Comparison of different heparin-free dialysis methods

To the Editor, Managing blood coagulation in the extracorporeal circuit during haemodialysis (HD) remains a challenge, particularly for patients with contraindication to systemic anticoagulation due to an increased bleeding risk [1]. In these situations, current guidelines recommend heparin-free dialysis without defining an optimal approach [2]. Saline flushes are no longer used due to limited or detrimental effect [3, 4], while regional citrate anticoagulation requires rigorous monitoring that is unsuitable for most maintenance HD centres [5]. Citrate-containing dialysate may reduce the risk of coagulation, particularly when combined with another alternative method such as predilution haemodiafiltration (HDF) or a heparin-grafted membrane [6]. We aimed to compare the effectiveness of predilution HDF, conventional HD with a polysulfone membrane, and conventional HD with a heparin-grafted membrane, all associated with a citrate-containing dialysate, in patients requiring heparin-free dialysis. We conducted a prospective, single-centre, observational study at Lyon-Sud University Hospital between December 2023 and December 2024. Patients undergoing maintenance HD with a temporary contraindication to systemic anticoagulation were included if they were aged |$\ge $|18 years and had vascular access allowing a blood flow of ≥300 ml/min. Exclusion criteria included systemic anticoagulation outside the HD session, single-needle HD, pregnancy, or severe liver disease. All sessions used a citrate-containing dialysate (with a concentration of 0.8 mmol/l) and were performed on Fresenius 5008/6008 dialysis machines. Conventional HD and predilution HDF used the same Fx60 (Fresenius®) polysulfone membrane (1.4 m²), while HD with a heparin-grafted membrane was performed with the Evodial (Baxter®) dialyser with a similar surface (1.3 m²). In predilution HDF sessions, the substitution flow rate was set at 70 mL/min according to the local routine protocol. Data collection included demographic and clinical characteristics, routine laboratory parameters, and dialysis session parameters. Bleeding risk classification was estimated using the Lohr and Schwab scale [7]. Dialysis prescriptions were allocated to the three groups in a non-randomized, rotating fashion, distributing them sequentially from one group to the next in a repeating cycle, unless specific contraindication existed (such as known intolerance to a membrane or heparin allergy). The primary outcome was a full 240-minute session completion without coagulation. Secondary outcomes included quality of restitution using a semi quantitative scale, dialysis efficacy (online clearance, online Kt/V), and tolerance of the dialysis sessions. Differences between groups were assessed by the Kruskal–Wallis test followed, when appropriate, by Dunn tests. Simple comparisons were performed using the Mann–Whitney U-test. Chi-squared tests were used to compare categorical variables. Blood circuit survival was analysed using Kaplan–Meier plots and the log-rank test to compare differences between techniques. P < .05 was considered as statistically significant.