Trauma patients who received compatible, but not ABO-identical plasma transfusions experienced higher rates of overall complications, particularly sepsis and acute respiratory distress syndrome (ARDS), according to results of a recent study [Arch Surg. 2010;145(9):899-906]. Stepwise increases in complication rates with greater exposure were also noted. When plasma transfusions are necessary, patients should ideally be given ABO-identical plasma (a donor plasma for blood group A patients), but to improve survival in trauma situations, aggressive replacement strategies are common and patients often receive compatible but nonidentical plasma. However, little is known about the consequences of this practice. Kenji Inaba, MD, FRCSC, of the University of Southern California, Los Angeles, and colleagues performed a retrospective study of trauma patients between 2000 and 2008 on the effect of compatible but nonidentical plasma transfusions, based on a hypothesis that these patients would experience worse clinical outcomes than patients given ABO-identical plasma. The researchers divided trauma patients accordingly into 2 groups, excluding those with blood group AB and those who received ABO-incompatible plasma. They used univariate analysis to compare differences in clinical characteristics and transfusion requirements between the 2 groups. After identifying 2788 trauma patients who had received plasma during the study period, the analysis ultimately included 284 matched pairs, of whom 230 had blood group O, 30 had blood group A, and 15 had blood group B. Most patients (80.6%) had an Injury Severity Score of ≥16. Patients who received compatible but not identical plasma were given a mean of 2.8 U within the first 6 hours of trauma, 4.5 U within the first 12 hours, and a total of 13.5 U during their hospital stay. Patients who received ABO-identical plasma received a mean of 3.1 U of plasma within the first 6 hours, 4.7 U in the first 12 hours, and a total of 11.9 U. The researchers also found that during the study period the use of ABO-compatible but not identical plasma had increased by 200%; the use of ABO-identical plasma increased 57% (P<.001 for both). Results showed no difference in mortality between the 2 groups (35.2% and 33.5% for ABO-compatible but not identical and ABO-identical plasma, respectively; P=.66). Significantly more complications, however, occurred among patients given compatible but not identical plasma compared with patients who received ABO-identical plasma (53.5% vs 40.5%; P=.002). Rates of ARDS were 19.4% for the compatible but not identical group and 9.2% for the ABO-identical group (P=.001), and rates for sepsis were 38.0% compared with 28.9% (P=.02). Patients receiving compatible but not identical plasma were also more likely to remain in intensive care longer (15.6 vs 12.7 days; P=.07) and be hospitalized longer (24.2 vs 20 days; P=.09) than patients who received ABO-identical plasma. Additionally, complications increased when greater volumes of compatible but not ABO-identical plasma were transfused. Compared with patients who received ABO-identical plasma, patients who received >6 U of ABO-compatible plasma experienced a complication rate of 70% with a 4-fold higher risk of ARDS, and patients who received between 4 and 6 U had a 3-fold greater risk of ARDS. Risk of sepsis increased as well, but was not found to be statistically significant. Patients of all blood groups who received compatible but not ABO-identical plasma experienced higher rates of overall complications, but this was more pronounced in patients having blood group O, who had a 2-fold higher risk of overall complications (50.9% vs 40%; P=.03) and a 3-fold greater risk of ARDS (17.4% vs 7.8%; P<.001) compared with patients of the same blood group who received ABO-identical plasma. The authors acknowledged the study’s retrospective design as a limitation as well as the possibility that although experienced nurses recorded the complications examined in the study in real time, errors may have been made when identifying those complications and entering information into the database. Also, exact reasons for patients receiving non–ABO-identical plasma could not be ascertained.