Introduction

Intrauterine adhesions (IUA) were first described and reported by Joseph Asherman in 1948 [1]. Asherman’s syndrome applies to partial or complete obliteration of the uterine cavity by adhesions secondary to trauma to a gravid or non-gravid uterine cavity [2]. A majority of patients of Asherman’s syndrome present with infertility, with associated menstrual problems [3]. The treatment of Asherman’s syndrome improved dramatically with the emergence of hysteroscopy [4]. Nowadays, hysteroscopy is the method of choice to diagnose, treat and follow up patients with Asherman’s syndrome [5]. There have been a number of studies that reported on the reproductive outcome after treatment of Asherman’s syndrome [5–9], but very few analyzed systematically the reproductive outcome of infertility patients. This retrospective clinical study analyzed 89 patients with infertility due to Asherman’s syndrome, to evaluate the fertility outcome, associated menstrual factors affecting conception, rate of conception (according to severity of adhesions) and pregnancy outcome.

Materials and methods

Initially, 134 patients with Asherman’s syndrome who presented with infertility were recruited from January 2002 to December 2006 for the study. The inclusion criterion for this study was hysteroscopically diagnosed intrauterine adhesions. In all patients, a comprehensive infertility workup was done. These included tubal patency test, pelvic ultrasonography, husband semen analysis and serum hormone measurements (FSH, LH, prolactin, estradiol, progesterone and androgen) on the 2nd to 5th day of the cycle or at a randomly chosen time in patients with amenorrhea. Any patient having abnormality in these tests, which may be responsible for reproductive failure, were excluded from the study.

In seven cases, husband’s semen parameters were abnormal and were excluded from the data. Two cases of hyperprolactinemia, one case of ovarian failure (FSH > 40 IU) and five cases of endometrial tuberculosis diagnosed from histology of endometrial biopsy were excluded for analysis. In 24 cases, tubes were found blocked on hysterosalpingography, but concurrent laparoscopy at the time of hysteroscopic adhesiolysis showed that 11 patients had tubal block and therefore excluded. In 12 cases, there were abnormal findings on pelvic ultrasonography (tubo-ovarian mass and endometriosis) and were also excluded from further analysis. There were seven patients who were lost to follow-up subsequently. Finally, 89 patients were included in the study.

The causative factors for Asherman’s syndrome analyzed were as follows: curettage for miscarriage, medical termination of pregnancy, curettage on postpartum uterine cavity, previous hysteroscopic surgery and previous myomectomy. The menstrual pattern in these patients in terms of normal period, hypomenorrhea and amenorrhea was recorded. Asherman’s syndrome, diagnosed hysteroscopically, was classified according to the Modified European Society of Hysteroscopy (ESH) and European Society of Gynaecological Endoscopy (ESGE) classification of intrauterine adhesions [10]. The modified classification of 89 patients with Asherman’s syndrome along with the number of patients with infertility in various grades of adhesions are depicted in Table 1.

Table 1 Modified classification based on European Society of Gynaecological Endoscopy (ESGE) classification of 89 patients with Asherman’s syndrome
Full size table

Hysteroscopic adhesiolysis under general anesthesia was performed for all patients by a single operator (first author), experienced in operative hysteroscopic procedure. The goal was to restore the normal anatomy of the uterus. The cervix was initially dilated with Hegar’s dilators up to a size of “10”. A 9 mm working element along with sheath and 4 mm 30° telescope (Karl Storz, Germany) equipped with a hysteroscopic monopolar (Collin’s) knife was introduced into the uterine cavity (Fig. 1). Glycine (1.5%) was used as the distending medium through an automated hysteroscopic distension pump. Adhesiolysis was performed by making incisions into the avascular adhesions starting from the point where adhesions were thickest. The procedure was continued in a cephalad direction and until the pink myometrium was visible. It was desired in all cases to visualize the ostial areas. But in cases of severe adhesions, the procedure was discontinued once the pink myometrium was visible, even if the ostial areas were not seen. Concomitant laparoscopy was done in all cases to guide the extent of adhesiolysis, to assess the tubal status and to overview the whole pelvis for any associated abnormality. A strict fluid input and output chart was maintained throughout the procedure. The procedure was stopped if the fluid deficit was more than 1.5 l. Serum electrolyte levels were checked in all patients at the end of the procedure. The occurrence of any intraoperative or postoperative complication was recorded. Postoperatively, antibiotics in the form of ciprofloxacin and metronidazole were given for a period of 5 days in all patients. An inert T-shaped intrauterine device as a stent was inserted into the uterine cavity at the end of the procedure to prevent adhesion reformation and kept in situ for 30 days. Hormonal treatment was started in all patients with estradiol valerate 2 mg once daily and was given for a period of 30 days. Thereafter in patients who had withdrawal bleeding, the same cyclical hormonal treatment was continued for 1 month. The intrauterine stent was removed in all cases at the end of 30 days. After 2 months, a second-look diagnostic office hysteroscopy was performed in the proliferative phase to assess the normalcy of uterine cavity or any adhesion reformation. In patients who remained amenorrheic in spite of hormonal treatment after 1 month, diagnostic office hysteroscopy was also performed 2 months after initial hysteroscopic adhesiolysis. In the presence of adhesion reformation, a repeat adhesiolysis procedure was performed. The same inert stent was reinserted into the uterine cavity and hormonal treatment was started again for 1 month. In the absence of adhesions on second-look hysteroscopy, patients were advised to try for pregnancy spontaneously. Patients who required repeat adhesiolysis procedure were advised to resume their effort to conceive following stent removal 1 month after the repeat procedure. No patient was subjected to more than two procedures. The duration of follow-up for these patients ranged from 3 to 48 months with a mean of 24.5 months. Patients were contacted routinely and followed up every 3 months. The menstrual pattern, conception rate, time interval to conceive and reproductive outcome were recorded. Statistical analysis was performed using χ2 test and P < 0.05 was considered to be statistically significant. The correlation between conception and menstrual pattern before and after hysteroscopic adhesiolysis was evaluated by “two-by-three” contingency table analysis. The conception rate in mild, moderate and severe IUA after hysteroscopic adhesiolysis was analyzed. Cumulative pregnancy rate (CPR) in women who underwent hysteroscopic adhesiolysis and the pregnancy outcome of these patients were analyzed.

Fig. 1
figure 1

Monopolar Colin’s knife in place during adhesiolysis

Full size image

Results

Out of the 96 patients who underwent hysteroscopic adhesiolysis, 7 were lost to follow-up subsequently and, therefore, 89 patients with symptoms of infertility were included in the final analysis. The age of the patients varied from 21 to 37 years (mean 28.4 years). Among 89 patients, 57 (64%) women had secondary infertility and 32 (36%) had primary infertility. The menstrual patterns of these patients were as follows: 27 (30.3%) patients had amenorrhea, 48 (53.9%) had hypomenorrhea and 14 (15.7%) had normal menstruation.

Among the various etiological factors analyzed, 29 (32.5%) patients had history of medical termination of pregnancy, 23 (25.8%) had curettage for miscarriage, 5 (5.6%) had curettage on postpartum uterus, 6 (6.7%) had endometrial curettage for menorrhagia, 4 (4.4%) patients had myomectomy where endometrial cavity was opened and 4 (4.4%) patients had hysteroscopic surgery including 2 cases of hysteroscopic septal resection and 2 cases of hysteroscopic myomectomy. In 18 (20.2%) patients, we did not find any obvious cause for Asherman’s syndrome.

A total of 77 patients showed no reformation of adhesions on second-look hysteroscopy. In the rest of the 12 patients, there was reformation of adhesions that needed a second procedure. No patient in this study was evaluated for a third procedure. Operative complications included two (2.2%) cases of uterine perforation toward the end of the procedure recognized during concomitant laparoscopy and did not require any specific treatment. There was one case of fluid overload, which was treated conservatively. The mean operating time for mild, moderate and severe intrauterine adhesions including the time taken for cervical dilatation was 32 ± 5, 35 ± 6 and 38 ± 3 min, respectively. The difference observed was of no statistical significance (P = 0.06).

The proportion of women with amenorrhea, hypomenorrhea and normal period before and after surgery were 27/48/14 and 10/12/67, respectively. The amenorrhea and hypomenorrhea prior to surgery were improved in 53 out of 75 patients (70.6%). “Two by three” contingency table analysis depicting correlation between conception and menstrual pattern before and after adhesiolysis is given in Table 2. There was no significant association between conception rate after adhesiolysis and preoperative menstrual pattern (P = 0.06). The likelihood of conception was 44.3% (35 out of 79) in those who continued to have improved menstrual pattern compared to only 10% likelihood conception rate (1 out of 10) in patients who continued to have amenorrhea after treatment. This difference was statistically significant (P = 0.03). The conception and reproductive outcome of 89 patients with infertility after hysteroscopic adhesiolysis is given in Table 3.

Table 2 Menstrual pattern before and after hysteroscopic adhesiolysis and relationship to conception rate
Full size table
Table 3 Conception and reproductive outcome in 89 infertile after hysteroscopic adhesiolysis in different grades of adhesion
Full size table

Among the 89 patients with infertility, 36 (40.4%) achieved pregnancy. Only two patients (2/36) conceived at more than 35 years of age at the time of treatment. In the rest of the 34 patients (34/36), pregnancies occurred in patients below 35 years of age at the time of treatment, reflecting a better conception rate when the age of the patient was <35 years. The mean time interval from the time of surgery to the time of conception was 11.5 months in mild, 12.8 months in moderate and 14.2 months in severe IUA, with overall mean conception time after surgery of 12.8 months. The difference in the time interval observed was not statistically significant (P > 0.05). The conception rate in mild intrauterine adhesion is higher (58%) compared to moderate IUA (30%) and severe IUA (33.3%) after surgery. None of 12 patients who required repeat adhesiolysis conceived. Out of 36 pregnancies, 31 (86.1%) achieved a live birth, 4 (11.1%) had spontaneous miscarriage and 1 (2.7%) pregnancy resulted in intrauterine death at 35 weeks of gestation. Four women (11.1%) had preterm delivery and five (13.8%) patients (three patients in moderate IUA and two patients in mild IUA group) had intrauterine growth restriction (IUGR). Among the 32 deliveries, there were 18 (56.2%) vaginal delivery and 14 (43.8%) patients had cesarean section (CS) for obstetric indication. Among the 28 term deliveries, there were 16 (16/28; 57.1%) vaginal deliveries, 4 (4/28; 14.2%) elective CS for obstetrical indication and 8 (8/28; 28.5%) emergency CS for fetal distress. In the rest of the four preterm deliveries, there were two vaginal deliveries and two emergency CS for prematurity.

Among 31 women who had live birth, 4 had adherent placenta (3 of them detected after normal vaginal delivery and 1 detected at CS). Three patients had adherent placenta where adhesiolysis was done in moderate IUA and one patient had severe IUA. Three of these four patients required manual removal of placenta and one patient required hysterectomy for uncontrolled postpartum hemorrhage. The cumulative pregnancy rates for the 89 women with infertility, who had undergone hysteroscopic adhesiolysis, is shown in Fig. 2. In total, 61.1% (22 out of 36) conception occurred within 1 year, and 97.2% (35 out of 36) of all conceptions occurred within 24 months after surgery. In only one patient in moderate IUA, pregnancy occurred after 24 months.

Fig. 2
figure 2

Cumulative frequency curve of pregnancy rate over time after surgery (months)

Full size image

Discussion

Asherman’s syndrome occurs due to intrauterine injury leading to disruption of the basal layer of the endometrium. The exact prevalence of the condition is difficult to determine, but the incidence has been increasing over the last few decades, probably due to increase in iatrogenic endometrial trauma as well as due to better diagnostic techniques like transvaginal ultrasound and hysteroscopy [11, 12]. Asherman’s syndrome typically presents with amenorrhea, hypomenorrhea, infertility and repeated abortions [13, 14]. The majority of patients with Asherman’s syndrome present with infertility [3]. In the present study, we analyzed 89 women who presented only with infertility due to Asherman’s syndrome, to evaluate the reproductive outcome in these patients. Therefore, any additional factors of infertility, including the presence of abnormal husband semen, which would undoubtedly affect the outcome, were excluded from the study. An accurate classification of Asherman’s syndrome is necessary to evaluate the prognosis in terms of menstrual function and fertility outcome. Numerous classifications have been proposed [3, 15–17], but none has received universal endorsement. All our cases were diagnosed and classified after diagnostic hysteroscopy, which is accepted as the gold standard worldwide. We had chosen to classify according to ESGE classification (1995 modified version) of intrauterine adhesions [10], as it clearly reflects the severity of adhesions. The mean age of patients in our study was 28.4 years, which is comparable to other studies [18, 19]. Regarding aetiology, postpartum curettage or abortion has been described to be the major cause of Asherman’s syndrome [20]. Other causes include genital tuberculosis [13], pelvic irradiation and uterine surgery including hysteroscopic surgery [20]. In the present study, in 64% of cases, the cause of Asherman’s syndrome was previous first trimester medical termination of pregnancy and curettage on a gravid uterus. Different techniques of hysteroscopic adhesiolysis have been described. Hysteroscopic scissors [17] or laser treatment [21] has been used to divide adhesions. Hysteroscopic resection with a monopolar probe was also found to be efficient [22, 23]. Fluroscopically guided adhesiolysis has also been reported and was found to reduce the complication rate [24, 25]. Some authors have used versa point hysteroscopic electrosurgical electrode using bipolar coagulation [26, 27]. The advantage of using bipolar energy is that this system requires no dilatation, does not require dispersive return electrodes and does not generate stray currents. Therefore, it minimizes complications and there is less chance of adhesion reformation. In our study, we used a monopolar knife (collin’s knife) to divide adhesions, which was also very effective, safe and less expensive in a limited resource setup. Some authors felt that hysteroscopic adhesiolysis under laparoscopic guidance is not necessary to reduce the risk of uterine perforation [5, 20], while some are in favor of the same [26]. The concurrent laparoscopy in our study to confirm the tubal patency and to rule out other pelvic pathology also helped to know the end point of adhesiolysis by observing the transillumination. There were only two (2.2%) cases of uterine perforation in our series and this low rate may be due to concurrent laparoscopy during the procedure. Equally important in the treatment of Asherman’s syndrome is the prevention of adhesion reformation. This is commonly done by physical means and stimulation of endometrial growth with hormone therapy. Intrauterine application of auto-cross-linked hyaluronic acid gel appeared effective in the prevention of intrauterine adhesions after hysteroscopic surgery [28] and after hysteroscopic adhesiolysis [29]. It also reduced postoperative adhesions. However, none of these studies evaluated fertility after the use of these gels and no data are available about their effects on the endometrium. The insertion of intrauterine device (IUD) or Foley’s catheter balloon has been advocated by various authors as an effective and widely used method to prevent adhesion reformation [14, 30–32]. We, therefore, in the present study used an inert IUD, kept in situ for 30 days, and found no complication with it. Simultaneously, we also started hormone treatment with estradiol valerate 2 mg daily for 30 days. A combination of estrogen and progestin therapy has also been reported [14, 33]. We preferred giving estrogen only in order to avoid the unacceptable side effects of progesterone. The mean duration of the hysteroscopic adhesiolysis procedure is reported to vary between 10 and 45 min [18, 24]. In our series, the mean operating time was 35 ± 3 min and the difference in the mean operating time for mild, moderate and severe Asherman’s syndrome was not statistically significant.

It has been reported that improvement of menstrual flow after hysteroscopic adhesiolysis ranges from 52.4 to 74.2% [8, 14, 18]. In our series, the improvement of menstrual function was 70.6% (53 out of 75) and this was similar to an earlier recent report [14]. We also found that there was no significant association between conception rate after adhesiolysis and preoperative menstrual pattern. However, the menstrual pattern after hysteroscopic surgery showed a significant impact (P = 0.03) on prognosis, with conception rate of only 10% (1 out of 10) in women with amenorrhea, 33.3% (4 out of 12) in hypomenorrhea and 46.2% (31 out of 67) in women with normal menstruation. These findings were similar to earlier reports [14, 20, 21] and clearly reflect that endometrial function (as reflected by menstrual pattern) after hysteroscopic adhesiolysis is an important factor in determining the reproductive outcome. As might be predicted, age is a contributing factor to fertility outcome after treatment of IUA. Earlier reports suggest that an age less than 35 years was associated with significantly better post-adhesiolysis pregnancy rate compared to pregnancy rate in women above 35 years of age [7, 16]. In our series, we also found that only two patients (2/36) conceived at more than 35 years of age. In the rest of the 34 patients (34/36), pregnancies occurred in patients below 35 years of age. This reflects that there is a better conception rate when the age of the patient is <35.

There was no significant difference in mean time interval from the time of surgery to the time of conception in mild (11.5 months), moderate (12.8 months) and severe (14.2 months) Asherman’s syndrome (p = 0.06). In a recently reported series the cumulative conception rate showed that 59% conception occurred within 1 year and 87.2% conception rate occurred within 2 years after hysteroscopic adhesiolysis [14]. In the present series, the analysis of cumulative pregnancy rate between adhesiolysis and conception intervals also showed 61.1% (22 out of 36) conception occurring within 1 year and 97.2% occurring within 2 years. One notable finding in our study showed that there was no conception in 12 patients who required repeat adhesiolysis. This could be explained by the fact that intrauterine adhesion reformation and concomitant endometrial atrophy are potential problems that may limit the success of hysteroscopic adhesiolysis [14, 18]. Therefore, these patients should be counseled about poor prognosis. It has been reported that success in terms of pregnancy and live birth cases are related to the severity of adhesions [14, 17, 34]. But on the contrary, other studies reported no significant correlation between severity of adhesions and live birth rates [9, 25]. In our series, the conception rate in mild intrauterine adhesions was higher (58%) compared to moderate IUA (30%) and severe IUA (33.3%). In the present series, the overall conception rate after hysteroscopic adhesiolysis was 40.4%, which was similar to earlier reports [14, 35]. The reported live birth rate after hysteroscopic adhesiolysis ranges from 64 to 79% [14]. The higher live birth rate (86%) in the present series was due to low miscarriage rate (11.1%) and reflects good antenatal care. Interestingly, the overall preterm delivery rate in our series is much lower (11.1%) compared to the reported preterm delivery rate of 17.9 to 50% [9, 14, 35]. This is due to the fact that there was no preterm delivery in the mild Asherman’s syndrome group in our series, compared to 16.6% preterm delivery rates in the moderate and severe group. This indicates that there is a higher risk of preterm delivery in moderate and severe Asherman’s syndrome after surgery. The CS rate is reported to be high, ranging from 25 to 80% [5, 24] and was mainly due to obstetric indications. In the present series, we also found that the CS rate was high (43.8%) and was performed mainly for obstetric indications. Placenta accreta is the most severe complication reported after treatment of IUA [20]. In our series, 12.5% (4 out of 32 deliveries) cases had adherent placenta, where 3 needed manual removal of placenta and 1 patient required hysterectomy for postpartum hemorrhage. Therefore this complication of adherent placenta should always be kept in mind in patients who conceived after hysteroscopic adhesiolysis.

Conclusion

In this study, we have analyzed the reproductive outcome of hysteroscopic adhesiolysis in women with Asherman’s syndrome presenting with infertility. There was no significant association between conception rate after adhesiolysis and preoperative menstrual pattern. The menstrual pattern after hysteroscopic surgery showed a significant impact on prognosis. In other words, women who experienced normal periods after hysteroscopic adhesiolysis and women who had normal uterine cavity on second-look hysteroscopy were more likely to conceive. The overall conception rate was 40.4%. The conception rate was higher in mild Asherman’s syndrome compared to moderate and severe groups. As much as 61% conceptions occurred within 1 year and 97.2% conception occurred within 2 years after surgery. There was no conception in patients who needed a repeat adhesiolysis due to adhesion reformation and therefore these patients should be counselled for poor prognosis. Though CS rate was very high, they were all performed for obstetric indications. It should be kept in mind that placenta accreta is a severe complication after hysteroscopic adhesiolysis, which may need hysterectomy.