Historical overview and review of current day treatment in the management of acute variceal haemorrhage

Table 1 Predictors of day 5 treatment failure

Variable	OR	95%CI
Transfusion in 24 h (units)	1.35	1.13-1.61
CTP class	2.27	1.22-4.22
AST (per 10 U increase)	1.03	1.01-1.06
PV thrombosis	2.75	1.25-6.04

Adapted from D’Amico et al[14]. AST: Aspartate aminotransferase; PV: Portal vein; CTP: Child–Turcotte–Pugh.

Table 2 Predictors of 6-wk mortality

Variable	OR	95%CI
Albumin (per 1 g reduction)	2.33	1.32-4.00
Bilirubin (per 1 mg increase)	1.23	1.10-1.37
Transfusion total (units)	1.40	1.19-1.66
Hepatocellular carcinoma	3.44	1.64-7.24
Encephalopathy	2.30	1.39-3.70

Adapted from D’Amico et al[14].

Table 3 Comparison of vasoactive pharmacological therapies used in variceal haemorrhage

	Octreotide	Somatostatin	Terlipressin
Mode of administration	Bolus followed by IV infusion	Bolus followed by IV infusion	IV bolus
Class	Somatostatin analogue		Synthetic analogue of Vasopressin
Indication	Variceal haemorrhage	Variceal haemorrhage	Variceal haemorrhage Hepatorenal syndrome
Proposed mechanism of action	Mechanism unclear Inhibition of glucagon-mediated splanchnic vasodilatation and reduction of postprandial gut hyperemia	Amino-acid peptide that reduced splanchnic blood flow (especially azygous). Prevent release of vasoactive peptides	V1 receptors blockade causing splanchnic vasoconstriction
Dose	Bolus of 50 μg, followed by an infusion of 50 μg per hour for up to 5 d	Infusion of 250-500 μg/h	2 mg bolus followed by 1 mg every 4 h for 3-5 d
Side effects/cautions	Vomiting, abdominal pain, nausea, hepatitis, abnormal LFTs, diahorrea, hypoglycaemia. Rarely arrhythmias, dyspnoea, pancreatitis, rash and alopecia	Loss of appetite, nausea, vomiting, abdominal, diahorrea and fatigue	Vasoconstrictive side-effects: myocardial ischemia, limb ischemia (avoid if peripheral vascular disease), nausea and diahorrea. Hyponatraemia

LFTs: Liver function tests.

Table 4 Summary of randomized controlled trials and meta-analysis of different therapies over time in variceal haemorrhage

Ref.	Trial design/therapy	Outcome/results
Surgical techniques
Inokuchi et al[19]	Randomised controlled trial (RCT)/prophylactic surgical intervention (n = 60) vs non surgical intervention (n = 52) for oesophageal varices	5-yr cumulative survival rate at 5 yr in the operated group was 72% vs 45% (P < 0.05). 5-yr cumulative variceal bleeding rate at 5 yr was 7% in the operated group v46% (P < 0.001)
Balloon tamponade
Terés et al[28]	RCT/comparison of SB vs Linton-Nachlas (LN)	Primary haemostasis rates of 86%. In oesophageal variceal bleeding SB tube achieved permanent haemostasis in 52% vs 30% in LN tube
Sclerotherapy
The Copenhagen esophageal varices sclerotherapy project[46]	Randomised multicentre trial/187 unselected patients with oesophageal variceal bleed randomly assigned to medical treatment including balloon tamponade or to medical treatment supplemented with paravariceal sclerotherapy	Overall mortality in the sclerotherapy group (hazard) was 76% (95%CI: 10%-54%) of that in the medical-regimen group ( relative mortality in the sclerotherapy group was 63% of that in the medical-regimen group)
Westaby et al[38]	RCT of sclerotherapy (n = 56) vs placebo (n = 60)	Survival was significantly better in those treated by sclerotherapy (P < 0.001)
Burroughs et al[39]	Randomised trial/a comparison of sclerotherapy (n = 5) with staple transection (n = 51) of the oesophagus for the emergency control of bleeding from oesophageal varices	Total mortality did not differ significantly between the two groups. Mortality at six wk was 44% among those assigned to sclerotherapy and 35% assigned to staple transection. Complication rates were similar for the two groups
D’Amico et al[126]	Cochrane database systematic/meta-analysis of 17 trials, assessing the benefits of sclerotherapy vs vasoactive drugs in patients with variceal bleeding	Authors concluded no convincing evidence to support the use of emergency sclerotherapy as the first, single treatment when compared with vasoactive drugs
Thakeb et al[62]	Randomised controlled trial/assess the role of the combined N-butyl-2-cyanoacrylate and ethanolamine oleate (n = 58) vs ethanolamine sclerotherapy (n = 56) for management of bleeding esophagogastric varices	Arrested acute bleeding in 66.7% of patients with gastric variceal bleeding. Recurrent bleeding in 8.6% in the combined therapy group vs 25% in the sclerosis group (P < 0.01). The mortality in the combined therapy group less than sclerosis group (3.5% and 8.8% respectively, P > 0.05)
Endoscopic variceal band ligation (EVBL)
Laine et al[65]	Meta-analysis of 7 RCTs/comparison of the effect of EVBL vs sclerotherapy in the treatment of patients with bleeding esophageal varices	EVBL (vs sclerotherapy) reduced the rebleeding rate (OR = 0.52, 95%CI: 0.37-0.74), the mortality rate (OR = 0.67, 95%CI: 0.46-0.98), and the rate of death due to bleeding (OR = 0.49, 95%CI: 0.24-0.996)
Garcia-Pagán et al[63]	Meta-analysis of 10 RCTs comparing sclerotherapy with EVBL	Non-significant benefit of EVBL in achieving initial haemostasis vs sclerotherapy (pooled relative risk of 0.53 with 95%CI: 0.28-1.01)
Radiological transjugular intrahepatic portosystemic stent-shunts (TIPSS)
Monescillo et al[11]	RCT of patients (116) divided into low risk/high risk of rebleeding based on hepatic venous pressure gradient (HVPG)	Early TIPSS placement in patients with HVPG > 20 within 24 h of admission reduced in-patient and 1 yr mortality
García-Pagán et al[4]	RCT/role of early TIPSS in patients with oesophageal variceal haemorrhage (n = 32) within 72 h of admission vs continuation of vasoactive Tx and B-blocker/EVBL (n = 31) thereafter	Rebleeding or failure to control bleeding in 14 patients in the pharmacotherapy-EVBL group vs 1 patient in the early-TIPS group (P = 0.001)
Garcia-Pagán et al[114]	Post-RCT surveillance study/retrospective review of patients admitted for acute variceal bleeding and high risk of treatment failure treated with early-TIPSS (n = 45) or drugs/endoscopic therapy (ET) (n = 30)	Early-TIPSS group had a much lower incidence of failure to control bleeding/rebleeding than drug + ET (3 vs 15, P < 0.001). 1-yr actuarial survival was 86% vs 70% respectively (P = 0.056)
Yang et al[127]	Mata-analysis of 6 studies of covered stents vs bare metal stents	Use of polytetrafluoroethylene-covered stent-grafts associated with improved shunt patency without increasing the risk of hepatic encephalopathy and with a trend towards better survival