Reproductive performance after hysteroscopic metroplasty in the hypoplastic uterus: a study of 29 cases


*Correspondence: Professor H. Fernandez, Department of Obstetrics and Gynaecology, Antoine Béclère Hospital, 157 rue de la Porte de Trivaux, 92140 Clamart Cedex, France.


Objective To evaluate the reproductive performance after hysteroscopic metroplasty performed for hypoplastic uterus.

Design Crossover study (15 patients) and descriptive analysis (14 patients).

Setting Tertiary care university hospital.

Population Twenty-nine women (mean age: 31.4 years; range: 27–38.5 years) with a hypoplastic malformed uterus and a history of primary infertility and/or recurrent abortion and/or preterm delivery were included in this study. Fourteen women suffered from primary infertility and 15 women had previous pregnancies. Twenty-three women had been exposed to diethylstilboestrol in utero.

Methods Women underwent hysteroscopic metroplasty between January 1996 and May 1999.

Main outcome measures Rate of pregnancies and live births, anatomical results.

Results The mean follow up was 40 months (range: 13–67 months). Twenty-one women (72.4%) experienced 30 pregnancies. Thirteen women gave birth to 16 live infants. At the time of the follow up, four women were pregnant in the second trimester. Compared with previous pregnancies, the rate of deliveries increased from 3.8% to 63.2%. No complications occurred during metroplasty. The hysteroscopic anatomical results were good in all cases.

Conclusions Our results suggest that hysteroscopic metroplasty, with its simplicity and minimal post-operative sequelae, seems to be an operation of choice in women with a hypoplastic malformed uterus and a history of severe infertility and/or recurrent pregnancy loss.


Congenital anomalies of the uterus can be associated with reproductive failures such as recurrent abortion and preterm birth1. Infertility has also been linked to uterine anomalies2. The reported incidence of congenital anomalies range from 0.1% to 0.5% of all deliveries1. The most common forms are septate, bicornuate and didelphic uterus. A hypoplastic uterus is a rare uterine malformation, except in exposure to diethylstilboestrol in utero3. The pathogenesis remains unclear and its cause is still unknown. Several studies showed very poor reproductive performances when the uterine malformation was not treated4,5. Reproductive performance after hysteroscopic metroplasty in women with a hypoplastic uterus has not been well established, concerning only three reports6–8, in contrast to women with a septate uterus9.

The aim of this study was to evaluate the influence of hysteroscopic metroplasty on reproductive performance of women with a hypoplastic uterus.


Between January 1996 and May 1999, 29 women wishing a pregnancy (mean age: 31.4 years, range: 27–38.5 years) with a hypoplastic malformed uterus were treated by hysteroscopic metroplasty. The malformation was classified by hysterosalpingography and hysteroscopy. According to the American Fertility Society Classification of Müllerian anomalies10, a hypoplastic uterus (type VII) was defined when a small cavity was seen in the hysterosalpingogram. All patients underwent transvaginal ultrasound permitting a better assessment in the diagnosis of the hypoplastic uterus. By sagittal ultrasound, the length of the uterus did not exceed 6 cm. The hysteroscopic findings revealed a cylindrical uterine cavity with a bulging of the uterine side walls and no possibility to visualise the tubal ostia.

The uterine malformation was coincident with tubal infertility in six cases, polycystic ovarian syndrome in two cases and male infertility in one case.

Fourteen women suffered from primary infertility and 15 women had previous pregnancies (Table 1). Of these 29 women, 23 were known to have been exposed to diethylstilboestrol in utero. Among these 23 women, 12 suffered from primary infertility.

Table 1.  Fertility before and after hysteroscopic metroplasty. Values are given as n or n (%).
 Women with previous pregnancies (n= 15)Women with primary infertility (n= 14)
 Pregnancies before metroplastyOutcome after metroplastyOutcome after metroplasty
  1. aP < 0.02.

  2. bP < 0.01.

No. of pregnancies261911
No. of pregnant women129
Pregnant at the time of follow up 2 (10.5)2
Ectopic pregnancies3 (11.5)01
Miscarriage before 12 weeks16 (61.6)5 (26.3)a4
Miscarriage between 12 and 26 weeks2 (7.7)00
Death in utero2 (7.7)00
Legal abortion2 (7.7)00
Live birth1 (3.8)12 (63.2)b4
Term deliveries084
Preterm deliveries between 32 and 37 weeks040
Preterm deliveries before 32 weeks100

The 15 women with previous pregnancies totalled 26 pregnancies with only one live birth at 29 weeks. The mean duration of infertility was 27.2 months (range: 12–48 months) for the 29 women and 33.3 months (range: 12–60 months) for the 14 women with primary infertility.

The indication for hysteroscopic metroplasty was a hypoplastic uterus in combination with a history of primary infertility, recurrent abortion and/or preterm delivery before 30 weeks with or without live birth. Oral informed consent was obtained from the women before this surgical procedure was performed. Each woman served as her own control.

Surgery was scheduled to be done early in the follicular phase. The surgical procedure was performed under general anaesthesia by a senior surgeon after cervical dilatation, with an operative hysteroscope (n= 27), fitted with a monopolar hook (Resectoscope 26 F, optical lens 2.9 mm, Ref. 260020FA; Iglesias's jacket, Ref. 26055 SL; monopolar hook, Ref. 26055 L, Karl Storz, Tuttlingen, Germany). The mean pre-operative hysterometry was 5.8 cm (range: 5–7 cm). The uterine cavity was distended using a glycocole solution (Glycocolle 1.5%, Aguettant Laboratory, Lyon, France), the flow of which was controlled electronically. In two cases, we used the Versapoint bipolar vaporisation system (Gynecare Laboratory, Issy-Les-Moulineaux, France), effective in saline solution, through the 5F operating channel of a 5.5-mm hysteroscope, as previously described by Fernandez et al.11 The hook was introduced into the uterine horn and the incision was performed under direct vision from the fundus to the isthmus, perpendicularly to the lateral wall of the uterus and decreasing the depth of the incision as the section advanced. The identical incision was repeated on the other lateral wall of the uterus. This procedure allowed the formation of a normal uterine triangular and symmetric cavity. The depth of incision did not exceed 5 to 7 mm. A single intravenous dose of an antibiotic (Ampicillin) was given during the surgical procedure. All women were discharged on the day of surgery with post-operative medication including sequential oestroprogestative medication for two months (50 μg of ethinyloestradiol and 2.5 mg lynestrol; Ovanon, Organon Laboratory, Saint-Denis, France). In all cases, diagnostic hysteroscopy was repeated two months post-operatively to identify a marginal synechia and to evaluate the configuration of the uterine cavity. The criteria for second procedure was the presence of synechia observed during the diagnostic hysteroscopy. Two women have had a second procedure to achieve the metroplasty two and six months after the initial procedure. To assess reproductive performance, the women were asked by phone to complete a health history questionnaire. This questionnaire concerned post-operative complications and pregnancies including their number, results (live birth, legal abortion, miscarriage and ectopic pregnancy), gestional age at delivery, mode of delivery and the use of assisted reproductive techniques. For women requiring a second metroplasty, follow up dates are derived from the time of the second procedure.

Survival analysis using the Kaplan–Meier estimator was performed to calculate cumulative pregnancy rate. According to the fertility studies, we studied the first pregnancy after the surgical procedure, regardless of its implantation, and also the first normally intrauterine pregnancy. Statistical analyses were performed using the χ2 test. For all statistical analyses, Statview Version 4.57 Software (Abacus Concepts, Berkeley, California) was used, and differences at P < 0.05 were considered statistically significant.


Subsequent fertility outcome was evaluated in all women. The mean follow up was 40 months (range: 13–67 months). Three women (10.3%) without diethylstilboestrol in utero exposure, who initially wanted to become pregnant, had no longer any pregnancy desire after the surgical procedure (two women with previous pregnancies and one with primary infertility). Figure 1 illustrates the distribution of the first pregnancies and live birth over the time after the surgical procedure.

Figure 1.

Cumulative rate of first pregnancies and live birth after hysteroscopic metroplasty.

Twenty-one women (72.4%) experienced 30 pregnancies after metroplasty (Table 1). Twelve of these were in cases of secondary infertility and 12 became pregnant without the use of assisted reproductive techniques. The mean time before the first conception was 14.6 months (range: 2–41 months) after the metroplasty.

Nine women with primary infertility experienced 11 pregnancies, leading to four live born babies. These pregnancies were spontaneous except in six cases, four after in vitro fertilisation and two after intrauterine insemination.

Of the 26 women wishing a pregnancy, 13 (50%) gave birth to 16 live infants, conceived spontaneously in nine cases (three women had given birth to two children). Nine of them delivered 12 viable term neonates, requiring six caesarean sections for five women. The modality of term pregnancy deliveries was caesarean section in 50% of the cases. At the time of the follow up, four women were pregnant in the second trimester. Compared with previous pregnancies, the rate of deliveries increased from 3.8% to 63.2% (P < 0.01) and the abortion rate decreased from 61.6% to 26.3% (P < 0.02). The neonatal courses were good in all cases. The mean birthweight was 3.196 kg (range: 2.100–4.170 kg).

Among 23 women exposed to diethylstilboestrol in utero, 17 experienced 26 pregnancies after the surgical procedure, leading to 13 live born babies.

Hysteroscopic anatomical results were good in all cases. No complications occurred during metroplasty. None of the women needed readmission. In two cases, a Stage I intrauterine adhesion (according to the Classification of the American Fertility Society10) was observed after metroplasty and was easily lysed at two months (time of control diagnostic hysteroscopy for all women). The first patient who had a synechia experienced two pregnancies eight months after the hysteroscopic metroplasty (one pregnancy ended in miscarriage and one ended in preterm delivery). The second patient did not want to become pregnant.


Different methods and instruments for the hysteroscopic metroplasty have been used, including scissors6 and a resectoscope with a monopolar hook7,8. The choice of the technique seems to depend on operating time, cost of instrumentation and rate of complication. For two more recent cases, we used the Versapoint bipolar electrode system, which seems to have multiple benefits. In contrast with the bipolar electrosurgical system, the normal saline used has ion concentrations similar to human plasma and may reduce electrolyte changes and hyponatraemia. A second advantage is that cervical dilatation is not required11, decreasing the risk of cervical incompetence.

In the case of uterine dysmorphism, infertility and obstetric complications are believed to be more common compared with those with a normal uterine cavity12.

Our results are encouraging in terms of fertility. Twenty-one women (72.4%) became pregnant after the metroplasty. Thirteen women gave birth to 16 live infants. Nine of them delivered 12 viable term neonates. Among these 13 women, one woman with primary infertility gave birth to two live infants. These results are in accordance with other studies using hysteroscopic metroplasty6–8. To the best of our knowledge, this is the largest series to date (with the longest follow up) which details the reproductive performance of patients with hypoplastic uterus after hysteroscopic metroplasty. Nagel and Malo6 published the first study of eight women with a history of recurrent pregnancy loss. Their results showed a successful outcome of term deliveries in three out of six women with recurrent miscarriages, and no success in two women with primary infertility. In the same way, Katz et al.7 published a study concerning eight women with a ‘T-shaped’ uterus. The post-operative performance available for seven of the eight women showed four term pregnancies and one ectopic pregnancy in three women with secondary infertility. The more recent series concerning 24 women showed that 10 women had 12 intrauterine pregnancies (10 term deliveries, 1 preterm delivery and 1 miscarriage before 12 weeks)8. Among these women, three with primary infertility gave birth to three live infants. These results suggest that an improved uterine contour may result in an improved pregnancy outcome and term deliveries in women with prior spontaneous pregnancy losses or primary infertility.

Homer et al.13 reviewed the literature on the reproductive performance of septate uterus after hysteroscopic metroplasty. The overall results concerning a total of 658 patients show an impressive improvement after hysteroscopy. Data obtained from retrospective series suggest that the hysteroscopic metroplasty for the septate uterus is associated with a favourable outcome. Evaluation of the efficacy of hysteroscopic metroplasty for the hypoplastic uterus presents a number of problems, being the small size of reported series and as the septate uterus the lack of any prospective, randomised, controlled trial. However, our study does provide some information on the reproductive performance of women who have undergone hysteroscopic metroplasty for hypoplastic uterus.

With regard to the modality of delivery, in our study, the caesarean section rate of the term pregnancies was high. Indeed, only seven women delivered vaginally, without complications. Nevertheless, in seven women, the indication for a caesarean section was neither the metroplasty itself nor an obstetric indication, but the preciousness of the pregnancy after a personal history of infertility. Delivery can be per vaginam after metroplasty, but in the light of the literature, we recommend not to hesitate to practice a caesarean section as a mode of delivery. This type of surgery probably induces a uterine fragility. The obstetric management should be careful, although no uterine rupture arose in the literature6–8. The cases of uterine rupture after hysteroscopic metroplasty concerned septate uterus13,14.

Post-operative diagnostic hysteroscopy showed that hysteroscopic metroplasty gives good anatomical results in the majority of cases according to different authors6–8.


Our results show that the hysteroscopic metroplasty seems to be an operation that improves the rate of live births for women with a hypoplastic uterus and a history of primary infertility and/or recurrent abortion and/or preterm delivery. These results also confirm the preliminary experiences previously published in retrospective, uncontrolled pilot studies6–8. In the past, however, correction of these uterine abnormalities was not undertaken. Ideally, in order to evaluate the efficiency of this new technique, a randomised study ought to be undertaken in multiple centres, taking into account a larger number of women suffering from this type of abnormality.