Open Access

Comparison of the effectiveness of 11 mainstay treatments for secondary prophylaxis of variceal bleeding in patients with cirrhosis: A network meta‑analysis

  • Authors:
    • Yu Kong
    • Liang Shi
  • View Affiliations

  • Published online on: April 1, 2020     https://doi.org/10.3892/etm.2020.8633
  • Pages: 3479-3496
  • Copyright: © Kong et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )
Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )


Abstract

The purpose of the present study was to compare the effectiveness of the transjugular intrahepatic portosystemic shunt (TIPS), endoscopic options, medications and mainstay combination therapies for patients with cirrhosis who have had at least one episode of variceal haemorrhage. The PubMed, Embase, Cochrane Library and Web of Science databases, as well as the reference lists of relevant articles, were searched to identify eligible studies. P‑scores, that were based solely on the point estimates and standard errors of the network estimates, were performed to rank all treatments, on a scale from 0 (worst) to 1 (best). The odds ratio (OR) was determined to assess effects on mortality, treatment failure and bleeding from gastroesophageal ulcers. A total of 43 randomized controlled trials comprising 3,787 adult patients were included. In total, 26 (61%) trials adopted concealed randomization, while most studies did not specify blinding. The drug combination of nadolol and isosorbide mononitrate (ISMN) ranked first for lowering risks of overall mortality (P‑score=0.8162), mortality due to liver failure (P‑score=0.7536) and bleeding from gastroesophageal ulcers (P‑score=0.7536). This combination was determined to be superior to endoscopic sclerotherapy (ES) alone (OR=0.63, 95% CI: 0.42‑0.94) and TIPS alone in reducing overall mortality (OR=0.62, 95% CI: 0.40‑0.96). ES was more likely to increase treatment failure compared with TIPS, endoscopic variceal ligation (EVL), ES plus EVL, EVL plus nadolol/propranolol plus ISMN and nadolol/propranolol plus ISMN. In conclusion, the present network meta‑analysis suggested that for a decreased mortality due to variceal rebleeding in patients with cirrhosis, nadolol plus ISMN may be a preferable choice, while ES is associated with a higher risk of unfavourable treatment outcomes. Further well‑controlled studies are required to further elucidate the appropriate treatment options.

Introduction

Approximately 30% of patients with cirrhosis have oesophageal varices at the time of diagnosis; this proportion increases with time and reaches 90% after ~10 years (1). Patients with oesophageal varices have a high tendency to develop bleeding. Only 10-20% of variceal bleeding occurs from gastric varices, but the associated outcome is worse than that of bleeding from oesophageal varices (2-5). Patients surviving a variceal bleed are at high risk of rebleeding (>60% in the first year) and the mortality of each rebleeding episode is ~20% (6). Therefore, prevention of recurrent variceal bleeding is important for patients with cirrhosis.

For secondary prophylaxis of variceal bleeding, the goal of improving outcomes is evolving, since therapy in these cases attempts to reduce the risk of death, and thus prevent the onset of complications of cirrhosis that may lead to death (1). The treatment effectiveness of secondary preventions, including endoscopic ligation or endoscopic sclerotherapy (ES), drug therapies [non-selective β-blockers (NSBB) with or without isosorbide mononitrate (ISMN)] and transjugular intrahepatic portosystemic shunts (TIPS) is an area of interest, but at present, a firm consensus as to the most effective treatment has not been reached. Several randomized controlled trials (RCTs) have investigated treatment outcomes in terms of mortality, complications and adverse effects (7-12). A previous study compared endoscopic variceal ligation (EVL) with a combination of EVL and nadolol and identified that adverse effects more frequently occurred in the EVL plus nadolol group (0.03 vs. 33%) (12). Another trial compared nadolol plus ISMN alone with EVL plus the drug combination and observed that the combination of EVL and drugs led to more adverse effects (62 vs. 32%), but there were no significant differences in either mortality or the causes of death (11). However, a previous direct meta-analysis comprising 925 patients comparing endoscopic therapy with a combination of BB and endoscopic therapy identified that mortality at 24 months was significantly lower in the combined treatment group (13). In addition to inconsistent results among the previous trials and analyses, to the best of our knowledge, there has been no previous network meta-analysis to compare treatment outcomes. Therefore, the present study was performed to compare the effectiveness of standard treatments for the secondary prevention of variceal bleeding in patients with cirrhosis through a network meta-analysis. The specific treatments studied were TIPS, endoscopic therapy (EVL alone or ES alone), a combination of EVL and ES, a combination of EVL/ES and NSBB (propranolol and nadolol) with or without ISMN, as well as a combination of NSBB and ISMN.

Materials and methods

Literature search

Searches were performed in the electronic PubMed, Cochrane Library, Embase and Web of Science databases in February 2018. The following search terms were used: ‘Cirrhotic patients’, ‘patients with cirrhosis’, ‘liver cirrhosis’, ‘haemorrhage’, ‘bleeding’, ‘rebleeding’, ‘variceal’, ‘oesophageal varices’, ‘endoscopic variceal ligation’, ‘endoscopic band ligation’, ‘endoscopic ligation’, ‘endoscopic sclerotherapy’, ‘sclerotherapy’, ‘endoscopic therapy’, ‘vasoconstrictors’, ‘venodilators’, ‘adrenergic beta antagonist’, ‘adrenergic-beta antagonist’, ‘adrenergic beta-antagonist’, ‘adrenergic-beta-antagonist’, ‘nitrate’, ‘beta-blocker’, ‘isosorbide mononitrate’, ‘placebo’, ‘TIPS’, ‘transjugular intrahepatic portosystemic shunt’ and ‘randomized controlled trial’. Manual searches of reference lists of relevant articles were also performed to identify additional studies. Only RCTs were included.

Inclusion and exclusion criteria

RCTs, irrespective of publication status, were included if they investigated endoscopic therapy with various combinations of NSBB and ISMN, or TIPS alone among adult patients with cirrhosis, who had at least one previous episode of upper gastrointestinal bleeding. Trials fulfilling the following criteria were excluded: i) Focus on primary prevention of variceal bleeding; ii) inclusion of pediatric patients or patients without cirrhosis; iii) comparison of only one of the aforementioned treatment regimens with other treatment(s), as it was impossible to make a network comparison; or iv) a clearly irrelevant topic, e.g. nutrition after variceal bleeding.

Study selection

Only RCTs whose reports were available in English or Chinese were included. If a trial was designed with more than two treatment arms, at least two of the arms had to match the scope of the present study.

Data extraction

According to the newly published guidelines of the American Association for the Study of Liver Diseases (AASLD) and consensus (14), therapies for secondary prophylaxis must account for the presence or absence of other complications of cirrhosis. In patients with a low risk of death (those with variceal haemorrhage as the sole complication of cirrhosis), the objective of therapy should be the prevention of an additional complication, whereas in patients with a high risk of death (those with variceal haemorrhage and other decompensating events), the objective of therapy should be to improve survival (15,16). Mortality (overall mortality, mortality due to rebleeding and mortality due to liver failure), treatment failure and complications (bleeding from gastroesophageal ulcer) were analyzed.

Data of treatment failure were analyzed when clearly stated in the literature, with exclusion of data that satisfied certain criteria but lacked declaration of treatment failure. Authors of the included trials were not approached for further data due to the large number of RCTs selected and acquisition of adequate data associated with treatment outcomes. The primary outcomes were overall mortality, mortality due to rebleeding, including but not limited to recurrent variceal bleeding and mortality due to liver failure. Overall mortality was defined as death that occurred during the trial treatment or follow-up caused by disease progression or treatment complications. Secondary outcomes were treatment failure and bleeding from gastroesophageal ulcers, including but not limited to post-banding ulcers.

Methodological quality

A bias assessment was performed for the included trials by evaluating randomization, completion of trials and blinding. The major targets were concealment randomization, participant blinding, health care provider blinding, data collector blinding, outcome assessor blinding and early trial cessation.

Randomization was considered concealed if it involved a third independent party or person not involved in the treatment of patients, opaque sealed envelopes or a similar method. Trials were not considered to feature early cessation unless premature termination was specifically announced in the article.

Statistical analysis

The odds ratio (OR) was used to denote the results with a 95% CI, indicating the strength of association between treatments and outcomes. An OR<1 represents the benefit of the comparison group compared with the control group. Pooled ORs and their 95% CIs were also calculated. Statistical significance was established with a two-sided P<0.05 or a CI that did not include a value of 1. The risk ratio was not used to measure outcomes due to limited data regarding the number of events among the selected trials.

To assess the comparability of the included trials, a heterogeneity analysis was performed. Since inconsistency is a source of heterogeneity in network meta-analyses, a generalized Cochran's Q statistic (Qtotal) and I2 statistic were adopted for assessment of homogeneity and consistency assumptions. Statistical heterogeneity was considered significant when P<0.10 for the Q-test or I2<50%. The network meta-analysis used fixed-effects models with I2 values of 0% for overall mortality, mortality due to rebleeding, mortality due to liver failure and bleeding from a gastroesophageal ulcer, and I2=29.4% for treatment failure.

All treatments were ranked according to P-score, which is on a scale from 0 (worst) to 1 (best). P-scores are based solely on the point estimates and standard errors of the most frequent network meta-analysis estimates under normality assumption, and can easily be calculated as means of one-sided P-values. They measure the extent of certainty that a treatment is better than another treatment, averaged over all competing treatments (17). Sensitivity analysis was performed by removal of trials with a mean follow-up of <6 months. The network meta-analyses were performed using R 3.3.1 along with the ‘netmeta’ package by Schwarzer et al (18).

Results

Search results

Electronic and manual searches identified 1,293 records in total. Following screening of titles and abstracts, 861 references were excluded and the remainder was subjected to full-text screening. Among the excluded studies were duplicates, non-RCTs, trials investigating other treatments, or those covering different topics or focusing on primary prevention of variceal bleeding, due to inadequate data for the present study or randomizing patients without cirrhosis. A previous trial investigating the effects of carvedilol plus EVL was excluded, as it assessed hemodynamic responses but not mortality (19). The screening process followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and is depicted in a flow chart in Fig. 1.

Characteristics of the studies included

A total of 43 trials (20-62) with a total sample size of 3,787 patients with cirrhosis were included for quantitative network meta-analysis. In total, 5 references were published or available as abstracts (22,35,39,42,43) and the remainder were available in full text. A previous trial had 4 treatment arms (44), among which 3 (EVL alone, EVL plus propranolol plus ISMN and propranolol plus ISMN) were included from the present study. Another had 3 arms (22), of which 2 (EVL alone and propranolol plus ISMN) were included, and the arm of carvedilol treatment alone was excluded. All of the other trials were designed with 2 treatment arms. The proportion of patients with cirrhosis was 100% in all of the trials. The baseline characteristics of the trials are presented in Table I. Patient inclusion and exclusion criteria varied slightly across trials, but patients were generally eligible if they were adults with cirrhosis with at least one episode of endoscopic-proven oesophageal or gastric variceal bleeding. Exclusion criteria included hepatocarcinoma, non-bleeding varices, existing multi-organ failure and lack of cirrhosis. A total of 30 of the 43 studies had a mean follow-up time of >2 years, as presented in Table I. In total, five trials were excluded from the sensitivity analysis due to follow-up times that were unknown or <6 months (43,47,55,60,61). TIPS alone was used as the comparative treatment in the forest plots, since it is a recommended surgery for secondary prophylaxis according to the newest UK guidelines (5).

Table I

Characteristic of randomized controlled trials included in the network meta-analysis.

Table I

Characteristic of randomized controlled trials included in the network meta-analysis.

First author (year)Treatment groupsPatients (n)Mean age (years)Females (%)Varices, grade or size, nChild-Pugh class (A/B/C)Mean Pugh scoreMean follow-up (days)No. of subjectsa(Refs.)
Sauerbruch (2015)TIPS905531NA45% A6.990627/NA/11/NA/NA(20)
  Propranolol+ISMN955534NA49% A790625/NA/9/NA/NA 
Luo (2015)TIPS375149NA0/25/128.7669012/1/9/NA/0(21)
  EVL+Propranolol365033NA0/24/128.8963717/3/10/NA/2 
Kumar (2015)EVL5644NANANA8.65009/NA/NA/NA/NA(22)
  Propranolol+ISMN3944NANANA8.65008/NA/NA/NA/NA 
Kong (2015)EVL205430F2:10; F3:109/11/0NA4880/NA/NA/NA/NA(23)
 ES185350F2:9; F3:118/10/0NA4880/NA/NA/NA/NA 
Kumar (2009) EVL+Propranolol+ISMN884215MG 3.1:8835/31/103.24582/NA/NA/NA/NA(24)
 EVL894112MG 3.2:8926/34/153.24583/NA/NA/NA/NA 
Sauer (2002)TIPS435437MG 2.4:4315/16/127.91,4988/1/4/1/NA(25)
 EVL425545MG 2.6:4210/19/138.21,3157/2/2/5/NA 
Gülberg (2002)TIPS285729NA11/15/2NA7304/1/2/NA/NA(26)
 EVL265627NA10/12/4NA7304/2/1/NA/NA 
Escorsell (2002) Propranolol+ISMN465620NA0/28/16NA47011/4/2/NA/NA(27)
 TIPS475730NA0/30/17NA43613/3/5/NA/NA 
Villanueva (2001)EVL725835G1: 1; G2:49; G3:2211/43/188.466630/10/12/23/7(28)
 Nadolol+ISMN726040G1: 2; G2:41; G3:2919/39/147.960523/4/10/12/0 
Pomier-LayrarguesEVL395431NANA9.858116/3/9/NA/NA(29)
(2001)TIPS415329NANA9.667817/0/10/NA/NA 
Narahara (2001)TIPS385116NANA6.81,11611/NA/7/NA/0(30)
 ES405525NANA7.41,0477/NA/NA/NA/NA 
Hou (2001)EVL475534NA14/17/168.23516/NA/4/NA/NA(31)
 EVL+ES475923NA11/23/1383367/NA/5/NA/NA 
Meddi (1999)TIPS186128NANA7.65454/NA/NA/NA/NA(32)
 ES205830NANA7.65452/NA/NA/NA/NA 
Villarreal (1999)TIPS225832NA5/10/78.67603/0/1/0/NA(33)
 ES24558NA3/14/78.85038/6/1/7/NA 
Sauer (1997)TIPS425464NA15/18/9753012/0/7/NA/0(34)
 ES416051NA12/18/11758411/5/3/NA/3 
Sanyal (1997)TIPS41NANANANANA1,00012/NA/2/NA/NA(35)
 ES39NANANANANA1,0007/NA/2/NA/NA 
Villanueva (1996)Nadolol+ISMN435833G1:3; G2:30; G3:109/27/7NA5454/0/2/3/1(36)
 ES436033G1:2; G2:31; G3:1011/22/10NA5459/2/5/16/7 
Li (2010)ES365458G1:0; G2:0; G3:36NA7.36181.57/NA/NA/4/7(37)
 EVL275126G1: 0; G2:0; G3:27NA6.86181.51/NA/NA/0/0 
Holster (2016) EVL+Propranolol355449NA13/18/47.37089/NA/0/12/2(38)
 TIPS375634NA13/19/57.570812/NA/3/14/1 
van Buuren (2000)TIPS19NANANANANA632NA/NA/NA/1/NA(39)
 EVL18NANANANANA632NA/NA/NA/0/NA 
Cabrera (1996)ES325628NA14/16/27.224545/3/NA/10/5(40)
 TIPS315635NA14/13/47.14546/0/NA/0/0 
Dobrucali (1998)EVL505522G2:1; G3:5; G4:4418/23/9NA908NA/NA/4/NA/NA(41)
 ES284243G2:2; G3:3; G4:237/13/8NA908NA/NA/0/NA/NA 
Holster (2013)TIPS355444NANA7.45458/NA/NA/NA/NA(42)
 EVL365444NANA7.45457/NA/NA/NA/NA 
Merli (1994)TIPS21NANANANANANA2/NA/NA/NA/NA(43)
 ES17NANANANANANA1/NA/NA/NA/NA 
Ahmad (2009) Propranolol+ISMN355240G1+G2:12; G3:232/19/149.111826/4/4/9/0(44)
 EVL395336G1+G2:15; G3:247/23/98.281828/2/3/12/3 
  EVL+propranolol+ISMN375019G1+G2:15; G3:224/27/68.321827/2/3/8/2 
Argonz (2000)EVL415322G2:27; G3:1414/23/4NA33716/9/1//1(45)
 EVL+ES395323G2:22; G3:1711/26/2NA38612/4/3//2 
Avgerinos (1997)EVL375516S:1; M:19; L:1723/11/37.74728/0/NA/5/2(46)
 ES405620S:3; M:19; L:1827/10/384548/2/NA/4/4 
Avgerinos (2004)ES255224G1:3; G2:13; G3:96/8/119.4427/4/NA/13/NA(47)
 EVL255620G1:3; G2:12; G3:107/6/129.2425/2/NA/6/NA 
Avgerinos (1993)ES405920S:3; M:19; L:1830/8/27.31,0049/2/3/NA/NA(48)
 ES+Propranolol455836S:4; M:19; L:2233/8/47.61,0358/3/3/NA/NA 
Baroncini (1997)ES546131F3:36; F2:1818/22/14853412/3/NA/NA/NA(49)
 EVL576333F3:42; F2:1517/24/167.749612/1/NA/NA/NA 
Chen (2016)EVL485633G1:18; G2:26; G3:419/29/0NA1821/1/NA/0/NA(16)
 EVL+ES485435G1:19; G2:23; G3:620/28/0NA1823/2/NA/0/NA 
García-Pagán (2009)Nadolol+ISMN785632L:7318/42/188.145415/8/1/22/0(50)
  EVL+Nadolol+ISMN805719L:7316/46/188.245416/5/4/17/4 
Hou (1995)ES676121F3:49; F2+F1:1815/29/238.229311/4/6/NA(51)
 EVL676019F3:55; F2+F1:1217/21/298.828714/0/13/NA/NA 
Hou (2000)ES706019F3:51; F2+F1:1917/34/1981,91811/4/NA/NA/2(52)
 EVL716021F3:57; F2+F1:1420/26/258.41,8189/1/NA/NA/3 
Lo (1995)ES595417NA7/24/288.929019/9/6/NA/NA(53)
 EVL615721NA9/22/309.531010/4/6/NA/NA 
Lo (1997)EVL375314F3:27; F2:102/13/22NA307/3/3/NA/NA(54)
 ES345512F3:26; F2:83/11/20NA3012/6/3/NA/NA 
de la Peña (2005)EVL+Nadolol436023G1:10; G2:26; G3:76/20/11NA5295/0/2/2/1(55)
 EVL376027G1:7; G2:17; G3:136/25/12NA4544/1/2/7/3 
Peña (1999)ES465935G1:10; G2:31; G3:512/22/13NA54510/NA/3/11/2(56)
 EVL425919G1:6; G2:25; G3:1110/22/10NA4848/NA/2/5/3 
Romero (2006)Nadolol+ISMN575135NA40/44/16734811/9/NA/18/0(57)
 EVL525333NA32/58/10736310/6/NA/12/3 
Stiegmann (1992)ES655322G1:9; G2:31; G3:2520/32/119.930329/8/NA/NA/NA(58)
 EVL645117G1:8; G2:35; G3:2122/30/129.430318/3/NA/NA/NA 
Umehara (1999)EVL+ES2558NAF3:9; F2:1610/11/4NA6113/0/NA/NA/NA(59)
 EVL2659NAF3:8; F2:187/13/6NA6294/0/NA/NA/NA 
Viazis (2002)EVL366442S:9; M:17; L:104/18/14NA50NA/NA/NA/NA/NA(60)
 ES376246S:10; M:15; L:126/16/15NA59NA/NA/NA/NA/NA 
Vinel (1992)ES355743F3:30; F2:5NA9.1925/3/NA/NA/NA(61)
 ES+Propranolol395523F3:20; F2:19NA9.21025/4/NA/NA/NA 

[i] G, Grade; Grade 1, visible only during 1 phase of respiration or on performance of valsalva maneuver; Grade 2, visible during both phases of respiration; Grade 3, 3-6 mm; Grade 4, >6 mm; S, small: Varix is flush with the wall of the oesphagus; M, medium: Protrusion of varix no further than halfway to the center; L, large: protrusion more than half-way to the center of the lumen; F, varicies; F1, straight, small-caliber varices; F2, moderately enlarged, beady varices; F3, markedly enlarged, nodular or tumor-shaped varices.

[ii] aNumber of subjects for overall mortality, mortality due to rebleeding, mortality due to liver failure and treatment failure, respectively. TIPS, transjugular intrahepatic portosystemic shunt; ES, endoscopic injection sclerotherapy; EVL, endoscopic variceal ligation; ISMN, isosorbide mononitrate; MG, mean grade; NA, not available.

Bias assessment

Risk of bias assessment for the RCTs included was performed following the PRISMA recommendations; the results are presented in Table SI. A total of 26 (61%) trials (20-22,24-30,34,36,38,44,47,16,50,56,61) adopted concealed randomization via sealed opaque envelopes, by using central randomization or through an independent person not involved in the treatment of the patients. Only one trial declared early cessation (54). A total of two trials were open labelled (20,24) and two trials reported using outcome assessors under blinded conditions (27,16). Blinding of the remaining trials was not specified.

Overall mortality

In total, 40 trials with a total of 3,599 patients reported overall mortality, involving all 11 treatment regimens. Fig. S1A illustrates the evidence networks connecting the regimens. Nadolol plus ISMN also had the highest P-score (P-score=0.8162, Table II) with the largest probability to reduce mortality when compared with the other treatments. No statistical heterogeneity was observed (Heterogeneity I2=0%; Cochran's test P=0.9618, Table III) in this outcome measure. The fixed-effects model analysis suggested that nadolol plus ISMN was significantly more effective than TIPS alone (OR=0.62, 95% CI: 0.40-0.96, Table III), as presented in Fig. 2A. Pairwise comparisons indicated that nadolol plus ISMN and EVL alone were significantly more effective than ES alone in reducing overall mortality (OR=0.63, 95% CI: 0.42-0.94; OR=0.80, 95% CI: 0.65-0.99, respectively, Table III), while differences among other treatments were not statistically significant.

Table II

Ranking for efficacy of 11 potential treatment options.

Table II

Ranking for efficacy of 11 potential treatment options.

 Overall mortalityMortality due to rebleedingMortality due to liver failureTreatment failureBleeding from gastroesophageal ulcer
RankP-scoreTreatmentP-scoreTreatmentP-scoreTreatmentP-scoreTreatmentP-scoreTreatment
10.8162Nadolol+ISMN0.8276TIPS alone0.7536Nadolol+ISMN0.8071 EVL+propranolol0.7536Nadolol+ISMN
20.7135 EVL+nadolol+ISMN0.7265EVL+nadolol0.6964ES alone0.7938TIPS alone0.6964ES alone
30.5983ES+EVL0.6931 EVL+nadolol+ISMN0.6651ES+propranolol0.7932EVL+nadolol0.6651ES+propranolol
40.5887EVL alone0.6833 Propranolol+ISMN0.6044 EVL+propranolol+ISMN0.5630 EVL+nadolol+ISMN0.6044 EVL+propranolol+ISMN
50.5843 EVL+propranolol+ISMN0.6047ES+EVL0.5755EVL+nadolol0.5189ES+EVL0.5755EVL+nadolol
60.5218ES+propranolol0.5041Nadolol+ISMN0.5230EVL alone0.4520 EVL+propranolol+ISMN0.5230EVL alone
70.4847EVL+nadolol0.4377 EVL+propranolol0.5226 Propranolol+ISMN0.4328Nadolol+ISMN0.5226 Propranolol+ISMN
80.4744 Propranolol+ISMN0.4331 EVL+propranolol+ISMN0.3439 EVL+propranolol0.3680 Propranolol+ISMN0.3439 EVL+propranolol
90.2811ES alone0.3934EVL alone0.3022TIPS alone0.2513EVL alone0.3022TIPS alone
100.2624TIPS alone0.1125ES alone0.2966ES+EVL0.0199ES alone0.2966ES+EVL
110.1747 EVL+propranolol0.0842ES+propranolol0.2167 EVL+nadolol+ISMN  0.2167 EVL+nadolol+ISMN

[i] Ranking indicates the probability to be the best treatment. Rank 1 with the highest P-score is the best treatment and rank N with the lowest P-score is the worst one. TIPS, transjugular intrahepatic portosystemic shunt; ES, endoscopic sclerotherapy; EVL, endoscopic variceal ligation; ISMN, isosorbide-5-mononitrate.

Table III

Network meta-analysis results of pairwise comparisons (Odds ratios with 95% CI).

Table III

Network meta-analysis results of pairwise comparisons (Odds ratios with 95% CI).

A, Overall mortality (Heterogeneity I2=0%, P=0.9618), fixed-effects model
GroupES aloneES+EVLES+propranololEVL aloneEVL+nadolol EVL+nadolol+ISMN EVL+propranolol EVL+Propranolol+ISMNNadolol+ISMN Propranolol+ISMNTIPS alone
ES alone-1.29 [0.79,2.10]1.21 [0.61,2.40]1.25 [1.01,1.54]1.16 [0.33,4.07]1.53 [0.72,3.23]0.85 [0.50,1.45]1.30 [0.60,2.83]1.59 [1.06,2.38]1.14 [0.77,1.71]0.98 [0.74,1.29]
ES+EVL0.78 [0.48,1.27]-0.94 [0.40,2.18]0.97 [0.62,1.52]0.90 [0.24,3.36]1.18 [0.50,2.78]0.66 [0.33,1.32]1.01 [0.42,2.43]1.23 [0.69,2.19]0.89 [0.49,1.60]0.76 [0.45,1.28]
ES+propranolol0.83 [0.42,1.64]1.06 [0.46,2.47]-1.03 [0.50,2.11]0.96 [0.23,4.01]1.26 [0.46,3.48]0.70 [0.30,1.68]1.08 [0.38,3.03]1.31 [0.59,2.90]0.95 [0.43,2.09]0.81 [0.39,1.70]
EVL alone0.80 [0.65,0.99]1.03 [0.66,1.63]0.97 [0.47,1.99]-0.93 [0.27,3.21]1.22 [0.59,2.53]0.68 [0.40,1.16]1.05 [0.49,2.22]1.27 [0.89,1.82]0.92 [0.63,1.35]0.79 [0.60,1.03]
EVL+nadolol0.86 [0.25,3.03]1.11 [0.30,4.16]1.04 [0.25,4.37]1.08 [0.31,3.71]-1.32 [0.31,5.54]0.74 [0.19,2.83]1.12 [0.26,4.79]1.37 [0.38,4.98]0.99 [0.27,3.61]0.85 [0.24,3.01]
EVL+nadolol+ISMN0.66 [0.31,1.39]0.84 [0.36,1.99]0.79 [0.29,2.19]0.82 [0.40,1.69]0.76 [0.18,3.19]-0.56 [0.23,1.37]0.85 [0.30,2.43]1.04 [0.55,1.96]0.75 [0.33,1.70]0.64 [0.30,1.39]
EVL+propranolol1.17 [0.69,2.00]1.51 [0.76,3.02]1.42 [0.60,3.38]1.46 [0.86,2.48]1.36 [0.35,5.23]1.79 [0.73,4.39]-1.53 [0.62,3.76]1.86 [0.99,3.52]1.34 [0.76,2.36]1.15 [0.73,1.81]
EVL+propranolol+ISMN0.77 [0.35,1.67]0.99 [0.41,2.38]0.93 [0.33,2.61]0.96 [0.45,2.03]0.89 [0.21,3.79]1.17 [0.41,3.33]0.65 [0.27,1.61]-1.22 [0.53,2.80]0.88 [0.40,1.92]0.75 [0.35,1.64]
Nadolol+ISMN0.63 [0.42,0.94]0.81 [0.46,1.44]0.76 [0.34,1.69]0.79 [0.55,1.12]0.73 [0.20,2.65]0.96 [0.51,1.81]0.54 [0.28,1.01]0.82 [0.36,1.89]-0.72 [0.43,1.21]0.62 [0.40,0.96]
Propranolol+ISMN[0.59,1.31] 0.871.13 [0.62,2.03]1.06 [0.48,2.34]1.09 [0.74,1.60]1.01 [0.28,3.71]1.33 [0.59,3.03]0.74 [0.42,1.31]1.14 [0.52,2.49]1.39 [0.82,2.34]-0.86 [0.61,1.20]
TIPS alone1.02 [0.77,1.35]1.31 [0.78,2.22]1.23 [0.59,2.58]1.27 [0.97,1.67]1.18 [0.33,4.20]1.56 [0.72,3.37]0.87 [0.55,1.37]1.33 [0.61,2.89]1.62 [1.04,2.52]1.17 [0.83,1.63]-
B, Mortaility due to rebleeding (Heterogeneity I2=0%, P=0.9963), fixed-effects model
GroupES aloneES+EVLES+propranololEVL aloneEVL+nadolol EVL+nadolol+ISMN EVL+propranolol EVL+Propranolol+ISMNNadolol+ISMN Propranolol+ISMNTIPS alone
ES alone-4.390.802.749.535.682.702.743.465.758.31
ES alone-4.39 [1.45,13.26]0.80 [0.27,2.41]2.74 [1.59,4.73]9.53 [0.38,237.86]5.68 [1.42,22.68]2.70 [0.23,31.58]2.74 [0.42,17.81]3.46 [1.44,8.31]5.75 [1.30,25.47]8.31 [2.85,24.22]
ES+EVL0.23 [0.08,0.69]-0.18 [0.04,0.87]0.62 [0.24,1.63]2.17 [0.08,59.65]1.29 [0.26,6.44]0.61 [0.04,8.50]0.62 [0.08,4.83]0.79 [0.24,2.60]1.31 [0.23,7.41]1.89 [0.46,7.77]
ES+propranolol1.25 [0.42,3.76]5.49 [1.15,26.14]-3.42 [1.00,11.71]11.91 [0.40,357.18]7.10 [1.21,41.67]3.37 [0.23,49.96]3.42 [0.39,30.05]4.33 [1.06,17.67]7.19 [1.13,45.80]10.39 [2.24,48.26]
EVL alone0.37 [0.21,0.63]1.60 [0.61,4.19]0.29 [0.09,1.00]-3.48 [0.15,82.91]2.07 [0.57,7.50]0.98 [0.09,11.36]1.00 [0.16,6.09]1.26 [0.62,2.57]2.10 [0.50,8.88]3.04 [1.08,8.54]
EVL+nadolol0.10 [0.00,2.62]0.46 [0.02,12.66]0.08 [0.00,2.52]0.29 [0.01,6.85]-0.60 [0.02,18.24]0.28 [0.01,15.52]0.29 [0.01,11.04]0.36 [0.01,9.36]0.60 [0.02,19.66]0.87 [0.03,24.50]
EVL+nadolol+ISMN0.18 [0.04,0.70]0.77 [0.16,3.85]0.14 [0.02,0.83]0.48 [0.13,1.74]1.68 [0.05,51.36]-0.47 [0.03,7.51]0.48 [0.05,4.42]0.61 [0.21,1.78]1.01 [0.15,6.97]1.46 [0.28,7.59]
EVL+propranolol0.37 [0.03,4.34]1.63 [0.12,22.53]0.30 [0.02,4.40]1.02 [0.09,11.71]3.53 [0.06,193.72]2.11 [0.13,33.29]-1.01 [0.05,19.49]1.28 [0.10,16.33]2.13 [0.17,27.32]3.08 [0.34,28.28]
EVL+propranolol+ISMN0.37 [0.06,2.38]1.60 [0.21,12.42]0.29 [0.03,2.57]1.00 [0.16,6.10]3.48 [0.09,133.93]2.08 [0.23,19.05]0.99 [0.05,18.94]-1.27 [0.18,8.80]2.10 [0.29,15.46]3.04 [0.43,21.47]
Nadolol+ISMN0.29 [0.12,0.69]1.27 [0.38,4.19]0.23 0.06,0.94]0.79 [0.39,1.61]2.75 [0.11,70.92]1.64 [0.56,4.80]0.78 [0.06,9.91]0.79 [0.11,5.50]-1.66 [0.33,8.26]2.40 [0.69,8.37]
Propranolol+ISMN0.17 [0.04,0.77]0.76 [0.13,4.31]0.14 [0.02,0.89]0.48 [0.11,2.01]1.66 [0.05,53.91]0.99 [0.14,6.79]0.47 [0.04,6.00]0.48 [0.06,3.49]0.60 [0.12,2.99]-1.45 [0.41,5.10]
TIPS alone0.12 [0.04,0.35]0.53 [0.13,2.17]0.10 [0.02,0.45]0.33 [0.12,0.93]1.15 [0.04,32.17]0.68 [0.13,3.54]0.32 [0.04,2.97]0.33 [0.05,2.32]0.42 [0.12,1.45]0.69 [0.20,2.44]-
C, Mortality due to liver failure (Heterogeneity I2=0%, P=0.8985), fixed-effects model
GroupES aloneES+EVLES+propranololEVL aloneEVL+nadolol EVL+nadolol+ISMN EVL+propranolol EVL+Propranolol+ISMNNadolol+ISMN Propranolol+ISMNTIPS alone
ES alone-0.51 [0.15,1.68]1.13 [0.24,5.26]0.80 [0.49,1.29]0.92 [0.13,6.63]0.29 [0.03,2.97]0.59 [0.23,1.50]0.94 0.23,3.90]1.15 [0.52,2.55]0.79 [0.34,1.79]0.59 [0.34,1.03]
ES+EVL1.96 [0.60,6.46]-2.21 [0.31,15.51]1.56 [0.53,4.65]1.82 [0.20,16.37]0.58 [0.05,7.24]1.15 [0.28,4.78]1.85 [0.32,10.65]2.26 [0.62,8.24]1.54 [0.41,5.87]1.15 [0.35,3.86]
ES+propranolol0.89 [0.19,4.16]0.45 [0.06,3.18]-0.71 [0.14,3.56]0.82 [0.07,10.03]0.26 [0.02,4.21]0.52 [0.09,3.17]0.84 [0.10,6.83]1.02 [0.18,5.79]0.70 [0.12,4.01]0.52 [0.10,2.69]
EVL alone1.26 [0.78,2.03]0.64 [0.22,1.90]1.41 [0.28,7.11]-1.16 [0.17,7.85]0.37 [0.04,3.62]0.74 [0.30,1.84]1.19 [0.30,4.66]1.44 [0.72,2.91]0.99 [0.46,2.14]0.74 0.44,1.24]
EVL+nadolol1.08 [0.15,7.75]0.55 [0.06,4.97]1.22 [0.10,14.84]0.86 [0.13,5.81]-0.32 [0.02,6.23]0.64 [0.08,5.27]1.02 [0.10,10.69]1.24 [0.16,9.50]0.85 [0.11,6.67]0.64 [0.09,4.60]
EVL+nadolol+ISMN3.39 [0.34,34.20]1.73 [0.14,21.63]3.82 [0.24,61.42]2.70 [0.28,26.39]3.14 [0.16,61.42]-1.99 [0.17,23.05]3.20 [0.23,45.57]3.90 [0.45,34.12]2.67 [0.24,29.42]1.99 [0.19,20.50]
EVL+propranolol1.70 [0.67,4.33]0.87 [0.21,3.59]1.91 [0.32,11.62]1.35 [0.54,3.37]1.57 [0.19,13.06]0.50 [0.04,5.80]-1.61 [0.33,7.70]1.96 [0.63,6.08]1.34 [0.49,3.64]1.00 [0.47,2.12]
EVL+propranolol+ISMN1.06 [0.26,4.39]0.54 [0.09,3.11]1.19 [0.15,9.71]0.84 [0.21,3.32]0.98 [0.09,10.28]0.31 [0.02,4.45]0.62 [0.13,2.99]-1.22 [0.26,5.64]0.83 [0.22,3.16]0.62 [0.16,2.47]
Nadolol+ISMN0.87 [0.39,1.93]0.44 [0.12,1.62]0.98 [0.17,5.55]0.69 [0.34,1.40]0.80 [0.11,6.16]0.26 [0.03,2.24]0.51 [0.16,1.59]0.82 [0.18,3.80]-0.68 [0.24,1.91]0.51 [0.22,1.20]
Propranolol+ISMN1.27 [0.56,2.90]0.65 [0.17,2.47]1.43 [0.25,8.23]1.01 [0.47,2.19]1.18 [0.15,9.24]0.38 [0.03,4.14]0.75 [0.27,2.04]1.20 [0.32,4.55]1.46 [0.52,4.09]-0.75 [0.39,1.45]
TIPS alone1.70 [0.97,2.97]0.87 [0.26,2.90]1.91 [0.37,9.87]1.35 [0.81,2.28]1.57 [0.22,11.39]0.50 [0.05,5.15]1.00 [0.47,2.11]1.61 [0.41,6.36]1.95 [0.83,4.58]1.34 [0.69,2.60]-
D, Treatment failure (Heterogeneity I2=29.4%, P=0.1739), fixed-effects model
GroupES aloneES+EVLEVL aloneEVL+nadolol EVL+nadolol+ISMN EVL+propranolol EVL+Propranolol+ISMNNadolol+ISMN Propranolol+ISMNTIPS alone
ES alone-3.72 [1.30,10.67]2.13 [1.31,3.45]8.65 [1.77,42.15]3.69 [1.65,8.27]9.09 [2.08,39.68]3.02 [1.22,7.53]2.78 [1.54,5.02]2.54 [1.06,6.12]8.24 [2.16,31.40]
ES+EVL0.27 [0.09,0.77]-0.57 [0.22,1.48]2.32 [0.39,13.82]0.99 [0.31,3.21]2.44 [0.43,13.94]0.81 [0.24,2.77]0.75 [0.27,2.11]0.68 [0.21,2.27]2.21 [0.43,11.29]
EVL alone0.47 [0.29,0.76]1.75 [0.68,4.53]-4.07 [0.90,18.39]1.74 [0.87,3.46]4.28 [0.99,18.49]1.42 [0.66,3.08]1.31 [0.86,1.98]1.20 [0.57,2.49]3.88 [1.03,14.61]
EVL+nadolol0.120.430.25-0.431.050.350.320.290.95
EVL+nadolol0.12 [0.02,0.56]0.43 [0.07,2.56]0.25 [0.05,1.11]-0.43 [0.08,2.24]1.05 [0.13,8.61]0.35 [0.06,1.91]0.32 [0.07,1.54]0.29 [0.05,1.58]0.95 [0.13,7.11]
EVL+nadolol+ISMN0.27 [0.12,0.61]1.01 [0.31,3.26]0.58 [0.29,1.15]2.34 [0.45,12.30]-2.46 [0.49,12.31]0.82 [0.29,2.31]0.75 [0.43,1.31]0.69 [0.25,1.89]2.23 [0.50,9.87]
EVL+propranolol0.11 [0.03,0.48]0.41 [0.07,2.34]0.23 [0.05,1.01]0.95 [0.12,7.78]0.41 [0.08,2.03]-0.33 [0.06,1.74]0.31 [0.07,1.39]0.28 [0.05,1.44]0.91 [0.49,1.68]
EVL+propranolol+ISMN0.33 [0.13,0.82]1.23 [0.36,4.19]0.70 [0.32,1.52]2.86 [0.52,15.57]1.22 [0.43,3.44]3.01 [0.57,15.74]-0.92 [0.38,2.21]0.84 [0.37,1.93]2.72 [0.59,12.65]
Nadolol+ISMN0.36 [0.20,0.65]1.34 [0.47,3.77]0.76 [0.50,1.16]3.11 [0.65,14.86]1.33 [0.77,2.30]3.27 [0.72,14.83]1.09 [0.45,2.61]-0.91 [0.39,2.12]2.96 [0.74,11.78]
Propranolol+ISMN0.39 [0.16,0.95]1.46 [0.44,4.86]0.84 [0.40,1.74]3.40 [0.63,18.20]1.45 [0.53,3.98]3.57 [0.70,18.38]1.19 [0.52,2.73]1.09 [0.47,2.54]-3.24 [0.71,14.76]
TIPS alone0.12 [0.03,0.46]0.45 [0.09,2.30]0.26 [0.07,0.97]1.05 [0.14,7.83]0.45 [0.10,1.98]1.10 [0.60,2.05]0.37 [0.08,1.71]0.34 [0.08,1.34]0.31 [0.07,1.41]-
E, Bleeding from gastroesophageal ulcer (Heterogeneity I2=0%, P=0.8354), fixed-effects model
GroupES aloneES+EVLEVL aloneEVL+nadolol EVL+nadolol+ISMN EVL+propranolol EVL+Propranolol+ISMNNadolol+ISMN Propranolol+ISMNTIPS alone
ES alone-0.55 [0.04,6.88]1.16 [0.48,2.82]4.05 [0.37,44.27]1.07 [0.04,28.38]2.71 [0.22,32.94]1.64 [0.24,11.51]9.41 [2.07,42.79]7.51 [0.35,160.14]8.02 [1.50,42.92]
ES+EVL1.81 [0.15,22.50]-2.10 [0.20,22.27][0.29,187.21]1.94 [0.03,109.86]4.89 [0.14,170.35]2.97 [0.16,55.53]17.01 [1.03,280.81]13.58 [0.32,583.96]14.51 [0.70,299.70]
EVL alone0.86 [0.35,2.09]0.48 [0.04,5.04]-3.49 [0.38,32.10]0.92 [0.03,24.40]2.33 [0.16,33.00]1.41 [0.25,7.99]8.09 [1.78,36.77]6.46 [0.35,120.78]6.90 [1.04,46.00]
EVL+nadolol0.25 [0.02,2.69]0.14 [0.01,3.48]0.29 [0.03,2.64]-0.26 [0.01,13.83]0.67 [0.02,21.21]0.41 [0.02,6.78]2.32 [0.16,34.08]1.85 [0.05,73.08]1.98 [0.11,36.71]
EVL+nadolol+ISMN0.93 [0.04,24.71]0.52 [0.01,29.25]1.09 [0.04,28.72]3.78 [0.07,197.91]-2.52 [0.04,155.57]1.53 [0.04,62.41]8.78 [0.48,160.33]7.01 [0.09,567.49]7.49 [0.19,297.05]
EVL+propranolol0.37 [0.03,4.50]0.20 [0.01,7.11]0.43 [0.03,6.09]1.50 [0.05,47.60]0.40 [0.01,24.40]-0.61 [0.03,14.43]3.48 [0.19,64.62]2.78 [0.05,144.26]2.97 [0.46,18.92]
EVL+propranolol+ISMN0.61 [0.09,4.26]0.34 [0.02,6.28]0.71 [0.13,4.00]2.47 [0.15,41.18]0.65 [0.02,26.51]1.65 [0.07,39.07]-5.72 [0.57,57.07]4.57 [0.23,91.89]4.88 [0.37,63.66]
Nadolol+ISMN0.11 [0.02,0.48]0.06 [0.00,0.97]0.12 [0.03,0.56]0.43 [0.03,6.33]0.11 [0.01,2.08]0.29 [0.02,5.35]0.17 [0.02,1.74]-0.80 [0.03,21.58]0.85 [0.09,8.18]
Propranolol+ISMN0.13 [0.01,2.84]0.07 [0.00,3.17]0.15 [0.01,2.89]0.54 [0.01,21.28]0.14 [0.00,11.55]0.36 [0.01,18.72]0.22 [0.01,4.40]1.25 [0.05,33.83]-1.07 [0.03,34.99]
TIPS alone0.12 [0.02,0.67]0.07 [0.00,1.42]0.14 [0.02,0.97]0.51 [0.03,9.37]0.13 [0.00,5.30]0.34 [0.05,2.15]0.20 [0.02,2.67]1.17 [0.12,11.23]0.94 [0.03,30.66]-

[i] TIPS, transjugular intrahepatic portosystemic shunt; ES, endoscopic sclerotherapy; EVL, endoscopic variceal ligation; ISMN, isosorbide-5-mononitrate.

Mortality due to rebleeding

A total of 27 trials with 2,447 patients investigated all 11 treatments and reported death due to rebleeding. The evidence network presented in Fig. S1B connects all of the treatments. Cochran's Q test did not identify any statistical heterogeneity among the selected trials for this outcome measure (Heterogeneity I2=0, P=0.9963, Table III). Compared with TIPS alone, ES plus propranolol increased the risk of mortality due to rebleeding (OR=10.39, 95% CI: 2.24-48.26, Fig. 2B; P-score=0.0842; Table II).

Pairwise comparisons indicated that ES plus EVL, EVL alone, EVL combined with nadolol plus ISMN, Nadolol plus ISMN, Propranolol plus ISMN and TIPS alone were significantly more effective than ES alone in reducing mortality due to rebleeding (OR=0.23, 95% CI: 0.08-0.69; OR=0.37, 95% CI: 0.21-0.63; OR=0.18, 95% CI: 0.04-0.70; OR=0.29, 95% CI: 0.12-0.69; OR=0.17, 95% CI: 0.04-0.77; OR=0.12, 95% CI: 0.04-0.35, respectively, Table III).

Mortality due to liver failure

A total of 24 trials with 2,258 patients investigating all 11 treatments reported on death due to liver failure. The evidence network presented in Fig. S1C connects all of the treatments. No statistical heterogeneity was observed (Heterogeneity I2=0%, P=0.8985; Table III). The results of the fixed-effects model analysis comparing with TIPS alone indicated that none of the other treatments were superior, though nadolol plus ISMN may be the next best option (OR=0.51, 95% CI: 0.22-1.20, Fig. 2C; P-score=0.7536; Table II). Furthermore, EVL combined with nadolol plus ISMN had the lowest P-score (P-score=0.2167, Table II), indicating the highest probability to increase mortality due to liver failure. Results of pairwise comparisons indicated no statistically significant differences when comparing with treatments other than TIPS (Table III).

Treatment failure

In total, 14 trials with a total of 1,445 patients reported on treatment failure. No data of this outcome were available for the treatment regimen ES plus propranolol. The evidence network in Fig. S1B connects the other 10 treatments for assessment of this outcome. Mild heterogeneity was identified for treatment failure (Heterogeneity I2=29.4%, P=0.1739; Table III). A fixed-effects model analysis was performed for comparing with TIPS alone. Differences were not statistically significant (Fig. 2D). The evidence network presented in Fig. S1D connects all of the treatments. EVL plus propranolol had the highest efficacy (P-score=0.8071), closely followed by TIPS (P-score=0.7938) and EVL plus nadolol (P-score=0.7932). ES alone ranked last (P-score=0.0199), suggesting that it was most likely to have the highest rate of treatment failure. Rankings are presented in Table II.

Pooled ORs suggested that ES alone was disadvantageous compared with the other 9 treatments with regard to treatment failure (OR=3.72, 95% CI: 1.30-10.67 compared with ES plus EVL; OR=2.13, 95% CI: 1.31-3.45 compared with EVL alone; OR=8.65, 95% CI: 1.77-42.15 compared with EVL plus nadolol; OR=3.69, 95% CI: 1.65-8.27 compared with EVL plus nadolol plus ISMN; OR=9.09, 95% CI: 2.08-39.68 compared with EVL plus propranolol; OR=3.02, 95% CI: 1.22-7.53 compared with EVL plus propranolol plus ISMN; OR=2.78, 95% CI: 1.54-5.02 compared with nadolol plus ISMN; OR=2.54, 95% CI: 1.06-6.12 compared with propranolol plus ISMN; OR=8.24, 95% CI: 2.16-31.40 compared with TIPS alone; Table III).

Bleeding from gastroesophageal ulcer

A total of 24 trials with a total of 2,258 patients investigated all 11 treatments and reported death due to bleeding from gastroesophageal ulcer. The evidence network is presented in Fig. S1E. There was no statistically significant heterogeneity for this outcome measure (Heterogeneity I2=0, P=0.8354; Table III). Results of the fixed-effects model analysis performed in comparison with TIPS alone indicated that none of the other 10 treatments were superior, but nadolol plus ISMN appeared to be the best among the compared treatments (OR=0.85, 95% CI: 0.09-8.18, Fig. 2E; P-score=0.7536, Table II).

Nadolol plus ISMN had the highest P-score (P-score=0.7536), indicating that it had the highest probability of reducing mortality due to rebleeding, followed closely by ES alone (P-score=0.6964) and ES plus propranolol (P-score=0.6651). The lowest P-score was obtained for EVL plus nadolol and ISMN (P-score=0.2167), indicating the lowest probability to reduce bleeding from gastroesophageal ulcer (Table II).

Pairwise comparisons among the treatments indicated that nadolol plus ISMN was associated with a relatively lower risk of causing bleeding from gastroesophageal ulcer when compared with ES alone (OR=0.11, 95% CI: 0.02-0.48), ES plus EVL (OR=0.06, 95% CI=0.00-0.97) or EVL alone (OR=0.12, 95% CI: 0.03-0.56) in Table III.

Sensitivity analysis

A sensitivity analysis was performed by removing several studies. The only criterion for removal was a mean follow-up time of <6 months, based on which 5 trials (43,47,55,60,61) were removed. The results were consistent with those of the primary meta-analysis (Table IV). Nadolol plus ISMN was still superior to TIPS with regard to overall mortality (OR=0.63, 95% CI: 0.40-0.98, Fig. 3A; Heterogeneity I2=0, P=0.9249, Table IV), while no significant differences were obtained for treatment failure (OR=2.97, 95% CI: 0.74-11.88, Fig. 3B; Heterogeneity I2=37.3%, P=0.1206, Table IV).

Table IV

Heterogeneity test results.

Table IV

Heterogeneity test results.

ItemI2 (%)QDFP-value
Overall mortality019.34320.9618
Overall mortality (sensitivity analysis)018.05280.9249
Mortality due to rebleeding05.97180.9963
Treatment failure29.412.7690.1739
Treatment failure (sensitivity analysis)37.312.7580.1206
Bleeding from gastroesophageal ulcer04.2380.8354
Mortality due to liver failure08.58150.8985

[i] DF, degrees of freedom.

Discussion

The present network meta-analysis included 43 randomized controlled trials to compare the treatment effectiveness of 11 mainstay secondary prophylaxes in patients with cirrhosis in terms of mortality, treatment failure and bleeding from gastroesophageal ulcers. The results suggested that nadolol plus ISMN was most likely to reduce the risk of overall mortality, mortality due to liver failure and bleeding from gastroesophageal ulcers, and was superior to ES and TIPS alone for reducing overall mortality. ES was inferior to 9 treatments for reducing treatment failure. The combination of endoscopic therapy and NSBB with or without ISMN was not significantly more effective than EVL or the drug combination alone.

The present study included 4 trials that investigated nadolol plus ISMN (26,36,51,57) with a total of 250 randomized patients. Nadolol plus ISMN was indicated to be more effective than TIPS alone in reducing overall mortality. The present results are consistent with those of Villanueva et al (28), which concluded that combination therapy was more effective than endoscopic ligation for the prevention of recurrent bleeding and was associated with lower rates of major complications. As a vasoconstrictor, nadolol is able to reduce portal pressure and blood flow in the porto-collateral system. The vasodilator ISMN has been demonstrated to decrease portal pressure in patients with cirrhosis by reducing intra-hepatic resistance (29). Despite adverse drug-associated effects, including hypotension, asthenia and headaches, proper dosage of this combination is most likely to reduce mortality and other complications of bleeding from ulcers.

In the newly published AASLD and UK guidelines, TIPS is recommended as a treatment option when endoscopic and pharmacologic treatments have failed (5). In the present study, a total of 16 trials provided data for TIPS in 590 patients with cirrhosis (20,21,25-27,29,30,32-35,38-40,42,43). The mean follow-up time for TIPS groups was 748.5 days [data for one trial (43) were not available]. Although TIPS is known to have the potential to increase hepatic encephalopathy (20,27), the present study demonstrated that it may reduce the risk of death due to rebleeding. The trials were contradictory regarding whether TIPS is superior to endoscopic or combination therapies. In one previous trial (26), TIPS alone was superior to EVL plus propranolol in the prevention of rebleeding, but this superiority did not result in improved survival. In this previous study (26), liver failure, hepatobiliary cancer and sepsis were the predominant causes of death. Clinically, it may be difficult to attribute death to rebleeding or to any one cause. Zheng et al (62) performed a meta-analysis of 12 RCTs to compare TIPS with endoscopic therapy, and the results suggested that TIPS reduced variceal rebleeding, but was associated with an increased risk of encephalopathy, although no differences in survival were observed. The present study indicated that TIPS was superior to ES alone, ES plus propranolol and EVL alone with a tendency for reduced mortality due to rebleeding. In clinical practice, TIPS has certain advantages in reducing portal pressure and reducing the risk of rebleeding. However, compared with endoscopic treatment, TIPS is more costly and technically more difficult.

Data from 4 previous studies (21,26,27,38) that used covered stents in their trials provided similar results among TIPS, EVL, EVL plus propranolol and propranolol plus ISMN in terms of overall mortality, although TIPS appeared to cause less mortality due to rebleeding than EVL plus propranolol (21).

Although ES was not recommended by the Baveno VI Consensus Workshop as a first-line treatment (15), it is still commonly used in China (61). Furthermore, the guidelines of the Chinese Society of Gastroenterology, Chinese Medical Association and Chinese Society of Endoscopy suggest that physicians choose EVL or ES for secondary prophylaxis based on their experience and the patients' clinical conditions (62). Therefore, RCTs on ES were not excluded from the present study.

Results of pooled ORs demonstrated inferiority of ES regarding treatment failure over the other 9 treatments. ES plus propranolol may increase the risk of mortality due to rebleeding. No statistically significant benefit of ES alone or ES plus drugs was identified in the present study. The results of the present study are consistent with those of previous studies (7-12,15) and support the most recent UK and AASLD guidelines, which do not recommend ES for secondary prevention of variceal bleeding in patients with cirrhosis (5).

EVL has been accepted as the preferred endoscopic treatment for the prevention of variceal rebleeding (37). Although EVL plus NSBB is now the first-line treatment, a review by Cotoras et al (64) reported that addition of β-blockers to EVL does not lead to a difference in mortality. In line with this, in the present study, the combination of EVL and NSBB with or without ISMN was not more effective than EVL alone or the drug combination of NSBB and ISMN.

The present study identified a tendency of EVL plus propranolol to increase mortality, which may be attributed to the data that were extracted from the included trials for this outcome measure. In a previous trial whose patients all had grade II-IV portal vein thrombosis (PVT), Luo et al (21) determined that the ability of EVL plus propranolol to reduce variceal bleeding may be counteracted by deteriorated PVT. Additional evidence from high-quality RCTs is required to address this issue.

The present study has several limitations. Therapy using drugs alone or using drugs other than NSBB or ISMN were beyond the scope of the present study, as regimens of single drugs are now seldom used in clinical practice for secondary prophylaxis. Although the included trials provided a solid foundation based on collected data, more data on serious adverse effects, the frequency and severity of drug-associated adverse effects, procedure-associated complications and consequent hospitalization may be helpful for further comparison. The quality of the present study depends on the RCTs that were included. A total of 4 RCTs focusing on acute variceal bleeding (37,38,47,54) were included, as they also assessed the outcomes of rebleeding and overall mortality. The included references were published between 1992 and 2015, which is a long period of time; there may be technical differences in the early stages; however, with the continuous standardization and maturity of operation technology, the differences are gradually narrowing. Therefore, the clinical value of the present results may be limited.

In conclusion, the present network meta-analysis suggested that for prophylaxis of variceal rebleeding in patients with cirrhosis, nadolol plus ISMN may be the preferred choice to decrease mortality and ES may be associated with a relatively higher risk of unfavourable treatment outcomes, particularly treatment failure.

Supplementary Material

Evidence network with regard to (A) overall mortality, (B) mortality due to rebleeding, (C) mortality due to liver failure, (D) treatment failure and (E) bleeding from gastroesophageal ulcer.
Risk of bias assessment in included randomized controlled trials.

Acknowledgements

The authors would like to thank Dr Sam Zhong and Dr Michelle Liu (Shanghai KNOWLANDS MedPharm Consulting Co.,Ltd.) for their support with data analyses in the present study.

Funding

No funding was received.

Availability of data and materials

The datasets used and/or analyzed in the present study are availabled from the correspoding author on reasonable request.

Authors' contributions

YK was responsible for the study design, data collection and analysis and manuscript review. LS reviewed the data collection and analysis and drafted the manuscript. Both authors read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References

1 

Bresci G: Portal hypertension: The management of esophageal/gastric varices, portal hypertensive gastropathy or hypertensive colopathy. Therapy. 4:91–96. 2007.

2 

Mansour L, El-Kalla F, EL-Bassat H, Abd-Elsalam S, El-Bedewy M, Kobtan A, Badawi R and Elhendawy M: Randomized controlled trial of scleroligation versus band ligation alone for eradication of gastroesophageal varices. Gastrointest Endosc. 86:307–315. 2017.PubMed/NCBI View Article : Google Scholar

3 

Sarin SK, Lahoti D, Saxena SP, Murthy NS and Makwana UK: Prevalence, classification and natural history of gastric varices: A long-term follow-up study in 568 portal hypertension patients. Hepatology. 16:1343–1349. 1992.PubMed/NCBI View Article : Google Scholar

4 

Trudeau W and Prindiville T: Endoscopic injection sclerosis in bleeding gastric varices. Gastrointest Endosc. 32:264–268. 1986.PubMed/NCBI View Article : Google Scholar

5 

Tripathi D, Stanley AJ, Hayes PC, Patch D, Millson C, Mehrzad H, Austin A, Ferguson JW, Olliff SP, Hudson M, et al: U.K. guidelines on the management of variceal haemorrhage in cirrhotic patients. Gut. 64:1680–1704. 2015.PubMed/NCBI View Article : Google Scholar

6 

Bosch J and García-Pagán JC: Prevention of variceal rebleeding. Lancet. 361:952–954. 2003.PubMed/NCBI View Article : Google Scholar

7 

Brunner F, Berzigotti A and Bosch J: Prevention and treatment of variceal haemorrhage in 2017. Liver Int. 37 (Suppl 1):S104–S115. 2017.PubMed/NCBI View Article : Google Scholar

8 

Gonzalez R, Zamora J, Gomez-Camarero J, Molinero LM, Bañares R and Albillos A: Meta-analysis: Combination endoscopic and drug therapy to prevent variceal rebleeding in cirrhosis. Ann Intern Med. 149:109–122. 2008.PubMed/NCBI View Article : Google Scholar

9 

Ravipati M, Katragadda S, Swaminathan PD, Molnar J and Zarling E: Pharmacotherapy plus endoscopic intervention is more effective than pharmacotherapy or endoscopy alone in the secondary prevention of esophageal variceal bleeding: A meta-analysis of randomized, controlled trials. Gastrointest Endosc. 70:658–664.e5. 2009.PubMed/NCBI View Article : Google Scholar

10 

Garcia-Tsao G, Abraldes J, Berzigotti A and Bosch J: Portal Hypertensive Bleeding in Cirrhosis: Risk Stratification, Diagnosis, and Management: 2016 practice guidance by the American association for the study of liver diseases. Hepatology. 65:310–335. 2017.PubMed/NCBI View Article : Google Scholar

11 

Thiele M, Krag A, Rohde U and Gluud LL: Meta-analysis: Banding ligation and medical interventions for the prevention of rebleeding from oesophageal varices. Aliment Pharmacol Ther. 35:1155–1165. 2012.PubMed/NCBI View Article : Google Scholar

12 

Puente A, Hernández-Gea V, Graupera I, Roque M, Colomo A, Poca M, Aracil C, Gich I, Guarner C and Villanueva C: Drugs plus ligation to prevent rebleeding in cirrhosis: An updated systematic review. Liver Int. 34:823–833. 2014.PubMed/NCBI View Article : Google Scholar

13 

Funakoshi N, Ségalas-Largey F, Duny Y, Oberti F, Valats JC, Bismuth M, Daurès JP and Blanc P: Benefit of combination β-blocker and endoscopic treatment to prevent variceal rebleeding: A meta-analysis. World J Gastroenterol. 16:5982–5992. 2010.PubMed/NCBI View Article : Google Scholar

14 

Garcia-Tsao G, Abraldes JG, Berzigotti A and Bosch J: Portal hypertensive bleeding in cirrhosis: Risk stratification, diagnosis, and management: 2016 practice guidance by the American Association for the study of liver diseases. Hepatology. 65:310–335. 2017.PubMed/NCBI View Article : Google Scholar

15 

de Franchis R: BavenoV Faculty: Expanding consensus in portal hypertension: Report of the Baveno VI Consensus Workshop: stratifying risk and individualizing care for portal hypertension. J Hepatol. 63:743–752. 2015.PubMed/NCBI View Article : Google Scholar

16 

Chen J, Zeng XQ, Ma LL, Huang XQ, Tseng YJ, Wang J, Luo TC and Chen SY: Long-term efficacy of endoscopic ligation plus cyanoacrylate injection with or without sclerotherapy for variceal bleeding. J Dig Dis. 17:252–259. 2016.PubMed/NCBI View Article : Google Scholar

17 

Rücker G and Schwarzer G: Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Med Res Methodol. 15(58)2015.PubMed/NCBI View Article : Google Scholar

18 

Schwarzer G, Carpenter J and Rücker G: Netmeta: Network meta-analysis with R (2016). R Package Version 0.9-2. netmeta, Cited February 22, 2016. Available from http://CRAN.R-project.org/package=netmeta.

19 

Gupta V, Rawat R and Shalimar Saraya A: Carvedilol versus propranolol effect on hepatic venous pressure gradient at 1 month in patients with index variceal bleed: RCT. Hepatol Int. 11:181–187. 2016.PubMed/NCBI View Article : Google Scholar

20 

Sauerbruch T, Mengel M, Dollinger M, Zipprich A, Rössle M, Panther E, Wiest R, Caca K, Hoffmeister A, Lutz H, et al: Prevention of rebleeding from esophageal varices in patients with cirrhosis receiving small-diameter stents versus hemodynamically controlled medical therapy. Gastroenterology. 149:p. 660–688.e1. 2015.PubMed/NCBI View Article : Google Scholar

21 

Luo X, Wang Z, Tsauo J, Zhou B, Zhang H and Li X: Advanced cirrhosis combined with portal vein thrombosis: A randomized trial of TIPS versus endoscopic band ligation plus propranolol for the prevention of recurrent esophageal variceal bleeding. Radiology. 276:286–293. 2015.PubMed/NCBI View Article : Google Scholar

22 

Kumar P, Saxena KN, Misra SP and Dwivedi M: Secondary prophylaxis of variceal hemorrhage: A comparative study of band ligation, carvedilol and propranolol plus isosorbide mononitrate. Ind J Gastroenterology. 34(A54)2015.

23 

Kong DR, Wang JG, Chen C, Yu FF, Wu Q and Xu JM: Effect of intravariceal sclerotherapy combined with esophageal mucosal sclerotherapy using small-volume sclerosant for cirrhotic patients with high variceal pressure. World J Gastroenterol. 21:2800–2806. 2015.PubMed/NCBI View Article : Google Scholar

24 

Kumar A, Jha SK, Sharma P, Dubey S, Tyagi P, Sharma BC and Sarin SK: Addition of propranolol and isosorbide mononitrate to endoscopic variceal ligation does not reduce variceal rebleeding incidence. Gastroenterology. 137:892–901. 2009.PubMed/NCBI View Article : Google Scholar

25 

Sauer P, Hansmann J, Richter GM, Stremmel W and Stiehl A: Endoscopic variceal ligation plus propranolol vs. transjugular intrahepatic portosystemic stent shunt: A long-term randomized trial. Endoscopy. 34:690–697. 2002.PubMed/NCBI View Article : Google Scholar

26 

Gülberg V, Schepke M, Geigenberger G, Holl J, Brensing KA, Waggershauser T, Reiser M, Schild HH, Sauerbruch T and Gerbes AL: Transjugular intrahepatic portosystemic shunting is not superior to endoscopic variceal band ligation for prevention of variceal rebleeding in cirrhotic patients: A randomized, controlled trial. Scand J Gastroenterol. 37:338–343. 2002.PubMed/NCBI View Article : Google Scholar

27 

Escorsell A, Bañares R, García-Pagán JC, Gilabert R, Moitinho E, Piqueras B, Bru C, Echenagusia A, Granados A and Bosch J: TIPS versus drug therapy in preventing variceal rebleeding in advanced cirrhosis: A randomized controlled trial. Hepatology. 35:385–392. 2002.PubMed/NCBI View Article : Google Scholar

28 

Villanueva C, Miñana J, Ortiz J, Gallego A, Soriano G, Torras X, Sáinz S, Boadas J, Cussó X, Guarner C and Balanzó J: Endoscopic ligation compared with combined treatment with nadolol and isosorbide mononitrate to prevent recurrent variceal bleeding. N Engl J Med. 345:647–655. 2001.PubMed/NCBI View Article : Google Scholar

29 

Pomier-Layrargues G, Villeneuve JP, Deschênes M, Bui B, Perreault P, Fenyves D, Willems B, Marleau D, Bilodeau M, Lafortune M and Dufresne MP: Transjugular intrahepatic portosystemic shunt (TIPS) versus endoscopic variceal ligation in the prevention of variceal rebleeding in patients with cirrhosis: A randomised trial. Gut. 48:390–396. 2001.PubMed/NCBI View Article : Google Scholar

30 

Narahara Y, Kanazawa H, Kawamata H, Tada N, Saitoh H, Matsuzaka S, Osada Y, Mamiya Y, Nakatsuka K, Yoshimoto H, et al: A randomized clinical trial comparing transjugular intrahepatic portosystemic shunt with endoscopic sclerotherapy in the long-term management of patients with cirrhosis after recent variceal hemorrhage. Hepatol Res. 21:189–198. 2001.PubMed/NCBI View Article : Google Scholar

31 

Hou MC, Chen WC, Lin HC, Lee FY, Chang FY and Lee SD: A new ‘sandwich’ method of combined endoscopic variceal ligation and sclerotherapy versus ligation alone in the treatment of esophageal variceal bleeding: A randomized trial. Gastrointest Endosc. 53:572–578. 2001.PubMed/NCBI View Article : Google Scholar

32 

Meddi P, Merli M, Lionetti R, De Santis A, Valeriano V, Masini A, Rossi P, Salvatori F, Salerno F, de Franchis R, et al: Cost analysis for the prevention of variceal rebleeding: A comparison between transjugular intrahepatic portosystemic shunt and endoscopic sclerotherapy in a selected group of Italian cirrhotic patients. Hepatology. 29:1074–1077. 1999.PubMed/NCBI View Article : Google Scholar

33 

García-Villarreal L, Martínez-Lagares F, Sierra A, Guevara C, Marrero JM, Jiménez E, Monescillo A, Hernández-Cabrero T, Alonso JM and Fuentes R: Transjugular intrahepatic portosystemic shunt versus endoscopic sclerotherapy for the prevention of variceal rebleeding after recent variceal hemorrhage. Hepatology. 29:27–32. 1999.PubMed/NCBI View Article : Google Scholar

34 

Sauer P, Theilmann L, Stremmel W, Benz C, Richter G and Stiehl A: Transjugular intrahepatic portosystemic stent shunt versus sclerotherapy plus propranolol for variceal rebleeding. Gastroenterology. 113:1623–1631. 1997.PubMed/NCBI View Article : Google Scholar

35 

Sanyal AJ, Freedman AM, Luketic VA, Purdum PP III, Shiffman ML, Cole PE, Tisnado J and Simmons S: Transjugular intrahepatic portosystemic shunts compared with endoscopic sclerotherapy for the prevention of recurrent variceal hemorrhage. a randomized, controlled trial. Ann Intern Med. 126:849–857. 1997.PubMed/NCBI View Article : Google Scholar

36 

Villanueva C, Balanzó J, Novella MT, Soriano G, Sáinz S, Torras X, Cussó X, Guarner C and Vilardell F: Nadolol plus isosorbide mononitrate compared with sclerotherapy for the prevention of variceal rebleeding. N Engl J Med. 334:1624–1629. 1996.PubMed/NCBI View Article : Google Scholar

37 

Li P, Kong RD, Xie HJ, Sun B and Xu JM: Efficacy and safety of endoscopic variceal ligation and endoscopic injection sclerotherapy in patients with cirrhosis and esophageal varices: A prospective study. World Chinese J Digestology. 18:3791–3795. 2010.

38 

Holster IL, Tjwa E, Moelker A, Wils A, Hansen BE, Vermeijden JR, Scholten P, van Hoek B, Nicolai JJ, Kuipers EJ, et al: Covered transjugular intrahepatic portosystemic shunt versus endoscopic therapy+β-blocker for prevention of variceal rebleeding. Hepatology. 63:581–589. 2016.PubMed/NCBI View Article : Google Scholar

39 

van Buuren HR, Groeneweg M, Vleggaar FP, Lesterrhuis W, Van Tillburg AJP, Hop WCJ, Peiterman H, Schalam SW, van Hout BA and Lameris JS: The results of a randomized controlled trial evaluating TIPS and endoscopic therapy in cirrhotic patients with gastro-esophageal variceal bleeding. J Hepatol. 32(71)2000.

40 

Cabrera J, Maynar M, Granados R, Gorriz E, Reyes R, Pulido-Duque JM, Rodriguez SanRoman JL, Guerra C and Kravetz D: Transjugular intrahepatic portosystemic shunt versus sclerotherapy in the elective treatment of variceal hemorrhage. Gastroenterology. 110:832–839. 1996.PubMed/NCBI View Article : Google Scholar

41 

Dobrucali A, Bal K, Tuncer M, Ilkova F, Çelik A, Uzunismail H, Yurdakul I and Oktay E: Endoscopic banding ligation compared with sclerotherapy for the treatment of esophageal varies. Turk J Gastroenterol. 9:340–344. 1998.

42 

Holster IL, Moelker A, Tjwa ET, et al: Early transjugular intrahepatic portosystemic shunt (TIPS) as compared to endoscopic treatment reduces rebleeding but not mortality in cirrhotic patients with a 1st or 2nd episode of variceal bleeding: A multicentre randomized controlled trial. United European Gastroenterology Journal. 1(A84)2013.

43 

Merli M, Riggio O, Capocaccia L, et al: TIPS vs sclerotherapy in preventing variceal rebleeding in cirrhotic patients: Preliminary results of a randomised controlled trial (EASL abstract). Journal of hepatology. 21(S41)1994.

44 

Ahmad I, Khan AA, Alam A, Butt AK, Shafqat F and Sarwar S: Propranolol, isosorbide mononitrate and endoscopic band ligation-alone or in varying combinations for the prevention of esophageal variceal rebleeding. J Coll Physicians Surg Pak. 19:283–286. 2009.PubMed/NCBI

45 

Argonz J, Kravetz D, Suarez A, Romero G, Bildozola M, Passamonti M, Valero J and Terg R: Variceal band ligation and variceal band ligation plus sclerotherapy in the prevention of recurrent variceal bleeding in cirrhotic patients: A randomized, prospective and controlled trial. Gastrointest Endosc. 51:157–163. 2000.PubMed/NCBI View Article : Google Scholar

46 

Avgerinos A, Armonis A, Manolakopoulos S, Poulianos G, Rekoumis G, Sgourou A, Gouma P and Raptis S: Endoscopic sclerotherapy versus variceal ligation in the long-term management of patients with cirrhosis after variceal bleeding. A prospective randomized study. J Hepatol. 26:1034–1341. 1997.PubMed/NCBI View Article : Google Scholar

47 

Avgerinos A, Armonis A, Stefanidis G, Mathou N, Vlachogiannakos J, Kougioumtzian A, Triantos C, Papaxoinis C, Manolakopoulos S, Panani A and Raptis SA: Sustained rise of portal pressure after sclerotherapy, but not band ligation, in acute variceal bleeding in cirrhosis. Hepatology. 39:1623–1630. 2004.PubMed/NCBI View Article : Google Scholar

48 

Avgerinos A, Rekoumis G, Klonis C, Papadimitriou N, Gouma P, Pournaras S and Raptis S: Propranolol in the prevention of recurrent upper gastrointestinal bleeding in patients with cirrhosis undergoing endoscopic sclerotherapy. A randomized controlled trial. J Hepatol. 19:301–311. 1993.PubMed/NCBI View Article : Google Scholar

49 

Baroncini D, Milandri GL, Borioni D, Piemontese A, Cennamo V, Billi P, Dal Monte PP and D'Imperio N: A prospective randomized trial of sclerotherapy versus ligation in the elective treatment of bleeding esophageal varices. Endoscopy. 29:235–240. 1997.PubMed/NCBI View Article : Google Scholar

50 

García-Pagán JC, Villanueva C, Albillos A, Bañares R, Morillas R, Abraldes JG and Bosch J: Spanish Variceal Bleeding Study Group: Nadolol plus isosorbide mononitrate alone or associated with band ligation in the prevention of recurrent bleeding: A multicentre randomised controlled trial. Gut. 58:1144–1150. 2009.PubMed/NCBI View Article : Google Scholar

51 

Hou MC, Lin HC, Kuo BI, Chen CH, Lee FY and Lee SD: Comparison of endoscopic variceal injection sclerotherapy and ligation for the treatment of esophageal variceal hemorrhage: A prospective randomized trial. Hepatology. 21:1517–1522. 1995.PubMed/NCBI

52 

Hou MC, Lin HC, Lee FY, Chang FY and Lee SD: Recurrence of esophageal varices following endoscopic reagent and impact on rebleeding: Comparison of sclerotherapy and ligation. J Hepatology. 32:202–208. 2000.PubMed/NCBI View Article : Google Scholar

53 

Lo GH, Lai KH, Cheng JS, Hwu JH, Chang CF, Chen SM and Chiang HT: A prospective, randomized trial of sclerotherapy versus ligation in the management of bleeding esophageal varices. Hepatology. 22:466–4671. 1995.PubMed/NCBI

54 

Lo GH, Lai KH, Cheng JS, Lin CK, Huang JS, Hsu PI and Chiang HT: Emergency banding ligation versus sclerotherapy for the control of active bleeding from esophageal varices. Hepatology. 25:1101–1104. 1997.PubMed/NCBI View Article : Google Scholar

55 

Peña J, Brullet E, Sanchez-Hernández E, Rivero M, Vergara M, Martin-Lorente JL and Garcia Suárez C: Variceal ligation plus nadolol compared with ligation for prophylaxis of variceal rebleeding: A multicenter trial. Hepatology. 41:572–578. 2005.PubMed/NCBI View Article : Google Scholar

56 

de la Peña J, Rivero M, Sanchez E, Fábrega E, Crespo J and Pons-Romero F: Variceal ligation compared with endoscopic sclerotherapy for variceal hemorrhage: Prospective randomized trial. Gastrointest Endosc. 49:417–423. 1999.PubMed/NCBI View Article : Google Scholar

57 

Romero G, Kravetz D, Argonz J, Vulcano C, Suarez A, Fassio E, Dominguez N, Bosco A, Muñoz A, Salgado P and Terg R: Comparative study between nadolol and 5-isosorbide mononitrate vs. Endoscopic band ligation plus sclerotherapy in the prevention of variceal rebleeding in cirrhotic patients: A randomized controlled trial. Aliment Pharmacol Ther. 24:601–611. 2006.PubMed/NCBI View Article : Google Scholar

58 

Stiegmann GV, Goff JS, Michaletz-Onody PA, Korula J, Lieberman D, Saeed ZA, Reveille RM, Sun JH and Lowenstein SR: Endoscopic sclerotherapy as compared with endoscopic ligation for bleeding esophageal varices. N Engl J Med. 326:1527–1532. 1992.PubMed/NCBI View Article : Google Scholar

59 

Umehara M, Onda M, Tajiri T, Toba M, Yoshida H and Yamashita K: Sclerotherapy plus ligation versus ligation for the treatment of esophageal varices: A prospective randomized study. Gastrointest Endosc. 50:7–12. 1999.PubMed/NCBI View Article : Google Scholar

60 

Viazis N, Armonis A, Vlachogiannakos J, Rekoumis G, Stefanidis G, Papadimitriou N, Manolakopoulos S and Avgerinos A: Effects of endoscopic variceal treatment on oesophageal function: a prospective, randomized study. Eur J Gastroenterol Hepatol. 14:263–269. 2002.PubMed/NCBI View Article : Google Scholar

61 

Vinel JP, Lamouliatte H, Cales P, Combis JM, Roux D, Desmorat H, Pradere B, Barjonet G and Quinton A: Propranolol reduces the rebleeding rate during endoscopic sclerotherapy before variceal obliteration. Gastroenterology. 102:1760–173. 1992.PubMed/NCBI View Article : Google Scholar

62 

Zheng M, Chen Y, Bai J, Zeng Q, You J, Jin R, Zhou X, Shen H, Zheng Y and Du Z: Transjugular intrahepatic portosystemic shunt versus endoscopic therapy in the secondary prophylaxis of variceal rebleeding in cirrhotic patients: meta-analysis update. J Clin Gastroenterol. 42:507–516. 2008.PubMed/NCBI View Article : Google Scholar

63 

Chinese Society of Gastroenterology, Chinese Medical Association, Chinese Society of Endoscopy: Guidelines for the diagnosis and treatment of esophageal and gastric variceal bleeding in cirrhotic portal hypertension. Chinese Society of Hepatology, Chinese Medical Association. Chinese Medial Association. J Clin Hepatol 32: 203-219, 2016 (In Chinese).

64 

Cotoras P, Faundez J and Candia R: Should we add beta-blockers to band ligation for secondary prophylaxis of variceal bleeding? Medwave. 17 (Suppl 1)(e6847)2017.PubMed/NCBI View Article : Google Scholar

Related Articles

Journal Cover

June-2020
Volume 19 Issue 6

Print ISSN: 1792-0981
Online ISSN:1792-1015

Sign up for eToc alerts

Recommend to Library

Copy and paste a formatted citation
x
Spandidos Publications style
Kong Y and Kong Y: Comparison of the effectiveness of 11 mainstay treatments for secondary prophylaxis of variceal bleeding in patients with cirrhosis: A network meta‑analysis. Exp Ther Med 19: 3479-3496, 2020
APA
Kong, Y., & Kong, Y. (2020). Comparison of the effectiveness of 11 mainstay treatments for secondary prophylaxis of variceal bleeding in patients with cirrhosis: A network meta‑analysis. Experimental and Therapeutic Medicine, 19, 3479-3496. https://doi.org/10.3892/etm.2020.8633
MLA
Kong, Y., Shi, L."Comparison of the effectiveness of 11 mainstay treatments for secondary prophylaxis of variceal bleeding in patients with cirrhosis: A network meta‑analysis". Experimental and Therapeutic Medicine 19.6 (2020): 3479-3496.
Chicago
Kong, Y., Shi, L."Comparison of the effectiveness of 11 mainstay treatments for secondary prophylaxis of variceal bleeding in patients with cirrhosis: A network meta‑analysis". Experimental and Therapeutic Medicine 19, no. 6 (2020): 3479-3496. https://doi.org/10.3892/etm.2020.8633