Characteristics and prognosis of small bowel tumors: A retrospective study

Abstract

BACKGROUND

Small bowel tumors (SBTs) are a heterogeneous group of difficult-to-diagnose tumors that account for 2%-5% of all gastrointestinal tumors. Single-balloon enteroscopy greatly enhances the diagnosis and treatment of SBTs. However, few epidemiological studies have been conducted in Taiwan to determine the clinical profile of SBTs.

AIM

To investigate the clinical characteristics, managements and prognosis of SBTs in a medical center in Taiwan.

METHODS

The study enrolled 51 patients aged 58.9 ± 8.8 years (range, 22-93) diagnosed with SBTs from November 2009 to July 2021. We retrospectively recorded clinical characteristics, indications, endoscopic findings, pathological results, management, and outcomes for further analysis.

RESULTS

A male preponderance was observed (56.8%). The most common indications were suspected small intestinal tumors (52.9%) and obscure gastrointestinal bleeding (39.2%). The most common tumor location was the ileum (41.2%). The performance of imaging studies (P = 0.004) and the types of findings (P = 0.005) differed significantly between malignant and benign SBTs. The most frequent imaging finding was a small intestinal mass (43.1%). The top three malignant tumor types were gastrointestinal stromal tumors (GISTs), adenocarcinomas, and lymphomas. Moreover, the proportions of benign and malignant tumors were 27.5% and 72.5%, respectively. The survival rates of patients with malignant tumors in the GIST and non-GIST groups differed significantly (P = 0.015). Kaplan–Meier survival analysis showed a significant difference in survival between patients in the malignant and benign groups (P = 0.04). All patients with lymphoma underwent chemotherapy (n = 7/8, 87.5%), whereas most patients with GISTs underwent surgery (n = 13/14, 92.8%).

CONCLUSION

Patients with GISTs have a significantly higher survival rate than those with other malignant SBTs. Therefore, a large-scale nationwide study is warranted to evaluate the population-based epidemiology of SBTs.

Key Words: Gastrointestinal bleeding; Intestinal neoplasms; Single-balloon enteroscopy, survival rate

Core Tip: Small bowel tumors (SBTs) are a heterogeneous group of difficult-to-diagnose tumors, accounting for 2%-5% of all gastrointestinal tumors. This study aimed to investigate the clinical characteristics and prognosis of SBTs detected using single-balloon enteroscopy. The most common tumor location was the ileum (41.2%). The top three malignant tumor types were gastrointestinal stromal tumors (GISTs), adenocarcinomas, and lymphomas. Moreover, the proportions of benign and malignant tumors were 27.5% and 72.5%, respectively. The survival rates of patients with malignant tumors in the GIST and non-GIST groups differed significantly. Therefore, a large-scale nationwide study is warranted to evaluate the population-based epidemiology of SBTs.

INTRODUCTION

Small bowel disease was thought to be rare approximately 20 years ago because of the lack of adequate diagnostic procedures. The development of capsule endoscopy has enabled visualization of the entire small intestinal mucosa. Subsequently, the introduction of balloon-assisted enteroscopy (BAE) enabled diagnostic and therapeutic procedures in the small intestine[1-4]. These advancements have greatly enhanced the evaluation of patients with suspected small bowel pathology. The incidence of primary small bowel tumors (SBTs) is increasing worldwide, and data from the Surveillance, Epidemiology, and End Results Program for the annual average age-adjusted incidence of small bowel cancer show 2.3 new cases of small bowel cancer and 0.4 deaths caused by it per 100000 population per year. These rates are age-adjusted based on the case and death data from 2014 to 2018. The average annual age-adjusted incidence of small bowel cancer has nearly doubled from 1975 to 2018[5,6]. Several kinds of histologic diagnoses have been reported for SBTs, which collectively account for approximately 2%-5% of all gastrointestinal tumors[7,8]. The most common types of SBTs include adenocarcinomas, lymphomas, carcinoid tumors, and stromal cell tumors[6-10]. We present the indications, diagnostic data, and outcomes of SBTs detected by single-balloon endoscopy (SBE) in one medical center.

MATERIALS AND METHODS

Study design

Successive anterograde and retrograde SBE were performed in 450 patients who underwent SBE for suspected small bowel lesions at [blinded for review] in Taiwan between November 2009 and July 2021. After being informed about the benefits and potential risks of SBE, all patients signed an informed consent form. The need to obtain patient consent for publication was waived because patient information was anonymized and de-identified prior to the analysis (Figure 1). The characteristics of the patients who underwent SBE, including age and sex, were recorded from their medical charts. An experienced gastroenterologist performed 450 SBE procedures for the following indications: Suspected intestinal tumor (68/450, 15%), obscure gastrointestinal bleeding (267/450, 59%), bowel obstruction (22/450, 5%), abdominal pain (47/450, 10%), chronic diarrhea (8/450, 2%), suspected inflammatory bowel disease (36/450, 8%), and foreign body in the jejunum (2/450, 0.4%). Patients initially suspected of having an intestinal tumor were excluded from the study if SBE confirmed no tumor. Fasting for > 8 hours was required prior to anterograde SBE. Bowel preparation for retrograde SBE was performed using sodium phosphate or polyethylene glycol solution. The SBE insertion route (oral or anal) depended on the clinical presentation and previous imaging results. SBE (SIF-Q260; Olympus Corp., Tokyo, Japan) was used to examine the small intestine using the push-and-pull method. A cap was also used at the tip of the enteroscope to improve the visual field and prevent mucosal hooking during the examination. The Institutional Review Board of [blinded for review] approved this study (IRB No. A202005031).

Figure 1 Flow chart of the selection of study participants. SBTs: Small bowel tumors.

Statistical analysis

Continuous data were expressed as mean (standard deviation), and discrete data were presented as numbers (percentages). The association of categorical and continuous variables between groups was examined using the χ² test or independent t-test, as appropriate. Differences in survival times were assessed among the different diagnostic groups using the Kaplan-Meier method. All statistical analyses were performed using the IBM Statistical Package for the Social Sciences software version 22 (IBM^® SPSS^® statistics 22, IBM Corp., Armonk, NY, United States).

RESULTS

SBE procedures were performed in 450 enteroscopic examinations (330 patients), and SBTs were detected in 51 patients enrolled in this study. The mean age (± SD, range) was 58.9 (± 18.8, 22-93) years. Men outnumbered women in the overall population (56.8% men and 43.2% women) and in the benign (71.4% men and 28.6% women) and malignant (51.4% men and 48.6% women) groups. The most common indications were suspected small intestinal tumors (52.9%) and obscure gastrointestinal bleeding (39.2%). The other indications included abdominal pain and bowel obstruction. The most common tumor location was the ileum (41.2%), followed by the jejunum. The tumor location and indications for enteroscopy showed no significant differences between benign and malignant SBTs. Most patients underwent imaging studies, including abdominal computed tomography (CT) or magnetic resonance imaging (MRI); however, some patients did not undergo any examination before enteroscopy. The performance of imaging studies (P = 0.004) and the type of findings (P = 0.005) differed significantly between malignant and benign SBTs (Table 1). The most frequent imaging finding was a small intestinal mass (43.1%). The top three malignant tumor types were gastrointestinal stromal tumors (GISTs), adenocarcinomas, and lymphomas. Moreover, the proportions of benign and malignant tumors were 27.5% and 72.5%, respectively. Representative images of malignant SBTs are shown in Figure 2, and benign SBTs are shown in Figure 3. Clinically, patients with GISTs had relatively better outcomes than those with other malignant SBTs. Therefore, we compared the survival rates between patients in the GIST and non-GIST groups and found a significant difference (P = 0.015) (Table 2). Kaplan-Meier survival analysis also showed a significant difference in survival between patients in the malignant and benign groups (P = 0.040) (Figure 4). We also analyzed the survival curves of all the malignant SBTs subtypes and found no significant differences (Figure 5).

Figure 2 Malignant small bowel tumors. A: Adenocarcinoma; B: Gastrointestinal stromal tumor; C: Neuroendocrine tumor; D: Diffuse large B cell lymphoma; E: Sarcomatoid carcinoma; F: Metastatic squamous cell carcinoma of the cervix.

Figure 3 Benign small bowel tumors. A: Adenoma; B: Inflammatory myofibroblastic tumor; C: Ectopic gastric mucosa; D: Lipoma; E: Hamartoma of Peutz-Jeghers syndrome; F: Lymphatic hyperplasia.

Figure 4 The Kaplan-Meier survival curve analysis of benign and malignant small bowel tumors. SBTs: Small bowel tumors.

Figure 5 The Kaplan-Meier survival curve analysis of subgroups of malignant small bowel tumors. GIST: Gastrointestinal stromal tumor; SBTs: Small bowel tumors.

Table 1 Demographic data and clinical manifestations of the enrolled patients (n = 51).

	n (%)	Benign	Malignant	P value
Sex	Male	29 (56.8)	10 (71.4)	19 (51.4)	0.329
	Female	22 (43.2)	4 (28.6)	18 (48.6)
Age		58.9 ± 18.8	45.9 ± 18.7	63.9 ± 16.5	0.002
Indication	Suspected small intestinal tumor	27 (52.9)	7 (50.0)	20 (55.3)	0.558
	Obscure GI bleeding	20 (39.2)	5 (35.7)	15 (39.5)
	Bowel obstruction	1 (2.0)	0 (0.0)	1 (2.6)
	Abdominal pain	3 (5.9)	2 (14.3)	1 (2.6)
Endoscopy route	Antegrade	24 (47.1)	3 (21.4)	21 (56.8)	0.074
	Retrograde	20 (39.2)	8 (57.1)	12 (32.4)
	Bidirectional	7 (13.7)	3 (21.4)	4 (10.8)
Imaging study	Not performed	4 (7.8)	4 (28.6)	0 (0.0)	0.004
	CT/MRI	47 (92.2)	10 (71.4)	37 (100)
Imaging finding	None	18 (35.3)	6 (42.9)	12 (32.4)	0.005
	Small intestinal tumor	22 (43.1)	3 (21.4)	19 (51.4)
	Wall thickening	7 (13.7)	1 (7.1)	6 (16.2)
	Not performed	4 (7.8)	4 (28.6)	0 (0.0)
Presentation	None	1 (2.0)	0 (0)	1 (2.8)	0.501
	Abdominal pain	12 (23.5)	5 (35.7)	7 (19.4)
	GI bleeding	29 (56.9)	6 (42.9)	22 (61.1)
	Other	9 (17.6)	3 (21.4)	6 (16.7)
Location	Duodenum	11 (21.7)	2 (14.3)	9 (24.3)	0.615
	Jejunum	19 (37.3)	5 (35.7)	14 (37.8)
	Ileum	21 (41.2)	7 (50)	14 (37.8)

CT: Computed tomography; GI: Gastrointestinal; MRI: Magnetic resonance imaging.

Table 2 Survival and mortality rate associated with benign and malignant lesions (n = 51).

			n (%)	Died	Survived	P value
Malignant tumors (n = 37)	GIST (n = 14)	Gastrointestinal stromal tumor	14 (27.5)	0 (0.0)	14 (100.0)	0.015a
	Non-GIST (n = 23)	Adenocarcinoma	9 (17.6)	4 (44.4)	5 (55.6)
		Lymphoma	8 (15.7)	2 (25.0)	6 (75.0)
		Neuroendocrine tumor	2 (3.9)	0 (0.0)	2 (0.0)
		Sarcomatoid carcinoma	2 (3.9)	1 (50.0)	1 (50.0)
		Metastatic tumors1	2 (3.9)	1 (50.0)	1 (50.0)
Benign tumors (n = 14)			14 (25.5)	0 (0.0)	14 (100.0)

¹Metastatic squamous cell carcinoma = 1, metastatic pancreas cancer = 1.

^aGIST group vs non-GIST group.

GIST: Gastrointestinal stromal tumor.

Management data were collected according to enteroscopic diagnosis (Table 3). All patients with lymphomas in our study received only chemotherapy, except for one patient who had bowel obstruction (n = 7/8, 15.4%), whereas most of the patients with GISTs underwent surgery (13/14, 26.9%). Other malignant SBTs in which SBE played an important role in the diagnosis also received adequate treatment.

Table 3 Outcomes and management of malignant small bowel tumors (n = 37).

Diagnosis	n (%)	Outcome (n)	Management (n)
Malignant small bowel tumors
GIST	14 (37.8)	Cure (14)	Operation (13) and Target therapy (1)
Adenocarcinoma	9 (24.3)	Cure (6) Expire (4)	Operation (4) Chemotherapy (2) Operation and chemotherapy (3) Hospice care (1)
Lymphoma	8 (21.6)	Cure (6) Expire (2)	Chemotherapy (7) Operation and chemotherapy (1)
Sarcomatoid carcinoma	2 (5.4)	Expire (1) Lost to follow-up (1)	Operation (1)
Neuroendocrine tumor	2 (5.4)	Cure (2)	Operation (2)
Metastatic squamous cell carcinoma	1 (2.7)	Expire (1)	Operation (1)
Metastatic pancreatic cancer	1 (2.7)	Follow-up (1)	Operation and chemotherapy (1)
Benign small bowel tumors
Hyperplastic polyp	3 (21.4)	Follow-up (4)	Biopsy (3)
Lipoma	3 (21.4)	Follow-up (3)	Unroofing polypectomy (1) Biopsy (2)
Peutz-Jeghers syndrome	2 (14.3)	Follow-up (2)	Polypectomy (2)
Adenoma	2 (14.3)	Cure (2)	Operation (1) Polypectomy (1)
Lymphatic hyperplasia	1 (7.1)	Follow-up (1)	Biopsy (1)
Ectopic gastric mucosa	1 (7.1)	Follow-up (1)	Biopsy (1)
Ectopic pancreas	1 (7.1)	Cure (1)	Operation (1)
Inflammatory myofibroblast tumor	1 (7.1)	Cure (1)	Operation (1)

GIST: Gastrointestinal stromal tumor.

DISCUSSION

To the best of our knowledge, malignant neoplasms in the small bowel are rare, accounting for less than 3% of all gastrointestinal tract cancers, even though the small intestine represents the largest part of the gastrointestinal tract[9,11-14]. Owing to the free intraperitoneal location, energetic contractility, and length of the small intestine, lesions from the small intestine are difficult to approach using conventional endoscopy. Moreover, SBTs show no specific clinical presentation, and different histological types of SBTs have distinct symptoms and require different treatment strategies. Thus, understanding the epidemiology of SBTs is important for raising physicians’ awareness about these diseases. A multicenter retrospective study conducted in 2018 described epidemiologic results in terms of sex, age, lesion location, and the association between malignant and benign SBTs. That study also presented pronounced static results regarding the initial presentation, pre-endoscopy imaging, and tumor morphology[9]. However, it did not discuss the relationship between the indications for SBE and the localization of malignant and benign tumors. We used the Kaplan-Meier method for survival curve analysis to compare malignant and benign tumors and lesions of different histological types. We also evaluated the management of SBTs and their outcomes.

In the United States, small intestine cancer is slightly more common among men than among women, with the highest incidence among patients aged 65-74 years. The median age of patients at diagnosis is 66 years[5]. In our study, patients with SBTs had a mean (± SD) age of 58.9 (± 18.8) years. Patients with malignant SBTs were significantly older than those with benign SBTs, with a mean (± SD) age of 63.9 (± 16.5) and 45.9 (± 18.7) years, respectively (P = 0.002). More men than women developed malignant (n = 19, 51.4%) and benign SBTs (n = 10, 71.4%) (Table 1). This result is consistent with the data from the United States and Asia.

Hyperplastic polyps, lipomas, adenomas, and Peutz-Jeghers syndrome are the most common benign SBT types. In our study, GISTs were the most common malignant SBTs, accounting for 37.8% of the malignant cases (n = 14). Adenocarcinoma and lymphoma accounted for 24.3% (n = 9) and 21.6% (n = 8) of cases, respectively (Table 3). Current clinical data in the United States have revealed that the most common small bowel malignancies are adenocarcinomas and neuroendocrine tumors[15]. Although this result was different from that of our study, in which only two neuroendocrine tumors were detected, our data are compatible with those of ethnic groups in Asia, where the most common malignant SBTs are GISTs, adenocarcinomas, and lymphomas[10,16]. This trend in malignant SBT distribution might be caused by differences in race, habitation, or diet. There was no statistically significant difference in the tumor location between malignant and benign SBTs. In the present study, malignant SBTs were mostly found in the ileum, whereas GISTs were most often found in the jejunum, adenocarcinomas in the duodenum, and lymphomas in the ileum. This result is compatible with those obtained in other studies conducted in Asia[7,9,10,16]. The pathophysiology may be that the duodenum is the first portion of the small bowel to be exposed to an abundant variety of ingested chemicals and pancreaticobiliary secretions. Thus, a higher rate of adenocarcinoma may indicate that the ingested chemicals and pancreaticobiliary secretions have carcinogenic properties[17,18].

The relationship between the indications for SBE and SBTs was also analyzed, which revealed no significant difference. Suspected small intestinal lesions (52.9%, n = 27) and obscure gastrointestinal bleeding (39.2%, n = 20) were the most common clinical presentations. This indicated that no specific clinical presentation could be an indication for SBTs with heterogeneous histological results. SBTs should always be considered if a patient shows refractory illness without adequate pathogens.

Some of our patients underwent imaging studies (abdominal CT or MRI), whereas others did not undergo any examination before SBE. Significant differences were observed between these two groups (P = 0.004), and the imaging findings showed a significant difference (P = 0.005). Thus, SBE was performed for further evaluation if the patient showed relevant findings in the imaging study, and imaging studies, including abdominal CT or MRI, were considered in patients suspected of having small bowel lesions.

We also analyzed the subtypes of malignant SBTs and found that patients in the GIST group had a better prognosis. All GISTs are now considered to have malignant potential; however, small tumors are generally considered benign, especially when the cell division rate is not rapid[19-22]. Our study also showed different survival rates between the patients in the GIST and non-GIST malignant SBT groups (P = 0.015). The data collected showed that most GISTs were resectable (92.9%, n = 13/14). Thus, BAE was considered to play an important role in the detection of intestinal GISTs and ensure that patients underwent proper surgery.

The Kaplan-Meier survival analysis showed a significant difference between malignant and benign tumors (P = 0.040) (Figure 4). We also evaluated the subtypes of malignant SBTs but obtained no significant findings, which may be attributed to the heterogeneity of malignant SBTs (Figure 5). Some patients showed a good prognosis if they received adequate treatment (e.g., GIST and lymphoma), but others showed a poor response to treatment within a short period of time, especially those with metastatic lesions.

We also evaluated the management of patients with SBTs. We found that almost all patients with GISTs underwent surgery (92.9%, n = 13/14), except for one patient who was too old to undergo surgery and showed a good response during the study period. All patients with lymphoma underwent chemotherapy (n = 7), and one patient underwent surgery due to bowel obstruction (Table 3). Since BAE can also yield biopsy samples, other interventions are not required to diagnose SBTs, and proper treatment can be initiated after a confirmed diagnosis to avoid the adverse effects of other tests or treatments.

The incidence of SBTs has increased worldwide over the past several years, probably owing to technological advancements in radiology, endoscopy, and physician awareness[2,3,5,6]. Nevertheless, many challenges persist in relation to the difficulty of early diagnosis and understanding treatment for different histological subtypes. This study provides epidemiological data on SBTs from a single-center BAE investigation. These findings highlight the differences in the survival rates among patients with various SBTs. We also evaluated the management and outcomes of all the patients with SBTs. However, two primary factors limit the significance of our findings. First, the relatively small sample size limited the generalizability of the results, indicating the need for large-scale nationwide studies to accurately assess the population-based epidemiology of SBTs. Second, the diagnostic yield of SBE was lower than that of double-balloon enteroscopy (DBE), which may have further limited our findings. The DBE system consists of an overtube and an endoscope equipped with balloons, which facilitate deep insertion into the small bowel by alternately inflating and deflating the balloons. In contrast, SBE features a balloon only at the tip of the overtube. In our study, we used SBE to detect the small bowel lesions. Studies comparing the efficacy and safety of DBE and SBE have yielded contradictory results. A prospective study revealed that DBE had higher completion rate and diagnostic yield[23]. In contrast, a meta-analysis of randomized studies showed the same diagnostic ability, therapeutic yield, and complication rate of both DBE and SBE[24]. However, DBE demonstrated superior visualization of the entire small bowel. A recent study indicated that retrograde DBE has a higher depth of insertion and shorter procedural duration, but a similar diagnostic yield and technical success than SBE[25]. To enhance the diagnosis, therapeutic yield, and completion rate, small bowel lesions should be approached using both anterograde and retrograde methods. Additionally, in some cases, imaging studies and capsule endoscopy can be valuable for accurately locating small bowel lesions, thereby improving the efficacy of SBE.

CONCLUSION

Further research on screening diagnostic modalities for SBTs is needed to improve patient outcomes through early detection, and the genetic or environmental factors that account for the development of SBTs should be investigated.