Thrombocytopenia is common in liver cirrhosis patients, which often complicates with patients' frequent issue with gastrointestinal bleeding and procedural needs. Based on biologic understanding of decreased thrombopoietin(TPO) level in liver cirrhosis patients, use of TPO agonists in liver cirrhosis have been actively studied. Over the period of time, new studies have come out about 2 FDA-approved TPO agonists, Avatrombopag and Lusutrombopag, for prophylaxis before procedure in liver cirrhosis patients with thrombocytopenia. In the past, there have raised concerns of increased risk of portal vein thrombosis and other thrombotic risks in other agents. In our study, we aimed to study the effectiveness and safety of TPO receptor agonists for pre-procedural use in patients with liver cirrhosis.

Study design

Study was conducted from August 2018 to July 2019. Previous studies were identified through database searching MEDLINE, CENTRAL, and google search. Randomized clinical trials with active treatment arm with TPO receptor agonists in the use of liver cirrhosis patients, with intention of pre-procedural prophylaxis, and having placebo for direct comparisons were included. One of the major exclusions was TPO receptor agonists to increase platelet counts for anti-viral treatment in cirrhosis patients. 14 studies were identified. Studies were reviewed and eligibility was determined by two independent clinically trained researchers. 5 duplicated studies were removed, and in total of 7 studies were included for quantitative and qualitative analysis. Details of studies were collected by 2 independent researchers and compared. When there is a discrepancy, repeat review of the study was conducted. Studies were conducted or published from 2012 to 2018. 1 trial from Eltrombopag, 3 trials from Avatrombopag, and 3 trials from Lusutrombopag were included.


Characteristics of included studies are summarized in table 1. Our studies found that 5.5(95% CI : 4.35-7.15) times higher risk ratio(RR) of reaching platelet target before procedure compared to placebo. Target platelet number goal was 50,000 to 80,000 depending on the study. Studies showed homogenous result with I-squared was 30.8% and q-statistics of p-value 0.193.(Figure 1) Subgroup analysis by FDA-approved medication of Avatrombopag and Lusutrombopag showed statistically significant higher risk ratio of 4.74(3.36-6.68) and of 5.52(3.65-8.34) each compared to placebo. Risk ratio for preventing platelet transfusion was not able to be calculated with heterogeneity of study with I-square higher than 90%. Lusutrombopag study showed significant benefits (RR 6.33, 95% CI 2.95-13.58) with heterogeneity inside the same medications, which might be explained with different doses in studies. No statistical significance in risk ratio in a study with Avatrombopag. Subgroup analysis limited to phase 3 studies showed risk ratio of preventing transfusion of 2.87(95% CI 2.15-3.82) but heterogeneity with q-statistics of p-value at 0.029.

Relative risk for adverse event related to portal vein thrombosis was not statistically significant with RR of 0.99(95% CI : 0.35-2.85) in total of 1,229 patients.(Figure 2) Study result was homogenous result by I-square 0%, q-statistics of p-value 0.794. Other severe adverse events, major bleeding risk, overall thrombosis risk were not statistically significant. Only increased risk without statistical significance was reported in trail with Eltrombopag which was early terminated.


Our meta-analysis of pre-procedural use of TPO agonist in liver cirrhosis patients showed statistically significant benefit of reaching platelet count goal by 5.58 times with risk ratio, but no benefit of preventing transfusion. Compared to prior studies including use of TPO agonists for Interferon-Ribavirin treatment, meta-analysis limited to pre-procedural use did not show statistically significant increase in portal vein thrombosis. Serious adverse events including thrombosis events and bleeding risk were not statistically significant. Most studies described that portal-vein thrombosis events were often related to high platelet counts about 200,000 and longer use of TPO agonist. In current era with lesser use of Interferon and Ribavirin as an anti-viral therapy, TPO agonists use in setting of pre-procedure mostly with lower platelet targets can be safely used.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution