The necessary and the unnecessary transfusion: a critical review of reported appropriateness rates and criteria for red cell transfusions

BACKGROUND: The purpose of this study was to evaluate the criteria for assessing the appropriateness of red cell transfusions. The data were obtained by a computer search of all English-language literature from 1966 to October 1992. STUDY DESIGN AND METHODS: Nine studies were selected, which dated from 1986 to 1989 and employed explicit criteria evaluating the appropriateness of red cell transfusion in adults. The following data were abstracted from all studies: study design, timing, location, criteria for evaluating appropriateness, and rate of appropriate or inappropriate transfusions. RESULTS: Five studies evaluated transfusion appropriateness. Appropriateness rates ranged from 88 to 99 percent in three studies, and inappropriateness rates ranged from 0.3 to 57.3 percent in two studies. Four studies evaluated transfusion inappropriateness and reported inappropriateness rates of 18 to 55 percent. Substantial variation was found in the criteria for an appropriate or an inappropriate transfusion. Appropriateness rates did not depend upon characteristics of the study design, location, or timing of data collection. Restrictiveness in the criteria used to determine appropriateness and the use of additional implicit evaluation after an initial explicit review affected appropriateness rates. CONCLUSION: In the 1980s, high rates of inappropriate transfusion and low rates of appropriate transfusion were still reported. Appropriateness rates varied widely, in part because of marked variation in the criteria for an appropriate transfusion. Newly derived standards for an appropriate red cell transfusion, published in 1992, appear to provide a simple and objective means of evaluating the appropriateness of a transfusion. Appropriateness rates resulting from the application of these new standards have not yet been determined.