RESUMO
BACKGROUND: More than a million surgeries are performed annually in the United States for hip or knee arthroplasty or hip fracture stabilization. One-fifth of these patients have blood transfusions during their hospital stay. Increases in transfusion rates have caused concern about increased adverse events from unnecessary transfusions. METHODS: We systematically reviewed randomized trials examining the effect of restrictive vs liberal transfusion thresholds on patients having major orthopedic surgery. Study results were meta-analyzed with a random-effects model and heterogeneity was tested with the I2 statistic. Study risk of bias was assessed using a modified Jadad scale and evidence strength was measured using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) system. RESULTS: A total of 504 published articles were screened, and 15 met inclusion criteria. The articles described 9 randomized trials, most comparing transfusion thresholds of 8 vs 10 g/dL hemoglobin. All involved hip or knee arthroplasty and/or hip fracture patients. Moderate-strength evidence suggested a reduction in need for transfusion (relative risk, 0.53; 95% confidence interval [CI], 0.39-0.71; I2 = 95%) and mean number of units transfused (-0.95 units, 95% CI, -1.48 to -0.41, I2 = 98%). There was a possible reduction in overall infections with more restrictive transfusion thresholds, although the result was not statistically significant (relative risk, 0.71; 95% CI, 0.47-1.06; I2 = 54%). Moderate-strength evidence suggested no differences in other clinical outcomes between the groups. Limitations included incomplete blinding, inconsistency, and imprecision. CONCLUSION: Moderate-strength evidence suggests that restrictive transfusion practices reduce utilization of transfusions and may decrease infections without increasing adverse outcomes in major orthopedic surgery.