Gastric varices management: Is clip-assisted glue injection a real-world alternative to endoscopic ultrasound-guided therapy?

Abstract

Gastric variceal (GV) bleeding remains a life-threatening complication of portal hypertension, with ongoing debate regarding the optimal endoscopic therapy. Conventional endoscopic cyanoacrylate injection (ECI) is effective but limited by the risk of ectopic embolism, particularly in the presence of gastrorenal shunts. Clip-assisted ECI (clip-ECI) has emerged as a novel modification designed to reduce embolic risk while maintaining hemostatic efficacy. We appraised the recent study by Xiong et al, which compared clip-ECI with endoscopic ultrasound-guided coil and cyanoacrylate injection in 108 propensity-matched patients with cardiofundal varices and shunts. Both techniques demonstrated comparable efficacy, with obliteration rates exceeding 90% and similar one-year rebleeding rates. Importantly, no embolic events were reported. These findings are consistent with prior studies, including multicenter cohorts and a recent randomized controlled trial, which highlight clip-ECI as a safe, effective, and efficient technique, with advantages of shorter procedure times, fewer sessions, and lower costs. While endoscopic ultrasound (EUS)-guided therapy offers precision in expert hands, clip-ECI provides a practical, accessible alternative, particularly in resource-limited settings. Larger prospective studies with standardized definitions and cost-effectiveness analyses are needed to refine treatment algorithms. Clip-ECI represents a promising “flow-control assisted” strategy and a real-world alternative to EUS-based therapies for GV.

Key Words: Gastric varices; Endoscopy; Gastrointestinal; Tissue adhesives; Hemostasis; Endoscopic; Endoscopic ultrasound

Core Tip: Clip-assisted cyanoacrylate injection (clip-ECI) is a safe, effective, and practical method for managing cardiofundal gastric varices, particularly in patients with gastrorenal shunts. Evidence from multicenter cohorts and a randomized trial shows comparable hemostasis to conventional and endoscopic ultrasound (EUS)-guided techniques, with a markedly lower risk of ectopic embolism. By restricting blood flow and improving variceal obliteration, clip-ECI reduces glue migration and often requires fewer sessions. Its shorter procedure time and lower cost make it ideal for centers with limited EUS resources.

TO THE EDITOR

We read with great interest the article by Xiong et al[1] comparing the outcome of clip-assisted endoscopic cyanoacrylate injection (clip-ECI) vs endoscopic ultrasound-guided coil and cyanoacrylate injection (EUS-C+G) for gastric varices (GVs). In this propensity score-matched analysis of 108 patients with fundal varices (gastroesophageal varices type 2 or isolated GV type 1) and gastrorenal shunt (GRS), clip-ECI demonstrated similar efficacy to EUS-C+G, with variceal obliteration rates of 91.7% vs 94.4% and comparable 1-year rebleeding rates (23.6% vs 19.4%). No cases of ectopic embolism occurred in either group[1]. The authors should be commended for undertaking this comparative analysis of two promising modalities in the management of GVs, an area where robust head-to-head data are scarce. While the study makes an important contribution to the field, several limitations merit discussion, particularly when interpreting the applicability of these findings to clinical practice.

Study limitations

As the study was retrospective in design and the treatment choice was at the discretion of the endoscopist, there was a chance of bias, even after propensity matching. The reduced post-matching sample size limits statistical power to detect rare but critical adverse events such as glue embolization, a major concern with endoscopic cyanoacrylate injection (ECI). Residual imbalances in baseline characteristics, including higher disease severity among clip-ECI patients, may have influenced outcomes despite nonsignificant differences. Additionally, variceal size assessment varied across patients, with some relying on visual endoscopic estimation rather than EUS measurement, potentially reducing accuracy in subgroup analyses. The cost comparison focused solely on procedural expenses, overlooking broader factors like hospitalization, transfusion needs, and reinterventions. Furthermore, subgroup analyses for patients with large varices were underpowered, especially in the EUS-C+G arm, limiting meaningful statistical comparison. Consequently, the absence of significant differences between treatment groups should be interpreted cautiously.

Comparison with previous literature

Table 1 summarizes the growing body of evidence on the efficacy and safety of clip-ECI for GVs. The earliest evidence came from retrospective cohort studies, such as Li et al[2], who demonstrated higher variceal obliteration (100% vs 72%) and lower rebleeding rates (4.3% vs 18%) with clip-ECI compared to conventional ECI. Subsequent multicenter observational studies, including those by Zhang et al[3], confirmed high technical success and complete obliteration without embolic events, while Yu et al[4], in a small case series, reported 100% hemostasis and no serious complications, though two patients experienced rebleeding. Subsequently, Zhang et al[5] conducted a large multicenter propensity-matched analysis, showing significantly lower six-month rebleeding rates (11.4% vs 26.3%) and fewer sessions required for obliteration with clip-ECI. The highest level of evidence at present comes from a randomized controlled trial by Wang et al[6], which demonstrated a substantial reduction in cyanoacrylate embolism (11.4% vs 42.9%) with similar rebleeding outcomes. This transition from retrospective to randomized data substantiates clip-ECI as a safer, flow-restrictive modification of conventional ECI, especially in patients with GRS. The present study by Xiong et al[1] complements this evidence by confirming that clip-ECI retains high efficacy and safety even when compared with the advanced EUS-guided coil-plus-glue technique, while offering procedural simplicity, shorter duration, and lower cost. These advantages make it a practical and accessible alternative in resource-limited settings.

Table 1 Summary of studies on the outcome of clip-assisted glue injection for gastric varices.

Ref.	Design/sample size	Baseline characteristics	Key results	Comparison with con-ECI
Li et al[2]	Multicenter retrospective (n = 96). Clip-ECI: 46; Con-ECI: 50	IGV1 only: GRS: 59/66 (89.4%); Child-Pugh A/B; No prior variceal therapy	Hemostasis: 100% both groups; Rebleeding: 4.3% (clip) vs 18% (con) (P = 0.007); Obliteration: 100% (clip) vs 72% (con) (P < 0.001); Embolism: 1 cerebral (con-ECI)	Clip-ECI superior in rebleeding prevention and obliteration; similar safety overall
Zhang et al[3]	Multicenter observational (n = 61) Clip-ECI only	GOV2/IGV1 with CT-confirmed GRS; Mostly cirrhotics; some with large varices; Child-Pugh A/B/C	Technical success: 100%; Rebleeding: 22.8% overall; Embolism: None; Minor side effects only	-
Yu et al[4]	Single-center case series (n = 9) Clip-ECI only	GOV1/GOV2: 4/9 had spontaneous shunts; Child-Pugh A/B/C; Acute bleeding cases	Hemostasis: 100%; Rebleeding: 2/9 (22%); Embolism: None; Cyanoacrylate extrusion: 2 cases	-
Zhang et al[5]	Multicenter retrospective (n = 274). Clip-ECI: 148; Con-ECI: 126	GOV2/IGV1 with GRS; comparable age, sex, MELD, Child-Pugh	Rebleeding-free at 6 months: 88.6% (clip) vs 73.7% (con) (P = 0.002); Survival: NS; Fewer sessions in Clip-ECI (P = 0.015)	Clip-ECI offers better long-term control with fewer sessions
Wang et al[6]	Randomized controlled trial (n = 70). Clip-ECI: 35; Con-ECI: 35	GOV2/IGV1 with GRS (confirmed CT/MRI); Comparable age, sex, MELD, Child-Pugh	Embolism: 11.4% (clip) vs 42.9% (con) (P = 0.003); Symptomatic PE: 4 (con) vs 0 (clip); Rebleeding: 14.3% both Mortality: 1 death (con)	Clip-ECI is significantly safer for embolism; equally effective for bleeding

Con: Conventional; ECI: Endoscopic cyanoacrylate injection; GOV: Gastroesophageal varices; GRS: Gastro-renal shunt; IGV: Isolated gastric varices; PE: Pulmonary embolism; GOV2: Gastroesophageal varices type 2; IGV1: Isolated GV type 1; CT: Computed tomography; MRI: Magnetic resonance imaging; MELD: Model for end-stage liver disease.

Implications for clinical practice and future directions

Taken together, the evidence suggests that both clip-ECI and EUS-C+G are effective and safe strategies for the management of GVs, with no significant difference in one-year rebleeding rates. Yet, their relative strengths lie in different domains: EUS-C+G in precision and embolic risk reduction in expert hands, and clip-ECI in feasibility, accessibility, and cost containment. Importantly, the absence of embolization events in both groups in this study may reflect the protective effect of using either clips or coils, lending support to the broader concept of “flow-control assisted” ECI.

Despite advances in the endoscopic management of GVs, the optimal approach remains unanswered, requiring larger prospective randomized controlled trials. These should stratify patients by variceal subtype and size, incorporate standardized outcome definitions, and ideally include cost-effectiveness analyses across diverse health care systems. Furthermore, recent studies have reported comparable efficacy of EUS-C+G as retrograde transvenous obliteration (RTO)[7,8]. Thus, comparative data for clip-CI against RTO would provide a more complete picture for multidisciplinary decision-making.

Conclusion

Despite the limitations, the study adds useful information to the management of GVs. The fact that clip-ECI was quicker, cheaper, and just as effective as EUS-C+G makes it an attractive option, particularly in urgent situations or in hospitals where EUS expertise is limited. In conclusion, this study provides encouraging evidence that clip-ECI can be a practical and effective alternative to EUS-C+G for GVs. Further, larger, prospective, randomized, multicentric studies with standardized techniques and longer follow-up are needed to confirm these results and to better guide treatment decisions.