Ligation-assisted endoscopic enucleation for the diagnosis and resection of small gastrointestinal tumors originating from the muscularis propria: a preliminary study

Some gastrointestinal tumors originating from the muscularis propria, such as gastrointestinal stromal tumors (GISTs), may be nonmalignant when diagnosed but have the potential to undergo malignant transformation. A majority of patients with gastrointestinal lesions originating from the muscularis propria prefer to undergo resection despite controversies over therapeutic decisions. Several endoscopic resection techniques have been proven feasible and safe for tumors originating from the muscularis propria, including endoscopic submucosal dissection [1–4], endoscopic enucleation [5, 6], endoscopic ligation [7–9], endoscopic ligation and resection [10], endoscopic full-thickness resection [11], and submucosal tunneling endoscopic resection [12, 13].

Ligation-assisted endoscopic enucleation (EE-L) was developed by combining endoscopic band ligation and endoscopic enucleation to fully exploit the advantages of each technique. The present study investigated the efficacy and safety of EE-L in the diagnosis and resection of gastrointestinal tumors originating from the muscularis propria.

All 43 gastrointestinal tumors were resected by EE-L. The mean operation time was 27.9 minutes (range, 16–40 minutes), and the mean enucleation time was 7.2 minutes (range, 5–11 minutes). One patient experienced self-limiting, non-life-threatening hemorrhage 5 days after EE-L. No perforation, massive hemorrhage, or peritonitis requiring further endoscopic or surgical intervention occurred. The mean follow-up time was 20.4 months (range, 14–38 months). No recurrence has occurred during the follow-up period.

The histological diagnoses obtained for the 43 lesions were 24 leiomyomas and 19 GISTs. Mitotic counts in all 19 GISTs were <5 per 50 high-power fields; thus, all were classified as very low-risk.

Gastrointestinal tumors of muscularis propria origin include leiomyomas, GISTs, neural tumors, and others. GISTs are the most common mesenchymal tumors of the gastrointestinal tract. Large GISTs with high mitotic rates are often associated with malignant behavior and display higher rates of recurrence and metastasis [14–18]. Therefore, the presence of such lesions can become a psychological burden to patients, even if the lesions are very small. For these reasons, some patients with such lesions prefer to undergo resection, although the management of small, incidentally discovered GISTs of <2 cm is controversial [19, 20]. A histologic diagnosis of these lesions is sometimes needed [19–21]. EUS-guided sampling, deep biopsy, and endoscopic partial resection with the unroofing technique are feasible and effective procedures with which to obtain a definitive pathological diagnosis of subepithelial tumors originating from the muscularis propria [22–26]. However, these biopsy procedures are difficult for small tumors [19, 20], especially those smaller than 10 mm. The true malignant potential for individual GISTs cannot be accurately determined using current imaging and noninvasive sampling methods [27].

Laparoscopic resection appears to be a safe and effective alternative method for the treatment of gastric tumors [28]. However, its application to small tumors is limited, especially those less than 10 mm. Several endoscopic resection techniques have potential advantages for small lesions originating from the muscularis propria. Endoscopic band ligation is a simple, effective, and safe procedure for treating gastrointestinal submucosal tumors of less than 10 mm, although it does not allow for a complete pathological examination because the tumor mass is exfoliated directly into the lumen and excreted [7, 8]. Endoscopic enucleation or endoscopic submucosal dissection using an insulated-tip knife, hook knife, forceps, or electrocautery snare may have the advantage of providing both a pathological diagnosis and clinical treatment for such lesions in selected patients [1–6]. However, these techniques usually require highly skillful manipulation by experienced specialists and relatively longer procedure times [1–6, 29]. In one study, the complete resection rate for small tumors, especially those smaller than 1 cm, was lower than that for larger tumors [5]. It was more difficult to strip the covering mucosa and dissect the submucosal layer in the small tumors [5].

Ligation-assisted endoscopic mucosal resection is one of several endoscopic treatment modalities employed in the treatment of esophageal neoplasia [30]. Endoscopic submucosal resection with a ligation device was successfully performed in all 25 esophageal subepithelial tumors localized within the muscularis mucosa or submucosa by Lee et al. [31], and the en bloc resection rate was 100% (25 of 25 tumors). Endoscopic ligation and resection shows promise as a safe and feasible technique with which to treat small EUS-suspected gastric GISTs [10].

EE-L was developed for the diagnosis and treatment of small gastrointestinal lesions originating from the muscularis propria. This technique combines endoscopic band ligation and endoscopic enucleation to fully exploit the advantages of each technique.

Endoscopic band ligation of the tumor simplifies the endoscopic enucleation procedure and reduces the time required because the elastic bands firmly ligate the lesions and cause them to assume a polypoid form with pseudostalks during the entire enucleation process. After the tumor has been ligated, the overall endoscopic enucleation process progresses easily and smoothly. The enucleation time in the present study was short (mean, 7.2 minutes; range, 5–11 minutes).

Perforation is a recognized complication during endoscopic resection, even in the hands of an expert endoscopist [1–6]. The EE-L technique may substantially decrease the risk of perforation during the enucleation process. First, the base below the tumor is firmly ligated by the elastic band. Second, precutting the overlying mucosa and submucosa above the tumor and then gradually enucleating the tumor maintains the major overlying mucosa, facilitating the wound closure procedure and promoting wound healing [2]. Third, closing the wound with clips and tissue adhesives prevents perforation and promotes wound healing. Clips have been widely applied for wound closure of the gastrointestinal tract to prevent perforation and bleeding [32–35].

Tissue adhesives are used for a variety of local applications including hemostasis, wound closure, and fistula repair. The most commonly utilized tissue adhesives in gastrointestinal endoscopy include cyanoacrylates, fibrin glues, and thrombin. Cyanoacrylates are widely used outside of the United States for gastric variceal bleeding and, to a lesser extent, ulcer bleeding and fistula closure [36]. Tissue adhesives sprayed onto the surface of the clips and wound firmly immobilize the clips onto the wound. This not only effectively prevents perforation, but also prevents bleeding from the iatrogenic ulcers [37, 38].

This study has some limitations. Like other endoscopic resection techniques for removal of tumors arising from the muscularis propria, EE-L is limited in that complete enucleation is defined solely by endoscopic observation. It was impossible to remove tumor tissue with a sufficiently safe margin using endoscopic resection. Therefore, long-term follow-up assessment should be performed to ensure the complete removal of GISTs. Although endoscopic resection cannot completely eliminate the need for continued surveillance in all patients, it can eliminate the need for surveillance in patients with leiomyomas. For patients with GISTs in this study, endoscopic resection achieved complete tumor resection based on gross observation, and no residual tumor or tumor recurrence was found during the long-term follow-up.

This study shows that EE-L appears to be a safe, effective, and relatively simple technique for the pathological diagnosis and resection of small gastrointestinal tumors originating from the muscularis propria. Controlled clinical trials with larger sample sizes and longer follow-up periods are necessary to further examine the value and limitations of this technique.