Aims. Orthopaedic surgeries are complex, frequently performed procedures associated with significant haemorrhage and perioperative blood transfusion. Given refinements in surgical techniques and changes to transfusion practices, we aim to describe contemporary transfusion practices in orthopaedic surgery in order to inform perioperative planning and blood banking requirements. Methods. We performed a retrospective cohort study of adult patients who underwent orthopaedic surgery at four Canadian hospitals between 2014 and 2016. We studied all patients admitted to hospital for nonarthroscopic joint surgeries, amputations, and fracture surgeries. For each surgery and surgical subgroup, we characterized the proportion of patients who received red blood cell (RBC) transfusion, the mean/median number of RBC units transfused, and exposure to platelets and plasma. Results. Of the 14,584 included patients, the most commonly performed surgeries were knee arthroplasty (24.8%), hip arthroplasty (24.6%), and hip fracture surgery (17.4%). A total of 10.3% of patients received RBC transfusion; the proportion of patients receiving RBC transfusions varied widely based on the surgical subgroup (0.0% to 33.1%). Primary knee arthroplasty and hip arthroplasty, the two most common surgeries, were associated with in-hospital transfusion frequencies of 2.8% and 4.5%, respectively. RBC transfusion occurred in 25.0% of hip fracture surgeries, accounting for the greatest total number of RBC units transfused in our cohort (38.0% of all transfused RBC units). Platelet and plasma transfusions were uncommon. Conclusion. Orthopaedic surgeries were associated with variable rates of transfusion. The rate of RBC transfusion is highly dependent on the surgery type. Identifying surgeries with the highest transfusion rates, and further evaluation of factors that contribute to transfusion in identified at-risk populations, can serve to inform perioperative planning and blood bank requirements, and facilitate pre-emptive transfusion mitigation strategies. Cite this article: Bone Jt Open 2021;2(10):850–857

This study aimed to identify factors associated with increased rates of blood transfusion in patients with adolescent idiopathic scoliosis (AIS) treated with posterior spinal fusion (PSF). A retrospective case-control study was performed for AIS patients treated at a large children's hospital between August 2018 and December 2020. All patients with a diagnosis of AIS were evaluated. Data on patient demographics, AIS, and transfusion parameters were collected. Univariate regression and multivariate logistic modeling were utilized to assess risk factors associated with requiring transfusion. Odds ratios (OR) and 95% confidence interval (CI) were calculated. Surgeries were done by three surgeons and thirty anesthesiologists. To quantify the influence of anesthesia practice preferences a categorical variable was defined as “higher-transfusion practice preference”, for the provider with the highest rate of transfusion. A total of 157 AIS patients were included, of whom 56 were transfused RBC units (cases), and 101 did not receive any RBC transfusion (controls). On univariate analysis, the following variables were significantly correlated with receiving RBC transfusion: “higher-transfusion practice preference,” “administration of crystalloids,” “receiving fresh frozen plasma (FFP),” “receiving platelets,” “pre-operative hemoglobin,” “cell saver volume,” and “surgical time.” On multiple regression modeling, “pre-operative hemoglobin less than 120 g/L” (OR 14.05, 95% CI: 1.951 to 135.7) and “higher-transfusion practice preference” (OR 11.84, 95% CI: 2.505 to 63.65) were found to be meaningfully and significantly predictive of RBC transfusion. In this cohort, we identified pre-operative hemoglobin of 120 g/L as a critical threshold for requiring transfusion. In addition, we identified significant contribution from anesthesia transfusion practice preferences. Our multivariate model indicated that these two factors are the major significant contributors to allogenic blood transfusion. Although further studies are required to better understand factors contributing to transfusion in AIS patients, we suggest standardized, peri-operative evidence-based strategies to potentially help reduce variations due to individual provider preferences

Stored red blood cells (RBCs) undergo a variety of changes that impair their post-transfusion viability, but the detrimental effect of such lesion at the clinical level is a matter of debate (1) and there is no data about the incidence of postoperative infection, a complication frequently associated with transfusion of stored RBCs (2). We reviewed 9906 patients who underwent a primary or revision arthroplasty between January 2000 and December 2012. Of these, 1153 (11.6%) received transfusion during surgery or within the first 6h after surgery (early transfusion, ET) and 920 (9.3%) received transfusion only between 24 and 96 hours after surgery (late transfusion, LT). Primary end-point was prosthetic joint infection (PJI) within the first year. Demographics, joint, type of surgery, duration of surgery, number and length of storage of transfused RBCs were collected. Ethical Committee approved the study. The median age was 74.9 (IQR:68.3–80.1) years and 1546 (74.6%) were female. There were 914 (44.1%) hip and 1117 (53.9%) knee arthroplasties and 428 (20.6%) were revision surgeries. The median duration of surgery was 105 (IQR:80–145) minutes. A total of 100 (4.8%) patients had a PJI. Figure 1 shows the PJI rate according to the number of RBC units transfused and the proportion of such units that had been stored for more than 14 days, both in the ET-group (Fig. 1A) and the LT-group (Fig. 1B). In the ET-group, the fact that >50% of transfused RBCs had been stored >14 days was an independent predictor of PJI (OR:2.50, 95%CI:1.44–4.33, Hosmer-Lemeshow test P=0.972). Stored RBC occlude the microcirculation (1), thereby precluding a good oxygenation of the surgical wound and the arrival of leukocytes and prophylactic antibiotics. Both factors are involved in the progression from wound bacterial contamination to wound infection and are particularly operative in the few hours following surgery (5). It is biologically plausible that transfusion of old RBC in this early, critical period results in more wound infections as compared to RBCs transfused later