**Background:** Fibrinogen is a critical coagulation factor, and levels below 200 mg/dL have been associated with increased perioperative bleeding in obstetric, trauma, and cardiac surgery patients. However, the role of preoperative fibrinogen as a predictor of transfusion risk in major orthopedic surgery—where both bleeding and thromboembolic risks are high—remains unclear. This study aimed to determine whether low preoperative plasma fibrinogen levels are associated with increased rates of blood-component transfusion in patients undergoing elective hip arthroplasty.
**Methods:** This prospective cohort study was conducted at Hospital São Vicente de Paulo, Passo Fundo, Brazil, from June 2018 to August 2021. Patients aged ≥18 years undergoing primary or revision hip arthroplasty for nontraumatic etiologies were included. Exclusion criteria were liver cirrhosis or active liver disease, thrombocytopenia <100,000/mm³, and use of full or prophylactic anticoagulation (except acetylsalicylic acid). Preoperative laboratory tests included blood count, platelet count, activated partial thromboplastin time, prothrombin time, international normalized ratio, and plasma fibrinogen measured by the Clauss method. The decision to transfuse was at the discretion of the anesthesiologist and surgical team. Patients were reassessed at 24 and 48 hours postoperatively, and 30- and 90-day survival was assessed via telephone. The initial sample size calculation estimated 103 participants, but after interim analysis showed a lower-than-expected prevalence of hypofibrinogenemia (10% vs. 30%), the sample was adjusted to 189 patients; 200 were enrolled to account for losses.
**Key Results:** Of 200 enrolled patients, 195 were analyzed (5 surgeries cancelled). Mean (SD) age was 62 years, and 24 patients (12.3%) received a blood-product transfusion. The mean (SD) plasma fibrinogen level was 315 (83) mg/dL. Only 13 patients (6.6%) had fibrinogen <200 mg/dL, and only 1 of these received a transfusion (absolute risk 7.69%; 95%CI: 1.37–33.31%). Among 182 patients with fibrinogen ≥200 mg/dL, 23 received transfusions. The relative risk of transfusion for fibrinogen <200 mg/dL was 0.608 (95%CI: 0.089–4.158, p=0.612). There was no association between preoperative fibrinogen and transfusion (p=0.745). Sensitivity of fibrinogen <200 mg/dL as a predictor of transfusion was 4.17% (95%CI: 0.11–21.12%), specificity was 93.57% (95%CI: 88.82–96.82%), positive predictive value was 7.69% (95%CI: 1.12–37.99%), and negative predictive value was 87.36% (95%CI: 84.47–89.79%). Test accuracy was 82.05% (95%CI: 75.93–87.17%). The positive likelihood ratio was 0.59 (95%CI: 0.08–4.36) and the negative likelihood ratio was 1.03 (95%CI: 0.94–1.13), indicating null to poor diagnostic value. Factors significantly associated with transfusion included preoperative hemoglobin (p=0.028) and hematocrit (p=0.006), surgery type (p=0.005), operative time (p=0.001), and ICU admission (p=0.006). No patient received cryoprecipitate. Of 158 patients contacted for follow-up, 1 died (due to SARS-CoV-2 infection).
**Clinical Implications:** This study provides evidence that preoperative plasma fibrinogen measurement is not a useful biomarker for predicting blood transfusion risk in patients undergoing elective hip arthroplasty for nontraumatic causes. The low sensitivity (4.17%) and poor positive likelihood ratio indicate that fibrinogen <200 mg/dL fails to identify patients at higher transfusion risk. The high specificity (93.57%) and negative predictive value (87.36%) suggest that normal fibrinogen levels may reassure clinicians, but the overall diagnostic performance is inadequate for clinical decision-making. These findings contrast with studies in cardiac surgery, trauma, and obstetrics where fibrinogen has shown predictive value, highlighting that the utility of fibrinogen testing is procedure- and population-specific. The study reinforces that preoperative hemoglobin and hematocrit, along with surgical factors (operative time, revision surgery), are more relevant predictors of transfusion in hip arthroplasty. Limitations include the single-center design, low prevalence of hypofibrinogenemia (6.6%), and lack of a standardized transfusion protocol, though the latter reflects real-world practice. The authors conclude that routine preoperative fibrinogen measurement is unnecessary for transfusion risk assessment in this patient population.