Pub Date : 2025-12-20DOI: 10.1016/j.transci.2025.104305
Nicholas Tong , Debra Masel , Phuong-Lan Nguyen , Akua Asante , Majed Refaai , Neil Blumberg
Leukoreduction is the process by which most donor leukocytes are removed from blood components either at the time of blood product manufacture or prior to transfusion. It has been shown to reduce the incidence of febrile nonhemolytic transfusion reactions, transfusion transmitted CMV infections, HLA alloimmunization and platelet refractoriness. Despite the reported benefits of leukoreduction, the FDA has not taken action to implement universal leukoreduction. In this review, we discuss the evidence from observational, retrospective and randomized controlled trials for and against leukoreduction. We advocate for the adoption of universal leukoreduction, which would improve patient outcomes and pay for itself.
{"title":"The case for universal leukoreduction of blood transfusions","authors":"Nicholas Tong , Debra Masel , Phuong-Lan Nguyen , Akua Asante , Majed Refaai , Neil Blumberg","doi":"10.1016/j.transci.2025.104305","DOIUrl":"10.1016/j.transci.2025.104305","url":null,"abstract":"<div><div>Leukoreduction is the process by which most donor leukocytes are removed from blood components either at the time of blood product manufacture or prior to transfusion. It has been shown to reduce the incidence of febrile nonhemolytic transfusion reactions, transfusion transmitted CMV infections, HLA alloimmunization and platelet refractoriness. Despite the reported benefits of leukoreduction, the FDA has not taken action to implement universal leukoreduction. In this review, we discuss the evidence from observational, retrospective and randomized controlled trials for and against leukoreduction. We advocate for the adoption of universal leukoreduction, which would improve patient outcomes and pay for itself.</div></div>","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104305"},"PeriodicalIF":1.2,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145821855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-20DOI: 10.1016/j.transci.2025.104302
Jerard Seghatchian
{"title":"Editorial commentary – Reflections on integrating some rapidly evolving essential developments into blood transfusion practice, with fresh eyes on innovation and operational excellence: Current status and future opportunities","authors":"Jerard Seghatchian","doi":"10.1016/j.transci.2025.104302","DOIUrl":"10.1016/j.transci.2025.104302","url":null,"abstract":"","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104302"},"PeriodicalIF":1.2,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145829039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-20DOI: 10.1016/j.transci.2025.104299
Sheharyar Raza , Garrett S. Booth , Jeremy W. Jacobs
{"title":"“Go with O Pos” – Blood transfusion during mass casualty in The Pitt","authors":"Sheharyar Raza , Garrett S. Booth , Jeremy W. Jacobs","doi":"10.1016/j.transci.2025.104299","DOIUrl":"10.1016/j.transci.2025.104299","url":null,"abstract":"","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104299"},"PeriodicalIF":1.2,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145829008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mobilization failure remains a challenge in autologous stem cell transplantation for multiple myeloma (MM). Chemotherapy-based mobilization protocols, though effective, are associated with toxicity, hospitalization, and delay in transplant timing. We developed an algorithm-based protocol for peripheral blood stem cell (PBSC) mobilization using G-CSF alone, with pre-emptive Plerixafor based on peripheral blood CD34+ cell counts.
Methods
This retrospective study included MM patients who underwent stem cell collection between July 2013 and November 2023 at Saskatoon Cancer Centre. The mobilization protocol aimed to collect ≥ 4 × 106 CD34+ cells/kg, using daily CD34 count monitoring to determine the use of Plerixafor. Predefined thresholds guided the decision for Plerixafor administration and apheresis timing.
Results
Out of 176 patients (median age 61 years), 100 % achieved successful mobilization. Preemptive Plerixafor was required in 13.6 % of patients (n = 24), with 1–3 doses administered. The median number of apheresis days was 2 (range: 1–4). The median CD34 yield was 6.4 × 106 cells/kg. All patients received high-dose melphalan and underwent autologous transplant, with median neutrophil and platelet engraftment at 12 and 15 days, respectively. All patients were alive at day 100 post-transplant.
Conclusion
An algorithm-driven, G-CSF-only mobilization with pre-emptive Plerixafor use based on CD34 counts resulted in 100 % successful collections, early engraftment, minimal plerixafor use, and no mobilization failures. This approach avoids chemotherapy toxicity, optimizes Plerixafor use, and prevents transplant delays.
{"title":"Safety and efficacy of G-CSF alone with pre-emptive plerixafor for autologous peripheral blood hematopoietic stem cell mobilization in newly diagnosed multiple myeloma","authors":"Youssef Elemary , Waleed Sabry , Julie Stakiw , Mark Bosch , Hadi Goubran , James Sanayei , Shruthi Kodad , Rebecca MacKay , Jill Lacey , Sabuj Sarker , Mohamed Elemary","doi":"10.1016/j.transci.2025.104304","DOIUrl":"10.1016/j.transci.2025.104304","url":null,"abstract":"<div><h3>Background</h3><div>Mobilization failure remains a challenge in autologous stem cell transplantation for multiple myeloma (MM). Chemotherapy-based mobilization protocols, though effective, are associated with toxicity, hospitalization, and delay in transplant timing. We developed an algorithm-based protocol for peripheral blood stem cell (PBSC) mobilization using G-CSF alone, with pre-emptive Plerixafor based on peripheral blood CD34+ cell counts.</div></div><div><h3>Methods</h3><div>This retrospective study included MM patients who underwent stem cell collection between July 2013 and November 2023 at Saskatoon Cancer Centre. The mobilization protocol aimed to collect ≥ 4 × 10<sup>6</sup> CD34+ cells/kg, using daily CD34 count monitoring to determine the use of Plerixafor. Predefined thresholds guided the decision for Plerixafor administration and apheresis timing.</div></div><div><h3>Results</h3><div>Out of 176 patients (median age 61 years), 100 % achieved successful mobilization. Preemptive Plerixafor was required in 13.6 % of patients (n = 24), with 1–3 doses administered. The median number of apheresis days was 2 (range: 1–4). The median CD34 yield was 6.4 × 10<sup>6</sup> cells/kg. All patients received high-dose melphalan and underwent autologous transplant, with median neutrophil and platelet engraftment at 12 and 15 days, respectively. All patients were alive at day 100 post-transplant.</div></div><div><h3>Conclusion</h3><div>An algorithm-driven, G-CSF-only mobilization with pre-emptive Plerixafor use based on CD34 counts resulted in 100 % successful collections, early engraftment, minimal plerixafor use, and no mobilization failures. This approach avoids chemotherapy toxicity, optimizes Plerixafor use, and prevents transplant delays.</div></div>","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104304"},"PeriodicalIF":1.2,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145828560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-20DOI: 10.1016/j.transci.2025.104298
Eric A. Gehrie, Garrett S. Booth, Burak Bahar
{"title":"Promoting technological advancement and innovation in transfusion medicine: Current approaches and future directions","authors":"Eric A. Gehrie, Garrett S. Booth, Burak Bahar","doi":"10.1016/j.transci.2025.104298","DOIUrl":"10.1016/j.transci.2025.104298","url":null,"abstract":"","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104298"},"PeriodicalIF":1.2,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145828503","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-19DOI: 10.1016/j.transci.2025.104301
Cynthia Sabrina Schmidt , Johannes Wutzkowsky , Henning Schäfer , Lea Reinartz , Haidar Chahin , Philipp Winnekens , Jamal Aldin Alsulaiman , Christian Temme , Veronika Lenz , René Hosch , Felix Nensa , Christoph M. Friedrich , Britta Böckmann , Peter A. Horn
Background
Efficient blood product utilization remains a critical operational challenge in transfusion medicine. One key aspect is the timely identification of products eligible for release and the monitoring of discharged patients with reserved blood products that may no longer be needed, as manual review processes for unused or misallocated units often delay their reintroduction into circulation.
Objective
To develop a real-time dashboard that integrates heterogeneous hospital data to identify blood products allocated to patients after discharge, transfer, or death.
Methods
Clinical and laboratory data from the Hospital Information System (HIS) and Laboratory Information System (LIS) were harmonized via an Extract-Transform-Load (ETL) pipeline into Fast Healthcare Interoperability Resources (FHIR) and stored in a structured database using a relational FHIR engine. A Grafana dashboard visualizes the information retrieved through SQL queries, consolidating patient data, blood product information and crossmatch status into a unified interface.
Results
The dashboard provides an overview of all blood products that could potentially be released back into stock and lists associated patients and departments. It is reviewed daily by blood depot staff as part of the clinical workflow, enabling faster verification and release of unused units. The integration of FHIR-based analytics ensures interoperability and real-time updates.
Conclusion
The dashboard demonstrates that structured FHIR data can be leveraged for operational decision support in transfusion medicine, enabling proactive stock management and improved resource utilization through seamless integration with existing clinical systems.
{"title":"A real-time dashboard for optimizing blood product utilization through a FHIR-based data integration pipeline","authors":"Cynthia Sabrina Schmidt , Johannes Wutzkowsky , Henning Schäfer , Lea Reinartz , Haidar Chahin , Philipp Winnekens , Jamal Aldin Alsulaiman , Christian Temme , Veronika Lenz , René Hosch , Felix Nensa , Christoph M. Friedrich , Britta Böckmann , Peter A. Horn","doi":"10.1016/j.transci.2025.104301","DOIUrl":"10.1016/j.transci.2025.104301","url":null,"abstract":"<div><h3>Background</h3><div>Efficient blood product utilization remains a critical operational challenge in transfusion medicine. One key aspect is the timely identification of products eligible for release and the monitoring of discharged patients with reserved blood products that may no longer be needed, as manual review processes for unused or misallocated units often delay their reintroduction into circulation.</div></div><div><h3>Objective</h3><div>To develop a real-time dashboard that integrates heterogeneous hospital data to identify blood products allocated to patients after discharge, transfer, or death.</div></div><div><h3>Methods</h3><div>Clinical and laboratory data from the Hospital Information System (HIS) and Laboratory Information System (LIS) were harmonized via an Extract-Transform-Load (ETL) pipeline into Fast Healthcare Interoperability Resources (FHIR) and stored in a structured database using a relational FHIR engine. A Grafana dashboard visualizes the information retrieved through SQL queries, consolidating patient data, blood product information and crossmatch status into a unified interface.</div></div><div><h3>Results</h3><div>The dashboard provides an overview of all blood products that could potentially be released back into stock and lists associated patients and departments. It is reviewed daily by blood depot staff as part of the clinical workflow, enabling faster verification and release of unused units. The integration of FHIR-based analytics ensures interoperability and real-time updates.</div></div><div><h3>Conclusion</h3><div>The dashboard demonstrates that structured FHIR data can be leveraged for operational decision support in transfusion medicine, enabling proactive stock management and improved resource utilization through seamless integration with existing clinical systems.</div></div>","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104301"},"PeriodicalIF":1.2,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145884459","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-19DOI: 10.1016/j.transci.2025.104303
Paulo Pereira
{"title":"Personalizing total quality management strategies for transfusion services: Integrating artificial intelligence, Big Data, and the SoHO framework","authors":"Paulo Pereira","doi":"10.1016/j.transci.2025.104303","DOIUrl":"10.1016/j.transci.2025.104303","url":null,"abstract":"","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104303"},"PeriodicalIF":1.2,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145835156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Peripheral blood stem cell (PBSC) collection is influenced by multiple factors. Present study is done to identify factors influencing successful stem cell collection in our setting and compare pre-peripheral blood CD34 + count (PBCC) based predictive protocols with current intra-procedure CD34 + count (IPC)-based protocol for dose prediction accuracy.
Methods
Retrospective analysis of data of PBSC collections performed over two years was done. Predictive algorithms were generated based on PBCC, CE2, IPC, blood volume processed (nTBV) and recipient body weight. For the prediction formulae, target dose (TD) of 5x106CD34 + cells/Kg for autologous and 10x106CD34 + cells/Kg for allogeneic collections was used.
Results
41 PBSC collections (31 allogeneic and 10 autologous) were included in the study. Correlation analysis showed FD was strongly associated with PBCC (r = 0.9,p < 0.001), IPD (r = 0.891,p = 0.001) in autologous collections, and only with IPD (r = 0.848,p < 0.001) in allogeneic collections. Prediction analysis revealed pFD as the most accurate protocol in autologous collections, with very strong correlation and low error. In allogeneic collections, pFD and prTBV correlated moderately, though pFD carried higher error. Re-evaluation using fractional procedure values to achieve target dose improved accuracy, with pceTBV and prTBV performing reliably, while pPV remained inconsistent despite strong correlation.
Conclusions
IPC-based algorithms showed the strongest correlations but were limited by poor error metrics, reducing reliability. Conversely, PBCC-based predictions provide more reliable balance of accuracy, error control and operational consistency, making it the preferred approach for minimizing donor burden and ensuring optimal processing.
{"title":"Balancing accuracy and efficiency in peripheral blood stem cell collection: A protocol comparison study","authors":"Nidhi Sharma, Khushboo Likhar, Sachin Sharma, Ashok Yadav","doi":"10.1016/j.transci.2025.104297","DOIUrl":"10.1016/j.transci.2025.104297","url":null,"abstract":"<div><h3>Background</h3><div>Peripheral blood stem cell (PBSC) collection is influenced by multiple factors. Present study is done to identify factors influencing successful stem cell collection in our setting and compare pre-peripheral blood CD34 + count (PBCC) based predictive protocols with current intra-procedure CD34 + count (IPC)-based protocol for dose prediction accuracy.</div></div><div><h3>Methods</h3><div>Retrospective analysis of data of PBSC collections performed over two years was done. Predictive algorithms were generated based on PBCC, CE2, IPC, blood volume processed (nTBV) and recipient body weight. For the prediction formulae, target dose (TD) of 5x10<sup>6</sup>CD34 + cells/Kg for autologous and 10x10<sup>6</sup>CD34 + cells/Kg for allogeneic collections was used.</div></div><div><h3>Results</h3><div>41 PBSC collections (31 allogeneic and 10 autologous) were included in the study. Correlation analysis showed FD was strongly associated with PBCC (r = 0.9,p < 0.001), IPD (r = 0.891,p = 0.001) in autologous collections, and only with IPD (r = 0.848,p < 0.001) in allogeneic collections. Prediction analysis revealed pFD as the most accurate protocol in autologous collections, with very strong correlation and low error. In allogeneic collections, pFD and prTBV correlated moderately, though pFD carried higher error. Re-evaluation using fractional procedure values to achieve target dose improved accuracy, with pceTBV and prTBV performing reliably, while pPV remained inconsistent despite strong correlation.</div></div><div><h3>Conclusions</h3><div>IPC-based algorithms showed the strongest correlations but were limited by poor error metrics, reducing reliability. Conversely, PBCC-based predictions provide more reliable balance of accuracy, error control and operational consistency, making it the preferred approach for minimizing donor burden and ensuring optimal processing.</div></div>","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104297"},"PeriodicalIF":1.2,"publicationDate":"2025-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-08DOI: 10.1016/j.transci.2025.104296
Alexandra Jimenez , Pooja Desai , Nitya Rajpal, Basimah Zahid, Jordan M. Brown, Dennis Chen, Priscilla Parra, Denden Benabdessadek, Tobias Cohen, Robert A. DeSimone, Melissa M. Cushing
{"title":"Comparative analysis of ABO titers in traditional cryoprecipitate versus INTERCEPT Fibrinogen Complex","authors":"Alexandra Jimenez , Pooja Desai , Nitya Rajpal, Basimah Zahid, Jordan M. Brown, Dennis Chen, Priscilla Parra, Denden Benabdessadek, Tobias Cohen, Robert A. DeSimone, Melissa M. Cushing","doi":"10.1016/j.transci.2025.104296","DOIUrl":"10.1016/j.transci.2025.104296","url":null,"abstract":"","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104296"},"PeriodicalIF":1.2,"publicationDate":"2025-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-28DOI: 10.1016/j.transci.2025.104287
Aqin Chen , Bing Cui , Chengcheng Xu , Yuan Xu
Objective
To develop and validate a prediction model integrating laboratory parameters and thromboelastography (TEG) for forecasting blood transfusion needs in orthopedic surgery.
Methods
This retrospective study enrolled 250 patients undergoing joint replacement, spinal fusion, or fracture fixation. Participants were randomized into training (n = 175) and validation (n = 75) sets. Preoperative demographics, laboratory indices, and TEG parameters were collected. Potential predictors were identified through univariate analysis and Least Absolute Shrinkage and Selection Operator (LASSO) regression. Significant variables were incorporated into multivariate logistic regression to identify independent factors. Multiple machine learning models—random forest (RF), k-nearest neighbors (KNN), and gradient boosting machine (GBM)—were constructed and evaluated using the area under the receiver operating characteristic curve (AUC).
Results
No significant differences were observed in baseline characteristics between the training and validation sets (P > 0.05). Univariate analysis revealed statistically significant differences between the non-transfusion and transfusion groups in the training set regarding preoperative Hb, PLT, TEG-R, TEG-K, TEG-MA, TEG-LY30, prothrombin time, and D-dimer levels (all P < 0.05). Multivariate logistic regression analysis demonstrated that preoperative Hb, preoperative PLT, and TEG-MA were independent protective factors against postoperative transfusion (all P < 0.05), whereas TEG-R, TEG-K, TEG-LY30, and preoperative prothrombin time were independent risk factors (all P < 0.05). The random forest model achieved the highest AUC (0.931), significantly outperforming the KNN (0.894) and GBM (0.813) models, thus being selected as the optimal predictive model.
Conclusion
The random forest model based on laboratory parameters and TEG parameters effectively predicts postoperative transfusion requirements in orthopedic surgery patients, providing an objective basis for preoperative blood risk assessment and individualized transfusion strategies.
{"title":"Development and validation of a novel orthopedic blood use prediction model incorporating lab tests and thromboelastography","authors":"Aqin Chen , Bing Cui , Chengcheng Xu , Yuan Xu","doi":"10.1016/j.transci.2025.104287","DOIUrl":"10.1016/j.transci.2025.104287","url":null,"abstract":"<div><h3>Objective</h3><div>To develop and validate a prediction model integrating laboratory parameters and thromboelastography (TEG) for forecasting blood transfusion needs in orthopedic surgery.</div></div><div><h3>Methods</h3><div>This retrospective study enrolled 250 patients undergoing joint replacement, spinal fusion, or fracture fixation. Participants were randomized into training (n = 175) and validation (n = 75) sets. Preoperative demographics, laboratory indices, and TEG parameters were collected. Potential predictors were identified through univariate analysis and Least Absolute Shrinkage and Selection Operator (LASSO) regression. Significant variables were incorporated into multivariate logistic regression to identify independent factors. Multiple machine learning models—random forest (RF), k-nearest neighbors (KNN), and gradient boosting machine (GBM)—were constructed and evaluated using the area under the receiver operating characteristic curve (AUC).</div></div><div><h3>Results</h3><div>No significant differences were observed in baseline characteristics between the training and validation sets (<em>P</em> > 0.05). Univariate analysis revealed statistically significant differences between the non-transfusion and transfusion groups in the training set regarding preoperative Hb, PLT, TEG-R, TEG-K, TEG-MA, TEG-LY30, prothrombin time, and <span>D</span>-dimer levels (all <em>P</em> < 0.05). Multivariate logistic regression analysis demonstrated that preoperative Hb, preoperative PLT, and TEG-MA were independent protective factors against postoperative transfusion (all <em>P</em> < 0.05), whereas TEG-R, TEG-K, TEG-LY30, and preoperative prothrombin time were independent risk factors (all <em>P</em> < 0.05). The random forest model achieved the highest AUC (0.931), significantly outperforming the KNN (0.894) and GBM (0.813) models, thus being selected as the optimal predictive model.</div></div><div><h3>Conclusion</h3><div>The random forest model based on laboratory parameters and TEG parameters effectively predicts postoperative transfusion requirements in orthopedic surgery patients, providing an objective basis for preoperative blood risk assessment and individualized transfusion strategies.</div></div>","PeriodicalId":49422,"journal":{"name":"Transfusion and Apheresis Science","volume":"65 1","pages":"Article 104287"},"PeriodicalIF":1.2,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145688428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号