ABSTRACT
PURPOSE: Sodium glucose cotransporter-2 inhibitors (SGLT2i) have been evaluated in children with type 2 diabetes mellitus (T2DM), type 1 diabetes mellitus (T1DM), and several other nondiabetic conditions. Potential tolerability issues have prevented the routine use of SGLT2i in children with diabetes. However, no meta-analysis to date has evaluated the safety and tolerability of SGLT2i in children. This systematic review and meta-analysis aimed to address this knowledge gap. METHODS: Databases were searched for randomized controlled trials (RCTs), case control, and cohort studies involving children receiving SGLT2i in the intervention-arm. Primary outcome was occurrence of treatment emergent adverse events (TAEs). Secondary outcomes were evaluation of glycemic efficacy and occurrence of severe adverse events (SAEs), hypoglycemia, ketosis, genital or urinary infections, and any other adverse events. RESULTS: From the 27 articles initially screened, data from 4 RCTs (258 children) were analyzed. In children with T2DM, occurrence of TAEs (odds ratio [OR], 1.77; 95% confidence interval [CI], 0.93-3.36; P=0.08; I2=0%), SAEs (OR, 0.45; 95% CI, 0.08-2.54; P=0.37; I2=0%), ketoacidosis (OR, 0.33; 95% CI, 0.01-8.37; P=0.50), urinary tract infections (OR, 2.34; 95% CI, 0.44-12.50; P=0.32; I2=0%), and severe hypoglycemia (OR, 4.47; 95% CI, 0.21-96.40; P=0.34) were comparable among the SGLTi group and placebo. Compared to placebo, T2DM children receiving SGLTi had significantly lower glycosylated hemoglobin at 24-26 weeks (mean difference [MD], -0.79%; 95% CI, -1.33 to -0.26; P=0.004; I2=0%). In T1DM children, ß-hydroxybutyrate levels were significantly higher in the SGLTi group than the placebo group (MD, 0.11 mmol/L; 95% CI, 0.05-0.17; P=0.0005; I2=53%). In T1DM, there was not a single report of an SAE, ketoacidosis, or severe hypoglycemia in either the placebo or treatment groups, but time-in-range was considerably greater in the SGLT2i group than the placebo group (68%±6% vs. 50%±13%, P<0.001). CONCLUSION: SGLT2i use in children and young adults appears to be both safe and tolerable based on our meta-analyses and review of the literature.
