Across 85 randomized trials covering 70,543 adults (mean age 62; median follow-up 24 weeks), SGLT2 inhibitor use raised hemoglobin by 0.60 g/dL (95% CI 0.55 to 0.66) and hematocrit by 2.36 percentage points versus control, which was mostly placebo. Erythropoietin rose modestly (about 2.1 mIU/mL difference), and four of six secondary markers of erythropoiesis and iron metabolism shifted significantly. Baseline hemoglobin averaged 13.6 g/dL, so the typical participant was not anemic.
This pooled estimate puts a firm number on a signal long seen in trials like EMPA-REG and CREDENCE, and the EPO rise fits the leading explanation: reduced tubular workload in the kidney relieves oxygen stress and lets EPO-producing interstitial cells recover, with iron mobilization following. Hemoconcentration from diuresis likely contributes early on, but the EPO and iron data argue that true erythropoiesis is also involved. The practical upside is that some of the cardiorenal benefit of these drugs may travel through better oxygen delivery, and people with anemia of chronic kidney disease or heart failure might gain a side benefit. The caveats matter: these are trial-level data with short follow-up, a hemoglobin rise is a surrogate rather than a proven mediator of fewer events, and the clinical meaning of a higher hematocrit in already-replete patients is unclear. Adults on these drugs should still have anemia and iron status evaluated on its own terms. This is a confirmatory but useful quantification, not a change in practice.