Plasma fibrinogen as a predictor of blood component transfusion in the perioperative period in patients undergoing major orthopaedic surgery: a cohort study.
Description
Background: There is a trend towards increased perioperative bleeding in patients with plasma fibrinogen levels < 200 mg.dL-1. This study aimed to assess whether there is an association between preoperative fibrinogen levels and blood components transfusion up to 48 hours after major orthopaedic surgery. Matherials and methods: This observational cohort study included 195 adult patients of both sexes who underwent primary hip arthroplasty or revision hip arthroplasty for non-traumatic aetiologies. Plasma fibrinogen, blood count, coagulogram, and platelet count were measured in the preoperative period. A plasma fibrinogen level of 200 mg.dL-1 was the cut-off value used to predict blood components transfusion. Results: The mean (SD) plasma fibrinogen value was 325 (83) mg.dL-1; only 13 patients had values < 200 mg.dL-1, and only one of these received a blood transfusion, with an absolute risk of 7.69% (1/13; 95% CI: 1.37% - 33.31%). There was no association between preoperative plasma fibrinogen values and blood transfusion (p = 0.745). The sensitivity and the positive predictive value of plasma fibrinogen < 200 mg.dL-1 as a predictor of blood transfusion were 4.17% (95% CI: 0.11% - 21.12%) and 7.69% (95% CI: 1.12% - 37.99%), respectively. The accuracy of the test was 82.05% (95% CI: 75.93% - 87.17%), but both positive and negative likelihood ratios were poor. Conclusion: Preoperative plasma fibrinogen level in hip arthroplasty patients was not associated with blood components transfusion and is ineffective as a biomarker for blood components transfusion risk.
Files
Steps to reproduce
In this prospective cohort study, patients with aged 18 years or older of both sexes undergoing hip arthroplasty or surgical revision of non-traumatic hip arthroplasty between June 2018 and August 2021 were studied. In the preoperative period, laboratory tests were performed, including plasma fibrinogen. Intraoperatively, related variables to bleeding and blood transfusion were studied. The patients were reassessed 24 and 48 hours after the procedure regarding blood component transfusion or referral to the intensive care unit. Considering the incidence of hypofibrinogenemia in patients at risk of surgical bleeding in trauma victims, liver transplantation and cardiovascular surgery, the proportion of exposed patients (ie, plasma fibrinogen < 200 mg dL-1) was initially estimated at 30% for sample size calculation. Since the evidence estimates a 23% to 54% probability of blood transfusion in patients with hypofibrinogenemia, we estimated the risk of transfusion in patients with fibrinogen deficiency to be 50%. In contrast, blood transfusion in the unexposed group (ie, fibrinogen > 200 mg dL-1) was estimated at 20%, based on transfusion risk in patients from the general population who undergo large joint arthroplasty. Therefore, considering a two-tailed α of 0.05 and a power of 80%, it was initially estimated that 103 participants would be needed to assess whether there is a relationship between preoperative plasma fibrinogen values and blood component transfusion in orthopaedic hip surgery. After a preliminary analysis of data, the magnitude of the factor under study (plasma fibrinogen < 200 mg dL-1) in this sample was lower than that found in the literature. Thus, the sample size was adjusted according to the incidence of hypofibrinogenemia, reducing the estimate from 30% to 10% in relation to the proportion of exposed subjects (fibrinogen < 200 mg dL-1) in the first stage. The estimate of the other variables was maintained. Given the adjustments and considering a two-tailed α of 0.05 and a power of 80%, 189 patients were required to determine whether there is an association (p < 0.05) between preoperative plasma fibrinogen levels and perioperative blood component transfusion in hip arthroplasty. Descriptive analyses are presented as mean (standard deviation) for quantitative variables and absolute and relative frequency (%) for categorical variables. The chi-square test or Fisher's exact test were used to compare categorical variables, while Student's t-test or the non-parametric Mann-Whitney U test were used for numerical variables. Differences were considered significant at p < 0.05. Since data collection involved diagnostic tests, the accuracy, likelihood ratios, sensitivity, specificity, and positive and negative predictive values of plasma fibrinogen measurement were calculated as predictors of perioperative blood component transfusion. The absolute and relative risks of blood component transfusion were calculated.