Sodium-glucose cotransporter 2 (SGLT2) inhibitors are widely prescribed for type 2 diabetes mellitus (T2DM) because of demonstrated survival and cardiovascular benefits. However, multiple reports have documented increases in hemoglobin (Hgb) and hematocrit (Hct) with these agents, raising concern for drug-associated erythrocytosis/secondary polycythemia and a possible attendant thrombotic risk. The clinical relevance of these hematologic changes remains unclear; this review aimed to synthesize available randomized and observational data on hematologic effects of SGLT2 inhibitors in adults with T2DM.
The authors conducted a systematic review and meta-analysis in accordance with PRISMA guidelines. PubMed, Scopus, and Embase were searched through February 2026 for studies assessing hematologic parameters after SGLT2 inhibitor therapy in adults with T2DM. The predefined primary outcomes were changes in Hgb and Hct. Secondary outcomes included incidence of erythrocytosis/polycythemia and reported thrombotic events. Patients with primary polycythemia, end-stage renal disease (ESRD), or heart failure were excluded from the review per the Methods section of the source.
Ten studies met inclusion criteria, comprising a mix of randomized controlled trials and observational (real-world) studies. Follow-up durations in included studies ranged from 8 weeks to 24 months. The SGLT2 inhibitors evaluated included empagliflozin, dapagliflozin, and canagliflozin.
Pooled randomized evidence from three randomized controlled trials (total n = 171) demonstrated statistically significant increases in hematologic measures with SGLT2 inhibitors versus placebo. The pooled mean difference for hematocrit was +2.29% (95% CI 1.48 to 3.09). For hemoglobin, the pooled mean difference was +0.51 g/dL (95% CI 0.28 to 0.74). Across the included studies, increases in Hgb and Hct were described consistently.
Observational and real-world datasets included larger numbers of patients. One pooled real-world sample (n = 9,646) reported an increase in polycythemia prevalence from 2.4% to 9.7% after SGLT2 inhibitor exposure, with severe cases noted in 1.4% of patients. Data addressing thrombotic outcomes were limited: one observational study of 100 patients with JAK2-unmutated erythrocytosis reported a 10% event rate for thrombotic events, suggesting a possible association, but broader evidence on thrombotic risk remains sparse in the available literature.
Hematologic changes were observed early after initiation of SGLT2 inhibitor therapy in the reviewed studies. The source reports that these increases in Hgb and Hct appeared reversible after drug discontinuation. Mechanistic hypotheses discussed in cited literature (reported within the source references) include hemoconcentration from diuresis and increased erythropoiesis mediated by erythropoietin and amelioration of renal hypoxia, but detailed mechanistic data were not synthesized in depth in the abstract.
The review emphasizes that while increases in Hgb and Hct may partially underlie some cardiovascular benefits associated with SGLT2 inhibitors, a subset of patients may develop clinically significant erythrocytosis/polycythemia. The authors recommend clinician awareness and monitoring of high-risk individuals. Specific monitoring protocols, thresholds for action, or management strategies were not detailed in the abstract; the source calls for heightened vigilance rather than prescribing a defined surveillance pathway.
Note: the Methods section of the review excluded patients with ESRD and heart failure, yet the Conclusions section of the abstract contains a sentence stating an association in "patients with type 2 diabetes, ESRD, or heart failure." This discrepancy is reported as presented in the source.
Limitations identified by the authors include a small number of randomized trials contributing to pooled estimates (three RCTs, 171 patients) and variable follow-up across studies. Observational datasets provided larger sample sizes but carry inherent limitations related to confounding and outcome ascertainment. Data on clinical endpoints such as thrombotic events were limited and inconsistent. The authors conclude that larger prospective studies are needed to clarify the prevalence, clinical significance, mechanisms, and outcomes (including thrombotic risk) of SGLT2-associated hematologic changes.
The systematic review and meta-analysis found that SGLT2 inhibitor therapy in adults with T2DM is consistently associated with increases in hemoglobin and hematocrit, with pooled randomized data showing modest but statistically significant rises. Real-world data indicate that a minority of patients may develop clinically meaningful polycythemia, and limited observational evidence raises concern for possible thrombotic complications in selected cohorts. Clinicians should be aware of this potential adverse effect and consider monitoring hematologic parameters in higher-risk patients. The source concludes that further large-scale prospective research is required to define best practices for surveillance and management.