The aim of our study is to compare the efficacy and safety of “outlet-shape” tailored stents with standard stents for the management of distal gastric cancer causing gastric outlet obstructions (GOOs) with varying gastric cavity shapes and sizes. To determine the shape and size of the GOOs, stomach opacifications were performed using contrast media before stenting. Two basic shapes of the residual cavity of the proximal GOO were observed: cup shaped or approximately cup shaped and funnel shaped or approximately funnel shaped. Other shapes were not found. In the GOO tailored group, the size and shape of the proximal ends of the tailored stent were suited for the residual cavity of the proximal GOO. The tailored stents included large cup-shaped stents and large funnel-shaped stents. GOO tailored covered stents led to less restenosis and reintervention rates compared to standard uncovered stents but with the same survival.
Various types of metal stents are used for the treatment of nonresectable gastric outlet obstruction (GOO) [
This study was conducted between May 2009 and March 2013 and was designed as a multicenter, controlled, prospective, observational, and randomized clinical trial involving three large hospitals, the First People’s Hospital of Yuhang District, the First Affiliated Hospital of Zhejiang University, and the First Affiliated Hospital of Henan College of Traditional Chinese Medicine. The study design was approved by the Infection Control and Ethics Committee of the above mentioned centers and was performed in compliance with the hospital policies related to the use of human subjects and human-derived material and informed consents regarding the study and procedures were obtained from all patients.
Inclusion criteria were (1) GOO defined by symptoms resulting in decreased oral intake (nausea, vomiting, and inability to eat), (2) the obstruction which was caused by primary distal stomach cancer, and (3) the site of stenosis which was between the gastric body and duodenum bulb. (4) All patients who were selected for stent placement had inoperable cancers or were unsuitable for surgery because of the presence of severe comorbid conditions. Exclusion criteria included the presence of only mild symptoms in patients who could tolerate a liquid diet, clinical evidence of perforation or peritonitis, and evidence of multiple small-bowel obstructions because of peritoneal seeding as well as diabetes or other diseases that affect gastric motility and use of promotility agents. Of the 75 patients, 10 patients refused stent implantation and the other 65 patients were randomly divided into two groups: a GOO tailored group (33 cases) in whom GOO tailored covered stents were used and a control group (32 cases) in whom uncovered standard stents were used. Patients were randomized to either the GOO tailored group or the control group, using a table of random numbers. Gastric outlet obstruction scoring (GOOS) was performed according to the scoring system introduced by Song et al. [
Stomach opacification was performed using contrast media less than three days before stent design in order to determine the shape of the GOO. Stomach opacification and stent design have been described by us previously [
Numbers of cases and obstruction dimensions.
GOO tailored group | Control group |
| |||
---|---|---|---|---|---|
Cup obstruction ( |
Funnel obstruction ( |
Cup obstruction ( |
Funnel obstruction ( | ||
Breadth (mm) | 53.3 ± 5.5 | 33.6 ± 3.6 | 53.5 ± 5.7 | 33.9 ± 3.7 | >0.05 |
Length (mm) | 15.1 ± 1.5 | 17.4 ± 1.8 | 15.3 ± 1.6 | 17.4 ± 1.9 | >0.05 |
Examples of proximal GOO lumen shapes and images of GOO tailored stents for GOO. The distal portion of the GOO tailored stents was semispherical, with a length of 20 mm and a diameter of 28 mm. The middle segment had a diameter of 20 mm. The overall length of the stents was 100 mm. (a) A cup-shaped obstruction and (b) a cup stent. The proximal ends were large cup-shaped (53.3 mm ± 5.5 mm in diameter, 15 and 20 mm in length). (c) A funnel-shaped obstruction and (d) a funnel stent. The proximal ends were funnel-shaped (33.6 mm ± 3.6 mm in diameters, 25 mm and 30 mm in lengths). (e) Standard uncovered stent. The length of the stents was 100 mm and the ends were semispherical with diameters of 28 mm and length of 20 mm.
Standard uncovered stents MTN-CG-s-20/100 (Micro-tech (Nanjing) Co., Ltd, Nanjing, Jiangsu, China) were used in the control group. The ends of the stents were semispherical with diameters of 28 mm and length of 20 mm. The length of the stents was 100 mm (Figure
All stent placements were performed by the same gastroenterologists. The stent cups with a proximal diameter of 53.3 mm ± 5.5 mm at full expansion were available in 2 lengths, 15 and 20 mm, which corresponded to obstructions <15 mm and ≥15 mm in length. The stent funnels with a proximal diameter of
X-ray images of stents. (a) Image showing a stent released. (b) An example of barium contrast radiography of a funnel stent.
Three to five days after stent placement, barium contrast radiography was performed to document the position and the function of the stents (Figure
Primary endpoints were the stent complications ingrowth/overgrowth and stent migration. Secondary endpoints were the adverse events including bleeding, abdominal pain and food impaction.
According to previous literature, reobstruction and migration occurred in 49.2% of self-expanding metal stents particularly in GO patients [
Table
Patient characteristics.
GOO tailored ( |
Control ( |
|
|
---|---|---|---|
Male/female | 18/15 | 17/15 | >0.05 |
Average age (y) | 76.4 ± 7.7 | 75.8 ± 7.6 | >0.05 |
Differentiated degree | >0.05 | ||
Moderately | 10 | 9 | |
Poorly | 23 | 23 | |
TNM staging | >0.05 | ||
IIIA | 6 | 8 | |
IIIB | 8 | 6 | |
IV | 19 | 18 | |
Comorbidities | 4 | 3 | >0.05 |
Chemotherapy | 2 | 1 | >0.05 |
GOOSS (mean) | 4.4 ± 0.4 | 4.2 ± 0.4 | >0.05 |
|
|||
Lost for followup | 1 | 1 |
TNM, tumor, nodes, metastasis; GOOSS, gastric outlet obstruction score.
The efficacy and complication rates of the two groups are shown in Table
Efficacy and complications.
GOO tailored group | Control group |
| |
---|---|---|---|
Technical success | 96.9% | 96.9% | >0.05 |
Clinical success | 93.8% | 93.5% | >0.05 |
GOOSS change | 3.2 ± 0.5 | 3.1 ± 0.4 | >0.05 |
Ingrowth + overgrowth | 1 | 7 | <0.05 |
Migration | 2 | 0 | >0.05 |
Adverse events | |||
Bleeding | 11 | 2 | <0.05 |
Abdominal pain | 13 | 1 | <0.05 |
Food impaction | 1 | 1 | >0.05 |
Reintervention rate | 9.4% | 22.6% | <0.05 |
Survival (d) | 231 ± 23 | 212 ± 22 | >0.05 |
GOOSS, gastric outlet obstruction score.
Clinical success was determined by resolution of obstructive symptoms and the ability to restart a low residue diet after stent placement. Two patients did not show improvement of obstruction symptoms in each group. There was no statistical difference in terms of symptom improvement. Follow-up contrast studies showed that the stents were completely open in the GOO tailored group, and there was poor expansion of stents in two patients of the control group.
Stent obstruction caused by tumor ingrowth or overgrowth (Figure
Images of stents taken by endoscopy. (a) An example of standard uncovered stent obstruction caused by tumor ingrowth and overgrowth. (b) An example of a funnel-shaped GOO tailored stent that had partly migrated into the stomach.
Bleeding and abdominal pain occurred in the two groups, but the incidence of bleeding and abdominal pain in the tailored group was significantly higher than that in the control group, respectively (Table
During the follow-up period after stent placement, 62 patients died. The mean survival time was
Stenting has been the preferred treatment method for inoperable malignant GOO caused by gastric cancer [
Restenosis even in recently improved stents has been reported to range from 8% to 10.3% [
A limitation of this study was the lack of individualization of stent lengths. Larger studies using stents in which the lengths are also GOO adjusted are planned for the future.
In summary, using covered GOO tailored stents for gastric cancer treatments led to significant less restenosis compared with uncovered standard stents, whereas the migration rate was not significantly different. The reintervention rate was also significantly less in the covered GOO tailored stent group. Mild adverse events like bleeding and abdominal pain were less in the conventional stent group but did not need interventions. Length variation needs further improvement for GOO tailored stents.
The authors declare that there is no conflict of interests regarding the publication of this paper.