SGLT-2 Inhibitor-induced erythrocytosis and risk of coronary artery disease in type 2 diabetes - a nationwide cohort study

Aidar Gosmanov; Darren Gemoets

doi:10.1530/endoabs.110.P455

JOINT608

Purpose of Study: Therapy of type 2 diabetes (T2D) with SGLT-2 inhibitors can lead to increased risk of new-onset erythrocytosis. On the other hand, studies conducted in pre-SGLT-2i era showed that incident erythrocytosis in non-diabetic individuals heightens cardiovascular risk. The aim of this investigation was to assess cardiovascular risk in persons who develop SGLT-2i-induced erythrocytosis.

Methods: This was a retrospective nationwide cohort study of US Veterans with T2D who received care between 3/2013–12/2023. Inclusion criteria were complete medical records data, normal baseline hematocrit (Hct)<50. 0% within 1 year before the study start, adequate adherence with SGLT-2i, no testosterone prescription, and no history of erythrocytosis or cardiovascular disease. We identified 2 groups of patients: control (no record of SGLT-2i use) and new SGLT-2i users; in all patients we followed hematocrit for 1 year to identify those who developed new erythrocytosis defined as hematocrit≥50%. Primary endpoint was odds ratio (OR) of incident non-fatal coronary artery disease (CAD) calculated by logistic regression model adjusted for case-mix.

Results: We identified 691104 patients mostly males (94. 5%) and Caucasians (Whites 70. 2% and Blacks 19. 4%) with mean age 65. 7yrs. The SGLT-2i use for 1 year resulted in significant HbA1c reduction and hematocrit increase compared with the control group (Table). In adjusted analyses, absolute incidence of erythrocytosis was about 5. 0-fold higher (P < 0. 0001) in SGLT-2i-treated than in control patients (10. 6% vs 2. 3%). In the models adjusted for baseline hematocrit, HbA1c, age, ethnicity, sex, BMI, eGFR, sleep apnea, concomitant hypoglycemic agent use, and smoking status, new erythrocytosis was associated with equally higher probability of incident CAD in control and SGLT-2i-treated patients (Table).


Study Outcomes	Control, Hct<50% (n = 605311)	Control, Hct≥50% (n = 14516)	SGLT-2i, Hct<50% (n = 63698)	SGLT-2i, Hct≥50% (n = 7579)
HbA1c, % (Mean±SD): At baseline	6. 9±1. 4	6. 9±1. 6	8. 3±1. 5	8. 4±1. 5
Δ after 1 yr	-0. 05±1. 1	-0. 04±1. 2	-0. 91±1. 5	-1. 06±1. 6
Hct, % (Mean±SD): At baseline	41. 6±4. 2	46. 5±3. 1	41. 7±3. 8	46. 1±2. 5
Δ after 1 yr	-0. 3±3. 3	2. 4±4. 3	2. 0±3. 2	4. 4±3. 2
CAD event OR (95% CI), adjusted	1. 0 (ref)	1. 76(1. 47, 2. 10)	0. 98(0. 90, 1. 06)	1. 36 (1. 10, 1. 66)

Conclusion: For the first time, we demonstrated that in a large T2D cohort, a new diagnosis of erythrocytosis is associated with incident CAD and SGLT-2i therapy does not offer cardiovascular protection in such patients. Considering markedly increased erythrocytosis incidence in the SGLT-2i-treated patients which in turn can lead to larger number of cardiovascular events, we recommend implementation of strategies to monitor hematocrit trends in T2D patients and particularly in individuals starting gliflozin therapy.

Endocrine Abstracts

P455