Heinrich Fritsch reported the first case of intrauterine adhesions (IUAs) at the end of the nineteenth century. Since 1948, a series of papers on this condition have been published by Joseph G. Asherman, which describe the frequency, aetiology and symptoms of IUA
Heinrich Fritsch reported the first case of intrauterine adhesions (IUAs) at the end of the nineteenth century. Since 1948, a series of papers on this condition have been published by Joseph G. Asherman, which describe the frequency, aetiology and symptoms of IUA [1, 2].
The terms IUA and Asherman syndrome are often used interchangeably, although IUA refers merely to the existence of adhesions in the uterine cavity without reference to any symptom. Indeed, from time to time, cases of IUA do not produce any symptoms. Asherman syndrome, by definition, is characterised by both the presence of IUAs and one or more symptoms arising from it, such as amenorrhoea, hypomenorrhoea, subfertility, pregnancy loss or dysmenorrhoea [1–3].
Trauma to the basalis layer of the endometrium is the major cause of IUAs. Such an event can occur in a gravid or non-gravid uterus, but it is more frequent in a gravid uterus. Trauma to the uterine mucosa can arise from miscarriages, with or without curettage, post-partum curettage, post-abortion or post-partum endometritis, ischaemic phenomena after post-partum haemorrhage, caesarean sections and uterine artery embolisation. For a non-gravid uterus, much less common causes of severe IUAs include the use of surgical hysteroscopy (resection or destruction of endometrium, fibroids, polyps and uterine septae), global endometrial ablation and myomectomy. IUAs may also occur secondary to infection (e.g. genital tuberculosis, endometritis or previous septic abortions). Predisposition to form IUAs may be linked to individual factors, such as age, race, nutritional status and the concomitance of other diseases or infectious processes [3–5].
The prevalence of IUAs is difficult to determine. It varies according to the populations examined, as well as the type of investigation used for the diagnosis of IUA. The incidence of IUA appears to be increasing in recent years, probably due to a combination of several factors. These include increased awareness among patients and clinicians, availability of more sensitive methods of diagnosis (such as hysteroscopy instead of hysterosalpingography (HSG)), and an increased number of induced abortions and hysteroscopic surgeries, such as endometrial ablation and removal of submucosal fibroids [3–5].
15.2.1 Treatment Considerations
Intrauterine adhesions that do not produce symptoms do not require treatment. However, when the adhesions have resulted in subfertility or other troublesome symptoms such as pain and abnormal bleeding, treatment should be considered to remove them (see Figure 15.1). Surgery is the most challenging in women with severe intrauterine adhesions resulting in obliteration of the entire uterine cavity. The skills required are great, complication rates are high and recurrence is common.
Figure 15.1 A thick band of intrauterine adhesions involving the anterior and posterior uterine wall.
Prior to the advent of hysteroscopic surgery, techniques for treating IUAs included cervical probing, cervical dilatation with or without endometrial curettage (‘D&C’), hysterotomy and blunt finger dissection . In the presence of complete cavity obliteration, the use of a cervical dilator sequentially directed from the cervical canal towards each tubal ostium has been described in a small series of patients . This technique created two lateral defects and a central fibrous septum which was then hysteroscopically incised under laparoscopic guidance. Complications were frequent in this series.
Several important surgical principles govern the hysteroscopic management of intrauterine adhesions. These include: (i) paying attention to the surgical technique, using careful steps to remove the adhesions safely and without causing complications, (ii) protecting the endometrium to prevent further damage and aggravation of the condition and (iii) preventing recurrence of adhesions using adjuvant therapies. As will become apparent, reliable evidence is lacking to guide practice, although the American Association of Gynecological Laparoscopists (AAGL) and the European Society for Gynaecological Endoscopy (ESGE) have recently published some joint practice guidelines [6, 8].
Technological advances – particularly miniaturisation of endoscopes and improvements in imaging and ancillary instrumentation – have facilitated safer and more precise hysteroscopic approaches to adhesiolysis. The aims of surgery are: (i) to lyse all adhesions, (ii) to restore the shape and dimensions of the uterine cavity and (iii) to identify both tubal ostia. In this way, communication is restored between the cervical canal and fallopian tubes via a normal volume uterine cavity, allowing normal menstrual flow and adequate sperm transportation for fertilisation and implantation to occur. Following surgery, further aims are: (iv) to stimulate endometrial repair and regeneration, (v) to prevent re-formation of IUAs and (vi) to restore normal reproductive function, ultimately confirming tubal patency. These objectives may not be possible to achieve in all cases and multiple surgeries may be required. In vitro fertilisation (IVF) becomes a viable treatment option for women seeking fertility if the cavity can be restored and the endometrium re-epithelialised. It is important to counsel all women about the risks of surgery, the need for multiple procedures, planned post-operative interventions and the likely clinical, and especially reproductive, outcomes.
This will vary according to the nature and extent of the adhesions. Several classification systems, summarised in Table 15.1, are in use, with no consensus as to which is preferred. The simplest cases involve the identification of filmy adhesions at the time of outpatient hysteroscopy that can be broken down with the tip of the hysteroscope; even stretching the cavity with the uterine distension media may be enough. In contrast, the treatment of severe, dense adhesions, where no part of the uterine cavity is recognisable, remains far more challenging. In such cases, the cavity may be completely occluded or too narrow to allow the insertion of a hysteroscopic sheath inside the cervical canal.
|Classification system||Criteria||Assessment||Clinical symptoms|
|March et al. ||Minimal, moderate or severe, based on the degree of uterine cavity involvement||Hysteroscopy||N/A|
|Hamou et al. ||Isthmic, marginal, central or severe||Hysteroscopy||N/A|
|Valle and Sciarra ||Mild, moderate or severe and based upon the extent of occlusion (partial or total)||Hysteroscopy and hysterosalpingogram||N/A|
|European Society for Hysteroscopy ||Hysteroscopy, hysterosalpingogram and clinical||Menstrual pattern|
|American Fertility Society ||Mild, moderate, or severe based on extent of endometrial cavity obliteration and appearance of adhesions characteristics||Hysteroscopy, hysterosalpingogram and clinical||Menstrual pattern|
|Donnez and Nisolle ||Hysteroscopy or hysterosalpingogram||N/A|
|Nasr et al. ||Prognostic score incorporating clinical and hysteroscopic assessment||Hysteroscopy and clinical||Menstrual pattern and obstetric history|
A variety of surgical techniques have been described, but the general approach is as follows. The removal of the adhesions should start from the lower part of the uterus and progress towards the upper part. Central and filmy adhesions should be separated initially to allow adequate distension of the uterine cavity and enhance visualisation. Dense and lateral adhesions should be treated last, bearing in mind the greater risk of uterine perforation and bleeding [3, 16].
There is no evidence to support the routine use of antibiotic therapy during hysteroscopic surgery . However, infection is thought to be involved in causing IUAs, and the role of antibiotics specifically in hysteroscopic adhesiolysis is unknown. For these reasons, some surgeons cover procedures with antibiotics, but no compelling evidence is available to support this practice.
Some investigators advocate the use of grasping forceps and scissors (cold steel), whereas others prefer to use laser or electrical energy (hot wire) to divide the adhesions (Figure 15.2). While the use of powered electrical instruments has proven efficient for hysteroscopic adhesiolysis , ‘cold’ mechanical approaches may prevent thermal damage of the residual endometrium [3, 19], and possibly reduce the rate and consequences of perforation during the procedure. The pros and cons of each of the two modalities are summarised in Table 15.2. Surgeons performing advanced hysteroscopic surgery should be able to use the two modalities flexibly, depending on the specific location of the adhesions. In any case, the most important consideration is to carry out the surgery safely and effectively, without further damage to the endometrium. Cararach et al. found no difference in outcome between the use of scissors and electrical energy to perform the surgery .
|Scissors (cold steel)||Electrical energy (hot wire)|
|Diameter of hysteroscopic system||≤5.5 mm||≤5.5 mmb or >7 mma|
|Complications||Lower risk of non-target injury to pelvic structures||Higher risk of non-target injury to pelvic structures from direct or indirect thermal damage|
|Efficiency||Can dissect as well as cut, but scissors may sometimes blunt or break (fragile)||Always cuts well and fast|
|Suitability||Particularly suitable for adhesions in cervical canal or isthmic region. Avoids thermal damage to endometrium||Particularly suitable for marginal adhesions|
a Monopolar/bipolar needles/hooks for use with a resectoscope
b Miniature bipolar electrodes (e.g. Versapoint® twizzle tip electrode (Gynecare, division of Ethicon, Inc., Somerville, NJ, USA))
One specific hysteroscopic electrosurgical technique, ‘myometrial scoring’, has been described to create a uterine cavity in the presence of severe cohesive IUAs that preclude conventional hysteroscopic adhesiolysis. The technique involves cervical dilatation under ultrasound guidance, followed by the creation of six to eight 4 mm deep, longitudinal myometrial incisions using an electrosurgical electrode activated in a retrograde fashion from the fundus to the uterine isthmus. Anatomic success and pregnancies have been reported following the surgery .
In the presence of moderate to severe IUAs, hysteroscopic division carries a significant risk of perforation of uterus, especially during dilatation of the cervical canal and introduction of the hysteroscope. Perforation can result in infection, bleeding and intra-abdominal trauma. However, when mechanical instruments have been used, serious consequences are unlikely if the perforation is recognised at the time. However, perforation precludes completion of the procedure, which should be ceased in the absence of adequate uterine distension owing to the risks of further uterine trauma, bleeding and fluid complications. For these reasons, concomitant guidance using fluoroscopy, ultrasound or laparoscopy during hysteroscopic adhesiolysis, especially during the introduction of the dilators and hysteroscope, has been promoted to improve surgical efficiency and safety.
Hysteroscopic adhesiolysis can be facilitated with intraoperative fluoroscopy [22, 23]. The technique involves placing the hysteroscope into the cervical canal and then advancing it until a blockage is reached, or to a point where there may be no obvious clear path along which to safely proceed. Contrast medium is injected via the inflow channel of the hysteroscope in order to delineate the direction of travel by revealing pockets of endometrium and part of the uterine contour. In severe cases, where there is occlusion of the cervical canal and lower uterine cavity, the injected contrast medium simply spills back into the vagina. Size 5 or 7 Fr hysteroscopy forceps are used to gently ‘punch’ a hole within the fibrous obstruction or expand a visible partial opening. Contrast is injected again and if the medium flows freely to fill a portion of the cavity then 5 or 7 Fr scissors are used to divide and dissect the adhesions, leaving healthy endometrium untouched while allowing advancement of the hysteroscope.
These steps are repeated every time a further obstruction is met, or where the operator is uncertain in which direction to advance the hysteroscope. The aim is to open up the obscured parts of the uterine cavity beyond the adhesions and restore the normal uterine size and shape. It should be noted that this is a vaginoscopic approach where blind dilatation can be avoided by using a small (3 mm) telescope, a continuous-flow sheath (4.5–6.5 mm) and size 5–7 Fr forceps and scissors. A saline distension medium should be used to minimise the risk of fluid overload complications, delivered via an automatic pump to regulate the pressure.
A variation of this technique uses a sharp, 16–18 gauge, 80 mm Tuohy needle (Portex Ltd., Hythe, Kent, UK) through which to inject contrast, which is introduced alongside a 5 mm hysteroscope. In cases where there is obliteration of the cavity, an ultrasound scan is used to identify any pockets of endometrium and the needle is inserted under direct vision into the adhesion in that direction. If the surgeon erroneously creates a false track and progresses into the underlying myometrium, contrast medium will rapidly extravasate into the uterine venous plexus, and the surgeon should then proceed in a different direction. Once the hysteroscopic graspers/needle tip is advanced past an adhesion into a pocket of endometrium, injected contrast medium will flow freely to fill that portion of the cavity. Thus, the technique enables early detection of a false passage or perforation, and also assessment of tubal patency, which is the indicator of full anatomic restoration of the uterine cavity.
Ultrasound may be used to guide the hysteroscopic procedure. A transabdominal probe may be used and the bladder filled up with about 300 to 400 ml of normal saline solution to facilitate visualisation of the cervical and endometrial canal. Ultrasound guidance is particularly useful when there are adhesions involving the cervical canal or the lower part of the uterus, causing difficulty in introducing the hysteroscope into the uterine cavity. Later on, during dissection of the adhesions, it can be used to constantly monitor the uterine wall thickness to ensure that the dissection has not gone too far into the muscle wall and to alert the surgeon if the thickness of a particular part of the uterine wall is significantly reduced when compared with adjacent areas. Ultrasound guidance is also useful in the detection and management of a false passage encountered during hysteroscopic surgery (see Section 15.2.4).
Laparoscopy can assess the pelvic anatomy, exclude endometriosis or other pathology and verify tubal patency. When laparoscopy is used to guide hysteroscopic adhesiolysis, the transillumination test can be used to ascertain if a particular part of the uterine wall has become unduly thin. This may be particularly helpful when the area concerned is over the fundus. In cases where the muscular wall of the uterus is severely weakened and perforation is about to occur, the serosa overlying the weakened area may be seen to bulge outwards. The procedure must be immediately stopped at this stage. Laparoscopic guidance may also facilitate the identification of the cornua when there is complete obliteration of the upper part of the uterine cavity including the cornua. A laparoscopic instrument may be used to touch and/or transilluminate the cornua, allowing it to be detected by the surgeon performing the hysteroscopic surgery and providing assistance in the dissection of adhesions around the tubal ostia.
In the management of severe Asherman syndrome, when the cavity is obliterated, a false passage may be created during an attempt to enter into the uterine cavity. The early diagnosis of a false passage is important to prevent further damage to the uterus. The features of a false passage include a narrow, round cavity surrounded by circular muscle fibres, with no recognisable endometrial tissue or tubal ostium. Fluoroscopy and ultrasound both allow early detection of a false passage. The typical ultrasonographic feature of a false passage is a fluid-filled space which is situated eccentrically within the uterus, unlike the endometrial echo, which is almost always situated centrally (unless displaced by intramural pathology such as a fibroid). If a false passage has been confirmed, the decision to continue or discontinue the hysteroscopic procedure should be considered carefully, depending on the experience of the surgeon and how easily the endometrial cavity can be visualised and entered under ultrasound guidance.
A full-thickness perforation of the uterine wall is usually easily recognisable hysteroscopically and should be managed as described in Chapter 9. If an automated pump has been used, the sudden drop in pressure and increase in flow rate are suggestive of a breach of the uterus. If fluoroscopy is used, loops of small bowel may be apparent (outlined by contrast medium) and instruments within the peritoneal cavity should be seen directly if laparoscopic guidance has been employed.
15.3.1 Surgical Considerations
Whatever form of energy is used to divide the adhesions, great care must be exercised to protect the endometrium from further damage. This requires operating under direct vision at all times and with precision, avoiding damage to the underlying myometrium. When using electrical energy, the power of the energy source must be set to the minimum required to effectively divide the adhesions. The use of excessive power carries a risk of further damage to the regenerating basal layer of the endometrium. In doing so, even if perfect anatomical normality can be restored, endometrial function would be permanently compromised, leading to an unsatisfactory reproductive outcome. Another important consideration is the choice of electrode needle or cutting loop. When using the needle, given the same power output, the power density will inevitably be greater. Therefore, it is often necessary to further reduce the power output. The electrode needle should generally be used to divide the adhesions, but the cutting loop may occasionally be required to carefully remove adhesion bands covering a wide area of the surface of the endometrium. Bipolar electrodes are now in common use for operative hysteroscopy and they allow the procedure to be used with physiological saline; some propose that they may reduce accidental electrical fluxes, making electrosurgery safer and reducing damage to the healthy myometrium .
Preventing adhesions from reforming is one of the most challenging aspects of hysteroscopic adhesiolysis. Recognition of the importance of this has resulted in a variety of interventions being recommended, used either in isolation or in synchrony with others. Several recent systematic reviews, including aggregated and network meta-analyses, have tried to make sense of the available evidence. These interventions are detailed below, followed by a summary.
Post-operative administration of oestrogen or progesterone is often adopted to enhance regrowth of the endometrium, stimulate re-epithelialisation of scarred surfaces and prevent recurrence. Oral progestin is given for a week or two during oestrogen therapy, to induce withdrawal bleeding . The damaged endometrium is perceived to be less responsive to hormone stimulation. For this reason, high-dose oestrogens may be used to stimulate rapid regeneration of the endometrium and, with prolonged use, to enhance continued regrowth of endometrium. For surgeons advocating post-procedural sex steroid hormone treatment, there is no consensus on the optimal duration, pattern (single or cyclical course), dosage or timing of sex steroids. Many regimens have therefore been suggested. The authors’ preferred regimen is to use a single post-operative course of oestradiol valerate 6 mg/day for six weeks, with the addition of dydrogesterone (Duphaston) 10 mg/day, or other progestogen, for the last two weeks of the oestrogen treatment.
The eclectic practice reflects the lack of clear evidence of benefit. In a randomised study reported by Farhi et al., women who were given oestrogen–progestin therapy after curettage for first-trimester abortions (n = 30) had significantly greater endometrial thickness (0.84 cm vs 0.67 cm; p <0.05) and endometrial volume (3.85 cm3 vs 1.97 cm3; p <0.05) than women without hormone treatment (n = 30) . However, this study focused on endometrial regeneration following intrauterine surgery and did not provide evidence to confirm the efficacy of hormone therapy administered following hysteroscopic surgery in reducing the formation of intrauterine adhesions. The value of oestrogen therapy in the management of intrauterine adhesions has been systematically reviewed by Johary et al., who analysed 26 studies, but still failed to draw a conclusion because of significant heterogeneity across the studies in their design and in the concurrent use of various adjuvant therapies . Thus, the value of post-operative hormonal treatment remains uncertain.
Intrauterine barriers include mechanical barriers (e.g. intrauterine devices, Foley balloon catheter, Cook Medical balloon uterine stent and amnion graft) and biochemical barriers (e.g. hyaluronic acid (HA)). The rationale for their use is to keep the dissected and denuded surfaces separated during the initial healing phase, so as to prevent the readherence of opposing uterine surfaces. Simultaneous administration of hormone therapy is usually used to promote epithelialisation of the exposed surface.
Many authors have advocated the insertion of an intrauterine contraceptive device (IUD) following hysteroscopic adhesiolysis, for 1–3 months [27–30]. Placement of the IUD can provide mechanical separation of the uterine cavity to reduce recurrence of adhesions after adhesiolysis and a number of IUDs have been studied. A loop IUD had been considered the best because of its large surface area, but it is no longer available in many countries . March et al. suggested that neither a T-shaped nor a copper IUD should be used for prevention of intrauterine adhesions, because a T-shaped IUD has a small surface area and a copper IUD may induce an excessive inflammatory reaction . The latter concern can be offset by manually removing the copper from the IUD. Levonorgesterol-releasing IUDs should also be avoided because continuous progestin release may prevent endometrial proliferation.
The effectiveness of IUDs in preventing reformation of IUAs is contentious because studies evaluating this outcome have been small and observational, with inconsistent populations, and often have employed co-interventions, making interpretation of the data problematic . One randomised study conducted after hysteroscopic septoplasty found no benefit of IUDs, with or without hormonal treatment, compared with either no treatment or hormone treatment alone .