TY - JOUR
T1 - The epidemiology of platelet transfusions
T2 - an analysis of platelet use at 12 US hospitals
AU - for the NHLBI Recipient Epidemiology and Donor Evaluation (REDS-III) Study
AU - Gottschall, Jerome
AU - Wu, Yan Yun
AU - Triulzi, Darrell
AU - Kleinman, Steven
AU - Strauss, Ronald
AU - Zimrin, Ann Butler
AU - McClure, Christopher
AU - Tan, Sylvia
AU - Bialkowski, Walter
AU - Murphy, Edward
AU - Ness, Paul
N1 - Funding Information:
These data from one of the largest reviews of platelet transfusion practice to date provide guidance for where to focus future clinical research studies and platelet blood management programs. Financial support: This work was supported by NHLBI contracts HHSN268201100001I, HHSN268201100002I, HHSN268201100003I, HHSN268201100004I, HHSN268201100005I, and HHSN268201100006I
Funding Information:
The infrastructure of the National Heart, Lung, and Blood Institute REDS-III domestic program has been described previously. REDS-III is a consortium of hubs, each consisting of a blood center and two to four affiliated hospitals, a single central laboratory, and a data-coordinating center. Data from 12 US hospitals comprising the REDS-III Recipient Database, which has been previously described in detail, were utilized to study recipient characteristics of platelet transfusions. A linked Components database was used to study blood donor and product characteristics. Data studied included those extracted electronic data from participating transfusion service/blood center (Components) and electronic health records from each participating hospital (Recipient) using a conserved specification created by the REDS-III program. All inpatient and outpatient platelet transfusions between January 2013 and December 2016 were included in the analysis, excluding transfusions among patient less than 18 years of age. Information on product manufacturing, including leukoreduction, irradiation, and donor blood group and type were extracted from the REDS-III Components database. Linkage of blood components to transfusion recipients was performed using an obfuscated blood product donation identifier assigned at the domestic hubs. Recipient factors were extracted from the REDS-III Recipient Database demographics table. Patient ABO and Rh blood type and pre- and posttransfusion laboratory data were collected from the laboratory table. Patient location within the hospital, medication use, and mortality were assessed using the transfers, medication, and encounter tables, respectively. Diagnosis codes were extracted from the table of the same name and collated into Healthcare Cost and Utilization Project categories (e.g., Diseases of the Circulatory System). Aggregated data were assessed for normality. Mean and standard deviation were used to describe normally distributed data, whereas median and interquartile range (IQR) were used to describe the nonnormally distributed data. Data manipulation was performed with computer software (SAS, SAS Institute).
Publisher Copyright:
© 2019 AABB
PY - 2020/1/1
Y1 - 2020/1/1
N2 - BACKGROUND: Using the Recipient and Donor Epidemiology Study-III (REDS-III) recipient and donor databases, we performed a retrospective analysis of platelet use in 12 US hospitals that were participants in REDS-III. STUDY DESIGN AND METHODS: Data were electronically extracted from participating transfusion service and blood center computer systems and from medical records of the 12 REDS-III hospitals. All platelet transfusions from 2013 to 2016 given to patients aged 18 years and older were included in the analysis. RESULTS: There were 28,843 inpatients and 2987 outpatients who were transfused with 163,719 platelet products (103,371 apheresis, 60,348 whole blood derived); 93.5% of platelets were leukoreduced and 72.5% were irradiated. Forty-six percent were transfused to patients with an International Classification of Diseases, 9th/10th Revision (ICD-9/10) diagnosis of leukemia, myelodysplastic syndrome (MDS), or lymphoma. The general ward and the intensive care unit (ICU) were the most common issue locations. Only 54% of platelet transfusions were ABO identical; and 60.6% of platelet transfusions given to Rh-negative patients were Rh positive. The most common pretransfusion platelet count range for inpatients was 20,000 to 50,000/μL, for outpatients it was 10,000 to 20,000/μL. Among ICU patients, 35% of platelet transfusion episodes had a platelet count of greater than 50,000/μL; this was only 8% for general ward and 2% for outpatients. The median posttransfusion increment, not corrected for platelet dose and/or patient size, ranged from 12,000 to 20,000/μL for inpatients, and from 17,000 to 27,000/μL for outpatients. CONCLUSIONS: These data from one of the largest reviews of platelet transfusion practice to date provide guidance for where to focus future clinical research studies and platelet blood management programs.
AB - BACKGROUND: Using the Recipient and Donor Epidemiology Study-III (REDS-III) recipient and donor databases, we performed a retrospective analysis of platelet use in 12 US hospitals that were participants in REDS-III. STUDY DESIGN AND METHODS: Data were electronically extracted from participating transfusion service and blood center computer systems and from medical records of the 12 REDS-III hospitals. All platelet transfusions from 2013 to 2016 given to patients aged 18 years and older were included in the analysis. RESULTS: There were 28,843 inpatients and 2987 outpatients who were transfused with 163,719 platelet products (103,371 apheresis, 60,348 whole blood derived); 93.5% of platelets were leukoreduced and 72.5% were irradiated. Forty-six percent were transfused to patients with an International Classification of Diseases, 9th/10th Revision (ICD-9/10) diagnosis of leukemia, myelodysplastic syndrome (MDS), or lymphoma. The general ward and the intensive care unit (ICU) were the most common issue locations. Only 54% of platelet transfusions were ABO identical; and 60.6% of platelet transfusions given to Rh-negative patients were Rh positive. The most common pretransfusion platelet count range for inpatients was 20,000 to 50,000/μL, for outpatients it was 10,000 to 20,000/μL. Among ICU patients, 35% of platelet transfusion episodes had a platelet count of greater than 50,000/μL; this was only 8% for general ward and 2% for outpatients. The median posttransfusion increment, not corrected for platelet dose and/or patient size, ranged from 12,000 to 20,000/μL for inpatients, and from 17,000 to 27,000/μL for outpatients. CONCLUSIONS: These data from one of the largest reviews of platelet transfusion practice to date provide guidance for where to focus future clinical research studies and platelet blood management programs.
UR - http://www.scopus.com/inward/record.url?scp=85076806386&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076806386&partnerID=8YFLogxK
U2 - 10.1111/trf.15637
DO - 10.1111/trf.15637
M3 - Article
C2 - 31850522
AN - SCOPUS:85076806386
SN - 0041-1132
VL - 60
SP - 46
EP - 53
JO - Transfusion
JF - Transfusion
IS - 1
ER -