Endoscopic Cyanoacrylate Injection with Lauromacrogol for Gastric Varices: Long-Term Outcomes and Predictors in a Retrospective Cohort Study

Abstract

Introduction:

The first-line treatment for gastric variceal rebleeding in cirrhotic patients is endoscopic cyanoacrylate injection. We focused on the efficacy and prognosis of cyanoacrylate combined with Lauromacrogol^® for gastric varices (GV) in a retrospective study of long-term follow-up.

Materials and Methods:

One hundred thirty patients with cirrhosis and GV from March 2011 to February 2013 were included. Sixty-eight patients underwent endoscopic cyanoacrylate injection with Lauromacrogol and 62 patients without Lauromacrogol. The median follow-up was 40.1 and 38.8 months, respectively.

Results:

The volumes of cyanoacrylate used for the GV eradication in the Lauromacrogol group were significantly lower than those in the Ethiodol^® group (1.6 ± 0.8 versus 2.1 ± 1.2 mL, P = .029). No ectopic embolisms were observed during follow-up. The 1- and 3-year rebleeding-free rate did not differ between groups (83.7% and 59.2% versus 75.8% and 62.5%; P = .797). The same was observed for mortality (86.6% and 83.5% versus 85.5% and 83.7%; P = .955). New portal venous thrombosis (PVT) and progression of previous partial PVT were independently associated with rebleeding (hazard ratio [HR] 5.127, 95% confidence interval [CI], 2.430–10.817, P = .000) and death (HR 10.093, 95% CI, 3.988–25.548, P = .000).

Conclusions:

Endoscopic cyanoacrylate injection with Lauromacrogol might minimize the required dosage of cyanoacrylate, but it did not improve rebleeding rate or survival. Exacerbation of PVT was associated with rebleeding and death.

Introduction

The morbidity and mortality of gastroesophageal variceal bleeding are high in cirrhotic patients with portal hypertension.¹ The prevalence of gastric varices (GV) was from 15% to 17% in patients with liver cirrhosis,² less than that of esophageal varices (EV). The bleeding incidence of GV was about 20% to 30% at 2 years with more blood loss than EV bleeding.³ The rebleeding and mortality rates were 40% and 50%, respectively,⁴ higher than that of EV. Treatments include vasoactive drugs and antibiotics, although those used in EV are less effective in GV.^5,6 Injection of N-butyl-cyanoacrylate (NBCA), a chemical tissue adhesive, is the most commonly used endoscopic treatment for GV^7,8 and is recommended by the Baveno VI, National Institute for Health and Care Excellence, and American Association for the Study of Liver Diseases guidelines for its higher hemostasis and lower rebleeding rates than other sclerosants and band ligations.^9–11

Endoscopic injection of this tissue adhesive is performed alone or in combination with an agent (e.g., Ethiodol^® (Bracco, Shanghai, China)) during or after the procedure.^12,13 The sandwich technique is the standard procedure of endoscopic cyanoacrylate injection in China, which involves tissue adhesive injection and a rinse of Ethiodol before and after the injection. The proper dosage of cyanoacrylate is difficult to ascertain¹⁴ and mostly decided by the endoscopist, taking into account the GV's number and size. However, a large dose of cyanoacrylate injection was demonstrated to increase the risk of ectopic embolism such as pulmonary and cerebral.^15–17 A cyanoacrylate dilution without Ethiodol adds to the difficulty of operation for doctors and nurses,¹⁸ with no difference in the volume of cyanoacrylate required. Lauromacrogol, an alcohol-based compound that has been used for the management of EV for many years, has drawn interest because it can produce sclerosis on any remaining small varices and submucosal fibrosis in the esophagus.^11,19 In our previous study, we found cyanoacrylate combined with Lauromacrogol prevents rebleeding in the short term, as well as the traditional procedure with Ethiodol.²⁰

The purpose of this study was to assess the efficacy and prognosis of cyanoacrylate injection with Lauromacrogol for individuals with cirrhosis and GV bleeding, compared with cyanoacrylate injection without Lauromacrogol, during a long follow-up. Predicting factors associated with rebleeds and survival among these patients were identified.

Patients and Methods

Study population

Consecutive cirrhotic patients with history of variceal bleeding were included between March 2011 and February 2013. Patients were excluded from the criteria for solitary esophageal varices without GV; portal-systemic shunt; GV treated with hypertonic glucose; hepatocellular carcinoma (HCC); acute portal venous thrombosis (PVT); intrahepatic arteriovenous fistula; systemic diseases that might reduce overall survival; previous treatment for esophagogastric varices (EGV), including surgical shunt, transjugular intrahepatic portosystemic shunt (TIPS), or liver transplantation. The use of clinical data was approved by the Clinical Research Ethics Committee of Zhongshan Hospital, Fudan University.

Cirrhotic patients with acute gastrointestinal bleeding were given vasoactive agents and antibiotic prophylaxis. Gastroscopy was performed to determine the status of gastroesophageal varices. GV classification was defined according to Sarin's classification and previous studies.^20–22

Treatment

Patients were treated by endoscopic cyanoacrylate injection with or without Lauromacrogol. Endoscope GIF-XQ240/H260^® (the injection needle Olympus NM-200L-423; Olympus), NCBA^® (Beijing Suncon Medical Adhesive Co. Ltd., Beijing, China), Lauromacrogol^® (Tianyu Pharmaceutical, Zhejiang, China), and Ethiodol were used for the treatment. The sandwich technique was used. Cyanoacrylate was injected in GV with Ethiodol or Lauromacrogol before and after tissue adhesive injection. Endoscopic band ligation was performed for patients with concurrent esophageal varices.

Follow-up and outcomes

If patients had symptoms of embolic symptoms, they would have X-ray or computed tomography (CT) scans. Proton pump inhibitors were prescribed to all patients for 8 weeks. Follow-up data were collected until death, liver transplantation, or study termination at February 1, 2016. Clinical and endoscopic follow-up schedules were 2, 4, and 6 months and then every 6 months. Medical history, physical examination, biochemistry, hematologic tests, abdominal ultrasound/CT/magnetic resonance imaging, and gastric endoscopy were recorded every 6 months. If varices reappeared, further endoscopic treatment was initiated. The major outcome variable was rebleeding from any source, which was defined as emerging hematemesis, melena, or both. The secondary outcome was survival.

Statistical analyses

Statistical analyses were performed with SPSS 16.0 software (SPSS Inc. Released 2007. SPSS for Windows, Version 16.0. Chicago, SPSS Inc.). Continuous variables were expressed as mean ± standard deviation and compared using a Student's t test. Qualitative data were described with constituent ratios, and intragroup comparisons were achieved using a χ² or Fisher exact tests. Patients were censored at the time of liver transplantation or at the last outpatient visit for loss-to-follow-up. Kaplan–Meier analysis was carried out to estimate the cumulative probability of rebleeding and survival. Factors associated with rebleeding or death were analyzed with the Cox proportional hazards model. Multivariate analyses were performed with backward stepwise regression, retaining in the model variables explaining a statistically significant proportion of the variance (P < .2). All statistical analyses were two-sided tests (P < .05).

Results

Clinical and endoscopic characteristics of study patients

In total, 522 patients with liver cirrhosis were admitted to our hospital between March 2011 and February 2013 for endoscopic treatment of EGV bleeding, of which 392 (75%) were excluded and 130 (25%) patients were included. The flow chart for the inclusion and exclusion is shown in Figure 1. Sixty-two patients had the endoscopic cyanoacrylate treatment with Ethiodol (the Ethiodol group), and 68 patients with Lauromacrogol (the Lauromacrogol group).

FIG. 1.

Flow chart. Flow diagram of the study. EGD, esophagogastroduodenoscopy; EGV, esophagogastric varices; HCC, hepatocellular carcinoma; OLT, orthotopic liver transplantation; TIPS, transjugular intrahepatic portosystemic shunt.

Basic clinical characteristics of the two groups are presented in Table 1. The most frequent etiology was viral hepatitis. Mean MELD score was 10 and 93% of the patients were Child–Pugh A or B class. Eighteen percent of the patients were diagnosed with PVT via computed tomography angiography (CTA) scan. Eleven patients (8%) were initially treated with propranolol. There was no statistical difference of clinical characteristics between the two groups.

Table 1.

Baseline Clinical Characteristics of Patients

Characteristics	Ethiodol^® group (n = 62)	Lauromacrogol^® group (n = 68)	P value
Age (mean ± SD), years	53.2 ± 13.7	53.2 ± 11.7	.988
Gender (male/female)	40/22	42/26	.745
Etiology of cirrhosis (hepatitis virus/alcohol/others)	39/4/19	40/8/20	.578
Child–Pugh score (A/B/C)	12/44/6	15/50/3	.486
MELD score (mean ± SD)	10.4 ± 2.4	9.6 ± 2.0	.102
Interval time (≤6 days/>6 days)	12/50	18/50	.336
Propranolol before treatment	3 (4.8%)	8 (11.8%)	.156
Splenectomy	5 ((8.1%)	3 (4.4%)	.387
Portal venous thrombosis	13 (21.0%)	11 (16.2%)	.482
Massive ascites	11 (17.7%)	12 (17.6%)	.989
Diabetes	10 (16.1%)	9 (13.2%)	.641
Platelets (median, range) 10⁹/L	59 (20–388)	66 (22–270)	.953

Interval time, time between variceal bleeding and randomization.

MELD, model for end-stage liver disease; SD, standard deviation.

Fifteen patients had isolated gastric varices, 8 in the Lauromacrogol group and 7 in the Ethiodol group. The volume of cyanoacrylate for the first endoscopic treatment was less in the Lauromacrogol group than that in the Ethiodol group (1.0 versus 2.0 mL, P = .021), as shown in Table 2. Endoscopic intervention was performed with 100% technical success. Two patients in each group experienced bleeding during the procedure. No serious adverse events such as ectopic embolism were found in either group during the follow-up period.

Table 2.

Endoscopic Characteristics, Treatment, and Outcomes in the Two Groups

	Ethiodol group (n = 62)	Lauromacrogol group (n = 68)	P value	95% CI
Endoscopic characteristics
Esophageal varices	56 (90.3%)	64 (94.1%)	.417	—
GOV classification (1/2/3)	46/6/2	57/2/1	.202	—
GV diameter (<1 cm/1–2 cm/≥2 cm)	1/44/17	2/49/17	.850	—
IGV 1	7 (11.3%)	8 (11.8%)	.933	—
PHG	14 (22.6%)	14 (20.6%)	.696	—
Endoscopic treatment
Volume of cyanoacrylate in first session (median, range) (mL)	2.0 (1.0–6.0)	1.0 (0.5–4.0)	.021	0.065–0.771
Bleeding during procedure	2 (3.2%)	2 (2.9%)	.925	—
Treatment failure	0 (0%)	0 (0%)	—	—
Endoscopy follow-up	44 (71.0%)	59 (86.8%)
Eradication rate for first session	39 (88.6%)	51 (86.4%)	.927	—
Postinjection ulcer	14 (31.8%)	22 (37.3%)	.565	—
Volume of cyanoacrylate for eradication (mean ± SD), mL	2.1 ± 1.2	1.6 ± 0.8	.029	0.048–0.869
No. of sessions for eradication	1.14 ± 0.41	1.17 ± 0.46	.706	−0.207–0.140

CI, confidence interval; GOV, gastroesophageal varices; GV, gastric varices; IGV, isolated gastric varices; PHG, portal hypertensive gastropathy; SD, standard deviation.

A total of 103 patients (79.2%) had a repeat endoscopy, of which 14 cases (31.8%) for the Ethiodol group and 22 cases (37.3%) for the Lauromacrogol group had postinjection ulcers (P = .565). For the first session, GV obliteration rates were 88.6% and 86.4% in the Lauromacrogol group and Ethiodol group, respectively (P = .927). The number of endoscopic sessions for GV eradication was 2.1 ± 1.2 in the Ethiodol group and 1.6 ± 0.8 in the Lauromacrogol group, and GV recurrence did not differ between the two groups. The volume of cyanoacrylate for GV eradication was less in the Lauromacrogol group than that in the Ethiodol group (P = .029) (Table 2).

The median follow-up did not constitute a significant difference (38.8 months versus 40.1 months). Eight patients (6%; 6 in the Ethiodol group, 2 in the Lauromacrogol group) were lost after a median of 20.7 (11.3–35.0) months. One patient underwent liver transplantation after 14.3 months. In the Lauromacrogol group, 2 patients presented with HCC during follow-up. Nine (14.5%) patients developed new PVT or progression of previous partial PVT in the Ethiodol group and 6 (8.8%) in the Lauromacrogol group.

Rebleeding

During follow-up, 52 patients experienced rebleeding from any source, 24 (38.7%) in the Ethiodol group and 28 (41.2%) in the Lauromacrogol group. Seven patients (11.3%) and 6 patients (8.8%) underwent TIPS of surgery treatment after rebleeding, respectively. The 1- and 3-year cumulative rebleeding-free rate was 75.8% and 62.5% in the Ethiodol group versus 83.7% and 59.2% in the Lauromacrogol group, respectively (P = .797) (Fig. 2D).

FIG. 2.

The cumulative rate of freedom from any source of rebleeding. (A) Stratified by massive ascites (P = .012); (B) stratified by portal vein thrombosis (P = .002); (C) stratified by new PVT or progression of previous partial PVT (P = .000); (D) endoscopy treatment (P = .797). PVT, portal venous thrombosis.

On univariable analysis, massive ascites, PVT and new PVT, and progression of previous partial PVT were found to be significantly associated with rebleeding (Fig. 2A–C). Lauromacrogol was not found to be statistically superior to Ethiodol in preventing rebleeding (hazard ratio [HR] 1.075, 95% confidence interval [CI], 0.619–1.866, P = .797). On multivariate analysis, massive ascites and new PVT or progression of previous partial PVT were found to be independently associated with rebleeding (HR 2.380, 95% CI, 1.238–4.576, P = .009; HR 5.127, 95% CI, 2.430–10.817, P = .000) (Table 3).

Table 3.

Factors Associated with Risk of Rebleeding

	Univariable analysis			Multivariable analysis
	HR	95% CI	P value	HR	95% CI	P value
Age (per year increase)	1.006	0.985–1.028	.589
Platelet count ( × 10⁹/L)	1.003	0.999–1.007	.211
Massive ascites (present versus absent)^a	2.221	1.195–4.129	.012	2.380	1.238–4.576	.009
Splenectomy (present versus absent)	1.127	0.351–3.620	.841
Portal vein thrombosis (present versus absent)^a	2.535	1.389–4.627	.002	1.850	0.921–3.713	.084
Child–Pugh score^a	1.189	0.960–1.471	.113
MELD score	1.062	0.939–1.200	.341
Diabetes (present versus absent)	1.305	0.635–2.684	.469
New PVT or progression of previous partial PVT^a	6.920	3.568–13.422	.000	5.127	2.430–10.817	.000
Endoscopy treatment (Ethiodol^® versus Lauromacrogol^®)	1.075	0.619–1.866	.797

Variables introduced in multivariable analysis.

CI, confidence interval; HR, hazard ratio; MELD, model for end-stage liver disease; PVT, portal venous thrombosis.

Survival

During the follow-up period, 21 patients died and 1 accepted liver transplant (Table 4). Here, 10 (16.1%) patients died in the Ethiodol group compared with 11 (16.2%) in the Lauromacrogol group (HR 0.975, 95% CI, 0.414–2.297, P = .955). The 1- and 3-year cumulative survival rates were 85.5% and 83.7% in the Ethiodol group versus 86.6% and 83.5% in the Lauromacrogol group (P = .955) (Fig. 3D). Rebleeding was the most frequent cause of death in both groups (8 patients and 7 patients, respectively). Other causes of death were liver failure (5 patients) and hepatorenal syndrome (1 patient).

FIG. 3.

The cumulative survival rate. (A) Stratified by Child–Pugh score (P = .000); (B) stratified by rebleeding (P = .000); (C) stratified by new PVT or progression of previous partial PVT (P = .000); (D) stratified by endoscopy treatment (P = .955). PVT, portal venous thrombosis.

Table 4.

Clinical Outcomes in the Two Groups

	Ethiodol group (n = 62)	Lauromacrogol group (n = 68)	P value
New-onset HCC	0 (0%)	2 (2.9%)	.174
New PVT or progression of previous partial PVT	9 (14.5%)	6 (8.8%)	.310
Treatment conversion, n (%)	7 (11.3)	6 (8.8)	.640
TIPS	1	1	—
Surgery	6	5	—
Need for rescue treatment (TIPS or surgical shunt)	2 (3.2%)	3 (4.4%)	.725
Rebleeding from any source, n (%)	24 (38.7)	28 (41.2)	.774
EGV rebleeding	9	14	—
Peptic ulcer	7	8	—
Portal hypertensive gastropathy	0	0	—
Unknown	8	6	—
Morbidity, n (%)	10 (16.1)	11 (16.2)	.994
Rebleeding	8	7	—
Liver failure	2	3	—
Hepatorenal syndrome	0	1	—
Clinical follow-up (months)	38.8 (0.9–55.2)	40.1 (0.9–53.6)	.493

EGV, esophagogastric varices; HCC, hepatocellular carcinoma; PVT, portal venous thrombosis; TIPS, transjugular intrahepatic portosystemic shunt.

At univariable analysis, PVT, MELD score, new PVT or progression of previous partial PVT, and recurrent bleeding were associated with survival (Fig. 3A–C). New PVT or progression of previous partial PVT and recurrent bleeding were independent predictors in multivariable analysis (HR 10.093, 95% CI, 3.988–25.548, P = .000) (Table 5).

Table 5.

Factors Associated with Risk of Death

	Univariable analysis			Multivariable analysis
	HR	95% CI	P value	HR	95% CI	P value
Age (per year increase)^a	1.030	0.993–1.070	.117	1.038	0.997–1.081	.068
Platelet count ( × 10⁹/L)^a	1.004	0.999–1.009	.150
Massive ascites (present versus absent)^a	2.023	0.784–5.217	.145
Splenectomy (present versus absent)	1.867	0.435–8.019	.401
Portal vein thrombosis (present versus absent)^a	6.337	2.680–14.985	.000
Child–Pugh score	1.181	0.844–1.652	.333
MELD score^a	1.200	1.006–1.432	.043
Diabetes (present versus absent)	1.444	0.486–4.292	.509
Endoscopy treatment (Ethiodol versus Lauromacrogol)	0.975	0.414–2.297	.955
New PVT or progression of previous partial PVT^a	22.977	9.320–56.646	.000	10.093	3.988–25.548	.000
Rebleeding (present versus absent)^a	36.910	4.949–275.266	.000	22.367	2.869–174.381	.003

Variables introduced in multivariable analysis.

CI, confidence interval; HR, hazard ratio; MELD, model for end-stage liver disease; PVT, portal venous thrombosis.

Discussion

The present retrospective cohort study compared the efficacy and prognosis of endoscopic cyanoacrylate treatment with or without Lauromacrogol for gastric variceal bleeding in a long follow-up period (median 38.8 and 40.1 months, separately). The 1- and 3-year cumulative rebleed-free rates were 75.8% and 62.5% in the Ethiodol group versus 83.7% and 59.2% in the Lauromacrogol group, which is consistent with the findings of a previous study.²³ Twenty-one patients died during the follow-up, of which rebleeding was the main cause of death in the two groups.

In this real-world study, one of the main findings of this study was that the volume of cyanoacrylate in the Lauromacrogol group was significantly lower than in the Ethiodol group for GV obturation, which is consistent with our previous research.²⁰ These results indicated that Lauromacrogol could replace Ethiodol with a reduced volume for GV injection without changing variceal eradication rates. No clinically significant ectopic embolization was observed in the present study. One possible cause may be that patients with spleen-renal or gastro-renal determined by preoperative CTA scan underwent balloon-occluded endoscopic gastric variceal obturation.

Unfortunately, rebleeding or survival rates had no statistically significant difference between the two groups. The lack of statistical significance was attributable to factors other than GV obliteration, most notably portal pressure and PVT, which may also affect rebleeding and survival. The measurement of hepatic venous pressure gradient (HVPG) has been strongly indicated as a useful predictor of rebleeding and survival.²⁴ However, it was not performed in the present study. Cirrhotics with ascites had a significantly higher HVPG than those without ascites.²⁵ Massive ascites were found to be a determinant of rebleeding in the present study.

In this study, the incidence of PVT in cirrhotic patients was 18%, which was consistent with previous reports.^26,27 The rebleeding rate was higher in patients with PVT or new PVT or progression of previous partial PVT during follow-up than in those without. Multivariate analysis of potential risk factors revealed that new PVT or progression of previous partial PVT was an independent predictor of rebleeding. When patients with cirrhosis have PVT, complications related to portal hypertension, such as ascites and variceal bleeding, become more serious.^27,28 Patients with PVT have a worse prognosis and a higher rate of rebleeding after endoscopy treatment than patients without PVT in the short-term follow-up.²⁹ Previous studies have shown that exacerbation of thrombosis not only increases the rate of rebleeding in patients but also increases mortality.^30,31 In the present study, new thrombosis and exacerbation of thrombosis were also an independent risk factors for death. At present, there is still no unified consensus regarding the best treatment for nontumor PVT patients with cirrhosis. Anticoagulant therapy is the main treatment for recanalization of chronic PVT.^26,32 Previous studies have suggested that anticoagulation after endoscopic therapy can reduce mortality in patients with EGV.³³ At present, no patients with thrombosis in the present study received anticoagulant therapy because of the risk of bleeding.

The advantage of the present trial is that it was a cohort study with a large sample size, long follow-up, and minimal loss rate. However, the disadvantage is that no measurement of preoperative HVPG was performed. HVPG can help treatment strategy compared with other therapy.^34,35 In this way, HVPG-guided trails of cyanoacrylate with Lauromacrogol are urgently needed.

In summary, results showed that the Lauromacrogol group significantly reduced the volume of cyanoacrylate use (2.1 ± 1.2 versus 1.6 ± 0.8 mL, P = .029), but there were no significant differences in rebleeding-free or survival rate between the two groups. Notably, exacerbation of PVT was the independent factor for both rebleeding and death in GV patients. All these findings suggest that endoscopic cyanoacrylate injection with Lauromacrogol offered no clear advantage over standard treatment with Ethiodol in long-term follow-up, but it might minimize the required dosage of cyanoacrylate, and this potential benefit, which still requires confirmation, might render the treatment more cost-effective. Further research is needed to illustrate the safety and efficacy of anticoagulation treatment in patients with EGV bleeding.

Funding Information

Shanghai Science and Technology Commission Science and Technology Support Project (19411970200) and Zhongshan Hospital Clinical Research Special Fund (2016ZSLC08).

Footnotes

Acknowledgment

We thank all the staff in the Endoscopy Center and Endoscopy Research Institute of Zhongshan Hospital for assistance.

Disclosure Statement

No competing financial interests exist.

References

Sharara

, Rockey

. Gastroesophageal variceal hemorrhage. N Engl J Med, 2001; 345:669–681.

Sarin

, Lahoti

, Saxena

, Murthy

, Makwana

. Prevalence, classification and natural history of gastric varices: A long-term follow-up study in 568 portal hypertension patients. Hepatology, 1992; 16:1343–1349.

de Franchis

, Primignani

. Natural history of portal hypertension in patients with cirrhosis. Clin Liver Dis, 2001; 5:645–663.

Greig

, Garden

, Anderson

, Carter

. Management of gastric variceal haemorrhage. Br J Surg, 1990; 77:297–299.

, Yeh

, Chen

. Pharmacologic efficacy in gastric variceal rebleeding and survival: Including multivariate analysis. J Clin Gastroenterol, 2002; 35:127–132.

Evrard

, Dumonceau

, Delhaye

, Golstein

, Deviere

, Le Moine

. Endoscopic histoacryl obliteration vs. propranolol in the prevention of esophagogastric variceal rebleeding: A randomized trial. Endoscopy, 2003; 35:729–735.

Saraswat

, Verma

. Gluing gastric varices in 2012: Lessons learnt over 25 years. J Clin Exp Hepatol, 2012; 2:55–69.

Seewald

, Ang

, Imazu

, et al. A standardized injection technique and regimen ensures success and safety of N-butyl-2-cyanoacrylate injection for the treatment of gastric fundal varices (with videos). Gastrointest Endosc, 2008; 68:447–454.

de Franchis

. Expanding consensus in portal hypertension: Report of the Baveno VI Consensus Workshop: Stratifying risk and individualizing care for portal hypertension. J Hepatol, 2015; 63:743–752.

10.

Dworzynski

, Pollit

, Kelsey

, Higgins

, Palmer

. Management of acute upper gastrointestinal bleeding: Summary of NICE guidance. BMJ, 2012; 344:e3412.

11.

Garcia-Tsao

, Sanyal

, Grace

, Carey

. Prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis. Hepatology, 2007; 46:922–938.

12.

Sarin

, Jain

, Gupta

. A randomized controlled trial of cyanoacrylate versus alcohol injection in patients with isolated fundic varices. Am J Gastroenterol, 2002; 97:1010–1015.

13.

Akahoshi

, Hashizume

, Shimabukuro

, et al. Long-term results of endoscopic Histoacryl injection sclerotherapy for gastric variceal bleeding: A 10-year experience. Surgery, 2002; 131:S176–S181.

14.

Hou

, Lin

, Lee

, Liao

, Lee

. A randomized trial of endoscopic cyanoacrylate injection for acute gastric variceal bleeding: 0.5 mL versus 1.0 mL. Gastrointest Endosc, 2009; 70:668–675.

15.

Fry

, Neumann

, Olano

, Malfertheiner

, Monkemuller

. Efficacy, complications and clinical outcomes of endoscopic sclerotherapy with N-butyl-2-cyanoacrylate for bleeding gastric varices. Dig Dis, 2008; 26:300–303.

16.

Wahl

, Lammer

, Conen

, Schlumpf

, Bock

. Septic complications after injection of N-butyl-2-cyanoacrylate: Report of two cases and review. Gastrointest Endosc, 2004; 59:911–916.

17.

Sato

, Yamazaki

. Evaluation of therapeutic effects and serious complications following endoscopic obliterative therapy with Histoacryl. Clin Exp Gastroenterol, 2010; 3:91–95.

18.

Kumar

, Singh

, Madan

, Garg

, Acharya

. Undiluted N-butyl cyanoacrylate is safe and effective for gastric variceal bleeding. Gastrointest Endosc, 2010; 72:721–727.

19.

Garcia-Tsao

, Bosch

, Groszmann

. Portal hypertension and variceal bleeding—unresolved issues. Summary of an American Association for the study of liver diseases and European Association for the study of the liver single-topic conference. Hepatology, 2008; 47:1764–1772.

20.

Zeng

, Ma

, Tzeng

, et al. Endoscopic cyanoacrylate injection with or without lauromacrogol for gastric varices: A randomized pilot study. J Gastroenterol Hepatol, 2017; 32:631–638.

21.

Cheng

, Wang

, Li

, Lin

, Yeo

, Jin

. Low incidence of complications from endoscopic gastric variceal obturation with butyl cyanoacrylate. Clin Gastroenterol Hepatol, 2010; 8:760–766.

22.

Chen

, Zeng

, Ma

, et al. Randomized controlled trial comparing endoscopic ligation with or without sclerotherapy for secondary prophylaxis of variceal bleeding. Eur J Gastroenterol Hepatol, 2016; 28:95–100.

23.

Rengstorff

, Binmoeller

. A pilot study of 2-octyl cyanoacrylate injection for treatment of gastric fundal varices in humans. Gastrointest Endosc, 2004; 59:553–558.

24.

Bosch

, Abraldes

, Berzigotti

, Garcia-Pagan

. The clinical use of HVPG measurements in chronic liver disease. Nat Rev Gastroenterol Hepatol, 2009; 6:573–582.

25.

Wadhawan

, Dubey

, Sharma

, Sarin

. Hepatic venous pressure gradient in cirrhosis: Correlation with the size of varices, bleeding, ascites, and child's status. Dig Dis Sci, 2006; 51:2264–2269.

26.

Tsochatzis

, Senzolo

, Germani

, Gatt

, Burroughs

. Systematic review: Portal vein thrombosis in cirrhosis. Aliment Pharmacol Ther, 2010; 31:366–374.

27.

Ertugrul

, Koklu

, Basar

, et al. Thrombosis of the portal venous system: A prospective study. J Clin Gastroenterol, 2008; 42:835–838.

28.

Congly

, Lee

. Portal vein thrombosis: Should anticoagulation be used?. Curr Gastroenterol Rep, 2013; 15:306.

29.

Amitrano

, Guardascione

, Scaglione

, et al. Splanchnic vein thrombosis and variceal rebleeding in patients with cirrhosis. Eur J Gastroenterol Hepatol, 2012; 24:1381–1385.

30.

Girleanu

, Stanciu

, Cojocariu

, Boiculese

, Singeap

, Trifan

. Natural course of nonmalignant partial portal vein thrombosis in cirrhotic patients. Saudi J Gastroenterol, 2014; 20:288–292.

31.

Englesbe

, Kubus

, Muhammad

, et al. Portal vein thrombosis and survival in patients with cirrhosis. Liver Transpl, 2010; 16:83–90.

32.

DeLeve

, Valla

, Garcia-Tsao

. Vascular disorders of the liver. Hepatology, 2009; 49:1729–1764.

33.

Amitrano

, Guardascione

, Manguso

, et al. The effectiveness of current acute variceal bleed treatments in unselected cirrhotic patients: Refining short-term prognosis and risk factors. Am J Gastroenterol, 2012; 107:1872–1878.

34.

Mishra

, Chander

, Kumar

, Sarin

. Endoscopic cyanoacrylate injection versus beta-blocker for secondary prophylaxis of gastric variceal bleed: A randomised controlled trial. Gut, 2010; 59:729–735.

35.

Procaccini

, Al-Osaimi

, Northup

, Argo

, Caldwell

. Endoscopic cyanoacrylate versus transjugular intrahepatic portosystemic shunt for gastric variceal bleeding: A single-center U.S. analysis. Gastrointest Endosc, 2009; 70:881–887.