Sodium-glucose co-transporter 2 inhibitors (SGLT2i) are standard-of-care treatment in heart failure (HF). The risk of ketosis in patients with HF is unclear, especially during hospitalisation. We aimed to evaluate the normal ketone concentration range in HF patients. We performed a cross-sectional study of inpatients with acutely decompensated HF and outpatients with stable HF. Ketone concentrations were measured and analysed based on SGLT2i use. Baseline demographic data (age, gender, body mass index [BMI]), time since last meal, HF type, type 2 diabetes status, insulin use, and blood parameters (creatinine, glycosylated haemoglobin A1c [HbA1c] and N-terminal pro-B-type natriuretic peptide) were collected from patients or medical records. The primary outcome was capillary blood ketone concentration in patients with acute decompensated HF and stable chronic HF stratified by SGLT2i use. Multivariate regression was also performed using ketones as the outcome variable, with age, gender, BMI, glucose levels, HbA1c, time since last meal and presence of insulin therapy as predictor variables. A total of 20 individuals with decompensated HF (n=5 SGLT2i treated) and 47 with stable chronic HF (n=22 SGLT2i treated) were recruited. Median ketone concentrations were similar in all groups irrespective of SGLT2i use and the presence of acute decompensation (0.1 mmol/L, biggest interquartile range 0.2 mmol/L, p=0.49). Apart from time from last meal, multivariate regression analysis showed no association of ketone concentration with SGLT2i use, age, gender, BMI, type 2 diabetes status, insulin use and blood glucose level. Ketone concentrations were low in individuals with HF regardless of SGLT2i use or the presence of acute decompensation.
Read full abstract