Blood transfusion exposure does not influence survival in patients with carcinoma of the breast.

There is abundant evidence of immune modulation induced by exposure to blood transfusions. Some studies have demonstrated a detrimental effect of transfusion on the recurrence of malignant disease and survival. We retrospectively studied the impact of blood transfusion exposure on 229 patients with breast cancer who were seen from July 1973 to September 1980, had at least 5 years' follow-up and had been randomized by therapy at the time of diagnosis. The patients were divided into four groups according to transfusion history: Group 1 (111 patients), no transfusion; Group 2 (34 patients), first transfusion after mastectomy; Group 3 (41 patients), first transfusion at mastectomy; and Group 4 (43 patients), first transfusion before mastectomy. All transfused patients received red cells or whole blood or both. At the time of analysis, 124 (54%) of the patients had died. Only Group 2 was statistically associated with decreased survival; recurrence of disease was 85 percent in this group, compared with 53 percent to 61 percent in the other three groups (p = 0.006, log-rank test). In general, Group 2 patients received transfusions because of recurrent disease. We conclude that transfusions before or at mastectomy are not associated with increased recurrence or reduced survival in patients with breast cancer.

Platelet crossmatch evaluation in refractory hematologic patients.

Although previous investigators have attempted to identify platelet crossmatching methods that could be used routinely for pretransfusion testing, such studies have excluded patients with underlying clinical conditions inherently associated with low corrected platelet increments. Because many refractory hematologic patients have such underlying conditions, we decided to assess the usefulness of platelet crossmatching (in addition to HLA matching) in determining the posttransfusion corrected platelet count increment in eight medically complicated patients who received a total of 35 single-donor platelet transfusions. In this limited study, the predictive value of a negative crossmatch was only 55%, whereas that of a positive crossmatch was 88%. The test used had a sensitivity of 65% and a specificity of 83%. The results of our study suggest that platelet crossmatching may be a useful additional study for predicting the outcome of transfusion--even in medically complex cases--if it can be done rapidly and relatively inexpensively.