Predictors and Outcomes of Postoperative Hemoglobin of <8 g/dL in Total Joint Arthroplasty

Abstract

Background: Restrictive transfusion practices have decreased transfusions in total joint arthroplasty (TJA). A hemoglobin threshold of <8 g/dL is commonly used. Predictors of this degree of postoperative anemia in TJA and its association with postoperative outcomes, independent of transfusions, remain unclear. We identified predictors of postoperative hemoglobin of <8 g/dL and outcomes with and without transfusion in TJA.

Methods: Primary elective TJA cases performed with a multimodal blood management protocol from 2017 to 2018 were reviewed, identifying 1,583 cases. Preoperative and postoperative variables were compared between patients with postoperative hemoglobin of <8 and ≥8 g/dL. Logistic regression and receiver operating characteristic curves were used to assess predictors of postoperative hemoglobin of <8 g/dL.

Results: Positive predictors of postoperative hemoglobin of <8 g/dL were preoperative hemoglobin level (odds ratio [OR] per 1.0-g/dL decrease, 3.0 [95% confidence interval (CI), 2.4 to 3.7]), total hip arthroplasty (OR compared with total knee arthroplasty, 2.1 [95% CI, 1.3 to 3.4]), and operative time (OR per 30-minute increase, 2.0 [95% CI, 1.6 to 2.6]). Negative predictors of postoperative hemoglobin of <8 g/dL were tranexamic acid use (OR, 0.42 [95% CI, 0.20 to 0.85]) and body mass index (OR per 1 kg/m2 above normal, 0.90 [95% CI, 0.86 to 0.94]). Preoperative hemoglobin levels of <12.4 g/dL in women and <13.4 g/dL in men best predicted postoperative hemoglobin of <8 g/dL. Overall, 5.2% of patients with postoperative hemoglobin of 7 to 8 g/dL and 95% of patients with postoperative hemoglobin of <7 g/dL received transfusions. Patients with postoperative hemoglobin of <8 g/dL had longer hospital stays (p < 0.001) and greater rates of emergency department visits or readmissions (p = 0.001) and acute kidney injury (p < 0.001). Among patients with postoperative hemoglobin of <8 g/dL, patients who received transfusions had a lower postoperative hemoglobin nadir (p < 0.001) and a longer hospital stay (p = 0.035) than patients who did not receive transfusions.

Conclusions: Postoperative hemoglobin of <8 g/dL after TJA was associated with worse outcomes, even for patients who do not receive transfusions. Optimizing preoperative hemoglobin levels may mitigate postoperative anemia and adverse outcomes.

Level of evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.