The effect of shortening vasoactive drug durations alongside endoscopic therapy in esophageal variceal bleeding: an updated systematic review and meta-analysis

Sudheer Dhoopa, Zohaib Ahmedb, Conner Lombardia, Mohammed Abu-Rumaileha, Syeda Faiza Arifa, Wasef Sayeha, Rayna Patela, Alborz Sherafatia, Wade Lee-Smithc, Mona Hassanb

University of Toledo College of Medicine and Life Sciences; University Libraries, University of Toledo, OH, USA

aDepartment of Internal Medicine, University of Toledo College of Medicine and Life Sciences (Sudheer Dhoop, Conner Lombardi, Mohammed Abu-Rumeileh, Syeda Faiza Arif, Wasef Sayeh, Rayna Patel, Alborz Sherafati); bDepartment of Gastroenterology and Hepatology, University of Toledo College of Medicine and Life Sciences (Zohaib Ahmed, Mona Hassan); cUniversity Libraries, University of Toledo (Wade Lee-Smith), Toledo, OH, USA

Correspondence to: Sudheer Dhoop, MD, Department of Internal Medicine, The University of Toledo College of Medicine and Life Sciences, 3000 Arlington Ave, Toledo, Ohio, 43614, USA, e-mail: Sudheer.Dhoop@rockets.utoledo.edu
Received 2 March 2024; accepted 15 July 2024; published online 19 August 2024
DOI: https://doi.org/10.20524/aog.2024.0906
© 2024 Hellenic Society of Gastroenterology

Abstract

Background The recommended duration of vasoactive drugs in esophageal variceal bleeding (EVB) spans 2-5 days. Prior meta-analyses of randomized trials include only a few studies that compared short vs. long vasoactive drug durations approximating this time range, including older management techniques, and only assessed variceal rebleeding at 5 days. We identified several additional randomized controlled trials (RCTs) assessing rebleeding at various durations, with updated management of EVB.

Methods We performed an updated systematic review and meta-analysis assessing the effect of shortening the vasoactive drug duration by 48-72 h. The primary outcome was rebleeding within 5 days. Secondary outcomes included rebleeding, mortality due to rebleeding, and all-cause mortality within 4-6 weeks (extended period) with subgroup analysis by vasoactive drug and type of endoscopic therapy. Length of stay, blood transfusion requirements and terlipressin-related adverse events were additional secondary outcomes.

Results Our comprehensive search strategy and screening process yielded 14 RCTs with 1060 patients (75.1% male): 7 trials used terlipressin, 4 octreotide, and 3 somatostatin. Shortened durations combined with band ligation led to similar rebleeding, with a trend towards less rebleeding when populations with more severe liver disease were excluded. There was greater rebleeding and mortality over an extended period when shorter durations were combined with sclerotherapy. Longer durations were associated with a longer hospital stay and, for terlipressin, more adverse events.

Conclusions Shorter vasoactive drug durations combined with band ligation in selected populations appear safe. Higher powered RCTs are needed, involving patients with different degrees of severity of EVB and liver disease.

Keywords Vasoactive, duration, esophageal variceal bleeding

Ann Gastroenterol 2024; 37 (5): 567-578

Introduction

The recommended duration of vasoactive drugs alongside endoscopic management in esophageal variceal bleeding (EVB) varies, with multiple societies recommending a broad range of 2-5 days [1-3]. Earlier studies [4,5] found a 5-day duration compared to placebo alongside endoscopic therapy effective in preventing EVB, but a subsequent study showed that a 2-day duration doubled treatment failures [6]. Since then, data from several randomized controlled trials (RCTs) [7-11] comparing shorter and longer durations of vasoactive drug durations have been pooled in systematic reviews and meta-analyses (SRMAs) [12,13]. These demonstrated similar rates of 5-day variceal rebleeding with shortened durations, approximating the 2-day vasoactive drug duration, compared to longer durations, leading the European Association for the Study of the Liver to suggest that a shortened duration should be considered [2].

Neither of these SRMAs compared the outcomes of short vs. longer courses across a single vasoactive drug, and only 1 SRMA [12] included a single study on octreotide [8], the vasoactive drug of choice due efficacy and safety [14,15], relative to terlipressin. Terlipressin, however, has recently gained approval for hepatorenal syndrome in the United States [16], which may increase its popularity and, while several relevant RCTs [7,10,11] have been pooled for meta-analysis [12,13], newer published RCTs with updated management are available [17,18] in addition to published RCTs on somatostatin [19].

Our updated SRMA aimed to add data to variceal rebleeding at 5 days, but also to pool data from multiple studies on rebleeding and mortality at time points within approximately 6 weeks, as recommended by portal hypertension guidelines [1]. It was designed to assess each vasoactive drug individually, to determine any differences between drugs over shorter durations. Additionally, the study assessed whether the effect on EVB of shortening the duration of vasoactive drug therapy was influenced by the endoscopic technique of sclerotherapy vs. band ligation (BL), the standard of care. Finally, Child-Pugh Class C (CPCC) has been cited as a significant predictor of variceal rebleeding [3,8], with prior RCTs focusing on higher CPCC distributions [7,10,11]. We sought to add studies across various distributions of CPCC to explore the effect of shortening vasoactive drug durations in various severities of liver disease.

Materials and methods

Search strategy

A comprehensive search was constructed in Embase (Embase.com, Elsevier) by an experienced health sciences librarian (WLS) on 13 February 2024, using truncated keywords, phrases, proximity searching and subject headings. This strategy was translated to MEDLINE (OVID, National Library of Medicine), the Cochrane Central Register of Controlled Trials (CochraneLibrary.com, Wiley), the Web of Science Core Collection, the Korean Citation Index, and SciELO (Web of Science platform, Clarivate) and Global Index Medicus (World Health Organization) (see Supplementary Table 1 for detailed search strategies). No limits were applied to publication date or language. All results were exported to EndNote 20 citation management software (Clarivate, Philadelphia, PA, USA) (Supplementary Table 2) and duplicates were removed by successive iterations of EndNote’s duplicate detection algorithms and manual inspection. Our systematic review process was conducted in accordance with the PRISMA guidelines (Supplementary Table 3) [20].

Study selection criteria

Two study authors (SD and MA) reviewed records and excluded duplicated studies not removed by the software, articles on animals, articles on children, review articles, case reports and case studies/series, study protocols, studies not involving exclusively EVB, and studies not studying the vasoactive drug duration as the primary comparison. We chose to include abstracts to increase the data available and to decrease publication bias [21]. One full text manuscript was written in Farsi [22], so the entire text was preliminarily translated by ChatGPT 4.0 and afterwards by a native Farsi speaker (AS). From the review of full texts or abstracts, we excluded studies that: 1) did not perform initial endoscopic therapy; 2) did not administer any vasoactive drug in a comparison arm; and 3) where sample sizes or type of endoscopic therapy could not be determined after attempts to contact authors. Only RCTs were included, as durations in retrospective studies are confounded by the severity of bleeding [3,23]. In terms of shorter and longer vasoactive drug duration, we included all studies with durations that closely approximated or intersected with the 2- to 5-day range recommended, with only 1 study [24] having a slightly longer duration (6.5 days), and we excluded 2 studies that compared 5 vs. 10 days of terlipressin. While guidelines [1] recommend BL over sclerotherapy for endoscopic hemostasis, we included studies with any endoscopic techniques, planning to carry out a subgroup analysis. Inclusion of each study was agreed upon by 2 authors (SD and MA) and another (CL) resolved any disagreements regarding study inclusion.

Baseline characteristics

We gathered demographic data in addition to any factors reported in studies that would influence EVB, based on guidelines [1,3], including Child-Pugh Class % distribution, varix grade distribution, the presence of active bleeding on endoscopy, and success in achieving initial endoscopic hemostasis (Table 1). Not all studies reported a model for end-stage liver disease (MELD) or a Child-Pugh score between comparison arms. Therefore, we calculated the pre-2016 MELD scores, using an online calculator [25] with reported means and standard deviations of total bilirubin, creatinine, and prothrombin (PT) values, and used them to compare liver disease severity between comparison arms. PT was converted to the international normalized ratio, assuming an International Sensitivity Index (ISI) of 1.3, which is between reported lab ranges [26], and we reported whether the pre-2016 MELD scores were significantly different across treatment arms (Table 1). Unfortunately, the wide ranges in ISI did not allow us to compare MELD scores across studies [26].

Table 1 Demographics and characteristics of studies included in the meta-analysis by vasoactive drug

thumblarge

Obtaining unpublished data

We noted studies that reported different measures for baseline characteristics, time periods of rebleeding, while 1 study [8] mixed sclerotherapy and BL without specifying how many were in the short- and long-duration groups. We emailed all corresponding authors listed in the study to obtain further data. We received responses from authors of 3 studies [8,17,18], with additional data on Child-Pugh Class distribution [18], mean blood products transfused [18], mean length of stay (LOS) [18], rebleeding at 5 days and 6 weeks [17], mortality due to rebleeding [17, 18] and all-cause mortality at 5 days and 6 weeks instead of 7 days and 8 weeks [17]. We also obtained a breakdown of patients who underwent sclerotherapy or BL in the 2-day and 5-day comparison, and which endoscopic treatment the patient who died received in the study [8] that mixed the treatments. All authors were notified of and consented to our plans to publish their unpublished data prior to providing it to us.

Extraction of primary and secondary outcome data

Our data were extracted into Microsoft Excel including previously unpublished data (Supplementary Table 4). We report variceal rebleeding within 5 days alongside BL, the endoscopic standard of care, as our primary outcome, despite guidelines [1] recommending 6-week mortality as the primary outcome in EVB studies. This is since neither rebleeding nor mortality was consistently reported at exactly 6 weeks, while of the studies that specified a primary outcome, most [7,10,11,18,27] chose 5-day rebleeding. Secondary outcomes were rebleeding at 5 days for sclerotherapy as well as rebleeding recorded at a duration between 4 to 6 weeks (henceforth to be referred to as the “extended period” [EP]), mortality due to rebleeding at the EP, all-cause mortality at the EP for both BL and sclerotherapy. This range accommodates the inclusion of 3 studies [7,8,18] that only measured rebleeding outcomes up to 1 month. It should be noted that rebleeding at 5 days was included in the number of rebleeds at the EP. Additionally, while the Baveno VII consensus [1] considers all mortality within 6 weeks as related to the initial variceal bleed, some included studies [7-9,22] and 1 study [18] author we asked made a distinction between death from variceal rebleeding directly and death from other causes, such as hepatic encephalopathy, so this was recorded as an additional outcome. Additional secondary outcomes were overall LOS, and blood transfusion requirements in units of packed red blood cells (pRBC), as well as adverse events specifically related to terlipressin, given concerns for its safety profile [15]. Note that the difference in short vs. longer vasoactive durations varied across studies from 2-3 days in BL, and given that LOS is partially dependent on the difference between short and long durations, we divided the LOS in each study by the difference in duration of vasoactive drugs in the study—henceforth known as corrected LOS (cLOS)—to allow the data to be pooled across studies. We also distinguished adverse events as total or severe, with the latter defined as those that had life-threatening consequences that required urgent intervention, that are potentially reversible with intensive treatment, or a death related to the drug as per classification of RCTs [28].

Meta-analysis

Outcome data were transferred from Microsoft Excel to Review Manager 5.4 software (Revman) for meta-analysis. Rebleeding, mortality, and adverse events were entered as dichotomous outcomes generating risk ratios and confidence intervals (CI). cLOS (days) and blood transfusion requirements (pRBC transfused) were continuous outcomes entered as mean and standard deviations, generating mean differences and confidence intervals. The random effects model was used, and a P-value of <0.05 was considered statistically significant. Statistical heterogeneity was assessed using Higgins I2 index, calculated in Revman. For our primary outcome, several studies had zero rebleeding events in both comparison groups, resulting in fewer than 10 studies where an effect size (risk ratio) that could be calculated. Since at least 10 effect sizes are recommended for generating a funnel plot to assess publication bias, we did not generate one. The Risk of Bias (RoB) 2.0 Cochrane Tool for individually randomized parallel controlled trials [29] was used to assess study bias and to make an overall judgment as whether there was a high risk of bias, some concerns, or a low risk of bias (Supplementary Table 5). Subgroup analysis was performed to assess the effects between individual drugs and between endoscopic therapies of BL and sclerotherapy. Sensitivity analysis was performed on studies where there was a high risk of bias, and/or concerns about less-than-optimal randomization, as there were more factors influencing variceal rebleeding in 1 comparison arm (Table 1). Subsequently, a second layer of sensitivity analysis was performed by excluding studies that were likely to have a wider CPCC distribution. This was determined by assessing the CPCC distribution across all studies and noting 2 studies [7,10] reporting CPCC distributions >30%, whereas the rest were <20%. Additionally, 1 study [11] was conducted 6 months after another [10] at the same hospital, with similar protocols, so it was assumed CPCC distributions were similar. We performed a final separate sensitivity analysis by pooling only studies with high CPCC distributions or high-risk varices. Stratification of variceal severity was carried out in a similar way as for CPCC distribution, by identifying 2 studies [7,22] with 24% and 39% active variceal bleed on endoscopy, compared to other studies [9,18] that reported 12% and 7%.

Results

Our comprehensive search strategy and screening process (Fig. 1) yielded 14 RCTs [7-11,17-19,22,24,27,30-32] that were published from 2004-2024 and included 1060 patients, with a mean age of 49 years, of whom 75.1% were male. There were no differences in age, sex distribution, etiology of liver disease, or distribution of Child-Pugh scores across comparison arms, but baseline characteristics related to rebleeding [1] were statistically higher in 4 studies in the longer vasoactive drug duration comparison arm, including higher PT [8], calculated pre-2016 MELD [17,24] and reported MELD [30]. In terms of data quality, our risk of bias analysis revealed that 11 of 14 RCTs had at least some risk (see supplementary table 5), with only 2 non-open label studies [7,9], some concerns about randomization bias [8,17,24,30], and limited methods [11,27,30-32].

thumblarge

Figure 1 PRISMA flow chart for study selection

EVB, esophageal variceal bleeding

Our primary outcome was rebleeding within 5 days in the short vs. long vasoactive drug duration arms, alongside BL only, involving 12 studies. There was no significant difference between short vs. long durations, with an overall risk ratio of 0.79 (95%CI 0.37-1.65; P=0.67; I2=0%; Fig. 2A). There was also no difference between individual drugs on subgroup analysis (Fig 2A).

thumblarge

Figure 2 Forest plots comparing variceal rebleeding and associated mortality in short vs. long durations in esophageal variceal band ligation for vasoactive drugs. (A) Rebleeding within 5 days. (B) Rebleeding within the extended period (30-42 days). (C) Mortality due to rebleeding within the extended period (30-42 days). (D) Overall mortality associated with variceal bleeding

M-H, Mantel-Haenszel; CI, confidence interval

In terms of secondary outcomes, at the EP, there was no significant difference in rebleeding, mortality due to rebleeding, or all-cause mortality between short and long vasoactive drug duration all alongside BL, with risk ratios of 0.77 (95%CI 0.46-1.30; P=0.33; I2=0%; Fig. 2B), 0.75 (95%CI 0.26-2.13; P=0.58; I2=0%; Fig. 2C), and 0.87 (95%CI 0.51-1.48; P=0.52; I2=0%; Fig. 2D), respectively.

2 studies [8,24] assessed patients who underwent sclerotherapy. In this group shortened durations led to more rebleeding at 5 days, with a risk ratio of 2.40 (95%CI 0.10-56.67; P=0.59; Fig. 3A), and at the EP, with a risk ratio of 4.24 (95%CI 0.96-18.78; P=0.06; I2=0%; Fig. 3B), though these differences were not statistically significant. Shortened durations in sclerotherapy did lead to a statistically significant increase in mortality due to rebleeding, and to all-cause mortality at the EP (all mortality in sclerotherapy was due to rebleeding), both with a risk ratios of 5.68 (95%CI 1.06-30.49; P=0.04; I2=0%; Fig. 3C, D). Finally, subgroup analysis comparing shortened durations in BL vs. sclerotherapy found that the latter led to significantly increased mortality (Fig. 3C, D).

thumblarge

Figure 3 Forest plots comparing variceal rebleeding and associated mortality in short vs. long durations by endoscopic therapy. (A) Rebleeding within 5 days. (B) Rebleeding within the extended period (30-42 days). (C) Mortality due to rebleeding with the extended period (30-42 days). (D) Overall mortality associated with variceal bleeding

M-H, Mantel-Haenszel; CI, confidence interval

There was a significantly greater number of total adverse events related to terlipressin in the long duration group, with a risk ratio of 1.66 (95%CI 1.23-2.26; P=0.001, I2=0%; Fig. 4A), but no statistically significant difference in severe adverse events, resulting in a risk ratio of 1.01 (95%CI 0.19-5.40; P=0.99; I2=0%; Fig. 4B). cLOS was significantly longer for the longer vasoactive drug duration in the BL subgroup, with a mean difference of 1.12 days (95%CI 0.71-1.53; P=0.003; I2=89%; Fig. 4C). Finally, there was no significant difference between blood transfusion requirements with a mean difference of 0.15 more pRBCs transfused (95%CI -0.10-0.41; P=0.24; I2=0%) in the longer duration group amongst patients who underwent BL; however, there was a difference between the BL and sclerotherapy groups in the blood transfusions required (Fig. 4D).

thumblarge

Figure 4 Forest plots comparing adverse events, corrected length of stay, and blood transfusion requirements in short vs. long durations for vasoactive drugs. (A) Total adverse events related to terlipressin. (B) Severe adverse events related to terlipressin. (C) Corrected length of stay. (D) Blood transfusion requirements

M-H, Mantel-Haenszel; CI, confidence interval

In terms of trends, prior to the sensitivity analysis there were statistically insignificant trends towards increased rebleeding, mortality due to rebleeding, and all-cause mortality in the longer vasoactive duration group who underwent BL. Octreotide and somatostatin both had similar trends of increased rebleeding within 5 days associated with the longer treatment durations (Fig. 2A), with only 1 study pooled on each, whereas shortening terlipressin had no notable trend and more studies were pooled. Study quality was assessed with the Risk of Bias 2.0 tool for all 14 studies (Supplementary Table 5), which revealed that 11 studies had some risk of bias. The sensitivity analysis, excluding studies with high CPCC distributions, revealed a trend towards more 5-day rebleeding with longer durations in patients who underwent BL (Supplementary Fig. 1A). Conversely, including only studies with high-risk variceal features or high CPCC distribution [7,10-,22] revealed a trend towards more 5-day rebleeding (Supplementary Fig. 1B) and a trend towards greater mortality due to rebleeding at the EP (Supplementary Fig. 1C).

Discussion

The conclusion of our SRMA aligns with those of prior studies [16,17] in that shortening vasoactive durations by 48-72 h in combination with BL does not lead to increased rebleeding in EVB. Our results, however, add more precision by analyzing new outcomes, including variceal rebleeding, mortality due to rebleeding, and overall all-cause mortality at an extended time duration of 4-6 weeks (EP) for rebleeding and mortality, and demonstrating the observation across 3 different vasoactive drugs. Our study also resulted in 3 original observations related to vasoactive drug durations.

First, longer durations led to a longer LOS and more total terlipressin-related adverse effects, but blood transfusions and severe adverse effects related to terlipressin remained similar. This is in agreement with another study where 5-day vasoactive drug vs. 5-day pantoprazole infusions in EVB were compared, showing longer stays for the vasoactive group, though without statistical significance, suggesting a potential trend towards a greater LOS related to the longer administration duration of a vasoactive drug, rather than the duration of the intravenous infusion. Regarding blood transfusions, 2 studies [8,22] in our review, which were not pooled because they reported dichotomous data, reported no significant differences, with 1 [8] noting slightly more transfusions in the longer duration group. This trend may be driven by a confounder, such as increased intensive care monitoring leading to more transfusions [33], or it might support the trend of more 5-day rebleeding due to a more vasoactive drug: it is difficult to determine which without more granular data. Our finding that shortening terlipressin courses decreased total adverse events, but not severe ones, is supported by 2 studies [34,35] excluded from our analysis, which found significantly more total adverse events in the longer duration group, also with no differences in severe adverse events, and only 1 severe adverse event in the shorter terlipressin duration arm across all 3 studies. Overall, this finding suggests that shortening vasoactive drug duration would mainly reduce LOS and total adverse effects related to terlipressin.

Second, our subgroup analysis revealed that a shorter duration of vasoactive therapy combined with sclerotherapy was associated with more rebleeding and greater mortality in the extended period (EP). This makes intuitive sense, as sclerotherapy has been associated with acute rises in portal pressures lasting for 5 days [36,37], which may increase rebleeding risk; thus, longer vasoactive durations for up to 5 days to potentially counteract that effect should remain the standard of care whenever sclerotherapy is performed [38].

Finally, our sensitivity analysis by liver disease and variceal bleeding severity showed a trend for more 5-day rebleeding with longer vasoactive durations in cases with less severity, and a trend for less 5-day rebleeding and lower mortality at the EP associated with longer durations in cases with higher severity. The trend that shortening durations alongside BL decreases rebleeding in patients with less severe bleeding, particularly with octreotide and somatostatin, but increases rebleeding in severe liver disease or variceal bleeding cases, warrants a cautious approach to shortening vasoactive durations in severe EVB, pending further investigation. It should be noted that the lower levels of rebleeding in shorter durations were more prevalent in the octreotide and somatostatin subgroups, which share similar mechanisms of action [39]. Data suggest that longer durations of octreotide lead to tachyphylaxis and less sustained drops in portal pressures compared to terlipressin [39,40], although how this relates back to greater rebleeding is unclear. Additionally, the trend for more 5-day rebleeding with longer durations of octreotide/somatostatin, but not terlipressin, could just be due to the limited sample size for the former drugs.

Our SRMA summarizes the highest level of data available (RCTs) on short vs. long vasoactive drug durations, incorporating subgroup analysis by vasoactive drug and endoscopic technique, with sensitivity analyses to account for the severity of liver disease and variceal bleed, but our analysis does have limitations. First, 4 of 14 studies were conference abstracts with limited methodology and peer review, although all 3 vasoactive drugs were represented in abstracts. While we reached out to all authors systematically for additional data points and clarification regarding methods, we received responses from authors of more recent studies [8,17,18], which introduces a reporting bias favoring recent data. Several forms of clinical heterogeneity existed in the methodologies per study (Table 2). These included exact durations of short vs. long courses of vasoactive drugs, and the vasoactive drug administration’s timing with respect to endoscopy, with 2 studies stopping vasoactive drugs after endoscopy [10,24], limiting exact day recommendations. Additionally, the time point for rebleeding and mortality after 5 days varied, and only 3 studies [8,18,22] mention β-blockers used for secondary prophylaxis to reduce portal pressures in EVB [1,41]. Our sample size was also limited, despite including 14 RCTs in our meta-analysis, as inconsistent methods of recording data such as transfusion requirements across studies resulted in several underpowered outcomes. Additionally, zero-events (no bleeds) in both arms limited our ability to detect differences in effect size, though they still suggest non-inferiority between short and long durations. In terms of publication bias, 11 of 14 studies showed some risk due to randomization and blinding issues. Our subgroup and sensitivity analyses faced limitations: previous meta-analyses [14,15,42] indicated minimal differences in vasoactive drug efficacies, making our study underpowered to detect any such differences. Furthermore, only 2 studies [8,24] were available for sclerotherapy subgroup analysis. While the sensitivity analysis improved data quality, it was limited by incomplete methodologies. Finally, the generalizability of our findings is limited, since 12 of 14 RCTs were conducted in Asia, with none from Europe, Canada or the United States, bringing into question their applicability to western healthcare settings.

Table 2 Summary of methodologies of studies included in the meta-analysis by vasoactive drug

thumblarge

More high quality RCTs with uniform methodology across different severities of liver disease and variceal risk profiles are required to determine whether shortening vasoactive drug duration is safe across all populations. In terms of outlook, high hepatic vein pressure gradient (HVPG) measures have been shown to be associated with early rebleeding [43], so measurements of portal pressures via HVPG that were made in the most recent RCT [18] may increase our understanding of how shortening each vasoactive drug reduces portal pressures, and how that affects variceal rebleeding.

Summary Box

What is already known:

  • Vasoactive drug regimens of 2-5-day durations are recommended, in combination with endoscopic band ligation, to reduce variceal rebleeding

  • There are data to suggest that decreasing vasoactive drug duration from 3-5 days to 1-3 days does not increase the risk of variceal rebleeding within 5 days

What the new findings are:

  • Decreasing vasoactive drug durations from 3-5 to 1-3 days alongside band ligation does not increase the risk of variceal rebleeding, all-cause mortality, nor mortality due to rebleeding at 1 month to 6 weeks across all vasoactive drugs

  • Longer vasoactive drug durations led to a statistically significant longer hospital stay, and trended towards more units of blood transfused, and possibly even greater variceal rebleeding in populations with low liver disease severity

  • Shorter vasoactive drug durations were associated with more rebleeding after sclerotherapy, and there was a trend towards slightly more rebleeding from shortened durations in populations with high-risk varices and/or high Child-Pugh C distributions

Acknowledgments

The authors would like to thank Dr. Manwar S. Ali, Dr. Deba Prasad and Dr. Shalimar, corresponding authors of the studies cited in the Methods section, for providing additional data that expanded the scope and increased the statistical power of our meta-analysis.

References

1. de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C;Baveno VII Faculty. Baveno VII - Renewing consensus in portal hypertension. J Hepatol 2022;76:959-974.

2. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on prevention and management of bleeding and thrombosis in patients with cirrhosis. J Hepatol 2022;76:1151-1184.

3. Garcia-Tsao G, Abraldes JG, Rich NE, Wong VW. AGA clinical practice update on the use of vasoactive drugs and intravenous albumin in cirrhosis:expert review. Gastroenterology 2024;166:202-210.

4. Sung JJ, Chung SC, Yung MY, et al. Prospective randomised study of effect of octreotide on rebleeding from oesophageal varices after endoscopic ligation. Lancet 1995;346:1666-1669.

5. Avgerinos A, Nevens F, Raptis S, Fevery J. Early administration of somatostatin and efficacy of sclerotherapy in acute oesophageal variceal bleeds:the European Acute Bleeding Oesophageal Variceal Episodes (ABOVE) randomised trial. Lancet 1997;350:1495-1499.

6. Avgerinos A, Viazis N, Vlachogiannakos J, et al. Two different doses and duration schedules of somatostatin -14 in the treatment of patients with bleeding oesophageal varices:a non-randomised controlled study. J Hepatol 2000;32:171-172.

7. Azam Z, Hamid S, Jafri W, et al. Short course adjuvant terlipressin in acute variceal bleeding:a randomized double blind dummy controlled trial. J Hepatol 2012;56:819-824.

8. Rengasamy S, Ali SM, Sistla SC, Lakshmi CP, Harichandra Kumar KT. Comparison of 2 days versus 5 days of octreotide infusion along with endoscopic therapy in preventing early rebleed from esophageal varices:a randomized clinical study. Eur J Gastroenterol Hepatol 2015;27:386-392.

9. Chitapanarux T, Ritdamrongthum P, Leerapun A, Pisespongsa P, Thongsawat S. Three-day versus five-day somatostatin infusion combination with endoscopic variceal ligation in the prevention of early rebleeding following acute variceal hemorrhage:A randomized controlled trial. Hepatol Res 2015;45:1276-1282.

10. Salim A, Malik K, Haq IU, Butt AK, Alam A. Comparison of 12-hour with 72-hour terlipressin therapy for bleeding esophageal varices. J Coll Physicians Surg Pak 2017;27:334-337.

11. Zaman M, Zaidi AR, Hyder A, Kumar M, Amin J, Malik K. Frequency of rebleeding between short course terlipressin (24 hours) and usual course (72 hours) terlipressin in adult cirrhotic patients presenting with acute variceal rebleeding. Med Forum 2019;30:130-133.

12. Yan P, Tian X, Li J. Is additional 5-day vasoactive drug therapy necessary for acute variceal bleeding after successful endoscopic hemostasis?:A systematic review and meta-analysis. Medicine (Baltimore) 2018;97:e12826.

13. Yeh JH, Lo GH, Huang RY, Lin CW, Wang WL, Perng DS. Short-course vasoconstrictors are adequate for esophageal variceal bleeding after endoscopic variceal ligation:a systematic review and meta-analysis. Sci Prog 2021;104:368504211031711.

14. Zhou X, Tripathi D, Song T, et al. Terlipressin for the treatment of acute variceal bleeding:A systematic review and meta-analysis of randomized controlled trials. Medicine (Baltimore) 2018;97:e13437.

15. Huaringa-Marcelo J, Huaman MR, Brañez-Condorena A, et al. Vasoactive agents for the management of acute variceal bleeding:a systematic review and meta-analysis. J Gastrointestin Liver Dis 2021;30:110-121.

16. Administration UFaD. Fda.gov. FDA approves treatment to improve kidney function in adults with hepatorenal syndrome. 2022. Available from:https://www.fda.gov/drugs/news-events-human-drugs/fda-approves-treatment-improve-kidney-function-adults-hepatorenal-syndrome [Accessed 20 July 2024].

17. Poudel RC, Dhibar DP, Sharma N, Sharma V, Taneja S, Prakash A. Rational for continuing terlipressin after endoscopic variceal ligation in acute variceal haemorrhage needs further evidence:a pilot study. Arq Gastroenterol 2022;59:89-96.

18. Vaishnav M, Biswas S, Shenoy A, et al. Comparison of 1-day versus 3-day intravenous terlipressin in cirrhosis patients with variceal bleeding:A pilot randomised controlled trial. Aliment Pharmacol Ther 2024;59:645-655.

19. Abdelghani Y, Yehia R. Continuing somatostatin does not prevent rebleeding after variceal ligation for acute variceal bleeding. J Pharmaceutical Negative Results 2022;13:477-482.

20. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement:an updated guideline for reporting systematic reviews. BMJ 2021;372:n71.

21. Scherer RW, Saldanha IJ. How should systematic reviewers handle conference abstracts?A view from the trenches. Syst Rev 2019;8:264.

22. Eskandar H, Hashemi SJ, Masjedizadeh R, Zarei F. Randomized prospective trial comparing a 48 hour infusion to a five-day infusion of octerotide in patients with acute variceal bleeding. Govaresh 2011;16:98-104.

23. Laws MB, Wahking R, Blackburn E, Williams W, Schadler A, Fritz MK. Outcomes in variceal bleeding associated with continuous octreotide in patients with delayed endoscopy. J Pharm Pract 2024:8971900241228948.

24. Yücesoy M, Başkol M, Keklik M, et al. Efficacy of five days of subcutaneous octreotide treatment after sclerotherapy in preventing rebleeding from esophageal varices. Turk J Gastroenterol 2004;15:137-143.

25. Kamath PS. Mdcalc.com. MELD Score (Original, Pre-2016, Model for End-Stage Liver Disease). 2022. Available from:https://www.mdcalc.com/calc/2693/meld-score-original-pre-2016-model-end-stage-liver-disease [Accessed 20 July 2024].

26. Dorgalaleh A, Favaloro EJ, Bahraini M, Rad F. Standardization of prothrombin time/international normalized ratio (PT/INR). Int J Lab Hematol 2021;43:21-28.

27. Solari J, Bandi JC, Redondo A, et al. Similar outcomes with 3 or 5 days therapy with terlipressin plus endoscopic band ligation in patients with esophageal variceal bleeding:1176. Hepatology 2012;56:754A-755A.

28. US Department of Health and Human Services. Ctep.cancer.gov. Common terminology criteria for adverse events (CTAE) v5.0. 2017. Available from:https://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/CTCAE_v5_Quick_Reference_8.5x11.pdf [Accessed 20 July 2024].

29. Sterne JAC, SavovićJ, Page MJ, et al. RoB 2:a revised tool for assessing risk of bias in randomised trials. BMJ 2019;366:l4898.

30. Choudhary NS, Chawla YK, Duseja A, Dhiman RK. PP23-58 Comparison of 2 days and 5 days terlipressin with endoscopic band ligation for esophageal variceal bleed in patients of cirrhosis. Hepatol Int 2011;5:359.

31. George A, Ponnudurai R, Sachithanandan S, et al. 217 Randomised prospective trial comparing a 48 hour infusion to a 5 day infusion of octreotide in patients with acute variceal bleeding. J Gastroenterol Hepatol 2006;21:A449-A450.

32. Yaras S, Ates F, Saritas B, et al. Short term somatostatin infusion therapy is as effective as long term therapy beside endoscopic ligation for treatment esophageal variceal bleeding. Hepatol Int 2013;7 (Suppl 1):S550-S551.

33. AkbaşT. Long length of stay in the ICU associates with a high erythrocyte transfusion rate in critically ill patients. J Int Med Res 2019;47:1948-1957.

34. Bruha R, Marecek Z, Prochazka V, et al. Double-blind randomized multicenter study comparing the efficacy and safety of 10-day to 5-day terlipressin treatment of bleeding esophageal varices. Hepatogastroenterology 2009;56:390-394.

35. Lo GH, Perng DS, Chang CY, Tai CM, Wang HM, Lin HC. Controlled trial of ligation plus vasoconstrictor versus proton pump inhibitor in the control of acute esophageal variceal bleeding. J Gastroenterol Hepatol 2013;28:684-689.

36. Avgerinos A, Armonis A, Stefanidis G, et al. Sustained rise of portal pressure after sclerotherapy, but not band ligation, in acute variceal bleeding in cirrhosis. Hepatology 2004;39:1623-1630.

37. Kumar A, Jha SK, Mittal VV, Sharma P, Sharma BC, Sarin SK. Addition of somatostatin after successful endoscopic variceal ligation does not prevent early rebleeding in comparison to placebo:a double blind randomized controlled trial. J Clin Exp Hepatol 2015;5:204-212.

38. Huang Y, Zhang W, Xiang H, et al. Treatment strategies in emergency endoscopy for acute esophageal variceal bleeding (CHESS1905):a nationwide cohort study. Front Med (Lausanne) 2022;9:∘81.

39. Baik SK, Jeong PH, Ji SW, et al. Acute hemodynamic effects of octreotide and terlipressin in patients with cirrhosis:a randomized comparison. Am J Gastroenterol 2005;100:631-635.

40. Escorsell A, Bandi JC, Andreu V, et al. Desensitization to the effects of intravenous octreotide in cirrhotic patients with portal hypertension. Gastroenterology 2001;120:161-169.

41. Cheung KS, Mok CH, Lam LK, et al. Carvedilol versus other nonselective beta blockers for variceal bleeding prophylaxis and death:a network meta-analysis. J Clin Transl Hepatol 2023;11:1143-1149.

42. Seo YS, Park SY, Kim MY, et al. Lack of difference among terlipressin, somatostatin, and octreotide in the control of acute gastroesophageal variceal hemorrhage. Hepatology 2014;60:954-963.

43. Shi Y, Shen W, Xu G, Wang X, Ning B. Hepatic venous pressure gradient and rebleeding risk of patients with nonalcoholic steatohepatitis cirrhosis after variceal bleeding. Front Med (Lausanne) 2023;10:1224506.

Notes

Conflict of Interest: None