Radioactive Stent for Malignant Esophageal Obstruction: A Meta-Analysis of Randomized Controlled Trials

Abstract

Purpose:

To compare the relative clinical efficacies of radioactive and normal stent insertion methods as a means of treating patients suffering from malignant esophageal obstruction (MEO).

Materials and Methods:

The Pubmed, Embase, and Cochrane Library databases were searched for relevant randomized controlled trials (RCTs) from the date of inception through to July 2020. RevMan v5.3 was used for all data analyses.

Results:

This meta-analysis included six RCTs that included a total of 194 patients who had undergone radioactive stent insertion and 209 who had normal stent insertion. There were no significant differences in pooled improvement of dysphagia scores (P = .40), rates of stent restenosis (24.7% versus 28.7%, P = .35), stent migration (3.3% versus 4.4%, P = .61), severe chest pain (22.8% versus 20.3%, P = .61), hemorrhage (11.0% versus 9.8%, P = .80), or fistula formation (6.1% versus 4.2%, P = .55) between two groups. The pooled time to restenosis (P < .00001) and survival (P < .00001) were significant longer in the radioactive stent group. Significant heterogeneity was detected in the endpoint of improvement of dysphagia score (I² = 89%; P = .0002). Funnel plot analyses did not detect any evidence of publication bias pertaining to the selected study endpoints.

Conclusions:

Our meta-analysis demonstrated that radioactive stent insertion can prolong stent patency and survival for patients with MEO compared with normal stent insertion.

Introduction

Malignant esophageal obstruction (MEO) is usually secondary to esophageal and other chest cancers,^1–4 and >80% of cases with MEO were found to be caused by esophageal cancer.² In one study, when patients were diagnosed with MEO, >80% of the cases missed the chance for curative resection.⁵ In addition, patients with MEO had a poor quality of life during their limited survival time because of dysphagia.

Stent insertion is typically employed as a first-line palliative treatment in patients with incurable MEO.^1–4 Normal stent insertion, however, does not directly affect the primary tumor. Thus, the prolongation of stent patency and overall survival time typically necessitates further intervention with anticancer treatments.^6,7 To combine the treatment effects of radiotherapy and stent insertion on MEO, researchers have developed a form of radioactive stent loaded with ¹²⁵I seeds.^8–10 Although many studies regarding of esophageal radioactive stents have been published, the number of randomized controlled trials (RCTs) are limited.^11–16

To come to a definite conclusion on the effectiveness of the esophageal radioactive stent, we conducted a meta-analysis of RCTs to compare the relative clinical efficacies of radioactive and normal stent insertion methods for the treatment of patients suffering from MEO.

Materials and Methods

Study selection

This meta-analysis was approved by our institutional review board. The Pubmed, Embase, and Cochrane Library databases were searched for relevant studies from the date of inception to July 2020 using the following search strategy: (((stent[Title/Abstract]) OR (SEMS[Title/Abstract])) AND ((((irradiation[Title/Abstract]) OR (seed[Title/Abstract])) OR (radioactive[Title/Abstract])) OR (iodine[Title/Abstract]))) AND ((esophageal[Title/Abstract]) OR (esophagus[Title/Abstract])).

The following study inclusion criteria were used:

(a) Type of study: RCTs

(b) Disease: MEO

(d) Language: any.

Studies were excluded if they were any of the following:

(a) Noncomparative studies

(b) Non-RCT studies

(d) Review articles.

Data extraction

Two investigators independently extracted the data from all identified articles, with the corresponding author helped to resolve any inconsistencies in the data. The extracted items included three data types: study baseline data, patient baseline data, and treatment-associated data.

Quality and bias assessment

The eight-point Jadad composite scale was used to gauge the quality of all the included RCTs,¹⁷ and RCTs with scores ≥4 points were considered to be high quality. Potential biases were assessed using the Cochrane risk of bias tool.

Endpoints

The primary endpoints for this meta-analysis were stent restenosis and survival, whereas the secondary endpoints included improvement of dysphagia score, stent migration, severe chest pain, hemorrhage, and fistula formation.

Statistical analyses

The software RevMan v5.3 was used for all data analyses. The Mantel–Haenszel method was used to measure pooled odds ratios (ORs) and 95% confidence intervals (CIs) for dichotomous variables, whereas continuous variables were analyzed based on the mean difference (MD) and 95% CIs. We used the hazard ratio (HR) with a 95% CI to measure the pooled survival times. Study heterogeneity was assessed through χ² and I2 tests, with I² > 50% being indicative of significant heterogeneity. Random-effects models were used in the presence of significant heterogeneity, whereas fixed-effects models were used for analyses when significant heterogeneity was not detected. The causes of heterogeneity were evaluated through sensitivity and subgroup analyses, and funnel plots were used to assess the risk of bias.

Results

Study characteristics

Our initial search strategy identified 154 potentially relevant studies, and 6 RCTs were ultimately included in our meta-analysis (Fig. 1). The Jadad composite scale of the six RCTs ranged from 2 to 5 (Table 1). Three RCTs had a high risk of random sequence generation and allocation concealment,^11,14,15 and none of the RCTs were double-blind studies (Fig. 2). Fluoroscopic-guided stent insertion was used in five studies,^{11,12,14–16} and endoscopic-guided stent insertion was used in one study.¹³ The six RCTs involved a total of 194 patients who had undergone radioactive stent insertion and 209 who had undergone normal stent insertion (Table 1). The outcome data are shown in Table 2.

FIG. 1.

The flowchart of this study.

FIG. 2.

Cochrane's risk of bias assessment tool.

Table 1.

Characteristics of the Included Studies

Study/year/country	Approaches	Cancer types	Tumor stage	Stent diameter (mm)	Groups	Sample size	Age (years)	Jadad scale
Dai et al./2013/China¹¹	Endoscopic	Multiple	Not given	Not given	Radioactive	31	68	2
Dai et al./2013/China¹¹	Endoscopic	Multiple	Not given	Not given	Normal	36	71	2
Guo et al./2008/China¹²	Fluoroscopic	Multiple	Not given	18–20	Radioactive	27	72	5
Guo et al./2008/China¹²	Fluoroscopic	Multiple	Not given	18–20	Normal	26	70	5
Lu et al./2014/China¹³	Endoscopic	Multiple	Not given	Not given	Radioactive	30	55	4
Lu et al./2014/China¹³	Endoscopic	Multiple	Not given	Not given	Normal	30	57	4
Xu et al./2011/China¹⁴	Fluoroscopic	Multiple	Not given	18–20	Radioactive	15	63 for all	2
Xu et al./2011/China¹⁴	Fluoroscopic	Multiple	Not given	18–20	Normal	17	63 for all	2
Zhao et al./2016/China¹⁵	Fluoroscopic	Multiple	III, IV	18–20	Radioactive	18	70 for all	2
Zhao et al./2016/China¹⁵	Fluoroscopic	Multiple	III, IV	18–20	Normal	25	70 for all	2
Zhu et al./2014/China¹⁶	Fluoroscopic	Multiple	II–IV	Not given	Radioactive	73	71	5
Zhu et al./2014/China¹⁶	Fluoroscopic	Multiple	II–IV	Not given	Normal	75	71	5

Table 2.

Characteristics of the Treatment Outcomes

Study	Groups	Restenosis	Migration	Severe chest pain	Hemorrhage	Fistula formation
Dai et al.¹¹	Radioactive	11/31 (35.5%)	Not given	Not given	Not given	Not given
Dai et al.¹¹	Normal	16/36 (44.4%)	Not given	Not given	Not given	Not given
Guo et al.¹²	Radioactive	8/27 (29.6%)	2/27 (7.4%)	8/27 (29.6%)	9/27 (33.3%)	1/27 (3.7%)
Guo et al.¹²	Normal	6/26 (23.1%)	3/26 (11.5%)	7/26 (26.9%)	7/26 (26.9%)	0/26 (0%)
Lu et al.¹³	Radioactive	3/30 (10%)	Not given	5/30 (16.6%)	0/30 (0%)	Not given
Lu et al.¹³	Normal	8/30 (26.7%)	Not given	4/30 (13.3%)	1/30 (3.3%)	Not given
Xu et al.¹⁴	Radioactive	3/15 (20%)	0/15 (0%)	3/15 (20%)	4/15 (26.7%)	0/15 (0%)
Xu et al.¹⁴	Normal	7/17 (41.2%)	0/17 (0%)	4/17 (23.5%)	4/17 (23.5%)	0/17 (0%)
Zhao et al.¹⁵	Radioactive	2/18 (11.1%)	0/18 (0%)	Not given	0/18 (0%)	Not given
Zhao et al.¹⁵	Normal	3/25 (12%)	0/25 (0%)	Not given	0/25 (0%)	Not given
Zhu et al.¹⁶	Radioactive	21/73 (28.8%)	Not given	17/73 (23.3%)	5/73 (6.8%)	6/73 (8.2%)
Zhu et al.¹⁶	Normal	20/75 (26.7%)	Not given	15/75 (20%)	5/75 (6.7%)	5/75 (6.7%)

The stent diameter data were reported in three articles and had values of 18–20 mm.^12,14,15 The lengths of the stents were chosen based on the obstructed length (obstructed length plus 2 cm at both ends of the tumor).^11–16

Dysphagia score

Two of the study reports provided data for the improvement of dysphagia scores,^12,16 and we found that the two groups exhibited similar Δdysphagia scores (MD: 0.01; 95% CI: −0.38, P = .40, Table 3). We detected significant dysphagia score heterogeneity between the two studies (I² = 89%; P = .0002).

Table 3.

Meta-Analytic Pooled Results of the Secondary Endpoints

	Number of studies	OR/MD (95% CI)	Heterogeneity
Dysphagia score	2 [12, 14]	0.01 (−0.38 to 0.40), P = .96	I² = 89%
Migration	3 [12, 14, 15]	0.61 (0.09–4.01), P = .61	Not applicable
Severe chest pain	4 [12–14, 16]	1.16 (0.66–2.03), P = .61	I² = 0%
Hemorrhage	5 [12–16]	1.10 (0.53–2.29), P = .80	I² = 0%
Fistula formation	3 [12, 14, 16]	1.42 (0.45–4.45), P = .55	I² = 0%

CI, confidence interval; MD mean difference; OR, odds ratio.

Stent restenosis

The data of stent restenosis rates were extracted from all studies. We found that there were comparable pooled stent restenosis rates between the two groups (24.7% versus 28.7%, OR: 0.81; 95% CI: 0.51–1.26; P = .35, Fig. 3), and no significant heterogeneity was apparent among the six studies (I² = 0%; P = .49).

FIG. 3.

The pooled stent restenosis rates were comparable between the two groups.

Time to restenosis

We were able to extract data for the time to stent restenosis from two reports,^12,14 and the pooled time to stent restenosis was significantly longer in the radioactive stent group (MD: 2.53; 95% CI: 1.95–3.11, P < .00001, Fig. 4). We found no significant heterogeneity among these studies (I² = 0%; P = .76).

FIG. 4.

The pooled time to stent restenosis was significant longer in radioactive stent group.

Stent migration

The data for the stent migration rates were available in three of the study reports,^12,14,15 and there were comparable pooled stent migration rates between the two stent groups (3.3% versus 4.4%, OR: 0.61; 95% CI: 0.09–4.01; P = .61, Table 3). Two of the studies^14,15 found that both the radioactive and normal groups had 0% stent migration rates; therefore, the evaluation of heterogeneity was not applicable.

Survival

We were able to calculate the HR for survival using the data in four articles.^11,12,15,16 The survival time was significantly longer in the radioactive stent group (HR: 1.70; 95% CI: 1.47–1.97, P < .00001, Fig. 5), and no significant heterogeneity among the four studies was observed (I² = 0%; P = .77).

FIG. 5.

The pooled survival was significant longer in radioactive stent group.

Severe chest pain

The data for severe chest pain rate were extracted from four studies.^12–14,16 We discovered that the comparable pooled severe chest pain rates were comparable between the two groups (22.8% versus 20.3%, OR: 1.16; 95% CI: 0.66, 2.03; P = .61, Table 3). Again, there was no significant heterogeneity found among the four studies (I² = 0%; P = .98).

Hemorrhage

Five studies provided data on the hemorrhage rates.^12–16 When the hemorrhage rates were pooled for each of the two groups, our analysis found them to be comparable (11.0% versus 9.8%, OR: 1.10; 95% CI: 0.53–2.29; P = .80, Table 3). Furthermore, there was no significant heterogeneity found among the five studies (I² = 0%; P = .88).

Fistula formation

The fistula formation rate data could be extracted from three studies,^12,14,16 and comparable pooled fistula formation rates were seen between the two groups (6.1% versus 4.2%, OR: 1.42; 95% CI: 0.45–4.45; P = .55, Table 3). No significant heterogeneity studies were observed between these studies (I² = 0%; P = .62).

Publication bias

We did not detect any evidence of publication bias pertaining to the selected study endpoints from our funnel plot analyses.

Discussion

In this meta-analysis, we assessed the reported clinical efficacies of the radioactive and normal stent insertion methods for treating patients with MEO. The clinical efficacies were primarily evaluated by comparing the instant effectiveness, long-term effectiveness, and stent-related complications.

Initially, we found similar Δdysphagia scores between two groups (P = .40), which indicate that both normal and radioactive stents can be used to provide rapid symptom relief to MEO patients.

Although various stent types can be used to provide short-term clinical benefits, stent restenosis remains a major problem because it limits long-term outcomes in patients with malignant esophageal, biliary, airway, or superior vena cava obstruction.^18–22 The major reasons for stent restenosis are tumor in-growth and overgrowth, followed by fibroepithelial hyperplasia and obstruction by food debris.¹⁸ However, a number of therapeutic strategies, including chemotherapy, external beam radiotherapy, and brachytherapy, have been employed in various efforts to prolong stent patency.⁵ Compared with traditional external beam radiation, radioactive seed brachytherapy has been shown to more effectively protect the surrounding tissues by precisely delivering radiation directly to the interior of the tumor.²³

Within the studies, the assessments of stent patency typically included two points: (i) the stent restenosis rate and (ii) time to stent restenosis. It was clear from our meta-analysis that, although the stent restenosis rates were comparable between the two groups (P = .35), the time to stent restenosis was significantly longer in the radioactive stent group (P < .00001). Furthermore, these results are consistent with findings from other studies on radioactive stent insertion in malignant biliary obstruction patients,^24,25 which suggests that radioactive seed brachytherapy cannot prevent tumor growth because not all patients were sensitive to this treatment. Nonetheless, radioactive seed brachytherapy was able to inhibit tumor growth and, thereby, prolong stent patency in treated patients.

Another problem with stent dysfunction is the phenomenon of migration. Stent migration usually results from tumor shrinkage after anticancer treatment. However, the stent migration rates were only observed as 3.3% and 4.4% in radioactive and normal groups, respectively (P = .61), which may be attributed to the antimigration design of the esophageal stents. The esophageal stents used are typically designed as a tubular configuration with a drum structure at both ends,² and the bilateral drum structure can help in fixing the stent to esophageal wall.

Our analysis of the pooled HR values indicated that radioactive stents significantly improve patient survival, and this finding is consistent with the results of other studies regarding radioactive stent insertion in malignant biliary obstruction patients.^24,25 However, a retrospective study into the use of radioactive stents for esophageal squamous cell carcinomas indicated that radioactive stents do not increase survival time,²⁶ as in that study, many patients died from massive hemorrhage.²⁶ However, there was a substantial imbalance between the patient numbers in the radioactive (n = 40) and normal (n = 91) groups.²⁶

The major complications associated with esophageal stents included severe chest pain, hemorrhage, and fistula formation,^11–16 and our meta-analysis indicated that radioactive stents do not increase the prevalence of these complications compared with normal stents. These results suggest that radioactive stent insertion is safe for the palliative management of patients with MEO.

This meta-analysis had some limitations. First, although we only included reports of RCTs, the quality of some of the RCTs was deemed low. We subsequently aim to identify higher quality RCTs and to perform further meta-analyses using those reports. Second, none of the included RCTs were double-blind studies; however, unlike in clinical drugs trials, it is almost impossible to use a double-blind study design for the surgical insertion of medical devices. Third, these studies enrolled MEO patients who had several different types of esophageal cancer, and thus, potentially constrained the applicability of our findings. Fourth, all the RCTs were conducted in China; the radioactive stent was developed by Guo et al.'s¹² and, currently, this type of stent is mainly used in China. A more general worldwide study may be possible in the future when the use of the stent has been adopted by more countries.

In conclusion, our meta-analysis demonstrated that radioactive stent insertion can prolong stent patency and the survival of patients with MEO in comparison with normal stent insertion.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received.

References

, Lv

, Xu

, et al. Double stent insertion for combined malignant airway and esophageal stenoses: Feasibility, safety, and long-term outcome. J Laparoendosc Adv Surg Tech A, 2016; 26:11–16.

Dobrucali

, Caglar

. Palliation of malignant esophageal obstruction and fistulas with self expandable metallic stents. World J Gastroenterol, 2010; 16:5739–5745.

Yamamoto

, Tada

, Kishi

, et al. Double stent for malignant combined esophago-airway lesions. Jpn J Thorac Cardiovasc Surg, 2002; 50:1–5.

Nomori

, Horio

, Imazu

, et al. Double stenting for esophageal and tracheobronchial stenoses. Ann Thorac Surg, 2000; 70:1803–1807.

Mariette

, Piessen

, Triboulet

. Therapeutic strategies in oesophageal carcinoma: Role of surgery and other modalities. Lancet Oncol, 2007; 8:545–553.

Bergquist

, Johnsson

, Nyman

, et al. Combined stent insertion and single high-dose brachytherapy in patients with advanced esophageal cancer—results of a prospective safety study. Dis Esophagus, 2012; 25:410–415.

Herth

, Peter

, Baty

, et al. Combined airway and oesophageal stenting in malignant airway–oesophageal fistulas: A prospective study. Eur Respir J, 2010; 36:1370–1374.

Zhongmin

, Xunbo

, Jun

, et al. Intraluminal radioactive stent compared with covered stent alone for the treatment of malignant esophageal stricture. Cardiovasc Intervent Radiol, 2012; 35:351–358.

Chen

, Shen

, Liu

. Radioactive self-expanding stents for palliative management of unresectable esophageal cancer: A systematic review and meta-analysis. Dis Esophagus, 2017; 30:1–16.

10.

Liu

, Liu

, Xiang

, et al. Radioactive self-expanding stents give superior palliation in patients with unresectable cancer of the esophagus but should be used with caution if they have had prior radiotherapy. Ann Thorac Surg, 2014; 98:521–526.

11.

Dai

, Zhou

, Hu

, et al. Clinical application of iodine-eluting stent in patients with advanced esophageal cancer. Oncol Lett, 2013; 6:713–718.

12.

Guo

, Teng

, Zhu

, et al. Self-expandable esophageal stent loaded with 125I seeds: Initial experience in patients with advanced esophageal cancer. Radiology, 2008; 247:574–581.

13.

, Wu

, et al. Clinical effect and safety of 125I particle stent on treatment of postoperative esophageal stricture for patients with advanced esophageal cancer. Chin J Radiol, 2014; 48:311–315.

14.

, Di

, Zhu

, et al. Complications related to conventional self-expandable metal stent insertion and internal irradiation stent insertion in patients with advanced esophageal carcinoma: An analysis of 32 cases. J Interv Radiol, 2011; 20:452–454.

15.

Zhao

, Zhang

, et al. Application of esophageal irradiation stents coated with 125I particles in advanced esophageal cancer. J BUON, 2017; 22:265–269.

16.

Zhu

, Guo

, Mao

, et al. Conventional stents versus stents loaded with (125)iodine seeds for the treatment of unresectable oesophageal cancer: A multicentre, randomised phase 3 trial. Lancet Oncol, 2014; 15:612–619.

17.

Jadad

, Moore

, Carroll

, et al. Assessing the quality of reports of randomized clinical trials: Is blinding necessary?. Control Clin Trials, 1996; 17:1–12.

18.

Dua

. History of the use of esophageal stent in management of dysphagia and its improvement over the years. Dysphagia, 2017; 32:39–49.

19.

Shim

, Gwon

, Han

, et al. Percutaneous metallic stent placement for palliative management of malignant biliary hilar obstruction. Korean J Radiol, 2018; 19:597–605.

20.

Casal

. Update in airway stents. Curr Opin Pulm Med, 2010; 16:321–328.

21.

Noor Khairiah

, Mohamad Nazrulhisham

, Hazman

. Malignant obstruction of superior vena cava: Endovascular stenting using Y-configuration stent in stent technique. Med J Malaysia, 2018; 73:407–409.

22.

Courtheoux

, Alkofer

, Al

Refaï

. M, Gervais R, Le Rochais JP, Icard P. Stent placement in superior vena cava syndrome. Ann Thorac Surg, 2003; 75:158–161.

23.

Wang

, Chai

, Wang

, et al. Expert consensus on computed tomography-assisted three-dimensional-printed coplanar template guidance for interstitial permanent radioactive ¹²⁵I seed implantation therapy. J Cancer Res Ther, 2019; 15:1430–1434.

24.

Jiao

, Wu

, Ren

, et al. Study of self-expandable metallic stent placement intraluminal ¹²⁵I seed strands brachytherapy of malignant biliary obstruction. Surg Endosc, 2017; 31:4996–5005.

25.

Zhu

, Guo

, Huang

, et al. Irradiation stents vs. conventional metal stents for unresectable malignant biliary obstruction: A multicentre trial. J Hepatol, 2018; 68:970–977.

26.

Tian

, Wen

, Fu

. Comparative study of self-expanding metal stent and intraluminal radioactive stent for inoperable esophageal squamous cell carcinoma. World J Surg Oncol, 2016; 14:1–8.