Matthew S Karafin1,2, Roberta Bruhn3, Nareg H Roubinian3,4, Dhuly Chowdhury5, Lirong Qu6, Edward L Snyder7, Edward L Murphy3,8, Don Brambilla5, Ritchard G Cable9, Joan F Hilton8, Elizabeth St Lezin8,10. 1. Versiti, Medical Sciences Institute, Milwaukee, Wisconsin. 2. Department of Pathology, Medical College of Wisconsin, Milwaukee, Wisconsin. 3. Vitalant Research Institute, San Francisco, California. 4. Division of Research Northern California, Kaiser Permanente, San Francisco, California. 5. Department of Pathology, Division of Transfusion Medicine, RTI International, Rockville, Maryland. 6. University of Pittsburgh, Pittsburgh, Pennsylvania. 7. Department of Laboratory Medicine, Yale University, New Haven, Connecticut. 8. Department of Laboratory Medicine, University of California, San Francisco, California. 9. American Red Cross, Farmington, Connecticut. 10. Laboratory Medicine, SF Veterans Affairs HCS, San Francisco, California.
Abstract
BACKGROUND: Patients with cancer or chronic hematologic disorders frequently receive red blood cell (RBC) transfusions. Based on long-standing assumptions, each RBC unit is thought to increase recipient hemoglobin by 1 g/dL, but smaller increments can occur. A better understanding of recipient factors affecting hemoglobin increments could help providers manage these patients. METHODS: Data were collected as a part of the observational Red Cells in Outpatients Transfusion Outcomes (RETRO) study of outpatients with hematologic or cancer-related diagnoses. Hemoglobin was measured before transfusion and 30 minutes after transfusion. A classification and regression tree (CART) analysis was performed to identify statistically significant associations with clinical variables. A corresponding prediction equation was developed and validated using linear regression. RESULTS: A total of 195 participants had both pre- and posttransfusion hemoglobin values for analysis. The median age was 66 years, and patients received one (73%) or two (27%) RBC units during the transfusion episode. The overall median change in hemoglobin was 0.6 g/dL per RBC unit. Both CART analysis and linear regression identified the following significant predictors of hemoglobin increment: number of units received (positive correlation), patient estimated circulating blood volume (negative correlation), pretransfusion hemoglobin (negative correlation), and patient age (negative correlation). CONCLUSION: In this study of outpatients with hematologic disease, most patients had a hemoglobin increment of less than 1 g/dL/unit. Recipient-specific factors influenced the hemoglobin increment at 30 minutes, and providers should consider circulating blood volume, pretransfusion hemoglobin, and recipient age, when developing patient-specific RBC transfusion plans for this unique cohort.
BACKGROUND:Patients with cancer or chronic hematologic disorders frequently receive red blood cell (RBC) transfusions. Based on long-standing assumptions, each RBC unit is thought to increase recipient hemoglobin by 1 g/dL, but smaller increments can occur. A better understanding of recipient factors affecting hemoglobin increments could help providers manage these patients. METHODS: Data were collected as a part of the observational Red Cells in Outpatients Transfusion Outcomes (RETRO) study of outpatients with hematologic or cancer-related diagnoses. Hemoglobin was measured before transfusion and 30 minutes after transfusion. A classification and regression tree (CART) analysis was performed to identify statistically significant associations with clinical variables. A corresponding prediction equation was developed and validated using linear regression. RESULTS: A total of 195 participants had both pre- and posttransfusion hemoglobin values for analysis. The median age was 66 years, and patients received one (73%) or two (27%) RBC units during the transfusion episode. The overall median change in hemoglobin was 0.6 g/dL per RBC unit. Both CART analysis and linear regression identified the following significant predictors of hemoglobin increment: number of units received (positive correlation), patient estimated circulating blood volume (negative correlation), pretransfusion hemoglobin (negative correlation), and patient age (negative correlation). CONCLUSION: In this study of outpatients with hematologic disease, most patients had a hemoglobin increment of less than 1 g/dL/unit. Recipient-specific factors influenced the hemoglobin increment at 30 minutes, and providers should consider circulating blood volume, pretransfusion hemoglobin, and recipient age, when developing patient-specific RBC transfusion plans for this unique cohort.
Authors: Robert A DeSimone; Colleen Plimier; Catherine Lee; Tamir Kanias; Melissa M Cushing; Bruce S Sachais; Steven Kleinman; Michael P Busch; Nareg H Roubinian Journal: Transfusion Date: 2020-05-01 Impact factor: 3.337
Authors: Nareg H Roubinian; Sarah E Reese; Hannah Qiao; Colleen Plimier; Fang Fang; Grier P Page; Ritchard G Cable; Brian Custer; Mark T Gladwin; Ruchika Goel; Bob Harris; Jeanne E Hendrickson; Tamir Kanias; Steve Kleinman; Alan E Mast; Steven R Sloan; Bryan R Spencer; Steven L Spitalnik; Michael P Busch; Eldad A Hod Journal: JCI Insight Date: 2022-01-11