Clinical results of a stringent policy on prophylactic platelet transfusion : non-randomized comparative analysis in 190 bone marrow transplant patients from a single institution

The threshold for prophylactic platelet transfusion remains controversial. Usually the decision is based on arbitrary numerical criteria. The classical 20x10 9 /l trigger could be safely reduced with considerable benefits. Few studies have evaluated the clinical impact of stringent policies. We have performed a retrospective analysis comparing major haemorrhages during hospitalization in 190 patients undergoing BMT in two different periods. In 87 patients transplanted from 1990 to 1991, the 20x10 9 /l trigger was used for prophylactic platelet transfusion. In 103 other patients transplanted from 1993 to 1994, we adopted a stringent prophylactic policy : <10x10 9 /l for stable patients and (20x10 9 /l when higher platelet consumption factors were present. In the stringent group, 12 patients presented 13 major haemorrhages and four died from haemorrhage. In the classical group 12 patients presented 14 major haemorrhages and three died from haemorrhage. Platelet consumption factors were present in 12 of 13 haemorrhages in the stringent group and in 12 of 14 in the classical group. By contrast, stable patients presented less haemorrhages (2/14 and 1/13, respectively). A statistically significant reduction in the use of platelet units was observed when comparing both groups : the median of platelet units administered in the first 100 days of transplant was 73 (3-943) and 54 (0-647) in the classical and in the stringent group, respectively (P < 0.01) ; and the median of platelet units received per day was 0.8 (0.03-30) and 0.5 (0-6.94) (P < 0.01). Our results emphasize the safety of a stringent prophylactic platelet transfusion policy after BMT, reducing the overall use of platelet transfusions. Further studies are necessary to confirm these results and to define optimal transfusion strategies.