Anales de la RANM

68 A N A L E S R A N M R E V I S T A F U N D A D A E N 1 8 7 9 POSTOPERATIVE BLOOD SALVAGE AFTER KNEE ARTHROPLASTY Valbuena I, et al. An RANM. 2022;139(01): 67 - 77 The incidence of arthritis and arthrosis is increasing at the same rate as the age and comorbidities of our population in high income countries. Consequently, the costs, risks and complications after total knee arthroplasty (TKA) are also on the rise (1). Postoperative bleeding after TKA is one of the important outcomes to control, to avoid postope- rative complications, especially in patients with comorbidity. Bleeding is greater during the first 6 postoperative hours and can reach up to 800 mL (2). Between 20 and 30% of patients undergoing major orthopedic surgery present moderate anemia, and an increased risk of requiring blood transfusion. For patients who undergo TKA, forty to fifty percent require allogenic blood transfusion (ABT) in the perioperative period (3). Transfu- sion complications have been widely recognized by many international organizations and commis- sions and include allergic reactions, transfusion- related acute lung injury (TRALI), ABO-incompa- tible transfusion and transfusion error s , i n f e c t i ons , immunomodulation/suppression (4) and is associated with an increased risk of short and long term mortality (5). In Europe, ETPOS (European Transfusion Practice and Outcome Study) group observed that restric- tive transfusion protocols were used only in 63% of hospitals and blood salvage (BS) are used in 6% (6). BS is an important system to implement the second pillar of Patient Blood Management approach. Its advantages are its few adverse effects (2,7) pro-inflammatory effect minimizing the immuno- suppression due to trauma and hemorrhage (8) and decreasing of the rate of ABT (9,10). The use of BS has dropped markedly during the last years, due to the implantation of other techni- ques of perioperative patient blood management. The aim of this study was to assess the interest of BS for the improvement of the postoperative results of ABT and postoperative anaemia in a population of patients scheduled for a total knee arthroplasty. After having obtained the approval of our local Ethics Committee, a retrospective, observational cohort study was conduced in consecutive patients scheduled for TKA throughout twelve months in a tertiary university hospital. The medical data of all consecutive patients scheduled for TKA during that period were systematically reviewed. Inclusion criteria were to have undergone primary or replacement knee arthroplasty, excluding those in whom tranexamic acid (TXA) was administered by any routes available, and those for whom data collected was incomplete. The study was carried out on patients scheduled for TKA who were exposed to BS or not. Demographic, comorbidity, surgical and anesthetic data were recorded until hospital discharge. Hematology and coagulation pre and postopera- tive results were collected at 6, 24, 48 hours and at hospital discharge from hospital. The volume of blood drained and reinfused from BS as well as the requirement of ABT were also recorded. Primary outcome was to assess whether the use of BS postoperative reduced ABT, influenced hemoglobin and hematocrit levels and length of hospital stay (LOS). Secondary outcomes were to define risk factors for bleeding and receiving ABT after knee arthroplasty and to assess the influence of ABT on LOS. After carrying out a preliminary study, the total incidence of ABT after knee arthroplasty was estimated to 23.3%. The sample size was calculated to 260 patients (130 per cohort), for a statistical power of 0.8 and an ⍺ error of 0.05 aiming at reducing ABT incidence by 50% and achieving an ABT incidence inferior or equal to 11.65% with the use of BS. Statistical analysis was performed using SPSS version 11 /version 21 (SPSS Inc. Chicago. Illinois. USA ® ). Parametric and non-parametric tests were employed. Homogeneity of the two groups for pre-BS variables were checked. Comparisons between patients having or not received BS were carried out with One-Way and Two-Way Anova and Chi-Square tests. p< 0.05 was accepted as significant. 260 were included in the study, in 130 BS was used and in 130 BS was not used (figure 1, flow diagram). Homogeneity was checked for both groups for pre-BS variables (Table 1). The most frequent indication for surgery was osteoarthritis in 90.7% of cases. Primary TKA was performed in 91.5% (88.5% cemented, 1.5% cementless, 1.5% computer assisted TKA) and revision in 8.5%. Age was 71.69 +8.6 years. 67,9% of patients was ASA II classification and 29,3% was ASA III. Hypertension was the most common comorbidity (70,4%). 18,5% received preoperative anti-platelet therapy (APT) (Table 2). Lower limb ischemia was used in 96,1% of cases, and in 84,3% was released before closing. Ischemia time was 82.62+22.7 min. Duration of surgery was 103,1+24,2 min. Loco-regional anesthesia was predominant used in 96.9% of cases, of which 79.5% were spinal anesthesia. Femoral block was the most frequent analgesic technique. General anesthesia was performed only in 3.1% of the cases. 93,5% of patients presented postopera- tive bleeding, mean bleeding volume was 713,1 +445,4 ml. In 90% of cases, recuperated blood was transfused, and the re-infused volume from BS was 619.96 +352.7 mL. INTRODUC TION MATERIALS AND METHODS RESULTS