BACKGROUND: Our 1100-bed referral hospital uses approximately 12,000 units of random-donor platelets (PLTs) and 1,900 units of single-donor apheresis PLTs per year with a mean of 23 percent outdating. An analysis of patterns of utilization has been undertaken to evaluate practice.

STUDY DESIGN AND METHODS: Over a 9-month period, data were collected on a total of 1682 transfusion episodes in 464 patients. When the pretransfusion count was greater than 10 × 109 per L an attempt was made to identify the specific indications for PLT transfusions such as bleeding.

RESULTS: The majority (78%) of PLTs were transfused when the counts were above 10 × 109 per L. The mean pretransfusion counts for different services were: bone marrow transplant (BMT) 17.4 × 109 per L, hematology-oncology 14.6 × 109 per L, the Heart Institute 3 × 109 per L, and other services 36 × 109 per L. The percentage of transfusions given to patients with a count greater than 10 × 109 per L varied by service with 79 percent in BMT, 60 percent in hematology and oncology, 98 percent at the Heart Institute, and 81 percent in other services. Routine monitoring of counts shows a mean increment of 10.2 × 109 per L per transfusion. One hour posttransfusion counts, 24-hour posttransfustion counts, and documentation of clinical justification for transfusions was often not available.

CONCLUSIONS: The data show that most patients who receive PLTs have pretransfusion counts of more than 10 × 109 per L and more than one-third have pretransfusion counts of greater than 20 × 109 per L. The medical literature supports prophylactic PLT transfusion based solely on the count when the PLT number is 10 × 109 per L or less. Above this level additional justification is needed although there are different points of view concerning the appropriate triggers. Our data suggest that there is a need for clear hospital transfusion guidelines and ongoing monitoring of PLT use.