Cai et al reported a model for predicting early death for a patient with acute promyelocytic leukemia (APML). This can help to identify a patient who may require more aggressive management. The authors are from Soochow University, Jinan University and affiliated hospitals.
Patient selection: acute promyelocytic leukemia (APML)
Outcome: early death (from before therapy initiation to end of induction therapy, which typically is within 30 days)
Causes of early death: differentiation syndrome, bleeding or severe infection
Parameters:
(1) age in years
(2) WBC count in 10^9/L
(3) platelet count in 10^9/L
(4) serum LDH in U/L
Parameter
Finding
Points
age in years
<= 52 years
0
> 52 years
1.5
WBC count
< 10 * 10^9/L
0
>= 10 * 10^9/L
2
platelet count
> 10 * 10^9/L
0
<= 10 * 10^9/L
1
serum LDH
<= 500 U/L
0
> 500 U/L
1
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 0
• maximum score: 5.5
Total Score
Risk Group
Early Death Rate
0
low risk
0%
1 or 2
intermediate risk
5.5%
2.5 to 4
high risk
22.7%
4.5 to 5.5
ultra high risk
61.5%
Performance:
• The area under the ROC curve is 0.88.
Continue Reading