Abstract
Background:
Endoscopic gastroduodenal stent (GDS) placement is widely used as a safe and effective method to rapidly improve gastrointestinal symptoms of malignant gastric outlet obstruction (MGOO). While previous studies reported the utility of chemotherapy after GDS placement for prognosis improvement, they did not fully address the issue of immortal time bias.
Objectives:
To examine the association between prognosis and clinical course following endoscopic GDS placement, using a time-dependent analysis.
Design:
Multicenter retrospective cohort study.
Methods:
This study included 216 MGOO patients who underwent GDS placement between April 2010 and August 2020. Data of patient baseline characteristics, including age, gender, cancer type, performance status (PS), GDS type and length, GDS placement location, gastric outlet obstruction scoring system (GOOSS) score, and history of chemotherapy before GDS were collected. The clinical course following GDS placement was evaluated by GOOSS score, stent dysfunction, cholangitis, and chemotherapy. A Cox proportional hazards model was used to identify prognostic factors after GDS placement. Stent dysfunction, post-stent cholangitis, and post-stent chemotherapy were analyzed as time-dependent covariates.
Results:
Mean GOOSS scores before and after GDS were 0.7 and 2.4, respectively, with significant improvement after GDS placement (p < 0.001). The median survival time after GDS placement was 79 [95% confidence interval (CI): 68–103] days. In multivariate Cox proportional hazards model with time-dependent covariates, PS 0–1 [hazard ratio (HR): 0.55, 95% CI: 0.40–0.75; p < 0.001], ascites (HR: 1.45, 95% CI: 1.04–2.01; p = 0.028), metastasis (HR: 1.84, 95% CI: 1.31–2.58; p < 0.001), post-stent cholangitis (HR: 2.38, 95% CI: 1.37–4.15; p = 0.002), and post-stent chemotherapy (HR: 0.01, 95% CI: 0.002–0.10; p < 0.001) significantly affected prognosis after GDS placement.
Conclusion:
Post-stent cholangitis and tolerability to receive chemotherapy after GDS placement influenced prognosis in MGOO patients.
Keywords: chemotherapy, cholangitis, gastric outlet obstruction, prognosis, proportional hazards model, stent
Introduction
Malignant gastric outlet obstruction (MGOO) can be caused by advanced pancreatic cancer, biliary tract cancer, gastric cancer, or lymph node metastasis of cancer. MGOO results in symptoms such as anorexia, vomiting, and swallowing difficulty, worsening the quality of life, and prognosis of patients. It has been approximately 30 years since endoscopic gastroduodenal stent (GDS) for MGOO was developed as a method to alleviate these symptoms.1 GDS has gained popularity as a simple and safe method to manage symptoms of gastrointestinal obstruction, with rapid outcome improvement.2,3 Endoscopic ultrasound-guided gastroenterostomy, a recently developed technique, has been reported to be more effective than GDS and less likely to cause stent dysfunction.4–6 However, it requires advanced endoscopic ultrasound techniques and cannot be performed at all facilities. Therefore, owing to its versatility and ease of execution, endoscopic GDS placement remains the primary treatment for MGOO.
Several factors affecting prognosis after endoscopic GDS placement were previously reported, including the presence of distant metastases7,8 and performance status (PS)9 at treatment time. In addition, there have been several reports on the prognostic impact of chemotherapy following GDS placement. In all of these reports, chemotherapy after GDS placement reportedly improved the prognosis of MGOO patients.7,9–15 However, no study has yet analyzed the effect of chemotherapy as a time-dependent covariate on survival after GDS placement. Furthermore, there are no reports on the prognostic impact of clinical disease course following GDS placement, such as stent dysfunction or complications of cholangitis associated with bile duct obstruction. Investigating the relationship between clinical course and prognosis after GDS placement would be useful for strategizing treatment of MGOO, allowing for better risk assessment and post-treatment outcome prediction.
Therefore, in this multicenter, retrospective study, we examined the clinical course after GDS placement as a time-dependent covariate in addition to baseline characteristics at the time of treatment as prognostic factors after GDS placement for MGOO.
Methods
Patients
For this study, 219 consecutive patients diagnosed with MGOO due to unresectable cancer and who underwent endoscopic GDS placement at the University of Kanazawa Hospital or its 14 affiliated hospitals were enrolled between April 2010 and August 2020. Unresectable cancer diagnosis was confirmed by either pathological or typical radiological findings. Three cases of altered gastroduodenal anatomy except those due to Billroth I reconstruction were excluded from the analysis. Data on patient baseline characteristics [age, gender, cancer type, PS, GDS type and length, GDS placement location, gastric outlet obstruction scoring system (GOOSS) score, and history of chemotherapy before GDS] were collected. We also reviewed the clinical outcome after GDS by evaluating the GOOSS score, complication, cholangitis, biliary drainage, stent dysfunction, and chemotherapy initiation and termination. All patient details were de-identified.
Endoscopic procedure
Informed consent of procedural modalities and risks was collected from each patient. GDS placement was performed under conscious sedation. The stricture was identified endoscopically, and the guidewire was passed through the stenosis. After confirming the location and length of duodenal stenosis using a contrast medium, the duodenal stent was positioned across the stricture under endoscopic and fluoroscopic guidance. Stent type and length were chosen according to stricture site and length. A WallFlex DS (6, 9, or 12 cm in length, 22 mm in body diameter; Boston Scientific, Marlborough, MA, USA), Niti-S DS (covered or uncovered types 6, 8, 10, or 12 cm in length, 22 mm in body diameter; Taewoong Medical, Seoul, Korea), or Evolution DS (6, 9, or 12 cm in length, 22 mm in body diameter; Cook Medical, Winston-Salem, NC, USA) was selected based on the physician’s judgment. Finally, the stent was deployed, and its patency was confirmed by injecting a contrast medium.
Definition of clinical course
Stent dysfunction was defined as the recurrence of MGOO symptoms due to tumor ingrowth and overgrowth, stent migration, and food impaction, as confirmed by radiological and/or endoscopic findings. Post-stent cholangitis was defined by the presence of biliary dilatation confirmed via imaging examination coupled with clinical symptoms and biochemical evidence of cholestasis. Post-stent chemotherapy was defined as systemic chemotherapy application after GDS placement as the first-line as well as the second- or third-line treatments. Chemotherapy was continued until disease progression, occurrence of intolerable adverse events, or the patient refused treatment. Survival after GDS was defined as the date of GDS placement through the date of patient death or last follow-up. Survival after post-stent cholangitis was defined as the time from post-stent cholangitis onset to the date of patient death or last follow-up.
Statistical analysis
Results were expressed as numbers and percentages of patients or as medians and interquartile ranges. Categorical variables were tested using Fisher’s exact test. The Wilcoxon signed-rank test was used to assess improvements in GOOSS score. The Kruskal–Wallis test was used to evaluate GOOSS score according to cancer type. The Kaplan–Meier method was used to evaluate survival time. Factors with p < 0.10 in a univariate Cox proportional hazards model built to evaluate the power of each factor in survival prediction were investigated further using a multivariate model. ‘Post-stent cholangitis’, ‘Post-stent chemotherapy’, and ‘Stent dysfunction’ were entered into the model as time-dependent covariates. ‘Stent dysfunction’ was entered into the model with a value of 0 before and a value of 1 after stent dysfunction occurred. ‘Post-stent chemotherapy’ was entered into the model as 0 before and 1 after chemotherapy administration, and 0 after chemotherapy termination. ‘Post-stent cholangitis’ was included as 0 before and 1 after cholangitis, and 0 if additional drainage was performed. Subgroup analysis was performed for patients who developed post-stent cholangitis or received post-stent chemotherapy. All analyses were performed using the R statistical software, version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria) and EZR package for R, version 1.55.16 The reporting of this study conformed to the Strengthening the Reporting of Observational Studies in Epidemiology Statement (Supplemental Table 1).17
Results
Patient characteristics
Data regarding the characteristics of 216 participants are summarized in Table 1. The primary disease diagnoses were gastric adenocarcinoma, pancreatic adenocarcinoma, biliary tract cancer, and others in 90, 74, 29, and 23 patients, respectively. Median patient age was 76 (68–86) years. A total of 119 patients (55%) had ascites, and 80 patients (37%) had metastasis.
Table 1.
Patient characteristics.
| Characteristics | Value, n (%) or median (IQR) |
|---|---|
| Number of patients | 216 |
| Gender | |
| Male | 110 (51%) |
| Female | 106 (49%) |
| Age (years) | 76 (68–84) |
| PS* | |
| 0–1 | 105 (49%) |
| ⩾2 | 111 (51%) |
| Primary cancer | |
| Gastric | 90 (42%) |
| Pancreas | 74 (34%) |
| Biliary tract | 29 (13%) |
| Other | 23 (11%) |
| Ascites, yes | 119 (55%) |
| Metastasis, yes | 80 (37%) |
| GOOSS score | |
| 0 | 117 (54%) |
| ⩾1 | 99 (46%) |
| Pre-stent biliary stent | 71 (33%) |
| Pre-stent chemotherapy | 79 (37%) |
| Stent | |
| Niti-S | 174 (81%) |
| WallFlex | 39 (18%) |
| Evolution | 3 (1%) |
| Stent length (cm) | 10 (8–12) |
| Location of stent placement | |
| Oral side of the papilla | 133 (62%) |
| Across the papilla | 53 (24%) |
| Anal side of the papilla | 30 (14%) |
PS was determined using the World Health Organization Classification.
GOOSS, gastric outlet obstruction scoring system; IQR, interquartile range; PS, performance status.
Pre-stent biliary stent and GDS placement
Pre-GDS biliary stent was placed in 71 cases. According to cancer type, pre-GDS biliary stents were placed in 3 cases for gastric cancer, 40 cases for pancreatic cancer, 24 cases for biliary cancer, and 4 cases for other cancers, with pancreatic cancer and biliary cancer being significantly more frequently (Supplemental Table 2).
Regarding GDS types, Niti-S was used in 174 cases, Wall Flex in 39 cases, and Evolution in 3 cases. Uncover stent was used in all but one case. The stent was placed on the oral side of the papilla in 133 cases, across the papilla in 53 cases, and on the anorectal side of the papilla in 30 cases (Table 1). The GDSs were placed on the oral side of the papilla in most cases of gastric cancer while they were placed more frequently across the papilla in pancreatic and biliary cancers (Supplemental Table 2).
Clinical course after GDS placement
The mean GOOSS scores before and after GDS placement were 0.7 and 2.4, respectively, with significant improvement after GDS placement (p < 0.001). The median survival after GDS placement was 79 [95% confidence interval (CI): 68–103] days. Stent dysfunction occurred in 37 patients, with a median time to occurrence of 109 (95% CI: 74–139) days. Complications occurred in 53 cases (24.5%).
Clinical outcomes by cancer type are presented in Supplemental Table 3. The median GOOSS score after GDS placement was 3 for all cancer types, with no significant differences among cancer types (p = 0.824). The median survival after GDS placement was 115 days for gastric cancer, 71.5 days for pancreatic cancer, 61 days for biliary cancer, and 48 days for other cancers, with no significant differences among cancer types (p = 0.18). Stent dysfunction occurred in 16 patients with gastric cancer, 14 patients with pancreatic cancer, 5 patients with biliary cancer, and 2 patients with other cancers, with no significant difference among cancer types (p = 0.77). Complications occurred in 12 cases of gastric cancer, 22 cases of pancreatic cancer, 14 cases of biliary tract cancer, and 6 cases of other cancers, with significant differences among cancer types (p = 0.001).
Post-stent cholangitis
Post-stent cholangitis occurred in 42 patients, with a median time to occurrence of 12.5 (95% CI: 5–19) days. Post-stent cholangitis was the most frequent complication in all cancer types. Post-stent cholangitis was more frequent in biliary and pancreatic cancers (Supplemental Table 3). The most common cause of post-stent cholangitis in all cancer types was bile duct obstruction due to tumor progression. There was no difference in the cause of post-stent cholangitis among cancer types (Supplemental Table 4). Drainage for cholangitis was performed in 19 cases, including self-expandable metallic stent placement in 8 cases, plastic stent placement in 5 cases, percutaneous transhepatic cholangio drainage in 4 cases, endoscopic ultrasound-guided biliary drainage in 1 case, and endoscopic nasobiliary drainage in 1 case.
Chemotherapy
Chemotherapy was administered in 79 cases before GDS. In all, 38 patients underwent chemotherapy after GDS, with a median start time to chemotherapy of 14.5 (1–70) days and a median duration of chemotherapy administration of 133 (6–638) days. Detailed administered regimens are presented in Supplemental Table 5. Cisplatin (CDDP) + tegaful/gimeracil/oteracil was administered as the first-line regimen in 11 gastric cancer cases, gemcitabine (GEM) in 18 pancreatic cancer cases, and GEM + CDDP in 10 biliary cancer cases, and these treatments were most frequently administered in each cancer type.
Predictors for survival after GDS placement
In univariate analysis, age >76 years, gastric cancer, PS 0–1, ascites, metastasis, Niti-S stent, stent position across the papilla, post-stent chemotherapy, post-stent cholangitis, and stent dysfunction were identified as predictors of survival after GDS placement (p < 0.1). Using these factors in the multivariate analysis revealed that PS 0–1 (hazard ratio [HR]: 0.55, 95% CI: 0.40–0.75; p < 0.001), ascites (HR: 1.45, 95% CI: 1.04–2.01; p = 0.028), metastasis (HR: 1.84, 95% CI: 1.31–2.58; p < 0.001), post-stent cholangitis (HR: 2.38, 95% CI: 1.37–4.15; p = 0.002), and post-stent chemotherapy (HR: 0.01, 95% CI: 0.002–0.01; p < 0.001) significantly affected survival after GDS (Table 2).
Table 2.
Cox proportional hazards models of univariate and multivariate analyses to identify prognostic factors after GDS placement.
| Characteristics | Univariate | Multivariate | ||
|---|---|---|---|---|
| HR (95% CI) | p Value | HR (95% CI) | p Value | |
| Age ⩾76 years | 1.34 (0.99–1.80) | 0.053 | 0.78 (0.56–1.09) | 0.15 |
| Gender, male | 0.99 (0.74–1.32) | 0.92 | ||
| PS, 0–1 | 0.55 (0.41–0.74) | <0.001 | 0.55 (0.40–0.75) | <0.001 |
| Primary cancer, gastric | 0.70 (0.52–0.94) | 0.019 | 0.98 (0.69–1.40) | 0.93 |
| Ascites, yes | 1.35 (1.00–1.81) | 0.049 | 1.45 (1.04–2.01) | 0.028 |
| Metastasis, yes | 1.73 (1.28–2.33) | <0.001 | 1.84 (1.31–2.58) | <0.001 |
| GOOSS score before GDS placement ⩾ 1 | 0.96 (0.71–1.28) | 0.77 | ||
| Pre-stent biliary stent | 1.22 (0.90–1.65) | 0.19 | ||
| Post-stent cholangitis* | 3.63 (2.38–5.54) | <0.001 | 2.38 (1.37–4.15) | 0.002 |
| Pre-stent chemotherapy | 1.22 (0.91–1.65) | 0.19 | ||
| Post-stent chemotherapy* | 0.029 (0.01–0.11) | <0.001 | 0.01 (0.002–0.10) | <0.001 |
| Stent dysfunction* | 1.52 (1.07–2.17) | 0.020 | 1.17 (0.72–1.90) | 0.53 |
| Stent, Niti-S | 1.54 (1.06–2.24) | 0.024 | 1.27 (0.92–1.79) | 0.15 |
| Stent length ⩾10 cm | 1.13 (0.84–1.51) | 0.43 | ||
| Stent location, across the papilla | 1.55 (1.11–2.14) | 0.009 | 1.16 (0.81–1.66) | 0.43 |
‘Post-stent cholangitis’, ‘Post-stent chemotherapy’, and ‘Stent dysfunction’ were entered into the model as time-dependent covariates.
CI, confidence interval; GDS, gastroduodenal stent; GOOSS, gastric outlet obstruction scoring system; HR, hazard ratio; PS, performance status.
Subgroup analysis in cases of post-stent cholangitis and post-stent chemotherapy
The median survival time after post-stent cholangitis was 38 days (95% CI: 19–59) in 42 cases of post-stent cholangitis. In univariate analysis, age >76 years and biliary drainage after cholangitis were potential predictors of prognosis after post-stent cholangitis (p < 0.1). In multivariate analysis, biliary drainage after cholangitis was significantly associated with prognosis after post-stent cholangitis (HR: 0.32, 95% CI: 0.14–0.73; p = 0.007) (Table 3).
Table 3.
Cox proportional hazards models of univariate and multivariate analyses to identify factors associated with prognosis in cases of cholangitis after GDS placement.
| Characteristics | Univariate | Multivariate | ||
|---|---|---|---|---|
| HR (95% CI) | p Value | HR (95% CI) | p Value | |
| Age ⩾76 years | 1.99 (1.04–3.82) | 0.038 | 1.19 (0.57–2.51) | 0.64 |
| Gender, male | 0.64 (0.34–1.22) | 0.18 | ||
| PS, 0–1 | 0.66 (0.35–1.26) | 0.21 | ||
| Primary cancer, gastric | 1.36 (0.59–3.12) | 0.47 | ||
| Ascites, yes | 1.27 (0.63–2.55) | 0.51 | ||
| Metastasis, yes | 0.85 (0.43–1.71) | 0.66 | ||
| GOOSS score before GDS ⩾ 1 | 0.70 (0.35–1.39) | 0.31 | ||
| Pre-stent biliary stent | 1.53 (0.78–3.01) | 0.21 | ||
| Pre-stent chemotherapy | 1.13 (0.58–2.21) | 0.72 | ||
| Stent, Niti-S | 0.57 (0.20–1.63) | 0.29 | ||
| Stent length ⩾10 cm | 1.37 (0.70–2.67) | 0.36 | ||
| Stent location, across the papilla | 0.84 (0.44–1.60) | 0.59 | ||
| Drainage for post-stent cholangitis* | 0.28 (0.13–0.61) | 0.002 | 0.32 (0.14–0.73) | 0.007 |
Drainage for post-stent cholangitis was entered into the model as a time-dependent covariate.
CI, confidence interval; GDS, gastroduodenal stent; GOOSS, gastric outlet obstruction scoring system; HR, hazard ratio; PS, performance status.
In univariate analysis, metastasis, pre-stent chemotherapy, and stent position across the papilla were potential risk factors for survival after GDS (p < 0.1) in 38 patients who underwent post-stent chemotherapy. In multivariate analysis, metastasis (HR: 2.67, 95% CI: 1.15–6.22; p = 0.022) and pre-stent chemotherapy (HR: 2.38, 95% CI: 1.05–5.37; p = 0.038) were significantly associated with survival after GDS placement (Table 4).
Table 4.
Cox proportional hazards models of univariate and multivariate analyses to identify factors associated with prognosis in patients undergoing chemotherapy after GDS placement.
| Characteristics | Univariate | Multivariate | ||
|---|---|---|---|---|
| HR (95% CI) | p Value | HR (95% CI) | p Value | |
| Age ⩾76 years | 0.82 (0.31–2.16) | 0.69 | ||
| Gender, male | 0.63 (0.30–1.32) | 0.22 | ||
| PS, 0–1 | 1.36 (0.60–3.06) | 0.46 | ||
| Primary cancer, gastric | 0.56 (0.26–1.21) | 0.14 | ||
| Ascites, yes | 0.75 (0.35–1.60) | 0.46 | ||
| Metastasis, yes | 1.94 (0.88–4.24) | 0.099 | 2.67 (1.15–6.22) | 0.022 |
| GOOSS score before GDS ⩾ 1 | 0.57 (0.28–1.19) | 0.13 | ||
| Pre-stent biliary stent | 1.32 (0.62–2.82) | 0.47 | ||
| Pre-stent chemotherapy | 2.44 (1.11–5.38) | 0.027 | 2.38 (1.05–5.37) | 0.038 |
| Stent, Niti-S | 1.24 (0.55–2.79) | 0.61 | ||
| Stent length ⩾10 cm | 0.90 (0.43–1.87) | 0.77 | ||
| Stent location, across the papilla | 2.16 (0.97–4.80) | 0.059 | 2.36 (0.99–5.64) | 0.053 |
CI, confidence interval; GDS, gastroduodenal stent; GOOSS, gastric outlet obstruction scoring system; HR, hazard ratio; PS, performance status.
Discussion
In this multicenter, retrospective cohort study, we investigated factors influencing clinical outcomes after GDS placement in cancer patients with MGOO, encompassing the initial clinical factors at the time of GDS placement as well as the influence of the subsequent clinical events after GDS placement on prognosis, in a time-dependent analysis. Our findings showed that in addition to disease progression (ascites, presence of distant metastases) and PS at the time of GDS placement, post-stent chemotherapy and post-stent cholangitis also affected prognosis after GDS placement.
In our study, the median survival time after GDS placement for MGOO was 79 days, which was consistent with the finding from a previous report.18 The relatively short prognosis in this study may be due to the inclusion of patients in poor general condition and with highly advanced disease, which reflects the real-world clinical situation in which endoscopic GDS placement is performed. In addition, uncover stents were used in almost all cases except one, and only four cases were complicated by stent migration in this study. In cases with severe prognosis, the advantage of using uncovered stents with a low risk of migration was considered to be greater than the disadvantage of stent dysfunction due to tumor progression, which also reflects the real-world clinical practice.
In some studies, gastric cancer reportedly had a better prognosis after stenting as compared to the other cancers.12,14 However, one study found no difference between the prognosis of pancreatic and gastric cancer post-stenting.7 We did not identify any difference in prognosis between gastric and non-gastric cancers, nor between pancreatic and non-pancreatic cancers. Furthermore, consistent with the findings of prior reports, factors related to tumor progression, such as the presence of ascites and distant metastases, as well as PS, were observed to be significantly associated with prognosis.7,13,19 These results suggested that in cases of highly advanced cancer-caused MGOO, tumor progression and systemic status may be more important than cancer type.
Immortal bias, commonly encountered in cohort studies examining the effects of specific drugs/treatments, is considered problematic because it can skew the results in a favorable direction.20–23 Previous retrospective studies investigating post-stent chemotherapy after GDS placement reported improved survival outcomes.7,9–15 However, none of these reports performed a time-dependent analysis, which might have accentuated the impact of chemotherapy on survival due to an immortal time bias. Time-dependent exposures, such as post-stent chemotherapy, have been suggested to be included as time-dependent variables in hazard-based analyses to minimize the risk of immortal bias in observational studies of survival outcomes.22 To our best knowledge, the current study is the first to use a time-dependent analysis to evaluate the association between post-stent chemotherapy and prognosis after GDS placement. Interestingly, in this time-dependent analysis with minimized immortal time bias, post-stent chemotherapy was associated with a better prognosis, as previously reported. It is well known that chemotherapy prolongs prognosis in patients with inoperable advanced cancer, even after GDS placement. In the analysis limited to post-GDS chemotherapy cases, metastasis and pre-GDS chemotherapy were associated with prognosis, which confirmed that tumor progression may have an impact on prognosis. The prognosis for patients who received chemotherapy before GDS might appear relatively poor because the cancer was already highly advanced at the time of diagnosis and more time had passed before GDS placement. Furthermore, cholangitis after GDS might also be associated with prognosis in this study. In this regard, patients with cholangitis after GDS might have originally been expected to have a severe prognosis because of the highly advanced condition in which biliary stricture was present along with gastroduodenal stenosis. Since cholangitis is a fatal condition, the observed effect of drainage on prognosis seems appropriate. In recent years, endoscopic ultrasound-guided biliary drainage using an echoendoscope and biliary stent placement have been used for obstructive cholangitis after duodenal stent placement,24 and drainage is performed aggressively unless the patient’s general condition deteriorates to the extent that endoscopic treatment is difficult.
The recently developed technique of endoscopic ultrasound-guided gastroenterostomy has been reported to be highly effective and safe, rarely requiring re-treatment.4–6 However, since this procedure is still not routinely performed in many centers, endoscopic GDS placement, which is conducted in most healthcare facilities, remains the most viable treatment option for MGOO. Future prospective studies are warranted to evaluate whether endoscopic GDS placement, surgery, or endoscopic ultrasound-guided gastroenterostomy is the better treatment for MGOO for various clinical conditions, such as cancer type, obstruction site, general condition, and schedule of aggressive treatment.
This study had several limitations. Although it was a retrospective study and time-dependent analysis was used to reduce bias, bias could not be completely eliminated. In addition, because of the long observation period in this study, the results were affected by ongoing changes in chemotherapy regimens. Furthermore, a variety of carcinomas and chemotherapy regimens were included in the analysis, and chemotherapy regimens as well as drainage methods for cholangitis were selected at the physician’s discretion. Therefore, no specific conclusions or advice can be extrapolated from our results of this retrospective and heterogeneous cohort. In routine clinical practice, all treatment options should depend on specific needs and clinical and anatomic status of each patient.
In summary, we performed a multicenter retrospective cohort study using time-dependent analysis to examine whether post-stent clinical course and clinical parameters at GDS placement time influence prognosis after GDS placement in MGOO patients. PS, ascites, and distant metastasis at GDS placement as well as post-stent chemotherapy and post-stent cholangitis during the clinical course after GDS placement were associated with prognosis.
Supplemental Material
Supplemental material, sj-docx-1-tag-10.1177_17562848231156279 for Prognostic impact of clinical outcome after endoscopic gastroduodenal stent placement for malignant gastric outlet obstruction: a multicenter retrospective cohort study using a time-dependent analysis by Yuji Hodo, Hajime Takatori, Takuya Komura, Yoshiro Asahina, Tomoyoshi Chiba, Hisashi Takabatake, Kenkei Hasatani, Ryuhei Nishino, Masaaki Yano, Yoshiko Takata, Tsuyoshi Suda, Haruhiko Shugo, Hiroyoshi Nakanishi, Kazutoshi Yamada, Kazuhiro Miwa, Kiichiro Kaji, Masaki Nishitani, Masaki Miyazawa and Taro Yamashita in Therapeutic Advances in Gastroenterology
Acknowledgments
We would like to thank Editage (www.editage.com) for English language editing.
Footnotes
ORCID iDs: Yuji Hodo 
https://orcid.org/0000-0003-4597-8750
Hajime Takatori
https://orcid.org/0000-0002-6680-2925
Supplemental material: Supplemental material for this article is available online.
Contributor Information
Yuji Hodo, Department of Gastroenterology, Saiseikai Kanazawa Hospital, Kanazawa, Ishikawa, Japan.
Hajime Takatori, Department of Endoscopy, Kanazawa University Hospital, 13-1 Takaramachi, Kanazawa, Ishikawa 921-8641, Japan.
Takuya Komura, Department of Gastroenterology, National Hospital Organization Kanazawa Medical Center, Kanazawa, Ishikawa, Japan.
Yoshiro Asahina, Department of Gastroenterology, National Hospital Organization Kanazawa Medical Center, Kanazawa, Ishikawa, Japan.
Tomoyoshi Chiba, Department of Gastroenterology, Public Central Hospital of Matto Ishikawa, Hakusan, Ishikawa, Japan.
Hisashi Takabatake, Department of Gastroenterology, Kanazawa Red Cross Hospital, Kanazawa, Ishikawa, Japan.
Kenkei Hasatani, Department of Gastroenterology, Fukui Prefectural Hospital, Fukui, Fukui, Japan.
Ryuhei Nishino, Department of Gastroenterology, Hakui General Hospital, Hakui, Ishikawa, Japan.
Masaaki Yano, Department of Gastroenterology, Toyama Prefectural Central Hospital, Toyama, Toyama, Japan.
Yoshiko Takata, Department of Gastroenterology, Tonami General Hospital, Tonami, Toyama, Japan.
Tsuyoshi Suda, Department of Gastroenterology, Kanazawa Municipal Hospital, Kanazawa, Ishikawa, Japan.
Haruhiko Shugo, Department of Gastroenterology, Keiju Medical Center, Nanao, Ishikawa, Japan.
Hiroyoshi Nakanishi, Department of Gastroenterology, Ishikawa Prefectural Central Hospital, Kanazawa, Ishikawa, Japan.
Kazutoshi Yamada, Department of Gastroenterology, Toyama City Hospital, Toyama, Toyama, Japan.
Kazuhiro Miwa, Department of Gastroenterology, Komatsu Municipal Hospital, Komatsu, Ishikawa, Japan.
Kiichiro Kaji, Department of Gastroenterology, Komatsu Sophia Hospital, Komatsu, Ishikawa, Japan.
Masaki Nishitani, Department of Gastroenterology, Kanazawa University Hospital, Kanazawa, Ishikawa, Japan.
Masaki Miyazawa, Department of Gastroenterology, Kanazawa University Hospital, Kanazawa, Ishikawa, Japan.
Taro Yamashita, Department of Gastroenterology, Kanazawa University Hospital, Kanazawa, Ishikawa, Japan.
Declarations
Ethics approval and consent to participate: The study was approved by the Medical Ethics Committee of Kanazawa University (2016-106; 7 March, 2017) and all affiliated hospitals, and written informed consent requirement was waived due to its retrospective nature.
Consent for publication: Not applicable.
Author contribution(s): Yuji Hodo: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing – original draft.
Hajime Takatori: Conceptualization; Data curation; Investigation; Project administration; Supervision; Writing – review & editing.
Takuya Komura: Data curation; Investigation; Writing – review & editing.
Yoshiro Asahina: Data curation; Investigation; Writing – review & editing.
Tomoyoshi Chiba: Data curation; Investigation; Writing – review & editing.
Hisashi Takabatake: Data curation; Investigation; Writing – review & editing.
Kenkei Hasatani: Data curation; Investigation; Writing – review & editing.
Ryuhei Nishino: Data curation; Investigation; Writing – review & editing.
Masaaki Yano: Data curation; Investigation; Writing – review & editing.
Yoshiko Takata: Data curation; Investigation; Writing – review & editing.
Tsuyoshi Suda: Data curation; Investigation; Writing – review & editing.
Haruhiko Shugo: Data curation; Investigation; Writing – review & editing.
Hiroyoshi Nakanishi: Data curation; Investigation; Writing – review & editing.
Kazutoshi Yamada: Data curation; Investigation; Writing – review & editing.
Kazuhiro Miwa: Data curation; Investigation; Writing – review & editing.
Kiichiro Kaji: Data curation; Investigation; Writing – review & editing.
Masaki Nishitani: Data curation; Investigation; Writing – review & editing.
Masaki Miyazawa: Data curation; Investigation; Writing – review & editing.
Taro Yamashita: Conceptualization; Supervision; Writing – review & editing.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
The authors declare that there is no conflict of interest.
Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
- 1. Topazian M, Ring E, Grendell J. Palliation of obstructing gastric cancer with steel mesh, self-expanding endoprostheses. Gastrointest Endosc 1992; 38: 58–60. [DOI] [PubMed] [Google Scholar]
- 2. Yim HB, Jacobson BC, Saltzman JR, et al. Clinical outcome of the use of enteral stents for palliation of patients with malignant upper GI obstruction. Gastrointest Endosc 2001; 53: 329–332. [DOI] [PubMed] [Google Scholar]
- 3. No JH, Kim SW, Lim CH, et al. Long-term outcome of palliative therapy for gastric outlet obstruction caused by unresectable gastric cancer in patients with good performance status: endoscopic stenting versus surgery. Gastrointest Endosc 2013; 78: 55–62. [DOI] [PubMed] [Google Scholar]
- 4. van Wanrooij RLJ, Vanella G, Bronswijk M, et al. Endoscopic ultrasound-guided gastroenterostomy versus duodenal stenting for malignant gastric outlet obstruction: an international, multicenter, propensity score-matched comparison. Endoscopy 2022; 54: 1023–1031. [DOI] [PubMed] [Google Scholar]
- 5. Krishnamoorthi R, Bomman S, Benias P, et al. Efficacy and safety of endoscopic duodenal stent versus endoscopic or surgical gastrojejunostomy to treat malignant gastric outlet obstruction: systematic review and meta-analysis. Endosc Int Open 2022; 10: E874–E897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sánchez-aldehuelo R, Carlos J, Iñigo S, et al. EUS-guided gastroenterostomy versus duodenal self-expandable metal stent for malignant gastric outlet obstruction: results from a nationwide multicenter retrospective study. Gastrointest Endosc 2022; 96: 1012–1020.e3. [DOI] [PubMed] [Google Scholar]
- 7. Oh SY, Edwards A, Mandelson M, et al. Survival and clinical outcome after endoscopic duodenal stent placement for malignant gastric outlet obstruction: comparison of pancreatic cancer and nonpancreatic cancer. Gastrointest Endosc 2015; 82: 460–468.e2. [DOI] [PubMed] [Google Scholar]
- 8. Sugiura T, Okamura Y, Ito T, et al. Prognostic scoring system for patients who present with a gastric outlet obstruction caused by advanced pancreatic adenocarcinoma. World J Surg 2017; 41: 2619–2624. [DOI] [PubMed] [Google Scholar]
- 9. Kobayashi S, Ueno M, Kameda R, et al. Duodenal stenting followed by systemic chemotherapy for patients with pancreatic cancer and gastric outlet obstruction. Pancreatology 2016; 16: 1085–1091. [DOI] [PubMed] [Google Scholar]
- 10. Miyabe K, Hayashi K, Nakazawa T, et al. Safety and benefits of self-expandable metallic stents with chemotherapy for malignant gastric outlet obstruction. Dig Endosc 2015; 27: 572–581. [DOI] [PubMed] [Google Scholar]
- 11. Takamatsu Y, Fujimori N, Miyagahara T, et al. The Glasgow Prognostic Score and stricture site can predict prognosis after endoscopic duodenal stent placement for malignant gastric outlet obstruction. Sci Rep 2022; 12: 9746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ye BW, Chou CK, Hsieh YC, et al. Metallic stent expansion rate at day one predicts stent patency in patients with gastric outlet obstruction. Dig Dis Sci 2017; 62: 1286–1294. [DOI] [PubMed] [Google Scholar]
- 13. Uemura S, Iwashita T, Iwata K, et al. Endoscopic duodenal stent versus surgical gastrojejunostomy for gastric outlet obstruction in patients with advanced pancreatic cancer. Pancreatology 2018; 18: 601–607. [DOI] [PubMed] [Google Scholar]
- 14. Ye BW, Lee KC, Hsieh YC, et al. Self-expandable metallic stent placement in malignant gastric outlet obstruction. Medicine (Baltimore) 2015; 94: e1208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wei TH, Ye BW, Wu PS, et al. Outcomes of patients with malignant duodenal obstruction after receiving self-expandable metallic stents: a single center experience. PLoS One 2022; 17: e0268920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’’ for medical statistics. Bone Marrow Transplant 2013; 48: 452–458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. PLoS Med 2007; 4: 1623–1627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Van Halsema EE, Rauws EAJ, Fockens P, et al. Self-expandable metal stents for malignant gastric outlet obstruction: a pooled analysis of prospective literature. World J Gastroenterol 2015; 21: 12468–12481. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Park CH, Park JC, Kim EH, et al. Impact of carcinomatosis and ascites status on long-term outcomes of palliative treatment for patients with gastric outlet obstruction caused by unresectable gastric cancer: stent placement versus palliative gastrojejunostomy. Gastrointest Endosc 2015; 81: 321–332. [DOI] [PubMed] [Google Scholar]
- 20. Lévesque LE, Hanley JA, Kezouh A, et al. Problem of immortal time bias in cohort studies: example using statins for preventing progression of diabetes. BMJ 2010; 340: 907–911. [DOI] [PubMed] [Google Scholar]
- 21. Targownik LE, Suissa S. Understanding and avoiding immortal-time bias in gastrointestinal observational research. Am J Gastroenterol 2015; 110: 1647–1650. [DOI] [PubMed] [Google Scholar]
- 22. Jones M, Fowler R. Immortal time bias in observational studies of time-to-event outcomes. J Crit Care 2016; 36: 195–199. [DOI] [PubMed] [Google Scholar]
- 23. Yadav K, Lewis RJ. Immortal time bias in observational studies. JAMA 2021; 325: 686–687. [DOI] [PubMed] [Google Scholar]
- 24. Tonozuka R, Itoi T, Sofuni A, et al. Endoscopic double stenting for the treatment of malignant biliary and duodenal obstruction due to pancreatic cancer. Dig Endosc 2013; 25: 100–108. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-tag-10.1177_17562848231156279 for Prognostic impact of clinical outcome after endoscopic gastroduodenal stent placement for malignant gastric outlet obstruction: a multicenter retrospective cohort study using a time-dependent analysis by Yuji Hodo, Hajime Takatori, Takuya Komura, Yoshiro Asahina, Tomoyoshi Chiba, Hisashi Takabatake, Kenkei Hasatani, Ryuhei Nishino, Masaaki Yano, Yoshiko Takata, Tsuyoshi Suda, Haruhiko Shugo, Hiroyoshi Nakanishi, Kazutoshi Yamada, Kazuhiro Miwa, Kiichiro Kaji, Masaki Nishitani, Masaki Miyazawa and Taro Yamashita in Therapeutic Advances in Gastroenterology