IMMUNOHEMATOLOGY Red blood cell alloantibody frequency, specificity, and properties in a population of male military veterans Christopher A. Tormey, John Fisk, and Gary Stack BACKGROUND: The prevalence of red blood cell (RBC) alloantibodies among general, hospital-based patients typically has averaged approximately 1 percent in various studies. The frequency and properties of RBC alloantibodies in military veterans has never been examined. STUDY DESIGN AND METHODS: Transfusion records of 18,750 military veterans at a Department of Veterans Affairs (VA) medical center were retrospectively reviewed. For patients with RBC alloantibodies, the following were collected: sex, race/ethnicity, decade of birth, transfusion history, alloantibody specificity, reaction phase(s), and whether alloantibodies were detected at the initial type and screen or later. RESULTS: The RBC alloantibody prevalence was 2.4 percent among predominantly male military veter- ans. Alloantibody prevalence varied with decade of birth, ranging up to 3.3 percent (1911-1920). The 10 most frequent alloantibodies in males, as a percentage of total male antibodies, were K (21.9%), E (19.4%), D (9.1%), Le a (7.4%), Fy a (5.4%), c (4.8%), C (4.6%), P1 (3.9%), Jk a (3.7%), and Le b (3.5%). Investigation of D alloimmunization in men revealed that antibody devel- opment occurred before VA care in 80 percent (39/49) of cases. For alloimmunization during VA care, anti-D was mostly associated with the transfusion of D+ plate- lets (7/10). The majority of alloantibodies in males reacted at the antiglobulin (AG) phase, even anti-Le a , -Le b , M, -Lu a , and -P1. CONCLUSION: Military veterans have a relatively high prevalence of RBC alloantibodies, including anti-D, despite a large male predominance and lack of pregnancy-related alloimmunization. Alloantibody preva- lence was highest in World War II veterans. The major- ity of male alloantibodies reacted with AG, even those traditionally considered to be clinically insignificant. B lood group alloantibody prevalence has been reported in a variety of study populations, including hospital-based patients, patients with hematologic disorders requiring chronic transfusion therapy, and blood donors. 1-8 Alloantibody prevalence varies approximately with the proportion of individuals previously transfused or pregnant in these various study groups. Accordingly, blood group alloanti- bodies have been detected in up to 0.8 percent of blood donors and generally approximately 1 to 2 percent of hospital-based patients. 1-4 A higher proportion of known transfusion recipients, i.e., approximately 2 to 9 percent, will form a new alloantibody after the transfusion of one or more units of RBCs. 9-11 The highest blood group allo- immunization rates, ranging from approximately 9 to 30 percent, have been reported in patients receiving chronic transfusion therapy, such as those with sickle cell anemia, b-thalassemia, or hematologic malignancies. 5-8 Military veterans constitute a large group of individ- uals whose blood group alloimmunization rate has not previously been studied. Veterans have a unique combi- nation of factors whose overall impact on alloimmuniza- tion rates is difficult to predict. For example, veterans are predominantly male, thereby eliminating pregnancy as a ABBREVIATIONS: AG = antiglobulin; IS = immediate spin; VA = (Department of) Veterans Affairs; VACHCS = VA Connecticut Healthcare System. From the Department of Laboratory Medicine, Yale University School of Medicine, New Haven, Connecticut; the VA Connecti- cut Healthcare System, West Haven Campus, West Haven, Connecticut; and the Department of Pathology and Anatomical Sciences, University at Buffalo School of Medicine, Buffalo, New York. Address reprint requests to: Christopher A. Tormey, MD, Department of Laboratory Medicine, Yale-New Haven Hospital, 20York Street, New Haven, CT 06504; e-mail: christopher.tormey@yale.edu. Received for publication February 22, 2008; revision received April 17, 2008, and accepted April 21, 2008. doi: 10.1111/j.1537-2995.2008.01815.x TRANSFUSION 2008;48:2069-2076. Volume 48, October 2008 TRANSFUSION 2069