgical evacuation of the uterus was performed almost consistently. However, medical treatment and expectant management (ie, monitoring until expulsion occurs spontaneously) were introduced recently. Expectant management produced variable success rates, of up to 75% after 1 week, without adverse events. Medical treatment with the progesterone receptor antagonist mifepristone and/or the synthetic prostaglandin E1 analog misoprostol was effective in about 80% of cases. Compared with surgery (vacuum aspiration), expectant management, or medical treatment, was less effective in emptying the uterus but led to lower infection rates and costs. Moreover, the rates of infection (3%) and 5-year infertility (20%) were acceptable with all three treatment methods, indicating that medical and expectant management were equivalent to surgery in terms of efficacy and safety.
However, in these studies, vacuum aspiration was performed in all patients with failed expectant management. Conceivably, initial expectant management (eg, for 1 week), followed by medical treatment, if needed, may combine the advantages of both methods and further decrease the need for surgery. For these benefits to occur, medical treatment after 1 week would have to be as effective as immediate medical treatment.
We conducted a pragmatic, open-label, randomized trial in patients with first-trimester miscarriage to assess whether delayed medical treatment strategy was associated with a lower vacuum aspiration rate compared with immediate medical treatment.
Materials and Methods
Trial design
We conducted a single-center, randomized, controlled, open-label, parallel-group trial to compare 2 management strategies in women with spontaneous first-trimester miscarriage. Informed consent was obtained from each patient before study inclusion. The study was approved by the appropriate ethics committee ( Comité Consultatif de Protection des Personnes en Recherche Biomedicale of Saint-Germain-en-Laye, France) and registered on ClinicalTrials.gov under no. NCT00190294 .
Patients
Eligibility criteria
We included patients with spontaneous miscarriage before 14 weeks of amenorrhea who agreed to participate in the study. Patients with any of the following clinical presentations were eligible: missed miscarriage, inevitable miscarriage (bleeding through the dilated cervical os), and incomplete miscarriage (retained product of conception). Transvaginal ultrasonography (Logic 700; General Electric, Velizy, France) was performed by a senior obstetrician/gynecologist. Miscarriage was diagnosed in patients with any of the 3 following ultrasound criteria: anembryonic gestation with an intact intrauterine gestational sac measuring 15-50 mm and containing no visible embryonic structures, confirmed after 5 days; early embryonic/fetal demise with an intrauterine gestational sac measuring 15-50 mm and containing a visible embryo having no cardiac activity and no growth after 48 hours if the crown-rump length was greater than 15 mm or after 5 days if the crown-rump length was 15 mm or less; and retained product of conception with intrauterine contents measuring more than 15 mm in thickness after spontaneous expulsion of a proven intrauterine pregnancy.
Clinical exclusion criteria were imminent miscarriage (pain, bleeding, and product of conception passing through the cervical os); unstable hemodynamics; severe pain (score >4/10 on a numerical rating scale); age younger than 18 years; gestational sac larger than 50 mm; multiple pregnancy; suspected molar or ectopic pregnancy; previous signs of cervical incompetence; evidence of infection; contraindication to mifepristone (allergy, corticosteroid deficiency, corticosteroid therapy, hemostasis disorder, thrombocytopenia <100,000/mm 3 , or blood hemoglobin <9 g/dL) or contraindication to misoprostol (allergy or severe asthma); patient living alone or more than an hour away from the hospital; and participation in another therapeutic trial.
Setting
This trial was performed at the gynecology emergency department of the teaching hospital of Poissy Saint Germain-en-Laye (France).
Intervention
Treatments were carried out on an outpatient basis. Under our current protocol for managing miscarriages, each woman received information about the progress of spontaneous miscarriage, including symptoms indicating a need for immediate medical attention (heavy bleeding and pain, fever), and a prescription for an analgesic (suppository containing acetaminophen, opium, and caffeine) and an antispasmodic agent (oral tablet of phloroglucinol and trimethylphloroglucinol). An injection of 250 IU of anti-D immunoglobulin was given if appropriate.
Women allocated to the immediate medical treatment arm were given 200 mg of mifepristone orally on the first study day (day 0) and 400 μg of misoprostol vaginally 2 days later (day 2). They then returned home for monitoring by the family. Transvaginal ultrasonography was performed on day 7. Patients with retained products of conception (gestational sac or intrauterine heterogeneous content >15 mm in thickness) underwent elective vacuum aspiration on the next day, and those with normal ultrasound findings were discharged.
Women allocated to the sequential management arm returned home with no intervention until day 7, when they came to the hospital for transvaginal ultrasonography. Patients with normal findings were discharged, and patients with retained products of conception were given the same medical treatment as in the immediate medical treatment group (200 mg of mifepristone orally on day 7 and 400 μg of misoprostol vaginally on day 9). Transvaginal ultrasonography was performed on day 14. Patients with normal findings were discharged, and those with retained products of conception underwent vacuum aspiration on day 15.
Patients had free access to our gynecology emergency department, which was open 24 hours a day, 7 days a week. Emergency vacuum aspiration could be performed at any time in patients with abundant bleeding, severe pain, or any other clinical situation requiring immediate evacuation of the uterus; as well as in patients who wanted to leave the study.
Patients who requested a change in treatment after randomization were analyzed as part of their initial group, on an intention-to-treat basis. Vacuum aspiration performed on patient request was recorded as an emergency procedure. Ectopic pregnancy and viable pregnancy were the only reasons that could lead to exclusion from the analysis. For each patient, we recorded unplanned visits and treatments, protocol deviations, and elective or emergency surgical vacuum aspiration.
Outcomes
The primary outcome was the incidence of vacuum aspiration in each group (ie, the incidence of nonsurgical treatment failure), determined without distinguishing between elective aspirations (retained product of conception 7 days after medical treatment given immediately or after 7 days of expectancy) and emergency aspirations.
Secondary outcomes were use of emergency services, total time to complete evacuation since randomization, efficacy of immediate or delayed medical treatment, and treatment side effects (intolerance and bleeding).
Sample size
Based on previously published trials, we expected spontaneous expulsion to occur in about 75% of patients after 1 week of expectant management and in about 80% after immediate medical treatment. According to our working hypothesis that medical treatment would have similar efficacy when used immediately or after expectant management, we expected sequential management to be followed by spontaneous evacuation in 75% of patients and medical treatment to be effective in 20% of patients (ie, 80% of 25%), with the remaining 5% of patients requiring vacuum aspiration. To detect a significant change in the expected differences in vacuum aspiration rates (ie, 5% vs 20%) with 80% power at 5% bilateral alpha risk, we needed 2 groups of at least 88 patients each.
Randomization
Patients were included by 14 senior obstetrician/gynecologists. For each patient, an inclusion form was faxed to a centralized statistics department for randomization into 2 groups, using a computer-generated list, in blocks of 4. Neither the patients nor the physicians were blinded to the treatment arm.
Statistical methods
Data were analyzed using the intention-to-treat approach. Differences between groups were assessed using the χ 2 test or Fisher’s exact test for categoric variables and the Student’s t test for continuous variables. P values smaller than .05 were considered significant. Results were expressed as percentage or mean, and 95% confidence intervals (95% CIs) were computed. The statistical analysis was performed using STATA version 11.0 (Stata Corp, College Station, TX).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
