Gastroduodenoscopy used for patients with gastrointestinal symptoms has identified a duodenal mass with increasing frequency. Duodenal adenomas, the most common manifestation of a duodenal neoplasm, are known to be precancerous lesions with a relatively high rate of malignant transformation [1, 2]. Most duodenal adenomas are resected surgically, although endoscopic mucosal resection (EMR) has comparable efficacy for the treatment of premalignant lesions as well as early malignant lesions of the esophagus, stomach, and colorectum [3, 4].

To date, the data on EMR for duodenal neoplasms are limited [5–7]. The safety and efficacy of EMR for the treatment of nonampullary duodenal neoplasms, predominantly duodenal adenomas, was evaluated for patients who had no familial polyposis syndrome.

Patients and methods

Study population

Between August 2002 and February 2009, 24 patients with a nonampullary duodenal neoplasm and no familial polyposis syndrome were included in the study and evaluated retrospectively. All the patients underwent diagnostic esophagoduodenoscopy whether they had gastrointestinal symptoms or not. In addition, data on the duodenal mass or polyp such as type and size were estimated, and a biopsy was performed using a conventional gastroscope or side-view duodenoscope (Olympus, Tokyo, Japan). Cases in which the lesion involved the ampulla or was confirmed histologically as an adenocarcinoma with submucosal invasion were excluded from the study.

Technical aspects of the EMR

After the adenoma or adenocarcinoma confined to the mucosa was confirmed by histology, EMR was performed. The EMR was carried out by experienced endoscopists (S.W.J., C.M.C.) who had managed more than 100 EMR cases over several years. With the patient under sedation using propofol and midazolam, conventional EMR and EMR with ligation (EMR-L) were performed using a snare-assisted technique and submucosal injection of normal saline solution with a single-use 21-gauge needle (Olympus). In two cases, a standard snare polypectomy without submucosal injection was performed. Two types of oval-shaped electrosurgical snares were used (diameters of 10 and 25 mm, Olympus).

The EMR technique was individualized by the endoscopist on a case-by-case basis. All duodenal lesions were trapped and resected as a whole lesion (en bloc method). Intraprocedural bleeding was defined as bleeding during the EMR. Hemostasis was provided using one or more of the following: treatment with a 1:10,000 epinephrine injection, clipping with hemoclips, or electrocoagulation with a coagulation probe (Fig. 1). The EMR was performed during a hospital admission or in the outpatient clinic.

Fig. 1
figure 1

Endoscopic resection of duodenal adenomas in the second portion of the duodenum. Electrocoagulation and hemoclipping were used for hemostasis

Full size image

Definition of terms and follow-up evaluation

A residual adenoma was defined as microscopically identified adenoma tissue with or without a remnant lesion identified macroscopically. A recurrent adenoma was defined as a microscopic adenoma at any time during the follow-up period after complete removal of the primary duodenal lesion. The duration of the follow up-period was defined as the number of months from the initial diagnostic endoscopy to the most recent endoscopy.

Results

Characteristics of the patients and the lesions

The mean age of the 17 men and 7 women included in the study was 57 years (range, 40–82 years). A total of 15 patients with comorbidities such as cardiopulmonary disease or diabetes were admitted to the hospital for the endoscopic resection (mean admission duration, 6.2 days; range, 3–9 days). The patients were followed up for a mean of 6 months (range, 3–36 months).

Outcome of the endoscopic resection

During the study period, 24 patients were included in the analysis. Duodenal adenoma was diagnosed for 23 patients and duodenal adenocarcinoma of the mucosa for 1 patient. For the patient with the mucosal adenocarcinoma, an abdominal computed tomography and positive emission tomography showed no evidence of metastatic or locally invasive disease.

In terms of tumor type, 17 tumors were tubular, 4 were villotubular, and 3 were villous. Of the 24 patients, 21 had mild dysplasia and 2 had high-grade dysplasia. All the patients had a single duodenal polyp except for one patient with three duodenal polyps. For 12 patients, the duodenal adenoma was located in the second portion of the duodenum, the most common site for its location. The lesion was located in the first portion of the duodenum for 11 patients and in the third portion for 1 patient. Although no objective measurements were obtained for the size of the polyp, it was estimated that 13 patients had polyps 1 to 2 cm in size, and 8 patients had a polyp was smaller than 1 cm (Table 1).

Table 1 Characteristics of the 24 patients, the features of the duodenal polyps, and the method and outcome of management
Full size table

With regard to management, a conventional EMR was performed in 19 cases, an EMR-L in 3 cases, and a snare polypectomy in 2 cases. A complete resection was performed for 17 patients after the first conventional EMR. Two patients who underwent a conventional EMR had residual lesions defined by the presence of adenoma tissue in the resected margins (2/2 positive vertical and lateral margins). The one patient was confirmed to have adenocarcinoma at the follow-up endoscopy 1 month after the EMR and was referred for surgical management. The other patient showed a visible macroscopic lesion, diagnosed as a recurrent adenoma at follow-up endoscopy 1 month after the EMR and resected by conventional EMR. No macroscopic lesion was identified at the next follow-up endoscopy. Two of three patients who underwent an EMR-L were completely resected, and one had involvement of the lateral resected margin. None of these patients showed recurrence at follow-up endoscopy. Two patients had a snare polypectomy and showed no recurrence at follow-up endoscopy (Fig. 2).

Fig. 2
figure 2

Flow chart of treatment for duodenal neoplasms. The numbers in parenthesis show how many patients are in each category

Full size image

Complications

Intraprocedural bleeding was noted in seven patients. Five of these patients underwent endoscopic hemostasis with hemoclips and electrocoagulation. Only electrocoagulation was needed for two patients with bleeding. Delayed bleeding was not detected, and other complications such as perforation or infection did not occur. There was no procedure-related death.

Discussion

Duodenal adenomas are precancerous lesions characterized by a high rate of malignant transformation. Despite considerable technical advances in endoscopic resection for superficial neoplasms of the gastrointestinal tract, few reports have described duodenal EMR. Due to the risk of perforation and difficulty with bleeding control, many investigators consider duodenal endoscopic treatment a high-risk procedure [5]. The duodenal adenoma causes nonspecific clinical manifestations such as epigastric pain (36%), jaundice (34%), weight loss (20%), anemia (12%), vomiting (5%), and diarrhea (2%). In many cases, the lesion is found incidentally [8]. However, these lesions occur more frequently with aging. The mean age at diagnosis is 60 to 72 years [9].

Although surgical resection has been considered the standard treatment for duodenal adenomas, a high rate of complications has been associated with the surgery due to the structures around the duodenum [10, 11]. For this reason, endoscopic resection of duodenal polyps may be a better approach and has been used more frequently in recent years.

The recurrence rate in the current series was 8.3% (2/24). Three cases had a villous adenoma, which is known to have a relatively high rate of recurrence [12]. Farnell et al. [13] reported that among 50 patients with benign tumors managed by transduodenal local excision, 17 had a recurrence. The recurrence rate was 32% at 5 years and 43% at 10 years. However, in the current study, none of the patients with a villous adenoma, including the mucosal adenocarcinoma, had a recurrence. However, the follow-up period was short. Two of the four villotubular adenomas recurred (patients 2 and 4 in Table 1). In the one case, the recurrence was managed with complete resection using the endoscopic procedure. However, in the other case, a snare polypectomy was performed at 6 months for a recurrent villous adenoma at the same site, and an adenocarcinoma with submucosal invasion was confirmed at the 12-month follow up evaluation. Although the follow-up period was short, the patients with nontubular adenomas seemed to show a tendency for a high recurrence rate.

The complete resection rate in this series was 87.5% (21/24) during the first 3 months of the follow-up period. The total length of the follow-up period varied widely. However, based on the short-term follow-up evaluation, the feasibility of endoscopic resection was comparable with that of more traditional procedures. Due to the en bloc resection of the neoplasm and no additional ablative thermal treatment such as argon plasma coagulation that could destroy adenomatous tissue, the histologic assessment of the resection margin and confirmation of a complete resection were reliable and accurate.

Two patients with residual lesions after the first endoscopic resection, excluding the patient with later development of an adenocarcinoma, had positive resection margins. Because they had no macroscopic lesion at follow-up assessment 1 to 3 months after the endoscopic resection, they received a diagnosis of “no recurrence.”

The complication rate in the current study was 26.2% (7/24). Ahmad et al. [6] reported a 33% frequency of bleeding during a duodenal EMR, and Lepilliez et al. [14] reported a 25% frequency of bleeding. In our series, all the patients with bleeding had intraprocedural bleeding. These findings are similar to those of previous reports. There was no case of massive bleeding requiring a transfusion. However, for patients with bleeding, additional hospitalization was required.

In conclusion, endoscopic resection had an outcome comparable with that of more traditional procedures and proved to be a safe and effective treatment for duodenal polyps. Therefore, EMR should be considered as an alternative to surgery, especially for patients that are not good candidates for surgery. Additional studies are needed to determine the long-term outcome for endoscopic resection of duodenal neoplasms.