Journal of Obstetrics and Gynaecology, 2014; Early Online: 14

2014 Informa UK, Ltd.

ISSN 0144-3615 print/ISSN 1364-6893 online
DOI: 10.3109/01443615.2014.936839

Does surgery improve live birth rates in patients with recurrent

miscarriage caused by uterine anomalies?
M. Sugiura-Ogasawara1, B. L. Lin2, K. Aoki3, T. Maruyama4, M. Nakatsuka5, N. Ozawa6, T. Sugi7,
T. Takeshita8 & M. Nishida9
1Department of Obstetrics and Gynecology, Nagoya City University, Graduate School of Medical Sciences, 2Department of Obstetrics

and Gynecology, Kawasaki Municipal Hospital, Kawasaki, 3Aoki Ladies Clinic, Nagoya, 4Department of Obstetrics and Gynecology, Keio
University School of Medicine, Tokyo, 5Faculty of Health Sciences, Okayama University Medical School, Okayama, 6Department of
Obstetrics and Gynecology, National Hospital Organization Tokyo Medical Center, Tokyo, 7Sugi Womens Clinic Laboratory for Recurrent
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Washington on 10/26/14

Pregnancy Loss, Kanagawa, 8Department of Obstetrics and Gynecology, Nippon Medical School, Tokyo and 9Department of Obstetrics and
Gynecology, National Hospital Organization Kasumigaura Medical Center, Ibaragi, Japan

We found that congenital uterine anomalies have a negative
impact on reproductive outcome in recurrent-miscarriage
couples, being associated with further miscarriage with a
normal embryonic karyotype. There has been no study compar-
ing live birth rates between patients with and without surgery.
We conducted a prospective study to prove that surgery for a
For personal use only.
Affected patients are offered surgery in an attempt to restore
the uterine anatomy, as shown in the literature (Jones and Jones
1953; Strassmann 1961; Goldenberg et al. 1995; Jacobsen and
DeCherney 1997; Grimbizis et al. 1998; Karande and Gleicher
1999; Lee and Hickok 2000; Patton et al. 2004; Kormayos et al.
2006). However, these studies are biased due to the fact that either

bicornuate or septate uterus might improve the live birth rate.
A total of 170 patients with congenital uterine anomalies
suffering two or more miscarriages were examined. The live
birth rate after ascertainment of anomalies, cumulative live
birth rate and infertility rate, were compared between patients
with and without surgery. In patients with a septate uterus, the
live birth rate (81.3%) at the first pregnancy after ascertainment
of anomalies with surgery tended to be higher than that (61.5%)
in those without surgery. The infertility rates were similar in
both groups, while the cumulative live birth rate (76.1%) tended
to be higher than without surgery (60.0%). Surgery showed no
benefit in patients with a bicornuate uterus for having a baby,
but tended to decrease the preterm birth rate and the low birth
weight. The possibility that surgery has benefits for having a
baby in patients with a septate uterus suffering recurrent
miscarriage could not be excluded.
Keywords: Bicornuate uterus, congenital uterine anomaly,
metroplasty, recurrent pregnancy loss, septate uterus
Introduction
Established causes of recurrent miscarriages are antiphospholipid
syndrome (APS), abnormal chromosomes in either partner and
in particular, translocations and chromosomal abnormalities
in the embryo (Farquharson et al. 1984; Ogasawara et al. 2000;
Sugiura-Ogasawara et al. 2004). We found that congenital uterine
anomalies have a negative impact on reproductive outcome in
recurrent-miscarriage couples, being associated with further mis-
carriage with a normal embryonic karyotype (Sugiura-Ogasawara
et al. 2010).
there were no controls, or patients with recurrent miscarriage
treated by surgery served as their own controls.
In our previous study, 78% of patients with anomalies who
did not undergo surgery could give live birth cumulatively
(Sugiura-Ogasawara et al. 2010). The general assumption is that
operations can increase successful pregnancies, but there have
been no randomised control trials comparing pregnancy outcome
between cases with and without surgery in patients with a history
of recurrent miscarriage (Kowalik et al. 2011).
Since information concerning prognosis in such women is
limited and there has been no casecontrol study comparing live
birth rates between patients with and without surgery for bicor-
nuate or septate uterus in individuals with a history of recur-
rent miscarriage, despite regular performance of this technique
worldwide, the present multicentre investigation was performed.
This was the first prospective study comparing the live birth rate
between patients with and those without surgery.
Materials and methods
Patients
This prospective study was performed in Nagoya City Univer-
sity Hospital, Kawasaki Municipal Hospital, Aoki Ladies Clinic,
Hirosaki University Hospital, Keio University Hospital, Okayama
University Hospital, Tokyo Medical Center, Toyama University
Hospital, Sugi Womens Clinic, Nippon Medical School Hospital
and National Hospital Organization Kasumigaura Medical Cen-
ter. Patients with a history of 2 (29) consecutive miscarriages
or 1 stillbirth, who visited the hospitals during the period
between January 2003 and June 2009, were enrolled in this study.
Hysterosalpingography (HSG), transvaginal ultrasound (US),
chromosome analysis for both partners, determination of aPL,

Correspondence: M. Sugiura-Ogasawara, Department of Obstetrics and Gynecology, Nagoya City University, Graduate School of Medical Sciences,
Mizuho-ku, Nagoya, 4678601, Japan. E-mail: og.mym@med.nagoya-cu.ac.jp
 2 M. Sugiura-Ogasawara et al.
and blood tests for hyporthyroidism, diabetes mellitus and hyper-
prolactinaemia were performed for all patients before their subse-
quent pregnancy.
HSG and/or 2D US were used as the initial screening tools.
Laparoscopy/laparotomy and/or MRI were performed to ascer-
tain the type of anomaly (investigating both the uterine cavity and
the external uterine contours) in accordance with the American
Fertility Society classification of Mllerian anomalies (American
Fertility Society 1988; Pellerito et al. 1992; Woelfer et al. 2001).
Tompkins Index 25% was the criterion used to distinguish
between arcuate uterus and mild septate or bicornuate uterus
(Tompkins 1962). A bicornuate uterus was distinguished from a
septate uterus when a fundal indentation dividing the two cornua
was detectable (American Fertility Society 1988).
Surgery
Patients with a bicornuate uterus wishing surgical treatment
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Washington on 10/26/14
before subsequent pregnancies underwent a Strassmann metro-
plasty mainly in the National Hospital Organization Kasumigaura
Medical Center (Strassmann 1961). The majority of patients with
a septate uterus underwent transcervical resection (TCR), or a
Jones modified metroplasty in the Kawasaki Municipal Hospital,
the National Hospital Organization Kasumigaura Medical Cen-
ter and Nippon Medical School Hospital (Jones and Jones 1953;
Goldenberg et al. 1995).
Follow-up
The outcomes of patients with a bicornuate or septate uterus were
followed up by telephone as far as possible. In the present study:
For personal use only.
(1) the first pregnancy outcome after ascertainment of anomalies
was determined for both septate and bicornuate uterus cases,
comparing patients with or without surgical correction; (2) the
infertility rates were compared between patients with and without
surgery; infertility was determined as not pregnant after 1 year
of trying to conceive; (3) all pregnancy outcomes after system-
atic examination were assessed and the final live birth rate of the
patients, including infertile patients calculated.
The study was approved by the Research Ethics Committee at
the Nagoya City University Medical School.
The analysis was carried out using the SPSS for Windows, ver-
sion 19.0. Multivariate logistic analyses were performed to exam-
ine the association of a subsequent live birth with the presence of
surgery, age and the number of previous miscarriages. A value of
p 0.05 was considered to be statistically significant.
Results
A total of 227 patients had congenital uterine anomalies: 56 had
partial or complete bicornis unicolli; 145 a septum; 12 a unicornis
and 14 a didelphys. None of them had hypoplasias/agenesis or
DES drug-related anomalies. One patient with a bicornuate uterus
also had a translocation in one partner.
Table I. Baseline characteristics of patients with congenital uterine anomalies (mean SD).
Septate (n 124)
With surgery Without surgery
(n 109) (n 15)
Age (years) 32.9 4.0 33.7 4.0
Previous miscarriages 2.74 1.48 2.93 1.16
Previous stillbirths 0.20 0.49 0.13 0.35
Previous live births 0.10 0.38 0.13 0.35
A total of 170 patients with a bicornuate or septate uterus could
be followed up. Baseline characteristics are shown in Table I. We
found no differences between the two groups.
Subsequent pregnancy outcomes are summarised in Table II.
A total of 78 of the 96 pregnant patients with a septate uterus
(81.3%) treated with surgery had a successful outcome, while
this was the case for eight of the 13 (61.5%) without surgery at
the subsequent first pregnancy (difference not significant). Eight
of 12 pregnant patients with a bicornuate uterus (66.7%) treated
with surgery and 22 of 28 patients (78.6%) without surgery gave a
live birth, respectively (not significant).
In total, 83 of 96 (86.5%) patients with a septate uterus treated
with surgery; nine of 13 (69.2%) patients with a septate uterus
and without surgery; nine of 12 (75.0%) patients with a bicor-
nuate uterus treated with surgery and 24 of 28 (85.7%) patients
without surgery could have a living baby cumulatively within the
follow-up period (not significant).
Infertility rates for patients with a septate uterus treated with
and without surgery, and with a bicornuate uterus treated with
and without surgery were 11.9% vs 13.3% and 14.3% vs 6.3%,
respectively. There were no significant differences.
Final live birth rates per person including infertility rates were
76.1% vs 60.0% in septate cases and 64.3% vs 75.0% in bicornuate
cases, respectively. Regarding a septate uterus, the live birth rate
of patients who received surgery tended to be higher than that of
patients without surgery after including infertile patients. Age at
diagnosis was found to be an independent risk factor (p 0.013,
OR, 1.148; 95% confidence interval (CI): 1.0291.280).
The frequency of preterm birth and low birth weight in
patients without surgery tended to be higher than that in patients
with surgery (p 0.058, p 0.023, Table III). The frequency of
caesarean section in patients with surgery was significantly higher
than that in patients without surgery (p 0.006). However, when
bicornuate and septate uterus cases were distinguished, the differ-
ences were not significant.
Discussion
The present multicentre study illustrated the possibility that sur-
gery might be effective for increasing the chance of having a live
baby in septate uterus cases suffering recurrent miscarriage. To
our knowledge, this is the first study of pregnancy outcomes of
patients with congenital uterine anomalies comparing surgery
and no surgery groups (Kowalik et al. 2011). Although the study
design was not a randomised control trial, the results are gener-
ally positive. Selection bias was relatively small because there was
no difference in the baseline characteristics.
Chan et al. (2011) conducted a systematic review and con-
cluded that the prevalence of uterine anomalies was not higher
in women with infertility (8.0%; 95% CI: 5.312.0, p 0.239)
but higher in women with a history of miscarriage (13.3%; 95%
CI: 8.920; p 0.011) compared with that in an unselected
Bicornuate (n 46)
With surgery Without surgery
p value (n 14) (n 32) p value
NS 33.1 4.2 31.0 3.9 0.097
NS 2.86 2.07 2.75 1.14 NS
NS 0.50 1.34 0.16 0.51 NS
NS 0.14 0.36 0.06 0.25 NS
 Recurrent miscarriage and uterine anomaly 3
Table II. Comparison of successful pregnancy outcomes between patients with and without surgery.

Septate (n 124) Bicornuate (n 46)

Without With Without

With surgery surgery surgery surgery
(n 109) (n 15) (n 14) (n 32)

n (%) n (%) p value OR 95% CI n (%) n (%) p value OR 95% CI

Live birth rate at the rst 78/96 81.3 8/13 61.5 0.289 2.034 0.5477.556 8/12 66.7 22/28 78.6 0.266 0.378 0.0682.095
pregnancy after
examination
Infertility rate after 13 11.9 2 13.3 NS 2 14.3 2 6.3 NS
diagnosis
Cumulative live birth rate 83/96 86.5 9/13 69.2 0.100 3.081 0.8711.157 9/12 75.0 24/28 85.7 0.610 0.614 0.0944.007
Cumulative live birth rate 83/109 76.1 9/15 60.0 0.189 2.214 0.6777.240 9/14 64.3 24/32 75.0 0.319 0.475 0.0112.052
per patient
One fetus of a patient with a bicornuate uterus was terminated at 20 weeks gestation because of a fetal anomaly. This case was excluded from the analysis. The case with both
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Washington on 10/26/14

a bicornuate uterus and translocation miscarried because of embryonic unbalanced chromosomes and was also excluded from the analysis.

population (5.5%; 95% CI: 3.58.5). The prevalence of anoma-
lies in women with two or more miscarriages (10.9%; 95% CI:
3.633.3) was not significantly different from that in women
with three or more miscarriages (15.4%, 95% CI: 10.323,
p 0.572). The prevalence of arcuate uterus, as diagnosed
by optimal tests, was not increased in infertile women (1.8%;
95% CI: 0.84.1) or in women with a history of miscarriage
(2.9%; 95% CI: 0.99.6), as compared with that in an unselected
population (3.9%; 95% CI: 2.17.1). Arcuate uterus seems to be
a normal variant and not a uterine anomaly Raga et al. (1997).
For personal use only.
Surgery here improved the live birth rate of patients with a
septate uterus from 61.5% to 81.3%, though the difference was
not significant. One of the most important complications of sur-
gery is speculated to be infertility caused by operative adhesion.
However, the infertility rates were found to be similar to that
without surgery in the present study. The cumulative live birth
rate also tended to be higher after including the infertile cases.
Furthermore, the rates for preterm birth and low birth weight
were significantly higher in patients without surgery.
A recent cohort study indicated that the presence of an

Thus, we excluded patients with arcuate uterus in the present
study. According to the systematic review carried out by Chan
et al. (2011), the prevalence of major uterine anomalies in
women with miscarriage was 10.4%.
Several studies concerning the obstetric outcome after removal
of a uterine septum have been reported (Goldenberg et al. 1995;
Jacobsen and DeCherney 1997; Grimbizis et al. 1998; Karande
and Gleicher 1999; Lee and Hickok 2000; Patton et al. 2004;
Kormayos et al. 2006). Lee and Hickok (2000) described a 77.3%
uncomplicated delivery rate per pregnant woman after hyst-
eroscopic septum resection. Kormanyos et al. (2006) compared
pregnancy outcome after removal of septum between cases with
and without a residual septum in patients with a history of two
or three miscarriages and concluded the live birth rate in cases
with no remnant was significantly higher. The live birth rate per
pregnancy after surgery was reported to be 6880%. However,
the live birth rate per patient could be calculated to be 5464%
(Sugiura-Ogasawara et al. 2013). Both live birth rates were lower
than those (81.3% per pregnancy and 76.1% per patient) in our
present study.
anomaly was associated with preterm birth 34 weeks (adjusted
odds ratio, OR, 7.4; 95% CI: 4.811.4; p 0.01); preterm birth
37 weeks (OR, 5.9, 95% CI: 4.3 8.1; p 0.01); primary
non-breech caesarean delivery (OR, 2.6; 95% CI: 1.74.0;
p 0.01); preterm premature rupture of membranes (OR, 3.2;
95% CI: 1.85.6; p 0.01) and breech presentation (OR, 8.6;
95% CI: 6.212.0; p 0.01) (Hua et al. 2011).
Regarding a bicornuate uterus, Strassmann metroplasty was
mainly performed here and surgery did not appear to influence
the live birth rate or the cumulative live birth rate. Thus, surgery
demonstrated no benefit in patients with a bicornuate uterus, for
having a baby. While, preterm birth was less frequent in patients
receiving surgery, it occurred from 32 to 35 weeks gestation.
It proved difficult to examine the significance of the distinc-
tion between septate and bicornuate uteri because of the absence
of internationally established criteria. However, it is important
to distinguish between the two, especially regarding selection
of surgical methods because TCR for the bicornuate uterus may
cause a risk of perforation of the uterine fundus. HSG is the diag-
nostic modality which most often leads to a tentative diagnosis

Table III. Preterm birth, low birth weight and caesarean section rates in patients with a bicornuate or a septate uterus.

Bicornuate or septate Septate Bicornuate

Without With Without

With surgery surgery With surgery Without surgery surgery surgery

n (%) n (%) p value n (%) n (%) p value n (%) n (%) p value

Preterm birth 4/87 4.6 5/32 15.6 0.058 4/79 5.1% 1/8 12.5 NS 0/8 0 4/24 16.7 NS
Low birth weight 4/87 4.6 6/32 18.8 0.023 4/79 5.1 1/8 12.5 NS 0/8 0 5.24 20.8 NS
Caesarean section 54/87 62.1 11/32 34.4 0.006 49/79 62.0 5/8 62.5 NS 5/8 62.5 6/24 22.7 0.068
Other remarks Abruptio
placenta
(31 w, 37 w)
Four patients with twin pregnancies suered preterm delivery and were excluded.
 4 M. Sugiura-Ogasawara et al.
of congenital anomalies. We ascertained the type of anomaly in
accordance with the American Fertility Society classification of
Mllerian anomalies, using laparoscopy or MRI for a final diag-
nosis (American Fertility Society 1988). Further study is needed
to establish criteria based on comprehensive evidence.
Since the live birth rate will clearly depend on the method of
metroplasty and the technique of each gynaecologist, we need to
examine the outcome per centre, but this was hampered in this
case by the small sample size. The live birth rate might depend on
the length of the septum.
Congenital uterine anomalies have a negative impact
on reproductive outcomes in recurrent-miscarriage couples
(Sugiura-Ogasawara et al. 2010; Chan et al. 2011; Sugiura-
Ogasawara et al. 2013). Our study showed the possibility that
surgery might improve the live birth rate, preterm birth rate and
low birth weight, with no influence on the infertility rates, in
patients with a septate uterus. However, the difference was not
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Washington on 10/26/14
statistically significant. This might be due to the small number of
patients not treated surgically (8/13). Surgery did not appear to
have any benefit for having a live birth in patients with a bicornu-
ate uterus, although it tended to reduce preterm birth. The differ-
ence was not statistically significant. This might also be due to the
fact that there were only eight patients for analysis without sur-
gery. A larger cohort study or a randomised control trial should
now be conducted to ratify the present findings.
Acknowledgements
We thank Dr Shunsaku Fujii (Department of Obstetrics
For personal use only.
and Gynecology, Hirosaki University Hospital) and Dr Shigeru
Saito (Department of Obstetrics and Gynecology, University of
Toyama Hospital) for collecting clinical data.
Declaration of interest: The authors report no conflicts of inter-
est. The authors alone are responsible for the content and writing
of the paper.
This study was supported by a grant-in-aid for Scientific
Research from the Ministry of Health, Labour and Welfare of
Japan.
References
American Fertility Society. 1988. The American Fertility Society classica-
tions of adnexal adhesions, distal tubal occlusion, tubal occlusion sec-
ondary to tubal ligation, tubal pregnancies, Mullerian anomalies and
intrauterine adhesions. Fertility and Sterility 49:944955.
Chan YY, Jayaprakasan K, Zamora J, Thornton JG, Raine-Fenning N,
Coomarasamy A. 2011. The prevalence of congenital uterine anomalies
in selected and high-risk populations: a systematic review. Human Repro-
duction Update 17:761771.
Farquharson RG, Pearson JF, John L. 1984. Lupus anticoagulant and preg-
nancy management. Lancet 28:228229.
Goldenberg M, Sivan E, Sharabi Z, Mashiach S, Lipitz S, Seidman DS.
1995. Reproductive outcome following hysteroscopic management of
intrauterine septum and adhesions. Human Reproduction 10:26632265.
Grimbizis G, Camus M, Clasen K, Tournaye H, De Munck L, Devroey P.
1998. Hysteroscopic septum resection in patients with recurrent abor-
tions or infertility. Human Reproduction 13:11881193.
Hua M, Odibo AO, Longman RE, Macones GA, Roehl KA, Cahill AG.
2011. Congenital uterine anomalies and adverse pregnancy outcomes.
American Journal of Obstetrics and Gynecology 205:558.e1e5.
Jacobsen LJ, DeCherney A. 1997. Results of conventional and hysteroscopic
surgery. Human Reproduction 12:13711387.
Jones HW, Jones GE. 1953. Double uterus as an etiological factor in repeated
abortion: indications for surgical repair. American Journal of Obstetrics
and Gynecology 65:325339.
Karande VC, Gleicher N. 1999. Resection of uterine septum using gynaeco-
radiological techniques. Human Reproduction 14:12261229.
Kormayos Z, Molar BG, Pal A. 2006. Removal of a residual portion of uter-
ine septum I women of advanced reproductive age: obstetrics outcome.
Human Reproduction 21:10471051.
Kowalik CR, Goddijn M, Emanuel MH, Bongers MY, Spinder T, de Kruif JH
et al. 2011. Metroplasty versus expectant management for women with
recurrent miscarriage and a septate uterus. Cochrane Database of Sys-
tematic Reviews (6):CD008576.
Lee R, Hickok LR. 2000. Hysteroscopic treatment of the uterine septum: a
clinicians experience. American Journal of Obstetrics and Gynecology
182:14141420.
Ogasawara M, Aoki K, Okada S, Suzumori K. 2000. Embryonic karyotype of
abortuses in relation to the number of previous miscarriages. Fertility and
Sterility 73:300304.
Patton PE, Novy MJ, Lee DM, Hickok LR. 2004. The diagnosis and repro-
ductive outcome after surgical treatment of the complete septate uterus,
duplicated cervix and vaginal septum. American Journal of Obstetrics
and Gynecology 190:16691678.
Pellerito JS, McCarthy SM, Doyle MB, Glickman MG, DeCherney AH.
1992. Diagnosis of uterine anomalies: relative accuracy of MR imaging,
endovaginal sonography, and hysterosalpingography. Radiology 183:
795800.
Raga F, Bauset C, Remohi J, Bonilla-Musoles F, Simon C, Pellicer A. 1997.
Reproductive impact of congenital Mullerian anomalies. Human Repro-
duction 12:22772281.
Strassmann EO. 1961. Operations for double uterus and endometrial atresia.
Clinical Obstetrics and Gynecology 4:240.
Sugiura-Ogasawara M, Ozaki Y, Kitaori T, Kumagai K, Suzuki S. 2010.
Midline uterine defect size correlated with miscarriage of euploid
embryos in recurrent cases. Fertility and Sterility 93:19831988.
Sugiura-Ogasawara M, Ozaki Y, Sato T, Suzumori N, Suzumori K. 2004. Poor
prognosis of recurrent aborters with either maternal or paternal recipro-
cal translocation. Fertility and Sterility 81:367373.
Sugiura-Ogasawara M, Ozaki Y, Suzumori N. 2013. Mllerian anomalies and
recurrent miscarriage. Current Opinion in Obstetrics and Gynecology
25:293298.
Tompkins P. 1962. Comments on the bicornuate uterus and twinning. Surgi-
cal Clinics of North America 42:1049.
Woelfer B, Salim R, Banerjee S, Elson J, Regan L, Jurkovic D. 2001. Repro-
ductive outcomes in women with congenital uterine anomalies detected
by three-dimensional ultrasound screening. Human Reproduction 98:
10991103.