Improvement in appropriate autologous donations with local education: 1987 to 1989.

Preoperative autologous blood donation for elective surgery patients at university hospitals was underused in the past. More recently, national educational efforts have been made. To test the impact of local surgeon interviews and education, in 1988 the same local educational program was instituted at three university hospitals; three community hospitals were used as controls. Donation by appropriate patients of interviewed surgeons (elective surgery, crossmatch recommended, no contraindications to donation) increased from 24 percent (44/180) to 40 percent (88/222) (p = 0.002) and 15 percent (21/143) to 32 percent (41/127) (p = 0.001) at two university hospitals where the investigator-educators were on site, but not at the three community hospitals. Between 1987 and 1989, donation rates at all six hospitals remained low among patients for whom autologous donation was (probably) less appropriate. Donation rates for type and screen procedures were 3.0 percent (131/4587) in 1987 and 3.0 percent (199/6606) in 1989 (p = 0.67). Donation rates for "no blood order" procedures were 0.2 percent (15/9429) in 1987 and 0.1 percent (9/11,239) in 1989 (p = 0.14). It can be concluded that appropriate autologous blood donations increased at university hospitals where surgeons were individually interviewed and educated by an investigator on site. However, despite this increase, apparently eligible elective surgery patients in 1989 still failed to donate. This situation deserves additional investigation.

Factors associated with successful autologous blood donation for elective surgery.

Not all donors can donate the number of autologous blood units requested by their physicians before surgery, and donors are more frequently unsuccessful as more units are requested. Therefore, 368 autologous blood donors who were requested to donate 4 or more units during the 6-week period before surgery at one community blood center were studied. More men were able to donate 4 units with no deferrals for anemia than were women (86% [181 of 211] compared to 42% [48 of 115], P less than 0.001). Greater success also was observed among donors with an initial hemoglobin level greater than 125 g/L (12.5 g/dL), those with higher weight, and those with more advanced age. Multiple logistic regression analysis showed that only higher initial hemoglobin levels (odds ratio, 3.3 per 10 g/L [1 g/dL] increment) and male sex (odds ratio, 2.7) were independent predictors of successful donation of 4 or more units.

Blood loss and replacement in total hip arthroplasty: a multicenter study. The Preoperative Autologous Blood Donation Study Group.

To determine blood loss, the number of transfusions, and the hemoglobin levels achieved in patients via transfusion in the course of total hip arthroplasty, 324 patient records from 1987 through 1989 were reviewed at three university and three community hospitals. Calculated blood loss was 3.2 +/- 1.3 units in primary procedures and 4.0 +/- 2.1 units in revision procedures (mean +/- SD). Of 777 red cell units transfused, 455 (59%) were autologous units. Transfused patients received 2.0 +/- 1.8 units for primary procedures and 2.9 +/- 2.3 units for revision procedures (mean +/- SD). The maximum number of units given to 95 percent of the transfused patients was 4 for primary procedures and 6 for revision procedures. The mean postoperative hemoglobin level after all transfusions was 103 to 110 g per L, regardless of patient age group of physical status, autologous donor status, or hospital. No difference in length of hospital stay was observed for patients less than 65 years old with hemoglobin concentrations of 80 to 139 g per L at discharge.

Probable reasons that autologous blood was not donated by patients having surgery for which crossmatched blood was ordered.

Preoperative autologous blood donation is used by only a small percentage of surgery patients for whom crossmatched blood is ordered. To document the reasons the patients failed to donate, the medical records of surgical patients at three university and three community hospitals were studied. All procedures for which crossmatched blood was ordered, but for which autologous blood was not available, were included (n = 8121). Probable reasons for nondonation were found in 72 percent of university hospital patients and 65 percent of community hospital patients (n = 6064 and n = 2057, respectively). The most frequent reasons for nondonation among university hospital patients were emergency surgery (27%) and age less than 12 years (17%), and those among community hospital patients were emergency surgery (42%) and American Society of Anesthesiologists physical status greater than or equal to 4 (20%). Surprisingly, anemia (hemoglobin less than 11 g/dL [less than 110 g/L]) as the only limitation to donation was rarely found: this was the sole reason in only 3.3 percent of university hospital and 4.5 percent of community hospital patients. Overall, of 8121 patients who failed to donate autologous blood, 5731 (71%) had legitimate medical reasons. The remaining 2390 (29%) had no identifiable reason for nondonation, and recruitment efforts should be focused on them and their surgeons.

Return of autologous blood donors as homologous blood donors.

Autologous blood donors (ABDs) have been reported to have favorable attitudes toward returning as homologous blood donors (HBDs), but the frequency of return has not been well documented. ABDs eligible by history to be HBDs were followed at one blood center: 255 donating for elective surgery and 234 donating during pregnancy were followed for an average of 18 months and 20 months, respectively, from time of eligibility after surgery or postpartum. Male ABDs had a higher rate of return as HBDs, as 34 percent (21/62) returned to donate an average of 3 units, whereas 13 percent (56/427) of female ABDs returned as HBDs to donate an average of 2 units. Although a history of donation was associated with a higher rate of return (30%, 34/113), 11 percent (43/376) of ABDs with no history as HBDs returned to donate homologous units, despite having been recruited less frequently than prior HBDs. Overall, all male ABDs and female ABDs with an HBD history returned most frequently. The extra effort required for an autologous donor program may result in the recruitment of new donors into the HBD pool.

Lack of increased bleeding after paracentesis and thoracentesis in patients with mild coagulation abnormalities.

To determine whether untreated mild coagulopathy in patients with no evidence of clinical bleeding is associated with an increased risk of hemorrhage after paracentesis or thoracentesis, retrospective examination was conducted of 608 consecutive procedures for which prothrombin time (PT), partial thromboplastin time (PTT), platelet (Plt) counts, and preprocedure and postprocedure hemoglobin concentrations were available. There was no increased bleeding in patients with mild to moderate coagulopathy (defined as PT or PTT up to twice the midpoint normal range or pit count of 50 to 99 x 10(3) per microL [50-99 x 10(9)/L]). However, patients with markedly elevated serum creatinine levels (6.0 to 14.0 mg/dL [530-1240 mumol/L]) had a significantly greater average hemoglobin loss (-0.82 +/- 1.3 g/dL [-8 +/- 13 g/L], n = 11) than patients with normal serum creatinine levels (-0.12 +/- 0.88 g/dL [-1 +/- 9 g/L], n = 450) (p = 0.011). Overall, the frequency of bleeding complications requiring red cell transfusions was very low: 0.2 percent of events. The most common diagnosis for patients who had paracentesis was alcoholic liver disease (72%); for those having thoracentesis, it was infection (37%). It can be concluded that, for these patients, prophylactic plasma or platelet transfusions are not necessary. Patients with markedly elevated serum creatinine deserve close postprocedure observation.

Lack of increased bleeding after liver biopsy in patients with mild hemostatic abnormalities.

Prophylactic transfusions of fresh frozen plasma and platelets are sometimes given to patients with mild elevations in prothrombin time (PT) and partial thromboplastin time (PTT) and mild thrombocytopenia before percutaneous liver biopsy. To determine whether PTs and PTTs 1.1-1.5 times midrange normal levels and platelet counts 50-99 x 10(9)/L are associated with increased bleeding complications, hospital records of all patients who underwent percutaneous liver biopsy during 56 consecutive months (n = 291) were reviewed. Complete information was available for 177 inpatient procedures (155 standard, 22 fine needle). Overall, the frequency of bleeding complications in patients with platelet counts greater than or equal to 50 X 10(9)/L was 3.4% (6 of 175), with no significant difference between patients with mild hemostatic abnormalities and patients with normal parameters. These data suggest that prophylactic transfusions may not be necessary. One factor was highly associated with bleeding complications: a patient diagnosis of malignancy, 14% (7 of 50) compared with 0.8% (1 of 127) among other patients (P less than 0.001). These patients should be monitored closely after biopsy.

Surgeons' knowledge, attitude, and use of preoperative autologous blood donation.

Before a comprehensive educational program on preoperative autologous blood donation was begun, 118 surgeons from three different areas of the country were tested to assess their baseline knowledge and attitude about this practice. Test results were correlated with the percentage of eligible patients that the surgeons actually referred for preoperative donation during a period of observation. The purpose of this preliminary effort was to identify areas in the educational program that required emphasis. Overall, the surgeons' attitude toward preoperative donation was quite favorable, but their depth of knowledge varied. Misunderstandings may have led to diminished use of this service (eg, about 50% didn't realize that many patients with medical conditions or low hematocrits are permitted to donate). However, it is not clear that simply bolstering surgeons' knowledge will increase their appropriate use of preoperative donation. When all 118 surgeons were studied, their knowledge and attitude were unrelated to the percentage of eligible patients referred. However, when 44 surgeons who managed the largest number of eligible patients were analyzed separately, their use of preoperative donation was directly correlated with their knowledge and attitude. The local awareness of AIDS also significantly influenced the use of this service. It is proposed that knowledge of preoperative donation may be important for inducing surgeons to begin referring patients for this service. Once a pattern of successful participation is established, referral seems to increase with the acquisition of working knowledge.

Donation reactions among autologous donors.

Studies of risk factors associated with reactions among autologous blood donors have been limited. Therefore, 2091 autologous and 4737 homologous donations were examined. Donors at greatest risk for reaction were autologous donors who had reactions at first donation; among 45 who made repeat donations for the same surgery, 17 (38%) had repeat reactions. The group least likely to experience reactions were the autologous donors greater than or equal to 66 years old; they experienced a 1.9 greater than or equal to percent (6/310) incidence of reactions. More reactions were seen in both autologous and homologous donors in the categories of first-time donor, female gender, decreasing age, and lower weight. Multiple logistic regression analysis showed that all of these variables were independent predictors of donor reaction, with first-time donation (odds ratio, 2.4) and female gender (odds ratio, 1.9) being the strongest predictors of reaction. Donor room personnel should be alerted that autologous donors who react at first donation are very likely to react at subsequent donations. Contrary to common concern, elderly autologous donors are least likely to have reactions.

Moderate and severe reactions during autologous blood donations are no more frequent than during homologous blood donations.

Because autologous donation is permitted for donors who do not meet homologous blood donation standards, referring physicians and blood center personnel may be concerned about autologous donor reactions. Small studies have determined that mild reactions do not occur more frequently, but the incidence of rarer, more serious, moderate and severe reactions is unknown. We therefore studied the frequency of reactions during 10,200 autologous and 219,307 concurrent homologous donations at four blood centers. No significant difference was seen for severe reactions: autologous 0.039% (4/10,200), homologous 0.037% (82/219,307) (p = 0.79); moderate reactions: autologous 0.19% (19/10,200), homologous 0.22% (473/219,307) (p = 0.60) or mild reactions: autologous 2.26% (231/10,200), homologous 2.26% (4946/219,307) (p = 0.98). We conclude moderate and severe donation reactions do not occur more frequently among autologous donors who are preselected by referring physicians and screened by blood center personnel.

Safety and use of autologous blood donation during the third trimester of pregnancy.

To examine autologous blood donation during the third trimester of pregnancy, records of 272 blood donors who donated by standard procedures were reviewed. The incidence of vasovagal reactions at autologous donation was 2.1% (7/341) versus 1.1% (1370/121,539) at homologous donation (p = 0.17). Follow-up was available for 199 of these donors and 10 additional donors with fetal monitoring. Infant morbidity and mortality rates were not higher than expected; 0.5% (1/214) was stillborn. Although 11% (24/209) of deliveries was associated with autologous transfusion, only 4.3% (9/209) would have had a hemoglobin level less than 8 gm/dl without transfusion. Patients with placenta previa had the highest incidence of transfusion, 83% (5/6). Transfusion incidence was higher for women donating within 1 week of delivery, 21% (8/38), versus 9.4% (16/171) for women donating earlier in pregnancy (p = 0.044). We recommend that donation be completed by 2 weeks before the estimated date of confinement to allow for recovery of red cell mass.

Evaluation of the indirect immunofluorescence assay as a confirmatory test for detecting antibodies to the human immunodeficiency virus.

One hundred and eighty-four serum specimens were assayed for antibodies to the human immunodeficiency virus. All specimens were screened with a commercial enzyme immunoassay and confirmed by two indirect immunofluorescence assays. Sera were also assayed by Western blot. Results from sera of 48 healthy heterosexual volunteers were all negative by EIA, IFA, and Western blot. Sera from 50 healthy homosexual men negative by EIA were also negative by IFA and Western blot. Sixty-two patients with persistent generalized lymphadenopathy or newly diagnosed AIDS all were positive by EIA, IFA, and Western blot. Of 24 sera from patients with autoantibodies, with no evidence of AIDS-related diseases, five appeared to be false-positive by EIA, since they were nonreactive by IFA and Western blot. In addition, three other samples contained both autoantibodies and human immunodeficiency virus antibodies. False-positive results were observed in both the EIA and IFA with monoclonal antibodies directed toward the MHC class II antigens DQ and DR. The reactivity of these antibodies could not be distinguished from positive patients' sera, in either EIA or IFA. We conclude that in general indirect immunofluorescence performed well as a confirmatory test after screening by enzyme immunoassay for human immunodeficiency virus antibodies.