- Email: [email protected]
Creation of a female rabbit model for intrauterine adhesions using mechanical and infectious injury
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Available online at www.sciencedirect.com
journal homepage: www.JournalofSurgicalResearch.com
Creation of a female rabbit model for intrauterine adhesions using mechanical and infectious injury Fang Liu, PhD (ABD),a,b Zhi-Jun Zhu, MM (ABD),c Peng Li, PhD (ABD),d and Yuan-Li He, MMa,* a
Department of Obstetrics and Gynecology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China Department of Obstetrics and Gynecology, Shihezi University School of Medicine, Shihezi, Xinjiang, China c Department of Cardiothoracic Surgery, Shihezi University School of Medicine, Shihezi, Xinjiang, China d Department of Neurosurgery, Key Laboratory on Brain Function Repair and Regeneration of Guangdong, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China b
article info
abstract
Article history:
Background: Intrauterine adhesions (IUA) are associated with secondary amenorrhea,
Received 29 June 2012
infertility, and recurrent pregnancy loss. An IUA animal model would contribute to research
Received in revised form
on the pathogenesis and pathologic changes of IUA and the exploration of new treatments.
25 October 2012
Materials and methods: Eighty female rabbits were randomly divided into five groups:
Accepted 8 November 2012
mechanical injury (16), infectious injury (16), dual injury (16), experimental control (16), and
Available online 24 November 2012
normal (16). Three methods were applied to establish the model: uterine curettage, uterine cavity left alone, lipopolysaccharide surgical suture in place for 48 h, and suture retention
Keywords:
for 48 h after curettage. A sterile surgical suture was left in the uterine cavity for 48 h in the
Intrauterine adhesions
experimental control group. Histologic changes were monitored at 0, 24, 48, and 72 h and 7,
Endometrium
14, and 28 d after operation.
Injury
Results: The experiments revealed that endometrium injured by simple curettage or
Repair
infection could be repaired. Although endometrial regeneration was severely impaired by
Fibrosis
dual injury, the ratio of the area with endometrial stromal fibrosis to total endometrial area
Animal model
significantly increased (P < 0.01), and the number of endometrial glands was significantly
Rabbit
reduced (P < 0.01). Conclusions: The method of dual injury can establish a stable rabbit IUA model. ª 2013 Elsevier Inc. All rights reserved.
1.
Introduction
Intrauterine adhesions (IUA) refer to the partial adherence of endometrial surfaces with fibrotic tissue. Presenting symptoms are related to the degree and location of IUA and include menstrual abnormalities ranging from irregular bleeding and hypomenorrhea to amenorrhea, infertility, and recurrent pregnancy loss [1]. In 1948, Asherman [2e4] published a series of articles describing the frequency, etiology, symptoms, and
roentgenologic picture of this condition, and Asherman syndrome has been used to describe the disease ever since. With the popularization of hysteroscopy and surgery, the incidence of IUA has increased and has become the second most common of female secondary infertility [5]. There are four aspects that relate to IUA: trauma [6], infection [7,8], endometrial repair disorders [9], and low estrogen levels [10]. Previous studies have focused on new treatments for preventing endometrial fibrosis and improving endometrial
* Corresponding author. Department of Obstetrics and Gynecology, Zhujiang Hospital, No. 253 Gongye Road, Guangzhou, Guangdong 510282, China. Tel.: þ86 133 188 07468; fax: þ86 020 616 43361. E-mail address: [email protected] (Y.-L. He). 0022-4804/$ e see front matter ª 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2012.11.009
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
regeneration [11e13]; however, animal models to research the mechanism and curative effects of these new treatments are still scarce. It is necessary to establish an animal model of IUA to intensively research the pathogenesis and pathologic changes of IUA and to explore new treatments. Lipopolysaccharide (LPS) is the major structural component of the cell wall of gram-negative bacteria and is also the endotoxin responsible for much of the inflammation and tissue injury associated with bacterial infections [14]. LPSinduced animal models highlight ways to explore mechanisms of multiple diseases and provide useful information on the discovery of novel biomarkers and treatment strategies [15e17]. However, reports of the effects of LPS on endometrium are rare. The present study aimed to establish an animal model of IUA in the New Zealand white rabbits using mechanical damage, infection, and double damage as is similar to the two main factors in the pathogenesis of IUA: trauma and infection. In this study, curettage was used as a method of mechanical damage and LPS as a predisposing factor for infection. Through this study, we hope to provide an ideal animal model for the study of the exact pathogenesis of and therapy for this difficult clinical condition.
four rabbits were killed at 7 d. Uterine tissue samples were collected.
2.3.1.
Materials and methods
2.1.
Animals
Experimental protocols were approved by the Institutional Animal Ethics Committee. Adult female New Zealand white rabbits (Guangzhou, Guangdong, China) weighing 2500e3500 g were fed standard chow and water ad libitum under standardized laboratory conditions in a temperature-controlled room and light conditions (12 h light and 12 h dark) for 2 wk.
2.2.
2.3.2.
Infectious injury group (n ¼ 32 uterine horns)
An LPS surgical suture was inserted in the uterine cavity through a uterine incision identical to that of the mechanical injury group. The tail end of the LPS surgical suture was placed through the abdominal wall, and a muscular layer was retained on the surface of the skin to facilitate removal after 48 h. Two rabbits (n ¼ 4 uterine horns) were killed for the collection of uterine tissue once the LPS surgical suture was removed.
Dual injury group (n ¼ 32 uterine horns)
After uterine curettage, performed in an identical manner to that of the mechanical injury group, an LPS surgical suture was placed in the uterine cavity (Fig. 1B). The LPS surgical suture was removed at 48 h after surgery, and two rabbits (n ¼ 4 uterine horns) were killed for the collection of uterine tissue.
2.3.4.
Experimental control group (n ¼ 32 uterine horns)
A sterile surgical suture was left in the uterus for 48 h as an experimental control group. Similarly, two rabbits (n ¼ 4 uterine horns) were killed when the surgical suture was removed.
LPS surgical suture preparation 2.3.5.
We obtained LPS from the manufacturer (derived from Escherichia coli 055:B5; Sigma, St. Louis, MO). A 6 mg/L solution of LPS in normal saline was stored in a 4 C refrigerator. The 10-0 medical sterile surgical sutures (10 cm) were soaked in an LPS saline solution for 24 h the day before use in the model.
2.3.
Mechanical injury group (n ¼ 32 uterine horns)
The rabbits were operated on through a 0.5-cm longitudinal incision in the lowest one-third of the connection between the middle and distal uterus and then the endometrial lining of the middle and upper two-third of the uterus was scraped using a 4-mm endometrial curette. Curettage ceased when the uterine cavity stopped feeling uneven and the uterine wall became rough. The uterine tissue from two rabbits (n ¼ 4 uterine horns) was collected immediately after curettage, and the abdominal incisions of the other rabbits were sutured after flushing the uterine and peritoneal cavities.
2.3.3.
2.
297
Method for establishing animal models
One hundred sixty uterine horns from 80 female New Zealand white rabbits were randomly divided into five groups: normal, experimental control, mechanical injury, infectious injury, and dual injury (mechanical and infection). Twenty-four hours before the creation of the model, the rabbits were injected with human chorionic gonadotropin (50 IU) by ear vein to reach synchronization of endometrial development. All the rabbits were anesthetized with urethane (1.5 g/kg) administered intravenously through the marginal ear vein and supplemented if necessary. Throughout the surgery, particular care was taken to confirm an adequate level of anesthesia such that noxious forelimb pinching evoked no response. Two rabbits (n ¼ 4 uterine horns) were killed in each group at 0, 24, 48, and 72 h and 14 and 28 d after surgery, and
Normal group (n ¼ 32 uterine horns)
In the normal group, rabbit abdomens were opened to expose the bilateral uteri but were not given any treatment.
2.4.
Histologic examination
The uterine tissues were fixed in 4% buffered formaldehyde, embedded in paraffin, and routinely stained with hematoxylineeosin and Masson stains. Histologic evaluation was performed by an experienced pathologist. On each hematoxylin- and eosin-stained slice, four high-power fields were selected, the number of glands per high-power field was counted, and the mean was calculated. On each Masson-stained slice, four high-power fields were selected, and the degree of endometrial fibrosis was calculated as the ratio between the area of endometrial stromal fibrosis and endometrial area per high-power field. The average of four fields was calculated.
2.5.
Statistical analysis
Data were expressed as the mean and standard errors of the mean. Statistical analysis was performed using SPSS for
298
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Fig. 1 e (A) Normal rabbit uterine appearance is didelphic and tubular in structure and 7 cm long and 3e4 mm wide on average. (B) Rabbit uterus damaged by the dual injury method; bilateral uterine cavities have a retained LPS surgical suture. (C) Rabbit uterus damaged by the dual injury method after 7 d (bar [ 1 cm). (Color version of figure is available online.)
Windows 13.0 (SPSS, Chicago, IL). One-way analysis of variance was used for comparison between two groups, and one-way analyses of variance were followed by StudenteNewmaneKeuls tests for comparisons between the four groups. A P value of < 0.05 was considered to denote a significant difference.
3.
Results
3.1.
Endometrial morphology in rabbit
3.1.1.
Normal endometrium
The uterine surface observed with the naked eye was gray and scraggly. The shape of the uterine cavity was irregularly star shaped and strongly extensible, and the endometrium was composed of four to six polypoid phymata with the surface endometrium covered with columnar epithelial cells (Figs. 1A and 2A). Round or oval endometrial glands were mainly located in the submucosa and basal layer; moreover, gland openings in the submucosa were at the endometrial surface (Fig. 3A). After Masson staining, the mucosa, submucosa, blood vessels, and muscle were dark red, but endometrial stromal fibers appeared blue (Fig. 3F).
3.1.2.
Endometrium by mechanical injury
Immediately after the lesions were produced, almost all the surface epithelium was denuded; scattered basal glands remained only in some deeper stromal fold regions. The stroma was bare and thinned; even the full thickness of the endometrium disappeared in the severely injured areas. Extensive recent hemorrhage and neutrophil infiltration were evident in the stroma. The capillaries were ruptured and had bled. Epithelial debris and red blood cells were visible in the uterine cavity. Twenty-four hours after the injury, part of the uterine cavity surface (20%e30%) was covered with a few flat regenerative epithelial cells. The stroma was hemorrhagic, edematous, and had lymphocytic infiltration. The capillaries had bled. Forty-eight hours after the injury, most uterine cavity surface (70%e80%) had regenerating epithelial cells, but there was no evidence of gland regeneration. Hemorrhage, edema, leukocyte infiltration, and coagulation necrosis were observed in the stroma and a fibrinous exudate in the bare stromal area.
Seventy-two hours after curettage, 90% of the regenerated surface epithelial cells were cuboidal or low columnar, and mitoses were occasionally observed. There were a few round glands beneath the regenerating epithelium. Except for areas of hemorrhage, edema, and lymphocyte infiltration, fibrous tissue proliferation was present in the stroma. Seven days after the injury, the surface was completely lined by epithelial cells, and no areas of denudation persisted. The regenerating cells were flattened (60%e70%), cuboidal, or low columnar cells. Occasional mitotic figures were observed in the taller cuboidal cells. Beneath the regenerating epithelium, only a few glands could be distinguished. The stroma prominently showed more fibrous tissue, collagen aggregation, focal hemorrhages, congestion, and leukocytic infiltration. A few newborn capillaries were visible (Figs. 2B and 3G). After 14 d, the surface was completely lined by epithelial cells and glands. Obvious mitosis was identifiable in the epithelial cells. The stroma showed a few polymorphonuclear leukocytes and infiltration of hemosiderinladen macrophages. Part of the area exhibited interstitial fibrosis and collagen aggregation. More newborn capillaries were visible. After 28 d, the histologic reconstruction of endometrium was largely completed, other than fibroses that could be observed in a minority of the stromal area. The stroma was neither congested nor infiltrated with leukocytes.
3.1.3.
Endometrium by infectious injury
Immediately after the LPS surgical suture was removed, 30%e40% of endometrial cells were shed. The stroma was exposed at the area of epithelial shedding and was accompanied by extensive hemorrhage, edema, neutrophil infiltration, telangiectasias, and congestion. After 24 h, 30% of the surface was lined with regenerating endometrial cells, which were flat with deeply stained large nucleoli, basophilic cytoplasm, and indistinct cell borders. Endometrial epithelial cells were completely regenerated, and all but 5% of the surface was uncovered, with endometrial, epithelial, and exposed stroma apparent. There were no glands visible underneath regenerating epithelial cells. The edema and neutrophil infiltration were more obvious in the stroma. After 48 h, the entire surface was covered with endometrial cells, and some of them (30%e40%) were low columnar or cuboidal but without glands beneath them. Other areas had
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
299
Fig. 2 e The profile of a rabbit uterus. (A) Normal uterine cavity with many polypoid phymata. (B) The endometrial surface was noted to have petechial hemorrhage by mechanical injury after 7 d. (C) Rabbit uterus damaged by the infectious method after 7 d. (D) The uterine cavity distinctly formed adhesions 7 d after dual injury (bar [ 0.5 cm). (Color version of figure is available online.)
normal endometrial morphology. The edema and neutrophil infiltration were still visible in the stroma. After 72 h, most endometrial cells were cuboidal or columnar epithelial cells, and a portion of the subepithelium was observed to have a few round regenerated glands. The edema and neutrophil infiltration were less prominent in the stroma. After 7 d, the surface was lined by columnar epithelial cells and, occasionally, high columnar epithelial cells. Many glands beneath the epithelium and their openings were visible at the surface. However, the stroma still featured edema and neutrophil infiltration (Figs. 2C and 3H). After 14 d, the epithelium and glands were completely regenerated. Distinct mitoses were recognizable in the epithelial cells. A few polymorphonuclear leukocytes and hemosiderin-laden macrophages were detected in the stroma. After 28 d, the histomorphology of the endometrium was completely normal. There was no congestion or lymphocytic infiltration in the stroma.
3.1.4.
Endometrium after dual injury
Immediately after dual injury, the endometrial epithelium was completely shed, except for an occasionally scattered gland. There was a large area of hemorrhage and neutrophilic infiltration in the naked stroma. Ruptured and hemorrhaging capillaries were prominent (Fig. 3B).
Twenty-four hours after dual injury, regeneration of the epithelial cells was barely visible, and there were marked fibrinous exudates in the stroma. Diffuse hemorrhage, edema, and dense lymphocytic infiltration were observed in the stroma. Forty-eight hours after the creation of the animal model, the regeneration of the flat epithelial cells was rarely observed on the surface, and a great quantity of fibrinous exudates and necrotic tissue debris emerged in the uterine cavity. Part of the stroma began to show fibrotic changes. Some capillary lumens were atretic and surrounded by fibrous tissue (Fig. 3C). Seventy-two hours after the creation of the model, a section of uterine cavity was atretic, and fibrous adhesions appeared in the surface. Five percent of the area without adhesions was observed to have sparse regenerated flat epithelial cells, and no gland regeneration was evident. The proliferation of stromal fibrosis was evident and accompanied by edema, lymphocytic infiltration, and capillary expansion (Fig. 3D). Seven days after the injury, the uterus was vertically cut, and on the axial view, some highly fibrous adhesions occurred in the uterine cavity. Microscopically, the fibrous adhesions thickened, and 70%e80% of the area without adhesions was lined by flattened or low columnar epithelial cells, and scarce rounded glands were present beneath the epithelial layer (Figs. 1C, 2D, and 3E).
300
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Fig. 3 e (AeE) Hematoxylineeosin sections of rabbit uterine tissue. (A) The normal rabbit endometrium protrudes into the uterine cavity, and many glands are located in the stroma (magnification, 3100). (B) Immediately after dual injury, the endometrial epithelium was completely shed, and a large area of hemorrhage was observed in the naked stroma (magnification, 3100). (C) Forty-eight hours after the dual injury, the uterine cavity showed distinct fibrinous exudates (magnification, 3100). (D) Seventy-two hours after dual injury, the uterine cavity appeared to have adhesions (magnification, 3100). (E) Seven days after dual injury, extensive fibrosis in the uterine cavity and rare glands in the stroma were observed (magnification, 3100). (FeI) Masson sections of rabbit uterine tissue. (F) High-power photomicrograph shows that endometrium, glands, and blood vessels are dyed dark red, but endometrial stromal fibers are blue (magnification, 3200). (G) Seven days after curettage, high-power photomicrographs showed that regenerated endometrium and glands were present, blood vessels were congested, and fibrous tissue of the stroma was increased (magnification, 3200). (H) Seven days after infectious injury, many regenerated glands are observed at the surface, and the blood vessels are congested (magnification, 3200). (I) Seven days after dual injury, a low-power photomicrograph shows interstitial fibrosis and no glands in the stroma (magnification, 3200) (bar [ 200 mm). (Color version of figure is available online.)
Fourteen days after the injury, histology revealed nonfunctional flattened epithelial cells covering the areas without adhesions, but the level of IUA resembled that seen after 7 d. Twenty-eight days after the injury, the endothelial morphology had failed to reconstruct and had formed a scar.
3.1.5.
Endometrium in the experimental control group
Less than 5% of the endometrial luminal epithelial cells were shed after the sterile suture was removed, and scattered red blood cells were visible in the uterine cavity. The endometrium in the remaining regions had the same histomorphology as the normal endometrium. After 24 h, the entire uterine surface was covered with endometrial epithelial cells. After 48 h, the
histomorphology of the endometrium returned to normal. Normal endometrium was also observed at 72 h, 7 d, 14 d, and 28 d.
3.2.
The degree of endometrial fibrosis in rabbits
The ratios of the area with endometrial fibrosis to total endometrial area were calculated before injury and 7 d after injury (Table 1), and the dual injury group had the highest ratio (P < 0.01) (Fig. 4A). The number of endometrial glands per high-power field was also counted, and the dual injury group had the fewest of the four groups (P < 0.01), whereas the mechanical injury group had fewer than the normal group (P ¼ 0.03) (Fig. 4B and Table 1).
301
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Table 1 e The percentage of fibrotic area and the number of glands in each group.
Fibrosis area (%) Number of glands (/HPF)
Normal
Curettage
LPS
Curettage þ LPS
Control
27.50 3.34 14.97 0.59
40.52 3.72 12.66 0.52y
37.80 3.79 13.59 0.55
68.03 4.10* 4.19 0.48z
28.33 3.58 15.00 0.44
HPF ¼ high-power field. Results are shown as the mean standard deviation. * P < 0.05 curettage þ LPS versus normal, curettage, and LPS. y P < 0.05 curettage versus normal. z P < 0.05 curettage þ LPS versus normal, curettage, and LPS.
4.
Discussion
A general consensus has been reached that the most common causes of IUA are trauma and infection [18]. Any trauma that destroys the endometrial basal layer could result in IUA, including abortion, curettage, and hysteroscopic surgery, among others [19]. After the endometrial basal layer is injured, the endometrium failed to regenerate and repair, and the uterine cavity appeared to form scars and adhesions. Studies have currently found no “consistent evidence” of the inflammatory processes causing IUA; nevertheless, some investigators [20] believe that inflammatory processes do contribute to the damaging effect of trauma and act synergistically to produce IUA [21]. At present, clinical treatment for IUA can restore the shape of the uterine cavity, but it remains difficult to repair uterine physiological function. Therefore, the rates of cure and pregnancy are low, particularly in moderate and severe IUA. Some possible treatments of IUA include direct delivery of stem cells or specific growth factors into the uterus [11,22]. Thus, it is essential to establish a new model of IUA to evaluate novel treatment options. In the present study, we designed three methods of creating animal models of IUA and have demonstrated that dual injury with curettage and LPS is a suitable method to establish a model. In this study, all layers of the rabbit endometrium were injured by undergoing curettage, and the endometrial cells began to regenerate in a short time. Although the repair process of the stroma showed was slow, the morphology of the endometrium was almost entirely repaired after 28 d. This
finding illustrates that the process of endometrial regeneration was unaffected by curettage injury [23]. LPS has many deleterious effects and plays a significant role in a number of disease processes by increasing inflammatory cytokine release [24,25]. In this study, we used LPS as an inflammation-inducing factor, and in consideration of the special anatomical structure of the rabbit uterus, we also used a surgical suture as a carrier of LPS to ensure the persistence of LPS in the uterine cavity. We demonstrated that the method of infectious injury could also ablate part of the surface, but the damaged endometrium could be completely rebuilt. Seven days after the injury, there were no significant differences between the infected injury and normal groups regarding the number of glands and the ratio of fibrotic tissue area to normal tissue area. Endometrial stromal fibrosis temporarily increased, but there were no fibrous scars or adhesions on the surface covered with epithelial cells. Present studies suggest that binding of LPS to Toll-like receptor 4 activates cellsignaling pathways leading to an inflammatory response and secretion of the cytokines interleukin-1b, interleukin-6, and tumor necrosis factor alpha among others. In addition, infected tissue and organs appear to undergo various pathologic changes, including edema, hemorrhage, fibrosis, and necrosis [26e29]. Although our studies did not examine these inflammatory cytokines, the pathologic changes of rabbit endometrium treated with LPS were compatible with other models induced by LPS. In our studies, the endometrium treated with LPS surgical suture alone could be fully repaired, perhaps because of a lower amount of inflammation induced by low-dose LPS [30]. The results of the study were in line with
Fig. 4 e (A) The value represents the area of stromal fibrosis quantified as a percentage of the total endometrial area in the visual field (magnification, 3200). *P < 0.05 curettage D LPS versus normal, curettage, and LPS. (B) We counted the number of glands in the visual field (magnification, 3200). *P < 0.05 curettage versus normal and **P < 0.05 curettage D LPS versus normal, curettage, and LPS. (Color version of figure is available online.)
302
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
the work by Polishuk [31] documenting 171 women who had undergone cesarean sections. Among these 171 cases, 28 patients developed significant endometritis. Afterward, hysterosalpingograms showed no difference in the occurrence of IUA between the endometritis and control groups. The authors concluded that endometritis is unlikely to be a major factor in the pathogenesis of IUA. An experimental control group was established to exclude the interference of surgical suture. The endometrium treated with surgical suture alone was only observed to have slight epithelial shedding and scattered red blood cells in the uterine cavity. These findings suggested that LPS was the main inflammatory inducer, not the surgical suture. There was no endometrial regeneration 72 h after endometrium was removed by the dual injury method, but then, fibrinous exudates obstructed the uterine cavity and formed adhesions. Finally, the endometrium was repaired through scar tissue. This finding suggests that the endometrium could achieve a perfect nonscarring self-repair with rapid epithelialization and a reduced inflammatory reaction [32]. Curettage to the endometrium destroyed the endometrial barrier, and the inflammatory response induced by LPS caused edema, inflammatory cell infiltration, epithelial cell necrosis, and stromal fibrosis hyperplasia [33]. The double hit of curettage and LPS interfered with the normal repair process after endometrial injury. It may be that the shedding of epithelial cells and the rupture of capillaries after curettage promoted the diffusion of the inflammatory response induced by LPS [30]. The delay in regeneration of endometrial epithelial cells may be important to the formation of IUA after endometrial damage [34]. Because of the deficiency of sufficient epithelial cells covering the uterine cavity, the stroma is bared, allowing fibroblast activity and connective tissue formation leading to endometrial fibrosis and possibly to IUA [21]. Since the 1970s, a succession of researchers have established the model of endometrial injury in the rabbit using various methods. Schenker and Polishuk [35] introduced the model of endometrial hypothermic impairment in the rabbit by applying a probe containing Freon and a model of endometrial chemical damage using an injection of 10% formalin into the uterine cavity [36]; Birkenfeld et al. [37] established endometrial damage models by Nd:YAG laser that retarded the regeneration of endometrium in the rabbit. Polishuk [31] transplanted polyethylene sponges carrying autologous fibroblasts in the injured uterine cavity of the rabbit or rat. Interestingly, the purpose of researchers developing a model of endometrial damage or adhesions is to explore the treatment of perimenopausal dysfunctional uterine bleeding, surgical contraindications in patients with hysteromyoma, or even as a contraceptive method rather than for research on IUA. These methods result in varying degrees of injury to the endometrium, including adhesion formation. However, these animal models mainly apply some physical or chemical lesion and are different from the pathogenesis of IUA. Therefore, we cannot confirm that the pathologic changes caused by these methods are the same as those that occur in vivo. In addition, chemical or physical materials retained in the uterine cavity may have affected the subsequent evaluations and analyses. IUA is the result of many factors [38]. It is difficult to establish an IUA model using only one method of damage
because of the strong regenerative capacity of endometrium, so the model may be easier to construct by combining two types of damage methods. To study the pathogenesis of IUA in the postpartum uterus after a uterine cavity operation, Chen et al. [39] established an IUA model in rabbit through curettage and bilateral oophorectomy. Li et al. [40] placed a copper wire in the uterine cavity after curettage and concluded that endometrial injury impeded blastocyst implantation in rabbit. In this study, we constructed a rabbit IUA model by means of curettage and infection to simulate the two main causes of IUAdtrauma and infection. The endometrial layer was completely ablated, and the LPS surgical suture was placed in the uterine cavity immediately after curettage and retained for 48 h. This method inhibits the regeneration of endometrium [41], and LPS contributes to scar repair by increasing the inflammatory reaction [42]. Timely removal of the LPS surgical suture from the uterine cavity has no influence on the formation of intimal fibrosis and adhesions. In summary, a stable IUA model can be established using the dual injury method.
5.
Conclusion
The combination of curettage and infection creates an IUA model in the rabbit that resembles the condition in humans, taking into account the features such as histologic changes in the uterine cavity. This rabbit model provides an opportunity to perform studies in a standardized animal model to explore new treatments and examine their effectiveness and safety.
Acknowledgment This study was supported by the National Nature Science Foundation of China (81270658 to Y.L.H.).
references
[1] Panayotidis C, Weyers S, Bosteels J, van Herendael B. Intrauterine adhesions (IUA): has there been progress in understanding and treatment over the last 20 years? Gynecol Surg 2009;6:197. [2] Asherman JG. Traumatic intra-uterine adhesions. J Obstet Gynaecol Br Emp 1950;57:892. [3] Asherman JG. Inversio uteri puerperalis. Gynaecologia 1954; 138:410. [4] Asherman JG. Uterine adhesions. Harefuah 1958;55:138. [5] Roge P, Cravello L, D’Ercole C, Brousse M, Boubli L, Blanc B. Intrauterine adhesions and fertility: results of hysteroscopic treatment. Gynaecol Endosc 1997;6:225. [6] Taylor PJ, Cumming DC, Hill PJ. Significance of intrauterine adhesions detected hysteroscopically in eumenorrheic infertile women and role of antecedent curettage in their formation. Am J Obstet Gynecol 1981;139:239. [7] Gupta N, Sharma JB, Mittal S, Singh N, Misra R, Kukreja M. Genital tuberculosis in Indian infertility patients. Int J Gynaecol Obstet 2007;97:135. [8] Al-Inany H. Intrauterine adhesions. An update. Acta Obstet Gynecol Scand 2001;80:986.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
[9] Reyniak JV, Gordon M, Stone ML, Sedlis A. Endometrial regeneration after voluntary abortion. Obstet Gynecol 1975; 45:203. [10] March CM. Intrauterine adhesions. Obstet Gynecol Clin North Am 1995;22:491. [11] Nagori CB, Panchal SY, Patel H. Endometrial regeneration using autologous adult stem cells followed by conception by in vitro fertilization in a patient of severe Asherman’s syndrome. J Hum Reprod Sci 2011;4:43. [12] Chason RJ, Levens ED, Yauger BJ, Payson MD, Cho K, Larsen FW. Balloon fluoroscopy as treatment for intrauterine adhesions: a novel approach. Fertil Steril 2008;90:2005. [13] Amer MI, Abd-El-Maeboud KH. Amnion graft following hysteroscopic lysis of intrauterine adhesions. J Obstet Gynaecol Res 2006;32:559. [14] Wang X, Quinn PJ. Lipopolysaccharide: biosynthetic pathway and structure modification. Prog Lipid Res 2010;49:97. [15] Hasegawa A, Iwasaka H, Hagiwara S, Noguchi T. Relationship between HMGB1 and tissue protective effects of HSP72 in a LPS-induced systemic inflammation model. J Surg Res 2011; 169:85. [16] Fujioka H, Akema T. Lipopolysaccharide acutely inhibits proliferation of neural precursor cells in the dentate gyrus in adult rats. Brain Res 2010;1352:35. [17] Saadani-Makki F, Kannan S, Lu X. Intrauterine administration of endotoxin leads to motor deficits in a rabbit model: a link between prenatal infection and cerebral palsy. Am J Obstet Gynecol 2008;199:651. [18] Yu D, Wong YM, Cheong Y, Xia E, Li TC. Asherman syndromedone century later. Fertil Steril 2008;89:759. [19] Dalton VK, Saunders NA, Harris LH, Williams JA, Lebovic DI. Intrauterine adhesions after manual vacuum aspiration for early pregnancy failure. Fertil Steril 2006;85:1821. [20] Czernobilsky B. Endometritis and infertility. Fertil Steril 1978; 30:119. [21] Schenker JG. Etiology of and therapeutic approach to synechia uteri. Eur J Obstet Gynecol Reprod Biol 1996;65:109. [22] Lin N, Li X, Song T, et al. The effect of collagen-binding vascular endothelial growth factor on the remodeling of scarred rat uterus following full-thickness injury. Biomaterials 2012;33:1801. [23] Salamonsen LA. Tissue injury and repair in the female human reproductive tract. Reproduction 2003;125:301. [24] Arima K, Nasu K, Narahara H, Fujisawa K, Matsui N, Miyakawa I. Effects of lipopolysaccharide and cytokines on production of RANTES by cultured human endometrial stromal cells. Mol Hum Reprod 2000;6:246. [25] Takeuchi O, Akira S. Pattern recognition receptors and inflammation. Cell 2010;140:805. [26] Sheldon IM, Roberts MH. Toll-like receptor 4 mediates the response of epithelial and stromal cells to lipopolysaccharide in the endometrium. PLoS One 2010;5:e12906.
303
[27] Robinson EK, Kennison SD, Suliburk JW, Mercer DW. Rat gastric injury after lipopolysaccharide: role of inducible nitric oxide synthase. Surgery 2005;138:523. [28] Adachi Y, Mio T, Striz I, et al. Lipopolysaccharide increases fibronectin production and release from cultured lung fibroblasts partially through proteolytic activity. J Lab Clin Med 1996;127:448. [29] Mu E, Ding R, An X, Li X, Chen S, Ma X. Heparin attenuates lipopolysaccharide-induced acute lung injury by inhibiting nitric oxide synthase and TGF-beta/Smad signaling pathway. Thromb Res 2012;129:479. [30] Schaffert CS, Duryee MJ, Hunter CD, et al. Alcohol metabolites and lipopolysaccharide: roles in the development and/or progression of alcoholic liver disease. World J Gastroenterol 2009;15:1209. [31] Polishuk WZ. Endometrial regeneration and adhesion formation. S Afr Med J 1975;49:440. [32] Liechty KW, Kim HB, Adzick NS, Crombleholme TM. Fetal wound repair results in scar formation in interleukin-10deficient mice in a syngeneic murine model of scarless fetal wound repair. J Pediatr Surg 2000;35:866. [33] Bannerman DD, Goldblum SE. Direct effects of endotoxin on the endothelium: barrier function and injury. Lab Invest 1999;79:1181. [34] Bigsby RM. Control of growth and differentiation of the endometrium the role of tissue interactions. Ann N Y Acad Sci 2002;955:110. [35] Schenker JG, Polishuk WZ. Regeneration of rabbit endometrium after cryosurgery. Obstet Gynecol 1972;40:638. [36] Schenker JG, Polishuk WZ. Regeneration of rabbit endometrium following intrauterine instillation of chemical agents. Gynecol Invest 1973;4:1. [37] Arie Birkenfeld, Shlomo Mor-Joseph, Joseph G. Schenker, Daniel Navot, Izak Levij, Okon E. Regeneration of the normal and pseudopregnant rabbit endometrium after neodymium YAG laser photocoagulation. J Gynecol Surg 1991;7:167. [38] Socolov R, Anton E, Butureanu S, Socolov D. The endoscopic management of uterine synechiae. A clinical study of 78 cases. Chirurgia (Bucur) 2010;105:515. [39] Chen F, Duan H, Zhang Y, Wu YH. Effect and mechanism of formation of intrauterine adhesion at different dose of estrogen. Zhonghua Fu Chan Ke Za Zhi 2010;45:917. [40] Li L, Shi J, Zhang QF, et al. Effect of curettage and copper wire on rabbit endometrium: a novel rabbit model of endometrial mechanical injury. Chin Med J (Engl) 2011;124:1708. [41] Schenker JG, Margalioth EJ. Intrauterine adhesions: an updated appraisal. Fertil Steril 1982;37:593. [42] Sheldon IM, Rycroft AN, Dogan B, et al. Specific strains of Escherichia coli are pathogenic for the endometrium of cattle and cause pelvic inflammatory disease in cattle and mice. PLoS One 2010;5:e9192.
Available online at www.sciencedirect.com
journal homepage: www.JournalofSurgicalResearch.com
Creation of a female rabbit model for intrauterine adhesions using mechanical and infectious injury Fang Liu, PhD (ABD),a,b Zhi-Jun Zhu, MM (ABD),c Peng Li, PhD (ABD),d and Yuan-Li He, MMa,* a
Department of Obstetrics and Gynecology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China Department of Obstetrics and Gynecology, Shihezi University School of Medicine, Shihezi, Xinjiang, China c Department of Cardiothoracic Surgery, Shihezi University School of Medicine, Shihezi, Xinjiang, China d Department of Neurosurgery, Key Laboratory on Brain Function Repair and Regeneration of Guangdong, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China b
article info
abstract
Article history:
Background: Intrauterine adhesions (IUA) are associated with secondary amenorrhea,
Received 29 June 2012
infertility, and recurrent pregnancy loss. An IUA animal model would contribute to research
Received in revised form
on the pathogenesis and pathologic changes of IUA and the exploration of new treatments.
25 October 2012
Materials and methods: Eighty female rabbits were randomly divided into five groups:
Accepted 8 November 2012
mechanical injury (16), infectious injury (16), dual injury (16), experimental control (16), and
Available online 24 November 2012
normal (16). Three methods were applied to establish the model: uterine curettage, uterine cavity left alone, lipopolysaccharide surgical suture in place for 48 h, and suture retention
Keywords:
for 48 h after curettage. A sterile surgical suture was left in the uterine cavity for 48 h in the
Intrauterine adhesions
experimental control group. Histologic changes were monitored at 0, 24, 48, and 72 h and 7,
Endometrium
14, and 28 d after operation.
Injury
Results: The experiments revealed that endometrium injured by simple curettage or
Repair
infection could be repaired. Although endometrial regeneration was severely impaired by
Fibrosis
dual injury, the ratio of the area with endometrial stromal fibrosis to total endometrial area
Animal model
significantly increased (P < 0.01), and the number of endometrial glands was significantly
Rabbit
reduced (P < 0.01). Conclusions: The method of dual injury can establish a stable rabbit IUA model. ª 2013 Elsevier Inc. All rights reserved.
1.
Introduction
Intrauterine adhesions (IUA) refer to the partial adherence of endometrial surfaces with fibrotic tissue. Presenting symptoms are related to the degree and location of IUA and include menstrual abnormalities ranging from irregular bleeding and hypomenorrhea to amenorrhea, infertility, and recurrent pregnancy loss [1]. In 1948, Asherman [2e4] published a series of articles describing the frequency, etiology, symptoms, and
roentgenologic picture of this condition, and Asherman syndrome has been used to describe the disease ever since. With the popularization of hysteroscopy and surgery, the incidence of IUA has increased and has become the second most common of female secondary infertility [5]. There are four aspects that relate to IUA: trauma [6], infection [7,8], endometrial repair disorders [9], and low estrogen levels [10]. Previous studies have focused on new treatments for preventing endometrial fibrosis and improving endometrial
* Corresponding author. Department of Obstetrics and Gynecology, Zhujiang Hospital, No. 253 Gongye Road, Guangzhou, Guangdong 510282, China. Tel.: þ86 133 188 07468; fax: þ86 020 616 43361. E-mail address: [email protected] (Y.-L. He). 0022-4804/$ e see front matter ª 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2012.11.009
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
regeneration [11e13]; however, animal models to research the mechanism and curative effects of these new treatments are still scarce. It is necessary to establish an animal model of IUA to intensively research the pathogenesis and pathologic changes of IUA and to explore new treatments. Lipopolysaccharide (LPS) is the major structural component of the cell wall of gram-negative bacteria and is also the endotoxin responsible for much of the inflammation and tissue injury associated with bacterial infections [14]. LPSinduced animal models highlight ways to explore mechanisms of multiple diseases and provide useful information on the discovery of novel biomarkers and treatment strategies [15e17]. However, reports of the effects of LPS on endometrium are rare. The present study aimed to establish an animal model of IUA in the New Zealand white rabbits using mechanical damage, infection, and double damage as is similar to the two main factors in the pathogenesis of IUA: trauma and infection. In this study, curettage was used as a method of mechanical damage and LPS as a predisposing factor for infection. Through this study, we hope to provide an ideal animal model for the study of the exact pathogenesis of and therapy for this difficult clinical condition.
four rabbits were killed at 7 d. Uterine tissue samples were collected.
2.3.1.
Materials and methods
2.1.
Animals
Experimental protocols were approved by the Institutional Animal Ethics Committee. Adult female New Zealand white rabbits (Guangzhou, Guangdong, China) weighing 2500e3500 g were fed standard chow and water ad libitum under standardized laboratory conditions in a temperature-controlled room and light conditions (12 h light and 12 h dark) for 2 wk.
2.2.
2.3.2.
Infectious injury group (n ¼ 32 uterine horns)
An LPS surgical suture was inserted in the uterine cavity through a uterine incision identical to that of the mechanical injury group. The tail end of the LPS surgical suture was placed through the abdominal wall, and a muscular layer was retained on the surface of the skin to facilitate removal after 48 h. Two rabbits (n ¼ 4 uterine horns) were killed for the collection of uterine tissue once the LPS surgical suture was removed.
Dual injury group (n ¼ 32 uterine horns)
After uterine curettage, performed in an identical manner to that of the mechanical injury group, an LPS surgical suture was placed in the uterine cavity (Fig. 1B). The LPS surgical suture was removed at 48 h after surgery, and two rabbits (n ¼ 4 uterine horns) were killed for the collection of uterine tissue.
2.3.4.
Experimental control group (n ¼ 32 uterine horns)
A sterile surgical suture was left in the uterus for 48 h as an experimental control group. Similarly, two rabbits (n ¼ 4 uterine horns) were killed when the surgical suture was removed.
LPS surgical suture preparation 2.3.5.
We obtained LPS from the manufacturer (derived from Escherichia coli 055:B5; Sigma, St. Louis, MO). A 6 mg/L solution of LPS in normal saline was stored in a 4 C refrigerator. The 10-0 medical sterile surgical sutures (10 cm) were soaked in an LPS saline solution for 24 h the day before use in the model.
2.3.
Mechanical injury group (n ¼ 32 uterine horns)
The rabbits were operated on through a 0.5-cm longitudinal incision in the lowest one-third of the connection between the middle and distal uterus and then the endometrial lining of the middle and upper two-third of the uterus was scraped using a 4-mm endometrial curette. Curettage ceased when the uterine cavity stopped feeling uneven and the uterine wall became rough. The uterine tissue from two rabbits (n ¼ 4 uterine horns) was collected immediately after curettage, and the abdominal incisions of the other rabbits were sutured after flushing the uterine and peritoneal cavities.
2.3.3.
2.
297
Method for establishing animal models
One hundred sixty uterine horns from 80 female New Zealand white rabbits were randomly divided into five groups: normal, experimental control, mechanical injury, infectious injury, and dual injury (mechanical and infection). Twenty-four hours before the creation of the model, the rabbits were injected with human chorionic gonadotropin (50 IU) by ear vein to reach synchronization of endometrial development. All the rabbits were anesthetized with urethane (1.5 g/kg) administered intravenously through the marginal ear vein and supplemented if necessary. Throughout the surgery, particular care was taken to confirm an adequate level of anesthesia such that noxious forelimb pinching evoked no response. Two rabbits (n ¼ 4 uterine horns) were killed in each group at 0, 24, 48, and 72 h and 14 and 28 d after surgery, and
Normal group (n ¼ 32 uterine horns)
In the normal group, rabbit abdomens were opened to expose the bilateral uteri but were not given any treatment.
2.4.
Histologic examination
The uterine tissues were fixed in 4% buffered formaldehyde, embedded in paraffin, and routinely stained with hematoxylineeosin and Masson stains. Histologic evaluation was performed by an experienced pathologist. On each hematoxylin- and eosin-stained slice, four high-power fields were selected, the number of glands per high-power field was counted, and the mean was calculated. On each Masson-stained slice, four high-power fields were selected, and the degree of endometrial fibrosis was calculated as the ratio between the area of endometrial stromal fibrosis and endometrial area per high-power field. The average of four fields was calculated.
2.5.
Statistical analysis
Data were expressed as the mean and standard errors of the mean. Statistical analysis was performed using SPSS for
298
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Fig. 1 e (A) Normal rabbit uterine appearance is didelphic and tubular in structure and 7 cm long and 3e4 mm wide on average. (B) Rabbit uterus damaged by the dual injury method; bilateral uterine cavities have a retained LPS surgical suture. (C) Rabbit uterus damaged by the dual injury method after 7 d (bar [ 1 cm). (Color version of figure is available online.)
Windows 13.0 (SPSS, Chicago, IL). One-way analysis of variance was used for comparison between two groups, and one-way analyses of variance were followed by StudenteNewmaneKeuls tests for comparisons between the four groups. A P value of < 0.05 was considered to denote a significant difference.
3.
Results
3.1.
Endometrial morphology in rabbit
3.1.1.
Normal endometrium
The uterine surface observed with the naked eye was gray and scraggly. The shape of the uterine cavity was irregularly star shaped and strongly extensible, and the endometrium was composed of four to six polypoid phymata with the surface endometrium covered with columnar epithelial cells (Figs. 1A and 2A). Round or oval endometrial glands were mainly located in the submucosa and basal layer; moreover, gland openings in the submucosa were at the endometrial surface (Fig. 3A). After Masson staining, the mucosa, submucosa, blood vessels, and muscle were dark red, but endometrial stromal fibers appeared blue (Fig. 3F).
3.1.2.
Endometrium by mechanical injury
Immediately after the lesions were produced, almost all the surface epithelium was denuded; scattered basal glands remained only in some deeper stromal fold regions. The stroma was bare and thinned; even the full thickness of the endometrium disappeared in the severely injured areas. Extensive recent hemorrhage and neutrophil infiltration were evident in the stroma. The capillaries were ruptured and had bled. Epithelial debris and red blood cells were visible in the uterine cavity. Twenty-four hours after the injury, part of the uterine cavity surface (20%e30%) was covered with a few flat regenerative epithelial cells. The stroma was hemorrhagic, edematous, and had lymphocytic infiltration. The capillaries had bled. Forty-eight hours after the injury, most uterine cavity surface (70%e80%) had regenerating epithelial cells, but there was no evidence of gland regeneration. Hemorrhage, edema, leukocyte infiltration, and coagulation necrosis were observed in the stroma and a fibrinous exudate in the bare stromal area.
Seventy-two hours after curettage, 90% of the regenerated surface epithelial cells were cuboidal or low columnar, and mitoses were occasionally observed. There were a few round glands beneath the regenerating epithelium. Except for areas of hemorrhage, edema, and lymphocyte infiltration, fibrous tissue proliferation was present in the stroma. Seven days after the injury, the surface was completely lined by epithelial cells, and no areas of denudation persisted. The regenerating cells were flattened (60%e70%), cuboidal, or low columnar cells. Occasional mitotic figures were observed in the taller cuboidal cells. Beneath the regenerating epithelium, only a few glands could be distinguished. The stroma prominently showed more fibrous tissue, collagen aggregation, focal hemorrhages, congestion, and leukocytic infiltration. A few newborn capillaries were visible (Figs. 2B and 3G). After 14 d, the surface was completely lined by epithelial cells and glands. Obvious mitosis was identifiable in the epithelial cells. The stroma showed a few polymorphonuclear leukocytes and infiltration of hemosiderinladen macrophages. Part of the area exhibited interstitial fibrosis and collagen aggregation. More newborn capillaries were visible. After 28 d, the histologic reconstruction of endometrium was largely completed, other than fibroses that could be observed in a minority of the stromal area. The stroma was neither congested nor infiltrated with leukocytes.
3.1.3.
Endometrium by infectious injury
Immediately after the LPS surgical suture was removed, 30%e40% of endometrial cells were shed. The stroma was exposed at the area of epithelial shedding and was accompanied by extensive hemorrhage, edema, neutrophil infiltration, telangiectasias, and congestion. After 24 h, 30% of the surface was lined with regenerating endometrial cells, which were flat with deeply stained large nucleoli, basophilic cytoplasm, and indistinct cell borders. Endometrial epithelial cells were completely regenerated, and all but 5% of the surface was uncovered, with endometrial, epithelial, and exposed stroma apparent. There were no glands visible underneath regenerating epithelial cells. The edema and neutrophil infiltration were more obvious in the stroma. After 48 h, the entire surface was covered with endometrial cells, and some of them (30%e40%) were low columnar or cuboidal but without glands beneath them. Other areas had
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
299
Fig. 2 e The profile of a rabbit uterus. (A) Normal uterine cavity with many polypoid phymata. (B) The endometrial surface was noted to have petechial hemorrhage by mechanical injury after 7 d. (C) Rabbit uterus damaged by the infectious method after 7 d. (D) The uterine cavity distinctly formed adhesions 7 d after dual injury (bar [ 0.5 cm). (Color version of figure is available online.)
normal endometrial morphology. The edema and neutrophil infiltration were still visible in the stroma. After 72 h, most endometrial cells were cuboidal or columnar epithelial cells, and a portion of the subepithelium was observed to have a few round regenerated glands. The edema and neutrophil infiltration were less prominent in the stroma. After 7 d, the surface was lined by columnar epithelial cells and, occasionally, high columnar epithelial cells. Many glands beneath the epithelium and their openings were visible at the surface. However, the stroma still featured edema and neutrophil infiltration (Figs. 2C and 3H). After 14 d, the epithelium and glands were completely regenerated. Distinct mitoses were recognizable in the epithelial cells. A few polymorphonuclear leukocytes and hemosiderin-laden macrophages were detected in the stroma. After 28 d, the histomorphology of the endometrium was completely normal. There was no congestion or lymphocytic infiltration in the stroma.
3.1.4.
Endometrium after dual injury
Immediately after dual injury, the endometrial epithelium was completely shed, except for an occasionally scattered gland. There was a large area of hemorrhage and neutrophilic infiltration in the naked stroma. Ruptured and hemorrhaging capillaries were prominent (Fig. 3B).
Twenty-four hours after dual injury, regeneration of the epithelial cells was barely visible, and there were marked fibrinous exudates in the stroma. Diffuse hemorrhage, edema, and dense lymphocytic infiltration were observed in the stroma. Forty-eight hours after the creation of the animal model, the regeneration of the flat epithelial cells was rarely observed on the surface, and a great quantity of fibrinous exudates and necrotic tissue debris emerged in the uterine cavity. Part of the stroma began to show fibrotic changes. Some capillary lumens were atretic and surrounded by fibrous tissue (Fig. 3C). Seventy-two hours after the creation of the model, a section of uterine cavity was atretic, and fibrous adhesions appeared in the surface. Five percent of the area without adhesions was observed to have sparse regenerated flat epithelial cells, and no gland regeneration was evident. The proliferation of stromal fibrosis was evident and accompanied by edema, lymphocytic infiltration, and capillary expansion (Fig. 3D). Seven days after the injury, the uterus was vertically cut, and on the axial view, some highly fibrous adhesions occurred in the uterine cavity. Microscopically, the fibrous adhesions thickened, and 70%e80% of the area without adhesions was lined by flattened or low columnar epithelial cells, and scarce rounded glands were present beneath the epithelial layer (Figs. 1C, 2D, and 3E).
300
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Fig. 3 e (AeE) Hematoxylineeosin sections of rabbit uterine tissue. (A) The normal rabbit endometrium protrudes into the uterine cavity, and many glands are located in the stroma (magnification, 3100). (B) Immediately after dual injury, the endometrial epithelium was completely shed, and a large area of hemorrhage was observed in the naked stroma (magnification, 3100). (C) Forty-eight hours after the dual injury, the uterine cavity showed distinct fibrinous exudates (magnification, 3100). (D) Seventy-two hours after dual injury, the uterine cavity appeared to have adhesions (magnification, 3100). (E) Seven days after dual injury, extensive fibrosis in the uterine cavity and rare glands in the stroma were observed (magnification, 3100). (FeI) Masson sections of rabbit uterine tissue. (F) High-power photomicrograph shows that endometrium, glands, and blood vessels are dyed dark red, but endometrial stromal fibers are blue (magnification, 3200). (G) Seven days after curettage, high-power photomicrographs showed that regenerated endometrium and glands were present, blood vessels were congested, and fibrous tissue of the stroma was increased (magnification, 3200). (H) Seven days after infectious injury, many regenerated glands are observed at the surface, and the blood vessels are congested (magnification, 3200). (I) Seven days after dual injury, a low-power photomicrograph shows interstitial fibrosis and no glands in the stroma (magnification, 3200) (bar [ 200 mm). (Color version of figure is available online.)
Fourteen days after the injury, histology revealed nonfunctional flattened epithelial cells covering the areas without adhesions, but the level of IUA resembled that seen after 7 d. Twenty-eight days after the injury, the endothelial morphology had failed to reconstruct and had formed a scar.
3.1.5.
Endometrium in the experimental control group
Less than 5% of the endometrial luminal epithelial cells were shed after the sterile suture was removed, and scattered red blood cells were visible in the uterine cavity. The endometrium in the remaining regions had the same histomorphology as the normal endometrium. After 24 h, the entire uterine surface was covered with endometrial epithelial cells. After 48 h, the
histomorphology of the endometrium returned to normal. Normal endometrium was also observed at 72 h, 7 d, 14 d, and 28 d.
3.2.
The degree of endometrial fibrosis in rabbits
The ratios of the area with endometrial fibrosis to total endometrial area were calculated before injury and 7 d after injury (Table 1), and the dual injury group had the highest ratio (P < 0.01) (Fig. 4A). The number of endometrial glands per high-power field was also counted, and the dual injury group had the fewest of the four groups (P < 0.01), whereas the mechanical injury group had fewer than the normal group (P ¼ 0.03) (Fig. 4B and Table 1).
301
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
Table 1 e The percentage of fibrotic area and the number of glands in each group.
Fibrosis area (%) Number of glands (/HPF)
Normal
Curettage
LPS
Curettage þ LPS
Control
27.50 3.34 14.97 0.59
40.52 3.72 12.66 0.52y
37.80 3.79 13.59 0.55
68.03 4.10* 4.19 0.48z
28.33 3.58 15.00 0.44
HPF ¼ high-power field. Results are shown as the mean standard deviation. * P < 0.05 curettage þ LPS versus normal, curettage, and LPS. y P < 0.05 curettage versus normal. z P < 0.05 curettage þ LPS versus normal, curettage, and LPS.
4.
Discussion
A general consensus has been reached that the most common causes of IUA are trauma and infection [18]. Any trauma that destroys the endometrial basal layer could result in IUA, including abortion, curettage, and hysteroscopic surgery, among others [19]. After the endometrial basal layer is injured, the endometrium failed to regenerate and repair, and the uterine cavity appeared to form scars and adhesions. Studies have currently found no “consistent evidence” of the inflammatory processes causing IUA; nevertheless, some investigators [20] believe that inflammatory processes do contribute to the damaging effect of trauma and act synergistically to produce IUA [21]. At present, clinical treatment for IUA can restore the shape of the uterine cavity, but it remains difficult to repair uterine physiological function. Therefore, the rates of cure and pregnancy are low, particularly in moderate and severe IUA. Some possible treatments of IUA include direct delivery of stem cells or specific growth factors into the uterus [11,22]. Thus, it is essential to establish a new model of IUA to evaluate novel treatment options. In the present study, we designed three methods of creating animal models of IUA and have demonstrated that dual injury with curettage and LPS is a suitable method to establish a model. In this study, all layers of the rabbit endometrium were injured by undergoing curettage, and the endometrial cells began to regenerate in a short time. Although the repair process of the stroma showed was slow, the morphology of the endometrium was almost entirely repaired after 28 d. This
finding illustrates that the process of endometrial regeneration was unaffected by curettage injury [23]. LPS has many deleterious effects and plays a significant role in a number of disease processes by increasing inflammatory cytokine release [24,25]. In this study, we used LPS as an inflammation-inducing factor, and in consideration of the special anatomical structure of the rabbit uterus, we also used a surgical suture as a carrier of LPS to ensure the persistence of LPS in the uterine cavity. We demonstrated that the method of infectious injury could also ablate part of the surface, but the damaged endometrium could be completely rebuilt. Seven days after the injury, there were no significant differences between the infected injury and normal groups regarding the number of glands and the ratio of fibrotic tissue area to normal tissue area. Endometrial stromal fibrosis temporarily increased, but there were no fibrous scars or adhesions on the surface covered with epithelial cells. Present studies suggest that binding of LPS to Toll-like receptor 4 activates cellsignaling pathways leading to an inflammatory response and secretion of the cytokines interleukin-1b, interleukin-6, and tumor necrosis factor alpha among others. In addition, infected tissue and organs appear to undergo various pathologic changes, including edema, hemorrhage, fibrosis, and necrosis [26e29]. Although our studies did not examine these inflammatory cytokines, the pathologic changes of rabbit endometrium treated with LPS were compatible with other models induced by LPS. In our studies, the endometrium treated with LPS surgical suture alone could be fully repaired, perhaps because of a lower amount of inflammation induced by low-dose LPS [30]. The results of the study were in line with
Fig. 4 e (A) The value represents the area of stromal fibrosis quantified as a percentage of the total endometrial area in the visual field (magnification, 3200). *P < 0.05 curettage D LPS versus normal, curettage, and LPS. (B) We counted the number of glands in the visual field (magnification, 3200). *P < 0.05 curettage versus normal and **P < 0.05 curettage D LPS versus normal, curettage, and LPS. (Color version of figure is available online.)
302
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
the work by Polishuk [31] documenting 171 women who had undergone cesarean sections. Among these 171 cases, 28 patients developed significant endometritis. Afterward, hysterosalpingograms showed no difference in the occurrence of IUA between the endometritis and control groups. The authors concluded that endometritis is unlikely to be a major factor in the pathogenesis of IUA. An experimental control group was established to exclude the interference of surgical suture. The endometrium treated with surgical suture alone was only observed to have slight epithelial shedding and scattered red blood cells in the uterine cavity. These findings suggested that LPS was the main inflammatory inducer, not the surgical suture. There was no endometrial regeneration 72 h after endometrium was removed by the dual injury method, but then, fibrinous exudates obstructed the uterine cavity and formed adhesions. Finally, the endometrium was repaired through scar tissue. This finding suggests that the endometrium could achieve a perfect nonscarring self-repair with rapid epithelialization and a reduced inflammatory reaction [32]. Curettage to the endometrium destroyed the endometrial barrier, and the inflammatory response induced by LPS caused edema, inflammatory cell infiltration, epithelial cell necrosis, and stromal fibrosis hyperplasia [33]. The double hit of curettage and LPS interfered with the normal repair process after endometrial injury. It may be that the shedding of epithelial cells and the rupture of capillaries after curettage promoted the diffusion of the inflammatory response induced by LPS [30]. The delay in regeneration of endometrial epithelial cells may be important to the formation of IUA after endometrial damage [34]. Because of the deficiency of sufficient epithelial cells covering the uterine cavity, the stroma is bared, allowing fibroblast activity and connective tissue formation leading to endometrial fibrosis and possibly to IUA [21]. Since the 1970s, a succession of researchers have established the model of endometrial injury in the rabbit using various methods. Schenker and Polishuk [35] introduced the model of endometrial hypothermic impairment in the rabbit by applying a probe containing Freon and a model of endometrial chemical damage using an injection of 10% formalin into the uterine cavity [36]; Birkenfeld et al. [37] established endometrial damage models by Nd:YAG laser that retarded the regeneration of endometrium in the rabbit. Polishuk [31] transplanted polyethylene sponges carrying autologous fibroblasts in the injured uterine cavity of the rabbit or rat. Interestingly, the purpose of researchers developing a model of endometrial damage or adhesions is to explore the treatment of perimenopausal dysfunctional uterine bleeding, surgical contraindications in patients with hysteromyoma, or even as a contraceptive method rather than for research on IUA. These methods result in varying degrees of injury to the endometrium, including adhesion formation. However, these animal models mainly apply some physical or chemical lesion and are different from the pathogenesis of IUA. Therefore, we cannot confirm that the pathologic changes caused by these methods are the same as those that occur in vivo. In addition, chemical or physical materials retained in the uterine cavity may have affected the subsequent evaluations and analyses. IUA is the result of many factors [38]. It is difficult to establish an IUA model using only one method of damage
because of the strong regenerative capacity of endometrium, so the model may be easier to construct by combining two types of damage methods. To study the pathogenesis of IUA in the postpartum uterus after a uterine cavity operation, Chen et al. [39] established an IUA model in rabbit through curettage and bilateral oophorectomy. Li et al. [40] placed a copper wire in the uterine cavity after curettage and concluded that endometrial injury impeded blastocyst implantation in rabbit. In this study, we constructed a rabbit IUA model by means of curettage and infection to simulate the two main causes of IUAdtrauma and infection. The endometrial layer was completely ablated, and the LPS surgical suture was placed in the uterine cavity immediately after curettage and retained for 48 h. This method inhibits the regeneration of endometrium [41], and LPS contributes to scar repair by increasing the inflammatory reaction [42]. Timely removal of the LPS surgical suture from the uterine cavity has no influence on the formation of intimal fibrosis and adhesions. In summary, a stable IUA model can be established using the dual injury method.
5.
Conclusion
The combination of curettage and infection creates an IUA model in the rabbit that resembles the condition in humans, taking into account the features such as histologic changes in the uterine cavity. This rabbit model provides an opportunity to perform studies in a standardized animal model to explore new treatments and examine their effectiveness and safety.
Acknowledgment This study was supported by the National Nature Science Foundation of China (81270658 to Y.L.H.).
references
[1] Panayotidis C, Weyers S, Bosteels J, van Herendael B. Intrauterine adhesions (IUA): has there been progress in understanding and treatment over the last 20 years? Gynecol Surg 2009;6:197. [2] Asherman JG. Traumatic intra-uterine adhesions. J Obstet Gynaecol Br Emp 1950;57:892. [3] Asherman JG. Inversio uteri puerperalis. Gynaecologia 1954; 138:410. [4] Asherman JG. Uterine adhesions. Harefuah 1958;55:138. [5] Roge P, Cravello L, D’Ercole C, Brousse M, Boubli L, Blanc B. Intrauterine adhesions and fertility: results of hysteroscopic treatment. Gynaecol Endosc 1997;6:225. [6] Taylor PJ, Cumming DC, Hill PJ. Significance of intrauterine adhesions detected hysteroscopically in eumenorrheic infertile women and role of antecedent curettage in their formation. Am J Obstet Gynecol 1981;139:239. [7] Gupta N, Sharma JB, Mittal S, Singh N, Misra R, Kukreja M. Genital tuberculosis in Indian infertility patients. Int J Gynaecol Obstet 2007;97:135. [8] Al-Inany H. Intrauterine adhesions. An update. Acta Obstet Gynecol Scand 2001;80:986.
j o u r n a l o f s u r g i c a l r e s e a r c h 1 8 3 ( 2 0 1 3 ) 2 9 6 e3 0 3
[9] Reyniak JV, Gordon M, Stone ML, Sedlis A. Endometrial regeneration after voluntary abortion. Obstet Gynecol 1975; 45:203. [10] March CM. Intrauterine adhesions. Obstet Gynecol Clin North Am 1995;22:491. [11] Nagori CB, Panchal SY, Patel H. Endometrial regeneration using autologous adult stem cells followed by conception by in vitro fertilization in a patient of severe Asherman’s syndrome. J Hum Reprod Sci 2011;4:43. [12] Chason RJ, Levens ED, Yauger BJ, Payson MD, Cho K, Larsen FW. Balloon fluoroscopy as treatment for intrauterine adhesions: a novel approach. Fertil Steril 2008;90:2005. [13] Amer MI, Abd-El-Maeboud KH. Amnion graft following hysteroscopic lysis of intrauterine adhesions. J Obstet Gynaecol Res 2006;32:559. [14] Wang X, Quinn PJ. Lipopolysaccharide: biosynthetic pathway and structure modification. Prog Lipid Res 2010;49:97. [15] Hasegawa A, Iwasaka H, Hagiwara S, Noguchi T. Relationship between HMGB1 and tissue protective effects of HSP72 in a LPS-induced systemic inflammation model. J Surg Res 2011; 169:85. [16] Fujioka H, Akema T. Lipopolysaccharide acutely inhibits proliferation of neural precursor cells in the dentate gyrus in adult rats. Brain Res 2010;1352:35. [17] Saadani-Makki F, Kannan S, Lu X. Intrauterine administration of endotoxin leads to motor deficits in a rabbit model: a link between prenatal infection and cerebral palsy. Am J Obstet Gynecol 2008;199:651. [18] Yu D, Wong YM, Cheong Y, Xia E, Li TC. Asherman syndromedone century later. Fertil Steril 2008;89:759. [19] Dalton VK, Saunders NA, Harris LH, Williams JA, Lebovic DI. Intrauterine adhesions after manual vacuum aspiration for early pregnancy failure. Fertil Steril 2006;85:1821. [20] Czernobilsky B. Endometritis and infertility. Fertil Steril 1978; 30:119. [21] Schenker JG. Etiology of and therapeutic approach to synechia uteri. Eur J Obstet Gynecol Reprod Biol 1996;65:109. [22] Lin N, Li X, Song T, et al. The effect of collagen-binding vascular endothelial growth factor on the remodeling of scarred rat uterus following full-thickness injury. Biomaterials 2012;33:1801. [23] Salamonsen LA. Tissue injury and repair in the female human reproductive tract. Reproduction 2003;125:301. [24] Arima K, Nasu K, Narahara H, Fujisawa K, Matsui N, Miyakawa I. Effects of lipopolysaccharide and cytokines on production of RANTES by cultured human endometrial stromal cells. Mol Hum Reprod 2000;6:246. [25] Takeuchi O, Akira S. Pattern recognition receptors and inflammation. Cell 2010;140:805. [26] Sheldon IM, Roberts MH. Toll-like receptor 4 mediates the response of epithelial and stromal cells to lipopolysaccharide in the endometrium. PLoS One 2010;5:e12906.
303
[27] Robinson EK, Kennison SD, Suliburk JW, Mercer DW. Rat gastric injury after lipopolysaccharide: role of inducible nitric oxide synthase. Surgery 2005;138:523. [28] Adachi Y, Mio T, Striz I, et al. Lipopolysaccharide increases fibronectin production and release from cultured lung fibroblasts partially through proteolytic activity. J Lab Clin Med 1996;127:448. [29] Mu E, Ding R, An X, Li X, Chen S, Ma X. Heparin attenuates lipopolysaccharide-induced acute lung injury by inhibiting nitric oxide synthase and TGF-beta/Smad signaling pathway. Thromb Res 2012;129:479. [30] Schaffert CS, Duryee MJ, Hunter CD, et al. Alcohol metabolites and lipopolysaccharide: roles in the development and/or progression of alcoholic liver disease. World J Gastroenterol 2009;15:1209. [31] Polishuk WZ. Endometrial regeneration and adhesion formation. S Afr Med J 1975;49:440. [32] Liechty KW, Kim HB, Adzick NS, Crombleholme TM. Fetal wound repair results in scar formation in interleukin-10deficient mice in a syngeneic murine model of scarless fetal wound repair. J Pediatr Surg 2000;35:866. [33] Bannerman DD, Goldblum SE. Direct effects of endotoxin on the endothelium: barrier function and injury. Lab Invest 1999;79:1181. [34] Bigsby RM. Control of growth and differentiation of the endometrium the role of tissue interactions. Ann N Y Acad Sci 2002;955:110. [35] Schenker JG, Polishuk WZ. Regeneration of rabbit endometrium after cryosurgery. Obstet Gynecol 1972;40:638. [36] Schenker JG, Polishuk WZ. Regeneration of rabbit endometrium following intrauterine instillation of chemical agents. Gynecol Invest 1973;4:1. [37] Arie Birkenfeld, Shlomo Mor-Joseph, Joseph G. Schenker, Daniel Navot, Izak Levij, Okon E. Regeneration of the normal and pseudopregnant rabbit endometrium after neodymium YAG laser photocoagulation. J Gynecol Surg 1991;7:167. [38] Socolov R, Anton E, Butureanu S, Socolov D. The endoscopic management of uterine synechiae. A clinical study of 78 cases. Chirurgia (Bucur) 2010;105:515. [39] Chen F, Duan H, Zhang Y, Wu YH. Effect and mechanism of formation of intrauterine adhesion at different dose of estrogen. Zhonghua Fu Chan Ke Za Zhi 2010;45:917. [40] Li L, Shi J, Zhang QF, et al. Effect of curettage and copper wire on rabbit endometrium: a novel rabbit model of endometrial mechanical injury. Chin Med J (Engl) 2011;124:1708. [41] Schenker JG, Margalioth EJ. Intrauterine adhesions: an updated appraisal. Fertil Steril 1982;37:593. [42] Sheldon IM, Rycroft AN, Dogan B, et al. Specific strains of Escherichia coli are pathogenic for the endometrium of cattle and cause pelvic inflammatory disease in cattle and mice. PLoS One 2010;5:e9192.