ABSTRACT
Background and Objectives
In portal hypertension, gastric varix–associated bleeding is known to have higher transfusion requirements, uncontrolled bleeding, rebleeding, intensive care unit requirements, and death. EUS–guided coil insertion is now an acceptable modality for endoscopic management in cases of gastric varices. With this study, we discuss our large single-center experience in the use of EUS for coil and glue injection in gastric varices. We also look into adverse events associated with and possibilities of using this modality as both primary prophylaxis and a rescue therapy.
Methods
The study was conducted in a tertiary care center in India. A total of 86 patients were included in the study. The indication for EUS-guided coil and glue was divided into 3 clinical situations, namely, rebleed, rescue, and primary. The technical success and clinical success, that is, control of bleed in patients, were confirmed by absence of Doppler signal on EUS, endoscopic view, and stabilized hemoglobin with no need of blood product transfusion to maintain hemoglobin.
Results
The mean Child-Turcotte-Pugh score and Model for End-Stage Liver Disease–Na score were 9.2 and 14.6, respectively. The mean size of the gastric varices was 18.9 mm. The mean number of coils used was 2.9, and the average quantity of glue required was 1.6 mL. The technical success was 100% across the patient group. Clinical success was seen in 90% of the patient group. Mean follow-up was seen for 175.2 days.
Conclusions
EUS-guided coil and glue therapy has a role in different clinical settings, as primary therapy, rebleed, and rescue therapy. It has significant technical and clinical success. Its role in treatment algorithms needs to be further studied in prospective studies. It may offer a cost advantage in comparison to interventional radiology-led interventions.
Key words: EUS, Embolization, Coil and glue, Gastric varices, Portal hypertension, Variceal bleed, Cyanoacrylate glue, EUS, Chronic liver disease, Rebleed
INTRODUCTION
Portal hypertension, irrespective of the cause, is associated with formation of portosystemic collaterals in the presence or absence of liver cirrhosis. It is associated with formation of varices, usually at multiple sites. These are commonly found in esophagus and gastric regions, and they are at significant risk of bleeding. The prevalence of varices in cirrhosis is between 40% and 85% depending on Child Class (Child-Pugh-Turcotte [CPT] scoring) of the patient.[1]
Gastric varices (GVs) are less common than esophageal varices, but they usually have higher risk and severity of bleeding. At 1, 3, and 5 years, the cumulative risks of bleeding from GVs are 16%, 36%, and 45% for patients who did not bleed at diagnosis.[2] GV-associated bleeding is known to have higher transfusion requirements, uncontrolled bleeding, rebleeding, intensive care unit requirements, and death. In GVs, the risk of bleeding is not entirely dependent on portal hypertension but also the size of varices, the wall tension, and presence of high-risk stigmata.[3] GVs are classified as gastroesophageal varices (GOV) or isolated GVs based on Sarin's classification. A larger size of the varix and child C cirrhosis are important risk factors for development of bleeds. A larger size of the fundal varix and CPT C cirrhosis are important risk factors for development of bleeding.[4]
Endoscopic cyanoacrylate (CYA) injection therapy is an accepted first-line therapy for bleeding GVs. It has higher hemostasis and lower rebleeding rates than other therapies like band ligation or sclerotherapy. Rebleeding rates vary among different studies. Some studies show it around 25% to 50%. There is no recommendation to inject nonbleeding varices; however, there are few studies that showed benefits in prophylactic glue injection when compared with nonselective β-blockers.[5] It is also associated with multiple adverse events, the fearsome being the embolization of the CYA glue into systemic circulation, which can have devastating complications. It is for this reason that EUS–guided coil insertion has come up in last few years as an acceptable modality for endoscopic management in cases of GVs. Coils with attached synthetic fibers function as a scaffold to hold CYA within the varix, and this leads to a decrease in the volume of glue needed and hence lower embolization rates.[6] One can also directly visualize occlusion of the varix in real time using the Doppler in EUS [Figure 1].
Figure 1.
A, Doppler image of gastric varices. B, Image of needle puncture. C, Image of coil being deployed. D, Obliterated varices.
EUS-guided coil and glue injection has been studied as a rescue therapy in cases of failed endoscopic intervention as well as in primary prophylaxis in high-risk cases previously. With this study, we discuss our large single-center experience for use of EUS-guided coil and glue injection in GVs. We also look into adverse events associated and possibilities of using this modality as both primary prophylaxis and a rescue therapy.
There are no studies from India comparing the cost-effectiveness of this approach as compared with the alternative radiological procedures like balloon-occluded retrograde transvenous obliteration (BRTO) or transjugular intrahepatic portosystemic shunt (TIPS). At our center, the EUS-guided coil and glue injection costs approximately 850 US dollars per procedure versus 3625 US dollars per procedure in BRTO or TIPS. This is similar to some West studies proving the cost-effectiveness of EUS-guided coil and glue injection against BRTO8 or TIPS[7] as well as glue injection only.[8] However, more regional studies from India are required to have a consensus on cost-effectiveness.
METHODS
From 2017 to 2021, prospectively maintained data were retrospectively analyzed. Patients presenting with bleeding GVs who underwent EUS-guided coiling and glue, patients with recurrent bleeding GVs after CYA injection who were undergoing EUS-guided coiling and glue, and patients with large GVs (>20 mm) with high-risk signs and Child C cirrhosis undergoing prophylactic EUS-guided coiling and glue injection were included. Patients who had multiorgan failure, refused consent, had prior gastrointestinal surgery, and/or were pregnant were excluded. Informed consent was taken from all the patients before the procedure.
There were various definitions used for EUS-guided coil and glue intervention. For the sake of this study, the indications for EUS-guided coil and glue were divided into 3 clinical situations, namely, rebleed, rescue, and primary.
Rebleed: Rebleed was defined as rebleed in patients of GVs or failure of conventional endoscopic means, who were initially controlled with endoscopic therapy (endoscopic glue injection), which happened within the first 5 days (as per Baveno 6 guidelines), and a decision was made to perform the EUS-guided coil and glue therapy.
Rescue: Rescue was defined as the use of EUS-guided coil and glue therapy when endoscopic glue injection fails in the primary control of bleed, and a decision was made to shift to EUS-guided coil and glue on the table.
Primary: Primary was defined as the use of EUS-guided coil with glue in patients who have not bled but are at high risk of GV bleed because of the following factors: size of the varix more than 20 mm, high-risk stigmata, and Child C status.
Technical success: Technical success was defined as EUS-guided placement of coils and glue with placement confirmed by fluoroscopy and EUS images.
Clinical success: Clinical success was defined as control of bleed in patients, confirmed by absence of Doppler signal on EUS, endoscopic view, and stabilized hemoglobin with no need of blood product transfusion to maintain hemoglobin.
Materials used and procedures
All procedures were done in a tertiary care hospital setting with adequate backup from interventional radiology, gastrointestinal surgery, and intensive care units. Most of the EUS were performed by an experienced interventional endoscopist with more than 15 years' experience. Interventional radiology backup was always ready. Total intravenous anesthesia using propofol (60–150 mg) was administered by the anesthesiologist without intubation in a majority of cases. In exceptional cases where the anesthesiologist deemed it necessary for safety of the patient, particularly in patients with acute bleeding, intubation and mechanical ventilation was done before the procedure. Midazolam (2 mg) and fentanyl (100 μg) were also administered. All the patients received intravenous ceftriaxone before the procedure. EUS was performed using an Olympus ME2 linear echoendoscope. Initially, the varices were evaluated with standard endoscopy, after which EUS examination was done. The stomach was filled with approximately 125 to 150 mL of saline, and the varices were assessed. The detailed analysis on their size, extent, flow rates on color Doppler, and location of perforating vessels supplying the varices was done. Then, the optimal method of EUS-guided coil ablation was decided—either the transesophageal–transcrural approach or the transgastric approach. In transesophageal approach, the scope is relatively stable and facilitates the puncture. It is beneficial only for varices near the cardia. For all other varices, transgastric approach is needed.
The choice of needle was made depending on the coil being used. Coils are synthetic strands attached to metal coils. They lead to coagulation in the varix and also provide a scaffold for the glue to act locally. Coil diameters range from 2 to 20 mm. A 19G needle is needed if we intend to use >10-mm coils. The size of coil used is approximately 1.2 times bigger than the diameter of the varix. N-actyl cyanoacrylate or N-octyl cyanoacrylate can be used as glue through the same needle used for puncture. N-actyl cyanoacrylate may require prior lipoidal injection along with the glue to delay polymerization.
We primarily used either Boston Scientific's 19G Expect Needle or Cook Medical's EchoTip Ultra Needle.
After ascertaining the appropriate approach, the puncture was made. The needle was primed with 5% dextrose. After puncturing the varix, aspiration can be done to confirm the presence of blood, which confirms the right position. The needle was flushed again with 5% dextrose to prevent clotting.
We used the embolization coils, which are between 5 and 18 mm in diameter and 7 and 14 cm in length (Cook Medical, Bloomington, IN). After inserting the coils into the working channel of the echoendoscope, the needle stylet was used to push the coils into the varix under echoendoscope and fluoroscopic guidance. Using the same needle, CYA glue diluted with lipoidal in a 1:1 ratio was injected in the lumen slowly.
After the procedure, hemostasis was ascertained in endoscopic view, and flow of the blood in the varix was checked using Doppler. There should be lack of active bleeding and absent flow during Doppler examination. The needle was withdrawn into the sheath, and the entire scope was pulled out gently. Diet was usually resumed 6 hours after the procedure, as it was done under anesthesia. Initially, clear liquid diet was started and upgraded as tolerated. Hemoglobin was monitored. Standard medical management was followed as per the clinical protocols for the underlying disease. Repeat endoscopy after 4 weeks was usually scheduled to assess the presence of residue varices.
Statistical analysis
The characteristics of the patients are reported with the use of descriptive statistics, which include percentages, median with range, and means with SDs. All continuous variables are compared with the Student t test and categorical variables with χ2 test. P values less than 0.05 were considered to indicate statistical significance. All P values were based on two-tailed tests. Data were analyzed using SPSS version 18.
RESULTS
A total of 86 patients were included in the study. Table 1 shows the demographic distribution and the study variables. The mean age of the patient was 53.5 years. The mean CPT score and Model for End-Stage Liver Disease–Na score were 9.2 and 14.6, respectively. The mean size of the GVs was 18.9 mm. The mean number of coils used was 2.9, and the average quantity of glue required was 1.6 mL.
Table 1.
Descriptive of the study variables (n = 86).
| Mean ± SD | Median (IQR) | Minimum | Maximum | |
|---|---|---|---|---|
| Age, y | 53.5 ± 14.4 | 52.5 (45–65) | 26 | 87 |
| Size of GV, mm | 18.9 ± 3.4 | 18 (16–20) | 12 | 34 |
| CTP score | 9.2 ± 1.8 | 9 (8–10) | 5 | 12 |
| Na MELD | 14.6 ± 5 | 12 (11–18) | 8 | 28 |
| No. coil | 2.9 ± 1 | 3 (2–4) | 1 | 6 |
| Glue, mL | 1.6 ± 0.8 | 1.5 (1–2) | 0.5 | 5.5 |
| Follow-up days | 175.2 ± 110.5 | 148 (90–240) | 30 | 365 |
CTP: Child-Turcotte-Pugh; GV: gastric varix; IQR: interquartile range; MELD: Model for End-Stage Liver Disease.
The technical success was 100% across the patient group. Clinical success was seen in 90% of the patient group. Mean follow-up was seen for 175.2 days.
Table 2 shows the various parameters of the patients involved in the study. The most common etiology was alcohol-related cirrhosis, with 51.2% of patients associated with it. Other etiologies seen included hepatitis B, hepatitis C, and non alcoholic steato-hepatitis-related cirrhosis. GOV2 GVs as per Sarin's classification were the most common type of fundal varix encountered. The rest of the varices were GOV1. Most of GOV1 fundal varices are managed by endoscopic band ligation or glue injection. The majority of GOV1, where endoscopic coil and glue was needed, was managed by rescue therapy. Less than half (20.9%) of patients had associated hepatocellular carcinoma, and 22.1% of patients had portal vein thrombosis or superior mesenteric vein thrombosis.
Table 2.
Distribution of the study variables (n = 86).
| n (%) | |
|---|---|
| Etiology | |
| Autoimmune | 4 (4.65) |
| Cryptogenic | 1 (1.1) |
| Ethanol | 40 (46.5) |
| HBV | 13 (15.1) |
| HCV | 8 (9.3) |
| NASH | 12 (13.9) |
| Wilson | 1 (1.1) |
| Noncirrhotic portal hypertension | 7 (8.13) |
| Sarin's GV class | |
| GOV1 | 18 (20.9) |
| GOV2 | 68 (79.1) |
| Primary/rebleed/rescue | |
| Primary | 18 (20.9) |
| Rebleed | 53 (61.6) |
| Rescue | 15 (17.4) |
| Associated PVT/SMV thrombosis | 19 (22.1) |
| HCC | 18 (20.9) |
GOV: gastroesophageal varix; GV: gastric varix; HBV: hepatitis B virus; HCC: hepatocellular carcinoma; HCV: hepatitis B virus; NASH: Non alcoholic steato-hepatitis; PVT: portal vein thrombosis; SMV: superior mesenteric vein.
For the sake of this study, the indication for EUS-guided coil and glue was divided into 3 clinical situations, namely, rebleed, rescue, or primary. These definitions are mentioned in the Methods section. EUS-guided coil and glue was used for rebleeds in a majority of cases, that is, 61.6% of patients. As rescue therapy, for immediate control of bleed, after the failure of conventional glue, it was used in 17.4% of patients. In 20.9% of patients, EUS-guided coil and glue was used for primary prophylaxis of bleed in patients who were at high risk of bleed from GV.
Thus, EUS-guided coil and glue intervention was used in 3 different clinical settings where its immediate use can be done.
There were no cases of immediate postprocedural bleed in our cohort. A total 4 adverse events were noted in our study. A patient also had esophageal varices, because of which we could not do the transesophageal puncture and had to use the transgastric route. The needle was stuck in the varix after glue injection in this case. However, there was no significant bleeding after the event, and the needle was confirmed to be extruded in the follow-up endoscopy.
Another adverse event was aspiration of blood in a patient with a history of rebleed; however, the patient was immediately intubated for airway protection and recovered soon after the procedure. Two patients complained of severe abdominal pain for 2 days after the procedure for which they were kept admitted, and they recovered with conservative management. Thus, EUS-guided coil and glue intervention was used in 3 different clinical settings. Because this is a retrospective analysis and our primary aim was to show the procedure's efficacy in all the 3 different scenarios, the outcome cannot be compared in terms of which group the procedure is more effective.
DISCUSSION
The aforementioned retrospective study highlights the multiple clinical situations where EUS-guided coil and glue is an acceptable alternative to other conventional means to control gastric variceal bleeds. EUS-guided coil and glue has been described multiple times in the literature previously.[9–11] We have used EUS-guided coil and glue injection in 3 clinical situations: in acute control, for control in cases of rebleed, and also prophylactically in high-risk cases. In the largest series from San Francisco,[11] there were around 152 patients, all of whom had combined coil and CYA injection. The study had a mix of cases, done prophylactically or after the first bleed. One to 4 coils were deployed, and mean quantity of glue used was around 2 mL. EUS-confirmed obliteration on Doppler imaging was around 93%. In their study, the authors concluded that the technique is highly effective. In our study, which is one the largest studies from India, we had 86 patients, all of whom underwent EUS-guided coil and glue placement in different clinical settings; 90% of patients showed clinical success as defined in Methods.
In a study from India by Mukkada et al.,[12] a retrospective analysis of a prospectively maintained database was done. Thirty patients who underwent CYA injection and EUS-guided glue injection were included. There was a comparison group of 51 patients who only underwent CYA injection. Both of the groups had a follow-up of 12 months. Twenty-percent of the patients in the coiling group had rebleeding during follow-up, whereas 51% had rebleeding in the glue-only group. They concluded that EUS-guided coil application resulted in significantly less rebleed than the glue-only group. This is one of the studies that showed superiority of the EUS-guided coil over conventional glue–only endoscopic intervention.
The strength of our study lies in demonstrating the use of EUS coil and glue therapy as one therapy that can help in various clinical settings that we see in our daily practice. In our study, it has shown success as primary therapy, in cases of rebleed and even in cases where a rescue therapy was needed. All these data lead to gradual acceptance of this therapy wherever available as the next therapy when endoscopic glue has failed or rescue is needed. It also shows that there are certain subgroups of patients, particularly who have a high risk of bleed from GVs, where EUS-guided coil and glue may be the first-choice therapy.
One of the limitations of our study is that being a retrospective study of prospectively maintained data, it comes with inherent retrospective data bias. The findings need to be confirmed in a prospective study. Another limitation is the lack of another group in which we could study the clinical outcomes in comparison to the standard procedure of endoscopic coil and glue to establish its noninferiority or superiority. We also need studies analyzing cost-effectiveness to establish its role in comparison to interventional radiology-led interventions. At our center, it is cheaper for us to do EUS-guided coil and glue in comparison to BRTO.
We hope that this study can lead to more multicentric prospective studies where EUS-guided coil and glue therapy can be studied as a first-line therapy in comparison to the endoscopic glue therapy and further identify patients who will be most benefitted from using EUS coil and glue as a first-line therapy. In the future, it may come out as one therapy to rule them all.
CONCLUSIONS
EUS-guided coil and glue therapy has a role in different clinical settings, as primary therapy, rebleed, and rescue therapy. It has significant technical and clinical success. Its role in treatment algorithms needs to be further studied in prospective studies. It may offer a cost advantage in comparison to interventional radiology-led interventions.
Author Contributions
Rajesh Puri conceptualized the study, edited the final proofs, was involved in procedures, and approved the final draft. Zubin Dev Sharma was involved in procedures, material preparation, data collection, and analysis and prepared the draft of the manuscript. He wrote the first draft of the manuscript and approved the final draft. Saurabh Patle was involved in data collection and analysis, proofread the first draft, and commented on multiple versions and approved the final draft. Suraj Bhagat was involved in data collection and analysis and in procedures and approved the final draft. Abhishek Kathuria was involved in data collection and analysis and approved the final draft.
Contributor Information
Rajesh Puri, Email: rajeshpuri69@gmail.com.
Saurabh Patle, Email: dr.saurabhpatle@gmail.com.
Suraj Bhagat, Email: bhagatsuraj@yahoo.com.
Abhishek Kathuria, Email: dr.saurabhpatle@gmail.com.
Conflicts of Interest
All authors have no conflict of interest to declare.
REFERENCES
- 1.Vine LJ, Subhani M, Acevedo JG. Update on management of gastric varices. World J Hepatol 2019;11(3):250–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Triantafyllou M, Stanley AJ. Update on gastric varices. World J Gastrointest Endosc 2014;6(5):168–175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tayyem O, Bilal M, Samuel R, Merwat SK. Evaluation and management of variceal bleeding. Dis Mon 2018;64(7):312–320. [DOI] [PubMed] [Google Scholar]
- 4.Sarin SK, Lahoti D, Saxena SP, Murthy NS, Makwana UK. Prevalence, classification and natural history of gastric varices: a long-term follow-up study in 568 portal hypertension patients. Hepatology 1992;16(6):1343–1349. [DOI] [PubMed] [Google Scholar]
- 5.Jun CH Kim KR Yoon JH, et al. Clinical outcomes of gastric variceal obliteration using N-butyl-2-cyanoacrylate in patients with acute gastric variceal hemorrhage. Korean J Intern Med 2014;29(4):437–444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Binmoeller KF, Weilert F, Shah JN, Kim J. EUS-guided transesophageal treatment of gastric fundal varices with combined coiling and cyanoacrylate glue injection (with videos). Gastrointest Endosc 2011;74(5):1019–1025. [DOI] [PubMed] [Google Scholar]
- 7.Wang TJ, Bazarbashi AN, Hathorn K, Thompson CC, Ryou M. Su1203 Eus-guided coil therapy is cost favorable over IR-guided therapies for treatment of gastric variceal bleeding: a cost utility analysis. Gastrointest Endosc 2019;89(6):AB314. [Google Scholar]
- 8.Robles-Medranda C Nebel JA Puga-Tejada M, et al. Cost-effectiveness of endoscopic ultrasound-guided coils plus cyanoacrylate injection compared to endoscopic cyanoacrylate injection in the management of gastric varices. World J Gastrointest Endosc 2021;13(1):13–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lee YT Chan FK Ng EK, et al. EUS-guided injection of cyanoacrylate for bleeding gastric varices. Gastrointest Endosc 2000;52(2):168–174. [DOI] [PubMed] [Google Scholar]
- 10.Romero-Castro R Pellicer-Bautista FJ Jimenez-Saenz M, et al. EUS-guided injection of cyanoacrylate in perforating feeding veins in gastric varices: results in 5 cases. Gastrointest Endosc 2007;66(2):402–407. [DOI] [PubMed] [Google Scholar]
- 11.Bhat YM Weilert F Fredrick RT, et al. EUS-guided treatment of gastric fundal varices with combined injection of coils and cyanoacrylate glue: a large U.S. experience over 6 years (with video). Gastrointest Endosc 2016;83(6):1164–1172. [DOI] [PubMed] [Google Scholar]
- 12.Mukkada RJ Antony R Chooracken MJ, et al. Endoscopic ultrasound-guided coil or glue injection in post-cyanoacrylate gastric variceal re-bleed. Indian J Gastroenterol 2018;37:153–159. [DOI] [PubMed] [Google Scholar]