Objective To systematically review the efficacy and safety of different SGLT2 inhibitors in the treatment of heart failure. Methods The Cochrane Library, Web of Science, PubMed and EMbase databases were searched for randomized controlled trials on the efficacy and safety of SGLT2 inhibitors in patients with heart failure from inception to July 2, 2021. Two researchers independently screened literature, extracted data and evaluated the risk of bias of the included studies. Network meta-analysis was then performed using Stata 16.0 software. Results A total of 16 randomized controlled trials, including 15 312 patients, involving 5 interventions, namely dapagliflozin, empagliflozin, canagliflozin, sotagliflozin and ertugliflozin were included. Results of network meta-analysis showed that there was no significant difference in the compound outcome of hospitalization for heart failure or cardiovascular death, hospitalization for heart failure, all-cause mortality, risk of cardiovascular mortality and serious adverse reactions among patients with heart failure among 5 different SGLT2 inhibitors (P>0.05). Compared with placebo, both selective and non-selective SGLT2 inhibitors improved the risk of hospitalization for heart failure, hospitalization for heart failure, or compound cardiovascular mortality (P<0.05), while only selective SGLT2 inhibitors improved the risk of cardiovascular mortality, all-cause mortality, and serious adverse events (P<0.05). However, there was no significant difference between them (P>0.05). The area under the cumulative ordering probability curve of selective and non-selective SGLT2 inhibitors ranked first and second, except for the combined outcome of heart failure or cardiovascular death. Conclusion The current evidence indicates that there is no significant difference in the efficacy and safety of the 5 different SGLT2 inhibitors in the treatment of heart failure, and there is no significant difference between selective SGLT2 inhibitors and non-selective SGLT2 inhibitors. Due to the limited quantity and quality of included studies, more high-quality studies are needed to verify the above conclusion.