Evaluation of PEscore performance in predicting platelet transfusion efficacy in hematological patients
10.13303/j.cjbt.issn.1004-549x.2025.06.008
- VernacularTitle:PEscore对血液病患者血小板输注疗效预测的应用效果评价
- Author:
Jun LI
1
;
Lan WANG
1
;
Yihua XIE
1
;
Ziqi CHEN
1
;
Gang ZHAO
2
Author Information
1. Blood Transfusion Branch of Kailuan General Hospital, Tangshan 063000, China
2. Department of Blood Transfusion, Affiliated Hospital of Traditional Chinese Medicine of Xinjiang Medical University, Urumqi 83000, China
- Publication Type:Journal Article
- Keywords:
PEscore;
platelet transfusion efficacy;
performance
- From:
Chinese Journal of Blood Transfusion
2025;38(6):797-802
- CountryChina
- Language:Chinese
-
Abstract:
Objective: To evaluate the performance of platelet efficacy score (PEscore) in predicting platelet transfusion efficacy in hematological patients. Methods: A total of 485 patients with hematological diseases, including 298 males (62.09±15.45 years) and 187 females (59.17±16.52 years) who received platelet transfusion from January 1, 2021 to December 31, 2024 were enrolled in this study. Clinical data of the patients such as diagnosis, gender, age, number of platelet transfusion, and platelet antibody data were analyzed to investigate the incidence and influencing factors of platelet transfusion refractoriness in hematological patients at our hospital. ROC curve was used to evaluate the performance of PEscore model in predicting platelet transfusion efficacy. The predictive performance of PEscore model was validated by calculating its sensitivity, specificity, and accuracy in 115 clinical cases. Results: The incidence of platelet transfusion refractoriness in 485 cases was 29.90% (145/485). Significant differences (P<0.05) were observed between the effective and ineffective platelet transfusion groups regarding the following factors: diagnosis: lymphoma [55.32% (26/47) vs 44.68% (21/47)], the number of previous platelet transfusions [≥25: 60.78% (31/51) vs 39.22% (20/51)], platelet antibody screening result [positive: 33.76% (53/157) vs 66.24% (104/157)], and platelet transfusion volume (×10
/L) [>6: 62.71% (74/118) vs 37.29% (44/118)]. The area under the ROC curve of PEscore was 0.876. The cut-off points and corresponding sensitivity and specificity were 19.90.59% and 69.44%, respectively. The results of clinical application showed that the sensitivity, specificity and accuracy of the PEscore model for predicting platelet transfusion were 87.50%, 93.41% and 92.17%, respectively. Conclusion: The incidence of platelet transfusion refractoriness in hematological patients is relatively high. PEscore prediction model has a good performance in predicting the effect of platelet transfusion, which can provide a reliable basis for predicting the effect of platelet transfusion in hematological patients before blood transfusion.
