The predictive power of baseline hemoglobin for transfusion risk in surgery patients.

Preoperative hemoglobin concentration may be an important predictor of transfusion risk in surgical procedures with significant expected blood loss. Contemporary studies investigating transfusion risk with regard to the relationship between perioperative administration of Epoetin alfa and baseline hemoglobin provide data to test this hypothesis. The predictive power of seven preoperative variables (hemoglobin concentration, age, erythropoietin level, ferritin concentration, serum iron, total iron-binding capacity, and predicted blood volume) on transfusion risk was examined via retrospective logistic regression analysis of 276 orthopedic surgical patients. In the two studies used to perform the regression analysis, patients were treated daily with either Epoetin alfa or placebo. Based on the retrospective analyses, a prospective study was conducted to validate the hypothesis. Of the seven variables evaluated, baseline hemoglobin concentration and predicted blood volume were significantly predictive of transfusion risk in both Epoetin alfa- and placebo-treated patients. Further, an inverse correlation between hemoglobin concentration and transfusion risk was demonstrated in placebo-treated patients. Placebo-treated patients with hemoglobin > 10 to < or = 13 g/dL had an approximately twofold greater risk of transfusion than patients with hemoglobin > 13 g/dL. In contrast to placebo treatment, Epoetin alfa significantly reduced transfusion risk in patients with hemoglobin > 10 to < or = 13 g/dL. Baseline hemoglobin concentration is an excellent predictor of transfusion risk in orthopedic surgical patients. As a result, hemoglobin testing should be considered a part of routine preoperative testing for orthopedic surgical patients.

Effects of heteroscedasticity upon certain analyses when regression lines are not parallel.

The two sample analysis of covariance model is considered where the regression lines are found not to be parallel. There are two analyses often utilized in this case. One is to obtain a standard confidence interval for the covariate value where the two lines intersect, and the other is to find a simultaneous confidence region for the difference between the regression lines. The asymptotic robustness to heteroscedasticity of the regressions' errors of the coverage probability of each of these two confidence procedures is considered. When the sample sizes are approximately equal, and the covariate means and the covariate variances are similar between the samples, the unequal regression variances are shown to have little effect on the asymptotic coverage probabilities of these two confidence procedures. Discussions concerning the effects of unequal sample sizes and inequitably distributed covariate values are also presented. These results are illustrated in application to a clinical trial of recombinant human erythropoetin versus placebo to treat patients with anemia secondary to advanced cancer, who were not receiving chemotherapy.