Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2022 Dec 22;17(12):e0279496. doi: 10.1371/journal.pone.0279496

Prophylactic antibiotics on patients with cirrhosis and upper gastrointestinal bleeding: A meta-analysis

Yanying Gao 1, Baoxin Qian 1, Xu Zhang 1, Hua Liu 1, Tao Han 2,*
Editor: Sanjiv Mahadeva3
PMCID: PMC9778565  PMID: 36548353

Abstract

Objective

To evaluate the effect of different prophylactic antibiotic treatments for cirrhosis patients with upper gastrointestinal bleeding (UGIB) and to investigate whether prophylactic antibiotics are equally beneficial to reducing the risk of adverse outcomes in A/B with low Child-Pugh scores.

Methods

Relevant studies were searched via PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Internet (CNKI), Wanfang, and VIP databases up to July 16, 2021. The heterogeneity test was conducted for each outcome measuring by I2 statistics. Subgroup analysis was performed regarding antibiotic types. Relative risk (RR) and 95% confidence interval (CI) were used to evaluate prophylactic antibiotics on the risk of adverse outcomes in cirrhosis patients with UGIB.

Results

Twenty-six studies involving 12,440 participants fulfilled our inclusion criteria. Antibiotic prophylaxis was associated with a reduced overall mortality (RR: 0.691, 95%CI: 0.518 to 0.923), mortality due to bacterial infections (RR: 0.329, 95%CI: 0.144 to 0.754), bacterial infections (RR: 0.389, 95%CI: 0.340 to 0.444), rebleeding (RR: 0.577, 95%CI: 0.433 to 0.767) and length of hospitalization [weighted mean difference (WMD): -3.854, 95%CI: -6.165 to -1.543] among patients with UGIB. Nevertheless, prophylactic antibiotics may not benefit to A/B population with low Child-Pugh scores. In our subgroup analysis, quinolone, beta-lactams alone or in combination reduced adverse outcomes in cirrhosis patients with UGIB.

Conclusion

Administration of antibiotics was associated with a reduction in mortality, bacterial infections, rebleeding, and length of hospitalization. Quinolone, beta-lactams alone or in combination can be used in cirrhosis patients with UGIB. Nevertheless, targeted efforts are needed to promote the appropriate use of antibiotics among patients with cirrhosis and UGIB.

Introduction

Cirrhosis is the 10th major cause of death worldwide, which is characterized by an increased portal inflow secondary to splanchnic arterial vasodilatation and high portal outflow resistance, both leading to portal hypertension [1, 2]. Upper gastrointestinal bleeding (UGIB) is a serious complication of cirrhosis [3], which most frequently caused by gastroesophageal varices development (65%-70%), isolated gastric varices (10%-15%), or less frequently portal hypertensive gastropathy [4]. Reports showed that about one-third of cirrhotic patients had gastroesophageal variceal bleeding (GOVB) and around 20% had peptic ulcer bleeding during follow-up [2, 5]. Bacterial infections during or immediately after bleeding episodes are associated with severe complications, which were related to the substantial medical expenses and high mortality [6, 7]. UGIB is one of the leading causes of hospital admission [3]. The survival of patients after an episode of UGIB has recently improved due to the standardization in supportive management and the advances in the treatment of portal pressure and endoscopic techniques [4, 8]. However, the management of UGIB in patients with cirrhosis remains a clinical challenge.

Current clinical guidelines recommend that antibiotic prophylaxis should be instituted for cirrhotic patients with UGIB [9, 10], with the advantages of decreasing not only bacterial infections but also mortality [6, 11, 12]. Despite the positive impact of antibiotic prophylaxis on the outcomes of cirrhotic patients with UGIB, there is scant data concerning the effect of different types of antibiotic prophylaxis on the outcomes of cirrhotic patients. To select the optimal prophylactic antibiotics in cirrhotic patients with UGIB patients, stratification of the antibiotic prophylaxis should be performed. In addition, patients with varying severity of cirrhosis may have disparate levels of risk of complications during UGIB [13, 14]. In patients with Child-Pugh class A cirrhosis and UGIB, a low risk of bacterial infection and mortality has been reported [15]. A balance of the risks and benefits of routine and indistinctive prophylactic antibiotic use in all patients with cirrhosis and UGIB is essential.

Herein, this meta-analysis was to explore the effect of different types of prophylactic antibiotics on cirrhosis patients with UGIB and determine whether antibiotics are equally beneficial in reducing the risk of adverse outcomes across different cirrhotic patients.

Methods

Search strategy

From inception to July 16, 2021, relevant studies were identified by literature in comprehensive databases including PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Internet (CNKI), Wanfang, VIP. Search strategies from PubMed database were as follows: (“Liver Cirrhosis” OR “Hepatic Cirrhosis” OR “Cirrhosis, Hepatic” OR “Cirrhosis, Liver” OR “Fibrosis, Liver” OR “Liver Fibrosis”) AND (“upper gastrointestinal bleeding” OR “Gastrointestinal Hemorrhage” OR “Hemorrhage, Gastrointestinal” OR “Gastrointestinal Hemorrhages” OR “Hematochezia” OR “Hematochezias”) AND (“Anti-Bacterial Agents” OR “Agents, Anti-Bacterial” OR “Anti Bacterial Agents” OR “Antibacterial Agents” OR “Agents, Antibacterial” OR “Antibacterial Agent” OR “Agent, Antibacterial” OR “Anti-Bacterial Compounds” OR “Anti Bacterial Compounds” OR “Compounds, Anti-Bacterial” OR “Anti-Bacterial Agent” OR “Agent, Anti-Bacterial” OR “Anti Bacterial Agent” OR “Anti-Bacterial Compound” OR “Anti Bacterial Compound” OR “Compound, Anti-Bacterial” OR “Bacteriocidal Agents” OR “Agents, Bacteriocidal” OR “Bacteriocidal Agent” OR “Agent, Bacteriocidal” OR “Bacteriocide” OR “Bacteriocides” OR “Anti-Mycobacterial Agents” OR “Agents, Anti-Mycobacterial” OR “Anti Mycobacterial Agents” OR “Anti-Mycobacterial Agent” OR “Agent, Anti-Mycobacterial” OR “Anti Mycobacterial Agent” OR “Antimycobacterial Agent” OR “Agent, Antimycobacterial” OR “Antimycobacterial Agents” OR “Agents, Antimycobacterial” OR “Antibiotics” OR “Antibiotic”).

Inclusion and exclusion criteria

Inclusion criteria were performed according to the PICOS principles: (1) population (P) patients diagnosed with cirrhosis complicated with UGIB; (2) intervention and control (I) and (C): patients treated with prophylactic antibiotics were regarded as the experimental group, and patients without prophylactic antibiotics treatment were served as the control group; (3) outcomes (O): overall mortality, mortality due to bacterial infections/gastrointestinal hemorrhage/liver failure/septic shock/multiple organs failure, bacterial infections, re-bleeding, hospitalization days, length of stay in the intensive care unit (ICU); (4) study design (S): randomized controlled trials (RCTs) and cohort studies; (5) literature published in English and Chinese.

Exclusion criteria were: (1) animal experiments; (2) studies in which data are incomplete or unable to be extracted; (3) conference abstracts, academic dissertations, case reports, letters, editorial materials, meta-analyses, and reviews.

Methodological quality assessment and data extraction

The modified Jadad scale was used to assess the quality of studies [16], using the risk assessment of bias tool from the Cochrane Collaboration’s tool, regarding the following four aspects: 1) methods for generating random series (0–2 points); 2) randomization concealment (0–2 points); 3) blind method (0–2 points); 4) assessment of withdrawal (0–1 points). A final score of 1–3 was regarded as low quality and 4–7 was regarded as high quality.

From all relevant articles, we extracted authors’ names, year of publication, country, intervention, age, gender, assessment of cirrhosis, clinical outcome. Two authors (YYG and BXQ) independently extracted the data from included studies. Any disagreements were resolved by discussion with a third author (TH).

Statistical analysis

Relative risk (RR) and hazard ratio (HR) were used as effect indicators for categorical data. Continuous data were analyzed calculating weighted mean difference (WMD), and the effect size was expressed by 95% confidence interval (CI). The heterogeneity of effects across studies was evaluated by I2 tests for heterogeneity. When the heterogeneity statistic I2≥25%, random effects model analysis was performed, otherwise, fixed effects model analysis was applied. P<0.05 was considered statistically significant. When the difference was statistically significant and I2≥25%, subgroup analysis was performed regarding antibiotic types. Sensitivity analysis was performed for all outcomes. Begg’ test was examined to evaluate the potential for publication bias. When publication bias occurred, the “trim-and-fill method” was adopted to adjust publication bias. Software Stata 15.1 (Stata Corporation, College Station, TX, USA) was used for statistical analysis.

Results

Literature search and characteristics of studies

The flowchart of the literature selection is summarized in Fig 1. Using this searching strategy, 3,690 studies were retrieved, 2,310 studies were left after duplicates removing, of which 47 articles were assessed after titles and abstracts screening. Eventually, 26 [6, 7, 14, 17–39] of these were identified, including 5,783 patients in the experimental group and 6,657 in the control group. The characteristics of each study are shown in Table 1.

Fig 1. The flowchart of the literature selection.

Fig 1

Open in a new tab

Table 1. Baseline characteristics of included studies.

Author Year Country Groups Intervention n Age (years) Male/female Ascites HCC Etiology of cirrhosis Child-Pugh A/B/C Child-Pugh score Clinical outcomes
Soriano 1992 Spain T Oral norfloxacin 800 mg ⁄ day during 7 days 60 61.0 ± 12.6 31/29 15 5 alcoholic:33, other:27 19/30/11 Overall mortality、mortality due to bacterial infections/gastrointestinal hemorrhage/liver failure、Bacterial infections、Hospitalization (days)
C No antibiotic prophylaxis 59 58.3 ± 13.5 30/29 14 2 alcoholic:34, other:25 21/25/13
Rolando 1993 UK T Intravenous imipenem + cilastin,500 mg before and after the sclerotherapy 47 54 (20–76) 30/17 30 NA NA NA Overall mortality、Bacterial infections
C Intravenous dextrosesaline solution 50 46 (18–84) 24/26 35 NA NA NA
Selby 1994 Australia T Intravenous cefotaxime, 1 g immediately before sclerotherapy 19 58.9 ± 14.2 15/4 5 NA Alcohol:11,HBV:2,HCV:3,Other:3 4/8/7 Overall mortality、Bacterial infections
C No antibiotic prophylaxis 20 49.5 ± 10.7 13/7 7 NA Alcohol:12,HBV:1,HCV:4,Other:3 4/10/6
Blaise 1994 France T Intravenous + oral ofloxacin,400 mg ⁄ day, 10 days;amoxicillin + clavulanic acid(bolus, 1 g) before each endoscopy procedure 46 52 ± 11 33/13 22 NA alcoholic:41, other:5 0/11/35 Overall mortality、mortality due to gastrointestinal hemorrhage/septic shock、Bacterial infections、Recurrence of hemorrhage、Length of stay in ICU (d)
C No antibiotic prophylaxis 45 54 ± 9 36/9 17 NA alcoholic:39, other:6 0/9/36
Pauwels 1996 France T Intravenous + oral ciprofloxacin 400 mg ⁄ day, amoxicillin-clavulanic acid 3 g ⁄ day, until 3 days after cessation of haemorrhage 30 51 ± 2 14/16 16 NA alcoholic:27, other:3 2/3/25 Overall mortality、mortality due to septic shock/gastrointestinal hemorrhage/liver failure、Bacterial infections、Recurrence of hemorrhage、Length of stay in ICU (d)
C No antibiotic prophylaxis 34 53 ± 3 24/10 13 NA alcoholic:21, other:13 0/10/24
Zacharof 1997 Greece T Oral ciprofloxacin 500 mg ⁄ day during 7 days 35 NA NA NA NA NA NA Overall mortality、Bacterial infections
C No antibiotic prophylaxis 30 NA NA NA NA NA NA
Hsieh 1998 Taiwan T Oral ciprofloxacin, 1 g ⁄ day, 7 days 60 58 ± 14 47/13 28 29 HBV/HCV:53,alcoholic:6,other:1 5/33/22 Overall mortality、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C Placebo 60 62 ± 13 42/18 30 19 HBV/HCV:48,coholic:9,other:3 6/31/23
Hou 2004 Taiwan T Intravenous ofloxacin 200 mg b.d. for 2 days followed by oral ofloxacin 200 mg b.d. for 5 days 59 60.02 ± 13.92 43/16 33 16 Viral/alcohol/mixed/others: 29/6/10/14 10/35/14 8.54 ± 1.90 Overall mortality、mortality due to bacterial infections/gastrointestinal hemorrhage/liver failure/multiple organs failure、Bacterial infections、Recurrence of hemorrhage
C No antibiotic prophylaxis 61 59.39 ± 14.85 48/13 29 14 Viral/alcohol/mixed/others:34/10/10/7 19/29/13 7.90 ± 2.04
Jun 2006 Korea T Intravenous cefotaxime 2 g t.d.s. for 7 days 62 54.7 ± 10.1 54/8 34 16 Esophageal/gastric varices:52/11 8.7 ± 1.9 Overall mortality、mortality due to bacterial infections/gastrointestinal hemorrhage/liver failure/multiple organs failure、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C No antibiotic prophylaxis 58 54.2 ± 11.9 56/2 33 10 Esophageal/gastric varices:50/8 8.3 ± 2.1
Tandon 2015 Canada T Antibiotic prophylaxis 206 55.0 ± 11.7 154/52 NA 14 Alcohol or HCV:121 16/97/93 9.4 ± 2.1 Overall mortality、Bacterial infections、Recurrence of hemorrhage
C No antibiotic prophylaxis 175 56.2 ± 13.4 114/61 NA 11 Alcohol or HCV:92 30/94/51 8.5 ± 2.0
Moon 2016 USA T 3rd generation cephalosporins 54.5%, Fluoroquinolones 31.9%, Penicillins 2.9%, Aminoglycosides 0.8%, other 9.9% 4210 60.0 ± 8.5 4100/110 1003 405 HCV:2103,Alcohol:1597,HBV:71,Non-alcoholic fatty liver disease:354,Other:85 NA Overall mortality
C No antibiotic prophylaxis 4445 61.2 ± 9.6 4333/112 459 290 HCV:1895,Alcohol:1790,HBV:72,Non-alcoholic fatty liver disease:545,Other:143 NA
Chang 2020 Taiwan T 19 received cefazolin, 4 received cefazolin plus gentamicin, 6 received cefuroxime, and 44 received ceftriaxone 73 55.22 ± 12.46 57/16 33 19 HBV:22.HCV:28,BC:8,NBNC:15 18/55/0 7.34 ± 1.20 Overall mortality、Bacterial infections、Recurrence of hemorrhage
C No antibiotic prophylaxis 840 59.86 ± 13.15 592/248 344 287 HBV:173.HCV:404,BC:77,NBNC:186 239/601/0 7.32 ± 1.21
Qin 2002 China T Intravenous infusion of ofloxacin (200mg, twice a day) for 7 consecutive days 50 51 ± 12 42/8 40 NA hepatitis:38, alcoholic:8, other:4 1/9/40 Overall mortality、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C No use of antibiotic prophylaxis 50 53 ± 10 43/7 42 NA hepatitis:40, alcoholic:7, other:3 2/10/38
Wu 2002 China T On the day of admission, ciprofloxacin or piperacillin was given intravenously, 0.4 g of ciprofloxacin or piperacillin 6–8 g was instilled twice a day, until 3 days after the bleeding stopped 54 47.6 ± 21.9 44/10 47 NA HBV NA 9.3 ± 2.7 Overall mortality、mortality due to bacterial infections、Bacterial infections
C No antibiotics were given after admission 48 45.1 ± 26.3 38/10 41 NA HBV NA 9.1 ± 2.2
Ma 2007 China T Ceftriaxone antibiotics were given immediately after admission, 2g once / 12h, until 3 days after the bleeding stopped 21 40.4 ± 17.4 33/9 21 NA NA NA Bacterial infections
C Conventional treatment 21 21 NA NA NA
Luo 2008 China T Quinolones or cephalosporin third-generation antibiotics were given for 1 week after admission 60 40.3 (21–73) 68/52 NA NA Alcohol:4,HBV:82,HCV:34 B/C Overall mortality、Bacterial infections
C No antibiotics 60 NA NA
Wang 2015 China T1 Anti-infection treatment with ceftriaxone sodium after admission 60 42–69 42/18 NA NA hepatitis:48, alcoholic:11, other:1 Overall mortality、Bacterial infections、Recurrence of hemorrhage
T2 Levofloxacin was used for anti-infection treatment for 7 days 60 40/20 NA NA hepatitis:45, alcoholic:14, other:1
C Antibiotics were not given on admission 60 38/22 NA NA hepatitis:47, alcoholic:13, other:0
Liu 2015 China T After admission, antibiotics were used on the basis of conventional treatment to prevent infection, ceftriaxone sodium was injected intravenously 1g, 1 time/d; for those who are allergic to it (2 cases in total), levofloxacin intravenously injected 0.4g, 1 time/d; prevention Infection treatment 7days 210 NA 160/50 NA NA HBV:147,HCV:25,Alcohol:14,autoimmune hepatitis:11,other:3 123/66/21 Bacterial infections、Recurrence of hemorrhage
C No antibiotics 173 NA 134/39 NA NA HBV:138,HCV:12,Alcohol:5,autoimmune hepatitis:2,other:16 115/38/20
Zheng 2017 China T After admission, cefotaxime sodium 2.0g was given intravenously, twice a day, for 7 days 60 56.5±12.4 51/9 NA NA hepatitis:45, alcoholic:14, other:1 44/13/3 Overall mortality、Bacterial infections、Recurrence of hemorrhage
C No antibiotics were administered on admission 60 52/8 NA NA hepatitis:47, alcoholic:13, other:0 45/13/2
Na 2017 China T Add human cefotaxime sodium 2.0g intravenously to the treatment plan of the control group, bid, 3 days after the bleeding stops 22 45.5 ±2.2 10/12 NA NA hepatitis:12, alcoholic:8, other:2 Overall mortality、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C Routine treatment was given first, pituitrin or octreotide were given, respectively, and intravenous drops were dropped until the bleeding stopped for 4 days. At the same time, auxiliary therapy of blood transfusion, electrocardiogram, hemostasis and oxygen inhalation was performed 28 47.5 ± 2.3 12/16 NA NA hepatitis:15, alcoholic:10, other:3
Zhang 2018 China T After admission, on the basis of the control group, intravenous infusion of ceftriaxone disodium, 1g/ time, once a day, for cephalosporin allergy, intravenous infusion of levofloxacin, 0.4g/ time, once a day, continuous application of antibiotics for 1 week 52 45. 19 ± 5. 34 33/19 NA NA hepatitis:50, alcoholic:2 Overall mortality、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C Conventional treatment 52 35/17 NA NA hepatitis:49, alcoholic:3
Wang 2014 China T Prophylactic use of antibiotics (piperacillin or ciprofloxacin) in addition to conventional treatment, piperacillin 5.0–9.0g/d, twice daily iv or ciprofloxacin 0.5g/d, twice daily IV, 3 days as a course of treatment 60 49±3.6 36/24 NA NA hepatitis:38, alcoholic:19, schistosomal cirrhosis:2,other:1 Overall mortality、Bacterial infections、Recurrence of hemorrhage、Hospitalization (days)
C Conventional treatment 60 31/29 NA NA hepatitis:38, alcoholic:20, schistosomal cirrhosis:1,other:1
Dong 2015 China T On the basis of conventional treatment, antibiotics were given to patients after blood, urine and ascites culture specimens were collected (levofloxacin 0.4g intravenously, once per day, or cefotaxime sodium 4-6g intravenously, once per 12h, 5 days as a course 49 NA NA NA NA NA C Overall mortality、Bacterial infections、Hospitalization (days)
C Conventional treatment 50 NA NA NA NA NA
Peng 2017 China T On the basis of conventional treatment, intravenous infusion of cephalosporin iii antibiotics or quinolones antibiotics was performed according to patients’ allergic constitution 30 40.49 ± 6.32 18/12 NA NA HBV:19,HCV:8,Alcohol:3 Overall mortality、Bacterial infections
C Routine treatments such as blood transfusion, hemostasis and liver protection were given 30 41.25 ± 7.15 17/13 NA NA HBV:20,HCV:8,Alcohol:2
Qin 2020 China T Prophylactic antibiotic therapy was given immediately after admission: norfloxacin 400mg orally, once a day; Ceftriaxone sodium for injection 1g was given intravenously, once a day; The third generation cephalosporin was selected for treatment for 5–7 days 45 54.76±8.69 26/19 NA NA NA 19/17/9 Overall mortality、Bacterial infections、Recurrence of hemorrhage
C Conventional treatment 45 56.31±8.83 29/16 NA NA NA 16/14/15
Xu 2021 China T On the basis of conventional treatment, patients were given cefoperazone sulbactam 3.0g intravenous infusion, twice a day; The allergic patients were treated with 0.4g levofloxacin hydrochloride injection 3.0g intravenously, twice a day 43 42.52±9.83 23/20 NA NA NA Overall mortality、Bacterial infections、Recurrence of hemorrhage
C Conventional treatment 43 42.55±10.15 22/21 NA NA NA
Open in a new tab

Notes: T: patients treated with prophylactic antibiotics; C: patients not treated with prophylactic antibiotics; NA represents that data is missing; MELD: model for end-stage liver disease; HCC: hepatocellular carcinoma; HBV: hepatitis B virus; HCV: hepatitis C virus; BC: presence of both HBV and HCV; NBNC: negative for both HBV and HCV.

Overall mortality

A total of 24 studies were assessed to investigate overall mortality between the experimental and control group. The heterogeneity test showed that the difference was statistically significant (I2 = 71.4%), so the random effects model was used. The result showed that antibiotic prophylaxis was associated with a significant reduction in overall mortality (RR: 0.691, 95%CI: 0.518 to 0.923, P = 0.012) (Table 2, Fig 2A). Subgroup analysis was applied based on antibiotic types. Combined quinolone and cephalosporins reduced the overall mortality of in cirrhosis patients with UGIB. Decreased overall mortality was also observed in cirrhosis patients with UGIB when quinolones were combined with beta-lactams (Fig 2B).

Table 2. Overall results and sensitivity analysis.

Indicators RR/WMD (95%CI) P I2
Overall mortality
Overall 0.691 (0.518, 0.923) 0.012 71.4
Sensitivity analysis 0.691 (0.518, 0.923)
Antibiotics
Quinolones 0.827 (0.608, 1.125) 0.227 0.0
Cephalosporins 0.663 (0.341, 1.289) 0.226 51.4
Quinolones+cephalosporins 0.406 (0.254, 0.650) <0.001 0.0
Quinolones+beta-lactams 0.553 (0.374, 0.816) 0.003 0.0
Others 1.173 (0.519, 2.650) 0.701 80.7
Unclear 1.467 (0.904, 2.383) 0.121 NA
Mortality due to bacterial infections
Overall 0.329 (0.144, 0.754) 0.009 12.7
Sensitivity analysis 0.329 (0.144, 0.754)
Mortality due to gastrointestinal hemorrhage
Overall 0.942 (0.524, 1.694) 0.842 0.0
Sensitivity analysis 0.942 (0.524, 1.694)
Mortality due to liver failure
Overall 0.811 (0.359, 1.832) 0.615 31.2
Sensitivity analysis 0.811 (0.359, 1.832)
Mortality due to septic shock
Overall 0.292 (0.050, 1.704) 0.171 0.0
Sensitivity analysis 0.292 (0.050, 1.704)
Mortality due to multiple organs failure
Overall 1.084 (0.436, 2.693) 0.863 0.0
Sensitivity analysis 1.084 (0.436, 2.693)
Bacterial infections
Overall 0.389 (0.340, 0.444) <0.001 16.4
Sensitivity analysis 0.389 (0.340, 0.444)
Publication bias Z = 1.98 0.047
Rebleeding
Overall 0.577 (0.433, 0.767) <0.001 64.5
Sensitivity analysis 0.577 (0.433, 0.767)
Publication bias Z = 1.07 0.284
Antibiotics
Quinolones 0.521 (0.353, 0.758) 0.001 10.8
Cephalosporins 0.517 (0.380, 0.703) <0.001 0.0
Quinolones+cephalosporins 0.480 (0.343, 0.671) <0.001 0.0
Quinolones+ beta-lactams 0.764 (0.264, 2.210) 0.619 82.2
Unclear 1.732 (1.133, 2.646) 0.011 NA
Hospitalization (days)
Overall -3.854 (-6.165, -1.543) 0.001 82.8
Sensitivity analysis -3.854 (-6.165, -1.543)
Antibiotics
Quinolones -4.115 (-8.202, -0.027) 0.049 52.5
Cephalosporins -1.083 (-3.354, 1.188) 0.350 0.0
Quinolones+cephalosporins -4.999 (-11.849, 1.851) 0.153 95.9
Quinolones+beta-lactams -5.500 (-6.881, -4.119) <0.001 NA
Length of stay in ICU (days)
Overall -0.272 (-1.545, 1.002) 0.676 89.3
Sensitivity analysis -0.272 (-1.545, 1.002)
Open in a new tab

Notes: RR: relative risk; WMD: weighted mean difference.

Fig 2. The forest plot of comparison on the overall mortality in experimental group and control group.

Fig 2

Open in a new tab

(a) overall; (b) subgroup.

Mortality due to bacterial infections

Four articles were used to evaluate mortality due to bacterial infections. The result showed a benefit of antibiotic prophylaxis over no antibiotic prophylaxis intervention regarding mortality due to bacterial infections (RR: 0.329, 95%CI: 0.144 to 0.754, P = 0.009) (Table 2, Fig 3).

Fig 3. The forest plot of comparison on the mortality due to bacterial infections in experimental group and control group.

Fig 3

Open in a new tab

Mortality due to gastrointestinal hemorrhage

Mortality due to gastrointestinal hemorrhage was assessed in 5 studies. The result could not prove a beneficial effect of antibiotic prophylaxis over mortality from bacterial infections (RR: 0.942, 95%CI: 0.524 to 1.694, P = 0.842) (Table 2).

Mortality due to liver failure

Four studies investigated mortality due to liver failure. The heterogeneity test showed I2 = 31.2%, so the random effects model was applied. There was no significant difference in mortality due to liver failure between the experimental and control group (RR: 0.811, 95%CI: 0.359 to 1.832, P = 0.615) (Table 2).

Mortality due to septic shock

Two studies were included to assess mortality due to septic shock between experimental and control group. No significant difference between experimental and control group was observed concerning mortality due to septic shock (RR: 0.292, 95%CI: 0.050 to 1.704, P = 0.171) (Table 2).

Mortality due to multiple organs failure

Mortality due to multiple organs failure was evaluated in 2 studies. No difference in mortality from multiple organ failure was observed between the antibiotic group and the control group (RR: 1.084, 95%CI: 0.436 to 2.693, P = 0.863) (Table 2).

Bacterial infections

Totally, 25 studies were included to evaluate the bacterial infections between 2 groups. Fixed effect model results showed that the bacterial infection rate was lower in the antibiotic prophylaxis group than in the control group (RR: 0.389, 95%CI: 0.340 to 0.444, P<0.001) (Table 2, Fig 4).

Fig 4. The forest plot of comparison on the bacterial infections in experimental group and control group.

Fig 4

Open in a new tab

Rebleeding

Rebleeding was investigated in 16 studies. The result reported that a decreased rate of rebleeding in the antibiotic prophylaxis group (RR: 0.577, 95%CI: 0.433 to 0.767, P<0.001) (Table 2, Fig 5A). Subgroup analysis was applied concerning antibiotic types. Subgroup analysis result showed that quinolone (RR: 0.521, 95%CI: 0.353 to 0.758, P = 0.001), cephalosporins (RR: 0.517, 95%CI: 0.380 to 0.703, P<0.001), quinolone in combination with cephalosporins (RR: 0.480, 95%CI: 0.343 to 0.671, P<0.001) reduced the rate of rebleeding (Fig 5B).

Fig 5. The forest plot of comparison on the rebleeding in experimental group and control group.

Fig 5

Open in a new tab

(a) overall; (b) subgroup.

Duration of treatment

Hospitalization

Eight studies compared hospitalization between the antibiotic group and the control group. The use of antibiotic prophylaxis reduced hospitalization in cirrhosis patients with UGIB (WMD: -3.854, 95%CI: -6.165 to -1.543, P = 0.001) (Table 2, Fig 6A). Subgroup analysis result demonstrated that the use of quinolones (WMD: -4.115, 95%CI: -8.202 to -0.027, P = 0.049), quinolone combined with beta-lactams were associated with shorter hospital stays (Fig 6B).

Fig 6. The forest plot of comparison on the hospitalization in experimental group and control group.

Fig 6

Open in a new tab

(a) overall; (b) subgroup.

Length of stay in ICU

Two articles were used to compare the length of stay in ICU between 2 groups. The random effect model result showed that there was no difference in length of stay in ICU between the antibiotic group and the control group (WMD: -0.272, 95%CI: -1.545 to 1.002, P = 0.676) (Table 2).

Child-Pugh’s score

Child-Pugh A/B

The effect of Child-Pugh A/B on antibiotic use was assessed in 2 studies. The result showed that antibiotic use in the Child-Pugh A/B patients had no significant effect on rates of mortality (RR: 1.819, 95%CI: 0.886 to 3.733, P = 0.103) and bacterial infection (RR: 0.826, 95%CI: 0.274 to 2.489, P = 0.735). The effect of prophylactic antibiotics for Child-Pugh A/B population is shown in Table 3.

Table 3. Prophylactic antibiotics for Child-Pugh A/B population.
Indicators RR (95%CI) P I2
Overall mortality
Overall 1.819 (0.886–3.733) 0.103 0.0
Sensitivity analysis 1.819(0.886–3.733)
Bacterial infections
Overall 0.826 (0.274–2.489) 0.735 65.9
Sensitivity analysis 0.826 (0.274–2.489)
Open in a new tab

Sensitivity analysis and publication bias

The sensitivity analysis for ensemble RR of each model was carried out. Sensitivity analysis result proofs that the findings are trustworthy (Table 2). We examined the publication bias of the included studies. The result showed that publication bias existed in bacterial infections. We used the “cut-and-fill method” to correct publication bias and adjust effect size. The combined RR of the fixed effects model prevalence merger was 0.389 (95%CI: 0.340 to 0.444) before the “trim-and-fill method”. It then showed the estimated number of missing studies was 8. Then, the missing studies were included in the model, and all the studies were re-meta-analyzed. The combined prevalence in the fixed effect model was 0.470 (95%CI: 0.345 to 0.595) (Fig 7).

Fig 7. Publication bias.

Fig 7

Open in a new tab

(a); unadjusted (b) adjusted.

Discussion

The bleeding events in cirrhosis patients were related to substantial medical expenses and high mortality [2]. The prophylactic use of oral or intravenous antibiotics has been recommended in several consensus guidelines [2, 40]. Nevertheless, studies to clarify the role of different types of antibiotic prophylaxis in cirrhosis patients with UGIB are scarce. And whether antibiotic prophylaxis is beneficial for all cirrhotic patients with UGIB is also unclear. The result of the current study proved the benefit of antibiotic prophylaxis for reductions in overall mortality and mortality due to bacterial infections among patients with UGIB. Besides, antibiotic prophylaxis was associated with reduced bacterial infections, rebleeding, and length of hospitalization. Nevertheless, prophylactic antibiotics may not benefit to A/B population with a low Child-Pugh score. In our subgroup analysis, quinolone, beta-lactams alone or in combination quinolone reduced the rate of rebleeding and shortened the length of hospital stay in cirrhosis patients with UGIB.

The benefits of antibiotic prophylaxis in cirrhosis patients with UGIB were observed in this study. Chavez-Tapia et al. [40] demonstrated that antibiotics were associated with a substantially lower 30-day all-cause mortality after adjusting for relevant confounders. Wu et al. [2] suggested that cirrhotic patients without major complications who suffered from UGIB benefited from the use of antibiotics to prevent rebleeding. A study showed that antibiotic prophylaxis reduced rates of bacterial infection, rebleeding, and mortality [41]. In addition, we found that quinolones could reduce the rate of rebleeding and shorten the length of hospital stay in cirrhosis patients with UGIB. However, some studies have presented evidence suggesting that oral quinolone administration may not be the best regime for the prevention of bacterial infections in cirrhotic patients with gastrointestinal hemorrhage. The prevalence of quinolone-resistant bacteria in the fecal flora [42, 43] and the incidence of spontaneous bacterial peritonitis (SBP) [44], and other infections [45] caused by these organisms have increased substantially. In this case, other classes of antibiotics are needed. Since the discovery of benzylpenicillin in the 1920s, thousands of new penicillin derivatives and related beta-lactam classes of cephalosporins, cephamycins, monobactams, and carbapenems have been discovered. The bactericidal mechanism of killing by beta-lactams is perceived to be a major advantage in the treatment of serious infections. When these agents were threatened by the rapid emergence of beta-lactamases, beta-lactamase-stable agents were developed, as well as potent beta-lactamase inhibitors (BLIs) [46]. Beta-lactams are vital antibiotics used to treat infection in patients with cirrhosis due to their spectrum of antibacterial activity and overall safety [47]. In those with cirrhosis, 3rd generation cephalosporins (TGC) has remained for decades the standard treatment of SBP [48]. Nevertheless, the spread of multiple drug resistance (MDR) bacterial infections reduces the effectiveness of commonly used antibiotics such as third-generation cephalosporins [49]. In addition, there is currently limited therapeutic use of penicillin as monotherapy. Ampicillin, amoxicillin, piperacillin, and ticarcillin have continued to be useful, primarily as a result of their combination with an appropriate beta-lactamase inhibitor [46]. This may indicate the necessity of combined antibiotics. We found quinolone in combination with beta-lactams could be associated with reduced mortality, rebleeding rate, and hospitalization.

It is well established that the increasing Child-Pugh class is a strong predictor of bacterial infection in patients with cirrhosis [14, 50]. Excessive prescription of antibiotics has several demerits, such as multidrug resistance of bacteria, Clostridioides difficile infection, and increased cost [51, 52]. To avoid antibiotic overuse, some authorities have proposed a risk-stratified approach. Tandon et al. [14] suggested antibiotic prophylaxis seems unnecessary in patients with good liver function (Child-Pugh A) because they have low rates of bacterial infection and mortality even without antibiotic prophylaxis. A study [18] by Hou et al. showed that antibiotic prophylaxis effectively prevented infection and rebleeding among the most advanced cirrhosis (Child-Pugh B or C). Given this fact, we believe that the patients who will benefit from antibiotic prophylaxis should be identified, and antibiotics need not be prescribed for all bleeding cirrhotic patients. The recommendation for routine antibiotic prophylaxis in this subgroup merits expert re-evaluation and ideally confirmation in large prospective multicenter RCTs.

Potential confounding factors may affect the result of this meta-analysis. Endoscopy is the primary diagnostic and therapeutic tool for UGIB [53]. Cirrhotic patients undergoing an endoscopic procedure to control bleeding are particularly at risk for infection [6, 54]. However, a study demonstrated that the use of therapeutic endoscopy has been shown to improve better prognosis in patients who present with severe acute UGIB [55]. As there have been several exciting recent advances in the endoscopic management of UGIB [56], future studies exploring prophylactic antibiotics in cirrhosis with UGIB should consider the role of therapeutic endoscopy.

To the best of our knowledge, this study is the first meta-analysis focusing on different prophylactic antibiotics on cirrhosis patients complicated with UGIB. A comprehensive analysis of different types of mortality was performed in this study. We attempted to investigate the application of prophylactic antibiotics to low Child-Pugh scores. However, some limitations of our study must be acknowledged. Firstly, even though we adjusted for publication bias using the “trim-and-fill method, certain biases could exist, and must be cautious in extrapolating the results. Secondly, at present, there are few studies on the dose-response relationship of prophylactic antibiotics, so we did not conduct further analysis on the dose of prophylactic antibiotics. Thirdly, due to the limitations of the included study, we were unable to further classify patients with different grades of cirrhosis or assess the effect of prophylactic antibiotic use on patients with Child-Pugh’s C cirrhosis. Finally, there were too small included studies in each part. Further RCTs investigating the effect of prophylactic antibiotics on the risk of adverse outcomes in cirrhosis patients complicated with UGIB are needed.

Conclusions

This study demonstrated that prophylactic antibiotics, especially quinolones and β -lactam alone or in combination were beneficial to cirrhosis patients with UGIB. Nevertheless, the use of antibiotic prophylaxis in Child-Pugh A/B patients with cirrhosis merits re-evaluation.

Supporting information

S1 Checklist. PRISMA 2020 checklist.

(DOCX)

Click here for additional data file. (31.8KB, docx)

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

The author(s) received no specific funding for this work.

References

  • 1.Li Y, Han B, Li H, Song T, Bao W, Wang R, et al. Effect of Admission Time on the Outcomes of Liver Cirrhosis with Acute Upper Gastrointestinal Bleeding: Regular Hours versus Off-Hours Admission. Can J Gastroenterol Hepatol. 2018;2018:3541365. Epub 2019/01/12. doi: 10.1155/2018/3541365 ; PubMed Central PMCID: PMC6304553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Wu CK, Yang SC, Liang CM, Li YC, Yeh WS, Tai WC, et al. The role of antibiotics in upper gastrointestinal bleeding among cirrhotic patients without major complications after endoscopic hemostasis. J Gastroenterol Hepatol. 2020;35(5):777–87. Epub 2019/11/02. doi: 10.1111/jgh.14873 . [DOI] [PubMed] [Google Scholar]
  • 3.Kapsoritakis AN, Ntounas EA, Makrigiannis EA, Ntouna EA, Lotis VD, Psychos AK, et al. Acute upper gastrointestinal bleeding in central Greece: the role of clinical and endoscopic variables in bleeding outcome. Dig Dis Sci. 2009;54(2):333–41. Epub 2008/07/12. doi: 10.1007/s10620-008-0364-1 . [DOI] [PubMed] [Google Scholar]
  • 4.Fouad TR, Abdelsameea E, Abdel-Razek W, Attia A, Mohamed A, Metwally K, et al. Upper gastrointestinal bleeding in Egyptian patients with cirrhosis: Post-therapeutic outcome and prognostic indicators. J Gastroenterol Hepatol. 2019;34(9):1604–10. Epub 2019/04/03. doi: 10.1111/jgh.14659 . [DOI] [PubMed] [Google Scholar]
  • 5.Yang SC, Wu KL, Wang JH, Lee CH, Kuo YH, Tai WC, et al. The effect of systemic antibiotic prophylaxis for cirrhotic patients with peptic ulcer bleeding after endoscopic interventions. Hepatol Int. 2013;7(1):257–67. Epub 2013/03/01. doi: 10.1007/s12072-012-9378-z . [DOI] [PubMed] [Google Scholar]
  • 6.Moon AM, Dominitz JA, Ioannou GN, Lowy E, Beste LA. Use of Antibiotics Among Patients With Cirrhosis and Upper Gastrointestinal Bleeding Is Associated With Reduced Mortality. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association. 2016;14(11):1629–37.e1. Epub 2016/10/19. doi: 10.1016/j.cgh.2016.05.040 . [DOI] [PubMed] [Google Scholar]
  • 7.Chang TS, Tsai YH, Lin YH, Chen CH, Lu CK, Huang WS, et al. Limited effects of antibiotic prophylaxis in patients with Child-Pugh class A/B cirrhosis and upper gastrointestinal bleeding. PloS one. 2020;15(2):e0229101. Epub 2020/02/23. doi: 10.1371/journal.pone.0229101 ; PubMed Central PMCID: PMC7034903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chalasani N, Kahi C, Francois F, Pinto A, Marathe A, Bini EJ, et al. Improved patient survival after acute variceal bleeding: a multicenter, cohort study. The American journal of gastroenterology. 2003;98(3):653–9. Epub 2003/03/26. doi: 10.1111/j.1572-0241.2003.07294.x . [DOI] [PubMed] [Google Scholar]
  • 9.Karstensen JG, Ebigbo A, Aabakken L, Dinis-Ribeiro M, Gralnek I, Le Moine O, et al. Nonvariceal upper gastrointestinal hemorrhage: European Society of Gastrointestinal Endoscopy (ESGE) Cascade Guideline. Endoscopy international open. 2018;6(10):E1256–e63. Epub 2018/10/12. doi: 10.1055/a-0677-2084 ; PubMed Central PMCID: PMC6175685. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.de Franchis R. Expanding consensus in portal hypertension: Report of the Baveno VI Consensus Workshop: Stratifying risk and individualizing care for portal hypertension. Journal of hepatology. 2015;63(3):743–52. Epub 2015/06/07. doi: 10.1016/j.jhep.2015.05.022 . [DOI] [PubMed] [Google Scholar]
  • 11.Garcia-Tsao G, Bosch J. Varices and Variceal Hemorrhage in Cirrhosis: A New View of an Old Problem. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association. 2015;13(12):2109–17. Epub 2015/07/21. doi: 10.1016/j.cgh.2015.07.012 ; PubMed Central PMCID: PMC4851858. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Fernández J, Ruiz del Arbol L, Gómez C, Durandez R, Serradilla R, Guarner C, et al. Norfloxacin vs ceftriaxone in the prophylaxis of infections in patients with advanced cirrhosis and hemorrhage. Gastroenterology. 2006;131(4):1049–56; quiz 285. Epub 2006/10/13. doi: 10.1053/j.gastro.2006.07.010 . [DOI] [PubMed] [Google Scholar]
  • 13.Conejo I, Augustin S, Pons M, Ventura-Cots M, González A, Esteban R, et al. Alcohol consumption and risk of infection after a variceal bleeding in low-risk patients. Liver Int. 2016;36(7):994–1001. Epub 2015/12/09. doi: 10.1111/liv.13038 . [DOI] [PubMed] [Google Scholar]
  • 14.Tandon P, Abraldes JG, Keough A, Bastiampillai R, Jayakumar S, Carbonneau M, et al. Risk of Bacterial Infection in Patients With Cirrhosis and Acute Variceal Hemorrhage, Based on Child-Pugh Class, and Effects of Antibiotics. Clinical gastroenterology and hepatology: the official clinical practice journal of the American Gastroenterological Association. 2015;13(6):1189–96.e2. Epub 2014/12/03. doi: 10.1016/j.cgh.2014.11.019 . [DOI] [PubMed] [Google Scholar]
  • 15.Kamath PS, Mookerjee RP. Individualized care for portal hypertension: Not quite yet. Journal of hepatology. 2015;63(3):543–5. Epub 2015/07/08. doi: 10.1016/j.jhep.2015.07.001 . [DOI] [PubMed] [Google Scholar]
  • 16.Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled clinical trials. 1996;17(1):1–12. Epub 1996/02/01. doi: 10.1016/0197-2456(95)00134-4 . [DOI] [PubMed] [Google Scholar]
  • 17.Zhou X, Tripathi D, Song T, Shao L, Han B, Zhu J, et al. Terlipressin for the treatment of acute variceal bleeding: A systematic review and meta-analysis of randomized controlled trials. Medicine (Baltimore). 2018;97(48):e13437. Epub 2018/12/05. doi: 10.1097/MD.0000000000013437 ; PubMed Central PMCID: PMC6283114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Hou MC, Lin HC, Liu TT, Kuo BI, Lee FY, Chang FY, et al. Antibiotic prophylaxis after endoscopic therapy prevents rebleeding in acute variceal hemorrhage: a randomized trial. Hepatology (Baltimore, Md). 2004;39(3):746–53. Epub 2004/03/05. doi: 10.1002/hep.20126 . [DOI] [PubMed] [Google Scholar]
  • 19.Hsieh WJ, Lin HC, Hwang SJ, Hou MC, Lee FY, Chang FY, et al. The effect of ciprofloxacin in the prevention of bacterial infection in patients with cirrhosis after upper gastrointestinal bleeding. The American journal of gastroenterology. 1998;93(6):962–6. Epub 1998/07/01. doi: 10.1111/j.1572-0241.1998.00288.x . [DOI] [PubMed] [Google Scholar]
  • 20.Jun CH, Park CH, Lee WS, Joo YE, Kim HS, Choi SK, et al. Antibiotic prophylaxis using third generation cephalosporins can reduce the risk of early rebleeding in the first acute gastroesophageal variceal hemorrhage: a prospective randomized study. Journal of Korean medical science. 2006;21(5):883–90. Epub 2006/10/18. doi: 10.3346/jkms.2006.21.5.883 ; PubMed Central PMCID: PMC2722000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Pauwels A, Mostefa-Kara N, Debenes B, Degoutte E, Lévy VG. Systemic antibiotic prophylaxis after gastrointestinal hemorrhage in cirrhotic patients with a high risk of infection. Hepatology (Baltimore, Md). 1996;24(4):802–6. Epub 1996/10/01. doi: 10.1002/hep.510240408 . [DOI] [PubMed] [Google Scholar]
  • 22.Rolando N, Gimson A, Philpott-Howard J, Sahathevan M, Casewell M, Fagan E, et al. Infectious sequelae after endoscopic sclerotherapy of oesophageal varices: role of antibiotic prophylaxis. Journal of hepatology. 1993;18(3):290–4. Epub 1993/07/01. doi: 10.1016/s0168-8278(05)80272-7 . [DOI] [PubMed] [Google Scholar]
  • 23.Selby WS, Norton ID, Pokorny CS, Benn RA. Bacteremia and bacterascites after endoscopic sclerotherapy for bleeding esophageal varices and prevention by intravenous cefotaxime: a randomized trial. Gastrointestinal endoscopy. 1994;40(6):680–4. Epub 1994/11/01. . [PubMed] [Google Scholar]
  • 24.Soriano G, Guarner C, Tomás A, Villanueva C, Torras X, González D, et al. Norfloxacin prevents bacterial infection in cirrhotics with gastrointestinal hemorrhage. Gastroenterology. 1992;103(4):1267–72. Epub 1992/10/01. doi: 10.1016/0016-5085(92)91514-5 . [DOI] [PubMed] [Google Scholar]
  • 25.Zacharof AKP, C. ; Soutos D.; Katsaros D.; Zacharof H.; Poulikakos J. Bacterial infection is prevented by ciprofloxacin in cirrhotics with gastrointestinal hemorrhage P538. Gut. 1997;41:189A. [Google Scholar]
  • 26.Juan-Hong Zheng JJ, Bing-Qu Yu, Jia-Yuan Chen. Efficacy of prophylactic antibiotics in cirrhosis patients with gastroesophageal variceal bleeding. China Journal of Modern Medicine. 2017;27:96–9. [Google Scholar]
  • 27.Na L. Study on the value of antibiotics in preventing nosocomial infection in patients with upper gastrointestinal hemorrhage of cirrhosis. Modem Digestion & Intervention. 2017;22(1):46–7. [Google Scholar]
  • 28.Wang Y. Prophylactic use of antibiotics for esophageal and gastric variceal hemorrhage. Pharmaceutical research. 2014;9:53–4. [Google Scholar]
  • 29.Xin-Mao Qin A-MX, Shi-Yun Guan, Yao-Hua Hu. Observation on the effect of ofloxacin in preventing nosocomial infection of upper gastrointestinal bleeding in liver cirrhosis. Chinese Journal of Nosocomiology. 2002;12(11):847–8. [Google Scholar]
  • 30.Chao Ma JH, Yi-Hui Rong. Observation on the effect of preventing intra-abdominal infection after liver cirrhosis complicated with gastrointestinal hemorrhage. Prophylactic medicine. 2007;50(7):408–9. [Google Scholar]
  • 31.Shi-Pin Wu JY, Hong Peng, Jian-Liang Zhou, Lin Jin, Wen-Yu Chen. Effect of Antibiotic Prophylaxis on Nosocomial Infection in Cirrhotic Patients with Gastrointestinal Hemorrhage. Chin J Nosocomiol. 2002;12:172–4. [Google Scholar]
  • 32.Lei Wang Z-YB, Xiao-Dong Yuan. The Effects of Preventing Use Antibiotics on the Prognosis of Cirrhotic Patients with Acute Upper Gastrointestinal Hemorrhage. Mark immune analysis and clinical. 2015;22(7):617–9. [Google Scholar]
  • 33.Wei Dong D-RT, Mei Zeng. Influence of curative effect prophylactic use of antibiotics on liver cirrhosis complicated with upper digestive tract hemorrhage. CHINESE COMMUNITY DOCTORS. 2015;31:16–8. [Google Scholar]
  • 34.Qin G-L. Analysis of the effect of prophylactic application of antibiotics on infection in cirrhotic patients with upper gastrointestinal bleeding. CHINESE COMMUNITY DOCTORS. 2020;36:48–9. [Google Scholar]
  • 35.Li Liu W-YZ, Shu Feng, Shen-Gnan Li, Jing Qian, Shu-Ren Liang, Wei Lu, Bin Gao. Effect of prophylactic antibiotics on infection in patients with upper gastrointestinal bleeding from cirrhosis. Shandong medicine. 2015;55:52–4. [Google Scholar]
  • 36.Xu T-T. Effect of prophylactic antibiotics on infection in patients with liver cirrhosis and upper gastrointestinal hemorrhage. China Prac Med. 2021;16:8–10. [Google Scholar]
  • 37.Xiao-Fei Zhang H-DM. Clinical impact analysis of prophylactic use of antibiotics in patients with liver cirrhosis upper gastrointestinal bleeding infection. J Medical Forum. 2018;39:70–2. [Google Scholar]
  • 38.Bin Luo Z-BG. Influence of early application of antibiotics on spontaneous bacterial peritonitis in patients with liver cirrhosis with upper gastrointestinal bleeding. J Clin Intem Med. 2008;25:128–9. [Google Scholar]
  • 39.Fang Peng MX. Effect of early antibiotics on spontaneous bacterial peritonitis of upper digestive tract in cirrhotic patients. Clinical medicine. 2017;13:56. [Google Scholar]
  • 40.Chavez-Tapia NC, Barrientos-Gutierrez T, Tellez-Avila F, Soares-Weiser K, Mendez-Sanchez N, Gluud C, et al. Meta-analysis: antibiotic prophylaxis for cirrhotic patients with upper gastrointestinal bleeding—an updated Cochrane review. Aliment Pharmacol Ther. 2011;34(5):509–18. Epub 2011/06/29. doi: 10.1111/j.1365-2036.2011.04746.x . [DOI] [PubMed] [Google Scholar]
  • 41.Chavez-Tapia NC, Barrientos-Gutierrez T, Tellez-Avila FI, Soares-Weiser K, Uribe M. Antibiotic prophylaxis for cirrhotic patients with upper gastrointestinal bleeding. Cochrane Database Syst Rev. 2010;2010(9):Cd002907. Epub 2010/09/09. doi: 10.1002/14651858.CD002907.pub2 ; PubMed Central PMCID: PMC7138054. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Chong Y, Shimoda S, Yakushiji H, Ito Y, Aoki T, Miyamoto T, et al. Clinical impact of fluoroquinolone-resistant Escherichia coli in the fecal flora of hematological patients with neutropenia and levofloxacin prophylaxis. PloS one. 2014;9(1):e85210. Epub 2014/01/28. doi: 10.1371/journal.pone.0085210 ; PubMed Central PMCID: PMC3898953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Chong Y, Shimoda S, Miyake N, Aoki T, Ito Y, Kamimura T, et al. Incomplete recovery of the fecal flora of hematological patients with neutropenia and repeated fluoroquinolone prophylaxis. Infect Drug Resist. 2017;10:193–9. Epub 2017/07/20. doi: 10.2147/IDR.S133333 ; PubMed Central PMCID: PMC5500534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Pericleous M, Sarnowski A, Moore A, Fijten R, Zaman M. The clinical management of abdominal ascites, spontaneous bacterial peritonitis and hepatorenal syndrome: a review of current guidelines and recommendations. Eur J Gastroenterol Hepatol. 2016;28(3):e10–8. Epub 2015/12/17. doi: 10.1097/MEG.0000000000000548 . [DOI] [PubMed] [Google Scholar]
  • 45.Schultalbers M, Tergast TL, Simon N, Kabbani AR, Kimmann M, Zu Siederdissen CH, et al. Frequency, characteristics and impact of multiple consecutive nosocomial infections in patients with decompensated liver cirrhosis and ascites. United European Gastroenterol J. 2020;8(5):567–76. Epub 2020/03/28. doi: 10.1177/2050640620913732 ; PubMed Central PMCID: PMC7268939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Bush K, Bradford PA. β-Lactams and β-Lactamase Inhibitors: An Overview. Cold Spring Harb Perspect Med. 2016;6(8). Epub 2016/06/23. doi: 10.1101/cshperspect.a025247 ; PubMed Central PMCID: PMC4968164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.EASL Clinical Practice Guidelines for the management of patients with decompensated cirrhosis. Journal of hepatology. 2018;69(2):406–60. Epub 2018/04/15. doi: 10.1016/j.jhep.2018.03.024 . [DOI] [PubMed] [Google Scholar]
  • 48.EASL clinical practice guidelines on the management of ascites, spontaneous bacterial peritonitis, and hepatorenal syndrome in cirrhosis. Journal of hepatology. 2010;53(3):397–417. Epub 2010/07/17. doi: 10.1016/j.jhep.2010.05.004 . [DOI] [PubMed] [Google Scholar]
  • 49.Piano S, Brocca A, Mareso S, Angeli P. Infections complicating cirrhosis. Liver Int. 2018;38 Suppl 1:126–33. Epub 2018/02/11. doi: 10.1111/liv.13645 . [DOI] [PubMed] [Google Scholar]
  • 50.Garcia-Tsao G, Bosch J. Management of varices and variceal hemorrhage in cirrhosis. N Engl J Med. 2010;362(9):823–32. Epub 2010/03/05. doi: 10.1056/NEJMra0901512 . [DOI] [PubMed] [Google Scholar]
  • 51.Mullish BH, Williams HR. Clostridium difficile infection and antibiotic-associated diarrhoea. Clin Med (Lond). 2018;18(3):237–41. Epub 2018/06/03. doi: 10.7861/clinmedicine.18-3-237 ; PubMed Central PMCID: PMC6334067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Ong GK, Reidy TJ, Huk MD, Lane FR. Clostridium difficile colitis: A clinical review. Am J Surg. 2017;213(3):565–71. Epub 2017/01/31. doi: 10.1016/j.amjsurg.2016.10.035 . [DOI] [PubMed] [Google Scholar]
  • 53.Schweizer U, Grund KE, Fundel J, Wichmann D, Königsrainer A. [Treatment of nonvariceal upper gastrointestinal bleeding: endoluminal-endovascular-surgical]. Der Chirurg; Zeitschrift fur alle Gebiete der operativen Medizen. 2019;90(8):607–13. Epub 2019/08/09. doi: 10.1007/s00104-019-0948-7 . [DOI] [PubMed] [Google Scholar]
  • 54.Kuo MT, Yang SC, Lu LS, Hsu CN, Kuo YH, Kuo CH, et al. Predicting risk factors for rebleeding, infections, mortality following peptic ulcer bleeding in patients with cirrhosis and the impact of antibiotics prophylaxis at different clinical stages of the disease. BMC gastroenterology. 2015;15:61. Epub 2015/08/14. doi: 10.1186/s12876-015-0289-z ; PubMed Central PMCID: PMC4533793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Popović D, Stanković-Popović V, Jovanović I, Krstić M, Djuranović S, Mijalković N, et al. [Endoscopic haemostasis of bleeding duodenal ulcer]. Acta chirurgica Iugoslavica. 2007;54(1):145–50. Epub 2007/07/20. doi: 10.2298/aci0701145p . [DOI] [PubMed] [Google Scholar]
  • 56.Awadalla MS, Desimone M, Wassef W. Updates on management of nonvariceal upper gastrointestinal bleeding. Current opinion in gastroenterology. 2019;35(6):517–23. Epub 2019/10/09. doi: 10.1097/MOG.0000000000000582 . [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Checklist. PRISMA 2020 checklist.

(DOCX)

Click here for additional data file. (31.8KB, docx)

Data Availability Statement

All relevant data are within the paper and its Supporting Information files.

Articles from PLOS ONE are provided here courtesy of PLOS

RESOURCES