Background: The benefits of SGLT2i in diabetes mellitus, heart failure and chronic kidney disease have increased their scope and led to widespread use. However, recent evidence linked SGLT2i to polycythaemia and erythrocytosis. Our aim is to study this effect and assess the safety of SGLT2i use.

Methods: We retrospectively analysed data from 230 patients with diabetes receiving treatment with SGLT2i and measured its effect on haematocrit. A cohort of patients within polycythaemia range was identified and studied for parameters including haemoglobin (Hb), Hct, renal function and incidence of diabetic ketoacidosis (DKA), major adverse cardiovascular events (MACE) and urinary tract infections (UTI). Data was also collected for potential confounders affecting Hct and Hb.

Results: Mean baseline Hct was 41.3% (n=230). This increased to 42.3% (MD 1%; p<0.001) and 42.6% (MD 1.3%; p<0.001) at 3 and 6-12 months. The prevalence of polycythaemia range Hct was 3.5% (n=8) with 62.5% (n=5) newly developed. In the high Hct cohort, mean baseline Hct was 48.6% (n=8). This increased to 49.5% (MD 0.9%; p=0.38) and 49.6% (MD 1%; p=0.34) at 12 and 18-24 months respectively. Mean Hb was 163.5, 166.9 (MD 3 g/L; p=0.31) and 166.9 g/L (MD 4.2 g/L; p=0.31) at baseline, 12 and 18-24 months respectively. Mean eGFR was 75.8, 71.9 (MD -3.9; p=0.17) and 76.3 ml/min/1.73m2 (MD 0.5; p=0.82) at baseline, 12 and 18-24 months respectively. Mean urea was 6.23, 7.83 (MD 1; p=0.009) and 5.45 mmol/L (MD -0.78; p=0.01) at baseline, 12 and 18-24 months respectively. DKA was observed in 25% (n=2), UTIs in 37.5% (n=3), MACE in 37.5% (n=3) and venesection was required for one (n=1) patient in the polycythaemia range cohort over the treatment period.

Conclusion: SGLT2i treatment led to small but statistically significant increase in Hct, which persisted over a 6-12 month period and 18-24 months in the high Hct cohort. The prevalence of polycythaemia range Hct was 3.5% and new in most cases. These patients had high incidence of MACE and DKA.

Disclosure

P. Pavlou: None.

© 2024 by the American Diabetes Association
2024
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.