Keywords

1 Introduction

Intrauterine adhesion (IUA) refers to partial or complete adhesions between the uterine walls which might result in several clinical manifestations such as hypomenorrhea, amenorrhea, dysmenorrhea, abdominal pain, infertility, habitual abortion, premature delivery, and abnormal placenta implantation.

The pathology also known as Asherman syndrome (AS) was first described in 1894 by Heinrich Fritsch. In 1927, Bass reported on 20 cases of cervical obstruction after induced abortions in Russia [1] and Stamer [2] in 1946 reported on a couple of cases of IUA after abortion or postpartum. However, the full characterization of the disease was carried out by Joseph Asherman [3]. He considered two different entities, a traumatic one and the stenosis of the cervical os. He also linked endometrial trauma and adhesion to menstrual disturbance, cyclical pelvic pain and even subfertility. Moreover, he recognized that the cause had to do with endometrial damage and not to pure obstruction of the menstrual flow. He described it as one type of complication of uterine curettage, where adhesions between the uterine walls were the result of an inflammatory process, mostly aseptic. Since then [4, 5] other underlying causes have been described.

2 Definition

IUA or synechia are composed of scar-fibrotic tissue that may result in the adherence of opposing surfaces of the endometrial lining. The adhesions can vary widely, if in one extreme they can be just simple thin bands of filmy tissue easily broken, in a small part of the uterus (Fig. 61.1); on the other hand, in cases of severe disease, all the endometrium can be obliterated by thick fibrous tissue (Figs. 61.2 and 61.3).

Fig. 61.1
figure 1

Filmy adhesion on hysteroscopy

Full size image
Fig. 61.2
figure 2

Thick adhesion on hysteroscopy

Full size image
Fig. 61.3
figure 3

Thick adhesion on hysteroscopy

Full size image

There is some discussion about the differences between the concept of IUA and AS. Often they have been used synonymously. However, syndrome is defined as a group of symptoms and medical signs that characterize a disease. Taking this into consideration, the term Asherman Syndrome should be avoided when no symptoms are present. However, other authors [6] have suggested that this term—AS, should be used exclusively in cases of IUA related to endometrium trauma of the gravid uterus.

3 Prevalence

It is very difficult to estimate the true prevalence of IUA, especially because in many cases they might be asymptomatic and undiagnosed. A 1982 review showed a wide variability on the geographical prevalence of adhesions, being especially high in Israel [4]. The authors tried to explain this, with the different incidence of legal and illegal abortion and the incidence of postpartum endometritis. The technique for puerperal and postabortion evacuation was also an issue. Other reasons had to do with the degree of awareness of the physicians, the criteria used for diagnosis, and the incidence of genital tuberculosis.

Taking this into consideration, the true incidence has been estimated to be around 0.3% in the general population, 1.5% of the patients undergoing hysterosalpingography [7], between 5 and 39% of women with recurrent pregnancy lost [8,9,10], up to 21% after postpartum curettage [11], and up to 40% of patients who have a history of surgical treatment for retained products of conception [12], while in infertile women, the incidence of IUA might be over 2%.

4 Etiology

4.1 Pregnancy Related

Most cases of synechiae are the result of intrauterine trauma of the gravid uterine cavity [13, 14] (Table 61.1). This trauma could be caused by curettage either after spontaneous or induced miscarriage or in the postpartum period. According to the publication of Schenker and Margalioth [4] who reviewed 1856 cases of Asherman syndrome, pregnancy was the predisposing factor in 90% of cases. In their description (Table 61.1), 67% of cases occurred after postabortion curettage, 22% after postpartum curettage, 2% after cesarean section, and around 1% after evacuation of a hydatidiform mole (Fig. 61.4).

Table 61.1 Asherman syndrome according to the review of Schenker and Margalioth (1982 based on 1856 cases) [4]
Full size table
Fig. 61.4
figure 4

Ultrasound of adhesions after first trimester curettage

Full size image

Westendorp et al. [12] reported on the prevalence of intrauterine adhesions after secondary removal of placental remnants after delivery and after a repeated curettage for incomplete abortion. A hysteroscopy was performed 3 months after the intervention in 50 women and they found adhesions in 40% of these, of which 75% had grade II–IV adhesions. The authors considered several risk factors and they concluded for a non significative risk increase in the presence of infection or lactation or history of previous abortion. Women with menstrual disorders after the procedure were more likely to have intrauterine adhesions as 13/25 with any menstrual disorder were found to have IUA.

In 2014, Hooker [15] published a methanalysis in over 900 women evaluated with hysteroscopy within 12 months following a spontaneous abortion and reported a prevalence of IUA in 19.1% (Figs. 61.5 and 61.6). He also concluded that the risk of adhesions seems to increase when two or more curettage procedures are performed, with a relative risk of 2.1. In a review by Fejgin [16], the author reports that repeated curettage for pregnancy loss increases the risk of adhesions from 8 to 35% between the first and the third procedure. Gilman [17] published a review on 884 patients submitted to active management of early pregnancy loss. According to their symptoms, six were found to have AS, all of them had undergone sharp curettage, and three had repeated curettage performed; there were no cases of AS after manual vacuum aspiration (0/191) or medical management with misoprostol (0/210). In a 2016 publication, the same author [18] concluded in favor of an association between IUA and mechanical suction dilatation and curettage (D&C), increased uterine size, particularly in the presence of multiple gestation, but not when manual vacuum aspiration was performed.

Fig. 61.5
figure 5

Ultrasound of adhesions after molar pregnancy evacuation: adhesions with an isthmic endometrial cyst

Full size image
Fig. 61.6
figure 6

3D-ultrasound of adhesions after curettage

Full size image

Other authors describe rates of adhesions between 17 and 30% after incomplete or missed abortions [19,20,21], while Lurie et al. found a prevalence of 39% after midtrimester abortion in women who underwent curettage [22] in contrast with the findings of Golan et al., who found a prevalence of intrauterine adhesions of 2% after manual removal of placenta in 48 patients [23]. It seems that the highest risk for a postpartum curettage to end up in adhesions in between 2 and 4 weeks postpartum, compared to when performed within 48 h postpartum, perhaps due to the different levels of circulating estradiol [24].

Women submitted to uterine compression sutures for postpartum hemorrhage also have an increased risk of IUA as it has been described in around 20–25% of them [25,26,27]. The reason is uncertain; however, the B-Lynch type creates a brace-like effect that apposes the anterior with the posterior wall, which in theory [28] can occlude uterine blood flow, leading to adhesions.

There is also a description of several case reports describing severe adhesions after uterine devascularization for the treatment of severe postpartum hemorrhage [29]. A 2014 review [30] reported on two cases of synechiae or necrotic uterus after bilateral uterine artery ligation followed by bilateral suspensory ligament of ovary ligation in two out of 12 patients. No complications were reported in ten patients submitted to bilateral uterine artery ligation.

4.2 Nonpregnant

Pregnancy is not the only factor related to uterine adhesions. According to the publication of Schenker and Margalioth [4] previously stated, out of the 1856 cases reported 30 (1.6%) had a previous history of a diagnostic curettage, while 24 an abdominal myomectomy, ten a cervical biopsy or polypectomy, and three the insertion of an intrauterine contraceptive device.

For long, the formation of adhesions has been described as a complication of hysteroscopic surgery, especially in cases of myomectomy. Taskin [31] described the rate of adhesion formation after a second-look hysteroscopy according to different procedures and concluded that adhesions were uncommon after treatment for polyps and uterine septa, while mild IUA formation were present in 31% with solitary fibroids and 45% with multiple ones. Most of them could be easily treated during the procedure. Other studies report different rates [32,33,34], Mazzon reported on 688 women with one or more type I–II fibroids submitted to Cold loop resectoscopic myomectomy to conclude that synechiae were found in 29 patients (4.2%) [35].

Bilateral uterine artery embolization (UAE) has also been associated with adhesion formation, with reports varying between 10 and 14% [36]. It also seems that AS after this treatment has a poorer prognosis with more severe adhesions than in other situations. Song hypothesized that adhesions after UAE performed for obstetric complications might carry a worse prognosis than when it is used for gynecological situations.

As expected, IUA are common complications of endometrial ablation. There are several techniques that can be used, but overall when properly performed, they all lead to the removal or destruction of the basal layer of the endometrium, followed by cicatricial fibrosis. Studies have shown that around 35% of women submitted to this procedures will have adhesions on hysteroscopy [37, 38].

Despite the fact that endometrial tuberculosis may contribute to the development of IUA [39, 40], this is more common in undeveloped areas, like the North of India. In the review of Schenker and Margalioth [4], only 4% of the cases were found to be related to tuberculosis. Schistosoma sp. have also been implicated [41], but as tuberculosis it is expected to be rare in developed countries.

Other situations have been related to IUA, like congenital anomaly of the Uterus. In a study by Stillman and Asarkof [42], seven out 43 (16%) infertile patients with Mullerian duct malformations had IUAs. They could not establish if the anomaly was the cause of the adhesions or if the anomaly could predispose to the adhesions, for example, by increasing the risk of recurrent pregnancy loss.

Pelvic irradiation can also be a rare cause of IUAs [4]. The so-called genetic predisposition might have its role. Possibly there might be an individual response to injury that leads to the formation of IUAs in some while it does not in others.

4.3 Pathogeny

Several explanations have proposed. One of this stated that the functional layer of the endometrium of the gravid uterus might somehow be more vulnerable to aggression and prone to be destroyed [5]. Another explanation is that endometrium depends on estrogens for regeneration and so a low estrogen status immediately after the procedure as it happens during lactation might predispose to adhesion formation.

The role of infection is not established. On the one hand, it is clear that tuberculosis or schistosomiasis might be responsible for IUAs; however, as told before these situations are quite uncommon in developed countries. On the other hand, the role of postpartum and postabortion endometritis is more controversial. Polishuk et al. using hysterography [43] reported on 171 women who had undergone cesarean sections. Of these, 28 had endometritis and only one developed adhesions. Charles [44] reported that less than 1% of his patients had a history of infection by the time of their initial surgery. However, as described by Yu et al. [5] some investigators believe in the synergy between the inflammatory processes and trauma to the damaging of the endometrium.

The natural history is unknown. There are women with a consistent clinical diagnosis of AS who resumed menses. Schenker and Margalioth [4] reported on the fertility outcome of 292 patients who received no treatment. Only a few had the diagnosis confirmed by hysteroscopy and so some of them might have resumed menses as a result of a cervical stenosis that was overcome by the pressure of a hematometra. Although 46% conceived, only half of them had a viable infant and 13% had placenta accreta.

4.4 Histopathology

Histologically , in IUA the normal endometrium is replaced by fibrous tissue. The intrauterine adhesions may involve endometrium, myometrium, or connective tissue.

Normal endometrium consists of simple columnar glands with or without cilia and vascular stroma composed of connective tissue and richly vascularized, with spiral arteries. In women of reproductive age, two layers can be distinguished. The functional layer (stratum spongiosum) is closer to the lumen and changes during the cycle according to the hormonal milieu established by estrogens and progesterone. The glands are regularly spaced. The basal layer (stratum basale) is adjacent to the myometrium and contrary to the functional layer does not shed during menstruation. This layer is less responsive to steroid hormones as typically the glands are irregularly shaped and the stroma is dense. The spiral arteries are different too, as they have thicker muscular walls [45].

With adhesions there is a loss of the distinction between the basal and the functional layer of the endometrium [46]. The functional layer is replaced by an inactive epithelial monolayer. This epithelia is much less or nonresponsive at all to hormone stimulation. Sparse and inactive glands can be identified. In most cases, the tissue is avascular although sometimes thin-walled telangiectatic vessels can be observed [47]. Yaffe [48] comparing biopsy specimens of the uterine wall from patients with intrauterine adhesions with controls concluded that the first ones contained 50–80% of fibrous tissue compared with 13–20%. Moreover, according to McCulloch [49] it seems that histological changes after transcervical resection of the endometrium are similar to those with Asherman syndrome. The changes are extended to the surrounding endometrium even in a nonaffected part of the uterus as this tissue also appears atrophic with an increased amount of connective tissue. This might help to explain why sometimes the hipomenorrhea symptoms seem disproportionally higher than the adhesions observed (Fig. 61.7).

Fig. 61.7
figure 7

Left uterine cornus after adhesiolysis showing atrophic endometrium and bleeding from the muscularis

Full size image

On electric microscopy [50], the glandular epithelial cells present with swelling, with a loss in the ribosomes and expansion of the endoplasmic reticulum. There are also changes in the mitochondria that have their cristae shortened and reduced and vacuolization. The interstitium is looser, with closed capillaries and tight junctions between endothelial cells. On the molecular level, there is a suggestion that adhesion-related cytokines, such as transforming growth factor β, platelet-derived growth factor, and b-fibroblast growth factor might be associated with IAU [51].

The use of pelvic angiography [11] has demonstrated a reduction in the myometrial vascular flow in patients with severe hypomenorrhea and amenorrhea.

Sometimes, calcification or even ossification of the stroma might be present.

VEGF [50] expression might have an important role. Chen, comparing patients pre- and post treatment, concluded that patients with IUA had microscopically vascular closure and hypoxic changes, with clear improvement after treatment. After treatment, they also concluded in favor of more expression of VEGF and increased microvascular density score.

Malhotra [52] compared endometrial thickness and blood flow impedance of the uterine spiral artery post-menstrual on day 2/3 pre- and post-hysteroscopic adhesiolysis and concluded in favor of a significant improvement in the endometrial thickness and a higher blood flow impedance of spiral artery after the procedure.

It seems that adhesions might progress, which means that they begin as thin endometrial bands to progress to thicker ones. When there are muscular adhesions and the endometrial basalis is lost, the prognosis is poor as there is no starting point to the proliferation of new endometrium after adhesiolysis [53].