Economic evaluation of a randomized clinical trial of haemodilution with cell salvage in aortic surgery

Background: This study evaluated the costs of acute normovolaemic haemodilution (ANH) and intraoperative cell salvage (ICS) versus homologous blood transfusion in aortic surgery in a prospective multicentre randomized trial. Methods: One hundred and forty‐five patients were randomized either to standard transfusion practice (homologous) or to a combination of ANH and ICS (autologous). Costs for each inpatient admission were identified. Cell salvage costs were assigned on the assumption that 50 operations were done each year employing a trained cell salvage operator. The results were analysed statistically using bias‐corrected bootstrap analysis. Results: Patients who had transfusion of homologous blood received some 251 units and those having a homologous transfusion received 103 units (P = 0·008). There was no difference in morbidity, mortality and duration of hospital stay. Transfusion‐related mean costs were similar at £340 for patients having a homologous transfusion and £357 for those receiving autologous blood (mean difference £17 (95 per cent confidence interval (c.i.) −£184 to £174); P not significant). There was also no significant difference in mean overall costs: £5859 for homologous and £5384 for autologous transfusion (mean difference −£475 (95 per cent c.i. −£2231 to £1342)). Sensitivity analysis showed that costs remained similar for 20 and 150 operations per annum. Exclusion of a dedicated cell salvage operator reduced autologous transfusion costs but did not have a significant impact on overall cost. Conclusion: Autologous transfusion is cost neutral in aortic surgery even when surgical activity is low. © 2002 British Journal of Surgery Society Ltd