Some gynaecologists advocate the use of a specially designed balloon which has a triangular shape instead of a round balloon, as it conforms to the normal shape of uterine cavity and maintains opposing walls apart, especially the margins of the uterine cavity [13]. The Cook^® balloon is particularly useful after division of intrauterine adhesions, while in cases of adhesions limited to endocervical canal or internal cervical os, a Foley catheter may be more useful.

Myers and Hurst [15] carried out a retrospective case series to investigate the combined use of Cook^® intrauterine balloon and intrauterine contraceptive device in women with severe intrauterine adhesions. After adhesiolysis, the Cook^® balloon was inserted into uterine cavity and was kept in place for about 3–7 days. Then a copper IUD was placed into the cavity and removed 4–6 weeks later. This combined approach achieved favourable outcomes with regards to resumed menses, pregnancy and delivery.

Human amnion has been developed as a biological dressing to promote healing of burned skin and skin ulcers [28, 29], as a graft in ophthalmology for ocular surface reconstruction [30, 31], and in gynecology in the formation of an artificial vagina, and as a barrier to prevent formation of postoperative adhesions in abdominal surgery [32]. The amnion membrane displays anti-inflammatory and anti-bacterial properties, as well as having a wound protection effect [33]. It also promotes epithelialisation by facilitating migration, proliferation or differentiation of epithelial cells [34–37], reinforcing adhesion of basal cells [38], and preventing apoptosis of epithelial cells [39].

Thirdly, it decreases reformation of postoperative adhesions by reducing inflammation, fibrosis and the extent and severity of vascularisation [40]. There is little immunologic reaction in response to the amnion membrane; so there is no need for the use of systemic immunosuppressant. Therefore, the amnion membrane is considered by some authors as an ideal adjunctive therapy following hysteroscopic adhesiolysis.

In 2006, Amer et al. [41] firstly advocated the use of amnion graft in uterine cavity following hysteroscopic adhesiolysis for prevention of intrauterine adhesion recurrence and found encouraging results. They carried out a pilot study involving 25 patients with moderate or severe intrauterine adhesions. After lysis of intrauterine adhesions, a paediatric 10 Fr Foley catheter, with amnion graft on its surface, was introduced into uterine cavity. The balloon was distended with 3.5–5 ml of saline. The stem of the catheter was cut off just below a knot near the balloon end, so that the balloon with the graft on its surface was kept in uterus. Oral antibiotics were given and the balloon was removed 14 days later.

At second look hysteroscopy, significant improvement was noted in uterine length and adhesion scores in both moderate group and severe group. The reformation of intrauterine adhesions was observed in 12 patients (12/25 = 48 %), all of whom were in the group with severe adhesions.

Four years later, the same team reported a prospective randomized trial to assess the efficacy of amnion graft in preventing the reformation of intrauterine adhesions [26]. Forty-five patients with severe intrauterine adhesions were randomized into three groups: insertion of an intrauterine balloon only (group 1), insertion of an intrauterine balloon with fresh amnion graft (group 2) or dried graft (group 3). At second look hysteroscopy, there was significant improvement in adhesion grade with amnion graft groups compared with intrauterine balloon alone group. Among 10 pregnancies achieved in all patients, 8 were in amnion graft groups and 2 in group without amnion. Therefore, the authors concluded that intrauterine administration of amnion graft after hysteroscopic lysis of adhesions was a promising adjunctive procedure for decreasing recurrence of adhesions.

Peng and Xia [42] applied a fresh amnion graft after lysis of intrauterine adhesions in 34 patients with moderate or severe intrauterine adhesions. Fresh amnion was cut and placed on the surface of a 14 Fr Foley balloon with the chorion side facing outwards. The Foley catheter with amnion graft on its surface was inserted into uterine cavity at the end of the surgery, and 3–5 ml saline was injected into balloon. Oral antibiotic was given and the balloon was kept in place for 7 days. Cyclic hormone treatment was given for 3 months. The results showed no operative complications and no clinical evidence of infections. Most of the patients experienced significant achievement of menstrual flow (31/34 = 91.2 %). At second look hysteroscopy, reformation of intrauterine adhesion was detected in 11 cases, and all of them originally had severe adhesions. Furthermore, no reformation of severe adhesions was observed. They concluded that the use of amnion as a biologic barrier in prevention of intrauterine adhesion reformation was safe and effective. Although these studies show promising results large prospective studies with long term follow up data are needed to confirm its clinical use.

Some investigators tried absorbable adhesion barriers such as hyaluronic acid following hysteroscopic surgery to prevent the reformation of intrauterine adhesions. Hyaluronic acid (HA), a water-soluble polysaccharide, is a natural component which is widely distributed in a variety of tissues, such as the extracellular matrix, the vitreous humour and synovial fluid of the joint. HA has been chemically modified to produce higher adhesive property and more prolonged in vivo residence time than unmodified HA [43] which is more suitable for use in the prevention of adhesion formation. It has been used as a barrier to reduce adhesion formation after abdominal and pelvic surgery [44], as well as intrauterine surgery [45, 46].

Modified HA preparations used to reduce the formation of adhesions following intrauterine surgery includes Seprafilm^® and auto-cross-linked HA (ACP) gel (Hyalobarrier^® gel).

Seprafilm^® is a bioresorbable membrane formulated from chemically modified HA (sodium hyaluronate) and carboxymethyl cellulose, and has been widely used in clinical practice. After placement of this membrane, it turns into a hydrophilic gel in 24 h, and provides a protective coating on tissue surface for up to 7 days.

ACP gel is a fully biocompatible cross-linked derivative of HA and is formed by an internal auto-cross-linking reaction of pure hyaluronan, which does not introduce any foreign substance into the molecule [47]. The viscosity of ACP gel can be modulated by adjusting the degree of crosslinking, which depends on its molecular weight and concentration [48]. The increased viscosity makes it possible to introduce the gel into uterine cavity and remain in situ for at least 72 h.

There has been only one study on the use of Seprafilm in preventing intrauterine adhesions [45]. In this randomized controlled study of the safety and efficacy of Seprafilm, a Seprafilm membrane was cut into two pieces, with each of them rolled up into a thin cylinder. One cylinder was placed in uterine cavity, while the other in cervical canal. Hysterosalpingography (HSG) was performed 8 months later to evaluate the uterine cavity. Fewer patients developed intrauterine adhesions in the Seprafilm^® group (1/10; 10 %) than those in the control group (7/14; 50 %).

There were several studies on the use of ACP gel in uterine cavity for preventing intrauterine adhesion formations after hysteroscopic surgery. De Iaco et al. [49] carried out a prospective randomized trial to evaluate the effectiveness of an ACP gel (Hyalobarrier^® gel) in preventing formation of intrauterine adhesions after hysteroscopic surgeries. At the end of the hysteroscopic procedure, the gel (5–20 ml) was introduced into uterine cavity with a cannula (length 20 cm, width 5 mm) to cover the entire uterine cavity. Second-look hysteroscopy was performed 9 weeks after the initial operation. Intrauterine adhesions were detected in 5 patients (5/18 = 27.8 %) in ACP gel group and 7 patients (7/22 = 31.8 %) in control group. There did not appear to be any difference between two groups.

In the same year, Acunzo et al. [46] reported the findings of a prospective randomized controlled study with ACP gel (Hyalobarrier® gel). After lysis of the intra uterine adhesions, the gel was introduced into the uterine cavity through the out-flow channel of the hysteroscope. The injection of the gel was monitored with hysteroscope and the uterine cavity was fully filled by the gel from tubal ostia to internal uterine orifice. Second-look hysteroscopy was performed 3 months later. At follow-up, the recurrence rate of intrauterine adhesions was significantly lower in ACP gel group (6/43 = 13.95 %) than that in control group (13/41 = 31.70 %). Furthermore, the severity of adhesions was significantly decreased in ACP gel group compared with control group.

In 2004, the same team reported on the use of Hyalobarrier gel in preventing intrauterine adhesions following hysteroscopic surgery (resection of myomas, polyps and septae). At second look hysteroscopy, intrauterine adhesions were found in seven patients with the ACP gel (7/67 = 10.44 %), compared with 17 patients without the gel (17/65 = 26.15 %) (p < 0.05). They concluded that Hyalobarrier^® gel was a safe and effective intrauterine barrier in preventing de novo adhesion formation following hysteroscopic surgery [50].

Ducarme et al. [51] conducted a case–control study in order to examine the efficacy of ACP gel in preventing adhesions after hysteroscopic surgery, but had opposite results. No differences were observed in both incidence and severity of intrauterine adhesions reformation between the group with the gel and the group without the gel.

Mais et al. [52] conducted a systematic review and meta-analysis of randomized controlled trials on efficacy of ACP gel for adhesion prevention in laparoscopy and hysteroscopy, and concluded that the limited data so far available suggested that ACP gel appeared to have benefits in preventing intrauterine adhesions after hysteroscopic surgery.

While several authors [53–55] suggested that infections might be a primary cause of intrauterine adhesions, many official guidelines (such as the American College of Obstetricians and Gynecologists guidelines) do not recommend the routine use of antibiotics in patients undergoing operative hysteroscopy [56, 57]. Bhattacharya et al. [58] carried out a prospective randomized study on the effect of prophylactic antibiotics with intravenous injection of 1.2 g Augmentin (co-amoxiclay) at induction of anaesthesia, on the incidence of bacteraemia following hysteroscopic surgery. They showed that incidence of bacteraemia was significantly higher in control group (10/61 = 16 %) than that in the antibiotic group (1/55 = 2 %). Nappi et al. [59] performed a double-blind, randomized, placebo-controlled study to evaluate the effectiveness of antibiotics and the incidence of infections following operative hysteroscopy. Antibiotics with 1 g of cefazolin were given intramuscularly in 523 patients during the procedures. Five days after operations, 5 patients were found with symptoms related to infections in antibiotic group (5/523 = 1.0 %) and 7 in the untreated group (7/523 = 1.3 %). They concluded that routine administration of antibiotics is not necessary for hysteroscopic surgery.

Prevention of both formation and reformation of intrauterine adhesions is a challenging issue in clinical practice. A variety of methods have been proposed including hormone treatment, intrauterine mechanical barriers, such as IUDs and intra uterine balloons, which all appear to have some benefit. Both Hyaluronic acid gel and amnion graft have promising results but further robust clinical trials are required to confirm their efficacy.