https://orcid.org/0000-0003-0737-3603
TO THE EDITOR:
Doi et al1 present a compelling post hoc analysis of the CANVAS and CREDENCE trials, exploring the relationship between canagliflozin, hematocrit, and thrombosis. Their key finding that baseline hematocrit modifies risk in a sex-specific manner represents an important advancement in understanding the hematologic effects of SGLT2 inhibitors (SGLT2is). They demonstrate that canagliflozin raises hematocrit regardless of baseline levels and that males with erythrocytosis may be at increased thrombotic risk, an effect not observed in females.
This analysis addresses an important evidentiary gap. Prior retrospective studies2-6 reported thrombotic events in patients with SGLT2i-associated erythrocytosis, occurring in 2.4% to 10% of patients (Table 1), but they were limited by small sample sizes, observational design, or lack of control groups. Although some have suggested that erythrocytosis may mediate the cardioprotective effects of SGLT2is, Doi et al1 show that such benefit is limited to patients with anemia and may confer thrombotic risk in those with baseline erythrocytosis, particularly males.
Summary of SGLT2i-associated erythrocytosis retrospective cohort studies
| Study . | N . | Population . | Male, n (%) . | Age∗, y . | Baseline Hb∗, g/L . | Peak Hb∗, g/L . | Hb change†, g/L . | Thrombotic events, n (%) . |
|---|---|---|---|---|---|---|---|---|
| Liu et al2 | 75 | Hematology | 56 (68.3) | 63 (30-85) | 161 (114-190) | 185 (162-216) | +24 | 5 (6.7) |
| Gangat et al3 | 30 | Diabetes | 20 (67) | 64 (29-81) | 158 (134-169) | 179 (153-198) | +21 | 2 (7) |
| Gangat et al4 | 100‡ | Diabetes | 78 (78) | 62 (29-87) | 155 (120-179) | 180 (151-210) | +25 | 10 (10) |
| Gill et al5 | 42 | Renal transplant | 32 (76) | 59 ± 12.3 | NA | NA | NA | 1 (2.4) |
| Lassen et al6 | 7926 | Nationwide | 3028 (38) | 60.7 ± 11.7 | 142 ± 15 | NA | +4.3 | NA |
| Study . | N . | Population . | Male, n (%) . | Age∗, y . | Baseline Hb∗, g/L . | Peak Hb∗, g/L . | Hb change†, g/L . | Thrombotic events, n (%) . |
|---|---|---|---|---|---|---|---|---|
| Liu et al2 | 75 | Hematology | 56 (68.3) | 63 (30-85) | 161 (114-190) | 185 (162-216) | +24 | 5 (6.7) |
| Gangat et al3 | 30 | Diabetes | 20 (67) | 64 (29-81) | 158 (134-169) | 179 (153-198) | +21 | 2 (7) |
| Gangat et al4 | 100‡ | Diabetes | 78 (78) | 62 (29-87) | 155 (120-179) | 180 (151-210) | +25 | 10 (10) |
| Gill et al5 | 42 | Renal transplant | 32 (76) | 59 ± 12.3 | NA | NA | NA | 1 (2.4) |
| Lassen et al6 | 7926 | Nationwide | 3028 (38) | 60.7 ± 11.7 | 142 ± 15 | NA | +4.3 | NA |
Hematology population included all undifferentiated patients presenting to a hematology clinic with erythrocytosis; diabetes population included consecutive patients with diabetes mellitus with erythrocytosis; renal transplant population included patients with erythrocytosis after kidney transplant; nationwide population included patients from the Danish national registry.
Hb, hemoglobin; NA, not available.
Median (range) or mean ± standard deviation.
Median or mean change.
Included the 30 patients from Gangat et al3 with additional follow-up.
With the expanding use of SGLT2is, erythrocytosis associated with this medication class is increasingly encountered by hematologists, found in >12% of patients referred for JAK2 testing according to real-world data.7 The findings by Doi et al1 support the need for complete blood count monitoring and individualized risk assessment, particularly in males with persistent or rising hematocrit, in whom emerging evidence suggests potential harm.
What is most needed are prospective studies to better characterize thrombotic risk across secondary causes of erythrocytosis to guide evidence-based management (eg, EVEREST [ClinicalTrials.gov identifier: NCT06785870]). Our recent systematic review in this journal8 highlights the growing relevance of drug-induced erythrocytosis in hematology practice and underscores the importance of recognizing SGLT2is as an increasingly common and clinically significant cause.
We commend Doi et al1 for contributing valuable new evidence to this discussion and advancing a more nuanced, data-informed approach to the evaluation and management of SGLT2i-associated erythrocytosis.
Contribution: B.C.-Y. and C.C.H. wrote the manuscript.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Cyrus C. Hsia, Division of Hematology, Department of Medicine, London Health Sciences Centre, 800 Commissioners Rd E, London, ON N6A 5W9 Canada; email: cyrus.hsia@lhsc.on.ca.
