Anticoagulation for people receiving long-term haemodialysis

Background: Haemodialysis (HD) requires safe and effective anticoagulation to prevent clot formation within the extracorporeal circuit during dialysis treatments to enable adequate dialysis and minimise adverse events, including major bleeding. Low molecular weight heparin (LMWH) may provide a more predictable dose, reliable anticoagulant effects and be simpler to administer than unfractionated heparin (UFH) for HD anticoagulation, but may accumulate in the kidneys and lead to bleeding.

Objectives: To assess the efficacy and safety of anticoagulation strategies (including both heparin and non-heparin drugs) for long-term HD in people with kidney failure. Any intervention preventing clotting within the extracorporeal circuit without establishing anticoagulation within the patient, such as regional citrate, citrate enriched dialysate, heparin-coated dialysers, pre-dilution haemodiafiltration (HDF), and saline flushes were also included.

Search methods: We searched the Cochrane Kidney and Transplant Register of Studies up to November 2023 through contact with the Information Specialist using search terms relevant to this review. Studies in the Register are identified through searches of CENTRAL, MEDLINE, and EMBASE, conference proceedings, the International Clinical Trials Registry Platform (ICTRP) Search Portal and ClinicalTrials.gov.

Selection criteria: Randomised controlled trials (RCTs) and quasi-randomised controlled studies (quasi-RCTs) evaluating anticoagulant agents administered during HD treatment in adults and children with kidney failure.

Data collection and analysis: Two authors independently assessed the risk of bias using the Cochrane tool and extracted data. Treatment effects were estimated using random effects meta-analysis and expressed as relative risk (RR) or mean difference (MD) with 95% confidence intervals (CI). Evidence certainty was assessed using the Grading of Recommendation, Assessment, Development and Evaluation approach (GRADE).

Main results: We included 113 studies randomising 4535 participants. The risk of bias in each study was adjudicated as high or unclear for most risk domains. Compared to UFH, LMWH had uncertain effects on extracorporeal circuit thrombosis (3 studies, 91 participants: RR 1.58, 95% CI 0.46 to 5.42; I² = 8%; low certainty evidence), while major bleeding and minor bleeding were not adequately reported. Regional citrate anticoagulation may lower the risk of minor bleeding compared to UFH (2 studies, 82 participants: RR 0.34, 95% CI 0.14 to 0.85; I² = 0%; low certainty evidence). No studies reported data comparing regional citrate to UFH on risks of extracorporeal circuit thrombosis and major bleeding. The effects of very LMWH, danaparoid, prostacyclin, direct thrombin inhibitors, factor XI inhibitors or heparin-grafted membranes were uncertain due to insufficient data. The effects of different LMWH, different doses of LMWH, and the administration of LMWH anticoagulants using inlet versus outlet bloodline or bolus versus infusion were uncertain. Evidence to compare citrate to another citrate or control was scant. The effects of UFH compared to no anticoagulant therapy or different doses of UFH were uncertain. Death, dialysis vascular access outcomes, blood transfusions, measures of anticoagulation effect, and costs of interventions were rarely reported. No studies evaluated the effects of treatment on non-fatal myocardial infarction, non-fatal stroke and hospital admissions. Adverse events were inconsistently and rarely reported.

Authors' conclusions: Anticoagulant strategies, including UFH and LMWH, have uncertain comparative risks on extracorporeal circuit thrombosis, while major bleeding and minor bleeding were not adequately reported. Regional citrate may decrease minor bleeding, but the effects on major bleeding and extracorporeal circuit thrombosis were not reported. Evidence supporting clinical decision-making for different forms of anticoagulant strategies for HD is of low and very low certainty, as available studies have not been designed to measure treatment effects on important clinical outcomes.