Sir—I have read with interest the article by Dr. Apelqvist et al. In their well-performed study they did not find pneumatic wound compression to reduce the need for blood transfusion after hip fracture surgery. In the discussion the authors mention several explanations for not receiving positive results found in two previous studies concerning the same wound compression device used after elective total hip arthroplasty (Hörnberg et al. 2002, Johansson 2005a). Of course, a mix of hip fracturepatients, operated on by different methods, differs from patients with hip osteoarthritis operated on with a total hip arthroplasty. However, the authors do not comment on the fact that all patients in their study received tranexamic acid. Tranexamic acid is probably the most powerful, simplest and most cost-effective treatment in minimizing blood loss after a total hip arthroplasty (Johansson et al. 2005b). This strong effect may even overshadow the possible positive effects of wound compression for hip fracturepatients.Sir—We thank Dr Johansson for his comment. The use of tranexamic acid was already in our clinical routine in hip fracture surgery and we made no exception for this study. Our trial was performed to evaluate the efficacy of the compression bandage in further reducing the need for postoperative blood transfusion. Of course we can not know whether the results would have been any different had tranexamic acid not been given prior to surgery. However, this seems unlikely since the small differences in mean blood units and in proportion of patients transfused, though statistically insignificant, tended to be in favor of the non compression group for all 3 surgical procedures. The mean number of blood units was in the compression group 1.3 and in the non-compression group 1.1. Blood transfusion was given to 84 (62%) in the compression group and to 85 (55%) in the non-compression group.