Renal replacement therapy for acute renal failure: a survey of practice in adult intensive care units in the United Kingdom.

This study surveyed current practice in adult intensive care units in the United Kingdom in three key areas of renal replacement therapy when used for acute renal failure: type of therapy used, typical treatment dose and anticoagulation. Responses were received from 303 (99%) of the 306 intensive care units. 269 units (89%) provide renal replacement therapy for acute renal failure. Most (65%) use continuous veno-venous haemofiltration as first-line therapy in the majority of patients, though continuous veno-venous haemodiafiltration is used by 31% of units. For haemofiltration, the median typical treatment dose (interquartile range [range]) is 32 ml.kg(-1).h(-1) (28.6-35.7 [14.3-85.7]), with 49% using a treatment dose of 35 ml.kg(-1).h(-1) or greater. For haemodiafiltration, the median typical treatment dose (interquartile range [range]) is 44 ml.kg(-1).h(-1) (28.6-57.1 [21.4-120.7]), with 67% using a treatment dose of 35 ml.kg(-1).h(-1) or greater. The vast majority of intensive care units use intravenous unfractionated heparin (96%) or epoprostenol (88%) for anticoagulation. Dosage and monitoring of these two agents vary markedly between units. No units use citrate anticoagulation. These results reveal a wide variety of practice in the delivery of renal replacement therapy between intensive care units in the United Kingdom.

Three case reports of the use of haemodiafiltration in the treatment of salicylate overdose.

Aspirin (acetylsalicylic acid) is widely available without prescription. Although self-poisoning is rare, if severe it may be life threatening. Haemodialysis has been recommended in severe cases when salicylate levels exceed 7.3 mmol l(-1). We describe three cases of severe salicylate poisoning, which were treated with continuous veno-venous haemodiafiltration (CVVHDF). All patients survived. The first case had already undergone haemodialysis before transfer to the ICU, where CVVHDF was commenced because salicylism persisted at 3 mmol l(-1). A small reduction in serum salicylate was noted. In the second case, serum salicylate decreased from 8.5 to 3.5 mmol l(-1) after 3 h of CVVHDF even though only minimal urine was produced. Our third case is a chronic overdose in whom serum salicylate decreased from 6.2 to 4 mmol l(-1) after 4 h and to 1.4 mmol l(-1) after a further 7 h. No bicarbonate was administered to this patient and elimination can only be attributed to CVVHDF and urinary clearance, which is known to be slow. We discuss the pathogenesis of severe salicylate toxicity and postulate that CVVHDF, which is widely used in the intensive care setting, may be a useful therapy in severely poisoned patients who are unstable and cannot undergo haemodialysis or in situations where haemodialysis is unavailable.