To summarize the evidence from systematic reviews (SRs) of randomized controlled trials (RCTs) evaluating the efficacy and safety of sodium-glucose cotransporter-2 (SGLT2) inhibitors versus placebo or active comparators for type 2 diabetes mellitus. We searched six databases between 2014 and 2021. We assessed the quality of evidence using Assessment of Multiple Systematic Reviews (AMSTAR 2) and Grading of Recommendations Assessment, Development and Evaluation(GRADE) and summarized the main outcome results according to their evidence of benefit (PROSPERO ID: CRD42019132431). We included 46 SRs, comprising 175 RCTs and 136 096 participants. The results showed "clear evidence of benefit" in relation to: myocardial infarction (odds ratio [OR]/hazard ratio [HR] 0.85 to 0.91); cardiovascular mortality (OR/HR 0.67 to 0.86); heart failure (OR/HR 0.64 to 0.69); albuminuria progression and composite renal outcome (relative risk [RR]/HR 0.55 to 0.63); glycated haemoglobin (HbA1c) versus placebo (mean difference [MD] -0.49% to -0.77% [5.4 to 8.4 mmol/mol]); and weight versus placebo (MD -1.09 kg to -2.99 kg). "Possible benefit" was observed in relation to major adverse cardiovascular events (OR/HR 0.80 to 0.89), all-cause mortality and nonalcoholic fatty liver disease. SGLT2 inhibitors showed "clear evidence of no effect or equivalence" in relation to stroke and fractures. "Clear evidence of harm" was observed in relation to genital infections (RR/OR 2.06 to 5.25) and ketoacidosis (HR/OR 1.36 to 2.20). Regarding amputation risk and urinary tract infections, we found "no conclusions possible due to lack of evidence". Our results showed that SGLT2 inhibitors have beneficial effects in relation to renal and cardiovascular outcomes (except for stroke), HbA1c and weight. Further studies are needed to assess urinary infections and amputation risk.