ENDOSCOPIC BAND LIGATION VERSUS ARGON PLASMA COAGULATION IN THE TREATMENT OF GASTRIC ANTRAL VASCULAR ECTASIA: A SYSTEMATIC REVIEW AND META-ANALYSIS

Background & Aims: Argon plasma coagulation (APC) is the established endoscopic treatment for gastric antral vascular ectasia (GAVE). Recently, the use of endoscopic band ligation (EBL) has gained recognition as an alternate approach. Data suggest that EBL might yield better clinical outcomes when compared to APC. The objective of our study was to compare the efficacy and safety of APC vs EBL in the treatment of GAVE.

Methods: We searched multiple databases including Medline, Web of Science, Cochrane Library, Pubmed, Embase and conference proceedings from inception through October 2023 for articles reporting outcomes of APC vs EBL in patients with GAVE. We studied the eradication rates, adverse events, transfusion requirements and recurrence rates among patients receiving these procedures. Standard meta-analysis methods were employed using the random-effects model, and heterogeneity was studied by I2 statistics.

Results: We analyzed 10 studies including 6 prospective and 4 retrospective studies involving 476 subjects (47.47% males and mean age of 56.25 years). 197 patients underwent EBL while 279 subjects received APC.
The number of sessions required for obliteration of lesions were 2.43±0.14 with EBL and 3.65±0.48 with APC. The pooled eradication rate of GAVE with EBL was 88.6% [95% Cl 79.9-81.5; I2=13.5%] and with APC was 57.9% [Cl 43.7-71; I2= 59%]; pooled RR 1.52 [1.16-2.02; I2=72%; P<0.001]. The EBL group was associated with lower bleeding recurrence, 6.6% [Cl 3.4-12.5; I2%=0%] compared to the APC group 39.7% [Cl 26.9-54.15; I2=55%]; RR 0.21 [0.09-0.44; I2=0%; p< 0.001]. The recurrence of GAVE with EBL was 7.3% [Cl 3.8-13.6; I2=0%]and with APC was 38.5% [Cl 24.4-54.9; I2=64%]; RR 0.22 [0.109-0.446; I2= 0%; p<0.01].
The pooled mean pre-treatment hemoglobin (Hb) in EBL and APC was 7.83 [Cl 6.9-8.7, I2=93%] and 7.90 [Cl 7.05-8.75; I2=55%]. The pooled mean post-treatment Hb in EBL and APC was 9.9 [Cl 9.1-10.7; I2=96%] and 9.68 [CI9.05-10.64; I2=96%]. The standard difference between mean pre-procedure Hb and post-procedure Hb between EBL and APC was 0.36± 0.124 [0.132-0.619; p <0.003].
The pooled transfusion requirements were lower with EBL 1.24 [Cl 0.82-1.66; I2=98%] compared to APC 3.098 [Cl 1.19-5.0; I2=99%]. Similarly, the pooled hospitalizations were lower with EBL 0.67 [Cl 0.31-1.04; I2=99%] in contrast to APC 1.673 [Cl 0.13-3.48; I2=99%]. The pooled rate of adverse events associated with EBL was 16.8% [Cl 6.6-36.7; I2= 83%] compared to APC 9.3% [Cl 5.6-15.1; I2=19%] with RR 2.11 [ 0.8-5.46; 58%; p =0.1].

Conclusion: Based on this meta-analysis, EBL demonstrated significantly better eradication (88.6% vs 57.9%) with fewer treatment sessions, recurrent bleeding, hospitalization and transfusion requirement as compared to APC. It is time that EBL be utilized as the first line endoscopic treatment for GAVE instead of APC.