Low-molecular-weight heparin (LMWH) has been suggested as providing safe, efficient, convenient and possibly more cost-effective anticoagulation for haemodialysis (HD) than unfractionated heparin, with fewer side-effects and possible benefits on uraemic dyslipidaemia. In this prospective, randomized, cross-over study we compared the safety, clinical efficacy and cost effectiveness of Clexane (enoxaparin sodium; Rhône-Poulenc Rorer) with unfractionated heparin in 36 chronic HD patients. They were randomly assigned to either Clexane (1 mg/kg body weight, equivalent to 100 IU) or standard heparin, and followed prospectively for 12 weeks (36 dialyses) before crossing over to the alternate therapy for a further 12 weeks. Heparin anticoagulation was monitored using activated coagulation times. Dialysis with Clexane resulted in less frequent minor fibrin/clot formation in the dialyser and lines than with heparin (P<0.001), but was accompanied by increased frequency of minor haemorrhage between dialyses (P<0.001). Clexane dose reduction (to a mean of 0.69 mg/kg) eliminated excess minor haemorrhage without increasing clotting frequencies. Mean vascular compression times were similar in both groups. Over 24 weeks, no changes in standard serum lipid profiles were observed. This study suggests that a single-dose protocol of Clexane is an effective and very convenient alternative to sodium heparin, but currently direct costs are about 16% more. We recommend an initial dose of 0.70 mg/kg.