Search

BACKGROUND: The American Red Cross (ARC) initiated a comprehensive donor hemovigilance program in 2003. We provide an overview of reported complications after whole blood (WB), apheresis platelet (PLT), or automated red cell (R2) donation and analyze factors contributing to the variability in reported complication rates in our national program. STUDY DESIGN AND METHODS: Complications recorded at the collection site or reported after allogeneic WB, apheresis PLT, and R2 donation procedures in 36 regional blood centers in 2006 were analyzed by univariate and multivariate logistic regression. RESULTS: Complications after 6,014,472 WB, 449,594 PLT, and 228,183 R2 procedures totaled 209,815, 25,966, and 12,282 (348.9, 577.5, and 538.3 per 10,000 donations), respectively, the vast majority of which were minor presyncopal reactions and small hematomas. Regional center, donor age, sex, and donation status were independently associated with complication rates after WB, PLT, and R2 donation. Seasonal variability in complications rates after WB and R2 donation correlated with the proportion of donors under 20 years old. Excluding large hematomas, the overall rate of major complications was 7.4, 5.2, and 3.3 per 10,000 collections for WB, PLT, and R2 procedures, respectively. Outside medical care was recorded at similar rates for both WB and automated collections (3.2 vs. 2.9 per 10,000 donations, respectively). CONCLUSION: The ARC data describe the current risks of blood donation in a model multicenter hemovigilance system using standardized definitions and reporting protocols. Reported reaction rates varied by regional center independently of donor demographics, limiting direct comparison of different regional blood centers.

Context.—Serious noninfectious complications are far more likely to occur than viral disease transmission from blood component transfusion.Objective.—To compile a comprehensive list of the noninfectious risks of transfusion, examples of published risk estimates, and summaries of recent information regarding cause, prevention, or management of noninfectious transfusion risks.Data Sources.—Information was obtained from peer-reviewed English-language medical journal publications since 1990.Conclusions.—Early complications, although potentially more serious, usually occur less frequently (

BACKGROUND: Hematocrit (Hct) values in healthy adult populations exhibit seasonal variation, with the lowest values occurring in the summer. The extent to which environmental temperature contributes to the seasonal trend in deferral rates for unacceptable Hct in the American Red Cross was further analyzed. STUDY DESIGN AND METHODS: A centralized database of donations during 2002 to 2004, constituting 24.3 million donor presentations, was further characterized. Data on mean monthly temperature in the United States were obtained for the same period from a government agency. Multivariate regression analyses were performed to determine the relationship between Hct deferral rates among blood donors and environmental temperature and donor characteristics. RESULTS: Hct deferral rates were associated with mean monthly temperature in the United States (R2 = 0.77). The relationship between the Hct deferral rate and environmental temperature was strongest in the region of the country with the highest seasonal variation in temperature, followed by regions with intermediate and low seasonal variation in temperature, respectively. The seasonal pattern in Hct deferral rates occurred in both sexes and across all age groups, with significantly higher Hct deferral rates occurring in June through August compared to other quarters (p

BACKGROUND: American Red Cross surveillance data on transfusion-related acute lung injury (TRALI) fatalities were analyzed to evaluate the association with components from donors with white blood cell (WBC) antibodies and to examine the potential impact of the selective transfusion of plasma from male donors. STUDY DESIGN AND METHODS: Suspected TRALI reports in 2003 through 2005 were identified and all fatalities were reviewed and classified by three physicians as “probable TRALI” or of “unrelated etiology,” with independent review of the associated serologic investigation. Hospital investigational and reporting biases could not be fully controlled in this retrospective study. RESULTS: A total of 550 reports of suspected TRALI, including 72 fatalities, were investigated. The number of reports increased each year and the rate varied by geographic region. Retrospective review of fatalities revealed 38 cases of probable TRALI, the majority (24 of 38 [63%]) after plasma transfusion. A female, WBC antibody–positive donor was involved in 71 percent (27 of 38) of cases and in 75 percent (18 of 24) of cases involving plasma transfusion. Female antibody-positive donors were more likely to be associated with probable TRALI than with unrelated cases (p = 0.0001; odds ratio [OR], 9.5; 95% confidence interval [CI], 2.9-31.1]. The rate of probable TRALI among recipient fatalities was higher for plasma components (1:202,673; OR, 12.5; 95% CI, 5.4-28.9) and apheresis platelets (PLTs; 1:320,572; OR, 7.9; 95% CI, 2.5-24.8) compared to red cells (1:2,527,437). Male donors contributed 64.5 and 52.0 percent of distributed apheresis PLTs and plasma components, respectively, in 2005. CONCLUSION: Plasma components linked to female donors with WBC antibodies were responsible for the majority of probable TRALI fatalities. Prudent measures to limit transfusion of WBC antibody–containing plasma components may prevent as many as six fatalities per year in the Red Cross system.

BACKGROUND: Routine quality control (QC) testing for bacterial contamination in apheresis platelet (PLT) products was implemented in all 36 regional blood centers of the American Red Cross in March 2004. STUDY DESIGN AND METHODS: PLT samples were cultured under aerobic conditions until the end of the product shelf life or when a positive reaction was indicated. To confirm the initial positive reaction, a new sample was taken from the unit for reculturing. All positive culture bottles were referred for bacterial isolation and identification. Bacterial testing data along with apheresis PLT collection information were collected for analysis. Reports and investigations of potential septic reactions to apheresis PLTs were reviewed. RESULTS: In the first 10 months of bacterial testing, 226 of 350,658 collections tested initially positive. Sixty-eight were confirmed on resampling to be bacterially contaminated for an overall confirmed-positive rate of 0.019 percent or 1 in 5157. Staphylococcus spp. (47.1%) and Streptococcus spp. (26.5%) were the most frequently isolated bacteria; Gram-negative bacteria accounted for 17.6 percent of the confirmed-positive products. Of the 354 apheresis PLT products derived from all 226 initial test–positive cases, 38 (10.7%) were transfused by the time the initial positive reaction was indicated. None of these transfused products, however, had a confirmed-positive bacterial screen and no patient who had been transfused with an unconfirmed-positive product had evidence of a septic transfusion reaction. Three high-probability septic transfusion reactions to screened, negative components were identified. In all three cases, a coagulase-negative Staphylococcus was implicated. CONCLUSION: Our experience demonstrates that bacterial testing of apheresis PLT products as a QC measure was efficiently implemented throughout the American Red Cross system and that this new procedure has been effective in identifying and preventing the transfusion of many, although not all, bacterially contaminated PLT units.

The protease-activated receptor (PAR)-2 is present on the smooth muscle and epithelium of human airways and can be activated by mast cell tryptase, trypsin, or the PAR-2 activating peptide (AP). Trypsin and the PAR-2 AP induced contractions in human isolated airways, and these contractions were potentiated in the presence of the cyclooxygenase (COX) inhibitor indomethacin. Trypsin also increased the contractions to histamine in airways from sensitized (allergic) patients but not from nonsensitized (nonallergic) patients. Tryptase purified from human lung, skin and lung recombinant β-tryptases, trypsin, and the PAR-2 AP all increased DNA synthesis in human airway smooth muscle (HASM) cells. Activation of PAR-2 by tryptase, trypsin, and the PAR-2 AP did not induce PGE2release from HASM cells. Trypsin and the PAR-2 AP increased the levels of intracellular calcium in HASM cells, with desensitization evident after treatment with either agonist. In conclusion, activation of PAR-2 can induce contractions of human airways, potentiate contractions to histamine, and induce proliferation and therefore may contribute to airway diseases such as asthma.

The medical conditions in which transfusion is required, types of transfusions, and host response to transfused cells are different in neonates and infants than in adults. As a result, certain features and complications of transfusion are more important. This article explores key aspects of these differences, indications for transfusion therapy, and possible actions blood banks can take to improve transfusion safety and efficacy.

Evaluation of a patient prompted the systematic review of causes of abrupt onset anemia in children and the development of a protocol for the immunohematologic workup of such cases. The patient, a 3-year-old boy, presented to our institution two weeks following a viral illness with gross hematuria, jaundice, and severe anemia (hematocrit 14.5%). He was found to have a positive direct antiglobulin test (DAT) due to primarily complement sensitization (2+) with smaller amounts of IgG (weak). A nonspecific cold agglutinin, anti-P1, and a hemolytic biphasic Donath-Landsteiner antibody were demonstrated in his serum, and he was diagnosed with paroxysmal cold hemoglobinuria (PCH). The hemolytic anemia resolved with packed red cell transfusions, intravenous immune globulin, and steroid treatment. It was concluded that a complete immunohematologic evaluation, including a DAT, cold agglutinin, and Donath-Landsteiner antibody tests, is pivotal to prompt and accurate assessment of the cause of an abrupt onset of severe anemia in children.

Objective. Extracorporeal membrane oxygenation (ECMO) has dramatically improved the survival of neonates with life-threatening respiratory and cardiac failure. However, ECMO requires numerous transfusions with significant risks. This study evaluated the effects of changing transfusion practices and blood component management on blood donor exposures in neonatal ECMO. Design. A 3-year retrospective chart review of all neonatal ECMO patients from December 1989 through November 1992 was undertaken. During this 3-year period, transfusion practices and blood product preparation were altered twice to reduce blood donor exposures. The use of apheresis platelets and fresh frozen plasma (FFP), and preserving the expiration date on packed red blood cells (PRBCs) through the use of a sterile connecting device, allowed multiple transfusions from individual donor components. In addition, education of the ECMO physicians was focused on standardizing and reducing transfused volumes. Sixty-four surviving neonatal patients (91.4%) were evaluated. Five patients had excessive bleeding and were excluded from analysis. The remaining 59 patients were divided into three protocols based upon the transfusion practice at the time of their ECMO course. Protocol 1 received PRBCs less than 5 days of age, volume-reduced platelet concentrates, and standard FFP units up to 24 hours after thawing. Changes in transfusion practice for protocol 2 included extended outdate for PRBCs to 10 days, and using single donor apheresis platelet aliquots. The third protocol entailed the use single donor apheresis FRP aliquots in addition to the protocol 2 changes. Results. Total PRBC transfusion volumes (721 ± 406 ml for protocol 1, 637 ± 172 ml for protocol 3) and associated blood donor exposures (5 ± 2.1 for protocol 1, 3.9 ± 0.9 for protocol 3) did not change substantially over the reviewed period. However, FFP transfusion volumes (478 ± 170 ml for protocol 1, 274 ± 63 ml for protocol 3), FFP-related donor exposures (4.5 ± 1.6 for protocol 1, 1.2 ± 0.4 for protocol 3) and platelet-related donor exposures (4.6 ± 3.6 for protocol 1, 2.5 ± 1.5 for protocol 3) were reduced progressively and significantly from protocol 1 to protocol 3 (P < .01, Tukey's B test adjusted for multiple comparisons). The total number of donor exposures per patient while on ECMO was decreased from 14.1 in protocol 1 to 10.0 in protocol 2 to 7.5 in protocol 3 (P < .01). Conclusions. We conclude that the changes in blood bank component selection and management as well as physician practice were effective in substantially reducing ECMO-related transfusion volumes and the resulting donor exposures.

A massive transfusion protocol (MTP) in which most non-RBC transfusions were laboratory result-driven was updated to a 1:1:1 RBC/plasma/platelet formula-driven protocol. Platelet count, fibrinogen level, and prothrombin time (PT) were monitored. In the patients who survived the first 12 hours, the results of coagulation tests were analyzed. Irrespective of the MTP or transfused RBC/plasma ratio, a majority of patients became coagulopathic, usually within the first 2 hours, and a fibrinogen deficiency (fibrinogen level

Limited red blood cell (RBC) regeneration often prevents collection of sufficient blood from autologous donors. We studied the effects of subcutaneous recombinant erythropoietin (rEPO) in subjects making frequent blood donations. Six healthy iron-replete male subjects took rEPO (200 U/kg) subcutaneously daily, and donated blood (450 mL) twice a week for 3 weeks. During a control study, these subjects also attempted twice-weekly blood donations without rEPO. Four other males given rEPO, including one with idiopathic hemochromatosis, waited until day 8 to begin blood donations. All healthy subjects took oral ferrous sulfate. Subcutaneous rEPO given with blood donations resulted in a marked reticulocytosis (mean peak value 568 +/- 159 x 10(9)/L v 235 +/- 77 x 10(9)/L, control study; P < .05), and enhanced RBC production at 28 days (1,208 +/- 227 mL v 719 +/- 161 mL, P < .05). rEPO in advance of blood donations was slightly less effective in normal subjects (941 +/- 139 mL, P < .05); however, the subject with hemochromatosis produced substantially more RBCs (1,764 mL) than any normal subject. rEPO-treated normal subjects (but not the rEPO-treated patient with hemochromatosis or untreated controls) produced iron-deficient RBCs with elevated zinc protoporphyrin levels and low hemoglobin content. These cells appeared within 1 week of rEPO administration and before laboratory confirmation of depleted iron stores. Thus, subcutaneous rEPO is an effective stimulant of erythropoiesis in nonanemic blood donors. However, in addition to eventual depletion of iron stores, early functional iron deficiency affects response to the drug.

Blood donor screening for hepatitis B core antibody and elevation of serum alanine aminotransferase level, surrogate markers of hepatitis C/non-A, non-B (NANB) infection, was implemented in 1986. Reported cases of posttransfusion hepatitis (PTH) from 1985 to 1988 were reviewed to ascertain the effect of surrogate testing on the number and character of cases and to compare any changes with those in the incidence of hepatitis in the general population. The reports of all PTH, NANB PTH, and type B PTH decreased 61.5%, 75.4%, and 84.5%, respectively. The rates of reported cases of NANB hepatitis and hepatitis B in the general community also fell during the period of review. The decrease in PTH and background hepatitis was similar between 1985 and 1986. In 1987, the first complete year of surrogate testing, the incidence of PTH decreased at a greater rate, to levels 50% below what would have been projected on the basis of background changes alone for NANB PTH, and to 35% below projected for type B PTH. There appears to have been a substantial decrease in the risk of both type B PTH and NANB PTH in the northeastern United States between 1985 and 1988 due to a combined effect of donor surrogate testing and a decrease in the background rates of hepatitis in the donor population. ( Arch Intern Med . 1991;151:2445-2448)

The mean and range values for leukocyte content in unmodified red blood cell and platelet components, as well as their leukocyte-reduced counterparts, have been included in multiple published reports. We compiled these data to facilitate comparisons between components and preparations, and to allow predictions of which methods of leukodepletion are likely to be efficacious for preventing leukocyte-mediated transfusion complications as the threshold values become better defined.

A 35-year-old woman developed acute intravascular hemolysis within five days of beginning a course of ceftazidime. The direct antiglobulin test became strongly positive with both anti-IgG and anticomplement. The serum contained an antibody that, in the presence of ceftazidime, sensitized unmodified reagent cells with IgG and complement (immune-complex type). The serum also agglutinated ceftazidime-pretreated cells at room temperature and 37 degrees C (drug-adsorption type). Retrospective testing disclosed that the drug adsorption antibody, which had been present before the current course of antibiotics, was not demonstrable during the hemolysis. The reactivity of the immune complex antibody, which developed by the second day of ceftazidime, paralleled the degree of hemolysis and the strength of the direct antiglobulin test. The authors believe that this patient had two separate ceftazidime-dependent antibodies and that the antibody reactive by immune complex mechanism mediated an episode of acute intravascular hemolysis.

The protease-activated receptor-2 (PAR-2) is present on human airway smooth muscle (ASM) cells and can be activated by mast cell tryptase, trypsin, or an activating peptide (AP). Trypsin induced significant increases in PGE2 release from human ASM cells after 6 and 24 h and also induced cyclooxygenase (COX)-2 mRNA expression and COX-2 protein. Tryptase and the PAR-2 AP did not alter PGE2 release or COX-2 protein levels, suggesting a lack of PAR-2 involvement. When we compared results in asthmatic and nonasthmatic muscle cells, both trypsin and bradykinin induced less PGE2 from asthmatic ASM cells, and bradykinin induced significantly less COX-2 mRNA in asthmatic cells. Significantly less PGE2 was released from proliferating ASM cells from asthmatic patients. In conclusion, trypsin induces PGE2 release and COX-2 in human ASM cells, which is unlikely to be via PAR-2 activation. In addition, ASM cells from asthmatic patients produce significantly less PGE2 and COX-2 compared with nonasthmatic cells. These findings may contribute to the increase in muscle mass evident in asthmatic airways.

ASIC CONCEPTS in platelet transfusion continue to be debated in the absence of reliable scientific observations with which to direct this therapy. The relationship between platelet count and bleeding risk, and the cofactors that affect that relationship, are only superficially understood. ~ Fundamental issues such as indications, component selection, and preparation are controversial. 2,3 What constitutes an optimal platelet dose remains an outstanding question, as does what type of platelet component to use. Recent publications have recommended exactly opposite practices--transfusion of larger numbers of platelets at longer intervals, or support with small, frequent doses--to achieve the best control of thrombocytopenic bleeding. 4-6 It is no wonder that platelet transfusion practices are still based on tradition and experience and are

Purpose A case of childhood acute hemolytic anemia following parvovirus infection provided an hypothesis for the high frequency of Donath-Landsteiner antibodies and inappropriately low reticulocyte counts in this disease. Patients and methods A 3-year-old boy with hematuria and jaundice was found to have autoimmune hemolytic anemia due to a biphasic IgG Donath-Landsteiner antibody. Despite profound anemia (hematocrit 14.5%), the reticulocyte count was low (1.0%) and examination of his normocellular bone marrow showed erythroid hypoplasia. Results A clinical diagnosis 2 weeks earlier of acute parvovirus B19 was serologically confirmed as the associated antecedent infection. Hemolytic anemia resolved with packed red cell transfusion, and intravenous immune globulin and steroid treatment. Conclusions The high-frequency red cell P antigen is both the unusual specificity of Donath-Landsteiner antibody and the viral receptor for parvovirus infection of red cell precursors. We speculate that interaction of the virus with its receptor may change antigenicity such that anti-P autoantibody forms. Parvovirus B19 may be a primary cause of reticulocytopenic postinfectious hemolytic anemia in children.

A closed-system filter-syringe set designed for preparation of prefiltered aliquots of packed red cells for neonatal transfusion was evaluated. In three experiments, filter-syringe sets were sterile-connected, and aliquots were prepared at six intervals during 35-day storage of CPDA-1 packed cells. Hemoglobin (HGB), supernatant potassium (K+), and free hemoglobin (fHGB) in the aliquot and primary storage container were compared. There was no tubing weld failure, filter-occlusion, or bacterial contamination of the units. Hemoglobin remained stable over 35 days of storage, with comparable values in the primary collection container and syringe aliquot. Supernatant potassium in the units increased to an average of 89.5 mEq/L at 35 days, and levels in the aliquots closely matched those in the primary containers at each storage interval (maximum average 91.8 mEq/L). Free hemoglobin also progressively and comparably increased in both the primary containers and syringe aliquots. The gentle negative pressure and turbulence during use of the filter-syringe set apparently causes little hemolysis. The set simplifies preparation of aliquots for neonatal small volume and syringe-pump transfusion at costs comparable to alternative preparation methods. When sterile-connected, the filter-syringe set facilitates "assigned unit" inventory management for neonatal transfusion, which has been shown to be highly effective in limiting donor exposures.

Leukopoor red cell components have a reduced incidence of febrile transfusion reactions. An analogous efficacy for leukopoor platelets has not been convincingly established. The authors analyzed the transfusion records of 36 patients (26 women, 10 men) who received leukopoor platelets following febrile reactions to unmodified platelets. These patients received 409 unmodified transfusions (mean/patient 11) with 84 febrile reactions (rate 20.5%). Reaction rates to unmodified, non-HLA-matched single donor platelets (14.0%) and HLA-matched platelets (6.5%) were significantly lower than to pooled concentrates (27.2%) (P less than .01 and P less than .001, respectively). Of 623 leukopoor transfusions (mean/patient 17), 84 resulted in reactions (13.5%). Although leukodepletion significantly lowered the overall reaction rate as compared with unmodified products (P less than .02), a majority of patients (28 of 36) continued to have reactions. When individual reaction rates to unmodified and leukopoor transfusions were compared, only two patients showed a significant decrease in their reaction rate with leukopoor platelets. It appears that most susceptible patients continue to react to leukopoor platelets, particularly when pooled concentrates are used, and many patients show no reduction in reaction rate.

THE IMPORTANCE of autologous blood to medical practice has fluctuated considerably over the last century. One hundred years ago, autologous blood transfusion practices offered the only hope for an exsanguinating patient. The principies of compatibility had not been discovered, and transfusions of blood from one individual to another appeared foolhardy; but surgeons could collect blood shed after delivery, strain it through eloth, and return it to the patient. In some cases, this salvaged autologous blood was lifesaving. Autologous blood donation in advance of surgery became a natural extension of this practice in the early part of the 20th century, once it was possible to anticoagulate and store blood in glass containers. The primary purpose of autologous blood was to provide a source of blood, at frrst because homologous blood transfusions were risky, and later to augment marginal homologous supplies. As the safety and availability of blood improved, interest in collecting autologous blood waned. Other advantages of autologous blood, such as the elimination of transfusion-transmitted diseases and immune-mediated reactions, were not seen as sufficiently important to warrant the effort inyolved in autologous donations. Sporadic reports conceming the use of autologous blood continued to appear, ineluding a useful monograph reviewing an assortment of applications for autologous blood collections. 1 Nonetheless, for most programs autologous blood was considered primarily for the patient with multiple alloantibodies, or ąntibodies to high-frequency antigens, for whom homologous donor blood was in limited supply. The identification of the human immunodeficiency virus in the 1980s, and the recognition that it was efficiently spread through blood transfusions, led to a rapid renewal of interest in autologous blood techniques. Between 1982 and 1988, American Association of BloOO Banks (AABB) member surveys found that the number of units of autologous blood collected annually increased each year (from 18,737 in 1982 to 337,773 in 1988 [Fig 1]). New advances, such as the use ofrecombinant human erythropoietin, may catalyze additional growth in autologous collections. Such rapid growth over a very short time period

Fifty-nine volunteers were screened, as if for blood donation, with the use of simultaneous fingerstick (FS), earlobe (EL), and venipuncture (VP) samples tested by copper sulfate density and two instruments for rapid assay, Statcrit® and HemoCue®. The “true” hemoglobin was determined on the venipuncture sample by cyanmethemoglobin assay. Eligibility (pass/fail) was established for each site/result with the use of standard and investigational cut-offs. Results were used to predict the performance of two-method protocols beginning with copper sulfate followed by FS or EL using Statcrit or HemoCue. Individually, the method sensitivities for hemoglobin below cut-off were low (12.5–62.5%) with the use of standard cut-offs. Assuming 5% prevalence of low hemoglobin, two-method protocols would inappropriately pass more than half of those with low hemoglobin, whereas the proportion of deferred donors with adequate “true” hemoglobin ranged widely (2.8–72.3%). For some methods, investigational cut-offs achieved improved sensitivity with no effect on specificity. Despite standard hemoglobin “requirements” for blood donation, the ability of the investigated methods to correctly classify donors is poor and varies considerably with method and protocol. With cut-offs as specified, the standards for FS and EL samples are not equivalent; the EL cut-offs are too low and the method is inherently less sensitive. Standardization may be better achieved by specifying both required hemoglobin and minimum performance for screening methods.

Microaggregate blood filters (MAFs) may be beneficial for patients at increased risk of ARDS receiving massive transfusion, although their use is not uniformly recommended. While new leukocyte adsorption filters (LAFs) are more efficient for white cell removal, both MAFs and LAFs successfully reduce the incidence of febrile reactions to red cell components by reducing the residual leukocyte content below 109. It is likely that leukocyte removal delays or prevents recipient HLA alloimmunization from transfusion. LAFs offer a simple means of providing both red cells and platelets, depleted of leukocytes to levels below 108 and may prevent or delay the onset of HLA alloimmunization. A promising area of application of LAFs is for prevention of transfusion-transmitted CMV where leukocyte-depleted components may be noninfectious. Several secondary benefits of leukocyte-depleted components are anticipated, including a decreased risk of HTLV-1 transmission, reduced degree of transfusion-mediated immunosuppression and a lowered risk of graft-vs-host disease.

To evaluate the risk of significant plateletpheresis-induced thrombocytopenia in normal volunteer donors undergoing plateletpheresis less frequently than every 56 days and to help understand factors influencing platelet yield in these donors, pre- and postapheresis platelet counts (X 10(3)/microliter) and platelet yields were analyzed from 916 plateletpheresis procedures (Fenwal CS-3000) on 607 donors (405 men, 202 women). The mean preapheresis platelet count was 265 +/- 59 (SD) (range 155-650) (men 256 +/- 55 [170-444]; women 284 +/- 65 [155-650]; P less than .001, t-test). After an average platelet yield of 4.08 X 10(11), the mean postplatelet count was 185 +/- 46 (range 72-420) (men 184 +/- 42 [80-345]; women 194 +/- 52 [72-500]; P less than 0.1). The percent decrease in platelet count (mean 29.4 +/- 13, range -28.1-65.5) had positive correlations with platelet yield (P less than .01, r = 0.10), predonation platelet count (P less than .001, r = 0.29), and age (P less than .01, r = .10) and a negative correlation with weight (P less than .001, r = 0.18) Both the percent decrease and platelet yield were significantly higher in women than in men (P less than .001, t-test), and platelet yield was best predicted by preapheresis platelet count in both sexes (r = 0.65). The effects of plateletpheresis on donor platelet count thus vary widely among donors. Although a decrease of greater than 50% was not a rare event (5.9% of all procedures), in only three instances (0.3%) was the actual postapheresis platelet count below 100 (72, 94, 95). Thus, plateletpheresis in normal donors appears unlikely to produce clinically significant thrombocytopenia immediately after apheresis.