Aims We undertook a systematic review and meta-analysis to assess the efficacy and safety of sodium-glucose cotransporter 2 inhibitors (SGLT2is) concerning kidney outcomes in patients with type 2 diabetes mellitus (T2DM), with or without prevalent kidney disease. Materials and Methods PubMed, Web of science, Embase and the Cochrane Library were systematically searched for randomized controlled trials (RCTs) to assess the efficacy and safety of treatment with SGLT2is versus placebo in patients with T2DM. The weighted mean difference (WMD) and its 95% confidence interval (CI) were applied for continuous variables, and the risk ratio (RR) and corresponding 95% CI were used for dichotomous outcomes. Patients were categorized according to whether the baseline mean estimated glomerular filtration rate (eGFR) was less or was more than 60 mL/min/1.73 m(2). Results A total of 25 eligible studies with 43 721 participants were included. There was an initial and small decrease in eGFR during the early treatment period (WMD, -4.63; 95% CI, -6.08 to -3.19 mL/min/1.73 m(2)), which was noted at 1-6 weeks and gradually narrowed over time, with a decline in protection from eGFR in the long term (WMD, 3.82; 95% CI, 2.80-4.85 mL/min/1.73 m(2)). SGLT2is significantly delayed albuminuria progression (RR, 0.71; 95% CI, 0.66-0.76), promoted albuminuria regression (RR,1.71; 95% CI, 1.54-1.90), improved the composite of >= 40% decrease in eGFR, in the need for renal-replacement and in death from renal causes (RR, 0.57; 95% CI, 0.49-0.66), and reduced all-cause mortality (RR, 0.84; 95% CI, 0.75-0.94). At the same time, they significantly increased the risk of genital infection (RR, 3.43; 95% CI, 2.87-4.10) vs placebo in patients with T2DM. Meta-regression analyses showed that eGFR-preservation effects were not significantly associated with basic patient characteristics (age, BMI, HbA1c, eGFR level), but were influenced by drug administration (treatment duration, type, dosage of SGLT2is). Subgroup analyses showed that the relative effects on renal outcomes of SGLT2is vs placebo were similar across eGFR subgroups (P heterogeneity >0.05). Conclusions SGLT2is slowed eGFR decline, lowered albuminuria progression, improved adverse renal endpoints and reduced all-cause mortality, but increased risk of genital infections vs placebo in patients with T2DM. The indication of consistent renal benefits across categories of baseline eGFR levels may allow additional individuals to benefit from SGLT2is therapy.