Inflammation on Dialysis: What It Means and What HelpsPamamaga Habang Nagda-dialysis: Ano ang Ibig Sabihin at Ano ang NakatutulongPanghubag Samtang Nag-dialysis: Unsa ang Buot Ipasabot ug Unsa ang MakatabangPamamaga kabang Magdialysis: Nanu ing Kayulugan at Nanu ing Makatulung

What “inflammation” means in kidney failure, and where it comes from

Acute inflammation is a protective, self-limiting response: pattern-recognition receptors on macrophages and neutrophils detect microbial products or damaged tissue, cytokines recruit cells, the threat is cleared, and resolution mediators switch the program off. Kidney failure distorts every step. Retained solutes, oxidative stress, and endotoxin translocation keep innate immune cells primed; renal clearance of cytokines falls; and adaptive immunity weakens at the same time (Speer et al., 2022). The result is a paradox clinicians see daily — a patient who is both persistently inflamed and immunocompromised, with a high CRP and a poor vaccine response.

The mechanism most relevant to drug development runs through the NOD-like receptor pyrin domain-containing 3 (NLRP3) inflammasome. Danger signals such as cholesterol crystals, urate, or uremic toxins activate NLRP3, which cleaves pro-IL-1β into active IL-1β. IL-1β induces IL-6, and IL-6 signaling in hepatocytes drives CRP, fibrinogen, serum amyloid A, and hepcidin production (Speer et al., 2022; Tuttle et al., 2026). CRP therefore sits at the end of the chain. It is a convenient read-out, not the whole inflammatory state, and a circulating concentration says nothing direct about inflammation inside a coronary plaque or the myocardium.

Sources: what can be found and fixed

Before any biology of uremia is invoked, a persistently elevated CRP in a dialysis patient is an instruction to look for a source. The contributors below differ in how often they occur and how firmly they are linked to inflammation, but each has a concrete assessment and an owner on the dialysis team. Modern water treatment and ultrapure dialysate have made circuit-derived inflammation less prominent than it was historically; an individual patient's high CRP is never, by itself, evidence that a unit's water is contaminated.

Measuring inflammation without over-reading it

High-sensitivity CRP (hsCRP) assays measure the low concentrations that matter in cardiovascular risk research; standard CRP assays are built for acute infection. Units differ between laboratories, and the conversion is simple but often missed: CRP in mg/dL × 10 = mg/L. A single value reflects the previous few days, so a dental abscess, a cannulation-site infection, or a viral illness can produce a misleading spike. When a measurement will change management, repeat it after intercurrent illness has settled and interpret it alongside the clinical story.

The threshold of hsCRP ≥2 mg/L that recurs in this literature is a trial enrichment criterion used by CANTOS, RESCUE, ZEUS, and the clazakizumab program. It identifies people in whom an anti-inflammatory drug has room to work. It is not a clinical definition of disease and does not, by itself, create an indication for any therapy. No guideline currently recommends routine hsCRP screening of every dialysis patient or a treat-to-CRP strategy.

Ferritin deserves particular care. Because ferritin rises as an acute-phase protein, a ferritin of several hundred ng/mL with a low TSAT can coexist with functional iron restriction. Albumin is similar: Kidney Disease Outcomes Quality Initiative (KDOQI) nutrition guidance treats it as one input among several in nutritional assessment, not a stand-alone diagnosis (Ikizler et al., 2020).

Outcomes, anemia, and symptoms: an endpoint taxonomy

“Cardiovascular benefit” is not one thing in dialysis. A large share of cardiovascular death in this population is sudden death, heart failure, and arrhythmia driven by left ventricular hypertrophy, fibrosis, volume and electrolyte shifts, and vascular calcification — processes only partly mediated by atherosclerotic plaque. An anti-inflammatory drug designed around atherothrombosis may therefore have less to work on in dialysis than in a post-MI population, even if the biology is shared. This is one reason trial results from CANTOS cannot be transplanted to dialysis, and why the POSIBIL6ESKD phase 3 endpoint (cardiovascular death or MI) differs from the 3-point MACE used in ZEUS.

The anemia link is the clearest mechanistic connection in the field. IL-6 induces hepatic hepcidin; hepcidin degrades ferroportin, trapping iron in enterocytes and macrophages; erythropoiesis becomes iron-restricted even when ferritin is high; and ESA requirements rise (Pergola et al., 2021). Anemia in kidney failure is still multifactorial, so the KDIGO 2026 guideline asks clinicians to address correctable causes, including iron deficiency, infection, and inflammation, before starting an ESA or HIF-PHI (Hedayati et al., 2026). Bleeding, inadequate dialysis, and high parathyroid hormone (PTH) levels compete with inflammation as explanations for a rising ESA requirement.

Read the ladder from the bottom. Observational cohorts tie IL-6 to death (Rao et al., 2005). Pilot and phase 2 trials show target engagement. The rilonacept study in stage 3–4 CKD improved brachial flow-mediated dilation (FMD) without changing aortic pulse-wave velocity (Nowak et al., 2017), a physiologic signal. RESCUE raised hemoglobin by roughly 0.6–1.0 g/dL in an exploratory analysis (Pergola et al., 2024), an anemia signal. ACTION found no patient-reported outcome (PRO) benefit (Dember et al., 2022). At the top rung, CANTOS succeeded outside dialysis, ZEUS failed in CKD, and PISCES succeeded in HD with an agent whose mechanism of benefit is unknown.

Confounding and reverse causation

A sicker patient generates more inflammation; inflammation also makes patients sicker. Cohort adjustment cannot fully separate the two, and it cannot remove the confounding from infection that is both a source of CRP and an independent cause of death. Randomized trials that lower a marker are the only way to test whether the marker is on the causal path, which is why the outcome trials in the next sections matter more than any hazard ratio from a cohort.

IL-1 blockade: proof of concept, and its limits

CANTOS — the trial that made the hypothesis credible (indirect evidence)

The Canakinumab Anti-inflammatory Thrombosis Outcomes Study (CANTOS) randomized 10,061 patients with prior MI and hsCRP ≥2 mg/L to placebo or one of three doses of canakinumab, a monoclonal antibody against IL-1β, given every 3 months. Over a median 3.7 years, the 150-mg dose reduced the primary composite of nonfatal MI, nonfatal stroke, or cardiovascular death (hazard ratio [HR] 0.85, 95% confidence interval [CI] 0.74–0.98), with event rates of 3.86 versus 4.50 per 100 person-years. Lipids did not change. Fatal infections were more frequent with canakinumab, and all-cause mortality did not differ (HR 0.94, 95% CI 0.83–1.06) (Ridker et al., 2017).

A secondary analysis examined the 1,875 participants (18.6%) with an estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m². MACE fell with randomized canakinumab (HR 0.82, 95% CI 0.68–1.00), and the effect was larger in those whose on-treatment hsCRP fell below 2 mg/L after the first dose (HR 0.68) (Ridker et al., 2018). Two cautions apply. These patients had moderate CKD, not kidney failure on dialysis. And the “responder” comparison is a post-randomization subgroup: people who achieve a low CRP on treatment differ in many ways from those who do not, so the analysis cannot validate a strategy of titrating therapy to a CRP target.

Direct dialysis and CKD studies

The first randomized IL-1 study in maintenance HD enrolled patients with three consecutive hsCRP values above 5 mg/L and gave recombinant human IL-1 receptor antagonist (IL-1ra) or placebo for 4 weeks. Of 22 randomized, 14 completed. Mean hsCRP fell 53% versus 1% and IL-6 fell 40% versus a 20% rise; the prealbumin difference was not significant (Hung et al., 2011). The agent was recombinant IL-1ra, the molecule marketed as anakinra; a photographed lecture slide that labeled this pilot as rilonacept appears to be mistaken.

Rilonacept, an IL-1 trap, was tested in 42 adults with stage 3–4 CKD who were not on dialysis. After 12 weeks it improved brachial FMD and lowered hsCRP but did not change aortic pulse-wave velocity (Nowak et al., 2017). That is a physiologic signal in a nondialysis population.

ACTION was the definitive pilot in dialysis. Eighty HD patients with hsCRP ≥2 mg/L received anakinra 100 mg three times weekly through the dialysis circuit, or placebo, for 24 weeks. Median hsCRP fell 41% with anakinra and 6% with placebo, a between-group difference that was not statistically significant; IL-6 fell significantly (25% vs. 0%). No effect on patient-reported outcomes was evident. Serious adverse events and deaths were similar, and the rate of prespecified adverse events of interest, including infections and cytopenias, was lower with anakinra (0.48 vs. 1.40 events per patient-year). Retention was 80% and 72% of doses were given (Dember et al., 2022). A 24-week pilot cannot establish long-term infection safety, but ACTION does show that an IL-1 trial in dialysis is feasible.

IL-6 blockade: target engagement versus benefit

IL-6 sits one step closer to the liver than IL-1β, which makes it the more direct handle on CRP and hepcidin. Two agents dominate the kidney literature, both monoclonal antibodies against the IL-6 ligand rather than the receptor: ziltivekimab and clazakizumab. Receptor blockers such as tocilizumab belong to rheumatology and carry their own safety data; infection or cytopenia rates from one agent should not be transferred to another.

Ziltivekimab in selected HD patients with anemia of inflammation

A phase 1/2 trial randomized 61 HD patients with IL-6 ≥4 pg/mL who carried the rs855791 variant of TMPRSS6, a polymorphism hypothesized to heighten IL-6–mediated effects on iron. Intravenous ziltivekimab every 2 weeks for 12 weeks lowered hsCRP, serum amyloid A, and fibrinogen; median ESA use fell by 15,000 to 33,000 international units (IU) per week across dose cohorts with no change on placebo; and TSAT, serum iron, and albumin rose. Four patients in the two higher-dose cohorts died of treatment-emergent adverse events (Pergola et al., 2021). In a trial this small and this selected, those deaths can neither be dismissed nor attributed to the drug. The anemia findings are biologically coherent and remain surrogate outcomes.

RESCUE (indirect: nondialysis CKD)

RESCUE randomized 264 adults with moderate-to-severe CKD and hsCRP ≥2 mg/L to subcutaneous ziltivekimab 7.5, 15, or 30 mg or placebo every 4 weeks. At 12 weeks median hsCRP fell 77%, 88%, and 92% versus 4% with placebo, with reductions in fibrinogen, serum amyloid A, and lipoprotein(a); the drug was well tolerated over 24 weeks (Ridker et al., 2021). An exploratory analysis of the same participants reported hemoglobin increases of 0.57 to 1.05 g/dL versus placebo at 12 weeks, with higher serum iron and TSAT; the trial ended early because of the coronavirus disease 2019 (COVID-19) pandemic (Pergola et al., 2024). These are two reports from one randomized population, not two trials, and both concern biomarkers.

Clazakizumab phase 2b in maintenance dialysis (direct)

The dose-finding stage of POSIBIL6ESKD randomized 127 adults on maintenance dialysis with cardiovascular disease and/or diabetes and hsCRP ≥2 mg/L to clazakizumab 2.5, 5, or 10 mg or placebo every 4 weeks. At week 12, hsCRP fell by 86%, 90%, and 92% relative to placebo; 79–82% of treated patients reached hsCRP below 2 mg/L versus none on placebo; and fibrinogen, serum amyloid A, and lipoprotein(a) fell while albumin rose. No sustained grade 3–4 neutropenia or thrombocytopenia occurred. Serious infections were similar to placebo at 2.5 and 5 mg but numerically more frequent at 10 mg (Chertow et al., 2024). This is the strongest direct evidence of IL-6 target engagement in dialysis, and it is still a biomarker trial.

How to interpret ZEUS

ZEUS tested the biomarker hypothesis about as fairly as it can be tested: a validated target, a population enriched for inflammation, an adequately powered event-driven design, and confirmed engagement. The result does not prove that inflammation is irrelevant to cardiovascular disease in CKD. It does show that lowering circulating IL-6 activity, in that population and over that time frame, did not translate into fewer atherothrombotic events, and that the infection cost was real. Several explanations remain open: the dominant cardiovascular mechanisms in advanced CKD may be nonatherosclerotic; tissue inflammation may not track circulating IL-6; competing risks may dilute benefit; and infection harm may offset any vascular gain. ZEUS was not a maintenance-dialysis trial, so it neither confirms nor refutes what POSIBIL6ESKD will find, but it removes any basis for assuming that a 90% hsCRP reduction predicts benefit.

POSIBIL6ESKD continues. The registry lists combined estimated enrollment of 3,110, estimated completion on May 22, 2029, and a phase 3 primary endpoint of time to cardiovascular death or MI (U.S. National Library of Medicine, 2026b). A trial operations page has listed different sample-size and timeline figures; until the sponsor reconciles them, quote the registry with its date. Registration is not evidence of efficacy.

Other strategies and the 2025–2026 outcome update

PISCES: fish oil in maintenance hemodialysis (direct)

PISCES randomized 1,228 adults on maintenance HD at 26 sites in Canada and Australia to daily fish oil — 4 g providing 1.6 g EPA and 0.8 g DHA — or corn-oil placebo, double-blind. The primary endpoint counted all serious cardiovascular events: sudden and nonsudden cardiac death, fatal and nonfatal MI, peripheral vascular disease leading to amputation, and fatal and nonfatal stroke. Over 3.5 years the event rate was 0.31 versus 0.61 per 1,000 patient-days (HR 0.57, 95% CI 0.47–0.70). Component HRs pointed the same way (cardiac death 0.55, MI 0.56, amputation 0.57, stroke 0.37), the extended endpoint including noncardiac death had an HR of 0.77, and adverse events and adherence did not differ meaningfully (Lok et al., 2026).

Three points of interpretation matter at the bedside. First, the primary result is expressed as an event rate across follow-up, so it should not be converted into a number needed to treat from the HR; the absolute risk of a first event, and the exact estimand, need the full text. Second, the benefit belongs to that product and dose. Typical over-the-counter fish-oil capsules contain far less EPA plus DHA per capsule, purity varies, and pure-EPA products are a different intervention. Third, fish oil has anti-inflammatory, antiarrhythmic, antiplatelet, and triglyceride-lowering effects, and PISCES did not show which of these, if any, produced the result. Calling it an anti-inflammatory success would be an overreach. Whether guideline bodies adopt it, and how it fits with bleeding risk on anticoagulants or antiplatelet therapy, is still to be settled; for now it supports an individualized conversation, with the trial regimen named precisely.

Colchicine and methotrexate: contrasts from cardiology (indirect)

Three nondialysis trials show how unpredictable anti-inflammatory strategies are. Low-dose methotrexate in stable atherosclerosis did not lower IL-1β, IL-6, or CRP and did not reduce events (HR 0.96, 95% CI 0.79–1.16) (Ridker et al., 2019). Colchicine 0.5 mg daily in chronic coronary disease (LoDoCo2) reduced the primary composite (HR 0.69, 95% CI 0.57–0.83), with a borderline excess of noncardiovascular death (HR 1.51, 95% CI 0.99–2.31) (Nidorf et al., 2020). Colchicine started soon after MI in CLEAR SYNERGY lowered CRP but did not reduce events (HR 0.99, 95% CI 0.85–1.16) (Jolly et al., 2025).

None of these trials studied dialysis. Colchicine is renally cleared, narrow in therapeutic index, and interacts with strong cytochrome P450 3A4 (CYP3A4) and P-glycoprotein (P-gp) inhibitors; the United States label for the cardiovascular product LODOCO contraindicates it when creatinine clearance is below 15 mL/min (AGEPHA Pharma USA, 2023). That labeling applies to the cardiovascular product and indication; colchicine use for gout flares in kidney failure is a separate decision with its own dosing and toxicity considerations. This guide does not support a dialysis cardiovascular colchicine regimen.

Dialysis modality: high-volume hemodiafiltration

CONVINCE randomized 1,360 patients already on high-flux HD who were candidates for a convection volume of at least 23 L per session and able to complete patient-reported outcome assessments. High-dose hemodiafiltration (HDF; mean 25.3 L per session) reduced all-cause death compared with high-flux HD over a median of 30 months (17.3% vs. 21.9%; HR 0.77, 95% CI 0.65–0.93) (Blankestijn et al., 2023). Better middle-molecule clearance and lower inflammation are plausible mediators, but the trial did not establish the mechanism, and the result applies to patients with access and hemodynamics able to deliver that volume, in units with the water quality and equipment to support it.

Nutrition, exercise, and other interventions

Nutrition support, exercise programs, probiotics, antioxidants, and expanded-hemodialysis membranes have all been studied with inflammatory markers as endpoints, mostly in small trials. Nutrition and physical activity are worth doing for function, appetite, and quality of life on their own terms (Ikizler et al., 2020); claims that they work by lowering CRP, or that they reduce mortality through that route, are not supported by adequately powered outcome trials. Small studies of surrogate markers should not receive the same weight as PISCES or CONVINCE.

Master trial matrix and benefit–harm ledger

Each row reports the prespecified primary result first. The directness column labels whether the population was maintenance dialysis (direct), nondialysis CKD, or a nonrenal population. The source column shows how the figures here were checked: an abstract verified on PubMed, a peer-reviewed full report, a sponsor headline, or a registry record.

Study regimens are listed to identify the intervention tested; they are not prescribing guidance. Confidence intervals are 95%. CV = cardiovascular; CVD = cardiovascular disease.

Benefit–harm ledger for anti-cytokine therapy in dialysis

A practical pathway for the inflamed dialysis patient

The entry point is a clinical concern: persistent unexplained elevation of CRP, ESA hyporesponsiveness, deteriorating nutrition, or burdensome symptoms. A screening CRP is not required for every patient, and nothing in the pathway below leads to an anti-cytokine prescription.

Established care with outcome evidence in dialysis

Two dialysis trials justify specific actions that sit alongside inflammation management, without implying that cytokine suppression explains their benefit. PIVOTAL randomized 2,141 HD patients to proactive high-dose IV iron sucrose (400 mg monthly unless ferritin exceeded 700 µg/L or TSAT reached 40%) or a reactive low-dose strategy. The proactive arm was noninferior and superior on the composite of nonfatal MI, nonfatal stroke, heart-failure hospitalization, or death (HR 0.85, 95% CI 0.73–1.00), used less ESA, and had the same infection rate (Macdougall et al., 2019). It supports the KDIGO 2026 preference for proactive IV iron in HD; it is not a license to give iron during active infection. CONVINCE supports high-volume HDF where feasible (Blankestijn et al., 2023).

Symptoms and patient-important outcomes

Fatigue, post-dialysis recovery time, pruritus, poor sleep, low appetite, pain, low mood, and declining physical function are what dialysis patients report most often and want addressed. The 2022 KDIGO Controversies Conference on symptom-based complications concluded that symptom assessment should be routine and individualized and that dialysis care has focused too narrowly on laboratory targets (Mehrotra et al., 2023). That report is consensus, not a graded guideline.

Inflammation may contribute to several of these symptoms through cytokine effects on the brain, muscle, and appetite, but ACTION is the instructive counterexample: a drug that lowered IL-6 did not make patients feel better (Dember et al., 2022). Symptom relief should never be made contingent on lowering CRP. Difelikefalin, a peripherally restricted kappa-opioid agonist, illustrates the alternative: in KALM-1 (378 HD patients), 49.1% versus 27.9% achieved at least a 3-point improvement on the Worst Itch Numerical Rating Scale (WI-NRS) at 12 weeks (Fishbane et al., 2020). That is symptom-specific benefit, not evidence for systemic anti-cytokine therapy; check local registration and availability before discussing it.

Check instrument wording, licensing, and validated cut-offs before adopting any tool unit-wide. Acute or severe symptoms, such as chest pain, syncope, new neurologic deficits, or suicidal thoughts, have an escalation route that bypasses this table entirely.

Peritoneal dialysis: a separate pathway

PD patients share the uremic drivers of inflammation but differ in sources and measurement. Peritonitis and exit-site or tunnel infection are the dominant infectious sources; the peritoneal membrane itself becomes inflamed with long exposure to glucose-based solutions; daily peritoneal protein losses lower serum albumin independently of inflammation; and residual kidney function is often better preserved. IL-6 predicted mortality similarly in incident PD and HD patients in a Swedish cohort (Pecoits-Filho et al., 2002), but no anti-cytokine trial has been designed around PD.

Use ISPD guidance rather than adapting HD catheter algorithms. The 2022 ISPD peritonitis recommendations set a target overall peritonitis rate of no more than 0.40 episodes per patient-year at risk and more than 80% of patients free of peritonitis per year, and address modifiable risks such as hypokalemia, pets, and training (Li et al., 2022). Cloudy effluent or abdominal pain is a same-day assessment.

Five teaching cases

These cases are composites written for teaching. They identify evidence, uncertainty, and the decision; they are not patient-specific prescriptions.

Translating the evidence into a dialysis service

The work described here belongs to the whole team. Nurses and access coordinators own exit-site surveillance and catheter-reduction plans; the nephrologist owns the source search, dialysis prescription, and anemia decisions; dietitians own nutrition assessment; and someone has to own symptom follow-up so that a patient's answer to “what bothers you most?” leads somewhere. Infection prevention follows the Centers for Disease Control and Prevention core interventions: surveillance, hand hygiene, catheter and exit-site care, staff competency, patient engagement, and catheter reduction (CDC, 2024).

Useful quality measures are processes and patient outcomes: bloodstream and access infection rates, catheter prevalence, completion of a source search when one is indicated, ESA hyporesponsiveness reviews, and documented symptom follow-up. Achievement of a target CRP is not a validated unit-performance measure, and a before-and-after change in a unit's average CRP does not show that care improved.

Philippine context

In resource-constrained units, the highest-yield actions are inexpensive: a structured exit-site check every session, documented delivered treatment time, dental and foot examination, a TB risk assessment where relevant, and a short symptom question. hsCRP and IL-6 assays may be unavailable or costly, and their results rarely change management in an individual patient. Investigational IL-6 antibodies are not approved products. Before discussing fish oil or difelikefalin, confirm local registration, the exact formulation available, and cost to the patient, which Philippine Health Insurance Corporation (PhilHealth) hemodialysis coverage may not include. Local infection-control protocols and the unit's antibiogram govern empiric therapy.

Open questions and the trials under way

The field needs trials that select patients by phenotype rather than a single CRP threshold, measure tissue as well as circulating inflammation, separate atherosclerotic from nonatherosclerotic cardiovascular death, account for infection as a competing harm, and collect patient-reported outcomes with complete follow-up. POSIBIL6ESKD can answer whether IL-6 ligand blockade reduces cardiovascular death or MI in a selected dialysis population; it cannot tell clinicians whether to treat inflammation in general, and it will not answer symptom questions it was not designed for.

Registry records were retrieved on November 2, 2026. Registration and status are not evidence of efficacy, and secondary news reports describing these programs have not always matched the registry; check the record before quoting a status.

Update log and unresolved items

• ZEUS: sponsor headline only. The full results are still to be published; this guide will be updated once the peer-reviewed report is available.
• ZEUS denominator: registry actual enrollment 6,385 versus 6,376 in a sponsor presentation. Retained as stated until the full report reconciles them.
• POSIBIL6ESKD: registry estimated enrollment and completion date differ from a trial operations page. Registry figures are quoted.
• PISCES: exact estimand, absolute first-event risks, and formulation details require full-text extraction.
• KDIGO 2026 anemia: statement numbers and grades should be checked against the full guideline before quoting them in local protocols.

Clinical pearls