Discontinue Farxiga (Dapagliflozin) When Dialysis Is Initiated
Once a patient reaches end-stage renal disease and starts dialysis, dapagliflozin should be discontinued, as there is no established benefit and the drug was not studied in dialysis patients during the pivotal trials.
Rationale Based on Guidelines and FDA Labeling
The 2022 ADA/KDIGO consensus guideline explicitly states that dapagliflozin "may continue if tolerated for kidney and CV benefit until dialysis" 1. This clearly indicates that dialysis initiation is the stopping point for the medication.
The FDA label for Farxiga reinforces this position: "Efficacy and safety trials with FARXIGA did not enroll patients with an eGFR less than 25 mL/min/1.73 m² or on dialysis" 2. Critically, while patients already enrolled in DAPA-CKD and DELIVER trials were not required to discontinue if they progressed to dialysis, this was a trial continuation protocol—not evidence supporting initiation or routine continuation in dialysis patients 2.
Why Continuation Makes No Sense on Dialysis
Loss of Mechanism of Action
Dapagliflozin works by inhibiting SGLT2 in the proximal renal tubules to promote urinary glucose excretion and sodium modulation 2. In dialysis patients, there is no functional tubular system for the drug to act upon—the mechanism of benefit is eliminated.
Pharmacokinetic Considerations
Recent research shows dapagliflozin is minimally dialyzable (only 0.10% of the administered dose recovered in dialysate) with a 26.7% accumulation ratio in dialysis patients 3. While the drug was well-tolerated in this small pharmacokinetic study, this does not translate to clinical benefit and raises concerns about unnecessary drug exposure without therapeutic effect.
No Evidence of Benefit Post-Dialysis
The landmark DAPA-CKD trial demonstrated profound benefits in delaying progression to kidney failure (delaying time to kidney failure by 6.6 years) 4, 5, but these benefits are predicated on having residual kidney function to preserve. Once on dialysis, there is no kidney function left to protect—the primary endpoint has already occurred.
Common Pitfall to Avoid
Do not confuse "patients were not required to discontinue" in the trials with "continuation is recommended." The trial design allowed continuation for data collection purposes, not because of demonstrated benefit in the dialysis population. The cardiovascular benefits seen in DAPA-HF and DELIVER trials 4 were in patients with heart failure who had residual kidney function, not in dialysis-dependent patients.
Clinical Algorithm
At dialysis initiation:
- Discontinue dapagliflozin immediately
- The drug provides no renal protective benefit (kidney failure has occurred)
- The glucose-lowering effect is minimal to absent without functioning kidneys
- For heart failure management in dialysis patients, rely on evidence-based therapies studied in ESKD populations