Impact of thrombocytopenia on failure of endoscopic variceal treatment in cirrhotic patients with acute variceal bleeding

Abstract

Background:

Acute variceal bleeding (AVB), a life-threatening complication of liver cirrhosis, can be effectively treated by endoscopy, but there is a risk of early rebleeding after endoscopic variceal treatment (EVT). Thrombocytopenia is the most common hemostatic abnormality in liver cirrhosis. However, it is still unclear about whether thrombocytopenia increases the failure of EVT in cirrhotic patients with AVB.

Objectives:

We investigated the association between thrombocytopenia and the failure of EVT in cirrhotic patients with AVB.

Design:

International multicenter, retrospective study.

Methods:

Overall, 2467 cirrhotic patients with acute gastrointestinal bleeding who were enrolled into an international multicenter study between September 30, 2020 and June 30, 2023 were retrospectively screened. Thrombocytopenia was defined as platelet count below 150 × 10⁹/L and further classified as mild (100 × 10⁹/L–150 × 10⁹/L), moderate (50 × 10⁹/L–100 × 10⁹/L), and severe (<50 × 10⁹/L). A 1:1 propensity score matching (PSM) analysis was performed. Five-day failure to control bleeding was evaluated.

Results:

Overall, 1079 patients were included, of whom 923 (85.5%) had thrombocytopenia, including mild (n = 241), moderate (n = 445), and severe (n = 237) thrombocytopenia. PSM analysis demonstrated that the rate of 5-day failure to control bleeding was not significantly different between patients with and without thrombocytopenia (mild: (12/153) 7.8% vs (7/153) 4.6%, p = 0.236; moderate: (9/155) 5.8% vs (7/155) 4.5%, p = 0.608; or severe: (5/132) 3.8% vs (7/132) 5.3%, p = 0.555).

Conclusion:

Thrombocytopenia may not influence the efficacy of EVT in cirrhotic patients with AVB.

Graphical abstract

Plain language summary

Effect of decreased platelet count on the failure to control bleeding after endoscopic therapy in cirrhotic patients presenting with hematemesis or melena

Why was the study done? Patients with cirrhosis who have hematemesis or melena are usually treated by endoscopy. However, in some patients, endoscopic therapy cannot completely control bleeding. Notably, cirrhotic patients may experience varying degrees of decreased platelet count, which may be associated with increased risk of bleeding. Until now, it is still unclear about whether decreased platelet count is associated with the failure to control bleeding after endoscopic therapy.

What did the researchers do? Cirrhotic patients with acute gastrointestinal bleeding from an international multicenter database were screened. We adjusted confounding factors that may potentially influence the efficacy of endoscopic therapy to further clarify the effect of decreased platelet count on the failure of endoscopic therapy.

What did the researchers find? Overall, 1079 patients were included, of whom 923 had decreased platelet count. Among them, platelet count was mildly, moderately, and severely decreased in 241, 445, and 237 patients, respectively. After eliminating the confounding factors, the incidence of failure of endoscopic therapy was not significantly different between patients with and without decreased platelet count.

What do the findings mean? Decreased platelet count may not influence the efficacy of endoscopic therapy in cirrhotic patients presenting with hematemesis or melena.

Keywords

endoscopy failure to control bleeding liver cirrhosis thrombocytopenia variceal bleeding

Introduction

Acute variceal bleeding (AVB) is a common and life-threatening complication of liver cirrhosis, which accounts for approximately 70% of acute gastrointestinal bleeding (AGIB) events.^1,2 AVB is attributed to the rupture of gastroesophageal varices secondary to portal hypertension.³ In spite of advances in diagnosis and treatment, the 6-week mortality of cirrhotic patients with AVB remains 15%–20%.⁴ Additionally, recurrent bleeding is frequent, ranging from 30% to 40% within the subsequent 6 weeks after an initial AVB episode, especially the first 5 days.⁵ Currently, endoscopic variceal treatment (EVT), together with medical therapy, is the golden standard for the management of AVB,⁶ but the incidence of failure to control bleeding within the initial 5 days after EVT is as high as 10%–15%.^7,8

Thrombocytopenia is the most common hemostatic abnormality occurring in cirrhosis. The pathogenesis of thrombocytopenia is multifactorial, including decreased synthesis of thrombopoietin, splenic sequestration of platelets secondary to portal hypertension, myelosuppression, and increased destruction of platelets.⁹ The prevalence and degree of thrombocytopenia is in parallel with the severity of cirrhosis. Approximately 80% of cirrhotic patients with Child-Pugh class B and C have thrombocytopenia, with a majority presenting with moderate thrombocytopenia.¹⁰

Current evidence remains very scarce about the association of thrombocytopenia with treatment failure of EVT in cirrhotic patients with AVB. To the best of our knowledge, only three previous studies have explored the association between platelet (PLT) count and short-term outcomes of cirrhotic patients with AVB after EVT. A prospective cohort study found that PLT was not significantly different between patients who failed to control bleeding after endoscopy and those who did not.¹¹ But the confounding factors were not adjusted in this study, which may influence the statistical validity of the results. Similarly, another two previous studies also demonstrated that PLT was not significantly different between patients who developed early rebleeding and those who did not.^12,13 However, the number of patients included in a study by Chau et al.¹² was very limited. Furthermore, all of them did not stratify the patients according to the severity of thrombocytopenia.^11
–13

Herein, we aimed to further explore the association of thrombocytopenia with 5-day failure to control bleeding after EVT in cirrhotic patients with AVB, based on the data from an international multicenter study.

Methods

Study design

We retrospectively screened the data of cirrhotic patients with AGIB who were enrolled into an international multicenter observational study between September 30, 2020 and June 30, 2023 (NCT04662918).¹⁴ Exclusion criteria were as follows: (i) patients who did not undergo endoscopy; (ii) patients who were not diagnosed with variceal bleeding on endoscopy; (iii) patients who did not undergo EVT; and (iv) patients who underwent splenectomy, liver transplantation, or transjugular intrahepatic portosystemic shunt. The study protocol was approved by the Medical Ethical Committee of the General Hospital of Northern Theater Command with an ethical approval number (Y (2024) 139) and was carried out following the rules of the 1975 Declaration of Helsinki. Patients’ informed consents were waived. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology statement.¹⁵

The following baseline data were collected: demographics, etiology of liver cirrhosis, decompensated complications at admission, and laboratory tests (i.e., white blood cell (WBC), PLT, prothrombin time (PT), and international normalized ratio (INR)). Child-Pugh score and Model for End-Stage Liver Disease (MELD) score at admission were calculated. According to the data collected in this study, EVT primarily included endoscopic variceal ligation (EVL), endoscopic cyanoacrylate glue injection (ECGI), and endoscopic injection sclerotherapy (EIS). Drugs mainly included vasoactive drugs (i.e., octreotide, somatostatin, and terlipressin), proton pump inhibitors, and antibiotics. Blood transfusions mainly included PLT, red blood cell (RBC), and fresh frozen plasma (FFP).

Patients were divided according to the presence of thrombocytopenia. Thrombocytopenia was defined as PLT <150 × 10⁹/L and further classified as mild (100 × 10⁹/L–150 × 10⁹/L), moderate (50 × 10⁹/L–100 × 10⁹/L), and severe (<50 × 10⁹/L).^10,16 The outcome of interest in this study was the rate of 5-day failure to control bleeding, which was defined according to the Baveno VII consensus.⁶

Statistical analyses

All statistical analyses were performed with IBM SPSS 25.0 (IBM Corp., Armonk, NY, USA) and Stata/SE 12.0 (Stata Corp., College Station, TX, USA). Continuous variables were expressed as mean ± standard deviation and median (range) and compared using the independent sample t-test or nonparametric Mann–Whitney U test. Categorical variables were expressed as frequency (percentage) and were compared using the Chi-squared test or Fisher’s exact test. Logistic regression analyses were conducted to explore whether thrombocytopenia was significantly associated with the risk of 5-day failure to control bleeding after EVT. Crude odd ratios (cORs) with their 95% confidence intervals (CIs) were calculated in univariate analyses. Adjusted odd ratios (aORs) with their 95% CIs were calculated in multivariate analyses after adjusting for age, gender, blood transfusions (i.e., PLT, RBC, and FFP), Child-Pugh score, and MELD score. A 1:1 propensity score matching (PSM) analysis was performed to match baseline characteristics between patients with and without thrombocytopenia. Matching factors included age, gender, blood transfusions (i.e., PLT, RBC, and FFP), Child-Pugh score, and MELD score. A two-tailed p < 0.05 was considered statistically significant.

Results

Patients

A total of 1079 patients were included (Figure 1), of whom 923 (85.5%) had thrombocytopenia, including mild (n = 241), moderate (n = 445), and severe (n = 237) thrombocytopenia. Patient characteristics are summarized in Table 1. The median age was 57 years (range: 19–92), and 762 (70.6%) patients were male. Hepatitis B virus infection (52.9%) was the most common etiology of liver cirrhosis followed by alcohol abuse (16.6%). Among them, 684 (63.4%) patients had ascites at admission, and 635 (58.9%) had history of variceal bleeding. The median Child-Pugh score and MELD score were 7.00 (range: 5.00–15.00) and 12.14 (range: 6.43–40.00), respectively. The rate of 5-day failure to control bleeding was 5.2% (n = 56/1079).

Figure 1.

Flowchart of patients’ screening and grouping.

Table 1.

Baseline characteristics of the study cohort.

Variables	Overall
Variables	No. Pts	Mean ± SD, median (range) or frequency (percentage)
Demographics
Age (years)	1079	56.94 ± 12.27 57.00 (19.00–92.00)
Male (%)	1079	762 (70.6%)
Etiology of liver cirrhosis
HBV (%)	1079	571 (52.9%)
HCV (%)	1079	74 (6.9%)
Alcohol (%)	1079	179 (16.6%)
Diabetes	1079	285 (26.4%)
Complications of liver cirrhosis (past and current)
Ascites (%)	1079	684 (63.4%)
HE (%)	1079	108 (10.0%)
History of variceal bleeding (%)	1079	635 (58.9%)
ACLF (%)	1079	56 (5.2%)
HCC (%)	1079	184 (17.1%)
Laboratory tests
WBC (10⁹/L)	1079	5.97 ± 3.97 4.92 (0.37–28.48)
HB (g/L)	1079	82.91 ± 23.56 81.00 (16.00–161.00)
PLT (10⁹/L)	1079	96.35 ± 69.65 78.00 (3.00–836.00)
TBIL (μmol/L)	1079	34.92 ± 41.71 23.90 (2.30–408.00)
ALB (g/L)	1079	31.12 ± 5.67 30.80 (12.40–55.00)
Scr (μmol/L)	1079	74.65 ± 49.98 64.80 (7.00–859.25)
PT (s)	1077	15.88 ± 3.49 15.20 (10.00–39.90)
INR	1079	1.38 ± 0.32 1.31 (0.90–4.13)
INR ⩾ 1.5 (%)	1079	251 (23.3%)
Child-Pugh score	1079	7.63 ± 1.78 7.00 (5.00–15.00)
MELD score	1079	13.40 ± 5.08 12.14 (6.43–40.00)
Endoscopy
EV (%)	1079	1055 (97.8%)
GV (%)	1079	752 (69.7%)
Medical therapy
Vasoconstrictors	1079	1029 (95.3%)
PPIs	1079	1049 (97.2%)
Antibiotics	1079	885 (82.0%)
Blood transfusions
PLT	1079	27 (2.5%)
RBC	1079	476 (44.1%)
FFP	1079	227 (21.0%)
Endoscopic variceal therapy
EVL	1079	454 (42.1%)
EIS	1079	52 (4.8%)
ECGI	1079	114 (10.6%)
EVL + EIS	1079	23 (2.1%)
EVL + ECGI	1079	169 (15.7%)
EIS + ECGI	1079	174 (16.1%)
EVL + EIS + ECGI	1079	76 (7.0%)
Others	1079	17 (1.6%)
5-Day failure to control bleeding (%)	1079	56 (5.2%)

ACLF, acute-on-chronic liver failure; ALB, albumin; ECGI, endoscopic cyanoacrylate glue injection; EIS, endoscopic injection sclerotherapy; EV, esophageal varices; EVL, endoscopic variceal ligation; FFP, fresh frozen plasma; GV, gastric varices; HB, hemoglobin; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HE, hepatic encephalopathy; INR, international normalized ratio; MELD, model for end-stage liver disease; No. Pts, number of patients; PLT, platelet; PPIs, proton pump inhibitors; PT, prothrombin time; RBC, red blood cell; Scr, serum creatinine; SD, standard deviation; TBIL, total bilirubin; WBC, white blood cell.

Mild thrombocytopenia versus normal PLT

In the overall analysis, patients with mild thrombocytopenia had significantly higher prevalence of esophageal varices (EV) (98.3% vs 94.9%, p = 0.049), but lower WBC (7.06 × 10⁹/L vs 9.44 × 10⁹/L, p < 0.001) than those with normal PLT. However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (5.4% vs 4.5%, p = 0.687; Table 2). Univariate logistic regression analysis showed that the presence of mild thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (cOR = 1.214, 95% CI = 0.473–3.112, p = 0.687). Multivariate logistic regression analysis showed that the presence of mild thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (aOR = 1.713, 95% CI = 0.618–4.748, p = 0.300).

Table 2.

Overall analysis and PSM analysis between patients with mild thrombocytopenia and normal PLT.

Variables	Overall analysis					PSM analysis
	With mild thrombocytopenia		With normal PLT		p Value	With mild thrombocytopenia		With normal PLT		p Value
	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value
Demographics
Age (years)	241	58.73 ± 11.95 58.00 (31.00–89.00)	156	58.66 ± 13.40 58.00 (25.00–92.00)	0.959	153	58.62 ± 12.44 59.00 (31.00–89.00)	153	58.69 ± 13.53 58.00 (25.00–92.00)	0.965
Male (%)	241	163 (67.6%)	156	122 (78.2%)	0.022	153	115 (75.2%)	153	119 (77.8%)	0.590
Etiology of liver cirrhosis
HBV (%)	241	98 (40.7%)	156	79 (50.6%)	0.051	153	67 (43.8%)	153	78 (51.0%)	0.208
HCV (%)	241	17 (7.1%)	156	6 (3.8%)	0.181	153	12 (7.8%)	153	6 (3.9%)	0.145
Alcohol (%)	241	48 (19.9%)	156	27 (17.3%)	0.517	153	31 (20.3%)	153	25 (16.3%)	0.375
Diabetes	241	81 (33.6%)	156	49 (31.4%)	0.648	153	54 (35.5%)	153	48 (31.4%)	0.467
Complications of liver cirrhosis (past and current)
Ascites (%)	241	148 (61.4%)	156	94 (60.3%)	0.818	153	89 (58.2%)	153	92 (60.1%)	0.727
HE (%)	241	23 (9.5%)	156	25 (16.0%)	0.053	153	15 (9.8%)	153	25 (16.3%)	0.090
History of variceal bleeding (%)	241	122 (50.6%)	156	90 (57.7%)	0.168	153	82 (53.6%)	153	89 (58.2%)	0.420
ACLF (%)	241	16 (6.6%)	156	9 (5.8%)	0.728	153	9 (5.9%)	153	9 (5.9%)	1.000
HCC (%)	241	57 (23.7%)	156	32 (20.5%)	0.464	153	46 (30.1%)	153	32 (20.9%)	0.066
Laboratory tests
WBC (10⁹/L)	241	7.06 ± 3.73 6.33 (1.70–22.27)	156	9.44 ± 4.30 8.82 (1.47–22.20)	<0.001	153	7.42 ± 4.02 6.90 (1.95–22.27)	153	9.46 ± 4.33 8.80 (1.47–22.20)	<0.001
HB (g/L)	241	85.56 ± 22.06 84.00 (22.00–161.00)	156	81.80 ± 27.09 80.50 (16.00–151.00)	0.149	153	85.28 ± 22.31 83.00 (34.00–161.00)	153	81.92 ± 27.33 81.00 (16.00–151.00)	0.239
PLT (10⁹/L)	241	121.34 ± 14.20 120.00 (100.00–149.00)	156	222.06 ± 91.67 193.00 (150.00–836.00)	<0.001	153	120.99 ± 14.66 119.00 (100.00–149.00)	153	222.50 ± 92.49 192.00 (150.00–836.00)	<0.001
TBIL (μmol/L)	241	36.29 ± 51.96 21.50 (6.70–391.50)	156	29.72 ± 35.75 19.90 (5.30–295.83)	0.173	153	41.38 ± 62.61 23.10 (6.70–391.59)	153	29.92 ± 36.06 19.90 (5.30–295.83)	0.093
ALB (g/L)	241	30.96 ± 5.64 30.30 (18.00–50.40)	156	30.82 ± 5.78 30.65 (16.60–47.20)	0.922	153	30.60 ± 5.77 30.00 (18.00–50.40)	153	30.70 ± 5.76 30.60 (16.60–47.20)	0.882
Scr (μmol/L)	241	72.57 ± 40.26 63.69 (26.52–477.36)	156	82.17 ± 50.07 67.50 (30.30–433.16)	0.062	153	75.21 ± 45.00 66.00 (28.00–477.36)	153	81.84 ± 50.31 67.18 (30.30–433.16)	0.316
PT (s)	241	15.42 ± 3.54 14.60 (10.00–32.90)	156	15.39 ± 3.36 14.90 (10.80–36.80)	0.840	153	15.68 ± 3.55 14.70 (11.20–32.90)	153	15.41 ± 3.39 14.90 (10.80–36.80)	0.561
INR	241	1.34 ± 0.31 1.27 (0.90–3.10)	156	1.34 ± 0.32 1.27 (0.99–3.39)	0.837	153	1.36 ± 0.31 1.28 (1.00–3.10)	153	1.34 ± 0.32 1.27 (0.99–3.39)	0.380
INR ⩾ 1.5 (%)	241	43 (17.8%)	156	28 (17.9%)	0.978	153	28 (18.3%)	153	28 (18.3%)	1.000
Child-Pugh score	241	7.60 ± 1.76 7.00 (5.00–13.00)	156	7.59 ± 1.85 7.00 (5.00–12.00)	0.888	153	7.63 ± 1.77 7.00 (5.00–13.00)	153	7.63 ± 1.85 7.00 (5.00–12.00)	0.959
MELD score	241	12.86 ± 5.49 11.02 (6.43–33.36)	156	13.60 ± 5.45 12.41 (6.43–32.27)	0.144	153	13.50 ± 5.88 11.17 (6.67–33.36)	153	13.53 ± 5.47 12.28 (6.43–32.27)	0.849
Endoscopy
EV (%)	241	237 (98.3%)	156	148 (94.9%)	0.049	153	151 (98.7%)	153	145 (94.8%)	0.054
GV (%)	241	148 (61.4%)	156	92 (59.0%)	0.628	153	90 (58.8%)	153	91 (59.5%)	0.907
Medical therapy
Vasoconstrictors	241	226 (93.8%)	156	149 (95.5%)	0.460	153	141 (92.2%)	153	147 (96.1%)	0.145
PPIs	241	226 (93.8%)	156	149 (95.5%)	0.460	153	145 (94.8%)	153	146 (95.4%)	0.791
Antibiotics	241	190 (78.8%)	156	130 (83.3%)	0.269	153	128 (83.7%)	153	127 (83.0%)	0.878
Blood transfusions
PLT	241	2 (0.8%)	156	1 (0.6%)	1.000	153	1 (0.7%)	153	1 (0.7%)	1.000
RBC	241	109 (45.2%)	156	81 (51.9%)	0.192	153	79 (51.6%)	153	78 (51.0%)	1.000
FFP	241	34 (14.1%)	156	28 (17.9%)	0.303	153	31 (20.3%)	153	28 (18.3%)	0.664
Endoscopic variceal therapy
EVL	241	114 (47.3%)	156	81 (51.9%)	0.368	153	75 (49.0%)	153	79 (51.6%)	0.647
EIS	241	10 (4.1%)	156	6 (3.8%)	0.881	153	6 (3.9%)	153	6 (3.9%)	1.000
ECGI	241	26 (10.8%)	156	20 (12.8%)	0.537	153	16 (10.5%)	153	20 (13.1%)	0.478
EVL + EIS	241	7 (2.9%)	156	2 (1.3%)	0.474	153	4 (2.6%)	153	2 (1.3%)	0.680
EVL + ECGI	241	33 (13.7%)	156	21 (13.5%)	0.948	153	19 (12.4%)	153	21 (13.7%)	0.734
EIS + ECGI	241	34 (14.1%)	156	20 (12.8%)	0.715	153	22 (14.4%)	153	20 (13.1%)	0.740
EVL + EIS + ECGI	241	13 (5.4%)	156	3 (1.9%)	0.086	153	8 (5.2%)	153	3 (2.0%)	0.125
Others	241	4 (1.7%)	156	3 (1.9%)	1.000	153	3 (2.0%)	153	2 (1.3%)	1.000
5-Day failure to control bleeding (%)	241	13 (5.4%)	156	7 (4.5%)	0.687	153	12 (7.8%)	153	7 (4.6%)	0.236

The values in bold mean statistically significant.

ACLF, acute-on-chronic liver failure; ALB, albumin; ECGI, endoscopic cyanoacrylate glue injection; EIS, endoscopic injection sclerotherapy; EV, esophageal varices; EVL, endoscopic variceal ligation; FFP, fresh frozen plasma; GV, gastric varices; HB, hemoglobin; HBV, Hepatitis B Virus; HCC, hepatocellular carcinoma; HCV, Hepatitis C Virus; HE, hepatic encephalopathy; INR, international normalized ratio; MELD, model for end-stage liver disease; No. Pts, number of patients; PLT, platelet; PPIs, proton pump inhibitors; PSM, propensity score matching; PT, prothrombin time; RBC, red blood cell; Scr, serum creatinine; SD, standard deviation; TBIL, total bilirubin; WBC, white blood cell.

In the PSM analysis, 153 patients were matched to each group. Patients with mild thrombocytopenia had significantly lower WBC (7.42 × 10⁹/L vs 9.46 × 10⁹/L, p < 0.001) than those with normal PLT. However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (7.8% vs 4.6%, p = 0.236; Table 2).

Moderate thrombocytopenia versus normal PLT

In the overall analysis, patients with moderate thrombocytopenia had significantly higher total bilirubin (TBIL) (35.53 μmol/L vs 29.72 μmol/L, p < 0.001), PT (15.98s vs 15.39 s, p = 0.023), INR (1.39 vs 1.34, p = 0.011), prevalence of EV (98.0% vs 94.9%, p = 0.044), and gastric varices (GV) (73.9% vs 59.0%, p < 0.001), but lower WBC (5.35 × 10⁹/L vs 9.44 × 10⁹/L, p < 0.001) than those with normal PLT (Table 3). However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (5.8% vs 4.5%, p = 0.522). Univariate logistic regression analysis showed that the presence of moderate thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (cOR = 1.321, 95% CI = 0.562–3.107, p = 0.524). Multivariate logistic regression analysis showed that the presence of moderate thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (aOR = 1.384, 95% CI = 0.567–3.380, p = 0.476).

Table 3.

Overall analysis and PSM analysis between patients with moderate thrombocytopenia and normal PLT.

Variables	Overall analysis					PSM analysis
	With moderate thrombocytopenia		With normal PLT		p Value	With moderate thrombocytopenia		With normal PLT		p Value
	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value
Demographics
Age (years)	445	56.85 ± 11.80 57.00 (19.00-90.00)	156	58.66 ± 13.40 58.00 (25.00-92.00)	0.112	155	59.86 ± 12.11 59.00 (32.00-90.00)	155	58.49 ± 13.27 58.00 (25.00-92.00)	0.342
Male (%)	445	308 (69.2%)	156	122 (78.2%)	0.032	155	119 (76.8%)	155	121 (78.1%)	0.786
Etiology of liver cirrhosis
HBV (%)	445	243 (54.6%)	156	79 (50.6%)	0.393	155	79 (51.0%)	155	79 (51.0%)	1.000
HCV (%)	445	33 (7.4%)	156	6 (3.8%)	0.119	155	12 (7.7%)	155	6 (3.9%)	0.145
Alcohol (%)	445	73 (16.4%)	156	27 (17.3%)	0.794	155	24 (15.5%)	155	27 (17.4%)	0.646
Diabetes	445	111 (24.9%)	156	49 (31.4%)	0.116	155	44 (28.4%)	155	49 (31.6%)	0.535
Complications of liver cirrhosis (past and current)
Ascites (%)	445	279 (62.7%)	156	94 (60.3%)	0.589	155	102 (65.8%)	155	94 (60.6%)	0.346
HE (%)	445	43 (9.7%)	156	25 (16.0%)	0.031	155	17 (11.0%)	155	25 (16.1%)	0.184
History of variceal bleeding (%)	445	265 (59.6%)	156	90 (57.7%)	0.685	155	92 (59.4%)	155	90 (58.1%)	0.818
ACLF (%)	445	19 (4.3%)	156	9 (5.8%)	0.444	155	10 (6.5%)	155	9 (5.8%)	0.813
HCC (%)	445	63 (14.2%)	156	32 (20.5%)	0.061	155	32 (20.6%)	155	31 (20.0%)	0.888
Laboratory tests
WBC (10⁹/L)	445	5.35 ± 3.52 4.56 (0.89–28.48)	156	9.44 ± 4.30 8.82 (1.47–22.20)	<0.001	155	5.48 ± 3.55 4.59 (1.20–24.30)	155	9.44 ± 4.32 8.80 (1.47–22.20)	<0.001
HB (g/L)	445	83.01 ± 23.30 81.00 (26.00–161.00)	156	81.80 ± 27.09 80.50 (16.00–151.00)	0.688	155	82.05 ± 20.78 81.00 (41.00–161.00)	155	81.90 ± 27.14 81.00 (16.00–151.00)	0.974
PLT (10⁹/L)	445	70.99 ± 13.86 70.00 (50.00–99.00)	156	222.06 ± 91.67 193.00 (150.00–836.00)	<0.001	155	71.82 ± 14.02 72.00 (50.00–99.00)	155	221.41 ± 90.61 192.00 (150.00–836.00)	<0.001
TBIL (μmol/L)	445	35.53 ± 43.44 23.95 (2.30–408.00)	156	29.72 ± 35.75 19.90 (5.30–295.83)	<0.001	155	40.05 ± 55.69 24.50 (5.20–383.90)	155	29.87 ± 35.81 19.90 (5.30–295.83)	0.004
ALB (g/L)	445	31.26 ± 5.48 31.00 (15.00–55.00)	156	30.82 ± 5.78 30.65 (16.60–47.20)	0.403	155	31.19 ± 5.15 31.00 (19.00–55.00)	155	30.81 ± 5.79 30.60 (16.60–47.20)	0.505
Scr (μmol/L)	445	74.67 ± 57.36 64.00 (9.10–859.25)	156	82.17 ± 50.07 67.50 (30.30–433.16)	0.043	155	82.57 ± 80.27 66.80 (27.70–859.25)	155	81.00 ± 48.05 67.30 (30.30–433.16)	0.576
PT (s)	443	15.98 ± 3.53 15.11 (10.30–39.90)	156	15.39 ± 3.36 14.90 (10.80–36.80)	0.023	154	15.99 ± 3.46 15.20 (10.30–35.00)	155	15.40 ± 3.37 14.90 (10.80–36.80)	0.052
INR	445	1.39 ± 0.34 1.32 (0.93–4.13)	156	1.34 ± 0.32 1.27 (0.99–3.39)	0.011	155	1.39 ± 0.33 1.30 (0.93–3.30)	155	1.34 ± 0.32 1.27 (0.99–3.39)	0.089
INR ⩾ 1.5 (%)	445	102 (22.9%)	156	28 (17.9%)	0.194	155	36 (23.2%)	155	28 (18.1%)	0.262
Child-Pugh score	445	7.61 ± 1.81 7.00 (5.00–15.00)	156	7.59 ± 1.85 7.00 (5.00–12.00)	0.942	155	7.72 ± 1.91 7.00 (5.00–15.00)	155	7.60 ± 1.85 7.00 (5.00–12.00)	0.706
MELD score	445	13.43 ± 5.12 11.89 (6.43–40.00)	156	13.60 ± 5.45 12.41 (6.43–32.27)	0.856	155	13.89 ± 5.61 12.09 (6.43–40.00)	155	13.56 ± 5.45 12.38 (6.43–32.27)	0.478
Endoscopy
EV (%)	445	436 (98.0%)	156	148 (94.9%)	0.044	155	152 (98.1%)	155	147 (94.8%)	0.125
GV (%)	445	329 (73.9%)	156	92 (59.0%)	<0.001	155	109 (70.3%)	155	92 (59.4%)	0.043
Medical therapy
Vasoconstrictors	445	428 (96.2%)	156	149 (95.5%)	0.714	155	149 (96.1%)	155	149 (96.1%)	1.000
PPIs	445	439 (98.7%)	156	149 (95.5%)	0.046	155	153 (98.7%)	155	148 (95.5%)	0.176
Antibiotics	445	368 (82.7%)	156	130 (83.3%)	0.856	155	131 (84.5%)	155	129 (83.2%)	0.757
Blood transfusions
PLT	445	6 (1.3%)	156	1 (0.6%)	0.783	155	1 (0.6%)	155	1 (0.6%)	1.000
RBC	445	189 (42.5%)	156	81 (51.9%)	0.041	155	75 (48.4%)	155	80 (51.6%)	0.570
FFP	445	88 (19.8%)	156	28 (17.9%)	0.619	155	30 (19.4%)	155	28 (18.1%)	0.771
Endoscopic variceal therapy
EVL	445	177 (39.8%)	156	81 (51.9%)	0.008	155	65 (41.9%)	155	80 (51.6%)	0.088
EIS	445	22 (4.9%)	156	6 (3.8%)	0.576	155	7 (4.5%)	155	6 (3.9%)	0.777
ECGI	445	39 (8.8%)	156	20 (12.8%)	0.143	155	8 (5.2%)	155	20 (12.9%)	0.017
EVL + EIS	445	8 (1.8%)	156	2 (1.3%)	0.945	155	1 (0.6%)	155	2 (1.3%)	1.000
EVL + ECGI	445	74 (16.6%)	156	21 (13.5%)	0.351	155	23 (14.8%)	155	21 (13.5%)	0.745
EIS + ECGI	445	84 (18.9%)	156	20 (12.8%)	0.085	155	34 (21.9%)	155	20 (12.9%)	0.036
EVL + EIS + ECGI	445	33 (7.4%)	156	3 (1.9%)	0.013	155	14 (9.0%)	155	3 (1.9%)	0.006
Others	445	8 (1.8%)	156	3 (1.9%)	1.000	155	3 (1.9%)	155	3 (1.9%)	1.000
5-Day failure to control bleeding (%)	445	26 (5.8%)	156	7 (4.5%)	0.522	155	9 (5.8%)	155	7 (4.5%)	0.608

The values in bold mean statistically significant.

ACLF, acute-on-chronic liver failure; ALB, albumin; ECGI, endoscopic cyanoacrylate glue injection; EIS, endoscopic injection sclerotherapy; EV, esophageal varices; EVL, endoscopic variceal ligation; FFP, fresh frozen plasma; GV, gastric varices; HB, hemoglobin; HBV, Hepatitis B Virus; HCC, hepatocellular carcinoma; HCV, Hepatitis C Virus; HE, hepatic encephalopathy; INR, international normalized ratio; MELD, model for end-stage liver disease; No. Pts, number of patients; PLT, platelet; PPIs, proton pump inhibitors; PSM, propensity score matching; PT, prothrombin time; RBC, red blood cell; Scr, serum creatinine; SD, standard deviation; TBIL, total bilirubin; WBC, white blood cell.

In the PSM analysis, 155 patients were matched to each group. Patients with moderate thrombocytopenia had significantly higher TBIL (40.05 μmol/L vs 29.87 μmol/L, p = 0.004), prevalence of GV (70.3% vs 59.4%, p = 0.043), but lower WBC (5.48 × 10⁹/L vs 9.44 × 10⁹/L, p < 0.001) than those with normal PLT. However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (5.8% vs 4.5%, p = 0.608; Table 3).

Severe thrombocytopenia versus normal PLT

In the overall analysis, patients with severe thrombocytopenia had significantly higher TBIL (35.83 μmol/L vs 29.72 μmol/L, p < 0.001), PT (16.49s vs 15.39s, p < 0.001), INR (1.44 vs 1.34, p < 0.001), and prevalence of GV (77.2% vs 59.0%, p < 0.001), but lower WBC (3.71 × 10⁹/L vs 9.44 × 10⁹/L, p < 0.001). However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (4.2% vs 4.5%, p = 0.898; Table 4). Univariate logistic regression analysis showed that the presence of severe thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (cOR = 0.938, 95% CI = 0.349–2.518, p = 0.898). Multivariate logistic regression analysis showed that the presence of severe thrombocytopenia was not independently associated with the rate of 5-day failure to control bleeding (aOR = 0.771, 95% CI = 0.256–2.315, p = 0.642).

Table 4.

Overall analysis and PSM analysis between patients with severe thrombocytopenia and normal PLT.

Variables	Overall analysis					PSM analysis
	With severe thrombocytopenia		With normal PLT		p Value	With severe thrombocytopenia		With normal PLT		p Value
	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value	No. Pts	Mean ± SD, median (range) or frequency (percentage)	No. Pts	Mean ± SD, median (range) or frequency (percentage)	p Value
Demographics
Age (years)	237	54.15 ± 12.20 54.00 (26.00–85.00)	156	58.66 ± 13.40 58.00 (25.00–92.00)	<0.001	132	56.35 ± 11.48 55.50 (28.00–85.00)	132	56.65 ± 12.54 57.50 (25.00–89.00)	0.838
Male (%)	237	169 (71.3%)	156	122 (78.2%)	0.127	132	103 (78.0%)	132	100 (75.8%)	0.661
Etiology of liver cirrhosis
HBV (%)	237	151 (63.7%)	156	79 (50.6%)	0.010	132	87 (65.9%)	132	75 (56.8%)	0.129
HCV (%)	237	18 (7.6%)	156	6 (3.8%)	0.129	132	12 (9.1%)	132	5 (3.8%)	0.079
Alcohol (%)	237	31 (13.1%)	156	27 (17.3%)	0.248	132	18 (13.6%)	132	22 (16.7%)	0.492
Diabetes	237	44 (18.6%)	156	49 (31.4%)	0.003	132	29 (22.0%)	132	39 (29.5%)	0.159
Complications of liver cirrhosis (past and current)
Ascites (%)	237	163 (68.8%)	156	94 (60.3%)	0.082	132	82 (62.1%)	132	81 (61.4%)	0.899
HE (%)	237	17 (7.2%)	156	25 (16.0%)	0.005	132	8 (6.1%)	132	20 (15.2%)	0.016
History of variceal bleeding (%)	237	158 (66.7%)	156	90 (57.7%)	0.071	132	88 (66.7%)	132	77 (58.3%)	0.162
ACLF (%)	237	12 (5.1%)	156	9 (5.8%)	0.761	132	4 (3.0%)	132	7 (5.3%)	0.355
HCC (%)	237	32 (13.5%)	156	32 (20.5%)	0.066	132	18 (13.6%)	132	27 (20.5%)	0.141
Laboratory tests
WBC (10⁹/L)	237	3.71 ± 2.70 2.87 (0.37–20.00)	156	9.44 ± 4.30 8.82 (1.47–22.20)	<0.001	132	3.93 ± 2.86 2.92 (0.68–20.00)	132	9.56 ± 4.45 8.82 (1.47–22.20)	<0.001
HB (g/L)	237	80.77 ± 22.89 79.00 (28.20–149.00)	156	81.80 ± 27.09 80.50 (16.00–151.00)	0.626	132	80.48 ± 24.25 77.50 (28.20–149.00)	132	83.49 ± 27.78 82.00 (16.00–151.00)	0.215
PLT (10⁹/L)	237	35.82 ± 9.74 38.00 (3.00–49.00)	156	222.06 ± 91.67 193.00 (150.00–836.00)	<0.001	132	35.45 ± 9.52 37.00 (10.00–49.00)	132	224.86 ± 96.40 195.00 (150.00–836.00)	<0.001
TBIL (μmol/L)	237	35.83 ± 28.12 27.70 (6.84–240.90)	156	29.72 ± 35.75 19.90 (5.30–295.83)	<0.001	132	33.64 ± 26.22 26.70 (6.84–240.90)	132	27.60 ± 26.36 19.25 (5.30–200.00)	<0.001
ALB (g/L)	237	31.21 ± 6.00 30.70 (12.40–49.20)	156	30.82 ± 5.78 30.65 (16.60–47.20)	0.525	132	31.39 ± 6.22 30.90 (12.40–49.20)	132	30.93 ± 5.69 30.65 (16.60–47.20)	0.531
Scr (μmol/L)	237	71.80 ± 43.33 64.97 (7.00–491.50)	156	82.17 ± 50.07 67.50 (30.30–433.16)	0.053	132	70.21 ± 25.25 66.85 (7.00–150.00)	132	79.99 ± 50.37 66.50 (30.30–433.16)	0.604
PT (s)	237	16.49 ± 3.35 15.70 (11.00–36.40)	156	15.39 ± 3.36 14.90 (10.80–36.80)	<0.001	132	16.33 ± 3.43 15.70 (11.80–36.40)	132	15.42 ± 3.58 14.70 (10.80–36.80)	0.001
INR	237	1.44 ± 0.29 1.37 (0.95–3.13)	156	1.34 ± 0.32 1.27 (0.99–3.39)	<0.001	132	1.42 ± 0.28 1.36 (1.03–3.13)	132	1.34 ± 0.33 1.25 (0.99–3.39)	<0.001
INR ⩾ 1.5 (%)	237	78 (32.9%)	156	28 (17.9%)	0.001	132	39 (29.5%)	132	24 (18.2%)	0.030
Child-Pugh score	237	7.73 ± 1.72 8.00 (5.00–13.00)	156	7.59 ± 1.85 7.00 (5.00–12.00)	0.341	132	7.52 ± 1.68 7.00 (5.00–13.00)	132	7.58 ± 1.83 7.00 (5.00–12.00)	0.944
MELD score	237	13.78 ± 4.25 13.07 (6.87–27.90)	156	13.60 ± 5.45 12.41 (6.43–32.27)	0.137	132	13.31 ± 3.98 12.52 (6.87–26.10)	132	13.44 ± 5.49 11.95 (6.43–32.27)	0.274
Endoscopy
EV (%)	237	234 (98.7%)	156	148 (94.9%)	0.050	132	130 (98.5%)	132	125 (94.7%)	0.090
GV (%)	237	183 (77.2%)	156	92 (59.0%)	<0.001	132	104 (78.8%)	132	82 (62.1%)	0.003
Medical therapy
Vasoconstrictors	237	226 (95.4%)	156	149 (95.5%)	0.943	132	125 (94.7%)	132	126 (95.5%)	0.776
PPIs	237	235 (99.2%)	156	149 (95.5%)	0.044	132	131 (99.2%)	132	127 (96.2%)	0.215
Antibiotics	237	197 (83.1%)	156	130 (83.3%)	0.956	132	108 (81.8%)	132	107 (81.1%)	0.874
Blood transfusions
PLT	237	18 (7.6%)	156	1 (0.6%)	0.002	132	0 (0.0%)	132	1 (0.8%)	1.000
RBC	237	97 (40.9%)	156	81 (51.9%)	0.032	132	65 (49.2%)	132	60 (45.5%)	0.538
FFP	237	77 (32.5%)	156	28 (17.9%)	0.001	132	35 (26.5%)	132	28 (21.2%)	0.312
Endoscopic variceal therapy
EVL	237	82 (34.6%)	156	81 (51.9%)	0.001	132	49 (37.1%)	132	66 (50.0%)	0.035
EIS	237	14 (5.9%)	156	6 (3.8%)	0.363	132	8 (6.1%)	132	5 (3.8%)	0.393
ECGI	237	29 (12.2%)	156	20 (12.8%)	0.864	132	18 (13.6%)	132	16 (12.1%)	0.713
EVL + EIS	237	6 (2.5%)	156	2 (1.3%)	0.622	132	2 (1.5%)	132	2 (1.5%)	1.000
EVL + ECGI	237	41 (17.3%)	156	21 (13.5%)	0.307	132	23 (17.4%)	132	19 (14.4%)	0.501
EIS + ECGI	237	36 (15.2%)	156	20 (12.8%)	0.511	132	15 (11.4%)	132	19 (14.4%)	0.462
EVL + EIS + ECGI	237	27 (11.4%)	156	3 (1.9%)	0.001	132	16 (12.1%)	132	3 (2.3%)	0.002
Others	237	2 (0.8%)	156	3 (1.9%)	0.635	132	1 (0.8%)	132	2 (1.5%)	1.000
5-Day failure to control bleeding (%)	237	10 (4.2%)	156	7 (4.5%)	0.898	132	5 (3.8%)	132	7 (5.3%)	0.555

The values in bold mean statistically significant.

ACLF, acute-on-chronic liver failure; ALB, albumin; ECGI, endoscopic cyanoacrylate glue injection; EIS, endoscopic injection sclerotherapy; EV, esophageal varices; EVL, endoscopic variceal ligation; FFP, fresh frozen plasma; GV, gastric varices; HB, hemoglobin; HBV, Hepatitis B Virus; HCC, hepatocellular carcinoma; HCV, Hepatitis C Virus; HE, hepatic encephalopathy; INR, international normalized ratio; MELD, model for end-stage liver disease; No. Pts, number of patients; PLT, platelet; PPIs, proton pump inhibitors; PSM, propensity score matching; PT, prothrombin time; RBC, red blood cell; Scr, serum creatinine; SD, standard deviation; TBIL, total bilirubin; WBC, white blood cell.

In the PSM analysis, 132 patients were matched to each group. Patients with severe thrombocytopenia had significantly higher TBIL (33.64 μmol/L vs 27.60 μmol/L, p < 0.001), PT (16.33s vs 15.42s, p = 0.001), INR (1.42 vs 1.34, p < 0.001), and prevalence of GV (78.8% vs 62.1%, p = 0.003), but lower WBC (3.93 × 10⁹/L vs 9.56 × 10⁹/L, p < 0.001) than those with normal PLT. However, the rate of 5-day failure to control bleeding was not significantly different between the two groups (3.8% vs 5.3%, p = 0.555; Table 4).

Discussion

Hepatic venous pressure gradient (HVPG) >20 mmHg is the most widely recognized predictor of 5-day failure after EVT.⁷ However, the applicability of HVPG measurement in patients with AVB is limited, because it is invasive and only available in specialized centers.^17,18 By comparison, low PLT serves as a convenient noninvasive predictor for the presence of large gastroesophageal varices,^19
–21 which are a consequence of high portal pressure and more prone to cause variceal rupture.²² Our previous study found that PLT was significantly lower in cirrhotic patients with AVB than those without, indicating a close association of PLT with portal hypertension-related bleeding in cirrhosis.²³ Recent studies have further investigated the association of PLT with the outcomes of cirrhotic patients after EVT.^24
–26 Most of them focused on patients who underwent prophylactic EVT and found that the risk of bleeding after prophylactic EVT was not associated with PLT.^24
–26 But there are very limited data on the association between thrombocytopenia and failure to control bleeding within 5 days after EVT. A previous Italian study by Amitrano et al.¹¹ prospectively evaluated the risk factors of 5-day failure in 185 cirrhotic patients with AVB after EVT, and showed that Child-Pugh class C, WBC count over 10 × 10⁹/L, and portal vein thrombosis were independent predictors of the 5-day failure, but not PLT (89 × 10⁹/L vs 105 × 10⁹/L, p = 0.184). Besides, a Chinese nationwide study by Huang et al.¹³ retrospectively included 1399 cirrhotic patients with AVB to explore the risk factors of rebleeding within 5 days after emergency endoscopy, and showed that PLT was not significantly associated with the risk of rebleeding within 5 days after EVT. Another study conducted by Chau et al.¹² in the United Kingdom prospectively included 20 cirrhotic patients with AVB to explore the predictors of rebleeding within 1–7 days after EVT and found that PLT was not significantly different between patients who developed early rebleeding and those who did not after therapeutic endoscopy (42 × 10⁹/L vs 71 × 10⁹/L, p = 0.60). Similarly, the main finding of our study was that thrombocytopenia may not be associated with failure to control bleeding within 5 days after EVT in cirrhotic patients with AVB. A major explanation for our finding should be that PLT function was often preserved in cirrhotic patients with thrombocytopenia, possibly due to elevated levels of von Willebrand factor.^27
–29 Another potential explanation could be that multiple risk factors for 5-day failure to control bleeding after EVT, such as hepatic encephalopathy, more severe cirrhosis, and presence of hepatocellular carcinoma,^30,31 were not significantly different between cirrhotic patients with and without thrombocytopenia in our patients. There is a seemingly counterintuitive phenomenon that the Child-Pugh score and MELD score were not significantly different between the patients with and without thrombocytopenia. Thus, thrombocytopenia may reflect an acute consumption of PLT in the setting of AGIB, but not sufficiently indicate the severity of liver cirrhosis.

As compared to previous studies, our study has several advantages in terms of study design. First, in our study, the data of cirrhotic patients with AGIB were prospectively and consecutively collected at 23 hospitals from 8 countries. Thus, such a large sample size in our study enhanced the credibility of our conclusions, and multiple sources of the included data allows for the generalizability of our conclusions. Second, the PSM analyses were performed to balance the severity of liver cirrhosis between the two groups. Notably, previous studies have demonstrated that the severity of liver cirrhosis evaluated by Child-Pugh class or MELD score was a strong predictor of 5-day failure in cirrhotic patients with AVB.^7,31 Accordingly, our statistical results should be more reliable to reflect the impact of thrombocytopenia on the outcomes of cirrhotic patients with AVB. Third, we stratified the severity of thrombocytopenia and further explored the association between different grades of thrombocytopenia and the rate of failure to control bleeding within 5 days.

Our study also had several limitations. First, our study was retrospective with selection bias. Second, EVT was performed at different hospitals. The difference in the levels of endoscopists’ skills could potentially influence the patients’ outcomes. Third, the long-term outcomes of cirrhotic patients with AVB were not evaluated. Fourth, the data regarding liver stiffness value, HVPG, and portal vein thrombosis had not been collected in our multicenter study. Thus, their impact on the failure of treatment could not be further explored.

Conclusion

In conclusion, the short-term outcomes of cirrhotic patients with AVB after EVT might not be dependent upon the presence and grade of thrombocytopenia. Thus, it seems that the correction of thrombocytopenia by platelet infusion and use of thrombopoietin before EVT, an invasive procedure, is not warranted.

Footnotes

Appendix

Acknowledgements

We are indebted to the colleagues from 23 participating centers of the V-CAGIB study, who are not listed as the authors of this paper, for their great contributions to establish and verify this prospective international multicenter database, including Zhaohui Bai, Yuhang Yin, Wentao Xu, Sijia Zhang, Xiaotong Li, Guo Lin, Di Sun, Yao Xiao, Xu Gao, Shoujie Zhao, Zhenhua Liu, Qinghe Cheng, Yunxin Wang, Hao Jiang, Yan Feng, Yaning Zhang, Botao Ning, Na Sun, Jinling Dong, Wenming Wu, Jian Zhang, Emine Mutlu, Stephany Castillo-Castañeda, Giuseppe Butera, and Alberto Maringhini.

Author’s note

Number of patients enrolled from 23 participating centers for the present sub-study is listed as follows: (1) Bimin Li: The First Affiliated Hospital of Nanchang University, Nanchang, China (N = 212); (2) Lei Liu: Tangdu Hospital, The Fourth Military Medical University, Xi’an, China (N = 193); (3) Su Lin: The First Affiliated Hospital of Fujian Medical University, Fuzhou, China (N = 97); (4) Fernando Gomes Romeiro: Botucatu Medical School, São Paulo State University, Botucatu, Brazil (N = 82); (5) Xingshun Qi: General Hospital of Northern Theater Command, Shenyang, China (N = 82); (6) Mingyu Sun: Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, Shanghai, China (N = 70); (7) Fanpu Ji: The Second Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China (N = 53); (8) Qiang Zhu: Shandong Provincial Hospital affiliated to Shandong First Medical University, Jinan, China (N = 51); (9) Yingli He: The First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, China (N = 43); (10) Dapeng Ma: The Sixth People’s Hospital of Dalian, Dalian, China (N = 40); (11) Shanshan Yuan: Xi’an Central Hospital, Xi’an, China (N = 40); (12) Xiaofeng Liu: The 960th Hospital of Chinese PLA, Jinan, Shandong, China (N = 29); (13) Cyriac Abby Philips: The Liver Institute, Center of Excellence in GI Sciences, Rajagiri Hospital, Kerala, India (N = 18); (14) Metin Basaranoglu: Bezmialem Vakif University, Istanbul, Turkey (N = 15); (15) Nahum Méndez-Sánchez: Medica Sur Clinic, National Autonomous University of Mexico, Mexico City, Mexico (N = 13); (16) Kanokwan Pinyopornpanish: Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand (N = 12); (17) Yiling Li: The First Affiliated Hospital of China Medical University, Shenyang, China (N = 11); (18) Yunhai Wu: The Sixth People’s Hospital of Shenyang, Shenyang, China (N = 6); (19) Ling Yang: Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China (N = 4); (20) Lichun Shao: Air Force Hospital of Northern Theater Command, Shenyang, China (N = 4); (21) Andrea Mancuso: Azienda di Rilievo Nazionale ad Alta Specializzazione Civico-Di Cristina-Benfratelli, Palermo, Italy (N = 3); (22) Yu Chen: Beijing You’an Hospital Affiliated to Capital Medical University, Beijing, China (N = 1); (23) Frank Tacke: Charité—Universitätsmedizin Berlin, Berlin, Germany (N = 0).

Declarations

ORCID iDs

Yiling Li

Xingshun Qi

References

Ginès

Krag

Abraldes

, et al. Liver cirrhosis. Lancet 2021; 398(10308): 1359–1376.

Stanley

Laine

Management of acute upper gastrointestinal bleeding. BMJ 2019; 364: l536.

Tapper

Parikh

ND.

Diagnosis and management of cirrhosis and its complications: a review. JAMA 2023; 329(18): 1589–1602.

Augustin

Altamirano

González

, et al. Effectiveness of combined pharmacologic and ligation therapy in high-risk patients with acute esophageal variceal bleeding. Am J Gastroenterol 2011; 106(10): 1787–1795.

García-Pagán

Reverter

Abraldes

, et al. Acute variceal bleeding. Semin Respir Crit Care Med 2012; 33(1): 46–54.

de Franchis

Bosch

Garcia-Tsao

, et al. Baveno VII—renewing consensus in portal hypertension. J Hepatol 2022; 76(4): 959–974.

Hernández-Gea

Berbel

Baiges

, et al. Acute variceal bleeding: risk stratification and management (including TIPS). Hepatol Int 2018; 12(Suppl, 1): 81–90.

Rudler

Bureau

Carbonell

, et al. Recalibrated MELD and hepatic encephalopathy are prognostic factors in cirrhotic patients with acute variceal bleeding. Liver Int 2018; 38(3): 469–476.

Saab

Brown

Jr.

Management of thrombocytopenia in patients with chronic liver disease. Dig Dis Sci 2019; 64(10): 2757–2768.

10.

Zanetto

Campello

Senzolo

, et al. The evolving knowledge on primary hemostasis in patients with cirrhosis: a comprehensive review. Hepatology 2024; 79(2): 460–481.

11.

Amitrano

Guardascione

Manguso

, et al. The effectiveness of current acute variceal bleed treatments in unselected cirrhotic patients: refining short-term prognosis and risk factors. Am J Gastroenterol 2012; 107(12): 1872–1878.

12.

Chau

Chan

Patch

, et al. Thrombelastographic changes and early rebleeding in cirrhotic patients with variceal bleeding. Gut 1998; 43(2): 267–271.

13.

Huang

Zhang

Xiang

, et al. Treatment strategies in emergency endoscopy for acute esophageal variceal bleeding (CHESS1905): a nationwide cohort study. Front Med 2022; 9: 872881.

14.

Yin

Romeiro

, et al. Impact of peptic ulcer bleeding on the in-hospital outcomes of cirrhotic patients with acute gastrointestinal bleeding: an international multicenter study. Expert Rev Gastroenterol Hepatol 2024; 18(8): 473–483.

15.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147(8): 573–577.

16.

Han

, et al. Association of coagulopathy with the risk of bleeding after invasive procedures in liver cirrhosis. Saudi J Gastroenterol 2018; 24(4): 220–227.

17.

Reverter

Tandon

Augustin

, et al. A MELD-based model to determine risk of mortality among patients with acute variceal bleeding. Gastroenterology 2014; 146(2): 412–419.e3.

18.

Ferlitsch

Reiberger

Hoke

, et al. von Willebrand factor as new noninvasive predictor of portal hypertension, decompensation and mortality in patients with liver cirrhosis. Hepatology 2012; 56(4): 1439–1447.

19.

Sarangapani

Shanmugam

Kalyanasundaram

, et al. Noninvasive prediction of large esophageal varices in chronic liver disease patients. Saudi J Gastroenterol 2010; 16(1): 38–42.

20.

Abd-Elsalam

Habba

Elkhalawany

, et al. Correlation of platelets count with endoscopic findings in a cohort of Egyptian patients with liver cirrhosis. Medicine 2016; 95(23): e3853.

21.

Ismail

Shah

Hamid

, et al. Noninvasive predictors of large varices in patients hospitalized with gastroesophageal variceal hemorrhage. Hepatol Int 2008; 2(1): 124–128.

22.

Garcia-Tsao

Bosch

Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med 2010; 362(9): 823–832.

23.

Deng

, et al. Association of conventional haemostasis and coagulation tests with the risk of acute upper gastrointestinal bleeding in liver cirrhosis: a retrospective study. Gastroenterol Rep 2016; 4(4): 315–319.

24.

Pfisterer

Schwarz

Jachs

, et al. Endoscopic band ligation is safe despite low platelet count and high INR. Hepatol Int 2023; 17(5): 1205–1214.

25.

, et al. Risk factors for predicting early variceal rebleeding after endoscopic variceal ligation. World J Gastroenterol 2011; 17(28): 3347–3352.

26.

Sun

Zou

, et al. Risk factors for 5-day bleeding after endoscopic treatments for gastroesophageal varices in liver cirrhosis. AME Med J 2017; 2: 39.

27.

O’Leary

Greenberg

Patton

, et al. AGA clinical practice update: coagulation in cirrhosis. Gastroenterology 2019; 157(1): 34–43.e1.

28.

de Oliveira Souza

D’Amico

É A

Flores da Rocha

, et al. Preservation of platelet function in patients with cirrhosis and thrombocytopenia undergoing esophageal variceal ligation. Hepatobiliary Pancreat Dis Int 2020; 19(6): 555–560.

29.

Roberts

Lisman

Stanworth

, et al. Periprocedural management of abnormal coagulation parameters and thrombocytopenia in patients with cirrhosis: guidance from the SSC of the ISTH. J Thromb Haemost 2022; 20(1): 39–47.

30.

Matei

Crisan

Procopet

, et al. Predictive factors of failure to control bleeding and 6-week mortality after variceal hemorrhage in liver cirrhosis—a tertiary referral center experience. Arch Med Sci 2022; 18(1): 52–61.

31.

Bambha

Kim

Pedersen

, et al. Predictors of early re-bleeding and mortality after acute variceal haemorrhage in patients with cirrhosis. Gut 2008; 57(6): 814–820.