Background: SGLT-2 inhibitors reduce the risk of hospitalization for heart failure as well as the progression of renal disease, regardless of the presence or absence of diabetes. While several hypotheses exist to explain their cardioprotective effect, the underlying mechanism has not been clarified. In experimental and clinical studies, increased tissue sodium Content led to enhanced hypertrophic stimuli, inflammatory processes as well as to exaggerated vasoconstrictive response in heart failure, thereby increasing afterload. Sodium (23Na)-MRI has been validated in experimental as well as human studies and allows to reliably and accurately quantify tissue (muscle and skin) sodium content.

Aim: We hypothesized that the SGLT-2 inhibitor empagliflozin reduces tissue sodium content in patients with chronic heart failure after 3 months therapy.

Methods: In a randomized, investigator initiated, double-blind, placebo controlled, parallel-group, prospective clinical study (NCT03128528), 74 patients with class II-III heart failure and an ejection fraction of 49% or less were randomized 2:1 to empagliflozin once daily or placebo for 3 months.  In each patient, tissue sodium content of the lower leg was assessed non-invasively by a 3.0T 23Na-MRI at baseline and after 3 months of treatment. In addition, we analyzed metabolic parameters, NTproBNP level, 24-hour ambulatory blood pressure (ABP) and 24-hour urinary sodium excretion.

Results: Our patients (men: n=62), aged 66±9 years, had a median NTproBNP level of 444pg/ml (interquartile range 247-1064pg/ml) and a mean ejection fraction (assessed by echocardiography) of 39±9%. Only 18 of 74 patients had type 2 diabetes. In patients treated by empagliflozin, a significant decrease of skin (22.9±6.1 vs. 21.6±6.1mmol/l, p=0.013) and tibial bone marrow sodium content (3.1±1.7 vs. 2.5±1.4mmol/l, 0.013) was observed after 3 months, while there was no significant change in muscle and water sodium content. In comparison to placebo, the 3-month decrease of skin sodium content was significantly greater in the empagliflozin vs. placebo group (-1.3±3.5 vs. 0.6±3.5mmol/l, p =0.022) and there was a trend towards greater sodium content reduction in tibial bone marrow (-0.6±1.4 vs. 0.1±1.1mmol/l, p =0.074). No significant differences regarding the change in muscle or water sodium content was observed after 3 months treatment with emapgliflozin compared to placebo. After 3 months of treatment with empagliflozin, fasting plasma glucose and 24-hour urinary glucose excretion showed the expected differences to placebo treatment and systolic as well as diastolic 24-hour ABP was lower in the empagliflozin than in the placebo group.

Conclusion: SGLT2-inhibition with empagliflozin significantly reduces skin sodium content after 3 months of treatment in patients with chronic heart failure. We believe that the decrease of tissue sodium content may be one of the mechanisms that lead to improved outcome in patients with chronic heart failure and reduced ejection fraction when treated with SGLT-2 inhibitors.