Thrombopoietin receptor agonists (TPO-RAs) play a crucial role in stimulating thrombopoiesis. However, conventional meta-analyses have shown inconsistent results regarding the efficacy of thrombopoietin receptor agonists versus placebo. Therefore, we performed a network meta-analysis to assess the effects of five TPO-RAs via indirect comparison. For this network meta-analysis, we considered randomized trials that included any of the following interventions: avatrombopag, lusutrombopag, eltrombopag, romiplostim, recombinant human thrombopoietin (rhTPO). We searched the Medline, PubMed, Embase, the Cochrane Library, and Web of Science databases for randomized controlled clinical trials from inception to January 31, 2021. We use randomized controlled clinical trials of TPO-RAs for treatment of immune thrombocytopenia in adults. The primary outcome was the number of patients achieving platelet response which was defined as the achievement of a platelet count of more than 30 or 50 cells × 109/L in the absence of rescue therapy, and the secondary outcome was the therapy-related serious adverse events and incidence of bleeding episodes. To obtain the estimates of efficacy and safety outcomes, we performed a random-effects network meta-analysis. These estimates were presented as odds ratios with 95% confidence intervals. We use surface under the cumulative ranking probabilities to rank the comparative effects and safety of all drugs against the placebo. In total, 2,207 patients were analyzed in 20 clinical trials. All preparations improved the point estimates of platelet response when compared with the placebo. Avatrombopag and lusutrombopag had the best platelet response compared to the placebo, the former had a non-significant advantage compared to the latter [odds ratio (OR) = 1.91 (95% confidence interval: 0.52, 7.05)]. The treatments were better than eltrombopag, romiplostim, rituximab, and rhTPO + rituximab, with corresponding ORs of 3.10 (1.01, 9.51), 9.96 (2.29, 43.29), 33.09 (8.76, 125.02), and 21.31 (3.78, 119.98) for avatrombopag and 1.62 (0.63, 4.17), 5.21 (1.54, 17.62), 17.34 (5.15, 58.36), and 11.16 (2.16, 57.62) for lusutrombopag. Regarding bleeding, the placebo group had the highest probability of bleeding, whereas lusutrombopag had the lowest risk of bleeding when compared to the placebo. Adverse events were slightly higher in patients receiving rituximab than in those receiving placebo or other treatments. Overall, this meta-analysis showed that avatrombopag may yield the highest efficacy because it has the most favorable balance of benefits and acceptability.