Friday, April 10th 2026, 12 pm Eastern- Space on X

Contemp Clin Trials. 2025 Jun:153:107911. doi: 10.1016/j.cct.2025.107911. Epub 2025 Apr 6.

System Interventions to Achieve Early and Equitable Kidney Transplants (STEPS): Protocol for STEPS, a randomized comparative effectiveness clinical trial

L Ebony Boulware, Patti L Ephraim, Tariq Shafi, Jamie A Green, Teri Browne, Tara S Strigo, Sarah Peskoe, Jonathan Wilson, Yuliya Lokhnygina, Aviel Alkon, George L Jackson, Matthew J Ellis, Debra Sudan, Blake Cameron, Pradeep K Vaitla, Ashley Cabacungan,  Lauren Brubaker, Emily L Obermiller, Clarissa J Diamantidis

PMID: 40199386

#NephTrials is an ongoing initiative between #NephJC and the ISN-ACT group. The ISN-ACT (Advancing Clinical Trials) is an International Society of Nephrology (ISN) initiative to encourage existing infrastructures within ISN to improve the global nephrology community’s participation in clinical trial research.

Introduction

Kidney transplantation remains the optimal therapy for kidney failure, yet access is neither early nor equitable. Despite decades of progress, only a minority of patients reach the transplant pathway before dialysis, and even fewer complete the steps required to receive a transplant. This gap reflects a fragmented process: patients are often identified late, inadequately informed, inconsistently referred, and infrequently able to complete the complex evaluation processes. As highlighted in the STEPS protocol, transplantation is not a single event but a multistep pathway, and failure at any single step can prevent patients from ever reaching the waitlist (Boulware et al,.Contemp Clin Trials, 2025).

Non-medical barriers, including health literacy, provider communication, system capacity, and logistical constraints, play a central role in limiting access to early transplant evaluation (Harding et al, Transplant Rev, 2021). At the same time, structural inequities remain deeply embedded in the system. Community-level vulnerability independently reduces access to living donor kidney transplantation (LDKT) and only partially explains racial disparities, with disproportionately worse effects among Black recipients (Killian et al, JAMA Surg, 2021). Meanwhile, these observations reinforce the need to move beyond simply describing inequities toward actively addressing them, as emphasized by Patzer, who highlighted bridging the gap between improved outcomes and unequal access to transplantation (Patzer R, JASN, 2024). More recently, these disparities have been reframed through an intersectional lens, recognizing how overlapping systems—race, geography, socioeconomic status, and healthcare infrastructure—interact to shape access long before transplant listing is even considered (Nonterah CW, Transpl Int, 2024).

Interventions targeting single components of the pathway have produced modest and inconsistent gains. Even multicomponent strategies, such as the pragmatic EnAKT LKD cluster randomized trial, improved selected steps but did not increase overall progression towards transplantation, underscoring the cumulative nature of barriers (Garg et al, JAMA Internal Medicine, 2023| NephJC summary | FF episode).

The STEPS trial (System Interventions to Achieve Equitable and Early Transplants) emerges in response to this gap. The trial tests a coordinated system-level intervention that combines automated electronic health record surveillance with proactive outreach and longitudinal patient navigation. Rather than improving isolated components, STEPS evaluated whether redesigning the transplant pathway itself can increase completion of the transplant evaluation and reduce disparities in access to kidney transplantation. 

How do you test a system-level intervention? The STEPS Trial Design

STEPS is a pragmatic, randomized comparative effectiveness trial designed to evaluate whether a coordinated health system intervention can improve access to kidney transplantation. The study examines whether redesigning how patients are identified, referred, educated, and navigated can change progression through the transplant pathway. The unit of intervention is the system, but the unit of inference is the patient, achieved through individual randomization embedded in routine care.

The STEPS intervention combines active electronic health record (EHR)–based surveillance with patient-centered outreach delivered by transplant social workers and coordinators. The surveillance registry was updated nightly within the EPIC electronic health record, identifying adults aged 18-74 years with CKD using objective criteria (two eGFR values < 60 ml/min/1.73 m²), followed by risk-based triggers: alerts at eGFR <30 ml/min/1.73 m² or ≥ 10% two-year risk using the Kidney Failure Risk Equation, and transplant referral prompts at eGFR ≤ 20 ml/min/1.73 m². This approach standardizes candidate identifications and reduces reliance on clinician recognition- one of the major sources of variability in previous studies. This multilevel approach targets key barriers across the transplant pathway- particularly among Black and rural individuals with advanced chronic kidney disease who are not yet on dialysis.

Importantly, the trial was conducted across three large and diverse U.S. health systems: Duke University Health System (DUHS), Geisinger Health System (GHS), and the University of Mississippi Medical Center (UMMC)- selected to capture different structural barriers to transplant access. DUHS and UMMC care for a high proportion of Black patients and rural populations, groups historically underrepresented in transplant access. Black individuals accounted for over 50% of kidney transplant recipients at DUHS and more than 70% at UMMC. In contrast, GHS has a highly rural population, enabling the study to also evaluate geographic disparities independent of race. By embedding the intervention within health systems that reflect real-world inequities in race and geography, STEPS is positioned not only to evaluate effectiveness but also to generate insights into how system-level interventions perform across populations disproportionately affected by limited access to transplantation. 

Can redesigning the health system improve access to kidney transplantation?

Following identification through the surveillance registry and enrollment, participants enter a structured pathway that reflects both methodological rigor and real-world applicability. Randomization was performed using a centralized, stratified block design, accounting for clinical site, race, and kidney function (eGFR ≤20 vs >20 ml/min/1.73m²), ensuring balance across factors closely linked to disparities in transplant access.

The trial successfully completed enrollment of 1,168 participants between March 2022 and March 2024. The trial was registered at ClinicalTrials.gov (NCT05014256). The planned sample size of approximately 1150 participants is expected to provide >80% power to detect a treatment effect for the primary outcome overall and specifically among Black participants, with conservative assumptions based on usual-care estimates across participating sites.

Once randomized, participants followed one of two care pathways. Participants assigned to arm 1, augmented usual care, continued to receive standard clinical management supported by EHR-based surveillance and best practice alerts, prompting clinicians to initiate nephrology referral and transplant evaluation when guideline thresholds are met. Participants in arm 2, STEPS intervention arm, receive a coordinated, multilevel intervention that integrated EHR-based identification, structured education, proactive outreach, and longitudinal patient navigation, designed to address barriers that traditional care does not systematically overcome.

A central component of STEPS is the integration of patient-centered educational and behavioral support strategies delivered by transplant social workers and nurse coordinators. Participants receive the “Living with Kidney Disease: All the Facts” booklet, developed to provide accessible, literacy-sensitive information about CKD progression and treatment options (Boulware et al, The PREPARE NOW Study, 2023). This is complemented by the TALK Social Worker Program, a structured intervention grounded in behavioral theory, which facilitates discussions about kidney transplantation, particularly LDKTs, among patients, their families, and providers (Boulware et al, AJKD, 2013). Through individualized and group sessions, the program helps patients identify personal, social, and logistical barriers while promoting informed decision-making and engagement in the transplant process.

Beyond education, STEPS incorporates active patient navigation, a critical element often missing in traditional care models. Transplant social workers and nurse coordinators maintain longitudinal contact with participants, assisting with scheduling, coordinating appointments, addressing financial and transportation barriers, and guiding patients through each step of the transplant evaluation. This coordinated approach reduces reliance on patient self-navigation, a well-recognized contributor to dropout along the transplant pathway.

Data collection in STEPS reflects its hybrid nature as both a clinical and behavioral intervention. Outcomes are assessed using electronic health records, administrative data, and centralized patient-reported measures collected at baseline and at 6, 12, and 18 months. The recruitment was centralized, having a dedicated data coordinating center to standardize the outcome ascertainment and minimize loss of follow-up.

The primary outcome is completion of the kidney transplant evaluation, defined using objective health system documentation, including evaluation completion, time to completion, listing decisions, or receipt of transplantation. This endpoint targets a critical bottleneck in the transplant pathway while avoiding confounding from organ availability. Secondary outcomes expand this perspective by examining intermediate steps along the pathway, including patient discussions with providers and family members, initiation and progression of transplant evaluation, and identification of potential living donors. Additionally, measures of patient knowledge, empowerment, and behavioral activation are included to better understand how the intervention influences decision-making and engagement.

Together, this design allows STEPS to move beyond traditional endpoints and evaluate whether a coordinated system-level intervention can meaningfully change both the trajectory and the experience of patients navigating the transplant pathway.

Statistical analysis follows an intent-to-treat framework, preserving randomization and minimizing bias from differential engagement with outreach and navigation. The primary comparison evaluates completion of transplant evaluation between groups using regression models adjusted for stratification variables (site, race, and baseline kidney function). Time-to-event analyses will assess progression through the pathway, accounting for variable follow-up. Secondary outcomes will be analyzed using generalized linear models appropriate for outcome type, with prespecified subgroup analyses by race, rural residence, and kidney function to examine effects on disparities. Missing patient-reported outcomes will be handled using multiple imputation, and sensitivity analyses will evaluate robustness to differential follow-up.

The study was funded by the Patient-Centered Outcomes Research Institute (PCORI) and the National Institute of Diabetes and Digestive and Kidney Disease (NIDDK), with total project costs of approximately 12.95 million $, split between NIH (66%) and PCORI (54%). Investigators reported limited competing interests: one author served as an editor for the publishing journal, and another disclosed a consulting relationship and financial interest in a health coaching company.

Why STEPS Matters in Context?

In the absence of outcome data from STEPS, prior interventions offer important insights into bridging LDKTs disparities. Across studies, a consistent pattern emerges: interventions that target a single barrier can improve individual steps in the transplant pathway but rarely turn into sustained, system-wide gains.

Educational strategies, particularly when structured and guided, have demonstrated improvements in knowledge, readiness, and evaluation completion. For example, standardized transplant education programs have been associated with reduced disparities in access, while tailored interventions have improved patient engagement and decision-making (Patzer et al, CJASN, 2012; Arriola et al, Prog Transplant, 2025). Similarly, the navigator-based and culturally tailored programs have demonstrated meaningful benefits in donor identification and evaluation completion, particularly in disparity reductions. However, these effects are highly context-dependent and often difficult to reproduce at scale (Sullivan et al, CJASN, 2018). This variability becomes more apparent in larger pragmatic studies. Multisite navigator trials and system-wide interventions, such as EnAKT (Garg et al, JAMA Internal Medicine, 2023|  NephJC summary) have demonstrated that even well-designed multicomponent strategies may not improve overall progression toward transplantation when implemented across heterogeneous settings. 

Importantly, several studies suggest that interventions may have a greater impact among populations experiencing baseline disparities, indicating that targeting structural barriers can both improve access and advance equity. However, these gains remain inconsistent, underscoring the need for approaches that are multilevel, integrated, and sustained across the entire care pathway (Sullivan et al, CJASN, 2018;  Garg et al, JAMA Internal Medicine, 2023; Bailey et al, Wellcome Open Research, 2024). 

STEPS represents a methodological shift. Rather than testing education, navigation, or referral improvement in isolation, the trial integrates automated identification, augmented usual care, structured outreach, and longitudinal navigation within a single coordinated framework. The design targets multiple failure points along the transplant pathway (candidate identification, referral, evaluation completion, and donor engagement), while embedding the intervention within routine care workflows across diverse health systems.

The transplant pathway is not defined by a single decision point, but by a sequence of interdependent steps shaped by clinical, social, and structural factors. STEPS represents a critical evolution in trial design. Rather than focusing on isolated components of care, it tests whether aligning early identification, education, referral, and navigation within a coordinated system can meaningfully change how patients move through the transplant pathway.

The question is no longer whether individual interventions can work, but whether the system itself can be redesigned to deliver them effectively, equitably, and at scale—ultimately addressing the persistent burden of CKD and the ongoing disparities in access to kidney transplantation.

Join us this week on the X-space to explore this trial design with Dr Boulware and the ISN-NephJC team, on April 10, 12 pm EST! 

Summary prepared by
Milagros Flores, Cristina Popa

Reviewed by 
Brian Rifkin, Swapnil Hiremath

Tuesday, April 7th 2026, 9pm Eastern

NEJM Evid. 2026 Mar;5(3):EVIDoa2500086. doi: 10.1056/EVIDoa2500086. Epub 2026 Feb 24.

A Randomized Trial of Targeted Hyponatremia Correction in Hospitalized Patients

Julie Refardt, Laura Potasso, Anissa Pelouto, Moritz Trappe, Claudia Gregorian, Markus Koster, Ivana Dora Vodanovic, Dario Norello, Svenja Ravioli, Sadrija Cukoski, Maria Boesing, Basil Ryser, Lana Sambula, Nikola Rapsch, Sophie Monnerat, Julia Beck, Sven Lustenberger, Deborah R Vogt, Laura Werlen, Joyce Santos de Jesus, Martina Bontognali, Philipp Schuetz, Adrienne A M Zandbergen, Alessandro Peri, Darko Kastelan, Gregor Lindner, Joerg Leuppi, Stefan Bilz, Beat Mueller, Volker Burst, Ewout J Hoorn, Mirjam Christ-Crain

PMID:  41733398

DOI: 10.1056/EVIDoa2500086

Introduction

Don’t get salty, but you don’t know as much about sodium as you think you do. 

In 2017, you knew that increased sodium intake leads to increased urine sodium output. Then, Dr. Roger Rodby opined about why everything we know about sodium might be wrong (Some Thoughts on Spooky Sodium| NephJC). Ninety-eight percent of total-body sodium is confined in the extracellular fluid (ECF) compartment in young, healthy humans. However, about 80% of exchangeable sodium is found in the interstitial and connective tissues, and only about 15% of exchangeable sodium is in plasma. When dietary salt intake increases, urinary sodium excretion increases, but it does not match intake immediately and thus generates a positive sodium balance until excretion again equals intake. Increasing NaCl intake also increases body water in most situations, although this does not result predominantly from more fluid consumption but rather from fluid retention. When dietary salt intake rises from low to moderate levels, body water increases. When salt intake rises further, sodium accumulation may occur without an increase in water (Rokova et al, J Clin Invest, 2017).

In 2021, you thought you knew that increased sodium intake led to increased thirst and elevated blood pressure. Doctors Ellison and Welling looked at controversies in salt handling and blood pressure. Increased salt doesn’t always lead to increased thirst but rather increased hunger and higher calorie burn. Over-restricting salt may lead to insulin resistance and higher blood pressure. The mechanistic basis for the link between ECF volume and vascular tone remains contentious. Drinking saline solutions to increase salt intake increases protein catabolism, glucocorticoid production, and urea generation, whereas salt loading through diet, with access to free water, does not. (Ellison et al, NEJM, 2021)

In 2022, you knew that lowering salt intake was better for heart failure and hypertensive patients. However, implemented dietary counseling (reducing sodium below 2 grams daily) did not show added benefit in the largest trial performed in this population (Ezekowitz et al, Lancet, 2022| SODIUM-HF NephJC). Don’t even get me started on fluid restriction in heart failure- hint: it isn’t helpful, stop torturing your patients (Hermann et al, Heart, 2026).

In 2023, sodium correction limits were emphasized to reduce the risk of osmotic demyelination syndrome. However, Seethapathy et al. showed that slower sodium correction rates were associated with higher mortality and longer hospital stays. Because the study evaluated observed correction rates rather than treatment intent, the association should not be interpreted as causal and may reflect greater illness severity in patients whose sodium corrected more slowly (Seethapathy et al, NEJM Evid, 2023| Standing Corrected? NephJC).

Nowadays, we know that hyponatremia is associated with osteoporosis, falls, fractures, gait disturbance, and subtle cognitive impairment (Verbalis et al, J Bone Miner Res, 2010| Brinkkoetter et al, Sci Rep, 2019). So, my homeostasis amigos, shouldn’t we fix it? Among patients with mild to moderate asymptomatic hyponatremia, it remains unclear if normalization of sodium levels is clinically beneficial. 

Confused? Yeah, even we nephrologists still have much to learn about the ion that sets the tone for extracellular water. Homeostasis ain’t no joke!

The Study

Methods

This is a pragmatic, randomized, controlled, parallel-group, international, multi-center superiority trial with blinded outcome assessment.

The trial was conducted at 9 centers in Europe from August 2018 through April 2024.

Both university and community-based hospitals were included.

Inclusion and exclusion criteria are as follows. Patients with severe symptomatic hyponatremia requiring urgent correction, hypertonic hyponatremia, or end-of-life care were excluded.

Screening, Inclusion, and Randomization

Plasma sodium levels were reviewed by local trial physicians. Patients were invited to be included once hypotonic hyponatremia was confirmed (Na < 130 mmol/L, plasma osmolality ≤ 300 mOsm/kg). Hyponatremia could have occurred prior to, or during hospitalization. Patients could be enrolled anytime during their hospitalization. Initial assessment included reason for hospitalization, comorbidities, and additional lab values. A cognitive assessment was done with the Trail Making Test (essentially a connect-the-dots in numerical order). 

The trial was open-label with respect to treatment assignment, but outcome adjudication and follow-up assessments were performed by investigators blinded to treatment allocation. Randomization was centralized and performed in a 1:1 ratio without stratification.

Targeted Correction of Hyponatremia

Patients were then randomly assigned 1:1 to either targeted plasma sodium correction (intervention) or standard treatment (control). Treating physicians in the control group were not restricted from correcting hyponatremia. Therefore, sodium correction also occurred in the standard-of-care arm, consistent with the pragmatic design.

The targeted treatment of hyponatremia was based upon the following protocol derived from specialty guidelines.

The predominant cause of hyponatremia was determined by history, clinical presentation, assessment of fluid status, and lab testing. Hyponatremia determined to be due to “mixed etiologies” was classified by the predominant cause. Each classification of hyponatremia had its own stepwise treatment approach, determined by severity and response to therapy.

Treatment response and adherence were monitored daily. Treatment was intensified if sodium correction was < 2 mmol/L/d, was maintained if the correction was 2-12 mmol/L/d, and stopped if the correction was > 12 mmol/L/d (or > 18 mmol/L in 48 hours, considered an overcorrection). Overcorrection could be treated with hypotonic fluids with or without desmopressin. Apart from assuring adherence to the hyponatremia protocol, the trial team did not influence treatment or discharge decisions. After discharge, the treatment of hyponatremia was left to the treating physician.

Targeted plasma sodium was treated to a level of 135-145 mmol/L and was maintained for the duration of the hospitalization until discharge. Targeted correction was also stopped if the patient was hospitalized for > 30 days after inclusion, and they failed to attain normalization of sodium.

In the control group, patients received usual care, meaning hyponatremia was managed as in routine practice. Although described as “based on the European guidelines” (Spasovski et al, Eur J Endocrinol, 2014), no structured algorithm or treatment targets were mandated. Clinicians followed general principles, but therapy could range from active treatment to observation, and escalation to achieve normonatremia was not required. This contrasts with the intervention group, where care followed a predefined, stepwise protocol aimed at sodium normalization. Treatment decisions and sodium values were recorded retrospectively from the medical record after discharge.

Outcomes

Primary Outcome

-A combined risk of death or rehospitalization within 30 days of trial inclusion. The authors used no hierarchical weighting (death and rehospitalization were treated as equivalent events).

Secondary Outcomes

• Mortality and rehospitalization within 30 days and 1 year of trial inclusion

• Absolute change and rate of change of sodium levels

• Maximum change in sodium levels from inclusion to discharge

• Rate and time to normalization of sodium

• Rate of recurrence of hyponatremia within 30 days

• Complications due to overcorrection

• Length of hospital stay

• Falls and fractures within 30 days

• Quality of life questionnaire

• Trail Making test at admission and discharge

• Corrected 30-day rehospitalization/mortality rates based upon etiology, age, sex, and treatment

Events during the trial period were recorded by the trial team. Additional assessments were done following the hospitalization by a team blinded to treatment assignment.

Statistical Analysis

The sample size was calculated assuming an event rate of 23% in the standard care and 18% in the treatment group. To achieve 80% power with an alpha error rate of 5%, 2050 patients were needed for evaluation. 

The primary analysis followed a modified intention-to-treat approach, including all randomized patients who remained in the study for at least 24 hours. Predefined subgroups included severe hyponatremia (Na < 120), etiology, age (< or > 70 years), and sex. Secondary outcomes were considered exploratory without plans for multiple comparisons or adjustment of confidence intervals.
Missing outcome data were handled using available-case analysis without imputations.

Funding

The study was supported by a grant from the Swiss National Science Foundation. The funders had no role in the design and conduct of the trial; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and the decision to submit the manuscript for publication.

Results

The per-protocol analysis included 2049 patients. In the full analysis set, 1079 were assigned to the intervention and 1094 to the control group.

The median age was 73, and 48% of the included patients were male. At inclusion the median plasma sodium level was 127 mmol/L in both groups. Roughly 6% of the entire cohort had severe hyponatremia (< 120 mmol/L).

Hyponatremia treatment strategies

Etiology-directed treatment was used more frequently in the intervention group, including fluid restriction for SIADH, isotonic saline for hypovolemia, and diuretic adjustments for hypovolemia. Hypertonic saline and desmopressin were used infrequently in both groups. Patients in the intervention group experienced a maximum absolute change of 10 +/- 5.6 mmol/L, while the control group had an 8.7 +/- 5.6 mmol/L change.

A normal sodium level was obtained by 60% of the intervention patients versus 46% of the control group. At discharge, 56% of intervention and 37% of control patients had normal sodium levels. Hyponatremia recurrence or persistence at 30-days was approximately 41% in both groups. The mean rate of sodium correction during hospitalization was modestly higher in the intervention group, but the overall sodium trajectories overlapped between groups. The median time to sodium normalization was shorter in the intervention arm.

Primary Outcome

Death or rehospitalization within 30 days of study inclusion occurred in 218 intervention patients versus 234 control patients (estimated absolute difference -1.3%; 95 CI: -4.9 to 2.2, p = 0.45). The observed event rate and sample size were consistent with the prespecified power assumptions. The finding was consistent when correcting for potential site effects, as well as best and worst-case scenario analyses, in which missing 30-day outcomes were assigned extreme assumptions favoring either the intervention or control group.

Secondary outcomes

Death occurred in 8% of both groups. In the same time frame, 13% of intervention patients and 14% of control patients were rehospitalized within 30 days. The intervention was not associated with time to the primary outcome, time to death, or time to rehospitalization. The median length of stay (7 days) was similar in both groups. There were no differences in cognitive assessment, quality of life, or rates of falls or fractures at 30-day follow-up. One-year mortality and rehospitalization rates likewise did not differ, indicating no sustained benefit of target sodium correction.

Subgroup analyses

Prespecified subgroups of etiology, severity, and sex did not suggest an association with the primary outcome. Tests for interaction were not statistically significant across prespecified subgroups, including baseline sodium severity, etiology, age, and sex. It is noteworthy that in patients > 70 years old, the intervention was associated with a lower risk of the primary outcome (OR 0.76, 95% CI, 0.57-1.00). This apparent age interaction was borderline and not adjusted for multiple comparisons. There was no evidence of correlation for reaching normal sodium levels and the primary outcome. However, reaching a normal sodium level (in either group) was associated with decreased odds of death or rehospitalization within 30 days (OR 0.74, 95% CI, 0.60-0.91). The association between normonatremia and improved outcomes appeared observational and was not modified by treatment assignment. 

Safety analysis

Overcorrection occurred in 2.3% of intervention and 1.4% of control patients. Most overcorrection events occurred within the first 48 hours of treatment and were managed with desmopressin and hypotonic fluids. Adverse events (e.g., prolonged hospitalization) occurred in 1.2% of intervention and 0.8% of control patients. No cases of osmotic demyelination syndrome were observed. Symptomatic worsening of hyponatremia occurred in 0.5% of intervention and 0.1% of control patients. All patients responded to therapy intensification. Rates of ICU transfer, neurologic complications, and treatment-related adverse events were low and similar between groups.

Weaknesses

• unblinded (design, with potential bias from increased awareness and monitoring of hyponatremia in the intervention arm)

• predominantly mild to moderate hyponatremia, limiting applicability to severe disease

• Intervention limited largely to the hospitalization period, with no standardized post-discharge management

• Modest separation between groups, with only small differences in achieved sodium level and substantial overlap in treatments

Strengths

• prospective, pragmatic, randomized design

• large sample size

• diverse (academic/community) hospital settings

• blinded outcome adjudication despite open-label treatment

• prespecified power calculation met with the expected event rate

• algorithm-based intervention grounded in physiology

• hard clinical primary endpoint (death or rehospitalization) rather than biochemical outcome 

Discussion

“To treat or not to treat?” is a profound question in asymptomatic hospitalized patients.

Therapeutic apathy and inertia can drive inactivity. However, there is precedent that intensifying treatments in the controlled hospital setting can lead to worse clinical outcomes and complications (e.g., hypoglycemia and hypotension) and often includes medication confusion following discharge. In the case of mild to moderate hyponatremia, the question arises- is euboxia and normonatremia the ultimate goal? It is interesting to note that in this well-run, large trial of hyponatremia, achieving a normal serum sodium (in either group) was associated with improved outcomes of death and rehospitalization at 30 days. However, this association does not establish causation and likely reflects improvement in the underlying disease rather than sodium correction itself. Hyponatremia represents disturbed water homeostasis and often serves as a marker of illness severity. This interpretation is consistent with pharmacological trials, such as vasopressin antagonists, which failed to show improved clinical outcomes despite normonatremia (Schrier et al, NEJM, 2006). Association versus causation appears to be in full effect for the hyponatremia correction conundrum.

This study found no significant difference in its primary outcome. However, the difference between treatment strategies was small (only 1.3 mmol/L higher in the intervention group), and therapies were not necessarily continued after discharge from the hospital setting. Despite greater normalization of sodium in the intensive group, there was a significant portion of patients who had recurrence of hyponatremia at 30 days (approximately 42% in both groups). Also, the use of therapies further down the algorithm was limited: only 8% were prescribed urea, and 3% vaptans. One could argue that urea is cheap, safe, and effective and could be used more (beyond what the guidelines call for). On the other hand, the vaptans are costly and do have significant side-effects (including liver toxicity and increased risk of GI bleeding) that might have limited their use, even in severe cases. One is left wondering if perhaps the intensive arm wasn’t intensive enough to make a significant difference. Especially since 40% of patients were discharged with uncorrected sodium levels. If we want to test if fixing the sodium makes a difference, we should use an algorithm that does a better job of fixing the sodium. 

The elderly (> 70 years old) appear to benefit the most from intensive interventions. This may be because this group is at higher risk for complications (cognitive, falls, fractures). Other than age, no other factors, including etiology and severity of hyponatremia, seem to have a significant association with the primary outcome. Because hyponatremia is common and ubiquitous, further large studies of subgroups might identify patients with a greater opportunity to benefit from intensification of sodium correction.

Hyponatremia is not a single disease but a manifestation of disturbed water homeostasis arising from multiple underlying conditions, including SIAD, diuretic use, hypovolemia, heart failure, cirrhosis, and endocrine disorders. Because these etiologies differ in pathophysiology, prognosis, and response to therapy, clinical studies of hyponatremia inherently pool biologically distinct populations. In this context, Hoorn and Zietse note that “the patient with hyponatremia does not exist”; because outcomes are largely characterized by the underlying disease complicated by hyponatremia rather than the sodium concentration itself (Hoorn, Zietse, JASN, 2017). Despite this heterogeneity, the present trial did not demonstrate differences in outcomes according to etiology. This might reflect both limited power for subgroup analysis and imprecision in diagnostic classification. Differentiating hypovolemic from euvolemic hyponatremia based on clinical assessment is unreliable, with reported sensitivities of 50-80% and specificities. 30-50%, and mixed etiologies are common. Patients frequently have multiple contributing mechanisms, further blurring etiologic distinctions. These diagnostic limitations dilute treatment effects in pragmatic trials and reduce the impact algorithm-based correction strategies, particularly when treatment is guided by imperfect volume classification. 

The HIT trial fits within this framework and provides randomized evidence addressing a central assumption underlying current guidelines. Both US (Verbalis JG, et al, Am J Med 2013) and European (Spasovski et al, Eur J Endocrinol, 2014) recommendations emphasize etiology-based therapy and cautious correction but differ in how strongly they prioritize biochemical normalization. The present trial shows that guideline-based intensification improves sodium normalization without improving short-term clinical outcomes. These findings suggest that, particularly in heterogenous populations with mild hyponatremia, normalization of sodium alone is unlikely to modify outcomes driven primarily by the underlying disease

Appropriately, the authors conclude that the trial does not suggest “do nothing” for patients with mild to moderate hyponatremia, but rather that intensive treatment does not confer clear or significant benefits over standard therapy. Increasing the number of therapeutics and intensifying treatment of moderate hyponatremia does not appear to be beneficial, at least in the short term. Standard treatment, which included doing nothing in 35% of patients, did not have any significant downside or safety issues.

Conclusion

In hospitalized patients with mild to moderate chronic hyponatremia, a more intensive targeted correction strategy modestly improved sodium levels but did not reduce 30-day mortality or rehospitalization, suggesting that in this population, where hyponatremia was generally mild, biochemical correction alone may be insufficient to influence short-term clinical outcomes.

Summary by

Brian Rifkin MD, FASN, FASDIN
Hattiesburg Clinic

Reviewed by

Cristina Popa, Joel Topf

The most fundamental advice for stone formers is simple: drink more.

How much? More. No matter how much urine you make, if you can make more you will do better, but more than 2.5 liters is the goal. We don’t actually care how much you drink—we care how much you pee. The goal is >2.5 liters of urine, because dilution fixes the chemistry that drives stones.

But patients don’t care about urine chemistries—they care about avoiding stones. And that’s where things get hard.

Stone trials with stone events are brutal. Events are slow, unpredictable, and unforgiving. They have a habit of wrecking dogma.

NOSTONE is the perfect example—thiazides improved urine chemistry and CT findings, but didn’t reduce patient facing stone events.

Now patient-centered outcomes come for the granddaddy of all advice: drink more water.

The PUSH trial (Lancet 2026) randomized patients to an aggressive, behaviorally engineered hydration strategy versus usual care (which, in stone disease, of course, is “MORE WATER”).
Once again, the RCT shows up and steals our favorite toy.

Let’s take a look at this trial before it is memory holed by every stone specialist.

Before we dive further into PUSH we should understand a bit of the prior art:

The most important study is an Italian RCT by Borghi, Urinary volume, water and recurrences in idiopathic calcium nephrolithiasis: a 5-year randomized prospective study (J Urol 1996) These guys enrolled patients, after their first kidney stone and randomized them to either:

1. Being told to drink a lot of water or

2. Being told, “Since it is your first kidney stone, there is no need to make changes to your diet.”

That’s it. That’s the intervention.

But the trialists must have been devastatingly persuasive.

People told to drink a lot of water immediately doubled their urine output and over time further increased their urine output to 2.6x their baseline urine volume. Patients in the control group had a modest 20% increase in urine volume for the first year and then reverted back to their baseline urine volume for years 2-5.

I hope I sound skeptical because I am. This sounds like bullshit to me. When Fred Coe (Changes in urine volume accomplished by physicians treating nephrolithiasis J Urol 2003) looked at how much urine volume increased following patients engaging a kidney stone doctor, he found a delta of 300 ml, not the 1600 ml the Italians “found.” Intrerestingly Coe found that the bump in urine volume was inversely proportional to the initial urine volume (reversion to mean or something else?).

Now back to PUSH

METHODS

Participants

The investigators targeted patients with a recent stone event (in the last 3-years, or in the last 5-years if there was also a new stone by U/S or CT).

Only patients with less than 2 liters of urine were randomized, exactly the patients we tell to drink more water. This setup raises a concern: regression to the mean could inflate urine improvements in both groups.

Intervention

Every nephrologist has their little advice they give people to get them to drink more and instagram has an entire cottage industry dedicated to convincing people to increase their water intake. PUSH didn’t just say “drink more”—it engineered behavior:

• Loss framed payments for drinking water. People were promised $1.50 a day unless they drank less than 2.5 liters a day. Very Kahneman and Tversky. This stimulus was gradually reduced over time

• Coaching with structured problem solving to get participants to DRINK MORE

• Text reminders and social nudges

The control got the usual, “drink more water” pep talk!

Assessment

Intake was tracked with Bluetooth water bottles. Very 2020s

Stone events were  investigated by a blinded adjudication committee. I love that patients qwere not trusted to adjudicate their own patient reported outcomes.

Secondary outcomes were exhaustive including serial CT scans and 24-hour urines.

Power analysis

The needed 821 group to reach 80% power to detect a 30% reduction in stone events at 2 years. Yikes, that is a large effect size. They assumed a 15% event rate with 20% attrition.

RESULTS

They nailed enrollment and retention, an under-appreciated win (1658 enrolled compared to 1642 estimated; 1367 retained, above the 1326 estimated).

Demographics

They had more women than men. We don’t see that often in studies. Maybe this was due to the requirement for low water intake for enrollment?

Baseline Urine

Baseline use of anti stone therapy was modest, 7% thiazide and 11% KCitrate.

Baseline urine chemistries were surprisingly normal. The standout abnormality was volume: 1.3 L/day.

Primary Outcome

After 2 years 19% in the intervention group had a stone events compared to 20% in the control group. No signal.

The intervention increased urine volume, but the effect faded over time and never reached the 2.5 liter target.

In the control group we have nice Hawthorne effect versus reversion to the mean.

The serial CT scans showed no difference in stone growth or reduction in the development of new stones.

Adverse effects

The intervention group drank ~28 extra gallons over 6 months.

And what was the payoff?

More urgency, more frequency, and more nocturia.

More water. More symptoms. No benefit.

The authors considered if the trial would have shown a different result if it went longer. They concluded, “The follow-up period was only 2 years; however, the PUSH trial exceeded the estimated event rate, and the Kaplan–Meier curves were nearly parallel, suggesting that a longer follow-up period for PUSH is unlikely to change the results.”

Conclusion

So what did PUSH actually show?

Not that water doesn’t work.

It showed that even an intensive, multi-layered, evidence-based behavioral intervention cannot increase hydration enough to prevent stones.

This is less a failure of physiology and more a failure of our ability to change behavior.

Ooof.

Kidney stones clinics take another one on the chin.

In this edition

• The Assist trial (Efficacy and Safety of Atrasentan in Patients with IgA Nephropathy Receiving Sodium-Glucose Cotransporter 2 Inhibitors: Placebo-Controlled, Crossover Trial; Heerspink et al, JASN 2026)

• 24-month data from APPLAUSE-IGAN (Iptacopan in IgA Nephropathy — Final 24-Month Data; Barratt et al NEJM 2026)

• Sexual Dimorphism and Age Effects in CKD and Type 2 Diabetes in the CONFIDENCE Trial (Agarwal et al, NDT, 2026)

• Acute eGFR Changes and Their Mediation of Albuminuria Reduction with Empagliflozin and Finerenone (Agarwal et al, JASN, 2026)

The Assist Trial

J Am Soc Nephrol. 2026 Mar 29. doi: 10.1681/ASN.0000001076. Online ahead of print.

Efficacy and Safety of Atrasentan in Patients with IgA Nephropathy Receiving Sodium-Glucose Cotransporter 2 Inhibitors: Placebo-Controlled, Crossover Trial

Hiddo J L Heerspink, Irene L Noronha, Jose Luis Górriz, Soo Kun Lim, Sradha S Kotwal, Gianna M Kirsztajn, José Barros Neto, Jessica Ryan, Mei Sian Fu, Sung-Gyun Kim, Jonathan Barratt, Yasmin Brahmbhatt, Greggory J Housler, Rong Jiao, Marion Dahlke, Amit Lodha, Amy K Mottl; Atrasentan and Sodium Glucose Co-transporter-2 Inhibitor Efficacy and Safety Trial (Assist) Investigator Group

PMID: 41904616

Why was this trial needed?

The natural history of IgA nephropathy (IgAN) is unforgiving: even after optimal RAS blockade, a proteinuria >0.5g/day predicts a 30-35% risk of end-stage kidney disease within two decades (KDIGO IgA group, Kidney Int, 2025 | NephJC summary). Recent “renoprotective” breakthroughs have turned flozins into the new backbone of supportive care (or “palliative” if you agree with Dr Barratt’s terminology) care. In the DAPA-CKD and EMPA-KIDNEY trials, patients with IgAN experienced a 26% reduction in albuminuria and a 71% decrease in the composite renal endpoint (Wheeler et al, Kidney Int, 2021| EMPA-KIDNEY Collaborative Group, Lancet Diabetes Endocrinol, 2024). Their mechanisms (tubulo-glomerular feedback-mediated reduction of intraglomerular pressure, natriuresis, anti-inflammatory, and podocyte-protective effects) soften the hemodynamic injury that fuels disease progression (Kim et al, NDT, 2026). Yet flozins alone leave a sizable residual risk. In both trials (which were not IgAN focused), median proteinuria after treatment remained >0.8 g/day, and the eGFR slope, although flatter, did not halt decline (Wheeler et al, Lancet Diabetes Endocrinol, 2021). Therefore, the sponsors and investigators believe IgAN calls for an additional, mechanistically distinct agent that targets the endothelin 1 (ET-1) pathway, a key driver of glomerular vasoconstriction, inflammation, and fibrosis. One could argue that targeting B Cells (or even complement - see the #NephMadness match-up) is more disease modifying than using Flozins and ETAs. Atrasentan, a selective endothelin-A receptor antagonist, achieves precisely this. In the phase-3 ALIGN study, atrasentan lowered the UACR by 38% versus a 3% rise with placebo (difference -36%, p<0.001) after 36 weeks, with a safety profile that lacked clinically relevant fluid overload (Heerspink et al, NEJM, 2025). Earlier SONAR data in DKD confirmed the proteinuria benefit and highlighted a manageable fluid retention signal when patients were carefully selected for low-risk phenotypes (Heerspink et al, Lancet, 2019 | NephJC summary).
The next logical question is whether adding atrasentan to a flozin amplifies the antiproteinuric effectS without compounding adverse effects. Real-world evidence with the dual endothelin-angiotensin blocker sparsentan (which shares the ETA antagonism of atrasentan) showed a 62% reductions in UPCR at 14 weeks when patients were already on maximal RAS blockers plus flozins (Schanz et al, CKJ, 2024). However, sparsentan confounds the interpretation because it simultaneously blocks the AT1receptor; a pure ETA antagonist, like atrasentan, is needed to isolate the endothelin contribution.

How was the trial done and what did it report?

The ASSIST phase II trial was a double-blind, placebo-controlled crossover study in which each participant receives atrasentan (0.75 mg daily) for 12 weeks, a 12-week washout, then placebo (or reverse sequence) while remaining on stable flozin dosing (Heerspink et al, Kidney Int, 2021). The crossover format maximizes statistical power- each subject acts as its own control - while limiting the sample size needed to detect modest but clinically meaningful changes in proteinuria and eGFR. Moreover, the trial stratifies enrollment by SGLT2i status (stable vs run-in) to ensure that any safety signal, such as the fluid retention observed in SONAR, is captured in the exact context of contemporary IgAN care. The proteinuria required for enrolment was >0.5g/day for flozinated patients, and > 0.85 g/day for flozin-naive, who underwent a flozin run-in period to ensure their post flozinated albuminuria was still high enough. Notable exclusions were type 1 DM (flozin’s risk for causing DKA), prior history of heart failure or BNP > 200 pg/ml, or anemia (Hgb < 9 g/dL), as ETAs can cause heart failure and anemia. The primary outcome was proteinuria reduction at 12 weeks, see figure 1 for details. The trial was powered for a 25% reduction in UPCR, requiring 52 participants. 

Overall, 54 patients were randomized, with a median GFR of 63 and UPCR of 1 g/g, and were mostly (65%) White. Only 13% were non-flozinated at baseline, requiring a flozin run-in phase. Atrasentan reduced proteinuria by 30% (versus 7% in placebo), thus hitting the powered for target and coming in at −25.3% (95% CI, −36.8 to −11.7; p < 0.001). Treatment sequence did not matter, and for those in sequence two who went on to receive atrasentan, there wasn’t much more of a proteinuria decrease at 24 weeks (treatment difference of -26.4% versus placebo).

The GFR slope at 12 weeks was a meaningless: 1.2 versus 1.5 ml/min/1.73m2. Though overall safety events seem similar, one patient had AKI deemed to be related to atrasentan, fluid retention was twice as common with atrasentan, BNP increased in sequence 1 but not sequence 2, and the hemoglobin dropped a bit with atrasentan. 

What does this trial change?

The trial proved atrasentan reduces proteinuria at 12 weeks in flozinated participants with UPCR > 0.5 g/day by about 25%. Does this matter, and is it really as safe as the authors claim? We would posit the answer to both is, “No”. 

Firstly, IgAN therapeutics are a crowded space with many efficacious disease modifying drugs (steroids, targeted-release budesonide, BAFF/APRIL, and complement inhibitors). Are ETAs disease-modifying or supportive/palliative therapies like RASi and flozins? Theoretical effects of ETAs aside, it is difficult to see them being disease modifying. Their effects are likely hemodynamic (the mechanistic lack of a huge effect on BP in an underpowered RCT should not be overinterpreted). Time will tell how good they are in establishing a position for themselves in this space. 

Unfortunately, safety remains a concern which cannot be allayed in a 54 patient RCT of 12 to 24 weeks of exposure. Even in this carefully selected (preserved GFR, no heart failure, low BNP) population, there were some concerning BNP signals (albeit in one sequence) and an associated hemoglobin drop. 

This trial is useful in showing the pharmacologically and volumetrically rationale combination of a flozin and ETA (note the ongoing trial with zibotentan and empagliflozin in DKD summarized here) is a promising option that attenuates the sodium retention abilities of ETAs while synergizing the anti-proteinuric effects. Will this supplant B-cell therapies? Again, the answer is, “No”. Is it good to have more options for the ‘most common primary glomerulonephritis’? To this we give a halfhearted, “Yes”. 

The APPLAUSE-IgAN Trial 24 month Data

N Engl J Med, 2026 Mar 29. doi: 10.1056/NEJMoa2600743. Online ahead of print.

Iptacopan in IgA Nephropathy — Final 24-Month Data

Jonathan Barratt, Necmi Eren, Naoki Kashihara, Bart Maes, Dana V Rizk, Brad Rovin, Hernán Trimarchi, Hong Zhang, Weiming Wang, Ismail Kocyigit, Chuanming Hao, Vladimir Tesař, Kenan Turgutalp, Li Yang, Guangqun Xing, Valter Duro Garcia, Seung Hyeok Han , Wanhong Lu, Antonio Pisani, Julia Weinmann-Menke, Annabel Magirr, Ronny Renfurm, Thomas Hach, Vlado Perkovi; APPLAUSE-IgAN Study Group

PMID: 41910396

Why was this trial needed?

Yet another IgAN trial? (in fact, ASSIST and APPLAUSE were part of a dedicated IGAN trial session, and sandwiched a fascinating phase 2/3 trial of targeted release budesonide - aka nefecon alternative - that we look forward to reading when published). But our cynical snark aside, the foundation for the APPLAUSE-IgAN study rests on an established unmet need, a strong mechanistic rationale, and the regulatory requirement for confirmatory evidence. 

• Unmet need: despite optimized supportive care with RAS inhibitors, a significant proportion of patients with proteinuria > 1 g/day progress to end-stage renal disease (Reich et al, JASN 2007| Liu et al, Exp Ther Med, 2016). The use of systemic corticosteroids is limited by substantial toxicity (Kim et al, CJASN, 2025). This created the need for targeted, disease-modifying therapies. Keep in mind this RCT began enrolling in 2020, when all we had in IgAN was the initial TESTING data (Lv et al JAMA 2017 | NephJC summary). 

• Mechanism: the alternative complement pathway has been reported to be a driver of glomerular inflammation and injury in IgAN (Medjeral-Thomas et al, Semin Immunopathol, 2021). As Jon Barratt points out in his NephMadness commentary, it has been universally noted that for every bit of IgA deposited within the glomerulus, there is accompanying C3. The deposition of galactose-deficient IgA1 (Gd-IgA1) immune complexes in the mesangium activates the alternative pathway, leading to the formation of C3 and C5 convertases and the terminal membrane attack complex (MAC), which mediate tissue damage (Teh et al, AJKD, 2026). Iptacopan is a first-in-class oral inhibitor of Factor B, an important component of the alternative pathway C3 convertase. By selectively blocking the alternative pathway, it targets a central step in the disease’s inflammatory cascade without affecting the classical or lectin pathways, theoretically offering a more precise therapeutic strategy (Schubart et al, Proc Natl Acad Sci USA, 2019| Risitano et al, Lancet Hematol, 2021). Since we all need reminders of (or have nightmares of trying to remember) the complement pathway, here it is, along with the various therapeutic targets developed or in development.

• Regulatory imperative: while the initial data (Perkovic et al, NEJM, 2025) showed that iptacopan reduced proteinuria at 9 months in a dose-dependent manner, full regulatory approval requires the 24 month GFR slope data from a large-scale phase 3 trial. Based on the 9 month data, iptacopan has already received accelerated approval from the FDA in August 2024.  

How was this trial done, and what did it report?

This was a large multicenter prospective placebo-controlled trial, sponsored by Novartis (the maker of iptacopan) and overseen by the academic team. It was the typical IgAN trial design that you have read already on NephJC: patients with IgAN with more than 1g/g UPCR and GFR > 30, receiving iptacopan or placebo, on a background of RASi with optional (not mandated) flozin use. Kidney biopsy was required within 2 years for those with GFR 30-45 and within 5 years for those with GFR > 45, and with less than 50% IFTA (though this was the reported and not a central pathology re-reading of biopsies). Pneumococcal and meningococcal vaccination was required, and Hib vaccination was recommended, given planned complement inhibition. The primary outcome was the annualized GFR loss over the 24-month follow-up period with several secondary outcomes including a composite kidney-failure endpoint (sustained decline in the eGFR of ≥30% from baseline, a sustained eGFR of <15, initiation of dialysis, kidney transplant, or death from kidney failure, whichever occurred first) and the change from baseline to 9 months in the score on the Functional Assessment of Chronic Illness Therapy (FACIT) – Fatigue, version 4 (on a scale from 0 to 52, with higher scores indicating less fatigue). 

478 patients were enrolled, and the discontinuation rates were higher with placebo (37%) than with iptacopan (19%). The mean age was about 40 years, just over half were men, just over half were Asian, with the rest mostly White, a UPCR of 1.7 g/g and a GFR just over 60 ml/min at baseline. Almost all were on RASi, about 20% flozinated, with just over a quarter having previously received steroids and 75% with hematuria at baseline. Iptacopan slowed GFR progression by half, from about 6 to 3 ml/min/year (between-group difference of 3.02 ml per minute per 1.73 m2 per year, 95% CI, 2.02 to 4.0; adjusted p<0.001). The composite kidney outcome was also positive (HR 0.57, 95% CI, 0.40 to 0.81; adjusted p = 0.003) but the FACIT score was not, and thus the p values for subsequent hierarchical outcomes were not reported. From the subgroups (enrollment stratified a priori for GFR 30-45 and > 45, Asian or not, UPCR >/< 2 g/g), the effects were quite consistent. Though adverse events were overall similar, ~ 60% of them were infections (consider enrolment started 2020) and serious infections were more common with iptacopan (6.7% versus 2.1% with placebo). 

What does this trial change?

Clearly, trial results deserved the APPLASE it received at the presentation. Iptacopan does slow kidney disease progression at 24 months in IgAN, and seems safe in this large phase 3 trial. Notable aspects of the trial results are the substantial 3 ml/min/year GFR decline that still occurred in the iptacopan group (compared with presented telitacicept data of almost no decline, while atacicept data is awaited), the rapid proteinuria reduction with iptacopan, and the higher infection rates despite prior vaccination and close care provided in the trial setting. 

It is useful to revisit the therapeutic arena in IgAN as covered by NephMadness (and the superb GN in 10 podcast episode) to frame this discussion. We have B-cell therapies (APRIL/BAFF inhibitors) to act at the proximal hits in the IgAN pathophysiology. Complement inhibitors are clearly distal to that mechanism - so would that make them redundant in someone already on B-cell therapies? Quite possibly - or at least lower down in therapeutic importance? In an ideal world with unlimited money (or seamless access) one could even consider using both these agents? Actual disease progression is not as clean as it is portrayed in these mechanistic figures, and combination therapy might well be the best approach. We are still in the early phase of drug development in this area, so we won’t have empirical data to make these decisions and are relying on mechanisms and pathophysiology to read the tea leaves. Hopefully we will eventually have this data to make better decisions - and a drug that costs less than half a million dollars annually. Until then, we can applaud the development of another therapeutic agent, which also helps us understand how the complement pathway is intricately involved in IgAN disease progression.  

CONFIDENCE: Age and Sex interactions

Nephrol Dial Transplant. 2026 Mar 27:gfag075. doi: 10.1093/ndt/gfag075. Online ahead of print.

Sexual Dimorphism and Age Effects in CKD and Type 2 Diabetes in the CONFIDENCE Trial

Rajiv Agarwal, Jennifer B Green, Hiddo J L Heerspink , Johannes F E Mann, Janet B McGill, Amy K Mottl, Masaomi Nangaku, Julio Rosenstock, Muthiah Vaduganathan, Charlie Scott, Li Li, Na Li, Carolina Aldworth, Meike Brinker, Peter Rossing

PMID: 41894225

Why was this analysis needed?

The data arrives clothed in certainty: a larger fall in albuminuria, a tolerable drip in filtration, a modest lowering of the pressure. Yet, at the bedside, this order dissolves- and biology refuses abstraction. The CONFIDENCE trial (Agarwal et al., NEJM, 2025 | NephJC summary) enrolled a broad DKD population and reported an average ≈ 30% extra UACR fall, a modest, reversible dip in eGFR, and a small systolic blood pressure (SBP) decline (Vaduganathan, NDT, 2025). Those aggregate numbers are compelling, yet they mask two biologically potent variables- sex and age- that shape the natural history of diabetic kidney disease. If the combination behaves differently in a 45-year-old man versus a 78-year-old woman, clinicians must know it before committing patients to a new regimen.

Sex is not a neutral descriptor. Men tend to experience earlier glomerular hyperfiltration and progression to end-stage renal disease, while women, despite higher CKD prevalence, progress more slowly, partly because estrogen confers vascular protection (Farahmand et al., BMC Endocr Disord, 2021). A pooled analysis of the FIDELITY data set (Bansal et al, BMJ Open, 2024) highlighted that finerenone’s reduction in heart-failure hospitalizations was more pronounced in men, whereas preliminary observations from CONFIDENCE hinted that women were more likely to achieve a >30% UACR drop (Mottl, et al, JASN, 2025). These divergent hints suggest that the finerenone-empagliflozin duet may deliver different magnitudes of benefit depending on the patient’s sex. Ignoring this possibility would be equivalent to prescribing a one-size-fits-all approach to patients with CKD. 

On the other hand, age governs both risk and tolerance. Elderly patients carry a higher baseline cardiovascular risk and a reduced renal reserve, making them especially vulnerable to the acute eGFR dip and hyperkalemia that follows mineralocorticoid-receptor antagonism. Younger patients, by contrast, have the potential to harvest a lifetime of protection if aggressive albuminuria lowering is instituted early. The primary CONFIDENCE manuscript assures us that safety was “consistent across KDIGO risk categories”, yet “consistent” can conceal clinically meaningful gradients- especially in the oldest quartile, where potassium handling and blood pressure autoregulation can differ markedly (Mottl, et al, JASN, 2025). An age-stratified analysis would therefore quantify whether an 80-year-old safely tolerates the same initial dose titration as a 55-year-old and whether the magnitude of UACR reduction, and its downstream impact on hard outcomes, remains comparable.

Mechanistic certainty already exists for SGLT2i and ns-MRAs to decrease albuminuria. A mediation analysis demonstrated that early UACR reduction accounts for roughly 84% of finerenone’s kidney-protective effect and about 37% of its cardiovascular benefit, establishing albuminuria as a validated surrogate endpoint (Agarwal et al, Ann Intern Med, 2023). However, it examined finerenone alone, not the synergistic combination now under scrutiny. The CONFIDENCE interplay analysis supplies the raw efficacy signal of the duet (larger UACR decline, modest eGFR and SBP changes) but does not translate those averages into probabilities of achieving a clinically meaningful albuminuria response for a specific demographic group. Only by overlaying the mediation framework with age- and sex-specific outcomes can we predict how a 30% UACR fall will affect an individual’s risk of end-stage renal disease or heart-failure hospitalization.

How was it done, and what did it report?
To move beyond the aggregate data, the investigators conducted a prespecified exploratory analysis of the 798 patients in the CONFIDENCE trial. They employed a linear mixed-effects model to dissect the longitudinal trajectories in UACR, allowing evaluation of how age and sex influence response over the 180-day treatment period. Age was modeled both as a continuous variable and categorically by quartiles, while sex was analyzed as a binary variable. The models tested for interaction between these demographic factors and the treatment arms to determine if the benefits were uniform. The analysis showed:

• A linear age-efficacy gradient: advancing age was a robust factor predicting a greater albuminuria-lowering response. The analysis quantified a significant 10.2% incremental reduction in UACR for every 10-year increase in age (95 CI: -15.5 to -4.6; p=0.001). This demonstrates that therapeutic efficacy does not wane, but scales positively with age.

• Pronounced sexual dimorphism: the data revealed a sex-based difference in treatment response. Women achieved an 18.6% greater reduction in UACR compared to men by day 180 (95% CI 6.3 to 28.2, p=0.008).

• Treatment-independent biological effect: the age and sex effects were independent of specific therapy. Whether a patient received finerenone, empagliflozin, or a combination, the magnitude of their response was similarly modulated by their age and sex. This points toward a fundamental biological mechanism rather than a drug-specific interaction. Importantly, the enhanced efficacy in older patients and women was not accompanied by a compromised safety profile; the incidence of adverse events, including hyperkalemia and acute kidney injury, remained consistent across all age and sex cohorts (see table 4).

What does this analysis change?
The analysis reframes the interpretation of the CONFIDENCE trial, shifting it from a straightforward report on additive efficacy to a more nuanced text on precision nephrology. However, this enthusiasm must be tempered by the analyst's inherent limitations.
First, it provides a compelling, if preliminary, counter-narrative to the “prescribing paradox” in geriatric medicine. The pervasive clinical inertia that leads to under-prescription in older patients is directly challenged by data suggesting that albuminuria reduction is enhanced with age. This provides an evidence-based rationale to treat the elderly more aggressively (or at least, just as aggressively). A crucial note of caution, however, is that this conclusion is based on a 180-day reduction in a surrogate marker. We are assuming that this ultimately translates into cardiorenal benefits. 

Second, it generates a powerful hypothesis regarding the “female paradox” in DKD. The superior response in women makes the authors say that the historical disparities in outcomes may stem from systemic undertreatment, rather than biological resistance. However, while statistically robust, these exploratory findings are from a trial not powered to definitively assess sex-specific outcomes. Though prespecified, these are from smaller subgroups looking at surrogate outcomes.  Never-the-less these signals may inform future studies.

Finally, while this analysis sharpens our clinical focus, it is important to interpret its findings with scientific precision. The lack of a significant treatment interaction means that age and sex are prognostic factors for the response to these classes of drugs, not that the synergy of the combination is uniquely amplified in these subgroups. An older woman responds better to empagliflozin, finerenone, and their combination. Furthermore, the reliance on albuminuria as a surrogate endpoint, though supported by mediation analyses, is not suitable for hard outcome data. History is replete with trials where promising changes in surrogate markers failed to translate into clinical benefit.

Conclusion: This analysis doesn’t provide a definitive new set of rules but rather a map. It changes the questions we must ask at the bedside: not “Does this patient need combination therapy?" but, "How might this patient’s age and sex profile predict the magnitude of their response?” It forces a reevaluation of clinical biases, while simultaneously highlighting the urgent need for long-term outcome data to validate these provocative, but as yet incomplete findings. Bottom line, treat older people and women just as aggressively as you treat younger men. No need for sexism or ageism while erecting your four pillars of DKD. 

CONFIDENCE: Mediation analysis

J Am Soc Nephrol. 2026 Mar 29. doi: 10.1681/ASN.0000001071. Online ahead of print.

Acute eGFR Changes and Their Mediation of Albuminuria Reduction with Empagliflozin and Finerenone

Rajiv Agarwal, Ricardo Correa-Rotter, Sankar D Navaneethan, Kei Fukami, Hiddo J L Heerspink, Johannes F E Mann, Janet B McGill, Amy K Mottl, Masaomi Nangaku, Julio Rosenstock, Peter Rossing, Muthiah Vaduganathan, Charlie Scott, Li Li, Carolina Aldworth, Jennifer B Green, Matthew R Weir 

PMID: 41905767

Why was this analysis needed?

The progression of CKD is propelled by a destructive interplay among declining eGFR, rising systolic blood pressure (SBP), and increasing UACR. The pathophysiological triad creates a vicious cycle where systemic and intraglomerular hypertension damage the glomerulus, leading to albuminuria; this protein leakage then incites tubular inflammation and fibrosis, accelerating nephron loss and a further decline in eGFR (Martinez Leon et al., Nat Rev Endocrinol, 2026). Therefore, effective therapy must address the entire disease axis.

While foundational RAS inhibition tempers this cycle, substantial residual risk remains. The past decade has armed us with two mechanistically distinct therapies: the non-steroidal finerenone, which targets aldosterone - and possibly also inflammation and fibrosis as shown in FIDELIO-DKD (NephJC Summary) and FIGARO-DKD trials, and the flozins, which correct glomerular hemodynamics, with profound benefits seen in trials such as CREDENCE (NephJC Summary), DAPA-CKD (NephJC Summary) and EMPA-KIDNEY (NephJC Summary). Post-hoc analyses hinted at a powerful synergy, setting the stage for the CONFIDENCE trial to prospectively test the simultaneous initiation of both agents (Rossing, et al, Diabetes Care, 2022).

This context makes a detailed interplay analysis of the CONFIDENCE trial essential (Agarwal, et al, NEJM, 2025 | NephJC). The trial’s primary findings - superior albuminuria reduction with combination therapy, without a safety signal with simultaneous start - is the first step to a more complete picture. A holistic analysis of all these markers is required to confirm the therapeutic synergy, clarifying whether the UACR reduction is achieved without an excessive initial eGFR dip or concerning blood pressure effect. By mapping the trajectory of eGFR, SBP, and UACR over time, we can determine if the combination produces a more favorable cardiac/renal state than either drug alone. 

Furthermore, dissecting this interplay may provide mechanistic and safety insight. It allows us to untangle the hemodynamic effect, driven by empagliflozin, from the anti-inflammatory benefits of finerenone. This moves beyond simple adverse-event reporting to reveal the dynamic physiological trade-offs of combination therapy, such as whether the natriuretic effect of empagliflozin mitigates the hyperkalemia risk posed by finerenone. 

How was it done, and what did it report?

This analysis is somewhat similar to the one published in Annals for FIDELITY (Agarwal et al, Annals of IM 2023), but now applied to CONFIDENCE. In this exploratory analysis of the CONFIDENCE trial, the purpose was to ascertain if the acute eGFR decline (defined as > 30% from baseline to day 14) mediated the treatment effect of the interventions on the primary outcome (UACR decrease at 180 days). A linear mixed model using fixed effects was used for acute eGFR decline estimations. For more on mediation analysis, check out this helpful tutorial (Lange et al Epid Health 2017).

Determinants of acute GFR decline

The acute decline in GFR (> 30 at 14 days) was, as expected, highest in the combination, followed by empa and then finerenone - but the nadir came earlier (at day 14) for the combo and empa, and much later (~ day 90) for finerenone. 

This acute GFR decline was associated in those with higher baseline GFR, and those on concomitant diuretics. Higher baseline UACR was associated with the chronic, not acute decline (noting that chronic is still only over 180 days). Notably there was not much of a GFR decline noted in those with lower levels of kidney function, GFR ~ 30.

Determinants of any GFR decline

Examining > 30% decline at any time point over 180 days, combo therapy (versus either agent alone), higher baseline UACR, eGFR, and diuretic use were associated with this (Table 2, below). 

Mediation analysis

In the mediation analysis (table 3, below), comparing the effect on UACR at 180 days that was mediated by the acute eGFR decline, this contribution was seen much more in combination versus finerenone (28%) than combination versus empagliflozin (5.2%), suggesting little of the additional benefit provided by finerenone is mediated by the acute eGFR decline.

Adverse effects by eGFR decline

These are shown in table 5, and it is notable that hyperkalemia was slightly more common in those who had an acute eGFR decline than in those who did not. 

What does this analysis change? 

This study provides rich granular data exploring the determinants of 30% GFR declines as well as mediation analysis of whether the acute GFR decline is important in albuminuria reduction. The decline only seen in higher GFR might reflect regression to mean or hemodynamic unloading of hyperfiltrating glomeruli. The authors posit that since only 5% of the UACR effect of finerenone mediated by eGFR decline reflects the antiinflammatory and antifibrotic (dare we say ‘pleiotropic’) effects rather than hemodynamics. This would support synergistic and independent effects of finerenone compared to flozins or RASi in DKD. 

These results should not make you more or less inclined to use these drugs together, but they do help in how we think about the creatinine bumps and changes happening upon initiation. The lack of GFR decline in those with low GFR is reassuring - one should perhaps be bold at initiating flozin + finerenone together even if GFR is 30 ml/min. The hyperkalemia seen with the combination group was somewhat disappointing in CONFIDENCE, especially for those of us who were hoping empa would attenuate the hyperkalemia seen with finerenone. As suspected, (listen to Joel on the pod) this was mostly mediated by the higher rates of acute GFR decline seen with the combination. What should one make of the mediation analysis? ¯\_(ツ)_/¯. Does it matter how a drug works, if we know it works? No, not when making decisions on whether to use it - but possibly on future research pathways more than anything else.

By Swap Hiremath & Cristina Popa

Reviewed by Brian Rifkin

Yesterday the NephJC Editorial Internship class graduated.

As part of that graduation we awarded the 2026 NephJC Kidneys, the best looking trophy in nephrology.

Here are the previous winners:

The first four awards are voted on by the entire #NephTwitter community

The manuscript of the year goes to PISCES by showing that we can make a difference in hemodialysis.

2026 NephJC Kidney for Manuscript of the Year: Fish-Oil Supplementation and Cardiovascular Events in Patients Receiving Hemodialysis.

NephJC coverage | PubMed

We love to promote new and ambitious social media projects in social media and we dedicate this award to the person who started this whole movement, Nathan Hellman.

2026 Nathan Hellman Social Media Award goes to Roger Rodby and RUSH Nephrology for Renal Biopsy Cases

Visual abstracts changed the way we share and discuss manuscripts in the short-attention-span era of social media. CJASN and NephJC are the pioneers for this media.

2026 Visual Abstract of the Year goes to Husam Alzayer

Social Justice and social media go together like CKD and SGLT2i/ARB/GLP1/nsMRAs. Since 2020 NephJC has had a Social Justice Award. In a world full of insane changes, we need a little more justice.

The 2026 Social Justice Award goes to World Kidney Day- Kidney Health for All

The second series of awards are voted on by the NephJC Working group.

NephJC is a volunteer organization and is dependant on people donating their time and energy to making this work. The MVP goes to the person who best exemplifies this ethos. NephJC only gets better and better because of this persons unyielding efforts to get the job done.

The 2026 NephJC MVP goes to co-editor in chief Brian Rifkin

NephJC depends on a constant supply of new people to help run the program. Many of these people do great work but we recognize one that is the greatest…

The 2026 Rookie of the Year goes to Akshaya Jayachandran

NephJC asks scientists to participate in the chats and podcasts and really depends on them to provide a lot of the content that makes NephJC work. The work group votes on the best of these participants. This year’s winner is sharp, funny, and friendly IgA expert with an accent…

The 2026 Engaged Scientist of the Year goes to Jonathan Barratt

And the last award is decided on just by the editorial care of NephJC, Swap, Joel, Cristina, and Brian. We already announced this one but wanted to re-recognize Paresh Jadav for his generosity for the ASN Kidney Week party.

That’s a wrap on the 2026 NephJC Kidneys! In the end, it’s not about the awards, it’s about the we do, the education we share, the friendships we build, and the inspiration driving nephrology forward. Let’s keep creating the #FOAMed to light the way for the next generation of nephrologists. Congratulations to all the winners!

The updated table of winners…

But a new development is that med-mastodon no longer is supported, and is likely to be shut down. It seems Nick Marks is unable to maintain it - for the last little while, Ryan Wild has been running this. See their blog post from January with more details. Ryan has tried to contact Nick Marks (and so have I) to transfer ‘ownership’ with no response, and Ryan will be shutting down Med-Mastodon soon. 

Many of you may just shrug. But for those of you who have a med-mastodon account and don’t want to walk away - there is an option. You can switch instances. The simple way is:

• Create an account on another instance (I went to mastodon.social , Ryan Wild runs Universeodon.com)

• Create an ‘alias’ there via preferences>Account settings, and enter your med-mastodon handle

• Go to your med-mastodon account and Preference>Account>export and follow those steps 

• Voila, your followers will come over. 

• There are ways to do a more sophisticated move e.g. here 

Not everything moves - see this excellent blog for more on what you lose and more advice

Swap (now at @hswapnil@mastodon.social)

Header image from the AMNH

#NephJC Chat

Tuesday, March 10th 2026, 9 pm Eastern on Bluesky

Clin J Am Soc Nephrol. 2026 Jan 2. doi: 10.2215/CJN.0000000969. Online ahead of print.

A Randomized Clinical Trial of Kidney Autologous Cell Therapy in Diabetic Kidney Disease

Čižman, Borut; Butler, Emily L; Stavas, Joseph; Prakash, Rachita; Saad, Theodore; Silva, Arnold; Wooldridge, Thomas; Aqeel, Ahmed; Yan, Hongxia; Barysauskas, Constance M.; Culleton, Bruce

PMID: 41481370   

‘Failure is the simple opportunity to begin again, this time more intelligently’
Henry Ford

Introduction

Around 850 million people around the world suffer from CKD; among them, 50% is due to diabetic kidney disease (DKD). For many decades, the only proven available therapy was renin-angiotensin system inhibitors (RASi). However, there has been a recent renaissance of therapies to slow DKD progression with newer molecules like sodium glucose transport inhibitors (SGLT2i), non-steroidal mineralocorticoid blockers (ns-MRAs), and glucagon-like peptide-1 receptor antagonists (GLP-1 RAs), creating the four pillars of DKD goal-directed medical treatment (GDMT) (Neuen et al, NDT, 2025).

These drugs can slow the progression of kidney disease but not fully reverse it, indicating that there is still an unmet need in DKD. Recently, cellular therapy (i.e., regenerative medicine) has opened up new possibilities in CKD treatment, and this sci-fi-esque technology is now within reach. Mesenchymal stem cells (MSCs), present in bone marrow and other tissues, can generate any tissue depending on their surrounding milieu. MSCs are attracted to injured organs (due to cytokine signaling) and, therefore, are potential therapeutics for many chronic diseases (Farini et al,  Stem Cells Int, 2014). Unfortunately, up until now, the ability of the kidney to repair glomerular injury was limited by the kidney’s low regenerative potential, and DKD led to an inevitable downward spiral to ESKD (Bussolati et al, Nat Rev Neph, 2016). 

Although initial studies with renal progenitor cells gave positive results in animal models of CKD, their short life span and lower potential to regenerate nephrons have limited their use in kidney disease. As we are aware, there are many types of kidney cells, and some do regenerate. Clinical medicine tells us that ATN and tubule regeneration is possible over a period of days to months. Surviving renal epithelial cells, even in the absence of renal progenitor cells, can regenerate the entire tubular system following acute tubular injury (Lin et al, J Clin Invest, 2005). In addition, animal models showed that selective renal cell injections can decrease the progression of kidney disease, halting the NfKB and TGF-ꞵ progressive damage (Stenvinkel et al, KI Rep, 2016); hence, renal autologous cell therapy (REACT therapy) has attracted interest for clinical use. REACT cells are collected through renal biopsy samples and grown in culture media before they are reinjected into the kidney directly (Stavas et al, Am J Neph, 2022).

If a kidney transplant (whole organ) is the best renal replacement therapy, then is REACT (partial liquid transplant) the best preventative strategy for progressive DKD?

The Study

Methods

The REGEN-007 trial was a phase 2, multicenter, randomized, open-label trial designed to evaluate the safety and efficacy of rilparencel in adults with type 1 or type 2 diabetes mellitus and advanced CKD. The trial enrolled individuals aged 30 to 80 with an eGFR of 20 to 50 ml/min/1.73 m², a UACR between 30 and 5000 mg/g, and an HbA1c below 10%.

Inclusion and exclusion criteria

Preparation of Autologous Cells (Rilparencel)
An ultrasound-guided or CT-guided renal biopsy specimen was taken and transported to the prokidney manufacturing unit. The harvested tissue undergoes enzymatic digestion and density gradient separation to isolate a heterogeneous population of renal epithelial lineage cells (SRCs). These cells are expanded ex vivo for approximately four to six weeks under “good manufacturing practice” conditions before formulation into the final product rilparencel. The cell suspension is embedded in a thermolabile hydrogel carrier to facilitate cortical retention following injection.

Unlike most investigational regenerative approaches in nephrology, which rely on systemically administered mesenchymal cells, rilparencel represents a locoregional autologous cell therapy. The therapeutic premise is that injection of renal lineage cells directly into the kidney cortex may exert regenerative or reparative effects through paracrine signaling, modulation of local inflammation, and antifibrotic pathways. Preclinical models have suggested that these SRC populations may contain epithelial progenitor-like cells derived from multiple nephron lineages, including tubular epithelial and glomerular epithelial cells, although the exact functional contribution of these cell populations in human kidneys remains uncertain.   

The required dose is calculated depending on the kidneys’ volume by MRI. For each gram of kidney,  3×106 cells are required for transfusion. The rilparencel concentration of 100×106  per 3 ml is sufficient for 100 g of kidney tissue. The final rilparencel product was injected percutaneously under CT guidance back into the patient’s kidney cortex. 

Randomization and endpoints
The study’s primary efficacy endpoint was the change in the total (acute and chronic) slope of eGFR, calculated with the CKD-EPI 2009 equation, comparing the pre-injection period to the post-last injection period. Secondary efficacy endpoints included time to ≥40% eGFR decline, dialysis, renal or cardiovascular death, and composite renal outcomes. Safety endpoints encompassed biopsy-related complications, injection-related AEs, and product-related AEs. Analyses employed linear mixed-effects models for slope comparison and Kaplan-Meier methods for time-to-event outcomes. 

Patients were randomized 1:1 into two treatment groups.

• Cohort 1 (scheduled dosing): received two REACT injections—the first into 1 kidney and the second into the contralateral kidney approximately 3 months later.  

• Cohort 2 (triggered dosing): received one initial injection for the biopsied kidney and then evaluations every 3 months. A second injection was only administered if the patient met a “redose trigger," defined as an eGFR decline of ≥ 20% or a sustained UACR increase of ≥ 30%. If there were no triggers over a 15-month observation period following the first injection, then those patients would only receive a single injection.

Outcome Measurements, Safety Assessments, and Statistical Analysis
Efficacy analyses were performed on the modified intent-to-treat population and safety analyses on the safety set. The primary efficacy endpoint—fall in eGFR slope—was made from a linear mixed model.

Various subgroups

Primary safety endpoint: The percentage of participants with procedure-related and investigational product-related treatment-emergent adverse events (AEs and TEAEs).  Secondary safety endpoint: The percentage of participants with procedure-related death.

Determinants of 5-year and 2-year risk of ESRD

Funding Source
Funding done by ProKidney (NASDAQ: PROK), a late-clinical-stage biotechnology company involved in autologous cell therapy which makes the product, and many of the authors (including first/last) are employees of the company. The company people did all the analysis and wrote the manuscript.

Results

Between July 2021 and March 2023, 77 participants were screened across five clinical sites in the United States. Of these, 53 participants underwent randomization, with 27 assigned to Cohort 1 and 26 to Cohort 2. Overall, 45 participants completed study treatment (23 in Cohort 1 and 22 in Cohort 2).

The study population reflected a typical cohort of patients with advanced diabetic kidney disease. The mean age was 60 years, and approximately 2/3rd of the trial participants were male. The majority had type 2 diabetes (78%), and the mean baseline eGFR was 33 ml/min/1.73 m², indicating moderate to advanced CKD. Median albuminuria across the cohort was 421 mg/g. Most patients were on a RASi, around 30-40% were flozinated, with smaller numbers on a GLP1RA or an MRA. 

In the first cohort, most patients (24) received the two doses of rilparencel, while in the second cohort, 15/25 (~ 67%) needed a second injection because of eGFR decline of ≥ 20% or a sustained UACR increase of ≥ 30%. 

Primary endpoint
In Cohort 1, the annual change in kidney function in the preinjection period, as measured by the slope of eGFR, was −5.84 ml/min per 1.73 m² (95% CI, −7.97 to −3.70). In the period after the last injection, the annual change was −1.27 ml/min per 1.73 m² (95% CI, −3.97 to 1.43). The difference in the slope of eGFR between treatment periods was 4.57 ml/min per 1.73 m² (95% CI, 1.95 to 7.18). This represents a 78% improvement in the rate of eGFR decline for this group. 

In Cohort 2, the annual change in kidney function in the preinjection period, as measured by the slope of eGFR, was −3.40 ml/min per 1.73 m2 (95% CI, −5.03 to −1.77). In the period after the last injection, the annual change was −1.71 ml/min per 1.73 m2 (95% CI, −3.78 to 0.36). The difference in the slope of eGFR between treatment periods was 1.70 ml/min per 1.73 m2 (95% CI, −0.24 to 3.63).

Subgroup analyses evaluated the primary endpoint across baseline characteristics (including CKD stage, body mass index, HbA1c level, albuminuria category, and background therapy with SGLT2i or GLP-1RA). No consistent interaction between baseline characteristics and treatment response was identified, although subgroup sample sizes were small.

Secondary outcomes
Clinical events were relatively infrequent during follow-up. For the 3-component composite outcome (≥40% decline in eGFR, eGFR <15 ml/min/1.73 m², or kidney/cardiovascular death), events occurred in 7 participants (29%) in cohort 1 versus 5 participants (20%) in cohort 2.

When a fourth component (increase in albuminuria ≥30%) was included, the four-component composite outcome occurred in 10 participants (42%) in Cohort 1 and 15 participants (60%) in Cohort 2. The median event-free time for this composite outcome in Cohort 2 was 17 months.  

Risk prediction using the Kidney Failure Risk Equation (KFRE, 8 variable equation) suggested that kidney failure risk stabilized or improved in a proportion of participants following treatment. This was a prespecified endpoint of the study.

At 12 months after the first injection: 29% (7 patients) Cohort 1 and 16% ( 4 patients) Cohort 2 had the same or a lower predicted risk of ESKD compared with baseline. At 18 months, this was unchanged for Cohort 1 and had increased further to 28% in Cohort 2 showing stabilization or reduction in their predicted risk of kidney failure.

Four patients in cohort 1 and two patients in cohort 2 experienced a > 40% sustained drop in eGFR at 30 days. Additionally, six patients in cohort 1 and twelve patients in cohort 2 had a > 30% increase in UACR sustained at 90 days. The actual change in UACR is not found in the paper or supplement.

Safety
A total of 87 rilparencel injections were performed during the trial. Safety outcomes included adverse events related to 3 distinct components of the intervention: the kidney biopsy, the injection procedure, and the rilparencel itself. Procedure-related treatment-emergent adverse events occurred in 16 participants (33%). The most frequently reported events included injection site pain, renal hematoma, chills, headache, nausea. Six participants (12%) experienced product-related treatment-emergent adverse events, most of which were mild systemic symptoms such as nausea, dizziness, fatigue, or headache. 

Biopsy-related serious adverse events were reported in three participants and included subcapsular renal hematoma, acute kidney injury, and hematuria with hydronephrosis. One participant experienced a procedure-related subcapsular hematoma following injection.

Importantly, no product-related serious adverse events and no procedure-related deaths were reported during the study.

Discussion

In this phase 2, clinical trial, the use of the rilparencel was associated with a slower decline of eGFR compared to the previous trend, in patients with advanced  DKD (GFR ~ 30s, UACR ~ 400 mg/g) and a pre-infusion eGFR loss of ~ 3.4 to 5.8 ml/min/1.73m2/year. This suggests that even in such advanced disease when we usually think fibrosis and a cycle of inexorably declining kidney function has set in, injecting renal autologous cells may promote recovery or suppress further damage. In this limited sample size and follow up, rilparencel also seems to be well tolerated and safe. 

Strengths
For a phase 2 RCT, this was quite well done to demonstrate that the investigational agent has some promise of efficacy and is relatively safe. 

Limitations
No placebo or sham procedure arm was included, a decision driven largely by the procedural nature of the intervention. Because all participants required an initial kidney biopsy to manufacture the cellular product, the investigators elected to compare outcomes against each participant’s historical disease trajectory rather than against a concurrently treated control group. Consequently, the trial’s primary efficacy analysis relied on within-subject slope comparison, contrasting the pre-intervention decline in kidney function- derived from historical eGFR measurements spanning up to 24 months- with the slope observed after treatment. While this design increases statistical efficacy in small exploratory studies, it introduces important interpretive challenges, including susceptibility to regression to the mean, the Hawthorne effect, variability in historical laboratory measurements, and changes in concomitant therapies over time. Though the authors report changes in KFRE, they do not report actual pre/post UACR to allow us to understand what drove this change. 

Rilparencel or GDMT?
The present study allowed all baseline GDMT for DKD. During the study 80% of participants were on ACE inhibitors or ARBs, however only 37% were receiving SGLT2 inhibitors and 39% were on GLP-1 receptor agonists at baseline. This is actually better than average for most patients with DKD in the US. So the change in slope of eGFR attributed to rilparencel injections is not the full story. Even in 2021-2023, our optimal use of DKD medications remains low, and less than 10% of patients who would qualify for such GDMT are actually receiving all four medications.

Hence, this was not a comparison that allows us to pit rilparencel against GDMT - since GDMT is already approved, one cannot do such a trial, and it is likely that in a phase 3 RCT many (or most) patients will be on GDMT as baseline standard of care. Nevertheless, as we have seen with reported data, it is almost possible to achieve remission - or bring eGFR down slope to a level expected for age-related decline - with full GDMT. GDMT also has other systemic benefits: reduction in blood sugars, blood pressure, obesity, and cardiovascular outcomes. It is unsure (and unlikely) that rilparencel would have such benefits. However, it represents a one or two shot intervention, with no ongoing pill burden. As uptake of  GDMT medications increases, the need and effects of invasive procedures like REACT may be reserved for specific high risk DKD patients. Ultimately, the placebo-controlled phase 3 trial will determine whether rilparencel represents a new treatment option for patients with advanced type 2 diabetes mellitus and CKD.

What do other studies in this area show?
Various studies done with different cell lines in patients with AKI and CKD have been attempted, including stem cells and progenitor cells (Salybekov et al, Front Cell Dev Biol, 2024). 
These studies are not without controversy, and unfortunately  incorporate many elements that may lead to bias in observations. Current research is hindered by several common limitations including: selection of patients at different CKD stages and different underlying etiologies of the CKD. Only a few trials, such as the REACT studies, specifically target diabetes-related cases. This is problematic because different etiologies can lead to distinct types of kidney damage, which may affect the efficacy of the treatment. In addition, there remains no consensus on the optimal cell dosage. The absence of a standardized dosing protocol may lead to inconsistent results across studies, difficulties in conducting meta-analyses and systematic reviews, and challenges in clinical implementation. Finally, the small sample sizes in many studies contribute to a lack of randomization, low statistical power, limited generalizability, a higher risk of false positives, and potential biases.

Studies on autologous renal cells
Of course there are many potential advantages of autologous cell therapy over RRT and even transplantation. Being able to regenerate one’s own tissue to prevent organ failure and/or exposure to lifelong immunosuppression would be preferable to many patients. Infusions are also less stressful than organ surgery on elderly patients who might not otherwise qualify for kidney transplantation. Various studies are in the pipeline; some are completed, some are recruiting, and the majority of the studies are in the US.

Conclusion

The REGEN-007  phase 2 trial suggests that autologous renal cell therapy may alter the trajectory of kidney function in patients with diabetic CKD. The magnitude of the reported change in eGFR slope is notable, although interpretation is limited by the open-label design, reliance on historical controls, and small sample size. Whether rilparencel ultimately becomes part of the DKD therapeutic armamentarium will depend on the results of the ongoing placebo-controlled phase 3 trial, which are needed to determine whether the observed signal translates into meaningful clinical outcomes such as delayed dialysis or improved survival.

Summary by
Dr Sai Vani Yellampalli,
Consultant nephrologist 
Kurnool Kidney Care

Dr Akshaya Jayachandran
Assistant Professor, Nephrology
Christian Medical College & Hospital
Vellore, Tamilnadu, India

NephJC Interns, Class of 2025

Reviewed by
Cristina Popa, Brian Rifkin, Swapnil Hiremath

Header Image created by AI, based on prompts by  Akshaya Jayachandran

Cost-effectiveness of protein-restricted diets in advanced CKD

Vincenzo Bellizzi, Mario Fordellone, Luca De Nicola, Roberto Minutolo, Simona Signoriello, Giuseppe Quintaliani, Biagio Di Iorio, Paolo Chiodini

PMID: 40815268
DOI: 10.1093/ndt/gfaf154

What do we actually know about protein restriction?

For decades, nephrology has kept dietary protein restriction (DPR) on life support: never quite dead, never convincingly alive. Physiologically, the case is neat: less protein means less glomerular hyperfiltration, lower nitrogenous waste, less acidosis, and phosphate load, and theoretically slower scarring of nephrons (Sallstrom et al, Am J Physiol Regul Integr Comp Physiol, 2010). Experimental models and tortured MDRD secondary analyses based on achieved (and not intention-to-treat) have suggested that lower achieved protein intake is associated with slower GFR decline and longer time to kidney failure. (Levey et al, AJKD, 1996)

Clinically, things are murkier. Large RCTs of low-protein diets (LPD) versus “normal” protein intakes did not deliver the sort of clean, hard outcome wins we now expect from drug trials. Meta-analyses show that very low-protein diets (VLPD 0.3-0.4g/kd/day) supplemented with ketoacids may reduce the risk of kidney failure in non-diabetic CKD, but have uncertain effects on mortality, GFR slope, and quality of life (Jiang et al, Int Urol Nephrol. 2016| Li et al,  Nutrients. 2019). In contrast, standard “moderate” LPDs (~0.55-0.6 g/kg/day) have shown at best modest or inconsistent effects on progression (Hahn et al, Cochrane Database Syst Rev, 2020| Bellizzi et al, Am J Clin Nutr, 2022 | Bawazir et al, J Braz Neph, 2025). 

Meanwhile, the practical obstacles are enormous and well known to anyone who has tried to run a serious renal diet program:

• Adherence is the Achilles’ heel. In real-life cohorts, only about one-third of patients prescribed an LPD and one-quarter on a VLPD hit their protein targets (Bellizzi et al, Am J Clin Nutr, 2022)

• Monitoring is intensive. Achieving and maintaining restriction without protein-energy wasting (PEW) needs regular dietitian input, frequent dietary assessment, and biochemical monitoring- luxuries in many clinics.

• Nutritional risk is real. Patients with CKD spontaneously reduce energy and protein intake as the disease progresses; overzealous restriction risks accelerating sarcopenia and PEW, particularly in older or frail patients (Kopple et al, Kidney Int, 2000). Longer term follow up of the MDRD trial reported twice the mortality with LPD (Menon et al, AJKD, 2009)

• Outdated Evidence: The LPD trials were run when we had no effective interventions for CKD progression, as we do now (see below). 

Guidelines reflect this tension:

• KDOQI (Ikizler et al, Am J Kidney Dis, 2020) takes a relatively assertive stance: for metabolically stable CKD G3-G5 (non-dialysis), it recommends an LPD of 0.55-0.6 g/kg/day, or a VLPD of 0.28-0.43g/kg/day plus ketoacid/amino acid analogues, under close supervision, to reduce ESKD/death risk and possibly improve quality of life (1A). 

• KDIGO is notably more cautious. It does not support low-protein diets alone (0.4-0.6g/kg/day) as a strategy to slow progression (KDIGO, Kidney Int, 2024| NephJC summary). Meta-analyses show little or no effect on kidney failure or eGFR change. Instead, KDIGO recommends 0.8 g/kg/day for most CKD G3-G5, and only “considers” very low, protein keto-supplemented diets for selected, motivated patients at high risk of kidney failure and under strict supervision. It explicitly warns against low/very-low protein diets in metabolically unstable patients or those with sarcopenia, cachexia, or undernutrition. 

All of this plays out in an era where guideline-directed medical therapy (GDMT) for cardiorenal protection (maximal RAS inhibition, flozins, non-steroidal MRAs, and GLP1RAs) delivers large, reproducible reductions in kidney failure, cardiovascular events, and mortality across multiple RCTs. EMPA-KIDNEY (The EMPA-KIDNEY Collaborative Group, NEJM, 2023| NephJC summary), DAPA-CKD (Heerspink et al, NEJM, 2020| NephJC summary), and CREDENCE (Perkovic et al, NEJM, 2019| NephJC summary)  alone show ~30-40% relative risk reductions in kidney disease progression and kidney failure, with benefits independent of diabetes in many patients (Nuffield Department of Population Health Renal Studies Group, Lancet, 2022).

Yet, in practice, both arms of this story underperform: strict protein restriction is rarely implemented to specifications, and GDMT uptake remains anemic (flozin eligible CKD cohort use often sits <30%) (Luyckx et al, Kidney Int, 2024). The question then, is not just “does a low-protein diet work?”, but “what is the most rational comparator and where should we spend our limited implementation capital?”

The study: a polished narrative constructed on an over-selected, hand-picked cohort

The cost-effectiveness article in question (Bellizzi et al, NDT, 2025) is, in many ways, the best-case scenario for protein-restricted diets. 

The study was a prospective observational economic analysis built on previously established cohorts. It included outpatients with CKD stages 4-5 from Italian nephrology clinics. Patients were divided into:

• Intervention (PRD): 62 patients with proven adherence to prescribed protein restriction: either LPD (0.6g/kg/day) or a VLPD (0.3g/kg/day plus ketoanalogues and protein-free special foods)

• Controls: 123 matched patients on a “free” diet who had never been prescribed restriction. 

All were followed for 5 years. Outcomes included time to ESKD, time to death, and a detailed accounting of direct and indirect costs during both non-dialysis CKD and dialysis. 

The study declared no industry funding. However, one author reports consultancy/ speaker roles with Fresenius Kabi, the manufacturer of ketoanalogues (ketosteril), a central therapeutic component of supplemented very-low protein diets. Additional author ties include companies active in CKD, dialysis, anemia, and metabolic therapeutics. 

The results follow a sequential structure that mirrors cohort construction, clinical outcomes, and downstream economic modelling. 

The study initially included 1471 patients with CKD: 223 exposed to protein-restricted diets and 1248 controls on a free diet. Only patients with proven dietary adherence were retained in the intervention arm. This reduced the analytic PRD cohort to 63 individuals; after propensity matching, the final comparison consisted of 62 PRD and 123 controls.

Baseline characteristics in the matched sample were balanced:~67 years, ~60% male, diabetes ~29%, cardiovascular disease ~35%, BMI ~26kg/m2, and eGFR of ~16-17mL/min/1.73 m2, with roughly 45% on CKD stage 5 (table 1).  

Follow-up duration was longer in the PRD cohort (~106 vs 89 months). Median time to kidney failure was longer with PRD 48.6 months vs 28.8 months (log-rank P=0.017).

Mortality outcomes followed a similar pattern. Median survival reached 107 months in PRD vs 86.6 months in control (P=0.004). Event proportions reflected these trajectories, with fewer deaths and later ESKD onset in the diet-adherent cohort.

Economic analyses were anchored in phase-specific cost structures. During the non-dialysis CKD phase, monthly costs were higher with dietary restriction: €383 for free diet, €507 for LPD, and €767 for VLPD. The incremental expenditure derived from protein-free foods, ketoanalogues supplementation, more frequent visits, and monitoring. 

Once dialysis began, costs rose sharply to approximately €4150 (~$4900) per patient per month, exceeding conservative-phase costs by several fold.

Two economic models were applied. In the theoretical model, which assumes survival to dialysis thresholds without accounting for attrition, protein restriction increased short-term costs but reduced long-term expenditures. Annual costs were ~80% higher early in CKD, but ~64% lower over longer horizons, producing an overall cumulative saving of ~43%.

The observed model, incorporating deaths, dialysis initiation, and follow-up losses, showed smaller savings. Annual costs were reduced by ~30% in the short follow-up and ~28% in the longer follow-up, with overall reductions of ~25%. Cost divergence emerged as more control patients entered dialysis earlier and remained longer in the high expenditure phase.

Across analyses, the economic signal tracked dialysis exposure. Delayed kidney failure translated into a prolonged time in the lower-cost conservative phase and reduced cumulative dialysis spending.

Discussions: 21st-century view on protein restriction in CKD

Bellizzi and colleagues have produced an impressive piece of work concerning the economics of a treatment that belongs to another century. They take a small, exquisitely selected subgroup of patients with stage 4-5 CKD who are willing to live for years inside a narrow biochemical corridor of protein intake, surround them with dieticians, protein-free foods, and ketoanalogues, and document what happens: dialysis comes later, remarkably death comes later, and the health system spends less on machines and membranes per unit of survival. Within that closed system, the logic is flawless. The unease comes from recognizing that the system itself is an artifact of a therapeutic worldview that has already been displaced. 

The lineage is not in doubt by far. Addis’s early experiments with severe nitrogen restrictions made it clear that, if one supply can be coaxed in a state of “minimum endogenous nitrogen metabolism” and held there, it will be at the cost of “mounting aversion and eventual loss of courage on the part of the subject” (Smith M, J Biol Chem, 1926| Obeid et al, Kidney360, 2022). The Giordano-Giovannetti diet translated into clinical practice 18-20 g of protein a day, carefully balanced amino acids, and in return a striking disappearance of classic gastrointestinal torments of uremia (vomiting, hiccups, anorexia, diarrhea), while blood urea levels fell and patients enjoyed a somewhat longer "comfortable life”…… before inexorable death (Berlyne et al, Lancet, 1965). The promise was never that the kidney would scar less; it was that the dying would be biochemically tidier. Barry Brenner’s hyperfiltration hypothesis then conferred upon this ascetic tradition a physiologist’s elegance. High protein intake was shown in animals to increase single nephron GFR and intraglomerular pressure; protein restriction mitigated these surges and delayed the development of glomerulosclerosis (Brenner et al, NEJM, 1982| Brenner et al, Kidney Int, 1996). In the absence of dialysis and any drug capable of altering glomerular hemodynamics, it was rational and almost inevitable, to turn the plate. 

What happened when this physiological conviction was finally submitted to the discipline of the randomized human trials is, retrospectively, not flattering. Across Rosman (Lancet, 1984), Ihle (NEJM, 1989), Locatelli (Lancet, 1991), Hansen (Kidney Int, 2002), Cianciaruso (AJKD, 2009), and, most decisively MDRD, the pre-specified endpoints are remarkably stubborn. Low-protein diets do not reliably reduce the rate of dialysis initiation, nor do they produce a sustained, clinically important slowing of GFR decline when GFR is measured directly rather than inferred from creatinine (Mehrotra, Miner Electrolyte Metab, 1999| Bawazir et al, J Braz Neph, 2025). In MDRD, low versus “usual” protein in patients with moderate CKD (0.58 vs 1.3 g/kg/day) and very low vs low protein in advanced CKD (0.28 vs 0.58 g/kg/day, with ketoacid supplements) failed to reduce time to ESKD or death in intention-to-treat analysis (Levey et al, AJKD, 1996). The methodological innovation of MDRD (measured iothalamate clearance) had the unfortunate virtue of showing that much of the apparent improvement seen in earlier creatinine-based studies was not a slowing of nephron loss at all, but a biochemical illusion created by reduced creatinine generation under protein restriction (Levey et al, AJKD, 1996| Bawazir et al, J Braz Neph 2025).

Meta-analysis has not rescued the paradigm. Kasiske et al (AJKD, 1998) combined more than 1900 patients and found that protein restriction slowed GFR decline by roughly 0.5 ml/min/year on average, a statistically significant but clinically modest effect that did not translate into a robust reduction in kidney failure. Hahn’s Cochrane review concluded that low-protein diets probably have little or no effect on the composite of kidney failure or death in non-diabetic CKD, and that the seemingly beneficial effect of VLPD over-simplifies avoiding high protein and is fragile and confounded (Hahn et al, Cochrane Database Syst Rev, 2020). Trials in diabetic nephropathy are even less forgiving: a small signal here, an equivocal slope there, none of it rising to the standard we routinely apply to therapies that claim to modify renal prognosis. The hyperfiltration story is not disproven so much as rendered insufficient: glomerular physics is not destiny when one is constrained by human behavior and the ethics of long-term semi-starvation.

Even this would be tolerable if the intervention itself were benign. From the earliest trials onward, the cost of sustained protein restriction has been paid in acceptance and flesh. In Rosman’s multicenter study, a third of patients rated the low-protein diet bad at 3 and 6 months, and substantial numbers withdrew for intolerance or non-cooperation (Rosman et al, Lancet, 1984). Locatelli’s comparison of 0.6 vs 1 g/kg/day reported 64 withdrawals, the majority for inability or unwillingness to comply with low-protein food (Locatelli et al, Lancet, 1991). Even with a carefully curated population, both protein and energy intake drifted downward, accompanied by reductions in body weight, arm muscle area, and transferrin, even as albumin remained in the normal range and gave the comforting appearance of safety (Chauveau et al, AJKD, 1999). Mehrotra’s review of these data is blunt: in the best hands, in non-diabetic, non-frail patients, long-term protein diets produce small but consistent signals of subclinical malnutrition, and there is every reason to believe that the margin of error is narrower in older and more comorbid patients who now populate our CKD clinics (Mehrotra, Miner Electrolyte Metab, 1999). To persist in calling such regimens conservative care as patients involuntarily lose lean mass feels, at some point, more like a euphemism than medicine. 

It is into this already crowded and ambiguous landscape that Bellizzi’s cost-effectiveness analysis arrives. The authors do something few have had the patience or infrastructure to do: they identify only those patients who actually adhere to a prescribed low- or very-low protein diet, achieve intake within 0.2 g/kg/day of the target, and follow them over five years, comparing propensity- matched patients who never received dietary restriction. Roughly a quarter of those originally prescribed the diet qualify as “adherent” by this definition. Under the care of Italian nephrology units with specialist dietitians and access to protein-free medical foods, these adherent patients enjoy significantly longer times to ESKD (48.6 vs 28.8 months) and death (107 vs 86.6 months) than their free eating counterparts. Their monthly costs during the non-dialysis phase are higher (€507 for LPD and €767 for supplemented VLPD vs €383 with no diet), but once dialysis begins at roughly €4150 per month, the arithmetic becomes trivial (Bellizzi et al, NDT, 2025). This kind of analysis is picking low protein cherries to the extreme. Data obtained from only adherent patients (a kind of per-protocol, not intention to treat), and even that from an observational study, should not satiate anyone. It only provides biased estimates. Patients adherent to a specific diet usually have other underlying characteristics (education, socioeconomic status, adherence to other therapeutic interventions) that make for the real mechanistic effect.

If one’s worldview is constructed around dialysis as the unavoidable abyss, and grams of protein as one of the few adjustable variables, this is triumphant. But it’s very coherence exposes how far removed it is from what now constitutes contemporary CKD care. The comparator is not “modern GDMT with or without serious dietary restrictions”. It is “diet plus a largely pre-GDMT standard of care” versus “no diet, plus same”. The physiological problem the diet is solving - hyperfiltration and its hemodynamic sequelae- is the same problem now addressed, more potently and reproducibly, by RAAS inhibitors and flozins (Brenner et al, Kidney Int, 1996| Iman et al, J Ren Nutr, 2025| Wanner et al, Adv Ther, 2025). ACE inhibitors and ARBs lower efferent arteriolar resistance, reducing intraglomerular pressure, proteinuria, and progression. Flozins enhance distal sodium delivery, engage tubulo-glomerular feedback, and constrict the afferent arteriole, producing an immediate drop in GFR followed by a slower decline- a kinetic pattern eerily reminiscent of the hoped-for effect of low protein diets, but achieved without asking the patient to consider every morsel of food. Large trials like EMPA-KIDNEY (The EMPA-KIDNEY Collaborative Group, NEJM, 2023) and DAPA-CKD (Heerspink et al, NEJM, 2020) show a 30-40% relative risk reduction in kidney failure and sustained GFR decline with parallel reductions in heart failure and cardio-vascular death, on top of RAS blockade, in diabetic and non-diabetic CKD alike (Nuffield Department of Population Health Renal Studies Group, Lancet, 2022). Ns-MRAs and GLP1RA add further, if more population-specific, renal and cardiovascular protection.

KDIGO 2024 recognizes this hierarchy explicitly. It recommends protein intake around 0.8 g/kg/day for most adults with CKD, warns against low- or very low protein diets in metabolically unstable patients, and judges the certainty of evidence for such diets on hard outcomes to be low and very low (KDIGO, Kidney Int, 2024| NephJC summary).

The same guideline confers its highest grade of recommendation (1A) to flozins as kidney-protective therapy and frames RAASi, SGLT2i, ns-MRA, and GLP1RA as the backbone of disease-modifying treatment in cardio-renal disease. KDOQI nutrition guidelines, by contrast, emerge from a different epistemic culture (Ikizler et al, AJKD, 2020). It grants strong recommendations to LPD and VLPD with or without ketoanalogues in CKD stages GA3 and G5, leaning heavily on small RCTs with surrogate endpoints, and embeds these diets as central tools of kidney protection rather than futile adjuncts. The divergence is not about data, is about what one wrongfully counts as a convincing signal, and how one weighs mechanistic plausibility against existence of therapies that have already cleared much higher evidentiary thresholds. A recent meta-analysis shows that adding ketoanalogues to LPD or VLPD improves GFR trajectories, reduces urea and phosphate, and maintains albumin and muscle indices over a median of 13 months (Chen et al, J Nephrol, 2024). There’s, however, no significant reduction in all-cause mortality and no statistically robust reductions in ESKD across the full CKD population. The trials are small, short, and methodologically heterogeneous. Safety is conditional on adequate caloric intake and careful monitoring, and the supplements themselves are expensive, in some settings approaching the cost of a year of dialysis when all inputs are counted (pills, dietitian time, repeat visits). It is telling that an expert RAND/UCLA panel, faced with this evidence, declared ketoanalogue supplementation “appropriate” in many scenarios on the basis of consensus rather than outcome data (Saavedra-Fuentes et al, Nutrients, 2024). This is not the profile of a foundational therapy, is the profile of an intricate scaffold built on still-damp concrete. 

What Bellizzi et al therefore show, perhaps more starkly than they intended, is not that protein restriction should reclaim the center of CKD management, but that it can still perform impressively under highly contrived conditions. If one is willing to build and fund an infrastructure of specialist dietitians, subsidized protein-free products, monthly visits, and detailed intake monitoring; if one is content to compare this against a backdrop in which RAASi and flozins are neither universal nor always optimized, and you choose to include data from patients who follow through all this - then yes, a few years of dialysis may be bought back and the balance sheets will look better. The absurdity, viewed from the vantage point of 2025, is that we continue to invest experimental and policy energy into perfecting this apparatus of controlled deprivation while simultaneously failing to deliver once-daily pills with high-certainty, trial-proven benefits to vast numbers of eligible patients. It’s easier to model the economics of a gram of protein than to solve the organizational problem of why so many patients with significant albuminuria never see a flozin.

LPD and VLPD are less marginal than outdated. In advanced CKD, asking patients to metabolically restrict themselves makes little sense when the same hemodynamic targets can be addressed pharmacologically. The so-called “ideal candidate”- stable, motivated, dietitian-supported- highlights how narrow, and not how relevant the strategy remains. At scale, it doesn’t hold. Adherence fails, infrastructure varies, and patients with advanced CKD are already nutritionally vulnerable. Dialysis delay that comes with muscle loss or frailty is a questionable trade. From that angle, Bellizzi et al reads like a detailed documentation of what dietary deprivation could achieve before pharmacological protection became routine. 

                                                                                                           Summary by

                                               Cristina Popa

Reviewed by

Swapnil Hiremath and Brian Rifkin

#NephJC Chat

Tuesday, February 25th 2026, 9 pm Eastern on Bluesky

Kidney Int 2026 Jan;109(1S):S1-S99. doi: 10.1016/j.kint.2025.06.006.

KDIGO 2026 Clinical Practice Guideline for the Management of Anemia in Chronic Kidney Disease (CKD)

Kidney Disease: Improving Global Outcomes (KDIGO) Anemia Work Group: Marcello Tonelli, Jeffery S Berns, Biykem Bozkurt, Rebecca S Cheung, Yarieli Cuevas, Emmanuel E Effa, Michele F Eisenga, Steven Fishbane, Yelena Z Ginzburg, Volker H Haase, S Susan Hedayati, Siah Kim, José A Moura-Neto, Evi V Nagler, Patrick Rossignol, Manisha Sahay, Tetsuhiro Tanaka, Angela Yee-Moon Wang, David C Wheeler, Jodie L Babitt

PMID: 41485812

DOI: 10.1016/j.kint.2025.06.006

Introduction

Anemia is one of the most common and troubling issues faced by people living with chronic kidney disease (CKD). Anemia is associated with higher cardiovascular risk and mortality, and drives healthcare utilisation plus erodes quality of life. Patients describe it in concrete terms: exhaustion that sleep does not repair, breathlessness with minor exertion, dizziness, headaches, cognitive slowing, loss of appetite, and low mood. These symptoms overlap with CKD itself, blurring attribution and often delaying targeted treatment. The result is a functional decline that feels disproportionate to laboratory values. 

Anemia advances quietly as kidney function recedes, following an insidious arithmetic. Overall, anemia is twice as prevalent as in the general population, with rates rising progressively as kidney function declines (Stauffer et al. PLOSOne, 2014). In the earliest stages, it is present in roughly 8-15% of the patients, by CKD stages 3-4, prevalence climbs to 30-50%, and in CKD stage 5 often 50-80% of patients are anemic. Within dialysis units, it is nearly universal, exceeding 90% in many cohorts (Stauffer et al, PLOS One, 2014; Hashimi et al, StatPearls Publishing, 2025; Kim et al, Nephrology (Carlton), 2025). Regional variability in anemia rates reflects demographic structure, socioeconomic gradients, nutritional status, comorbidity burden, and access to screening (Gardner et al, Lancet Global Health, 2023). Treatment intensity mirrors these disparities. Some health systems deploy iron supplementation and erythropoiesis-stimulating agents (ESAs) early and systematically; others struggle with delayed recognition and therapeutic inertia. Therefore, anemia in CKD is also shaped by the systems of care, as well as a myriad of other biologic factors.  These challenges underscore the critical importance of evidence-based management strategies to mitigate anemia's impact on patient outcomes. The Kidney Disease: Improving Global Outcomes (KDIGO) has recently released the 2026 Clinical Practice Guideline for the Management of Anemia in Chronic Kidney Disease, the first comprehensive update to the KDIGO guidelines since 2012. This guideline incorporates over a decade of new evidence on anemia pathophysiology, diagnostic approaches, iron therapy, erythropoiesis-stimulating agents (ESAs), hypoxia-inducible factor–prolyl hydroxylase inhibitors (HIF-PHIs), and red blood cell transfusions. It emphasizes individualized, patient-centered care—balancing benefits, risks, symptoms, and preferences—while providing practical recommendations and practice points applicable across CKD stages, including dialysis and transplant populations.

And let's give a massive shoutout to the science flex behind those HIF-PHIs. Back in 2019, the Nobel Prize in Physiology or Medicine went to Gregg Semenza, Sir Peter Ratcliffe (a nephrologist), and William Kaelin Jr. for cracking the code on how cells sense and adapt to oxygen levels—straight-up discovering and decoding the hypoxia-inducible factor (HIF) pathway. It’s not just decreased erythropoietin production, but rather a lack of localized hypoxia due to loss of kidney cellular activity, driving anemia.

Chapter 1 

Diagnosis and evaluation of anemia in people with CKD

Anemia definition

KDIGO 2026 retains the WHO-derived hemoglobin thresholds Hb <13 g/dl in men, and <12g/dl in women (recommendation 1.1.1). The definition is explicitly not graded. The work group states that there’s no new evidence to justify the revision. This preserves epidemiologic continuity with KDIGO 2012, and NICE 2021.

The retained thresholds are age-agnostic and largely statistical rather than mechanistic. A hemoglobin of 12.5 g/dL in a young woman is biologically different from the same value in an octogenarian with multimorbidity, yet KDIGO applies a single cut-off. Nor does the guideline address the conceptual tension between diagnosis and treatment: a patient with CKD stage G3 and one with G5D with identical hemoglobin carry the same diagnostic label, but therapeutic implications differ radically. Pregnancy-specific thresholds are not incorporated (see table 1 from Popa, Piccoli, Kidney Int Rep, 2024). 

Diagnostic framework

Practice points 1.2.1-1.2.3 expand the initial work-up to include CBC, with reticulocyte count, ferritin, TSAT, CRP, vitamin B12, folate, and TSH, with peripheral smear when unrevealing. This is an upgrade to a much narrower 2012 focus on iron and EPO deficiency. The 2026 approach recognizes anemia in CKD as layered: inflammatory, nutritional, endocrine, hemorrhagic, and marrow-related.

Yet, the algorithm remains underspecified. The sequence of testing is not evidence-derived. CRP is listed conditionally, although inflammation is central to hepcidin-mediated iron sequestration. Reticulocyte hemoglobin content and percentage of hypochromic red cells are acknowledged as mechanistically superior to ferritin and TSAT for assessing iron availability, but are not recommended for routine use due to cost and availability. Hepcidin, the biological pivot of the chapter, is not measured. The guideline invokes a mechanism without equipping clinicians to confirm it.

Updated KDIGO Nomenclature for Iron-Deficient States

Anemia in CKD is more than simply limited iron availability for red blood cell (RBC) production. This arises from multiple factors: ongoing blood loss, nutritional shortfalls, drugs impairing iron absorption (e.g., PPIs), elevated hepcidin (the iron-regulatory hormone that blocks dietary iron uptake and sequesters iron in stores), and increased iron demand during ESA-driven erythropoiesis. The complexities of iron absorption and utilization, EPO production and bone marrow output are visualized below.

These processes lead to two distinct iron-deficient states in CKD. The KDIGO Work Group updates the terminology for greater pathophysiological precision:

• Previously “absolute iron deficiency” → now systemic iron deficiency (low TSAT + low ferritin; true depletion of circulating and storage iron).

• Previously “functional iron deficiency” → now iron-restricted erythropoiesis (low TSAT despite normal/high ferritin; iron trapped in stores, often hepcidin-driven, limiting erythroid use despite apparent adequacy).

This shift better explains why iron supplementation can boost hemoglobin and reduce ESA needs, even when stores appear replete.

Monitoring

KDIGO 2026 recommends annual hemoglobin assessment in CKD G3, biannual in G4-G5 non-dialysis, and every 1-3 months in G5D. Iron parameters should be reassessed during therapy, with withholding considered when ferritin exceeds 700 ng/mL or TSAT ≥40%. These intervals reflect clinical logic: anemia accelerates as kidney function declines. No randomized data define these frequencies. The “1-3 months” range leaves room for practice variation. The guideline specifies absolute thresholds but doesn’t address rate-of-change or lab variability. Monitoring is sensible and empirically derived.

Chapter 2

Use of iron to treat iron deficiency and anemia in people with CKD

​​Two issues dominate this chapter: when to start iron and how to give it. KDIGO 2026 answers both, but with explicit acknowledgment that much of the evidence remains of low certainty. 

This is the updated threshold from recent KDIGO anemia guidelines—a higher ferritin cutoff than older versions, reflecting real-world data on iron repletion in dialysis without excessive risk of overload. The 2D grade means a weak recommendation (benefits likely outweigh risks/burdens in most, but evidence quality is low-moderate), so individualization remains key: think symptoms, ESA responsiveness, inflammation, infection history, etc. 

KDIGO’s 2.1 adopts the same ferritin ≤500 ng/mL/ TSAT ≤30% cut-off that existed in the 2012 guideline, but now couples it with a proactive IV-iron regimen (400 mg monthly, unless ferritin >700 ng/mL). The logic rests almost entirely on the PIVOTAL trial (Macdougall et al, NEJM 2019| NephJC summary). PIVOTAL (2141 patients with incident HD) compared proactive IV iron with a reactive protocol that gave iron only when ferritin fell <200ng/mL or TSAT <20%. The proactive arm used less ESA (~25% lower) and achieved a modest hemoglobin advantage. The primary composite cardiovascular endpoint showed an HR of 0.85, but the 95% CI (0.73-1.00) demonstrating a CV benefit with proactive iron approach (whether driven by more iron, less ESA, or higher Hgb will remain a topic of debate). KDIGO extrapolates the trial’s protocol (withholding iron only when ferritin >700ng/mL), which is reasonable given that PIVOTAL demonstrated it is safe to go beyond the arbitrary 500 to 700. Interestingly, the smaller and shorter DRIVE trial is ignored (Coyne et al, JASN 2007), which demonstrated the safety of IV iron to a ferritin of 1200, though without any CV outcome data.

The 2.2 recommendation that IV iron is superior to oral in HD is supported by small RCTs showing superior hemoglobin response and lower ESA doses with IV iron. Yet, those studies also share KDIGO’s limitation: they measure surrogate outcomes (Hb, ESA dose) and are short-term; no trial has shown a mortality or hard cardiovascular benefit from IV versus oral iron in dialysis.

The guidelines prefer the intravenous over the oral route of administration among patients receiving hemodialysis. 

The permissive non-dialysis thresholds

Recommendation 2.3 dramatically lowers the ferritin bar for non-dialysis CKD (to <100 ng/mL) or 100-299 ng/mL with a tighter TSAT<25%. The change is driven by the FIND-CKD (Macdougall et al, NDT, 2014) and REVOKE (Agarwal et al, Kidney Int, 2015| NephJC summary) trials, both tiny. FIND-CKD showed that IV ferric carboxymaltose raised Hb by ~0.7 g/dL more than oral iron over 12 weeks; REVOKE showed a similar modest benefit, but reported more adverse events with IV Iron. Neither trial tested the specific thresholds that KDIGO now recommends, and both were underpowered to detect hard outcomes. The 2D grade, therefore, reflects that KDIGO took the inclusion criteria of these trials (ferritin < 300ng/mL) and turned them into treatment cut-offs.

This recommendation acknowledges the uncertainty by urging shared decision-making. It correctly notes that oral iron is cheaper, less invasive, and often sufficient, while IV iron gives a bigger Hb boost. The recommendation’s strength (2D) mirrors the modest efficacy gap (~0.3-0.7g/dL) and the lack of mortality data. In practice, many clinicians will try oral iron first, reserving IV for intolerance or failure after 1-3 months- exactly what KDIGO suggests, albeit in a more verbose, less decisive fashion.

Upper safety limits- where KDIGO’s caution may be too cautious

KDIGO’s practice point to stop iron when ferritin is  >700 ng/mL or TSAT≥ 40% is taken straight from the PIVOTAL protocol.  Observational dialysis cohorts do show higher infection and mortality rates at ferritin >700-800 ng/mL, but ferritin is an acute-phase reactant; the association might be confounded by inflammation. The DRIVE trials (Coyne et al, JASN, 2007), including patients with ferritin 500-1200 ng/mL, demonstrated no excess serious adverse events with short-term high-dose iron, hinting that the 700 ng/mL ceiling might be arbitrary.

What do other guidelines say?

• NICE (London: National Institute for Health and Care Excellence (NICE); 2021) avoids numeric cut-offs altogether, simply recommending high-dose iv iron to spare ESA. 

• ERBP (Locatelli et al, NDT, 2013; Stoumpos et al, NDT, 2024) prefers the older, more conservative thresholds (absolute deficiency: ferritin <100ng/mL, TSAT <20%). The 2026 commentary (Del Vecchio et al NDT 2026) endorses KDIGO’s proactive HD approach but cautions against liberal iron in patients with high CRP, effectively re-introducing inflammation-adjusted decision-making that KDIGO 2026 glosses over. 

• UKKA is the most conservative: iron only for absolute deficiency or when ESA is already required, and it sets the upper ferritin limit at 600 ng/mL (Bhandari et al, BMC Nephrol, 2025). 

Chapter 3

ESA/HIF-PHI Initiation, Targets & Management: Personalized & Cautious Approach.

For four decades, Erythropoiesis-Stimulating Agents (ESAs) have dominated CKD anemia management. They reliably raise hemoglobin, reduce transfusion need, and improve quality of life, outcomes demonstrated in many trials. Then came the hypoxia-Inducible Factor–Prolyl Hydroxylase Inhibitors (HIF-PHIs), oral agents that stabilize HIF- ⍺, upregulate endogenous erythropoietin, and reduce hepcidin-driven iron sequestration. They promise convenience, mechanistic elegance, and potentially reduced iron requirements. Yet their long-term safety remains uncertain.  The 2026 KDIGO update attempts to reconcile the long-term safety data of ESAs with the newer, oral, but safety-uncertain HIF-PHIs. The basic clinical dictum of correcting potentially reversible causes of anemia before initiation of these agents remains constant and has been well emphasized by the guidelines.  Many patients won't need additional medications if iron is optimized. Discuss symptoms, transfusion risk, and medication risks (stroke, thrombosis, cancer progression) with patients when trying to make an anemia plan.

ESA recommendations

KDIGO 2026 maintains the same hemoglobin threshold for ESA initiation in dialysis (HD or PD) patients as did the 2012 guideline: start therapy when hemoglobin falls to or below 9-10 g/dL. This recommendation carries grade 2D (weak recommendation/very low certainty), signaling that while evidence supports starting ESA in this range, the certainty is limited. The rationale draws from observational and trial data showing that untreated dialysis patients rapidly progress to Hb <8g/dl, with associated transfusion requirement and poor functional status. An early landmark Canadian trial randomised patients on dialysis with Hb <9 g/dL to placebo, low-target ESA (Hb 9.5-11 g/dL), or high-target ESA (Hb >11 g/dL) and found that 58% of placebo recipients required transfusion within 8 weeks, compared with 2.5% in each ESA arm (Canadian Erythropoietin Study Group, BMJ, 1990). This trial established the principle that ESA prevents transfusion dependency, but it doesn’t answer whether Hb 9.5%g/dL is superior to 10.5g/dL as an initiation threshold, nor does it validate the exact 9-10g/dL window. 

For non-dialysis CKD, KDIGO 2026 takes a different posture, abandoning a fixed threshold in favor of shared decision-making. The guideline suggests considering the presence of symptoms attributable to anemia, the potential benefits of higher hemoglobin, and the potential harms of RBC transfusion or ESA therapy. This echoes the KDIGO 2012 recommendation, but is now graded 2D (vs 1B in the 2012 recommendation 3.2, and 3.4.2- 2C), making explicit the very low certainty. In practice, this means ESA initiation might occur at Hb 8.5 g/dl in high-risk patients (recent stroke, thrombosis history, active cancer with curative intent) or at Hb 10 g/dL in a symptomatic patient pursuing a kidney transplant. The guideline does not specify exact thresholds for these subgroups; instead, it defers to clinician judgement informed by shared discussion. This transparency about uncertainty is commendable, but creates ambiguity. Two nephrologists facing a non-dialysis patient with a Hb of 9.5 g/dL, dyspnea, or prior stroke could reach opposite decisions under this framework, and both could be guideline-concordant. 

KDIGO 2026 recommends maintaining Hb below 11.5 g/dl in adults on ESA therapy. This is the guideline’s strongest recommendation (1D- strong recommendation/very low certainty), a rare combination reflecting both mechanistic conviction and explicit acknowledgment of weak evidence. It’s very logical: ESA trials targeting higher hemoglobin (≥12.6-14 g/dL) consistently increased stroke, vascular access thrombosis, and cardiovascular events compared with lower targets (10-11 g/dL). The CHOIR trial randomized 1432 patients with non-dialysis CKD to Hb targets of 13.5 vs 11.3 g/dL and was terminated early after detecting a hazard ratio of 1.3 for the composite of death, MI, heart failure hospitalization, or stroke in the high-target arm (Singh et al, NEJM, 2006). The TREAT trial in diabetic CKD similarly found increased stroke with the darbepoetin targeting Hb ≥ 13g/dL (Pfeffer et al, NEJM, 2009). Notably, CHOIR and TREAT were trials of ESA dosing strategies, in addition to Hb targets (TREAT required rescue doses of darbepoetin if Hb < 9 g/dL). The achieved hemoglobin in the “high target” arms was often lower than intended (e.g. CHOIR achieved 12.6 g/dL in the high target group, despite aiming for 13.5), and one still cannot help but ask: was it the ESA or the “high” hemoglobin target that was problematic? Why does it matter? If the culprit is ESA dose rather than Hb level, then a non-dialysis patient who achieves Hb 13 g/dL on very low ESA doses might face less risk than current evidence suggests. This was also the era of lower iron targets (see above) so that also should be taken into considerations. With more (proactive) iron use, ESA doses are typically lower.

The guidelines also suggest that pediatric and transplant populations may tolerate higher Hb targets without equivalent toxicity, citing lower baseline cardiovascular risk and distinct clinical priorities (avoiding allosensitization, optimizing growth) (Portoles et al, Clin Kidney J, 2025). However, no pediatric RCTs have tested ESA safety at different Hb targets, so the recommended range for children (implicitly similar to adults, but individualized) rests on observational data and extrapolation.

ESA hyporesponsiveness

In clinical practice, ESA hyporesponsiveness is defined by the inability to achieve target hemoglobin levels despite significant dose escalations or the requirement of persistently high doses to maintain stability. This condition affects approximately 12.5% to 30.3% of the CKD population, though prevalence varies globally (Macdougall et al, Am J Nephrol, 2023).

Crucially, hyporesponsiveness is not merely a laboratory hurdle; it is a potent clinical marker. Clinically, it creates a significant testing burden of invasive testing for patients. Additionally, patients who exhibit a poor response to ESAs face significantly higher risks of cardiovascular events, progression to kidney failure, and all-cause mortality.

Key characteristics of ESA hyporesponsiveness:

• Temporal nature: It can be acute or chronic (defined as lasting >4 months). It is often dynamic and transient, making it a difficult clinical target to treat.

• Etiology: The underlying causes are complex and multifactorial. Common drivers include absolute or functional iron deficiency, chronic inflammation (the "anemia of inflammation"), and secondary hyperparathyroidism.

• The "unexplained" gap: Despite thorough investigation, an evident cause may not be identified in approximately 30% of cases, highlighting the limitations of current management strategies.

Managing these patients requires a shift from simply "chasing the number" with higher doses, which may itself increase the risk of toxicity. Potentially reversible causes of anemia—particularly iron deficiency—should first be identified and treated.

If hyporesponsiveness persists despite evaluation and correction, management depends on anemia severity and symptoms: continue monitoring if mild and asymptomatic, or consider a 3–4 month trial of a HIF-PHI  (stop HIF-PHI if Hb unchanged, ungraded) and/or transfusion as clinically indicated.

ESAs as first line, HIF-PHI as alternative

This is a weak recommendation (2D) that places higher value on ESA’s long track record of safety and efficacy than on HIF-PHIs’ putative advantages. The rationale acknowledges that heat-to-head RCTs have shown non-inferiority of HIF-PHIs may carry increased thrombotic risk, particularly vascular access thrombosis and acute coronary syndrome, in certain populations. The very-low-certainty grade reflects genuine uncertainty: the data do not conclusively show that ESAs are better than HIF-PHIs for hard outcomes, only that they have more extensive safety tracking. This is a conservative but defensible position in the face of incomplete information. 

HIF-PHI role

If ESAs are the reliable "old guard," Hypoxia-Inducible Factor–Prolyl Hydroxylase Inhibitors (HIF-PHIs) are the shiny new oral disruptors. But with new mechanisms come new rules. Here’s the "CliffsNotes" version of how to use them without getting burned.

HIF-PHIs serve as a strategic pivot rather than a first-line default, offering a vital oral alternative for patients with ESA hyporesponsiveness, intolerance, or logistical barriers to injections. When switching, clinicians should utilize the same hemoglobin triggers and targets established for ESAs, following a "start low, go slow" titration to the minimum dose required for symptom relief and transfusion avoidance. Crucially, these agents must be used as a monotherapy; combining ESAs and HIF-PHIs is strictly avoided due to a lack of safety data.

Vigilance is required regarding off-target effects, necessitating extreme caution in patients with Polycystic Kidney Disease, proliferative retinopathy, or pulmonary arterial hypertension, and a definitive avoidance during pregnancy. Monitoring extends beyond monthly hemoglobin checks; if a meaningful response is not achieved within 3–4 months, the drug should be discontinued rather than escalating to potentially toxic doses. Finally, Roxadustat specifically requires thyroid monitoring during the first trimester of treatment to screen for central hypothyroidism. Ultimately, success with HIF-PHIs relies on a clear exit strategy and strict adherence to safety protocols. As the global availability of HIF-PHI molecules varies significantly by region, ongoing local post-marketing surveillance will be essential to clarifying their long-term safety and efficacy profiles across diverse populations. 

Some of the shiny new drugs image has been tarnished with the rocky Roxadustat rollercoaster (See NephJC summary and Podcast). Subsequent -dustats have demonstrated non-inferiority quite well, but not clear superiority hence these measured suggestions.

Chapter 4

 Red blood cell transfusions to treat anemia in people with CKD

While modern erythropoietic agents and iron therapy have reduced blood bank reliance, RBC transfusions remain a vital but risky necessity in CKD. A restrictive strategy is now favored, moving away from arbitrary hemoglobin triggers toward clinical necessity. This shift balances immediate stability against significant risks like circulatory overload lung injury (TRALI), and—most critically for CKD—immunologic sensitization. This alloimmunization can devastate transplant prospects by increasing antibody levels, narrowing donor pools, and raising rejection risks. Therefore, clinicians must weigh acute life-saving benefits against long-term transplant-threatening harms through shared decision-making (practice point 4.1). 

Transfusions are clearly indicated for life-threatening crises, such as acute hemorrhage or unstable coronary disease. In chronic settings, they are reserved for cases where ESAs or HIF-PHIs are ineffective—due to bone marrow failure—or contraindicated, such as active malignancy. Crucially, the decision should be symptom driven rather than numerical. While guideposts like <7 g/dL for asymptomatic inpatients or <8 g/dL for those with cardiovascular disease exist, the patient's actual clinical trend and symptoms, like dyspnea or angina, take precedence.

Practice point 4.5 specifies that transfusion is indicated when the benefit is clearly outweighed by the risk in acute crises (massive hemorrhage, unstable coronary syndromes, or perioperative anemia correction). The wording mirrors the older KDIGO 2012 stance (2C), but was now downgraded to a practice point. Minimizing transfusion requirements involves proactive systemic and bedside strategies. This includes standardized iron deficiency correction, consistent use of maintenance therapies during hospitalization, and limiting unnecessary phlebotomy. By educating patients and utilizing decision aids, clinicians ensure that every transfusion is a targeted, value-aligned choice that protects both current stability and future transplant eligibility.

Discussion

We stand, as ever, on the narrow ledge peering into the dim recesses of CKD anemia. The 2026 KDIGO compass points us toward familiar landmarks (Hb thresholds inherited from a bygone WHO decree, ferritin and TSAT as the only lanterns we trust, and the old names of ESAs and iron as the twin pillars that have steadied practice for decades. Yet the markers, like the warning moon, are dimmed by inflammation, the hidden hand of hepcidin, and the shadows of age, gender, and pregnancy that the guideline refuses to illuminate. 

In the dialysis ward, the guideline urges a “proactive” infusion of iron when ferritin slips beneath 500ng/mL and TSAT beneath 30%. The edifice of this counsel is the PIVOTAL trial. In the non-dialysis world, the threshold is lowered still, to ferritin <100 ng/mL with TSAT <40% or the more tentative ferritin 100-299 ng/mL with TSAT <25%. Here, the numbers are not drawn from outcomes but from the very doors through which patients were allowed to enter the FIND-CKD study, a subtle reminder that we sometimes mistake the map for the territory.

The ESAs, those old companions, are summoned when hemoglobin sinks to 9-10 g/dL in the dialyzed and are left to the discretion of the clinicians in the non-dialyzed, a range that swells to 8.5-10 ng/mL for those who suffer the weight of symptoms or the dread of cardiovascular peril.

The ceiling (11.5g/dL) stands as a stark reminder of the CHOIR and TREAT tragedies:  to chase a higher tide is to invite storms of stroke and hypertension. Yet, the recommendation to hold fast at this ceiling is itself a thin thread, graded only “strong” in name while the certainty beneath it is whisper-thin. ESA hyporesponsiveness is highlighted as a strong prognostic marker (2- to 3-fold higher cardiovascular risk). The guideline rightly urges correction of reversible causes (iron, inflammation, PTH, infection) before escalating ESA dose. However, the algorithm for managing refractory cases (switch to a HIF-PHI for 3-4 months or consider transfusion) relies on very limited head-to-head data. 

HIF-PHI agents are placed as a “second-line” oral alternative. The 2D recommendation is cautious: use the same Hb thresholds as ESAs, start slow, titrate slowly, and stop after 3-4 months if the response is inadequate. Safety concerns (thrombotic events, VEGF-mediated angiogenesis, thyroid dysfunction) are flagged, but the guideline doesn’t differentiate between agents that have shown divergent safety signals in recent phase 3 trials. The overall confidence is low, reflecting the paucity of long-term data. 

Transfusion is finally framed as a symptom-driven, transplant-protective decision rather than a numeric trigger. The guideline advises transfusion only in life-threatening bleeding, unstable coronary disease, or when ESA/HIF-PHI therapy has failed. Reference values (Hb <7 g/dL asymptomatic, <8 g/dL with cardiovascular disease) are retained as background, not mandates. The evidence for reducing allo-sensitisation by restricting transfusion comes from older observational studies; no contemporary RCT has proven that a restrictive strategy improves transplant outcomes. 

Where should anemia management head?

• Mechanistic biomarkers: replace ferritin/TSAT with hepcidin, reticulocyte-hemoglobin content, or % hypochromic RBCs, and validate these markers in CKD cohorts. Until we can reliably differentiate systemic iron deficiency from iron-restricted erythropoiesis, the current threshold will continue to be arbitrary.

• Outcome-based thresholds- conduct RCTs that test the clinical impact of ferritin/TSAT cut-offs, not merely their ability to raise Hb. Likewise, evaluate whether a lower ESA dose to achieve the same Hb target reduces cardiovascular risk.

• Clarify HIF-PHI role: head-to-head, long-term studies comparing HIF-PHI with ESAs in non-dialysis CKD, with explicit safety endpoints (thrombosis, cancer progression, thyroid dysfunction), are needed before HIF-PHIs can be recommended beyond a narrow rescue indication.

• Population-specific guidance: develop separate algorithms for women of reproductive age, pregnancy, children, and the elderly. Current one-size-fits-all hemoglobin definitions ignore well-documented physiological differences.

• Equity and implementation: address the documented 68% treatment gap in the US non-dialysis CKD (CKD-DOPPS data) by integrating decision-support tools into electronic health records, tracking ESA and iron utilization, and targeting resources to underserved populations. 

• Transfusion stewardship: build concrete decision aid calculators that incorporate symptom burden, CV risk, and allo-sensitisation probability, turning the guideline’s “shared decision-making” principle into an actionable process.

KDIGO 2026 is a methodologically transparent document: it grades most recommendations, openly acknowledges very low certainty, and emphasizes correction of reversible causes before adding pharmacologic therapy. This is honestly commendable. However, the substance of many recommendations is little more than a rebranding of old numeric thresholds, now couched in softer language. The guideline leans heavily on post-hoc interpretations of trials (PIVOTAL, FIND-CKD) and on observational data for transfusion safety, offering little in the way of new, high-quality evidence. In practice, clinicians will find a clearer philosophical framework: ”treat what you can, use minimal effective dose, involve the patient”, but will still be left to decide how to apply vague hemoglobin or iron cut-offs in the absence of solid outcome data. In short, KDIGO 2026 moves the conversation forward in tone and structure, yet the road ahead remains paved with uncertainty. Until the recommended thresholds and therapeutic hierarchies are proven by robust trials, clinicians must treat the guideline as a cautionary compass rather than a definitive map. 

Summary by

Jeyakumar Meyyappan
Transplant Nephrologist 

Reviewed by 

Cristina Popa, Sayali Thakare, Pallavi Prasad,
Swapnil Hiremath, Brian Rifkin

Header Image created by AI, based on prompts by Brian Rifkin