Misoprostol versus curettage in women with early pregnancy failure after initial expectant management: A randomized trial

Abstract

Little is known about the effectiveness of misoprostol treatment in women with early pregnancy failure who have been managed expectantly. We therefore performed a randomized trial on this subject.
Women with early pregnancy failure, who had been managed expectantly for at least 1 week, were assigned randomly (using a computer-generated list) to receive either vaginal misoprostol 800 microg or curettage. If, after 24 h, there had not been complete evacuation, a further 800 microg of misoprostol was administered. In the absence of complete evacuation after > days, curettage was performed. The primary end-point was the complete evacuation of the products of conception, with secondary end-points being the occurrence of side effects, pain intensity, need for analgesics, and intensity and duration of bleeding.
In the misoprostol group, 47% (37/79) required curettage due to incomplete evacuation. After 1 week, evacuation was complete in 85% (67 out of 79) of the misoprostol group and 93% (70 out of 75) of the curettage group. Severity of pain, bleeding and emergency evacuation was higher in the misoprostol group. The complication rate was 0% for misoprostol and 4% for curettage.
Curettage is superior to misoprostol in the evacuation of early pregnancy failure after failed expectant management. However, misoprostol could be clinically useful since it reduces the need for curettage by half and has a lower complication rate, at the expense of increased pain, vaginal bleeding and emergency evacuation.

Full text

Misoprostol versus curettage in women with early
pregnancy failure after initial expectant
management: a randomized trial
G.C.M.Graziosi
1,7
, B.W.J.Mol
2,3
, P.J.H.Reuwer
4
, A.Drogtrop
5
and H.W.Bruinse
6
1
Department of Obstetrics and Gynaecology, St Antonius Hospital, Koekoekslaan 1, Nieuwegein,
2
Department of Obstetrics
and Gynaecology, Academic Medical Centre, Meibergdreef 9, Amsterdam,
3
Department of Obstetrics and Gynaecology, Maxima
Medical Centre, De Run 4600, Veldhoven,
4
Department of Obstetrics and Gynaecology, St Elisabeth Hospital, Hilvarenbeekseweg
60, Tilburg,
5
Department of Obstetrics and Gynaecology, Tweesteden Hospital, Dr Deelenlaan 5, Tilburg and
6
Department of
Obstetrics and Gynaecology, University Medical Centre, Heidelberglaan 100, Utrecht, The Netherlands
7
To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, St Antonius Hospital, PO Box 2500,
3430 EM, Nieuwegein, The Netherlands. E-mail: p.graziosi@Antonius.net
BACKGROUND: Little is known about the effectiveness of misoprostol treatment in women with early pregnancy
failure who have been managed expectantly. We therefore performed a randomized trial on this subject.
METHODS: Women with early pregnancy failure, who had been managed expectantly for at least 1 week, were
assigned randomly (using a computer-generated list) to receive either vaginal misoprostol 800 mg or curettage. If,
after 24 h, there had not been complete evacuation, a further 800 mg of misoprostol was administered. In the
absence of complete evacuation after > 3 days, curettage was performed. The primary end-point was the complete
evacuation of the products of conception, with secondary end-points being the occurrence of side effects, pain
intensity, need for analgesics, and intensity and duration of bleeding. RESULTS: In the misoprostol group, 47%
(37/79) required curettage due to incomplete evacuation. After 1 week, evacuation was complete in 85% (67 out of
79) of the misoprostol group and 93% (70 out of 75) of the curettage group. Severity of pain, bleeding and emer-
gency evacuation was higher in the misoprostol group. The complication rate was 0% for misoprostol and 4% for
curettage. CONCLUSIONS: Curettage is superior to misoprostol in the evacuation of early pregnancy failure after
failed expectant management. However, misoprostol could be clinically useful since it reduces the need for cure-
ttage by half and has a lower complication rate, at the expense of increased pain, vaginal bleeding and emergency
evacuation.
Key words: curettage/early pregnancy failure/expectant management/misoprostol/randomized trial
Introduction
Early pregnancy failure is a common complication of preg-
nancy, as approximately one in four women will experience
a miscarriage (Warburton and Fraser, 1964; Edmonds et al.,
1982). For . 50 years, the standard management of early
pregnancy failure has been dilatation and curettage. Curettage
bears a small risk of pelvic infection, cervical injury, uterine
perforation, excessive bleeding and the risk associated with
anaesthesia (Chipchase and James, 1997; Chung et al.,
1997). Incomplete evacuation occurs in 2 – 3% of curettage
procedures (Hurd et al., 1997).
In recent years, it was shown that expectant management
results in a complete abortion rate of 50–60%, reducing the
need for curettage and its complications (Nielsen and Hahlin,
1995; Ankum et al., 2001). Based on these data, it is com-
mon practice in The Netherlands to wait at least 1 week after
the diagnosis of early pregnancy failure before curettage is
planned.
Treatment with misoprostol, as an alternative to curett-
age, causes expulsion in 50 –99% of women with early
pregnancy failure up to the 14th week of gestation (Hinshaw
et al., 1993; Creinin et al., 1997; Herabutya and Prasertsawat,
1997; Zalanyi, 1998; Chung et al., 1999; Demetroulis et al.,
2001; Ngai et al., 2001; Sahin et al., 2001; Kovavisarach and
Sathapanachai, 2002; Muffley et al., 2002). However, most
data on the effectiveness of misoprostol are based on series of
women treated shortly after diagnosis of early pregnancy fail-
ure was established and also included cases with incomplete
miscarriage. Data on the effectiveness of misoprostol in
women with early pregnancy failure who were initially mana-
ged expectantly are lacking. Therefore, the aim of this study
was to compare the effectiveness of misoprostol treatment
Human Reproduction Page 1 of 6 DOI: 10.1093/humrep/deh344
Human Reproduction
q
European Society of Human Reproduction and Embryology 2004; all rights reserved Page 1 of 6
Hum. Reprod. Advance Access published June 10, 2004
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from

with the standard method of curettage in women who had
been managed expectantly for at least 1 week after the diag-
nosis of early pregnancy failure.
Materials and methods
The study was performed in three teaching hospitals in The
Netherlands (St Antonius Hospital Nieuwegein, St Elisabeth Hos-
pital Tilburg and Tweesteden Hospital Tilburg) between November
2001 and June 2003. All women gave written informed consent
before study entry. The institutional review boards of the three hos-
pitals had approved the study protocol.
All women aged 18 –45 years with a diagnosis of early pregnancy
failure between 6 and 14 weeks of gestation who had been managed
expectantly for at least 1 week were eligible for the trial.
A diagnosis of early pregnancy failure was made by transvaginal
ultrasonography, and defined as an intrauterine gestational sac (sac
diameter . 15 mm or , 15 mm not showing any growth after a 7 day
interval) with or without an embryonic pole and absence of cardiac
activity. Women with incomplete abortion were not eligible for the
study. Exclusion criteria were haemodynamic instability, a history of
Caesarean section, known uterine anomalies, multiple pregnan-
cies, infection, suspicion of extra-uterine pregnancy, coagulopathies,
allergy to misoprostol, severe pulmonary disease, congenital or
acquired heart disease, liver diseases, glaucoma, sickle cell disease,
prolonged use of corticosteroids or adrenal gland insufficiency.
Women were asked to participate in the study by their treating
gynaecologist at the time early pregnancy failure was diagnosed,
and written informed consent was obtained. At the time women
requested intervention (after at least 7 days of expectant manage-
ment), randomization was performed by their treating gynaecologist
using a computer program with a block randomization sequence,
thus guaranteeing the concealment of allocation. Randomization
was stratified for previous vaginal birth, duration of amenorrhoea
(, 10 or . 10 weeks) and participating centre.
Participants were allowed to use analgesic medication. Curettage
consisted of evacuation of the uterus by suction curettage under gen-
eral anaesthesia in a day care setting, and had to be performed
within a week after randomization.
Misoprostol treatment was given in an out-patient setting, and
consisted of four tablets of 200 mg of misoprostol placed in the pos-
terior fornix using a speculum (Creinin et al., 1997; Demetroulis
et al., 2001; Muffley et al., 2002; Wood and Brain, 2002). The
effect of this treatment was evaluated 24 h after the first misoprostol
dose, using transvaginal sonography. Treatment was considered suc-
cessful if no signs of retained products of conception, i.e. a focal
hyperechoic intrauterine mass with an anterior – posterior diameter
exceeding 15 mm (Nielsen et al., 1999), were observed. In the pre-
sence of signs of residual conception products, a second dose of
800 mg misoprostol was administered vaginally. Women who were
treated with a second dose were re-evaluated 24 h later, i.e. 48 h
after the initial dose.
Misoprostol treatment was considered to have failed where abnor-
mal bleeding and signs of retained products of conception (i.e. a
focal hyperechoic intrauterine mass with an anterior–posterior
diameter exceeding 15 mm at sonography) (Chung et al., 1998a,b;
Nielsen et al., 1999) were observed. In these situations, in the
absence of complete evacuation after . 3 days following initial mis-
oprostol administration, curettage was performed. Similarly, curet-
tage was considered to have failed where intervention was needed
because of abnormal bleeding and signs of retained products of con-
ception visible at sonography. However, sonography was performed
systematically in all patients who were allocated to misoprostol,
whereas in the curettage group this was only done in the case of
clinical symptoms suggestive of incomplete abortion. In all women
that needed curettage, a pathologist examined the obtained tissue.
Women were asked to complete a questionnaire after 2 days and
2 weeks, including duration and intensity of bleeding, intensity of
pain, need for analgesics and side effects of treatment. Intensity of
pain and severity of bleeding were measured using a visual analogue
scale (VAS); the VAS score zero indicates no pain or no bleeding
and 10 indicates severe pain or severe bleeding. Complications were
defined as infection, need for transfusion and surgery-related com-
plications such as perforation and cervical tear.
The primary end-point was defined as complete evacuation of the
products of conception. Kaplan–Meier curves were constructed
indicating time to complete evacuation. Cox regression analysis was
used to calculate a hazard rate ratio with a 95% confidence interval
(CI). Secondary end-points were occurrence of side effects, intensity
of pain and need for analgesics, and intensity and duration of bleed-
ing. Occurrence of these end-points was tabulated, and relative risks
and their 95% CIs were calculated. Analysis was performed accord-
ing to the ‘intention to treat’ principle. Continuous variables, i.e.
pain and vaginal bleeding expressed at the VAS scores, were com-
pared using the Wilcoxon test.
We were willing to accept a 20% incomplete evacuation rate after
the use of misoprostol, considering an incomplete evacuation rate of
4% after initial curettage. A sample size of 150 women, randomized
in a 1:1 ratio, would allow us to detect a difference from curettage
of 15% with a two-sided x
2
test with a P-value of 0.05 and a power
of 80%.
Results
Of the 241 eligible women, 87 (36%) declined to participate
in the trial and chose curettage. Therefore, 154 women were
included in the trial (Figure 1). Seventy-nine were allocated
to misoprostol treatment, and 75 were allocated to curettage.
Table I shows the baseline characteristics of the two groups.
In the group allocated to misoprostol, 26 women aborted
without additional treatment within 24 h after the first miso-
prostol dose (Table II). Among these 26 women, there were
four women in whom misoprostol was considered to be suc-
cessful initially, but who needed curettage during follow-up.
One of these four women had an emergency curettage. Four
other women needed emergency evacuation because of per-
sistent bleeding after the first dose of misoprostol. In two
women, a complete evacuation had not been reached after the
first dose; a second dose was not given which was a violation
of the protocol. Curettage was performed in both women.
The remaining 47 women (59.5%) needed a second miso-
prostol dose. Of these women, 20 reached complete evacua-
tion after the second dose. Seven women needed emergency
evacuation because of persistent bleeding after the second
dose of misoprostol. The remaining 20 women were sched-
uled for curettage. Of these 20 women, one needed a repeat
curettage after a failed curettage (and a failed second dose of
misoprostol). Finally, 37 women (46.8%) required curettage
for incomplete evacuation in the misoprostol group, two of
which were protocol violations since curettage was scheduled
before a second dose of misoprostol was administered. In the
misoprostol group, there were no surgical complications in
women who needed curettage after failed misoprostol.
G.C.M.Graziosi et al.
Page 2 of 6
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from

In the group allocated to curettage, a spontaneous evacua-
tion occurred in four women (5.3%) after randomization, but
before the scheduled curettage. Two of these women needed
emergency curettage because of residual trophoblast and per-
sistent bleeding. The other 71 women had a scheduled curet-
tage (Table II). In three (4%) of these 71 women, repeat
curettage was needed because of residual trophoblastic tissue
and persistent bleeding (Figure 1). The number of women
needed to treat with misoprostol to prevent one curettage was
two (95% CI 1.6 – 2.7, P , 0.01). Complications occurred in
three women (4%) allocated to curettage (Table III). One
woman had a uterine perforation and was managed expec-
tantly without further complications. One woman needed a
transfusion after a haemorrhage of 800 ml during curettage
that was treated with uterotonic agents. One woman had
amenorrhoea after curettage due to intrauterine synechia
(Asherman syndrome) that needed hysteroscopic resection.
The number of women needed to treat with misoprostol
to prevent one complication was 25 (95% CI 12–1,
P ¼ 0.07).
In all cases that underwent a curettage or repeat curettage,
i.e. secondary to misoprostol or curettage, pathological exam-
ination confirmed the presence of retained conception
products.
Vaginal misoprostol treatment caused mainly gastrointest-
inal side effects. Nausea and diarrhoea were reported in
14 and 27%, respectively, but were not severe (Table III).
There were no side effects reported in women allocated to
curettage.
Table I. Baseline characteristics: misoprostol compared with curettage
Characteristic Misoprostol
(n ¼ 79)
Curettage
(n ¼ 75)
Age (years), mean ^ SD 32.5 ^ 4.8 32.1 ^ 4.1
Range 23–42 22– 41
Nulliparous (%) 33 (42) 36 (48)
Gestational age (weeks)
mean ^ SD
10.2 ^ 1.8 10.1 ^ 1.8
Range 6.4–14.5 6–13.6
Clinical symptoms (% bleeding) 27 (34) 21 (28)
Previous miscarriage (%) 15 (19) 10 (13)
Duration expectant management
(days) mean ^ SD
11.4 ^ 7.4 10.1 ^ 4.1
Range 5– 65 6–20
Figure 1. Flow chart showing the flow of participants allocated to misoprostol or curettage.
Table II. Outcome measures of misoprostol versus curettage
Misoprostol
n ¼ 79 (%)
Curettage
n ¼ 75 (%)
RR 95% CI
Complete evacuation 42/79 (53.2) 72/75 (96) 0.63 0.52–0.77
Complete evacuation
after first dose
22/79 (27.8) – – –
Complete evacuation
after second dose
20/47 (42.5) – – –
Emergency curettage 12/79 (15) 2/75 (2.7) 5.7 1.3–25
Repeat curettage 1/79 (1.3) 3/75 (4) 0.32 0.03–3
RR ¼ relative risk.
Table III. Secondary outcome measures of misoprostol versus curettage
Misoprostol
n ¼ 79
Curettage
n ¼ 75
P-value
Side effects
Nausea 11/79 (14%) 0/75 NA
Diarrhoea 21/79 (27%) 0/75 NA
Bleeding characteristis
VAS score for severity
of bleeding, mean (SD)
6.2 (2.6) 3.9 (2.2) , 0.001
Duration of bleeding in
days, mean (SD)
10.4 (5.6) 8.7 (5.1) 0.12
Pain characteristics
VAS score for severity
of pain, mean (SD)
5 (3) 3 (2.4) , 0.001
Analgesia need 24/79 (30%) 16/75 (21%) 0.2
Complications 0/79 3/75(4%) NA
VAS ¼ visual analogue scale; scores range from 0 to 10, with higher scores
indicating more bleeding or pain; NA ¼ not applicable.
Misoprostol versus curettage for early pregnancy failure
Page 3 of 6
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from

Figure 2 shows Kaplan–Meier curves indicating time to
complete evacuation. Median time to complete evacuation
was 2 days in both groups (log rank test P ¼ 0.66, hazard
rate ratio 1.1, 95% CI 0.77–1.5). After 1 week, 67 women
(85%) treated with misoprostol had reached complete abor-
tion versus 65 women (93%) allocated to curettage.
Table III shows the mean duration of bleeding, severity of
bleeding and intensity of pain in the two groups. Women
allocated to misoprostol had significantly more pain
(P , 0.001).
Severity of bleeding was significantly higher in women
allocated to misoprostol as compared with curettage
(P , 0.001), whereas duration of bleeding was equal in
both groups (10.4 versus 8.7 days, respectively, P ¼ 0.12)
(Table III).
Discussion
This study compared misoprostol and curettage in women
with early pregnancy failure managed expectantly for at least
1 week. Women treated with misoprostol had a 53% com-
plete evacuation rate without additional surgery, whereas in
the curettage group 96% had complete evacuation after the
initial procedure. The number of women needed to treat with
misoprostol to prevent one curettage was two. Pain and
bleeding were more severe in the misoprostol group.
Thirty-six percent of the eligible participants were not
recruited into the study because of refusal to provide consent.
It is important to realize that misoprostol was only available
for women in the trial. Thus women who did not participate
were always treated with curettage. Furthermore, our study is
different from other studies due to the general policy in The
Netherlands to wait for at least 1 week after the diagnosis of
early pregnancy failure before treatment is started.
Our study had two other methodological limitations. First,
the doctor who saw the patients at follow-up was not blinded
to the treatment allocation, which could influence their
decision of subsequent curettage after misoprostol treatment.
However, since incomplete abortion was defined as clinical
symptoms combined with sonography and different doctors
in a multicentre setting performed the follow-up, we think
this would not greatly affect the outcome and conclusions.
Moreover, blinding of the patient is virtually impossible in
the comparison between curettage and misoprostol treatment,
thus hampering the possibility to mask treatment allocation
to the doctors involved.
Secondly, although the protocol required patients to be
randomized only after at least 1 week of expectant manage-
ment, two patients were randomized after a waiting period of
5 and 6 days, respectively. However, the intention to treat
principle that we followed does not allow exclusion of these
patients from the study.
The 53% complete evacuation rate of misoprostol in our
study is slightly lower compared with the 60–88% success
rates described in other trials, using comparable misoprostol
treatment protocols and criteria for diagnosis of incomplete
abortion (Creinin et al., 1997; Demetroulis et al., 2001;
Muffley et al., 2002; Wood and Brain, 2002). The fact that
expectant management occurred before misoprostol treat-
ment was started might explain the lower evacuation rate in
our study since selection of a adherant type of early preg-
nancy failure or uterus insensitive to prostaglandins could
diminish the efficacy of misoprostol in this particular group
of women.
However, as sonography after evacuation was only part of
the misoprostol strategy, and definition of 15 mm anterior –
posterior diameter is an arbitrary one, incomplete evacuation
in the misoprostol group could have been overestimated. Fur-
thermore, a longer interval between misoprostol treatment
and the conclusion that misoprostol failed (i.e. incomplete
abortion) could have increased complete evacuation rates.
Bagratee et al. (2004) showed that complete evacuation rates
increased from 73% 2 days after misoprostol treatment to 89%
1 week after treatment with misoprostol. It must be noted,
however, that, assuming that the pharmacological effect of
misoprostol is related to its concentration in the plasma, the
therapeutic effect of misoprostol is expected to last for , 8h.
Thus it is not unlikely that the increased evacuation rate in
that study could be attributed to the expectant management
followed by the initial misoprostol treatment, rather than a pri-
marily misoprostol effect (Zieman et al., 1997). Conside-
ring that , 40% of women with early pregnancy failure will
have a spontaneous abortion after expectant management
(Wieringa-de Waard et al., 2002), and that misoprostol was
effective in 53% of women who did not abort spontaneously
in our study, the overall reduction in the number of cases
requiring curettage after an expectant management followed
by misoprostol treatment could be as high as 75%. Therefore,
misoprostol after a period of expectant management seems to
be clinically useful if safety, side effects and patient satisfac-
tion with the treatment strategy are acceptable.
Women who underwent curettage after misoprostol treat-
ment had no recorded surgical complications. This was also
reported in a study with 635 participants comparing miso-
prostol and curettage for treatment of early pregnancy failure,
showing complication rates in women treated with curettage
after failed misoprostol of 0% versus 5.4% in women allo-
cated to curettage (Chung et al., 1999). This reduction in
Figure 2. Kaplan –Meier curves indicating the time to complete
evacuation for women allocated to misoprostol or curettage.
G.C.M.Graziosi et al.
Page 4 of 6
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from

surgical complications could be explained by the cervical
priming effect of misoprostol allowing easy surgical access
to the uterine cavity. This effect is well documented in thera-
peutic abortions (el Refaey et al., 1994).
Furthermore, drug-related side effects including nausea
and diarrhoea occurred in about one-quarter of women, but
seemed only moderately troublesome. Since the process of
abortion will give uterine contractions and bleeding, whereas
curettage was performed under general anaesthesia, pain and
bleeding were more severe in women treated with misopro-
stol. Emergency evacuation was seen more frequently in
women treated with misoprostol. In women allocated to cure-
ttage, only two women needed emergency evacuation due to
the waiting time until scheduled curettage.
A practical advantage of misoprostol over curettage seems
to be its direct and easy applicability, offering women instant
therapy without further waiting. However, in our study, time
to complete evacuation was not decreased in women treated
with misoprostol as compared with curettage, and the number
of women that had not reached complete evacuation after
1 week was doubled. The time to complete abortion in
women treated with misoprostol was influenced by the failure
rate (47%) and delayed diagnosis of incomplete abortion. In
our study, a hyperechoic intrauterine mass , 15 mm was not
always conclusive for the absence of residual tissue, which
led to a delay in additional curettage in four women who
were thought to have aborted completely after misoprostol
treatment.
Women with early pregnancy failure might be willing to
accept some disadvantages of misoprostol to avoid curettage.
Only one study comparing misoprostol with curettage has
looked at women’s satisfaction as an outcome parameter
(Lee et al., 2001). In that study, 218 women randomly allo-
cated to curettage or misoprostol were compared for psycho-
logical well-being. Complete evacuation was achieved in
50% of women treated with misoprostol. No differences in
psychological well-being were observed. Understandably,
patients who required curettage after misoprostol were less
satisfied compared with women allocated to curettage or
women with successful misoprostol treatment. The authors
stated that cultural differences could possibly influence their
study results as the study was conducted in China. A recently
published study (Bagratee et al., 2004) compared expectant
management with misoprostol treatment and showed that
women were equally satisfied in both groups. However, sig-
nificantly more women allocated to misoprostol compared
with women allocated to expectant management (90 versus
73%, odds ratio 1.26, 95% CI 1.05 –1.50) would elect to
have the same treatment in the future rather than curettage.
This indicates that misoprostol treatment is well tolerated
and could be clinically useful. It is likely that the high com-
plete evacuation rate for misoprostol (89%) in this study is
attributed to this preference for misoprostol treatment at the
expense of curettage. Since misoprostol was compared with
expectant management and not with curettage, this could
cloud their results.
In conclusion, curettage is significantly superior in terms
of evacuation rates as compared with misoprostol in treat-
ment of early pregnancy failure. Furthermore, treatment
inconvenience is to be considered using misoprostol in terms
of increased time to complete abortion, gastrointestinal side
effects, more pain, bleeding and emergency evacuation.
Women treated with curettage have to consider a higher com-
plication rate due to the surgical procedure.
Our study does not show whether misoprostol treatment
should be preferred over curettage in women with early preg-
nancy failure in whom expectant management failed. Thus
individual women’s preferences will be of vital importance
in the choice between misoprostol and curettage.
The results of our study should be combined with results
of future studies to provide more precise estimates on the
subject. Furthermore, research on optimal dose finding in
order to increase evacuation rates for misoprostol treatment
are warranted.
References
Ankum WM, Wieringa-de Waard M and Bindels PJE (2001) Management
of spontaneous miscarriage in the first trimester: an example of putting
informed shared decision making into practice. Br Med J
322,1343–1346.
Bagratee JS, Khullar V, Regan L, Moodley J and Kagoro H (2004) A ran-
domized controlled trial comparing medical and expectant management of
first trimester miscarriage. Hum Reprod 19,266 – 271.
Chipchase J and James D (1997) Randomised trial of expectant versus surgi-
cal management of spontaneous miscarriage. Br J Obstet Gynaecol
104,840–841.
Chung T, Leung P, Cheung LP, Haines C and Chang AM (1997) A medical
approach to management of spontaneous abortion using misoprostol.
Extending misoprostol treatment to a maximum of 48 hours can further
improve evacuation of retained products of conception in spontaneous
abortion. Acta Obstet Gynecol Scand 76,248 –251.
Chung TK, Cheung LP, Sahota DS, Haines CJ and Chang AM (1998a) Spon-
taneous abortion: short-term complications following either conservative
or surgical management. Aust NZ J Obstet Gynaecol 38,61– 64.
Chung TK, Cheung LP, Sahota DS, Haines C and Chang AM (1998b) Evalu-
ation of the accuracy of transvaginal sonography for assessment of
retained products of conception after spontaneous abortion. Gynecol
Obstet Invest 45,190– 193.
Chung TK, Lee DT, Cheung LP, Haines CJ and Chang AM (1999) Spon-
taneous abortion: a randomized, controlled trial comparing surgical eva-
cuation with conservative management using misoprostol. Fertil Steril
71,1054–1059.
Creinin MD, Moyer R and Guido R (1997) Misoprostol for medical evacua-
tion of early pregnancy failure. Obstet Gynecol 89,768–772.
Demetroulis C, Saridogan E, Kunde D and Naftalin AA (2001) A prospective
randomized control trial comparing medical and surgical treatment for
early pregnancy failure. Hum Reprod 16,365 –369.
Edmonds DK, Lindsay KS, Miller JF, Williamson E and Wood PJ (1982)
Early embryonic mortality in women. Fertil Steril 38,447–453.
el Refaey H, Calder L, Wheatley DN and Templeton A (1994) Cervical
priming with prostaglandin E1 analogues, misoprostol and gemeprost.
Lancet 343,1207–1209.
Herabutya Y and Prasertsawat P (1997) Misoprostol in the management of
missed abortion. Int J Gynaecol Obstet 56,263–266.
Hinshaw K, Cooper K, Henshaw R, el Refaey H, Rispin R and Smith N
(1993) Management of uncomplicated miscarriage. Randomized trials are
possible. Br Med J 307,259.
Hurd WW, Whitfield RR, Randolph JF Jr and Kercher ML (1997) Expectant
management versus elective curettage for the treatment of spontaneous
abortion. Fertil Steril 68,601– 606.
Kovavisarach E and Sathapanachai U (2002) Intravaginal 400 microg miso-
prostol for pregnancy termination in cases of blighted ovum: a randomised
controlled trial. Aust NZ J Obstet Gynaecol 42,161– 163.
Lee DT, Cheung LP, Haines CJ, Chan KP and Chung TK (2001) A compari-
son of the psychologic impact and client satisfaction of surgica,l treatment
Misoprostol versus curettage for early pregnancy failure
Page 5 of 6
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from

with medical treatment of spontaneous abortion: a randomized controlled
trial. Am J Obstet Gynecol 185,953–958.
Muffley PE, Stitely ML and Gherman RB (2002) Early intrauterine preg-
nancy failure: a randomized trial of medical versus surgical treatment. Am
J Obstet Gynecol 187,321– 325.
Ngai SW, Chan YM, Tang OS and Ho PC (2001) Vaginal misoprostol as
medical treatment for first trimester spontaneous miscarriage. Hum Reprod
16,1493–1496.
Nielsen S and Hahlin M (1995) Expectant management of first trimester
spontaneous abortion. Lancet 345,84– 86.
Nielsen S, Hahlin M and Platz-Christensen J (1999) Randomised trial com-
paring expectant with medical management for first trimester miscarriages.
Br J Obstet Gynaecol 106,804 –807.
Sahin HG, Sahin HA and Kocer M (2001) Randomized outpatient clinical
trial of medical evacuation and surgical curettage in incomplete miscar-
riage. Eur J Contracept Reprod Health Care 6,141– 144.
Warburton D and Fraser FC (1964) Spontaneous abortion risks in man: data
from reproductive histories collected in a medical genetics unit. Hum
Genet 16,1–25.
Wieringa-de Waard M, Vos J, Bonsel GJ, Bindels PJ and Ankum WM
(2002) Management of miscarriage: a randomized controlled trial of
expectant management versus surgical evacuation. Hum Reprod
17,2445–2450.
Wood SL and Brain PH (2002) Medical management of missed abortion:
a randomized clinical trial. Obstet Gynecol 99,563–566.
Zalanyi S (1998) Vaginal misoprostol alone is effective in the treatment of
missed abortion. Br J Obstet Gynaecol 105,1026–1028.
Zieman M, Fong SK, Benowitz NL, Banskter D and Darney PD (1997)
Absorption kinetics of misoprostol with oral or vaginal administration.
Obstet Gynecol 90,88–92.
Submitted on January 5, 2004; accepted on May 5, 2004
G.C.M.Graziosi et al.
Page 6 of 6
by guest on June 1, 2013http://humrep.oxfordjournals.org/Downloaded from