[The Influencing Factors of Platelet Transfusion Efficacy in Hematological Disease Chemotherapy Patients and the Construction and Validation of a Nomogram Risk Prediction Model].

OBJECTIVE

To explore the influencing factors of platelet transfusion efficacy in hematological disease chemotherapy patients, construct a nomogram for predicting platelet transfusion refractoriness (PTR), and verify its predictive effect.

METHODS

A total of 158 hematological disease chemotherapy patients who underwent platelet transfusion in Haian People's Hospital from January 2021 to December 2023 (469 transfusions) were included. The platelet transfusion effects were reviewed and they were separated into effective group (362 transfusions) and PTR group (107 transfusions). The 16 items of data of two groups were obtained and compared. Logistic regression model was used to screen for risk factors of PTR in patients. A nomogram prediction model was constructed based on the screening results, and its discriminability (using the receiver operating characteristic curve) and consistency (using calibration curve) were evaluated.

RESULTS

There were statistically significant differences between PTR group and effective group in terms of the proportions of fever, infection, splenomegaly, chemotherapy frequency≥6 times, combined chemotherapy, infusion frequency≥3 times, and using platelet raising drugs (all P <0.05). The risk factors affecting PTR in hematological disease chemotherapy patients mainly included infection (OR=2.781, 95%CI: 1.661-4.657), splenomegaly (OR=2.682, 95%CI: 1.527-4.713), chemotherapy frequency≥6 times (OR=2.296, 95%CI: 1.432-3.682), combination chemotherapy (OR=1.844, 95%CI: 1.159-2.935), and blood transfusion frequency≥3 times (OR=1.784, 95%CI: 1.105-2.879). Internal validation of the model showed that the area under the ROC curve was 0.834 (95%CI: 0.790-0.879), and the calibration curve fitted well, the goodness of fit of the HL test showed χ2=13.151, P=0.107.

CONCLUSION

A nomogram model constructed based on infection status, spleen condition, chemotherapy frequency, combination chemotherapy, and platelet transfusion frequency can effectively predict PTR in hematological disease chemotherapy patients, which is beneficial for screening high-risk individuals for PTR and targeted prevention.