Does preemptive transjugular intrahepatic portosystemic shunt improve survival after acute variceal bleeding? Systematic review, meta‐analysis, and trial sequential analysis of randomized trials.

Background and Aim: A preemptive transjugular intrahepatic portosystemic shunt (p‐TIPSS) after acute variceal bleeding (AVB) is advocated. However, when compared with the current standard of care, the survival benefit of p‐TIPSS is questionable. We performed a systematic review, meta‐analysis, and trial sequential analysis of randomized control trials (RCTs) to assess the survival benefit of p‐TIPSS in patients with cirrhosis and AVB. Methods: Comprehensive literature search of three bibliographic databases (MEDLINE, Embase, and Cochrane) was conducted from inception till May 2021. All study types evaluating the survival benefit of p‐TIPSS in AVB were considered for inclusion. The relative risk (RR) of mortality and rebleeding at 6 weeks and mortality at 1 year with a random‐effects model was computed. Trial sequential analysis was performed for the primary outcome of 6‐week mortality. Results: A total of nine studies (four RCTs and five cohort studies) comprising 2861 patients with AVB were included. The overall pooled risks of mortality at 6 weeks and 1 year were 17.9% (95% confidence interval [CI]: 16.5–19.3%) and 26.7% (95% CI: 25.0–28.3%), respectively. Although p‐TIPSS was associated with lower 6‐week rebleeding risk (RR = 0.20; 95% CI = 0.13–0.29, I2 = 0%), data from pooled RCTs showed no significant difference in mortality at 6 weeks (RR = 0.33; 95% CI = 0.08–1.36, I2 = 63%) or at 1 year (RR = 0.76; 95% CI = 0.51–1.14, I2 = 30%). Using trial sequential analysis, required sample size to detect a 20% relative risk reduction in mortality at 6 weeks with p‐TIPSS was estimated to be 6317, which is beyond the total number of patients available for analysis. Conclusions: This meta‐analysis found that the available data from RCTs are insufficient to confer 6‐week mortality benefit with p‐TIPSS compared with standard of care; thus, adequately powered RCTs are required. [ABSTRACT FROM AUTHOR]