GLP-1 in Dialysis and Stage 5 CKD: What the Evidence Shows

Stage 5 CKD and dialysis patients sit in a clinical blind spot for GLP-1 therapy. They are excluded from almost every pivotal trial — STEP-1, SURMOUNT-1, SELECT, and even FLOW (Perkovic 2024 NEJM^[2]) which enrolled CKD but capped at eGFR 25 mL/min/1.73m². The published evidence in dialysis itself is one randomized PK study (Idorn 2016 Diabetes Care^[1]) plus a handful of retrospective cohorts (Long 2024 Endocrine Practice^[3]) and narrative reviews (Clemens 2023^[4], Thomas 2023 ^[5]). Yet the population is large, growing, and carries two competing pressures: a transplant-list BMI ceiling that can be lowered by GLP-1 weight loss (Gill 2014 CJASN ^[10]), and a protein-energy wasting phenotype (Fouque 2008^[6]) that makes rapid weight loss dangerous. This article walks through the PK, the evidence, the nutritional guardrails, and the practical dosing approach.

The honest summary

• The pivotal trials excluded dialysis. SURMOUNT-1, STEP-1, SELECT, and FLOW (Perkovic 2024 ^[2]) all excluded patients on renal replacement therapy. The only RCT in true ESRD is Idorn 2016 ^[1] — a 12-week investigator-initiated liraglutide study in 24 hemodialysis patients.
• PK is favorable for semaglutide, liraglutide, and dulaglutide. These three are highly protein-bound, metabolized by endopeptidases rather than renally cleared, and are not removed by hemodialysis. Exenatide is the opposite (~80% renal clearance) and is contraindicated in dialysis per its label.
• Protein-energy wasting is the dominant safety concern. The ISRNM diagnostic criteria (Fouque 2008 ^[6]) and the KDOQI Nutrition 2020 update (Ikizler 2020 AJKD^[7]) require a minimum protein intake of 1.0–1.2 g/kg/day on dialysis. GLP-1-induced appetite suppression can drop intake below that floor quickly.
• Transplant candidacy is the strongest indication. Most U.S. kidney programs decline candidates with BMI > 35–40 (Gill 2014 CJASN^[10]). GLP-1 weight loss can move a patient from non-listable to listable status — an outcome with mortality benefit far larger than any GLP-1 trial endpoint.

What FLOW did and did not tell us

FLOW (Perkovic 2024 NEJM^[2]) is the largest GLP-1 trial in CKD: 3,533 patients with type 2 diabetes and established kidney disease randomized to semaglutide 1.0 mg weekly or placebo, with a primary composite of major kidney disease events. The trial was stopped early for efficacy; semaglutide reduced the primary endpoint by 24% (HR 0.76, 95% CI 0.66–0.88). That headline has been used to justify GLP-1 prescribing across the CKD spectrum, including in dialysis. The trial does not support that.

FLOW enrolled patients with eGFR 25–75 mL/min/1.73m² and explicitly excluded those on dialysis. None of the kidney events that drove the primary endpoint occurred in patients already on renal replacement therapy. The mechanistic case for benefit in dialysis is plausible — semaglutide reduces albuminuria, blood pressure, and weight, all of which affect cardiovascular outcomes — but it has not been tested.

The pharmacokinetic picture in dialysis

Three properties of a GLP-1 receptor agonist determine whether it can be used in dialysis without dose adjustment: degree of renal clearance, protein binding, and removal by dialysis.

• Semaglutide. Greater than 99% protein-bound; minimal renal clearance (proteolysis is the dominant elimination route); not removed by hemodialysis. The Long 2024 Mayo Clinic cohort (Endocrine Practice^[3]) followed semaglutide use in advanced renal failure including dialysis and reported a tolerability profile broadly consistent with the general T2D population.
• Liraglutide. Protein-bound, no renal clearance, not removed by hemodialysis. The Idorn 2016 Diabetes Care RCT^[1] randomized 24 hemodialysis patients with T2D to liraglutide or placebo for 12 weeks. Steady-state plasma concentrations were 49% higher in the ESRD group than in matched controls, supporting starting at the lowest dose (0.6 mg) and titrating cautiously.
• Dulaglutide. Hepatic clearance via proteolysis; not renally cleared in any meaningful fraction. Used off-label in dialysis case series with no dose adjustment specified.
• Tirzepatide. PK profile similar to semaglutide — protein-bound, minimal renal clearance, not dialyzable. Limited published dialysis-specific data but the labeling does not require dose adjustment in renal impairment.
• Exenatide. ~80% renal clearance. Half-life extends from ~2 hours in normal renal function to > 6 hours in severe impairment. The FDA label contraindicates use in ESRD (CrCl < 30 mL/min) due to accumulation and increased GI adverse events. Avoid in dialysis.

Protein-energy wasting and why rapid GLP-1 weight loss can hurt

Protein-energy wasting (PEW) is a clinically distinct syndrome on dialysis: low serum albumin, low BMI, unintentional weight loss, reduced muscle mass, and reduced protein/energy intake (Fouque 2008 Kidney International^[6], the ISRNM consensus that remains the canonical diagnostic framework). PEW prevalence in maintenance dialysis is roughly 28–54% depending on the criteria used; mortality is approximately double that of dialysis patients without PEW.

The mechanism that helps GLP-1 patients lose fat — appetite suppression — is the same mechanism that precipitates PEW in dialysis. The KDOQI Nutrition 2020 update (Ikizler 2020 AJKD^[7]) recommends 1.0–1.2 g/kg/day of dietary protein for adults on maintenance hemodialysis or peritoneal dialysis (higher than the 0.6–0.8 g/kg recommendation in non-dialysis CKD). A 70 kg dialysis patient needs 70–84 g of protein per day. GLP-1-induced appetite suppression can cut that intake in half within weeks. Baseline albumin trending below 3.8 g/dL, or any decline of > 0.3 g/dL between dialysis-day measurements, is the practical trigger to slow titration or hold the dose.

The obesity paradox in dialysis

Higher BMI is associated with better survival on maintenance dialysis — the so-called obesity paradox documented across DOPPS and USRDS cohorts and reviewed by Kalantar-Zadeh and Kopple in 2006^[9]. The mortality benefit attributed to higher BMI in dialysis is concentrated in patients with preserved muscle mass; when sarcopenic obesity is separated out, the paradox attenuates substantially.

The practical implication is not that GLP-1 weight loss is forbidden — it is that the goal is fat loss with muscle preservation, not aggressive total body-weight reduction. A dialysis patient with BMI 38 who is otherwise transplant-list eligible may benefit from getting to BMI 32–33 (the usual program ceiling). A dialysis patient with BMI 32 and baseline sarcopenia probably should not be on a GLP-1 at all unless transplant access depends on it.

Transplant-list eligibility: the strongest indication

Most U.S. kidney transplant programs decline candidates with BMI > 35–40 because of higher rates of surgical complications and delayed graft function. Gill 2014 (CJASN ^[10]) showed that BMI 35–40 reduced the rate of transplant access by roughly 55% relative to normal-BMI candidates, with even larger reductions above BMI 40. The mortality benefit of receiving a kidney transplant versus remaining on dialysis is large — the most-cited long-term estimate is a 50–70% reduction in mortality across the first decade post-transplant. That benefit dominates any GLP-1 trial endpoint and is the reason transplant centers have begun referring listed-but-overweight candidates for GLP-1 therapy.

The Clemens 2023 narrative review^[4] and the Long 2024 Mayo cohort^[3] both describe semaglutide use in this exact transplant-candidacy indication. Long 2024 followed 38 patients with advanced kidney disease (including dialysis-dependent) treated with semaglutide; the cohort achieved meaningful weight loss with a tolerability profile consistent with the general T2D population, and several patients were successfully relisted after reaching BMI < 35.

Magnitude: weight change at 6 months in dialysis patients with obesity

Practical dosing in dialysis

1. Confirm the indication. Transplant-list candidacy is the strongest. Glycemic control in T2D on insulin is a reasonable secondary indication per KDIGO 2022 (Rossing 2022^[8]). Pure weight loss for cosmetic or comfort reasons is hard to justify against PEW risk.
2. Confirm the agent. Semaglutide, liraglutide, dulaglutide, or tirzepatide. Never exenatide in dialysis (FDA-label contraindication).
3. Baseline nutrition assessment. Albumin, pre-dialysis weight trend over the prior 3 months, dietary recall, grip strength if available. Engage the dialysis unit dietitian before starting.
4. Start low, titrate slow. Semaglutide 0.25 mg weekly for 8 weeks (not 4); liraglutide 0.6 mg daily for 4 weeks before any uptitration; tirzepatide 2.5 mg weekly for 8 weeks. The Idorn 2016 finding of 49% higher steady-state liraglutide concentrations in ESRD^[1] is the rationale for slower titration.
5. Inject between dialysis sessions. No mechanistic reason to coordinate with the dialysis schedule, but most patients prefer non-dialysis days for home injection.
6. Hold for severe interdialytic dehydration. Persistent vomiting or volume overload between sessions warrants temporary discontinuation. Coordinate with the nephrology team on dry weight reassessment.
7. Monitor monthly. Pre-dialysis weight trend (target a slow, steady loss, not a steep one), albumin (hold for > 0.3 g/dL drop), GI tolerability, T2D patients on insulin for hypoglycemia (especially on dialysis days when carbohydrate intake is reduced).

Insurance and access

Patients with ESRD qualify for Medicare regardless of age under the ESRD entitlement. Medicare Part D covers GLP-1s for T2D indications (Ozempic, Mounjaro, Trulicity, Victoza, Bydureon) with the usual prior authorization and step therapy. Coverage for obesity-only indications (Wegovy, Zepbound) remains limited by the historic Medicare exclusion on weight-loss drugs — though the 2026 policy environment is in flux. Transplant programs occasionally secure peri-transplant GLP-1 coverage as part of bundled transplant readiness care.

Related research and tools

• FLOW trial and GLP-1 kidney protection — the Perkovic 2024 NEJM trial in CKD (eGFR 25–75)
• GLP-1 with transplant immunosuppressants — tacrolimus, cyclosporine, mTOR inhibitors, gastric stasis stacking
• GLP-1 in older adults at sarcopenia risk — many dialysis patients sit in this overlap population
• GLP-1 muscle loss prevention protocol — the protein and resistance-training playbook for preserving lean mass