Sodium-Glucose Cotransporter 2 Inhibitor Use and Levels of Hemoglobin, Markers of Iron Metabolism, and Erythropoiesis

Importance In randomized clinical trials, sodium-glucose cotransporter 2 (SGLT2) inhibitor use was associated with increased hemoglobin and hematocrit compared with placebo. However, the magnitude and importance of this association remain unclear. Objective To evaluate the associations between SGLT2 inhibitor use and levels of hemoglobin, hematocrit, and markers of iron metabolism and erythropoiesis compared with placebo or active comparator. Data Sources PubMed, Web of Science, and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched from inception to July 22, 2026. Study Selection Randomized, double-masked clinical trials comparing SGLT2 inhibitor use with placebo or active comparators in patients with 1 or more of the following conditions: cardiovascular disease, diabetes, heart failure, and chronic kidney disease. Data Extraction and Synthesis Three reviewers independently extracted data and results were pooled using random-effects meta-analysis. Main Outcomes and Measures Primary outcomes were changes in hemoglobin and hematocrit. Secondary outcomes were changes in red blood cell concentration, erythropoietin (EPO), and iron measures (serum iron, ferritin, transferrin saturation, and hepcidin). Results A total of 85 randomized clinical trials were analyzed (median [IQR] duration, 24 [12-32] weeks), including 70 543 participants (mean [SD] age, 62.4 [9.4] years; 25 334 female [35.9%]). Mean (SD) baseline hemoglobin was 13.6 (1.5) g/dL; baseline hematocrit was 41.5% (4.9). Seventy-eight clinical trials (67 208 participants) compared SGLT2 inhibitors with placebo, and 7 clinical trials (3335 participants) used active comparators, most commonly sitagliptin (2 trials [1158 participants]). From baseline to follow-up, hemoglobin changed by 0.55 g/dL (95% CI, 0.49 to 0.61 g/dL; P < .001) in the SGLT2 inhibitor group, and −0.08 g/dL (95% CI, −0.13 to −0.03 g/dL; P = .003) in the control group (difference, 0.60 g/dL; 95% CI, 0.55 to 0.66 g/dL; P < .001). From baseline to follow-up, hematocrit changed by 2.17% (95% CI, 1.95% to 2.39%; P < .001) in the SGLT2 inhibitor group and −0.13% (95% CI, −0.28% to 0.01%; P = .07) in the control group (difference, 2.36%; 95% CI, 2.19% to 2.52%; P < .001). Of the 6 secondary outcomes, 4 showed significant associations with SGLT2 inhibitors, including increased EPO (2.35 mIU/mL in SGLT2 inhibitor group vs 0.04 mIU/mL in control group; difference, 2.11 mIU/mL [95% CI, 0.5-3.71 mIU/mL]; P = .01) and decreased transferrin saturation (−2.2% in SGLT2 inhibitor group vs −0.10% in control group; difference, −0.84% [95% CI, −1.21% to −0.47%]; P < .001). Conclusions and Relevance In a meta-analysis of randomized clinical trials, compared with control, SGLT2 inhibitor use was associated with small statistically significant increases in hemoglobin and hematocrit levels. The clinical significance of these findings is unclear.