Fixation of biomaterial to metallic stent and fixation of stents after circular endoscopic dissection in the esophagus on an animal model

Summary

Introduction:
Complete circular endoscopic (submucosal) resection (CER) performed using the endoscopic submucosal dissection (ESD) technique is burdened with a high incidence of post-operative strictures in the esophagus. The most effective method of preventing them is not known so far; one of the possible methods is to prevent these strictures directly at the site of their formation by covering the defect with a stent. The aim of the study was to find a way to fix a selected biomaterial to a stent, and subsequently, to fix the stent at the CER site to prevent esophageal strictures in an animal model.

Method:
Miniature piglets from the Czech Academy of Sciences’ breeding unit in Libechov (N=10) were used. Endoscopy was performed using a single-channel endoscope. First, we made two circular mucosal cuts spaced 5 cm apart in the middle esophagus and we performed the CER between them using the endoscopic submucosal dissection technique. After that, we covered the defect with a stent coated with biomaterial (Xe-Derma^®) while we tried to prevent stent migration into the stomach. For stent fixation, we tested endo-clips (N=3), the Apollo endoscopic system (N=1) and the suspension technique using two polyamide threads (N=6) anchored in the nasal septum. We performed a control endoscopy, stent removal and subsequent autopsy after 1−2 weeks.

Results:
All procedures were completed successfully without serious complications or deaths. Although stents covered with Xe-Derma^® were applied to the entire resection area, one case of mediastinitis and one paraesophageal abscess were found during autopsy, most likely due to microperforations caused during the procedure. Histological analysis showed that after contact with the biomaterial, re-epithelisation took place within one week of application to the resection area. Stent migration occurred in each case when the stent was attached to the esophageal wall by endo-clips or with the endoscopic suture system (Apollo). There was no stent dislocation in the cases where the stent was suspended by two polyamide threads.

Conclusion:
We developed a technique of fixing biomaterial to the surface of metallic stents which we used to prevent the formation of esophageal strictures after CER. Distal suspension fixation of the stent with a polyamide thread proved to be the most effective, while fixations by endo-clips or with the endoscopic suture system were ineffective.

Key words:
benign esophageal strictures − circular endoscopic resection − endoscopic submucosal dissection − complication − prevention

INTRODUCTION

Endoscopic en-bloc (R0) resection of superficial distal esophageal tumours provides a great therapeutic potential and may replace surgical procedures in the future. Its benefit consists in mini-invasive approach while preserving the therapeutic scope of the resection. The repertory of advanced endoscopic techniques, originally based on the NOTES concept, includes endoscopic submucosal dissection (ESD) that even supports complete circular endoscopic resection (CER) of esophageal mucosa for circular superficial lesions. Both techniques are recognized by the European Society of Gastrointestinal Endoscopy, although they are considered as technically very demanding. Therefore their implementation in the clinical practice is gradual and takes place in stages to ensure safety (training structure according to Gotoda and Deprez with determination of the order of operated segments of the gastrointestinal tract [1]; gaining skills on a simulator (EASIE – Erlangen Active Simulator for Interventional Endoscopy), in ex-vivo models [2,3] and in live pigs [4]; determination of the training duration and of the minimum number of ESD procedures under expert supervision and of the number of ESD procedures by a single endoscopist annually).

Postoperative esophageal strictures (esophageal narrowing to <1 cm with impossibility of passing the endoscope, Fig. 1) are a problem of successful (R0) CER using ESD. No ideal way of preventing the strictures is known; the used methods include systemic corticotherapy or covering the defect with a metallic stent (self-expandable metallic stent – SEMS) [5]. Esophageal strictures after CER without adequate prevention form in 100% cases if more than 90% of mucosal circumference is removed. Their formation thus poses an important problem in the clinical practice, often approached by repeated dilations of the refractory stricture. Most endoscopically removable lesions occupy less than 1/3 of the circumference where strictures are not formed at all or only rarely after endoscopic removal of the lesion [6,7]. However, some endoscopically removable lesions (usually early squamous cell carcinomas) cover the major part of the circumference, and CER is necessary for radical removal of the lesion, associated with formation of esophageal stricture almost with certainty [8]. In our experiment we therefore used a CER model assuring that the stricture would form in all cases, which enabled us to test preventive insertion of a metallic stent; however, the stent usually migrates to the stomach without fixation. The purpose of this experimental study was to develop an ideal fixation method of the stent (coated with biomaterial) after CER as a measure to prevent the formation of esophageal stricture, and additionally, to find the most suitable method of fixing the biomaterial to the stent.

METHOD

Miniature piglets from the Libechov breeding unit of the Czech Academy of Sciences (CAS) (N=10) were used for the experimental model. The piglets were chosen as a suitable model given their considerable similarity to human anatomy, physiology and pathophysiology of the esophagus [9]. The animals were monitored and premedicated with intramuscular application of TKX combinations (containing Tiletamine 4 mg/kg, Zolazepam 4 mg/kg (Zoletil 100, Virbac), Ketamine 5 mg/kg (Narketan 10, Chassot) and Xylazine 1 mg/kg (Rometar 2%, Spofa)) and Atropine 0.2 mg (BB Pharma). After intravenous cannulation and intubation, Isoflurane (1.5%) combined with i.v. Fentanyl (3–5 ml/h) was used for inhalation anaesthesia. The single-channel endoscope (GIF H180J; Olympus, Tokyo, Japan) was used. After the procedure, the piglets received only liquid diet for 10 days. Subsequently, control endoscopy and then also autopsy (depending on how the animals thrived) were done 1–2 weeks from the procedure.

Circular endoscopic resection (CER)

First, coagulation marks were made to indicate the lower and upper margins of the incision (5 cm apart) in the middle esophagus. The lower circular incision was made, followed by the upper circular incision. Subsequently, both incisions were connected by endoscopic submucosal dissection and the tissue lifted en-bloc was removed from the esophagus. Physiological solution with methylene blue (1%) were used for injections under the mucosa, and atmospheric air was used for insufflation. Two types of electrocoagulation knives were used for the dissection in the hybrid mode (30-W, monopolar): dual or IT (KD-650L, KD-611L; Olympus Medical Co, Tokyo, Japan).

Stents

Fully coated, 10 cm long esophageal stents (Wallflex Prox CVD, 25 mm, Boston Scientific, Natick, Mass, USA) were used to cover the created defects. Moistened plates of biodegradable Xe-Derma^® (Medicem, Prague, Czech Republic) were applied to the surface of the stents. Thus prepared stents were applied under endoscopic control over the resection area, making sure that the biological cover would exceed the lesion margins on both sides. However, contact of the defect surface with Xe-Derma^® moistened with physiological solution reduced friction, increasing the tendency of the stent to migrate. The stents were therefore fixed to the esophageal wall using endo-clips (N=3, HX-610- 90L; Olympus Medical Co., Tokyo, Japan), using an endoscopic suture system (N=1, OverStitch; Apollo Endosurgery, Austin, Texas, USA) or using a suspension technique with two polyamide threads (N=6).

RESULTS

Biomaterial fixation to the stent

The Xe-Derma^® plates were sutured with each other using Vicryl (2-0, Ethicon; Conelia, Georgia, USA) to obtain a square (approx. 13x10 cm) and the edges of the square were then sutured with each other to obtain a cylinder, longer than the stent in all cases. The stent was temporarily removed from its sheath and the biomaterial was fixed to its distal (i.e. aboral) part of the mesh (using Vicryl 2-0) on both sides (Fig. 2). Before application of the stent, the biomaterial was bathed in physiological solution to make sure that it was supple and hydrophilic during insertion.

Stent insertion and fixation

The coated stent was inserted over a pre-inserted guide-wire and was unfolded under endoscopic control to make sure that the entire length of the stent with the biomaterial safely covers the defect after CER. Fixation of the upper stent end using endo-clips (5–6 endo-clips for both the upper and lower ends of the stent) and using the Apollo system showed to be inefficient because the stents migrated to the stomach in all cases (Fig. 3). On the contrary, stent suspension using two threads was functional as no stent dislocation was observed during the follow-up period. First, the threads were fixed using surgical suture at the upper (i.e. oral) end of the stent at opposite sides (Fig. 4). After fixing the threads, the stent was carefully inserted into the sheath. With the stent properly unfolded over the defect, the threads were anchored by passing them through the nostrils. When extracorporeal knotting of the threads was used, the piglets often pulled out the threads from the snout in several hours. Therefore we approached transfixation by piercing the nasal septum with a screw (Fig. 5), which was used as an anchoring bar. Both threads were adequately tightened under endoscopic control so as to fix both threads with an adequate pulling force exerted on the suspended stent (Fig. 6).

Complications

All CER procedures were completed successfully without any serious complications. No visible perforations of the esophagus were observed during the procedures, although one suppurative mediastinitis and one paraesophageal abscess were found during the autopsy (Tab. 1). Perioperative bleeding was not a significant complication; it occurred in two cases (2/10) and could be easily managed by electrocoagulation and using injections with adrenaline.

All suspended stent fixations healed without complications (Tab. 1). Insignificant bleeding in the nasal sheath was observed in one case (1/6). No rupture of the polyamide threads was observed.

In all cases where the exposed submucosal tissue was protected with Xe-Derma^® macroscopic healing was smooth, integral, and free of any hypergranulation and ulceration. As shown by histological analysis, re-epithelisation occurs in the resection area in contact with the biomaterial as early as one week from application (Fig. 7). Deeper layers of the defect showed no dysplasia, no disintegration of the layers and no presence of foreign cell granuloma.

Our experiment did not evaluate the meaning of biomaterial-coated stents with respect to prevention of esophageal stricture formation after CER, which will be the subject of a subsequent experimental study.

DISCUSSION

Endoscopic resection and endoscopic submucosal dissection are efficient endoscopic techniques used to treat early neoplasia in the esophagus as well as in other parts of the GIT. When used in the esophagus, these methods are specifically associated with the problem of postoperative stricture formation. This problem is not present provided that less than 75% of the circumference is removed [8]. Some superficial lesions, particularly squamous cell neoplasms, tend to exhibit circumferential involvement; mucosal resection along the entire circumference (in variable lengths) is thus the only option of their endoscopic removal. The advantage of mini-invasive endoscopic curative therapy goes hand in hand with the disadvantage of almost 100% probability of stricture formation, subsequently requiring repeated (and sometimes even risky) endoscopic dilations.

The question of how to prevent the formation of postoperative esophageal strictures is thus important. Uncovered muscle and possibly also submucosal tissues are unable to repair on their own. Strictures form almost in all cases due to scarring, despite administration of systemic corticosteroids in high doses, which are still considered as the most efficient preventive measure (Tab. 2). Other tested prevention methods include the use of metallic stents, implantation of autologous epithelial cells, local application of corticoids, application of other substances or even implantation of another (for example, gastric) mucosa. For example, defect coverage with autologous cells collected from oral mucosa has been tested especially in Japan [10]. However, usually, tissue engineering techniques based on ex-vivo culture of cell plates are highly financially demanding.

So far, none of these methods has achieved universal application. Locally acting, biological degradable covers on metallic stents used as carriers could also provide causal prevention [4]. Xe-Derma^® with 3D matrix of collagenous and elastic fibres namely stimulates the proliferation of keratinocytes (through expression of p63, involucrin and CD29), thereby resulting in high-quality, multi-layered re-epithelisation [11].

To make sure that the stenting method is successful (irrespective of the biomaterial), the stent must remain in situ throughout the healing period and it must not migrate. Clearly, no stenosis exists at the resection site immediately after CER and the stents migrate, similarly as, for example, stents inserted for the treatment of anastomotic leaks. Spontaneous migration of fully coated stents has been reported in 58–71% [4,12]. With the perspective of further experimental testing, three stent fixation methods were evaluated in our study – endo-clips, Apollo endoscopic suture system, and fixation with polyamide threads to a bar between the nostrils of the animal. We also tried to fix the biomaterial (Xe-Derma^®) to the stent in a manner so that the biomaterial could be unfolded along the entire circumference of the esophagus between the resection area and the stent.

Local transfixing of the stents did not prove to be good. When closed, the endo-clips did not provide sufficient fixation to the mucosa (Fig. 3). Based on our subjective experience, the Apollo suture system was not suitable for anchoring the coated stents in the esophagus, either. As follows from our experience, this system requires much more operating space, which is not available in the esophagus. Additonal limitations in the esophagus include insufficient folding of the wall (due to pressure exerted by the stent) and an inadequate (perpendicular and central) direction of the needle with respect to optic equipment. As shown by our single Apollo experience, esophageal fixation was very difficult to achieve, moreover providing a suboptimal perioperative result. As shown by the group of Prof. Swanstrom (ex vivo), the force of 4.8 N (Newtons) is needed to release coated stents without fixation, 6.1 N is needed to release stents anchored with endo-clips, and up to 20.4 N to release stents fixed using the Apollo method [13]. As follows, unlike endo-clips that are almost meaningless for fixation, the suture system actually may provide a certain fixation potential (at least 15.6 N). However, this method is not well usable in the esophagus, and transcutaneous fixation described for fixation of tracheal stents cannot be adapted in the esophagus, either [14]. The high price of the Apollo system (approx. CZK 80,000 per suture) should also be emphasized.

Distance fixation inspired by the Shim “lasso” method [15] can be used safely, is inexpensive, and it effectively prevents sliding of the stent to the stomach. One advantage of using extended threads fixed in the nasal septum is that the animals cannot damage the suspension system; another advantage is that after re-epithelisation, the stent can be pulled out using the threads (preferably under endoscopic control). The threads can even be used to modify the stent position in the oral direction in the course of inserting the stent (Fig. 8).

Stent fixation using an OTSC clip (OVESCO) has also been used as a nontraditional method [16]. Although we have experience with this method within the scope of other NOTES techniques (closures of large rectal and gastric defects [17,18]), we believe that this method is rather unadapted for this indication, i.e. esophageal stent fixation.

Generally, methodological experimental studies with animal models may pose an ethical burden in projects. Therefore we always have followed the Russell’s concept of 3Rs in order to use as few animals as possible. The project was approved by the Animal Protection Committee (authorisation No. 7779/2004–1020) in compliance with applicable legal regulations (No. 246/1992, 207 / 2004). Our set is thus limited by the low number of animals and by the missing control group.

CONCLUSION

This experimental methodological study focused on the prevention of strictures after CER in distal esophagus of animals using stents. To open up the possibility of using the potential of SEMS stents, we developed a technique for fixing the Xe-Derma^® biomaterial to the stents and a suspension technique for fixing the stents at the site of the defect after CER.

Based on our observations we will continue the experiment to determine whether the biomaterial-coated metallic stent can be efficient in the prevention of esophageal stricture development after CER. We shall study for how long the defect should remain treated with the stent, and we shall seek to ascertain the effect of the biomaterial in situ.

Abbreviations

AV CR (CAS) − Czech Academy of Sciences

CD − cluster of differentiation

CER − circular endoscopic resection

EASIE − erlangen active simulator for interventional endoscopy

ESD − endoscopic submucosal dissection

GIT − gastrointestinal tract              

NOTES − natural orifice transluminal endoscopic surgery

OTSC − over-the-scope-clip

SEMS − self-expandable metallic stent

Supported by the program project of the Ministry of Health of the Czech Republic, Reg. No. NV16-27653A, and project No. LO1609, with financial support of the Ministry of Education, Youth and Sports of the Czech Republic in the scope of NPU I (National Programme of Sustainability I), DRO: 67985904 and IP of the Ministry of Defence of the Czech Republic 1012.

Conflict of interest

The authors declare no conflict of interest with respect to this paper. The authors declare that this paper has not been published in any other journal.

MUDr. Radek Doležel

Department of Surgery,

2^nd Faculty of Medicine,

Charles University and Central Military 

  Hospital, Prague

U Vojenské nemocnice 1200

169 02 Praha 6

e-mail: dolezrad@uvn.cz