Gastroenterology Research, ISSN 1918-2805 print, 1918-2813 online, Open Access
Article copyright, the authors; Journal compilation copyright, Gastroenterol Res and Elmer Press Inc
Journal website https://gr.elmerpub.com

Original Article

Volume 18, Number 3, June 2025, pages 139-148

Comparison of Prophylactic Transcatheter Arterial Embolization and Standard Therapy in High-Risk Non-Variceal Upper Gastrointestinal Bleeding: A Meta-Analysis

Figures

Figure 1.
Figure 1. Forest plot of overall rebleeding rate in comparison between prophylactic transcatheter arterial embolization (P-TAE) and no embolization.
Figure 2.
Figure 2. Forest plot of overall rebleeding rate in comparison between prophylactic transcatheter arterial embolization (P-TAE) and no embolization (only RCTs).
Figure 3.
Figure 3. Forest plot of overall rebleeding rate in comparison between prophylactic transcatheter arterial embolization (P-TAE) and therapeutic arterial embolization (TAE).
Figure 4.
Figure 4. Forest plot of all-cause mortality in comparison between prophylactic transcatheter arterial embolization (P-TAE) and no embolization.
Figure 5.
Figure 5. Forest plot of all-cause mortality in comparison between prophylactic transcatheter arterial embolization (P-TAE) and therapeutic arterial embolization (TAE).
Figure 6.
Figure 6. Forest plot of procedure-related adverse events in comparison between prophylactic transcatheter arterial embolization (P-TAE) and therapeutic arterial embolization (TAE).

Table

Table 1. Characteristics of Included Studies
 
StudyDesignCountryYear of studyDefinition of high-risk gastrointestinal bleedingStudy groups, N (%)Forrest classification (%)
NEP-TAETAEIaIbIIaIIb
P-TAE: prophylactic transcatheter arterial embolization; NE: no embolization in the absence of angiographic evidence of bleeding; TAE: therapeutic arterial embolization; NVUGIB: non-variceal upper gastrointestinal bleeding.
Padia et al [12]Retrospective studyUnited States2009Initial endoscopy fails to arrest the NVUGIB.-72 (67)36(33)----
Ichiro et al [13]Retrospective studyJapan2011Persistent and substantial NVUGIB and hemodynamic instability despite endoscopic intervention.-36 (61)23(39)----
Arrayeh et al [14]Retrospective studyIsrael, United States2012Initial endoscopy fails to arrest the NVUGIB.17 (14.7)56 (48.7)42(36.6)----
Dixon et al [15]Retrospective studyUK2012Initial endoscopy fails to arrest the NVUGIB.7 (17.5)20 (50)13(32.5)----
Laursen et al [16]Randomized controlled trialDenmark2013Bleeding from ulcers classified as Forrest I-IIb.68 (70.5)31 (29.5)-1032507
Sildiroglu et al [17]Retrospective studyUnited States, Turkey2014Initial endoscopy fails to arrest the NVUGIB.22 (29.7)18 (24.3)31 (41.9)----
Mille et al [18]Retrospective studyGermany2014If patients exhibited at least 1 endoscopic as well as 1 clinical risk factor, then they were defined as high-risk patients.47 (40)55 (47)15 (13)19 (16)39 (33)14 (12)10 (9)
Kaminskis et al 2017 [19]Prospective StudyLatvia2017Forrest I-IIb type of ulcer and the Rockall score ≥ 5.50 (66.7)25 (33.3)-5 (20)4 (16)11 (44)5 (20)
Lau et al [20]Randomized controlled trialHong Kong, Thailand, Netherlands, China2018Patients with actively bleeding gastroduodenal ulcers (Forrest type I) or ulcers with non-bleeding visible vessels (Forrest IIa).123 (51)118 (49)-42 (17.3)90 (37.4)109 (45.2)0
Kaminskis et al 2019 [21]Prospective studyLatvia2019Patients with Forrest Ia-IIc ulcers and the Rockall score ≥ 6.341 (85.5)58 (14.5)-6 (10.3)8 (13.8)22 (37.9)22 (37.9)