Blay et al reported a model for predicting the need for platelet transfusion during cancer chemotherapy. The authors are from multiple institutions in France.
Patient selection: after cancer chemotherapy
Outcome: platelet transfusion
Parameters from a logistic regression model:
(1) platelet count per microliter on day 1 before chemotherapy
(2) absolute lymphocyte count on day 1 before chemotherapy
(3) type of cancer chemotherapy
(4) ECOG performance status
High risk chemotherapy regimens - one of the following:
(1) > 90 mg per square meter doxorubicin per course
(2) > 90 mg per square meter epirubicin per course
(3) > 100 mg per square meter cisplatin per course
(4) > 9 g per square meter ifosfamide per course
(5) > 1 g per square meter cyclophosphamide per course
(6) > 500 mg per square meter etoposide per course
(7) > 1 g per square meter cytarabine per course
Parameter |
Finding |
Points |
platelet count |
>= 150,000 per µL |
0 |
|
< 150,000 per µL |
1 |
absolute lymphocyte count |
> 700 per µL |
0 |
|
<= 700 per µL |
1 |
type of chemotherapy |
high risk |
1 |
|
not high risk |
0 |
ECOG performance stagus |
0 or 1 |
0 |
|
2 to 4 |
1 |
total score =
= SUM(points for all 4 parameters)
Interpretation:
• minimum score: 0
• maximum score: 4
• The higher the score the greater the need for platelet transfusion.
Total Score |
Percent Receiving Platelet Transfusions |
0 |
1.5% |
1 |
7% |
2 |
13% |
>= 3 |
45% |
Specialty: Clinical Laboratory, Hematology Oncology