The aim of this study is to learn about the safety of empagliflozin in dialysis patients as a preparation for a future large clinical trial. Empagliflozin has been approved by the Food and Drug Administration for the treatment of either type 2 diabetes, heart failure, or chronic kidney disease among patients not on dialysis. The use of empagliflozin has not been studied or approved among patients on dialysis for kidney failure because empagliflozin acts on the kidneys. However, recent experimental studies have indicated that empagliflozin may provide direct heart benefits. Some dialysis patients have substantial residual kidney function, which may be protected by empagliflozin.

Participants will be given empagliflozin for three (3) months on top of the standard of care (usual medical care for participants' condition) and will be followed up until one (1) month after the last dose. The investigators will collect information about participants' general health, obtain blood, urine, and imaging studies, check home blood pressure, monitor home blood sugar levels, and ask health-related questions to assess the safety and potential benefits of empagliflozin over four (4) months, including one month before the three (3)-month empagliflozin treatment.

The incidence of end-stage kidney disease (ESKD) in the US ranks among the highest in the world. ESKD is the last phase of chronic kidney disease when the kidneys are functioning below 10-15% of normal capacity, and the patient is on dialysis. According to the US Renal Data System (USRDS), 120,834 individuals started dialysis and nearly 524,000 people were living on dialysis in 2017.1 Although advancement in technology and general medical care has led to a modest decrease in mortality among dialysis patients, their mortality rate remains extremely high at approximately 16.5 per 100 patient-years. The leading cause of death among dialysis patients is cardiovascular disease (CVD), accounting for almost 45% of deaths. Unfortunately, established therapies to prevent incident CVD in the general population, such as renin-angiotensin system inhibitors or statins, have not been shown to be effective in the dialysis population.

Sodium-glucose transporter type 2 (SGLT2) inhibitors are originally approved by FDA for the treatment for type 2 diabetes. SGLT2 is localized to the brush border of the early proximal tubule, and hence, SGLT2inhibitors induce osmotic diuresis and natriuresis but do not activate the systemic renin-angiotensin-aldosterone system.2 Recent clinical trials have consistently shown their potent renal and cardiovascular benefits in both diabetic and non-diabetic patients, which cannot be explained only by their glucose-lowering and diuretic properties. In fact, diuretics have not been shown to reduce cardiovascular mortality and such benefits of SGLT2 inhibitors are clear even among non-diabetic populations.3-5 Their renoprotective effect potentially extends to the dialysis population where residual kidney function (RKF) still plays a major role in solute clearance and volume control and has a strong association with patient outcomes.6 Patients who retain greater RKF can consume a more liberal diet and have better nutritional status, less pill burden, better blood pressure, and less interdialytic fluid gain with less frequent intradialytic hypotension, as well as greater quality of life and better survival.6 The pathophysiology underlying the cardiovascular benefits of SGLT2 inhibitors are yet to be fully elucidated, but a recent in-vitro studies indicate its direct effects on cardiomyocytes. Therefore, the investigators hypothesize that dialysis patients also benefit from SGLT2 inhibitors even if they do not have any RKF.

Efficacy and safety studies with SGLT2 inhibitors did not enroll end-stage kidney disease (ESKD) patients on dialysis. Empagliflozin, canagliflozin, and dapagliflozin can be started if the glomerular filtration rate is more than 20-25 mL/min per 1.73 m2 and can be continued until dialysis initiation or kidney transplant. From a pharmacokinetics standpoint, those SGLT2 inhibitors are extensively metabolized by glucuronidation into inactive metabolites, and are not likely to cause dose-dependent toxicity even in ESKD. Nevertheless, extra caution is necessary for their use in the setting of ESKD because SGLT2 inhibitors are not well dialyzable due to large distribution volumes and high protein binding rates.

Our overall goal is to conduct a non-randomized feasibility clinical trial of empagliflozin in the dialysis population to obtain data that will help plan future larger, sufficiently powered efficacy clinical trials. The investigators plan to enroll a total of 24 dialysis patients (18 patients on hemodialysis and 6 patients on peritoneal dialysis). After one month of the run-in period, participants will take oral empagliflozin for 3 months.

*Hemodialysis is a form of renal replacement therapy that utilizes an external filter (dialyzer) to remove wastes from the bloodstream. Peritoneal dialysis utilizes the peritoneum as a filter to remove wastes.

Inclusion Criteria:

1. age ≥18 years;
2. diagnosis of end-stage kidney disease requiring dialysis, and
3. ability to provide informed consent.

Exclusion Criteria:

1. systolic blood pressure <100 mm Hg (pre-dialysis for HD patients)
2. two or more episodes of urinary tract infection within the last 12 months
3. history of urinary retention or urinary tract obstruction
4. liver cirrhosis
5. advanced heart failure requiring heart assist device or inotropic support
6. heart or liver transplant recipient
7. major surgery performed within the last 3 months ("major" per the investigator's assessment)
8. major surgery scheduled within 3 months after screening ("major" per the investigator's assessment)
9. active cancer
10. pregnant or lactating women
11. known allergy or hypersensitivity to any SGLT2 inhibitors
12. history of ketoacidosis during the last 12 months
13. any other medical condition considered unappropriated by their nephrologists or a study physician (i.e., cachexia, short life expectancy, or uncontrolled personality/phycological disorder).

The data that support the findings of this study will be available on request from the corresponding author, Dr. Yoshitsugu Obi, MD. The data are not publicly available due to restrictions e.g. their containing information that could compromise the privacy of research participants.

Data will become available upon request after one month until 60 months of the publication of the manuscript.