Four-Hour Hemoperfusion with Stepwise Blood Flow Rates for Toxin Clearance and Safety in Hemodialysis Patients: A Randomized Controlled Trial

Summary

Extending hemoperfusion to 4 h enhances β2-microglobulin removal and protein preservation. Increasing blood flow up to 260 mL/min optimizes clearance of small molecules and hs-CRP via a significant dose-response relationship. This 4-h, high-flow regimen was associated with hemodynamic stability in this cohort using modern adsorbents.

Abstract

Optimal hemoperfusion (HP) parameters combined with hemodialysis (HD) for removing middle-molecule and protein-bound toxins remain undefined. This prospective randomized controlled trial investigated the effects of extending HP duration and increasing blood flow rate. Eighty-eight maintenance hemodialysis patients were randomized to 2-h or 4-h HP with 4-h HD at 220 mL/min (Phase 1). After a 2-week washout, patients received 4-h HP+HD at 180, 220, or 260 mL/min (Phase 2). Analysis of covariance (ANCOVA) adjusted for baseline imbalances, and linear trend tests evaluated dose-response effects. In Phase 1, the 4-h regimen achieved significantly superior adjusted β2-microglobulin clearance (65.11% ± 4.73% vs. 45.93% ± 4.71%; Adjusted P < 0.001) and reduced loss of hemoglobin and albumin (both P < 0.01). In Phase 2, a significant linear dose-response relationship was observed for all toxins (P for trend < 0.001). A flow rate of 260 mL/min achieved optimal reduction ratios for small molecules and hs-CRP (43.12%). Restricted cubic spline analysis suggested that apparent negative hs-CRP reduction ratios may reflect hemoconcentration-related artifacts (Model P = 0.049). All regimens maintained hemodynamic stability with mean systolic pressure drops of 4.7 ± 24.3 mmHg and 4.2 ± 22.9 mmHg in Phases 1 and 2, respectively. Adverse events were low (0–3.45%) and mild. In conclusion, a 4-h HP session combined with HD, particularly at 260 mL/min, enhances the removal of middle molecules, protein-bound toxins, and inflammatory markers without apparent hemodynamic instability in this single-session study, providing evidence for optimizing operational parameters.

Introduction

Hemodialysis (HD) serves as a cornerstone of renal replacement therapy for patients with uremia¹. However, conventional HD modalities exhibit limited efficacy in removing protein-bound and middle-to-large molecular weight uremic toxins, such as intact parathyroid hormone (iPTH) and β2-microglobulin (β2-MG)^1,2. The accumulation of these toxins is not benign; substantial evidence links them to debilitating complications in maintenance hemodialysis (MHD) patients, including refractory pruritus, malnutrition, dialysis-related amyloidosis, and carpal tunnel syndrome, which collectively severely impair survival rates and quality of life^3,4,5,6,7,8. To address this clearance gap, the combination of hemoperfusion (HP) with HD has been employed and shown to be superior to HD alone in eliminating accumulated middle and large molecule toxins, such as iPTH and inflammatory markers^6,9. This synergistic approach leverages the adsorptive capacity of HP cartridges to effectively remove a range of solutes, including middle molecules (such as iPTH and β2-MG), protein-bound toxins (such as homocysteine [Hcy]), and inflammatory cytokines more which may lead to clinical improvements such as significant alleviation of uremic pruritus^5,6.

Despite its demonstrated efficacy, the optimal operational parameters for HD+HP therapy to maximize clinical benefit remain incompletely defined and are subjects of ongoing clinical inquiry. Current clinical practice and much of the existing research, often constrained by earlier adsorbent materials, typically employ an HP treatment duration of 2–2.5 h and blood flow rates ranging from 180–250 mL/min⁹. Some foundational studies even suggested that the adsorption columns might saturate after 2–3 h, implying limited benefit from prolonging the perfusion time⁴. However, with advancements in adsorbent technology, such as the development of enhanced resins (e.g., the SR130 cartridge used in this study), the rationale for these conventional parameters warrants critical re-evaluation. Emerging evidence hints at the importance of parameter optimization. For instance, studies on the timing of HP during a combined session have found that performing HP in the latter half of a dialysis session (a de facto extension of effective adsorption time relative to blood toxin concentration) results in better clearance of iPTH and β2-MG compared to early HP⁷. This challenges the old saturation dogma and suggests that longer or differently timed HP regimens with modern adsorbers may yield superior outcomes. Conversely, while higher blood flow rates could theoretically enhance the convective mass transfer of solutes to the adsorbent surface, potentially improving clearance efficiency, they may also raise practical concerns regarding circuit coagulation and hemodynamic stability^2,8.

Therefore, a significant and clinically relevant knowledge gap persists. There is a lack of robust, controlled evidence evaluating the acute clearance efficacy and safety profile of systematically extending HP duration to 4 h using modern adsorbents, and of employing systematically graded, higher blood flow rates within a single treatment session^3,6. This study was specifically designed to bridge this gap. Utilizing the SR130 hemoperfusion device, we conducted a stratified cohort study to evaluate the hypothesis that a 4-h HP regimen, compared to the conventional 2-h regimen, would provide superior clearance of key middle molecules such as β2-MG, as well as protein-bound toxins such as Hcy, while maintaining a favorable safety profile. Furthermore, we aimed to explore the dose-response effects of graded, higher blood flow rates (180, 220, and 260 mL/min) during 4-h HP+HD to determine the optimal operational range for maximizing the clearance of inflammatory markers, protein-bound solutes, and small molecule toxins without increasing adverse events, thereby offering novel evidence to refine and optimize this vital extracorporeal therapy for MHD patients.

Protocol

This was a two-phase, prospective, randomized controlled trial conducted at the Blood Purification Center of Ziyang People's Hospital between January 2025 and December 2025. The study protocol was strictly approved by the Ethics Committee of Ziyang People's Hospital (Approval No: 2025-K-2-40), and written informed consent was obtained from all participants before enrollment.

Study design and participants
A total of 88 maintenance hemodialysis (MHD) patients were recruited. For both phases, random sequence generation was performed using a computer-generated random number table. Allocation concealment was ensured through the use of sequentially numbered, opaque, sealed envelopes. A dedicated clinical research coordinator, who was not involved in direct patient care or outcome assessment, implemented the allocation and assigned participants to their respective groups. Inclusion criteria were defined as: (1) adults (≥18 years) with end-stage renal disease; (2) dialysis vintage of at least 6 months via stable vascular access (autogenous arteriovenous fistula or long-term cuffed catheter); (3) elevated baseline levels of uremic toxins, specifically intact parathyroid hormone (iPTH) > 300 pg/mL and β2-microglobulin (β2-MG) > 20 mg/L; and (4) presence of clinical complications including refractory pruritus or bone pain.

Exclusion criteria were strictly applied to ensure patient safety and data integrity. Patients were excluded if they met any of the following: severe cardiopulmonary insufficiency (New York Heart Association [NYHA] Class III or IV); acute renal failure or acute-on-chronic kidney disease; active bleeding disorders or a high risk of hemorrhage; active systemic infections; severe malnutrition (serum albumin < 30 g/L); or concurrent participation in other clinical studies.

Treatment protocol and procedures
The study comprised two distinct treatment phases separated by a 2-week washout period using standard HD. In Phase 1, the 88 participants were randomly assigned (1:1) to evaluate hemoperfusion (HP) duration (2-h vs. 4-h). The control group received 2-h HP followed by 2-h HD. The experimental group received 4-h HP concurrent with 4-h HD. Both groups maintained a total treatment time of 4 h at a fixed blood flow rate of 220 mL/min. Throughout each session, patients' vital signs, including blood pressure, heart rate, and oxygen saturation, were monitored continuously and recorded at 30-min intervals by trained nurses to ensure hemodynamic stability.

Following the 2-week washout period to minimize carry-over effects and within-subject correlation, the same cohort of 88 patients was re-randomized into three parallel groups for Phase 2 to assess blood flow rates (180, 220, and 260 mL/min) during a 4-h HP+HD session. All treatments utilized hemodialysis machines and resin cartridges. Before use, the HP cartridges were primed with 2000 mL of heparinized saline (20 mg heparin/500 mL) followed by a 30-min static soak. This priming procedure is essential to ensure the biocompatibility of the resin surface and prevent the activation of the coagulation cascade upon blood contact.

Measurements and assessments
Blood samples were collected immediately before (0 h) and after (4 h) each session. The samples were drawn from the arterial line pre-treatment and from the venous line at completion (using a 15-s slow-flow method) to avoid saline dilution. The primary efficacy endpoints were the reduction ratios (RRs) of: small molecules: Creatinine (µmol/L), Urea (mmol/L), Uric Acid (µmol/L); middle-to-large molecules: β2-MG (mg/L), iPTH (pg/mL); inflammatory markers: hs-CRP (mg/L); others: Homocysteine (µmol/L), Calcium (mmol/L), Potassium (mmol/L).

The RR was calculated as: (Pre-value - Post-value)/Pre-value × 100%. To detect potential blood loss during treatment or adsorption-induced loss of essential components, we also monitored serum hemoglobin (g/L), platelets (10⁹/L), and albumin (g/L). Safety monitoring and adverse events (AEs) were assessed continuously throughout each session and recorded at 30-min intervals by nurses, including hypotension (defined as a drop in SBP >20 mmHg), muscle cramps, and allergic reactions. At the end of each session, dialyzer and cartridge coagulation were assessed using a visual scale (Grade 0: no clotting; Grade I: <5% fibers clotted; Grade II: 5%–50% fibers clotted; Grade III: >50% fibers clotted)

Statistical analysis
Statistical analyses were performed using R 4.2.2. The normality of data was assessed using the Shapiro-Wilk test. Continuous variables were expressed as mean ± SD for normally distributed data, allowing the use of independent samples t-tests (Phase 1) and one-way ANOVA (Phase 2). For non-normally distributed data, median (IQR) and Mann-Whitney U or Kruskal-Wallis tests were employed. In Phase 1, despite randomization, an incidental imbalance in baseline ultrafiltration rate was detected; therefore, an Analysis of Covariance (ANCOVA) was performed to adjust for this covariate when comparing RRs. In Phase 2, a linear trend test was applied to evaluate the dose-response relationship of blood flow rates. Bonferroni post-hoc tests were strictly applied for all pairwise comparisons in Phase 2 when significant overall differences were detected via ANOVA or Kruskal-Wallis tests to control for Type I error inflation. Additionally, a restricted cubic spline (RCS) model with four knots was utilized to explore the potential non-linear relationship between baseline hs-CRP levels and their reduction ratio. P < 0.05 was considered significant.

Results

Phase 1: Comparison of hemoperfusion duration (2-h vs. 4-h HP)
A total of 88 MHD patients were enrolled and stratified into two cohorts for the first phase of this study. The patient selection and cohort flow diagram are presented in Figure 1. The baseline demographic and clinical characteristics of the two groups were comparable (Table 1), except for a statistically significant difference in the ultrafiltration rate (10.75 ± 3.65 vs. 8.74 ± 3.06 mL/kg/h...

Discussion

This study demonstrates that within a single treatment session, extending the hemoperfusion duration to 4 h and utilizing a blood flow rate as high as 260 mL/min in combination with hemodialysis significantly enhances the clearance of key uremic toxins in maintenance hemodialysis patients, while maintaining a favorable safety profile¹⁰. The findings here challenge the conventional parameters of HD+HP therapy and provide novel, evidence-based data for optimizing this extracorporeal detoxification s...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Hemodialysis Machine	Fresenius Medical Care	4008S	Hemodialysis delivery system used for all treatment sessions.
Resin Adsorption Cartridge	Chongqing Healthcom Blood Purification Equipment Research&Development Co.,Ltd.	SR130	Resin adsorption cartridge (130 mL) containing hyper-cross-linked styrene-divinylbenzene resin.
Hemodialyzer	Chengdu OCI MEDICAL Devices Co., Ltd.	OCI-HD150	High Flux Polyethersulfone Hollow Fiber Hemodialyzer with a membrane area of 1.5m² and ultrafiltration coefficient of 48mL/h/mmHg.
Low-molecular-weight Heparin	Pfizer	Fragmin	Anticoagulant used for systemic anticoagulation during the procedure.