Ultrasound Obstet Gynecol 2018; 51: 24–32

Published online 5 December 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.18940

Misoprostol treatment vs expectant management in women

with early non-viable pregnancy and vaginal bleeding: a
pragmatic randomized controlled trial
A. FERNLUND , L. JOKUBKIENE , P. SLADKEVICIUS and L. VALENTIN
Department of Obstetrics and Gynecology, Skåne University Hospital, Lund University, Malmö, Sweden

K E Y W O R D S: first-trimester pregnancy; misoprostol; pregnancy complications; randomized controlled trial; spontaneous

miscarriage

ABSTRACT
Objective To compare vaginal misoprostol treatment
with expectant management in early non-viable pregnancy
with vaginal bleeding with regard to complete evacuation
of the uterine cavity within 10 days after randomization.
Methods This was a parallel randomized controlled,
open-label trial conducted in Skåne University Hospital,
Sweden. Patients with anembryonic pregnancy or early
fetal demise (crown–rump length ≤ 33 mm) and vaginal
bleeding were randomly allocated to either expectant
management or treatment with a single dose of
800 μg misoprostol administered vaginally. Patients
were evaluated clinically and by ultrasound until
complete evacuation of the uterus was achieved (no
gestational sac in the uterine cavity and maximum
anteroposterior diameter of the intracavitary contents
< 15 mm as measured by transvaginal ultrasound on
midsagittal view). Follow-up visits were planned at 10,
17, 24 and 31 days. Dilatation and evacuation (D&E)
was recommended if miscarriage was not complete
within 31 days, but was performed earlier at patient’s
request, or if there was excessive bleeding as judged
clinically. Analysis was by intention to treat. The main
outcome measure was number of patients with complete
miscarriage without D&E ≤ 10 days.
Results Ninety-four patients were randomized to miso-
prostol treatment and 95 to expectant management. After
exclusion of three patients and withdrawal of consent
by two patients in the expectant management group, 90
women were included in this group. Miscarriage was
complete ≤ 10 days in 62/94 (66%) of the patients in the
misoprostol group and in 39/90 (43%) of those in the
group managed expectantly (risk difference (RD) = 23%;
95% CI, 8–37%). At 31 days, the corresponding figures
were 81/94 (86%) and 55/90 (61%) (RD = 25%; 95%
CI, 12–38%). Two patients from each group under-
went emergency D&E because of excessive bleeding and
one of these in each group received blood transfusion.
The number of patients undergoing D&E at their own
request was higher in the expectantly managed group,
15/90 (17%) vs 3/94 (3%) in the misoprostol group
(RD = 14%; 95% CI, 4–23%), as was the number of
patients making out-of-protocol visits, 50/90 (56%) vs
27/94 (29%) (RD = 27%; 95% CI, 12–40%). Compared
with the expectant management group, more patients in
the misoprostol group experienced pain (71/77 (92%) vs
91/91 (100%); RD = 8%; 95% CI, 1–17%) and used
painkillers (59/77 (77%) vs 85/91 (93%); RD = 17%;
95% CI, 5–29%). No major side effect was reported in
any group.
Conclusions In women with early non-viable pregnancy
and vaginal bleeding, misoprostol treatment is more
effective than is expectant management for complete
evacuation of the uterus. Both methods are safe but
misoprostol treatment is associated with more pain than
is expectant management. Copyright © 2017 ISUOG.
Published by John Wiley & Sons Ltd.
INTRODUCTION
Almost one in five clinically recognized pregnancies
miscarries spontaneously in the first trimester1,2 . On
ultrasound, early miscarriage can be classified as (1)
early fetal or embryonic loss, i.e. a gestational sac
with a visible embryo without cardiac pulsations3 ,
(2) anembryonic gestation, i.e. empty sac or sac with
minimal embryonic debris without cardiac activity3 , or (3)
incomplete miscarriage, i.e. no visible gestational sac but
ultrasound signs of retained products of conception4 . In

Correspondence to: Dr A. Fernlund, Department of Obstetrics and Gynecology, Skåne University Hospital, SE 20502 Malmö, Sweden
(e-mail: anna.fernlund@skane.se)
Accepted: 23 October 2017

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. RANDOMIZED CONTROLLED TRIAL
Misoprostol vs expectant management for miscarriage
this work, early fetal or embryonic loss and anembryonic
pregnancies are referred to as non-viable pregnancies.
To ascertain non-viability in early pregnancy can be
difficult, and new criteria of non-viable early pregnancy
have been suggested5–7 and incorporated into national
guidelines8–12 .
Treatment of miscarriage should be individualized.
Surgical evacuation and medical treatment are alternative
methods to empty the uterus after miscarriage, but one
can also await spontaneous emptying13–15 .
Medical and expectant management avoid the small
risks of surgery and anesthesia16,17 . Misoprostol, a
synthetic prostaglandin E1 analog, is the most studied
medical treatment13 , but should be avoided if there
are contraindications to its use. The success rate of
any treatment depends on the type of miscarriage
(non-viable pregnancy or incomplete miscarriage), length
of follow-up and definition of complete miscarriage. It
may be underestimated if the follow-up period is very
short. In incomplete miscarriage, the success rate of
expectant management is similar to that of misoprostol
treatment4 . In non-viable pregnancies some studies report
very low success of expectant management or treatment
with placebo, but this may be explained by short
follow-up18,19 .
The primary aim of this randomized controlled trial
was to compare expectant management with vaginal
misoprostol treatment in early non-viable pregnancies
with vaginal bleeding with regard to complete evacuation
of the uterine cavity ≤ 10 days after randomization.
Secondary aims were to compare the two methods
with regard to complications and complete evacuation
≤ 17 days, ≤ 24 days and ≤ 31 days after randomization.
METHODS
This was a randomized controlled, open-label trial with
individual randomization into two parallel groups 1:1,
i.e. treatment with vaginal misoprostol or expectant
management (ClinicalTrials.gov ID: NCT01033903). We
deliberately chose an open label design because we wanted
to study how these two management strategies work in
clinical practice when both doctors and patients know
which treatment is given. We did not aim to study the
effect of misoprostol per se by comparing it with placebo
treatment.
Patients were recruited from the emergency clinic of
the Department of Obstetrics and Gynecology, Skåne
University Hospital, Malmö, Sweden. Local Ethical
Committee approval was obtained (Dnr 83/2008), and
all participants gave written informed consent. Eligible
were hemodynamically stable patients with vaginal
spotting or bleeding in early pregnancy. Patients with
heavy bleeding judged clinically to necessitate immediate
surgical evacuation of the uterine cavity were not
eligible.
Inclusion criteria were: ≥ 18 years old, ability to
understand written and spoken Swedish, hemoglobin
concentration (Hb) > 80 g/L, no contraindications to
Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.
25
treatment with misoprostol, fulfilling ultrasound criteria
for early non-viable pregnancy and fetal crown–rump
length (CRL) ≤ 33 mm. Because of new recommendations
for ultrasound criteria of early pregnancy failure5–7 , we
changed our ultrasound criteria of non-viability in 2014.
Before 2014 the criteria of non-viability in our study were
one of the following: (1) intrauterine gestational sac with a
mean diameter > 16 mm with no embryonic pole20 or (2) a
gestational sac with an embryo with CRL ≥ 5 mm without
cardiac pulsations20 , or (3) if the above criteria were not
fulfilled, a gestational sac with or without an embryo
that showed no significant development at a repeat scan
after 7 days20 . After April 2014 the non-viability criteria
in our study were one of the following: (1) intrauterine
gestational sac with a mean diameter ≥ 25 mm with no
embryonic pole or (2) a gestational sac with an embryo
with CRL ≥ 7 mm without cardiac pulsations, or (3) if
the above criteria were not fulfilled, a gestational sac
with or without an embryo that showed no significant
development at a repeat scan after 7 days5,6,8,9 . Since
all patients were managed as outpatients we did not
include patients with a fetus with CRL > 33 mm because
of the risk of heavy bleeding at expulsion of the
pregnancy.
Inclusion and follow-up of all patients recruited to the
study were done by one of three doctors in the trial
team. The flow of patients through the trial is shown
in Figure 1. Each patient was followed up with visits
planned on day 10, 17, 24 and 31 after randomization,
i.e. after initiation of treatment. As soon as complete
expulsion of the pregnancy was achieved as assessed with
transvaginal ultrasound, the patient was discharged with
no more follow-up visits planned. Complete miscarriage
was defined as no gestational sac in the uterine cavity and
maximum anteroposterior diameter of the intracavitary
contents < 15 mm as measured by transvaginal ultrasound
on a midsagittal view.
Primary outcome was number of patients with
complete miscarriage without dilatation and evacuation
(D&E) within 10 days after randomization. Secondary
outcomes were number of patients with complete
miscarriage without D&E within 17, 24 and 31 days after
randomization, number of patients with complications
from randomization until 4 weeks after the uterine cavity
was judged to be empty (with or without D&E) or, if
expulsion did not occur within 31 days, 4 weeks after
day 31. Secondary outcomes were also self-reported
pain, use of analgesics, duration of vaginal bleeding
and side effects (nausea, diarrhea, vomiting, headache
and dizziness) until complete miscarriage or up to
31 days. The complications assessed were pelvic infection
(endometritis/salpingitis), number of D&Es (emergency
and others), blood transfusion and change in Hb
concentration from inclusion until complete miscarriage.
The number of patients treated with antibiotics (including
those prescribed antibiotics), making unplanned visits,
being hospitalized and taking sick-leave was also
recorded. A diagnosis of endometritis was assigned to
patients with low abdominal or pelvic pain and/or
Ultrasound Obstet Gynecol 2018; 51: 24–32.
26 Fernlund et al.

Randomization (n = 189)

Allocation

Misoprostol treatment (n = 94) Expectant management (n = 95)

Received allocated intervention (n = 94) Received allocated intervention (n = 93)
Withdrew consent (n = 2)

D&E before first D&E before first

follow-up (n = 2) follow-up (n = 5)
(Bleeding (n = 2)) (Patient’s request (n = 3);
bleeding (n = 2))
Follow-up

Follow-up day 10 (5–13)
(n = 92)
Failed to appear (n = 0)
D&E (n = 1)
(Patient’s request (n = 1))
D&E (n = 0) Follow-up day 10 (5–13) Complete miscarriage
Complete miscarriage (n = 86) (n = 39)
(n = 62)* Failed to appear (n = 2) Excluded (n = 2)†

D&E (n = 3) Follow-up day 17 (14–21) D&E (n = 5)

Follow-up day 17 (14–21)
(Patient’s request (n = 44) (Patient’s request (n = 5))
(n = 29)
(n = 3)) Failed to appear (n = 2) Complete miscarriage
Failed to appear (n = 1)
Complete miscarriage (n = 5)
(n = 11)

Follow-up day 24 (22–28)
(n = 15)
Failed to appear (n = 1)
D&E (n = 6)
D&E (n = 0) Follow-up day 24 (22–28) (Patient’s request (n = 6))
Complete miscarriage (n = 34) Complete miscarriage
(n = 5) Failed to appear (n = 2) (n = 7)
Excluded (n = 1)‡

D&E (n = 2) D&E (n = 8)
Follow-up day 31 (30–37)
Follow-up day 31 (30–37) (Failure of treatment
(Failure of treatment (n = 17)
(n = 11) (n = 2)) (n = 6); infection
Failed to appear (n = 5)
Complete miscarriage (n = 2))
(n = 3) Complete miscarriage
(n = 4)
Continued expectant Continued expectant
management management
(patient’s request) (patient’s request)
(n = 6)§ (n = 10)¶

D&E (n = 6)
Analysis D&E (n = 3)

Retrospective review of patient records 4 weeks after Retrospective review of patient records 4 weeks after
complete miscarriage or after day 31 complete miscarriage or after day 31
Analyzed (n = 94) Analyzed (n = 90)

Figure 1 Flow of patients through the trial. *One patient returned with vaginal bleeding after 2 months and retained pregnancy tissue was
confirmed after surgical evacuation. †Did not fulfill inclusion criteria (viable pregnancy, n = 1; molar pregnancy, n = 1). ‡Did not fulfill
inclusion criteria (scar pregnancy). §Underwent dilatation and evacuation (D&E) after 2, 6 and 8 weeks of additional observation (n = 3);
complete miscarriage after 2 and 4 additional weeks (n = 2); treated with second dose of misoprostol at last trial follow-up and was judged
to have complete miscarriage 1 week later (n = 1). ¶Underwent D&E after 1, 3, 3, 4, 6 and 8 weeks of additional observation (n = 6);
complete miscarriage after 2 days, 1 week and 3 weeks (n = 3); did not return for follow-up until after 2 months when there was suspicion of
arteriovenous shunts and was followed up with serum β-human chorionic gonadotropin which had normalized after 5 months (n = 1).

abnormal vaginal discharge, uterine tenderness and
pyrexia ≥ 38 ◦ C and/or elevated level (> 5 mg/L) of
C-reactive protein (CRP). The definition of salpingitis
was acute abdominal pain, cervical motion tenderness
and adnexal tenderness at pelvic bimanual examination
in addition to one of the following: (1) pathological
wet smear or pathological discharge, (2) elevated CRP,
or (3) pyrexia ≥ 38 ◦ C21 . Antibiotics were prescribed
when an infection was suspected on the basis of
symptoms, clinical findings and laboratory results as
recommended by the US Centers for Disease Control
and Prevention22 .
Sample size calculation was based on the assumption of
a 70% success rate (complete miscarriage within 10 days
after randomization) in the misoprostol group and 50%
success rate in the expectantly managed group23–25 . To

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2018; 51: 24–32.
Misoprostol vs expectant management for miscarriage
achieve 80% power with an alpha error of 0.05 the
total required sample size was 93 patients in each group.
To compensate for loss to follow-up, exclusions and
incomplete data, we planned for 120 subjects in each
group.
Randomization was computer-generated in blocks of
six and was accomplished by an independent consultant
at an off-site research service center. Allocation to
misoprostol or expectant management was hidden in
sealed opaque envelopes prepared and sequentially
numbered by staff not involved in the trial. All enrolment
was done by one of three clinicians in the trial team.
After informed consent had been obtained, the consecutive
envelope was opened and allocation was revealed to the
clinician and the patient, i.e. neither the clinician nor the
patient was blinded to treatment.
The day of randomization was defined as day 1. At
inclusion, the patients were examined clinically and by
ultrasound. The ultrasound equipment used was a Sequoia
512 (Acuson Inc., Mountain View, CA, USA) with a
4–7.5-MHz transvaginal transducer. Each patient was
randomly allocated to expectant care or to treatment
with a single vaginal dose of 800 μg misoprostol (four
tablets at 200 μg) that was placed in the posterior vaginal
fornix by the trial clinician. All patients got written
information on the study and the management to which
they had been assigned, along with the telephone number
of the responsible trial clinician. A structured diary to
record bleeding, pain, use of painkillers and side effects
(nausea, vomiting, headache, dizziness and diarrhea) was
handed out and participants were instructed to fill it in
at home every day from day 1 until the miscarriage was
complete or for at most 31 days, and to bring it to each
scheduled follow-up visit. Participants self-rated bleeding
and pain as none, mild, moderate or severe. All patients
were offered a prescription of paracetamol with codeine
to use in case of severe pain. On day 1, blood was
drawn for analysis of Hb, human chorionic gonadotropin
(hCG), progesterone and blood type. Rh-negative women
received anti-D rhesus prophylaxis (1500 IU) within
72 h after inclusion. All participants were managed as
outpatients.
The first follow-up visit was scheduled on day
10 as shown in the flowchart (Figure 1). On every
follow-up visit, transvaginal ultrasonography and clinical
examination were performed by the trial clinician, and
Hb concentration was analyzed. If the gestational sac
was still present, or if the uterine contents measured
≥ 15 mm in maximum anteroposterior diameter, the
patient was scheduled for a subsequent follow-up visit
after 7 days. If the miscarriage was considered to be
complete no more visits were planned and the diaries
were collected. If complete evacuation was not achieved
on day 31, treatment was considered a failure and the
patient was recommended D&E. Participants could ask
for D&E at any time during the trial, if they did not
want to await spontaneous emptying of the uterine
cavity. The method of D&E was chosen by the doctor
carrying out the procedure, the routine procedure being
Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.
27
vacuum aspiration followed by blunt curettage without
ultrasound guidance. Products of conception were not
routinely analyzed histologically, and serum β-hCG levels
were not checked after the miscarriage was considered
complete.
Data were collected from the prospectively filled in
research protocols, the patients’ diaries and medical
records. Prospectively collected data included information
on results of clinical and ultrasound examinations at inclu-
sion and follow-up visits and the planned management
at each follow-up visit. Additional information regard-
ing out-of-protocol visits, complications and interventions
from inclusion (day 1) until 4 weeks after complete mis-
carriage (or 4 weeks after day 31 if miscarriage was not
complete on day 31) was retrieved retrospectively from the
patient records. According to the Swedish Medical Prod-
ucts Agency, assignment to treatment had to be noted
in the patients’ medical records, i.e. the person retrieving
information from the patient records retrospectively could
not be blinded to the assignment group.
Statistical analysis was performed using SPSS Statistics,
version 21 (IBM Corp., Armonk, NY, USA). Analysis
was by intention-to-treat. Main outcome measure was
absolute risk difference with 95% confidence interval (CI).
Differences in proportions were calculated with 95% CI
including continuity correction using a calculator on www
.vassarstats.net (©Richard Lowry). The somersd package
in Stata version 14 (StataCorp LLC, College Station, TX,
USA) was used to calculate differences in continuous
variables with 95% CI.
The manuscript was prepared following the Con-
solidated Standards of Reporting Trials (CONSORT)
guidelines26 .
RESULTS
Between September 2008 and December 2015, 189
patients were recruited to the trial. Ninety-five patients
were allocated to expectant management and 94
to treatment with misoprostol. Twenty-one patients
(expectant group: 11, misoprostol group: 10) were
recruited after the new ultrasound criteria of non-viable
early pregnancy were adopted. Recruitment was slower
than anticipated, and the number of patients recruited
was lower than the planned 240 despite a prolonged
recruitment period.
An overview of our main results are presented in
Figure 1. Two patients in the expectantly managed group
withdrew consent shortly after enrolment and three were
excluded because they were found not to fulfill our inclu-
sion criteria at follow-up visits (viable pregnancy (n = 1),
molar pregnancy (n = 1) and Cesarean scar pregnancy
(n = 1); Appendix S1). Thus, 184 patients were analyzed
with regard to treatment outcome, 90 receiving expectant
management and 94 receiving misoprostol treatment.
Baseline characteristics of the study population are
presented in Table 1. Gestational age according to the
last menstrual period ranged from 6 to 16 weeks. The
distribution of the different types of miscarriage was
Ultrasound Obstet Gynecol 2018; 51: 24–32.
28 Fernlund et al.

Table 1 Baseline characteristics of 184 patients with early non-viable pregnancy with vaginal bleeding, randomized to vaginal misoprostol
treatment or expectant management, included in randomized controlled trial

Misoprostol treatment Expectant management

Characteristic (n = 94) (n = 90)

Age (years) 32.1 ± 5.6 31.9 ± 5.5

Gestational age according to last menstrual period (days) 76.5 ± 14.3 75.5 ± 12.3*
Nulliparous 49 (52.1) 33 (36.7)
Previous miscarriage 24 (25.5) 33 (36.7)
Previous termination of pregnancy 22 (23.4) 22 (24.4)
Hemoglobin concentration (g/L) 131.2 ± 8.4 129.2 ± 9.3†
Plasma hCG (IU/L)¶ 5303.5 (6–77 286)‡ 8483.0 (107–80 317)
Plasma progesterone (nmol/L)¶ 18 (1–63)§ 19 (3–190)§
Type of miscarriage at inclusion ultrasound exam
Fetal pole without cardiac activity (CRL ≥ 5 mm/≥ 7 mm**) 52 (55.3) 39 (43.3)
Anembryonic (gestational sac ≥ 16 mm/> 25 mm**) 13 (13.8) 20 (22.2)
Abnormal progress of pregnancy (after 7 days’ observation) 29 (30.9) 31 (34.4)
Anembryonic 15 14
Fetal pole without cardiac activity 14 17
Visible fetal pole (includes cases with abnormal progress of pregnancy) 66 (70.2) 56 (62.2)
CRL (mm) 11.0 (1.9–31.0) 6.4 (3.0–26.0)
Days from start of bleeding to inclusion 5.0 (1–42) 6.0 (1–60)§

Data are given as mean ± SD, n (%) or median (range). *Five missing values. †Four missing values. ‡Two missing values. §One missing
value. ¶Before mid May 2013, human chorionic gonadotropin (hCG) and progesterone were analyzed in serum and not in
plasma – reference intervals unaffected. **Two different cut-offs pertaining to before and after change, in 2014, of ultrasound criteria of
non-viability.
CRL, crown–rump length.

similar in the two groups, but there was a slight imbalance
between the groups with regard to parity, hCG levels
and CRL.
Outcome with regard to success rate and complications
are presented in Table 2 and Figure 1. The success rate
was significantly higher in the misoprostol group: 62/94
(66.0%) vs 39/90 (43.3%) in the expectantly managed
group had achieved complete miscarriage within 10 days
(risk difference (RD) = 22.6%; 95% CI, 7.5–36.5%).
However, one patient in the misoprostol group judged
to have miscarried completely on her first follow-up visit
returned with bleeding and dizziness after 2 months and
underwent emergency D&E. Trophoblastic tissue was
confirmed on histological examination. The cumulative
rate of complete miscarriage (without D&E) within 17,
24 and 31 days was also higher in the misoprostol group
than in the expectantly managed group. The success rate
within 31 days was 81/94 (86.2%) vs 55/90 (61.1%)
(RD = 25.1%; 95% CI, 11.6–37.5%).
D&E was significantly less common in the misoprostol
group than in the expectantly managed group: 10/94
(10.6%) vs 31/90 (34.4%) (RD = −23.8%; 95% CI,
−35.8 to −11.1%). The number of patients undergoing
emergency D&E was similar, 2/94 (2.1%) in the
misoprostol group vs 4/90 (4.4%) in the expectantly
managed group (RD = −2.3%; 95% CI, −9.7 to 4.5%),
but the number of patients undergoing D&E at their
own request was much higher in the group receiving
expectant management, 15/90 (16.7%) vs 3/94 (3.2%)
(RD = 13.5%; 95% CI, 4.1–23.4%). Also, the number of
patients undergoing D&E because of failed treatment was
higher in the expectant group, although the difference was
not statistically significant, 12/90 (13.3%) vs 5/94 (5.3%)
(RD = 8.0%; 95% CI, −1.4 to 17.8%). The number of
patients making unscheduled visits was also significantly
higher in the expectantly managed group. One patient
from each group had profuse blood loss and received
a blood transfusion. No other major complications
occurred in any of the groups. There was one case of
repeat D&E because of retained products of conception
in the misoprostol group. Two patients in the expectantly
managed group underwent hysteroscopy, one because of
retained products of conception after one D&E and one
because of difficulties with evacuating the cavity at D&E
due to a uterine septum.
Analysis of change in Hb concentration was not
meaningful because different blood sampling methods
were used at inclusion (venous) and follow-up (capillary).
Capillary Hb concentration at last follow-up was similar
in the two groups.
One hundred and sixty-eight patients completed and
returned the form for self-report (diary) of bleeding, pain
and side effects. The results are presented in Table 3.
Three patients (3.2%) in the misoprostol group and 13
patients (14.4%) in the expectantly managed group did
not complete the diary and were excluded from the
analysis of diary data. Duration of vaginal bleeding
differed between the groups with 2.3 more days of
bleeding in the expectantly managed group. Six patients
in the expectantly managed group did not report any
pain at all, while all patients in the misoprostol group
reported at least some pain, and more patients in the
misoprostol group took oral painkillers, 85/91 (93.4%)
vs 59/77 (76.6%) (RD = 16.8%; 95% CI, 5.2–28.7%).
Nausea, vomiting, headache, diarrhea and dizziness were
equally common in the two groups.

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2018; 51: 24–32.
Misoprostol vs expectant management for miscarriage 29

Table 2 Outcome of 184 patients with early non-viable pregnancy with vaginal bleeding, randomized to vaginal misoprostol treatment or
expectant management

Misoprostol treatment Expectant management Difference

Outcome (n = 94) (n = 90) (95% CI)
Complete miscarriage without D&E
≤ 10 days 62 (66.0)* 39 (43.3) 22.6 (7.5 to 36.5)
≤ 17 days 73 (77.7) 44 (48.9) 28.8 (14.2 to 41.9)
≤ 24 days 78 (83.0) 51 (56.7) 26.3 (12.4 to 39.1)
≤ 31 days 81 (86.2) 55 (61.1) 25.1 (11.6 to 37.5)
Total number of patients undergoing D&E 10 (10.6) 31 (34.4) −23.8 (−35.8 to −11.1)
D&E before first follow-up 2 (2.1) 5 (5.6) −3.4 (−11.2 to 3.6)
Emergency (bleeding) 2 (2.1) 2 (2.2) −0.1 (−6.3 to 6.7)
Patient request 0 3 (3.3) −3.3 (−10.1 to 2.2)
D&E during follow-up 8 (8.5) 26 (28.9) −20.4 (−32.0 to −8.4)
Patient request 3 (3.2) 12 (13.3) −10.1 (−19.6 to −1.3)
Failure of treatment 5 (5.3) 12 (13.3) −8.0 (−17.8 to 1.4)
Emergency (bleeding) 1 (1.1)* 0 1.1 (−4.1 to 6.6)
Emergency (endometritis) 0 2 (2.2) −2.2 (−8.6 to 3.0)
Endometritis 1 (1.1) 3 (3.3) −2.3 (−9.1 to 3.8)
Prescribed antibiotics 3 (3.2) 7 (7.8) −4.6 (−13.0 to 3.2)
Hemoglobin at last follow-up (g/L) 124.2 ± 12.3† 123.5 ± 12.2‡ 0.7 (−3.6 to 5.1)
Received blood transfusion 1 (1.1) 1 (1.1) 0.0 (−5.9 to 5.6)
Out-of-protocol visit§ 27 (28.7) 50 (55.6) −26.8 (−40.3 to −11.8)
Number of out-of-protocol visits 0.5 ± 1.0 1.0 ± 1.2 −0.5 (−0.8 to −0.2)
Hospitalized 3 (3.2) 4 (4.4) −1.2 (−8.8 to 5.9)
Days of hospitalization 0.0 ± 0.2 0.1 ± 0.4 −0.05 (−0.1 to 0.04)
Sick leave 1 (1.1) 4 (4.4) −3.4 (−10.6 to 2.9)
Sick-leave days (n) 0.1 ± 0.5 0.8 ± 4.6 −0.7 (−1.7 to 0.2)

Data are given as n (%) or mean ± SD. *One patient judged to have had a complete miscarriage at first follow-up returned with vaginal
bleeding after 2 months and underwent dilatation and evacuation (D&E); she is not included in total number of patients undergoing D&E,
because she could not be classified both as having had a complete miscarriage and as failure of treatment. †17 missing values. ‡37 missing
values. §From inclusion until 4 weeks after complete miscarriage or after 31 days if not complete miscarriage on day 31.

Table 3 Self-reported bleeding, pain and side effects according to completed diaries of patients with early non-viable pregnancy with vaginal
bleeding, randomized to vaginal misoprostol treatment or expectant management

Misoprostol treatment Expectant management Difference

(n = 91) (n = 77) (95% CI)
Self-reported bleeding and pain
Vaginal bleeding, number of days after inclusion 12.7 ± 6.6 15.0 ± 8.2 −2.3 (−4.6 to −0.6)
Patients with pain 91 (100.0) 71 (92.2) 7.8 (1.0 to 16.8)
Patients with severe pain 64 (70.3) 47 (61.0) 9.3 (−5.8 to 24.2)
Number of days with severe pain 1.3 ± 1.6 1.3 ± 1.5 0.02 (−0.5 to 0.5)
Patients taking painkillers 85 (93.4) 59 (76.6) 16.8 (5.2 to 28.7)
Number of tablets of paracetamol and codeine 3.7 ± 5.2 2.6 ± 5.2 1.1 (−0.5 to 2.6)
Number of tablets of other painkillers 6.6 ± 11.0 5.0 ± 6.0 1.6 (−1.1 to 4.4)
Reported side-effects
Nausea, number of patients 57 (62.6) 45 (58.4)* 4.2 (−11.3 to 19.6)
Number of days with nausea 1.5 ± 2.2 2.4 ± 4.3* −0.9 (−1.9 to 0.1)
Vomiting, number of patients 7 (7.7) 10 (13.0) −5.3 (−15.9 to 4.7)
Number of days with vomiting 0.1 ± 0.5 0.2 ± 0.5 −0.1 (−0.2 to 0.1)
Diarrhea, number of patients 26 (28.6) 17 (22.1) 6.5 (−7.7 to 20.0)
Number of days with diarrhea 0.5 ± 1.0 0.6 ± 1.8 −0.1 (−0.6 to 0.3)
Dizziness, number of patients 50 (54.9) 33 (42.9)* 12.0 (−4.0 to 27.3)
Number of days with dizziness 1.4 ± 2.1 1.9 ± 4.2* −0.4 (−1.4 to 0.6)
Headache, number of patients 56 (61.5) 51 (66.2)* −4.7 (−19.7 to 10.7)
Number of days with headache 2.5 ± 3.1 2.8 ± 4.2* −0.3 (−1.4 to 0.8)

Results are shown as n (%) or mean ± SD. *One patient in expectantly managed group reported nausea, dizziness and headache for 31 days,
i.e. all days during study period.

DISCUSSION bleeding. Complete miscarriage was achieved within

This randomized controlled trial shows that treatment
with vaginal misoprostol is more effective than is
expectant management in evacuating the uterus in
patients with early non-viable pregnancy and vaginal
10 days in 66% of the patients treated with misoprostol
vs 43% of those managed expectantly. After 1 month,
the success rate was 86% vs 61%. More patients in
the expectantly managed group underwent D&E and
made out-of-protocol visits, but more patients treated

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 2018; 51: 24–32.
30
with misoprostol experienced pain and took painkillers.
Major complications were rare in both groups. Our results
are generalizable to patients similar to those included in
this trial. They are not generalizable to patients with
incomplete miscarriage (i.e. no gestational sac in the
uterine cavity) or to patients without vaginal bleeding.
Our study adds information to the existing literature,
because our study population is well defined (only
non-viable pregnancies with bleeding were included), and
because D&E was not recommended before 31 days of
observation. These are the strengths of our study. In many
randomized trials comparing misoprostol treatment with
placebo or expectant management, the study populations
are less well defined than in our study (including patients
both with and without bleeding or not stating if bleeding
was present, or including both non-viable pregnancies
and incomplete miscarriages without showing results
separately for the two groups). In addition, in many
trials, D&E was carried out after a very short period of
observation, so that the chance of successful treatment
with long observation cannot be evaluated properly.
Results of published randomized trials comparing
misoprostol treatment with expectant management or
placebo are summarized in Table 418,19,23,25,27–31 .
It is a limitation of our study that we did not attain
the planned 240 patients. However, the number finally
analyzed almost reached the number needed according
to our sample size calculation. Difficulty with recruiting
patients to randomized trials on management of early
pregnancy failure is not unique to our study25,30,32 .
Our experience, in accordance with that of others, is
that many patients have a strong preconceived treatment
preference32 . Some might find the open-label design of
our trial a limitation. However, as explained above, we
deliberately chose this design, because our aim was to
compare two treatment regimens as they work in clinical
practice when both doctors and patients know which
treatment is given. It is a weakness that the person
retrieving data retrospectively from patient records could
not be blinded to the allocated treatment. However, the
risk of bias is low, since most of the variables assessed
retrospectively were well defined. It is also a limitation
that we changed our inclusion criteria at the end of the
recruitment period. We felt obliged to do so because
new criteria for defining a non-viable early pregnancy
on ultrasound examination were recommended5–7 in
several national guidelines8–12 . Because chills and fever
are known side effects of misoprostol treatment33 ,
it is a limitation that the patients did not record
these side effects in their diaries, meaning that we
cannot provide reliable information on their prevalence.
However, misoprostol-induced fever rarely lasts for more
than 24 h after administration33 , and neither prospectively
recorded nor retrospectively retrieved information implied
the occurrence of this complication. Another limitation is
that, for logistic reasons, we are unable to report the total
number of eligible patients during the study period.
There was a slight imbalance between the treatment
groups in our trial, in that more women in the expectantly
Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.
Fernlund et al.
managed group were parous, had slightly higher hCG
levels (while progesterone levels were very similar) and
their fetuses had smaller CRL values. It is uncertain if
these imbalances, explained by chance, affect the results
of our trial. In a study by Schwärzler et al.34 , hCG
and progesterone levels seemed to be associated with
complete miscarriage < 7 days in women with non-viable
early pregnancy managed expectantly, while parity had
no effect, and the effect of CRL was not examined.
We used the same or a similar definition of complete
miscarriage as several other research teams, i.e. maximum
sonographic anteroposterior diameter of the uterine
cavity contents < 15 mm or ≤ 15 mm23,27,29,35 . This
definition has been questioned24,36–39 . It is likely that
the success rate and the number of D&Es would have
been different if we had used other criteria of success,
e.g. absence of a gestational sac in the uterine cavity or
cessation of vaginal bleeding.
Our results confirm those of others, that misoprostol
treatment shortens the time to complete miscarriage in
women with a non-viable early pregnancy18,19,23,25,27,31
(see Table 4), but they also show that expectant
management is a viable alternative to medical treat-
ment. Many women find expectant management highly
attractive24,34,40 . An important advantage of a ‘watch and
wait’ approach is that it avoids the risk of terminating
a normal pregnancy misdiagnosed as a failed one41,42 .
One patient with a viable pregnancy was included in our
trial, and randomization of viable pregnancies was also
reported by Trinder et al.25 .
Our results indicate that approximately 50% of women
with an early non-viable pregnancy who have started
to bleed will miscarry spontaneously and completely
within 2–3 weeks, but that waiting beyond this period
is probably not worthwhile. Waiting beyond 2–3 weeks
also seems futile in women treated with misoprostol.
The number of unplanned visits and D&Es at patient’s
request was higher in women randomized to expectant
management than to treatment with misoprostol, but the
number is likely to be lower if women actively choose
expectant management in accordance with their own
preference. It is important to stress, that risks and benefits
of all treatment options, including surgery (which was not
evaluated in our trial but was in others25,30 ), should be
considered and discussed with the patient to individualize
treatment. The risks of surgical vacuum evacuation may
be small in expert hands if performed under ultrasound
guidance43 but may be higher in patients with risk factors
(e.g. previous Cesarean section). While prostaglandin
treatment avoids surgical and anesthesiological risks, it
should be used with caution in certain situations, e.g.
in patients with cerebrovascular disease, coronary artery
disease, or inflammatory bowel disease. When balancing
advantages and disadvantages of different treatments
against each other, the time from diagnosis to complete
miscarriage, treatment-associated discomfort such as pain
and limitation of conduct of life while awaiting complete
miscarriage should be taken into account.
Ultrasound Obstet Gynecol 2018; 51: 24–32.
 Table 4 Randomized controlled trials comparing different treatments for early miscarriage (surgery, medical treatment, expectant management)

Inclusion Medical Outcome measures (definition Success

Reference Size of trial criteria Bleeding treatment of successful treatment) rate

Medical treatment vs placebo

Kovavisarach 27 medical, Anembryonic, well defined Not stated Misoprostol 400 μg vaginally Complete miscarriage (no definition) Medical: 63%
(2002)18 27 placebo without D&E within 24 h Placebo: 18%
Lister 18 medical, Non-viable pregnancy, well No bleeding or only Misoprostol 800 μg vaginally Complete miscarriage (absence of Medical: 83%
(2005)19 16 placebo defined spotting (repeat after 24 h if no gestational sac) without D&E within 48 h Placebo: 13%
expulsion)
Blohm 64 medical, ‘Gestational residue (AP ‘Signs of miscarriage’ Misoprostol 400 μg vaginally Complete miscarriage (AP diameter Medical: 81%
(2005)27 62 placebo diameter) 15–50 mm’, Open or closed cervix < 15 mm) without D&E within 6–7 days Placebo: 52%
not well defined
Bagratee 52 medical, ‘Incomplete or early Bleeding or no bleeding Misoprostol 600 μg vaginally Complete miscarriage (endometrial Early pregnancy
(2004)23 52 placebo pregnancy failure’, (repeat after 24 h if no thickness ≤ 15 mm) without D&E within failure:
well defined expulsion) 7 days Medical: 87%
Placebo: 29%
Incomplete:
Medical: 100%
Placebo: 86%
Wood 25 medical, ‘Non-viable pregnancy’, No bleeding or only Misoprostol 800 μg vaginally Complete miscarriage (no definition) Medical: 80%
Misoprostol vs expectant management for miscarriage

(2002)31 25 placebo well defined spotting (repeat after 24 h if expulsion without D&E within 4 weeks (D&E Placebo: 16%

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.

not complete) offered after 48 h if not complete
miscarriage, not always ultrasound
diagnosis)
Medical treatment vs expectant management
Nielsen 62 medical, Incomplete or non-viable Bleeding, closed cervix Mifepristone 400 mg orally + Complete miscarriage (AP diameter Medical: 82%
(1999)29 60 expectant pregnancy, well defined misoprostol 400 μg orally < 15 mm) without D&E within 5 days Expectant: 76%
Ngai 30 medical, Incomplete or ‘missed Not stated, ‘clinical Misoprostol 400 μg vaginally Complete miscarriage (no definition) Missed miscarriage:
(2001)28 30 expectant miscarriage’, well defined diagnosis of on day 1, 3 and 5 without D&E within 43 days (D&E if Medical: 80%
spontaneous gestational sac was present on day 15) Expectant: 32%
miscarriage’ Incomplete:
Medical: 100%
Expectant: 80%
Medical treatment vs expectant management vs surgery
Shelley 15 expectant, ‘Incomplete or inevitable Bleeding Misoprostol 400 μg vaginally Complete miscarriage (not defined) without Medical: 80%
(2005)30 13 medical, miscarriage’, not well (repeat after 4–6 h if unplanned D&E within 8 weeks (D&E if Expectant: 79%
12 surgery defined incomplete miscarriage) RPOC on day 15) Surgery: 100%
Trinder 399 expectant, Incomplete or non-viable Not stated Misoprostol 800 μg vaginally No unplanned D&E within 8 weeks (D&E Non-viable:
(2006)25 398 medical, pregnancy, well defined (incomplete) or mifepristone if expulsion had not started 8 h after Medical: 62%
403 surgery 200 mg orally + misoprostol misoprostol + D&E at 10–14 days if Expectant: 50%
800 μg vaginally (non-viable) RPOC seen on ultrasound) Surgery: 94%
(Primary outcome: gynecological Incomplete:
infection ≤ 14 days) Medical: 71%
Expectant: 75%
Surgery: 98%

AP, anteroposterior; D&E, dilatation and evacuation; RPOC, retained products of conception.

Ultrasound Obstet Gynecol 2018; 51: 24–32.

31
32
To improve selection for expectant or medical treatment
of non-viable early pregnancy, predictive factors for
successful evacuation within a certain time should be
sought. Moreover, the definition of complete miscarriage
may need revision. It might be better to assess the
appearance of the uterine cavity contents on ultrasound
images than to rely on its anteroposterior diameter.
Alternatively, the ultrasound definition of complete
miscarriage could be replaced with a clinical definition,
e.g. cessation of vaginal bleeding.
ACKNOWLEDGMENT
This study was supported by grants from the Swedish gov-
ernment (‘ALF-medel’, ‘regionalt forskningsstöd Region
Skåne’).
REFERENCES
1. Hemminki E. Treatment of miscarriage: current practice and rationale. Obstet
Gynecol 1998; 91: 247–253.
2. Warburton D, Fraser FC. Spontaneous abortion risks in man: Data from reproductive
histories collected in a medical genetics unit. Am J Hum Genet 1964; 16:
1–25.
3. Farquharson RG, Jauniaux E, Exalto N. Updated and revised nomenclature for
description of early pregnancy events. Hum Reprod 2005; 20: 3008–3011.
4. Kim C, Barnard S, Neilson JP, Hickey M, Vazquez JC, Dou L. Medical
treatments for incomplete miscarriage. Cochrane Database Syst Rev 2017; 1:
Cd007223.
5. Abdallah Y, Daemen A, Kirk E, Pexsters A, Naji O, Stalder C, Gould D, Ahmed
S, Guha S, Syed S, Bottomley C, Timmerman D, Bourne T. Limitations of current
definitions of miscarriage using mean gestational sac diameter and crown–rump
length measurements: a multicenter observational study. Ultrasound Obstet Gynecol
2011; 38: 497–502.
6. Abdallah Y, Daemen A, Guha S, Syed S, Naji O, Pexsters A, Kirk E, Stalder C, Gould
D, Ahmed S, Bottomley C, Timmerman D, Bourne T. Gestational sac and embryonic
growth are not useful as criteria to define miscarriage: a multicenter observational
study. Ultrasound Obstet Gynecol 2011; 38: 503–509.
7. Jeve Y, Rana R, Bhide A, Thangaratinam S. Accuracy of first-trimester ultrasound
in the diagnosis of early embryonic demise: a systematic review. Ultrasound Obstet
Gynecol 2011; 38: 489–496.
8. National Guideline Clearinghouse. ACR Appropriateness Criteria® first trimester
bleeding. https://www.guideline.gov/summaries/summary/43883 [Accessed 24 April
2017].
9. National Institute for Health and Care Excellence (NICE). Ectopic pregnancy and
miscarriage: diagnosis and initial management. https://www.nice.org.uk/guidance/
cg154 [Accessed 24 April 2017].
10. Committee on Practice Bulletins—Gynecology. The American College of Obstetri-
cians and Gynecologists. Practice Bulletin No. 150. Early pregnancy loss. Obstet
Gynecol 2015; 125: 1258–1267.
11. Australasian Society for Ultrasound in Medicine. Guidelines for the perfor-
mance of first trimester ultrasound. http://www.asum.com.au/files/public/SoP/
D11-Guidelines-for-the-Performance-of-First-Trimester-Ultrasound.pdf [Accessed
22 August 2017].
12. Morin L, Cargill YM, Glanc P. Ultrasound evaluation of first trimester complications
of pregnancy. J Obstet Gynaecol Can 2016; 38: 982–988.
13. Neilson JP, Hickey M, Vazquez J. Medical treatment for early fetal death (less than
24 weeks). Cochrane Database Syst Rev 2006: Cd002253.
14. Chen BA, Creinin MD. Contemporary management of early pregnancy failure. Clin
Obstet Gynecol 2007; 50: 67–88.
15. Dempsey A, Davis A. Medical management of early pregnancy failure: how to treat
and what to expect. Semin Reprod Med 2008; 26: 401–410.
16. Lemmers M, Verschoor MAC, Hooker AB, Opmeer BC, Limpens J, Huirne JAF,
Ankum WM, Mol BWM. Dilatation and curettage increases the risk of subsequent
SUPPORTING INFORMATION ON THE INTERNET
The following supporting information may be found in the
online version of this article:
Appendix S1 Description of three patients excluded from the
expectantly managed group because of incorrect diagnosis
of miscarriage.
Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd.
Fernlund et al.
preterm birth: a systematic review and meta-analysis. Hum Reprod 2016; 31: 34–45.
17. Tristan SBM, Gilliam MMM. First trimester surgical abortion. Clin Obstet Gynecol
2009; 52: 151–159.
18. Kovavisarach E, Sathapanachai U. Intravaginal 400 μg misoprostol for pregnancy
termination in cases of blighted ovum: a randomised controlled trial. Aust N Z J
Obstet Gynaecol 2002; 42: 161–163.
19. Lister MS, Shaffer LET, Bell JG, Lutter KQ, Moorma KH. Randomized, double-blind,
placebo-controlled trial of vaginal misoprostol for management of early pregnancy
failures. Am J Obstet Gynecol 2005; 193: 1338–1343.
20. Levi CS, Lyons EA, Lindsay DJ. Ultrasound in the first trimester of pregnancy. Radiol
Clin North Am 1990; 28: 19–38.
21. Romosan G, Bjartling C, Skoog L, Valentin L. Ultrasound for diagnosing acute
salpingitis: a prospective observational diagnostic study. Hum Reprod 2013; 28:
1569–1579.
22. Centers for Disease Control and Prevention. Sexually Transmitted Diseases
Guidelines: Pelvic inflammatory disease. https://www.cdc.gov/std/tg2015/pid.htm
[Accessed 22 August 2017].
23. Bagratee JS, Khullar V, Regan L, Moodley J, Kagoro H. A randomized controlled
trial comparing medical and expectant management of first trimester miscarriage.
Hum Reprod 2004; 19: 266–271.
24. Luise C, Jermy K, May C, Costello G, Collins WP, Bourne TH. Outcome of expectant
management of spontaneous first trimester miscarriage: observational study. BMJ
2002; 324: 873–875.
25. Trinder J, Brocklehurst P, Porter R, Read M, Vyas S, Smith L. Management of
miscarriage: expectant, medical, or surgical? Results of randomised controlled trial
(miscarriage treatment (MIST) trial). BMJ 2006; 332: 1235–1240.
26. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: Updated guidelines
for reporting parallel group randomised trials. J Clin Epidemiol 2010; 63: 834–840.
27. Blohm F, Fridén BE, Milsom I, Platz-Christensen JJ, Nielsen S. A randomised double
blind trial comparing misoprostol or placebo in the management of early miscarriage.
BJOG 2005; 112: 1090–1095.
28. Ngai SW, Chan YM, Tang OS, Ho PC. Vaginal misoprostol as medical treatment
for first trimester spontaneous miscarriage. Hum Reprod 2001; 16: 1493–1496.
29. Nielsen S, Hahlin M, Platz-Christensen J. Randomised trial comparing expectant with
medical management for first trimester miscarriages. BJOG 1999; 106: 804–807.
30. Shelley JM, Healy D, Grover S. A randomised trial of surgical, medical and expectant
management of first trimester spontaneous miscarriage. Aust N Z J Obstet Gynaecol
2005; 45: 122–127.
31. Wood SL, Brain PH. Medical management of missed abortion: a randomized clinical
trial. Obstet Gynecol 2002; 99: 563–566.
32. Wieringa-de Waard M, Vos J, Bonsel GJ, Bindels PJ, Ankum WM. Management of
miscarriage: a randomized controlled trial of expectant management versus surgical
evacuation. Hum Reprod 2002; 17: 2445–2450.
33. Blum J, Winikoff B, Gemzell-Danielsson K, Ho PC, Schiavon R, Weeks A. Treatment
of incomplete abortion and miscarriage with misoprostol. Int J Gynaecol Obstet
2007; 99: S186–S189.
34. Schwärzler P, Holden D, Nielsen S, Hahlin M, Sladkevicius P, Bourne TH. The
conservative management of first trimester miscarriages and the use of colour
Doppler sonography for patient selection. Hum Reprod 1999; 14: 1341–1345.
35. Nielsen S, Hahlin M. Expectant management of first-trimester spontaneous abortion.
Lancet 1995; 345: 84–86.
36. Chen BA, Creinin MD. Medical management of early pregnancy failure: efficacy.
Semin Reprod Med 2008; 26: 411–422.
37. Creinin MD, Harwood B, Guido RS, Fox MC, Zhang J. Endometrial thickness after
misoprostol use for early pregnancy failure. Int J Gynaecol Obstet 2004; 86: 22–26.
38. Reeves MF, Lohr PA, Harwood BJ, Creinin MD. Ultrasonographic endometrial
thickness after medical and surgical management of early pregnancy failure. Obstet
Gynecol 2008; 111: 106–112.
39. Sawyer E, Ofuasia E, Ofili-Yebovi D, Helmy S, Gonzalez J, Jurkovic D. The value
of measuring endometrial thickness and volume on transvaginal ultrasound scan
for the diagnosis of incomplete miscarriage. Ultrasound Obstet Gynecol 2007; 29:
205–209.
40. Wieringa-de Waard M, Bindels PJ, Vos J, Bonsel GJ, Stalmeier PF, Ankum WM.
Patient preferences for expectant management vs. surgical evacuation in first-trimester
uncomplicated miscarriage. J Clin Epidemiol 2004; 57: 167–173.
41. Hately W, Case J, Campbell S. Establishing the death of an embryo by ultrasound:
report of a public inquiry with recommendations. Ultrasound Obstet Gynecol 1995;
5: 353–357.
42. Ledger WL, Turner MJ. Implementation of the findings of a national enquiry into the
misdiagnosis of miscarriage in the Republic of Ireland: impact on quality of clinical
care. Fertil Steril 2016; 105: 417–422.
43. Acharya G, Morgan H, Paramanantham L, Fernando R. A randomized controlled
trial comparing surgical termination of pregnancy with and without continuous
ultrasound guidance. Eur J Obstet Gynecol Reprod Biol 2004; 114: 69–74.
This article has been selected for Journal Club.
A slide presentation, prepared by Dr Fiona Brownfoot,
one of UOG's Editors for Trainees, is available online.
Chinese translation by Dr Yang Fang.
Spanish translation by Dr Ruben Dario Fernandez.
Ultrasound Obstet Gynecol 2018; 51: 24–32.
Ultrasound Obstet Gynecol 2018; 51: 24–32
Published online 5 December 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.18940

Tratamiento con misoprostol versus tratamiento expectante en mujeres con embarazos no viables
tempranos y hemorragia vaginal: un ensayo controlado aleatorizado pragm ático
RESUMEN
Objetivo Comparar el tratamiento vaginal con misoprostol con el tratamiento expectante en embarazos no viables
tempranos con hemorragia vaginal respecto al vaciado completo de la cavidad uterina en los 10 dı́as posteriores a la
aleatorización.
Métodos Este fue un ensayo paralelo, aleatorizado, controlado y abierto, realizado en el Hospital Universitario de
Skåne en Suecia. Las pacientes con embarazo anembriónico o con muerte del feto temprana (longitud céfalo-caudal ≤
33 mm) y hemorragia vaginal se asignaron al azar, al tratamiento expectante o al tratamiento con una dosis única de
800 μg de misoprostol administrado por vı́a vaginal. Las pacientes fueron evaluadas clı́nicamente y por ecografı́a hasta
que se logró el vaciado completo del útero (ausencia de saco gestacional en la cavidad uterina y máximo diámetro
anteroposterior de los contenidos intracavitarios <15 mm, medido por ecografı́a vaginal en la proyección sagital media).
Las visitas de seguimiento se planificaron a los 10, 17, 24 y 31 dı́as. Se recomendó una dilatación y vaciado (DyV)
cuando el aborto no se completó dentro de los 31 dı́as, pero se realizó antes a solicitud de la paciente, o si hubo una
excesiva hemorragia según lo estimado clı́nicamente. El análisis fue por intención de tratar. La medida de resultado
principal fue el número de pacientes con aborto completo sin DyV ≤ 10 dı́as.
Resultados Noventa y cuatro pacientes fueron aleatorizadas al tratamiento con misoprostol y 95 al tratamiento
expectante. Después de la exclusión de tres pacientes y la retirada del consentimiento por parte de dos pacientes en el
grupo de tratamiento expectante, se incluyeron 90 mujeres en este grupo. El aborto fue completo en ≤ 10 dı́as en 62/94
(66%) de las pacientes en el grupo de misoprostol y en 39/90 (43%) de aquellas en el grupo de tratamiento expectante
(diferencia de riesgo (DR) = 23%; IC 95%, 8–37%). A los 31 dı́as, las cifras correspondientes fueron 81/94 (86%) y
55/90 (61%) (DR = 25%, IC 95%, 12–38%). Dos pacientes de cada grupo fueron sometidas a DyV de emergencia
debido a una excesiva hemorragia y una de estas en cada grupo recibió una transfusión de sangre. El número de
pacientes sometidas a DyV a petición propia fue mayor en el grupo con tratamiento expectante, 15/90 (17%) vs 3/94
(3%) en el grupo de misoprostol (DR = 14%; IC 95%, 4–23 %), al igual que el número de pacientes que realizaron
visitas fuera del protocolo, 50/90 (56%) vs 27/94 (29%) (DR = 27%, IC 95%, 12–40%). En comparación con el grupo
de tratamiento expectante, fue mayor el número de pacientes que experimentaron dolor en el grupo de misoprostol
(71/77 (92%) vs 91/91 (100%), DR = 8%, IC 95%, 1–17%) y utilizaron analgésicos (59/77 (77%) vs 85/91 (93%),
RD = 17%, IC 95%, 5–29%). No se reportó ningún efecto secundario importante en ninguno de los grupos.
Conclusiones En mujeres con embarazos no viables tempranos y hemorragia vaginal, el tratamiento con misoprostol
es más efectivo que el tratamiento expectante para el vaciado completo del útero. Ambos métodos son seguros, pero el
tratamiento con misoprostol está asociado con más dolor que el tratamiento expectante.

Copyright © 2017 ISUOG. Published by John Wiley & Sons Ltd. RANDOMIZED CONTROLLED TRIAL