ORIGINAL ARTICLE

Management of Anembryonic Pregnancy Loss:

An Observational Study
Ying-Ti Huang, Shang-Guo Horng, Fa-Kung Lee, Ying-Tzu Tseng*
Division of Obstetrics and Gynecology, Hsinchu Cathay General Hospital, Hsinchu, Taiwan, R.O.C.

Background: This study was undertaken to determine if expectant management with a longer waiting period is an effec-
tive and safe option for women with anembryonic pregnancy.
Methods: Women with an ultrasound diagnosis of anembryonic pregnancy were offered the option of expectant manage-
ment with a 3-week waiting period or surgical evacuation according to their preference.
Results: A total of 121 women with anembryonic pregnancies participated in the study; 45 of them elected expectant
management. The overall success rate was 83.3% in the expectant group and 97.3% in the surgical group. No significant
complications were noted in either group.
Conclusion: Expectant management with a 3-week waiting period is an efficacious and safe option with a low risk of
infection and hemorrhage. However, it is difficult to predict the exact time period before spontaneous abortion. [J Chin
Med Assoc 2010;73(3):150–155]

Key Words: abortion, anembryonic pregnancy, expectant management, uterine curettage

Introduction anesthesia-related complications, have always been of

concern to patients. In 1995, Nielsen and Hahlin con-
Pregnancy loss is a common obstetric complication and ducted a small randomized study of expectant and sur-
affects > 30% of conception.1 The majority of these gical management of early nonviable pregnancies, which
losses occur in the 1st trimester, including spontaneous suggested that the outcomes were similar, including
abortion, anembryonic gestation and embryonic or complications and the need for a second or emergent
fetal death.2 The prevalence of early pregnancy failure curettage.8 Additional studies have supported the role
was 2.8% in a study involving 17,810 women at of expectant management as a treatment option in early
10–13 weeks’ gestation, and anembryonic pregnancies pregnancy loss; however, reported success rates have
accounted for 37.5% of the pregnancies lost.3 Loss ranged widely from 25% to 76%.9,10
before the development of an embryo is more likely According to Luise et al’s study in 2002, expectant
to be associated with genetic abnormalities than those management in patients with anembryonic pregnancies
later in gestation,4 and imparts a considerable influ- had a less favorable success rate compared to patients
ence on recurring risk in subsequent pregnancies. with incomplete abortion or embryonic fetal death.10
Anembryonic pregnancy is defined as a gestational Patients with failure of expectant management eventu-
sac (GS) containing no fetal pole with a mean diam- ally required surgical evacuation to remove the products
eter ≥ 15 mm, or a GS < 15 mm not showing any of conception. In contrast, based on a randomized trial
growth in 7 days.5,6 Dilatation and curettage has been conducted by Wieringa-de Waard et al, up to 40% of
the primary treatment option for early pregnancy loss surgeries can be avoided by a waiting period of 7 days,
in many countries since it was first introduced into which could be offered to well-informed women.6
clinical practice in the 1930s.7 However, the risks of Patients were willing to accept the waiting period with
uterine evacuation, including sepsis, hemorrhage and adequate counseling and help.

*Correspondence to: Dr Ying-Tzu Tseng, Division of Obstetrics and Gynecology, Hsinchu Cathay
General Hospital, 678, Chunghua Road, Section 2, Hsinchu 300, Taiwan, R.O.C.
E-mail: [email protected] Received: October 22, 2009
● Accepted: December 23, 2009
●

150 J Chin Med Assoc • March 2010 • Vol 73 • No 3

© 2010 Elsevier Taiwan LLC and the Chinese Medical Association. All rights reserved.
 Management of anembryonic pregnancy

The success rate of surgical treatment has been If there was evidence of retained tissue or if the endome-
superior to that of expectant management in both trial lining was ≥ 15 mm, then a 200-μg tablet of miso-
reviewed literature and clinical practice. However, prostol was given 3 times a day orally for the next 3 days
expectant management can be justified as an alternative (maximum dose, 1,800 μg). Patients with retained tissue
treatment option if the success rate can reach 80%. were considered a treatment failure.
The purpose of this study was to determine whether If the expulsion was still incomplete after taking
or not expectant management with a 3-week waiting misoprostol or if expulsion failed to occur after 21 days
period is an effective and safe option for women with of expectant management, then surgical uterine evacu-
anembryonic pregnancy, and if outcomes differ from ation was performed. Patients had 1 scheduled follow-
those of surgical evacuation. up visit in 7 days, and ultrasonography was performed
to assess the endometrial lining.
The symptoms and signs during the waiting period
Methods of expectant management before spontaneous loss
were thoroughly explained to the patients in the expec-
Subjects tant group, who also received a contact phone num-
Women presenting to the outpatient clinic and emer- ber in case further care or information was necessary.
gency department of Hsinchu Cathay General Hospital Emergent admission and surgery were arranged if
(CGH) in Taiwan between July 1, 2008 and June 30, necessary.
2009 with anembryonic pregnancy were included in
the study. The trial was approved by the Institutional Outcome measures
Review Board of Hsinchu CGH (number CT 9813), The primary outcome measure was complete expul-
and all of the subjects gave informed consent. sion of the products of conception within 21 days after
Diagnosis of anembryonic pregnancy was made the diagnosis of anembryonic pregnancy was made.
when, as seen on transvaginal ultrasound, the mean Treatment success was defined as endometrial thick-
diameter of the GS was ≥ 15 mm without a visible ness < 15 mm in the follow-up visit after spontaneous
embryo or the GS was < 15 mm and with no growth in abortion or surgical curettage had been performed.
7 days.5,6 All of the sonographic scans were performed For women with incomplete expulsion or endometrial
by obstetricians in Hsinchu CGH using a 6.5-MHz thickness ≥ 15 mm in both groups, we also determined
transducer (Aloka Co. Ltd., Tokyo Japan). Endometrial whether or not additional management with medica-
thickness was measured in the sagittal plane from the tion improved the overall success rate. The secondary
interface of the endometrium and myometrium across outcomes assessed were the number of patients requir-
the uterine cavity. ing a second curettage and immediate surgery due to
The inclusion criteria were hemodynamic stability, bleeding > 500 mL, and the incidence of pelvic in-
temperature < 37.5°C in the past 24 hours, no history flammatory disease within 4 weeks after evacuation of
of current serious systemic disease, ≥ 20 years of age, uterus and pain severity.
singleton pregnancy, no intrauterine device present, Patients were offered nonsteroidal anti-inflammatory
and no contraindication to the use of prostaglandin and drugs for pain relief. If the pain was unbearable, they
ergonovine. Patients were excluded if they were expe- were required to return to the clinic, where narcotic
riencing severe pain, fever or heavy bleeding requiring or opiate medications were provided if necessary.
immediate surgery. Patients could opt out of the study Infection was diagnosed if patients presented with
at any time on request. lower abdominal pain or leukorrhea, and 1 of any of the
following criteria was met: temperature > 38°C, white
Protocol blood cell count ≥ 11,000/mm3, and administration
Patients who met the inclusion criteria were informed of antibiotics within 4 weeks of expulsion, whether by
of the risks and benefits of expectant management and expectant management or surgical evacuation.
surgical treatment. Patients then elected expectant or
surgical management. The patients in the expectant Statistical analysis
management group were seen in the outpatient clinic Statistical analysis was performed using SPSS version
on days 14 and 21 for reevaluation before the prod- 13 (SPSS Inc., Chicago, IL, USA). Student’s t test,
ucts of conception were expelled. Once the tissue was Fisher’s exact test and χ2 test were used to analyze the
passed, patients returned to the outpatient clinic in 7 differences in patient characteristics and outcomes.
days and ultrasonography was performed to determine A p value < 0.05 was considered to indicate statistical
if there was any retained tissue in the uterine cavity. significance.

J Chin Med Assoc • March 2010 • Vol 73 • No 3 151

 Y.T. Huang, et al

Results 16. All 8 patients subsequently received uterine curet-

tage. Also, 1 patient did not return after being allo-
A total of 121 eligible women who presented to cated to the expectant management group, so she was
Hsinchu CGH between July 1, 2008 and June 30, excluded as well. In the 36 women who were included
2009 were enrolled into this study (Figure 1). There for analysis, 22 experienced spontaneous loss within 14
were no significant differences in patient characteris- days of diagnosis and 8 within 21 days. Of the remain-
tics between the expectant and surgical management ing 6 patients, 2 expelled the products of conception
groups except for the size of the GS (Table 1). The on days 3 and 4, respectively, but an endometrial
mean size of the GS in the expectant management thickness ≥ 15 mm was revealed by ultrasound when
group was 2.7 cm, versus 2.1 cm in the surgical man- they returned to the clinic. These 2 patients experi-
agement group. enced vaginal bleeding for more than 14 days and were
Of the enrolled women, 76 underwent surgical administered oral misoprostol, after which the bleeding
evacuation and 45 elected to have expectant manage- stopped within 7 days. In the last 4 patients, ultrasound
ment. The overall success rate of expectant manage- revealed visible GS on day 21. These 4 women only
ment was 83.3%, compared with 97.3% for surgical experienced minimal vaginal spotting; they underwent
evacuation. surgical evacuation and were classified as treatment
Of the 45 women in the expectant management failure of expectant management.
group, 8 withdrew from the study and were excluded In the surgical management group, 2 patients re-
from the analysis. Five of these 8 women dropped out quired a second curettage due to persistent vaginal bleed-
on day 7, and the other 3 dropped out on days 6, 9 and ing for > 21 days (Table 2) despite the administration

Eligible and consented

(n = 121)

By patient’s preference

Allocation Expectant management Curettage

(n = 45) (n = 76)

Expectant Dropped out Loss to All 76 patients

Treatment management for surgery follow-up underwent surgery
(n = 36) (n = 8) (n = 1)

Successful EM ≥ 15 mm Successful A second

0–14 days treatment within Use of misoprostol treatment curettage
14 days (n = 22) (n = 2) (n = 74) (n = 2)

Successful treatment
0–21 days within 21 days
(n = 30)

Failure and arranged

Follow-up
evacuation (n = 4)

Analysis

Figure 1. The allocation of patients based on their preferences. EM = endometrial thickness.

152 J Chin Med Assoc • March 2010 • Vol 73 • No 3

 Management of anembryonic pregnancy

Table 1. Patients’ characteristics*

Surgical management Expectant management

p†
(n = 76) (n = 36)

Age (yr) 32.7 ± 0.5 (26–39) 32.9 ± 0.48 (24–42) NS

Parity
0 43 (56.5) 22 (59.4) NS
≥1 15 (40.6) 33 (43.5) NS
Previous miscarriage 21 (27.6) 3 (8.1) NS
GS size (cm) 2.1 ± 0.1 2.7 ± 0.2 0.02
Time to return to clinic (d) 6.7 ± 0.3 8.2 ± 0.5 NS
Successful treatment
EM < 15 mm 74 (97.3) 30 (83.3) 0.013
Vaginal bleeding present 23 (30.2) 17 (47.2) NS
EM thickness at follow-up (mm) 8±2 6±5 NS
*Data presented as mean ± standard deviation (range) or n (%) or mean ± standard deviation; †success rate was analyzed by Fisher’s exact test and differ-
ences in patient characteristics analyzed by c2 test. GS = gestational sac; EM = endometrial thickness; NS = not significant.

Table 2. Outcomes*

Surgical management Expectant management

(n = 76) (n = 36)

Successful treatment with EM < 15 mm 74 (97.3) 30 (83.3)

Treatment with medication due to 2 (2) 2 (6.2)
EM ≥ 15 mm after abortion
Successful treatment with medication† 4 (97.3) 32 (88.85)
Second curettage 2 (2) 0
Emergent curettage 0 0
Failure of expulsion – 4 (11.1)
Complications
Hemorrhage > 500 mL 0 0
Pain requiring opiates or narcotics 0 0
Infection within 4 wk of abortion 5 (6.5) 1 (2.7)
Uterine perforation 0 –
Bleeding status before treatment 23 (30.2) 17 (47.2)
Time to spontaneous abortion (d)‡ – 9.15 ± 2.9 (2–21)
*Data presented as n (%) or mean ± standard deviation (range); †success rate was assessed after patients with incomplete abortion or EM ≥ 15 mm took
misoprostol; ‡from time of diagnosis of anembryonic pregnancy. EM = endometrial thickness.

of misoprostol. Pathological reports revealed residual drugs for pain relief, and none of them underwent
gestational tissue, and bleeding only stopped on days emergent curettage because of intolerable pain or heavy
5 and 6 after the second operation. bleeding. Five patients in the surgical group experi-
The presence of vaginal bleeding before treatment enced lower abdominal pain after uterine curettage
had no significant effect on the success of expulsion; and received antibiotic treatment. The incidences
47.2% (17/36) of patients in the expectant manage- of pelvic inflammatory disease were 6.5% and 2.5%
ment group had vaginal spotting, but the success rate in the surgical and expectant management groups,
reached 83.3% (30/36; p = 0.662, Fisher’s exact test). respectively.
Three patients with failed expectant management had
intermittent vaginal bleeding, but the tissue was not
expelled within the 3-week waiting period. Discussion
No woman in either group required blood trans-
fusion. No patient in the expectant management group The causes of pregnancy loss vary with gestational
required more than nonsteroidal anti-inflammatory age. The expression “blighted ovum” has been replaced

J Chin Med Assoc • March 2010 • Vol 73 • No 3 153

 Y.T. Huang, et al

with anembryonic or preembryonic pregnancy loss, patients with endometrial thickness ≥ 15 mm required
defined in terms of developmental biology, and possibly surgical intervention.17 Based on literature reports, it
occurs when there are genetic problems.11 Recently, the would appear that it was not necessary to use miso-
management of early pregnancy loss has changed from prostol in our 2 patients with endometrial thickness
a surgical approach to conservative treatment. The rate ≥ 15 mm as spontaneous resolution would likely have
of spontaneous expulsion varies depending on the type occurred. However, the medication may reduce the
of miscarriage; with anembryonic pregnancy, a lower suc- length of vaginal bleeding. In our surgical management
cess rate may be due to an intact sac and closed cervix.12 group, 2 patients underwent a second curettage due to
Rates as low as 24.7% have been reported by Jurkovic prolonged bleeding. When we compared the intrauter-
et al,9 whereas Nielsen and Hahlin reported the success ine sonographic pattern of these 4 women (2 in the
rates of expectant management to be 91% for incom- expectant management group; 2 in the surgical man-
plete miscarriage, 76% for missed abortion, and 66% agement group) with endometrial thickness ≥ 15 mm,
for anembryonic pregnancy.8 In our study, the success the echogenicity was obviously more heterogeneous
rate by day 14 was 61.1%, consistent with the study of in the surgical management patients. Bleeding persisted
Nielsen and Hahlin. Our success rate reached 83.3% despite the administration of misoprostol. Operative
by day 21, indicating that a longer waiting period can reintervention is indicated in symptomatic abnormal
reduce the need for surgical evacuation.6 endometrial content.18
Successful outcome in the current study was defined In this study, mean GS was 2.7 cm and 2.1 cm in
as an endometrial thickness < 15 mm without retained the expectant and surgical management groups, respec-
products of conception after natural expulsion or sur- tively. Although GS size was a significantly different
gical evacuation. If we were to include the 2 patients factor in the patient characteristics of the 2 groups,
with endometrial thickness ≥ 15 mm who took miso- the group allocation in the study was based on the
prostol (after which their bleeding stopped) and the requests of the patients rather than on randomization.
patient with endometrial thickness < 15 mm within The 4 patients who had expectant management fail-
10 days in the expectant group, the overall success rate ure had GS > 3 cm in diameter (3.5, 3.28, 3.54 and
would be 88.8%. In that case, successful treatment was 3.1 cm). The presence of a GS that is deformed or flat-
considered as no need for uterine curettage. According tened, with largest diameter > 5 cm without a detectable
to the systematic review of 18 studies by Geyman embryo, is often an indicator of miscarriage.19 Due to
et al,13 expectant management for 1st trimester preg- our small sample size, we could not determine if GS
nancy loss has an overall success rate of 93%. The > 3 cm is associated with a higher failure rate of spon-
surgical success rate reported by Creinin et al14 was taneous expulsion.
98%, which is consistent with our finding. There was no increase in complications such as
In Zhang et al’s study of 1st trimester pregnancy infection or bleeding in the expectant management
failure including anembryonic gestation, embryonic or group, or in the 4 patients with treatment failure during
fetal death, incomplete abortion or inevitable sponta- the 3-week waiting period. The 8 patients who dropped
neous abortion, misoprostol 800 mg vaginally was out of the expectant management group did so because
given to the women on day 1 and a second dose on day they lost patience and were tired of waiting. With thor-
3 if expulsion was incomplete. By day 3, 71% had expe- ough explanation and adequate consultation, the waiting
rienced complete expulsion; by day 8, the success rate time for tissue expulsion might be extended on their
had reached 84%.15 Immediate administration of med- preference. Retained tissue in the endocervical canal
ication saves waiting time and its success rate is similar with pain and bleeding sometimes occur in patients
to that of expectant management. managed expectantly, but this did not occur in our
In our study, we used a cut-off value for endome- study group. In most cases, the tissue can be removed
trial thickness of < 15 mm to define treatment success. with ring forceps without further treatment.17 We assumed
However, there is evidence to suggest that 15 mm is that anembryonic pregnancy often presents with an
too stringent.10,15–17 Patients with endometrial thick- empty sac, or only a minimal fetal pole, which allows the
ness 15–30 mm after misoprostol treatment usually small amount of tissue to pass smoothly.
expel the products of conception completely without In conclusion, our results indicate that expectant
complications.17 Creinin et al analyzed 80 women management with a 3-week waiting period is a safe and
with early pregnancy loss who were treated with miso- effective option with a low risk of infection and hem-
prostol, and suggested that there was no obvious rela- orrhage for the management of anembryonic preg-
tionship between increasing endometrial thickness and nancy in early gestation. The success rate is higher than
the need for surgical intervention as none of their previously reported, though lower than for surgical

154 J Chin Med Assoc • March 2010 • Vol 73 • No 3

 Management of anembryonic pregnancy

curettage. Adequate counseling and help are important 7. Hertig AT, Livingstone RG. Spontaneous, threatened, and habit-
for women who choose expectant management of preg- ual abortion: their pathogenesis and treatment. N Engl J Med
1944;230:797–806.
nancy loss as it is difficult to predict the exact time 8. Nielsen S, Hahlin M. Expectant management of first trimester
when expulsion may occur and so the optimal follow- spontaneous abortion. Lancet 1995;345:84–6.
up time is not yet standardized. 9. Jurkovic D, Ross JA, Nicolaides KH. Expectant management
of missed miscarriage. Br J Obstet Gynaecol 1998;105:670–1.
10. Luise C, Jermy K, May C, Costello G, Collins WP, Bourne TH.
Outcome of expectant management of spontaneous first
Acknowledgments trimester miscarriage: observational study. BMJ 2002;324:873–5.
11. Silver RM. Fetal death. Obstet Gynecol 2007;109:153–67.
12. Condous G, Okaro E, Bourne T. The conservative manage-
The authors gratefully acknowledge the generous assis-
ment of early pregnancy complications: a review of the literature.
tance of all the nurses in the Department of Obstetrics Ultrasound Obstet Gynecol 2003;22:420–30.
and Gynecology, Hsinchu CGH, Taiwan. 13. Geyman JP, Oliver LM, Sullivan SD. Expectant, medical, or
surgical treatment of spontaneous abortion in first trimester of
pregnancy? A pooled quantitative literature evaluation. J Am
Board Fam Prac 1999;12:55–64.
References 14. Creinin MD, Schwartz JL, Guido RS, Pymar HC. Early preg-
nancy failure—current management concepts. Obstet Gynecol
1. Wilcox AJ, Weinberg CR, O’Connor JF, Baird DD, Schlatterer Surv 2001;56:105–13.
JP, Canfield RE, Armstrong EG, et al. Incidence of early loss of 15. Zhang J, Gilles JM, Barnhart K, Creinin MD, Westhoff C,
pregnancy. N Engl J Med 1988;319:189–94. Frederick MM. Management of Early Pregnancy Failure Trial.
2. Alberman E. Spontaneous abortions: epidemiology. In: A comparison of medical management with misoprostol and sur-
Grudzinskas G, Chard T, eds. Spontaneous Abortion: Diagnosis gical management for early pregnancy failure. N Engl J Med
and Treatment. London: Springer-Verlag, 1992:19–20. 2005;353:761–9.
3. Pandya PP, Snijders RJ, Psara N, Hilbert L, Nicolaides KH. 16. Harwood B, Meckstroth KR, Mishell DR, Jalin JK. Serum
The prevalence of non-viable pregnancy at 10–13 weeks of ges- beta-human gonadotropin levels and endometrial thickness after
tation. Ultrasound Obstet Gynecol 1996;3:170–3. medical abortion. Contraception 2001;63:255–6.
4. Klein J, Stein Z. Epidemiology of chromosomal abnormalities 17. Creinin MD, Harwood B, Guido RS, Fox MC, Zhang J.
in spontaneous abortion: prevalence, manifestation and deter- Management of Early Pregnancy Failure Trial. Endometrial
minants. In: Bennett MJ, Edmonds DK, eds. Spontaneous and thickness after misoprostol use for early pregnancy failure. Int J
Recurrent Abortion. Oxford: Blackwell Scientific, 1987:29–50. Gynaecol Obstet 2004;86:22–6.
5. Coulam CB, Goodman C, Dorfmann A. Comparison of ultra- 18. Debby A, Golan A, Sadan O, Rotmensch S, Malinger G.
sonographic findings in spontaneous abortions with normal and Sonographic characteristics of the uterine cavity following first-
abnormal karyotypes. Hum Reprod 1997;12:823–6. trimester uterine evacuation. Ultrasound Obstet Gynecol 2008;
6. Wieringa-de Waard M, Vos J, Bonsel GJ, Bindels PJ, Ankum 31:555–9.
WM. Management of miscarriage: a randomized controlled 19. Cho FN, Chen SN, Tai MH, Yang TL. The quality and size of
trial of expectant management versus surgical evacuation. Hum yolk sac in early pregnancy loss. Aust N Z J Obstet Gynaecol
Reprod 2002;17:2445–50. 2006;46:413–8.

J Chin Med Assoc • March 2010 • Vol 73 • No 3 155