To the Editor: We would like to comment on the analysis of the results and the conclusions of the study recently published by Trujillo et al. [1], The utility of the perioperative autologous transfusion system OrthoPAT in total hip replacement surgery: a prospective study. The authors’ aim was to investigate the eVectiveness of an intraoperative blood recovery system for reducing allogeneic blood transfusion in patients undergoing primary arthroplasty of the hip. Nonetheless, although it was not their objective, the authors attempt to estimate the cost-beneWts of the system and conclude that there is a reduction of 85 D per patient. In our opinion, this conclusion is not supported by the data presented, for several reasons. First, the amount of blood obtained for reinfusion was insuYcient in 18% of patients in this study. The authors did not specify whether these patients were included in the “cost-beneWt” calculation. If they were included, the saving would be less than 85 D per patient. Unfortunately, in our experience, it is diYcult to predict preoperatively whether reinfusion will be possible, because it is impossible to know the severity of perioperative blood loss and because there may be some unknown factor that will complicate proper blood recovery and processing. The presence of low hemoglobin levels might be an indication that the erythrocyte mass recovered will be less than for a similar volume of blood loss and consequently, the volume recovered will be lower [2 , 3]. Second, as is shown in Fig. 2, the majority of patients had hemoglobin values greater than 14 g/dL, and many had values between 15 and 16 g/dL. We wonder whether these latter patients were really at risk of requiring a transfusion. The “cost-beneWt” would be highly debatable in these patients because of their low transfusion risk [4 , 5]. Patients were consecutively enrolled in the study, except for those that presented a disease that could have an eVect on surgical bleeding or the possibility to receive a transfusion, with no further considerations related to hematimetry. The authors did not consider a hemoglobin (Hb) level greater than 14 g/ dL to be an exclusion criteria for applying the blood-sparing technique; thus even patients with higher Hb values would be candidates. From the viewpoint of transfusion practice, the question raised by this observation is whether application of a transfusion technique would be justiWed in these patients, since their risk of transfusion is low. The patients’ total perioperative blood loss [1, Table 2] showed no statistical diVerences between the groups, and we were surprised that one of the groups, speciWcally, the control patients, received more transfused blood “overall”. It is logical that the study group would receive less allogeneic blood, but overall transfusion should be similar in the two groups; if it is not, we can assume that the control group was overtransfused with respect to the WSB group. The authors mention all the advantages of the OrthoPAT system, but none of the potential disadvantages. The blood-processing capacity of the system (800–1,000 mL/h) may not be suYcient when saline solution is used for washing tissues or when bleeding is intense. In these cases, the system can become saturated and there is no intermediate solution. An alternative aspiration system would have to be available and the circuit would be interrupted. In our opinion, the 1,000-mL capacity is insuYcient for orthopedic surgery, particularly hip arthroplasty and spine surgery. A. Biarnes · M. J. Colomina · L. Mora Anaesthesia Department, Hospital Universitario Vall d’Hebron, Area de Traumatologia, Barcelona, Spain