Malaria seroprevalence in blood bank donors from endemic and non-endemic areas of Venezuela

In Venezuela, a total of 363,466 malaria cases were reported between 1999-2009. Several states are experiencing malaria epidemics, increasing the risk of vector and possibly transfusion transmission. We investigated the risk of transfusion transmission in blood banks from endemic and non-endemic areas of Venezuela by examining blood donations for evidence of malaria infection. For this, commercial kits were used to detect both malaria-specific antibodies (all species) and malaria antigen (Plasmodium falciparum only) in samples from Venezuelan blood donors (n = 762). All samples were further studied by microscopy and polymerase chain reaction (PCR). The antibody results showed that P. falciparum-infected patients had a lower sample/cut-off ratio than Plasmodium vivax-infected patients. Conversely, a higher ratio for antigen was observed among all P. falciparum-infected individuals. Sensitivity and specificity were higher for malarial antigens (100 and 99.8%) than for antibodies (82.2 and 97.4%). Antibody-positive donors were observed in Caracas, Ciudad Bolívar, Puerto Ayacucho and Cumaná, with prevalences of 1.02, 1.60, 3.23 and 3.63%, respectively. No PCR-positive samples were observed among the donors. However, our results show significant levels of seropositivity in blood donors, suggesting that more effective measures are required to ensure that transfusion transmission does not occur.     

Epidemiological aspects of hepatitis C virus infection among renal transplant recipients in Central Brazil

An investigation was conducted involving 255 renal transplant recipients in the state of Goiás, Central Brazil, to determine the prevalence of hepatitis C virus (HCV), its risk factors, the genotypes involved, and the level of alanine aminotransferase (ALT) present in the patients. All serum samples were tested for anti-HCV antibodies and HCV RNA. Forty-one patients were anti-HCV and/or HCV RNA positive, resulting in an overall HCV infection prevalence of 16.1% (95% CI: 11.9-21.3). A multivariate analysis of risk factors showed that a history of blood transfusions without anti-HCV screening, the length of time spent on hemodialysis, and renal transplantation before 1994 are all associated with HCV positivity. In HCV-positive patients, only 12.2% had ALT levels above normal. Twenty-eight samples were genotyped as genotype 1, subtypes 1a (62.5%) and 1b (31.3%), and two samples (6.2%) were genotype 3, subtype 3a. These data show a high prevalence of HCV infection and low ALT levels in the studied population. The risk factor analysis findings emphasize the importance of public health strategies such as anti-HCV screening of candidate blood and organ donors, in addition to the stricter adoption of hemodialysis-specific infection control measures. The present study also demonstrates that HCV genotype 1 (subtype 1a) is predominant in this population.     

Post-transfusion hepatitis. Sudden drop of its incidence associated with hyperbilirubinemic donors disqualification

The lacking impact of various precautions on the incidence of post-transfusion hepatitis (PTH) in the last 15 years is presented. In 1984, however, PTH dropped from 0.12 to 0.04%. This PTH drop did not coincide with the disqualification of HBsAg carriers, nor with that of high ALT, nor with the introduction of complete voluntary donorship in 1982, but with the systematic exclusion of hyperbilirubinemic donors. Circulating immune complexes (CIC) were found in 28.9% of PTH compromised donors, and CIC are therefore supposed to be a marker of PTH risk even if HBsAg was not demonstrable. Re-examining donors compromised in 120 PTH found 12.4% HBsAg carriers who had not been detected by CIEP prior to the transfusions. Accumulation of more than one HBsAg positive transfusion in 56.5% of the PTH patients suggests a cumulative effect. The inability of specific tests to identify sources of infection in 50% of PTH suggests that doubts must be raised as to their post-transfusional origin. This thesis is also supported by the relative increase of this group after the considerable drop of PTH. The importance of viral sources non-B and B undetectable by sensitive specific methods must be emphasized. The latter is in accordance with the observed PTH drop due to a non-specific marker effect. The bilirubin level screening in each donor before stored blood collection is recommended by means of the described extended AST screening test and, besides HBsAg screening, this is supposed to be the most effective precaution for preventing PTH.     

Transfusion-transmitted infectious diseases

A spectrum of blood-borne infectious agents is transmitted through transfusion of infected blood donated by apparently healthy and asymptomatic blood donors. The diversity of infectious agents includes hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency viruses (HIV-1/2), human T-cell lymphotropic viruses (HTLV-I/II), Cytomegalovirus (CMV), Parvovirus B19, West Nile Virus (WNV), Dengue virus, trypanosomiasis, malaria, and variant CJD. Several strategies are implemented to reduce the risk of transmitting these infectious agents by donor exclusion for clinical history of risk factors, screening for the serological markers of infections, and nucleic acid testing (NAT) by viral gene amplification for direct and sensitive detection of the known infectious agents. Consequently, transfusions are safer now than ever before and we have learnt how to mitigate risks of emerging infectious diseases such as West Nile, Chikungunya, and Dengue viruses.     

Prevention of transfusional Trypanosoma cruzi infection in Latin America

Trypanosoma cruzi is a protozoan infection widely spread in Latin America, from Mexico in the north to Argentina and Chile in the south. The second most important way of acquiring the infection is by blood transfusion. Even if most countries of Latin America have law/decree/norms, that make mandatory the screening of blood donors for infectious diseases, including T. cruzi (El Salvador and Nicaragua do not have laws on the subject), there is usually no enforcement or it is very lax. Analysis of published serologic surveys of T. cruzi antibodies in blood donors done in 1993, indicating the number of donors and screening coverage for T. cruzi in ten countries of Central and South America indicated that the probability of receiving a potentially infected transfusion unit in each country varied from 1,096 per 10,000 transfusions in Bolivia, the highest, to 13.02 or 13.86 per 10,000 transfusions in Honduras and Venezuela respectively, where screening coverage was 100%. On the other hand the probability of transmitting a T. cruzi infected unit was 219/10,000 in Bolivia, 24/10,000 in Colombia, 17/10,000 in El Salvador, and around 2-12/10,000 for the seven other countries. Infectivity risks defined as the likelihood of being infected when receiving an infected transfusion unit were assumed to be 20% for T. cruzi. Based on this, estimates of the absolute number of infections induced by transfusion indicated that they were 832, 236, and 875 in Bolivia, Chile and Colombia respectively. In all the other countries varied from seven in Honduras to 85 in El Salvador. Since 1993, the situation has improved. At that time only Honduras and Venezuela screened 100% of donors, while seven countries, Argentina, Colombia, El Salvador, Honduras, Paraguay, Uruguay and Venezuela, did the same in 1996. In Central America, without information from Guatemala, the screening of donors for T. cruzi prevented the transfusion of 1,481 infected units and the potential infection of 300 individuals in 1996. In the same year, in seven countries of South America, the screening prevented the transfusion of 36,017 infected units and 7, 201 potential cases of transfusional infection.     

Frequencies of ABO, MNSs, and Duffy phenotypes among blood donors and malaria patients from four Brazilian Amazon areas

We compared the serological phenotypic frequencies of ABO, MNSs, and Duffy in 417 blood donors and 309 malaria patients from four Brazilian Amazon areas. Our results suggest no correlation between ABO phenotype and malaria infection in all areas studied. We observed significant correlation between the S +s +, S +s -, and S -s + phenotypes and malaria infection in three areas. Some of the Duffy phenotypes showed significant correlation between donors and malaria patients in different areas. These data are an additional contribution to the establishment of differential host susceptibility to malaria.     

Hepatitis C prevalence and risk factors in hemodialysis patients in Central Brazil: a survey by polymerase chain reaction and serological methods

An hemodialysis population in Central Brazil was screened by polymerase chain reaction (PCR) and serological methods to assess the prevalence of hepatitis C virus (HCV) infection and to investigate associated risk factors. All hemodialysis patients (n=428) were interviewed in eight dialysis units in Goiania city. Blood samples were collected and serum samples screened for anti-HCV antibodies by an enzyme-linked immunosorbent assay (ELISA). Positive samples were retested for confirmation with a line immunoassay (LIA). All samples were also tested for HCV RNA by the PCR. An overall prevalence of 46.7% (CI 95%: 42-51.5) was found, ranging from 20.7% (CI 95%: 8.8-38.1) to 90.4% (CI 95%: 79.9-96.4) depending on the dialysis unit. Of the 428 patients, 185 were found to be seropositive by ELISA, and 167 were confirmed positive by LIA, resulting in an anti-HCV prevalence of 39%. A total of 131 patients were HCV RNA-positive. HCV viremia was present in 63.5% of the anti-HCV-positive patients and in 10.3% of the anti-HCV-negative patients. Univariate analysis of risk factors showed that the number of previous blood transfusions, transfusion of blood before mandatory screening for anti-HCV, length of time on hemodialysis, and treatment in multiple units were associated with HCV positivity. However, multivariate analysis revealed that blood transfusion before screening for anti-HCV and length of time on hemodialysis were significantly associated with HCV infection in this population. These data suggest that nosocomial transmission may play a role in the spread of HCV in the dialysis units studied. In addition to anti-HCV screening, HCV RNA detection is necessary for the diagnosis of HCV infection in hemodialysis patients.     

Risk of HIV infection from transfusion with blood negative for HIV antibody in a west African city.

OBJECTIVE--To estimate the risk of infection with HIV (HIV 1 or HIV 2, or both) from transfusion of a screened unit of blood in a high prevalence area in west Africa. DESIGN--Retrospective cohort study for January-July 1991. SETTING--National Blood Transfusion Centre, Abidjan, Côte d''Ivoire. SUBJECTS--Repeat donors (5831 units of blood) and first time donors (5076 units) in the first five months of 1991. MAIN OUTCOME MEASURES--Prevalence and estimated incidence of HIV infection in repeat and first time donors; estimated rate of potentially infected, HIV antibody negative units; and rate of (false negative) potentially infected units assuming a laboratory test sensitivity of 99%. RESULTS--Overall HIV prevalence was 11.0% in first time donors and 2.1% in repeat donors. In the first seven months of 1991, 29 HIV antibody positive (27 HIV 1, 1 HIV 2, 1 dually reactive) donors with a seronegative unit of blood earlier in the year were identified; 26 had donated blood eight weeks or less before their estimated dates of seroconversion and may have been infectious (minimum rate 26/5831 (4.5/1000 potentially infected units)). Estimated incidence of infection in repeat donors was 1.2-2.5%. Laboratory test insensitivity would result in an estimated 1.1/1000 false negative units from first time donors and 0.2/1000 units from regular donors. The overall rate of potentially infected units (all donors, seroconversions, and errors) was estimated at 5.4-10.6/1000. CONCLUSIONS--The risk of HIV infection from a single unit of blood remains substantial (5.4-10.6/1000 units). To prevent infection from blood transfusion in areas of high incidence and prevalence of HIV all but absolutely essential transfusions should be avoided, and donors with low incidence of HIV infection should be selected.     

Seroprevalence and Incidence of hepatitis E in Blood Donors in Upper Austria

Background

In recent years various studies showed, that hepatitis E virus (HEV) is a growing public health problem in many developed countries. Therefore, HEV infections might bear a transmission risk by blood transfusions. The clinical relevance still requires further investigations. The aim of this study was to provide an overview of acute HEV infections in Upper Austrian blood donors as well as a risk estimation of this transfusion-related infection.

Methods and Findings

A total of 58,915 blood donors were tested for HEV RNA using a commercial HEV RT-PCR Kit. 7 of these donors (0.01%) were PCR-positive with normal laboratory parameters in absence of clinical signs of hepatitis. Viral load determined by quantitative real-time PCR showed a HEV nucleic acid concentration of 2,217 293,635 IU/ml. At follow-up testing (2–11 weeks after donation) all blood donors had negative HEV RNA results. Additionally, genotyping was performed by amplification and sequencing of the ORF1 or ORF2 region of the HEV genome. All HEV RNA positive donor samples revealed a genotype 3 isolate. For the antibody screening, anti-HEV IgM and IgG were detected by ELISA. Follow up serological testing revealed that no donor was seropositive for HEV IgM or IgG antibodies at time of donation. Moreover, we verified the prevalence of anti-HEV IgG in 1,203 of the HEV RNA negative tested blood donors. Overall 13.55% showed positive results for anti-HEV IgG.

Conclusions

In the presented study, we investigated HEV infections in blood donations of Upper Austria over 1 year. We concluded that 1 out of 8,416 blood donations is HEV RNA positive. Seroprevalence of anti HEV IgG results in an age-related increase of 13.55%. Therefore, based on this data, we recommend HEV-PCR screening to prevent transmission of hepatitis E virus by transfusion.     

Prevalence of hepatitis A and B infections in multiply transfused thalassaemic patients.

Evidence of hepatitis B virus (HBV) and hepatitis A virus (HAV) infections was south in 148 multiply transfused patients with thalassaemia and in healthy controls (2040 for HBV and 217 for HAV). The prevalence of the HBV surface antigen or antibody to it was significantly higher in patients than in controls and increased with the number of blood transfusions. In contrast, the prevalence of antibody to HAV was significantly lower in patients than in controls and decreased with the number of blood transfusions. These results support the view that blood transfusion does not play any appreciable part in transmitting HAV. Indeed, regular blood transfusion, where donors almost all have HAV antibody, seems to give protection against infection.     

Evaluation of a malaria antibody enzyme immunoassay for use in blood screening

Transfusion-transmitted malaria is rare, but it may produce severe problem in the safety of blood transfusion due to the lack of reliable procedure to evaluate donors potentially exposed to malaria. Here, we evaluated a new enzyme-linked immunosorbent assay malaria antibody test (ELISA malaria antibody test, DiaMed, Switzerland) to detect antibodies to Plasmodium vivax (the indigenous malaria) in the blood samples in the Republic of Korea (ROK). Blood samples of four groups were obtained and analyzed; 100 samples from P.vivax infected patients, 35 from recovery patients, 366 from normal healthy individuals, and 325 from domestic travelers of non-endemic areas residents to risky areas of ROK. P.vivax antibody levels by ELISA were then compared to the results from microscopic examination and polymerase chain reaction (PCR) test. As a result, the ELISA malaria antibody test had a clinical sensitivity of 53.0% and a clinical specificity of 94.0% for P.vivax. Twenty out of 325 domestic travelers (6.2%) were reactive and 28 cases (8.6%) were doubtful. Of the reactive and doubtful cases, only two were confirmed as acute malaria by both microscopy and PCR test. Thus we found that the ELISA malaria antibody test was insufficiently sensitive for blood screening of P.vivax in ROK.     

Success of a program of routine prenatal screening for hepatitis B surface antigen: the first 2 years.

Prenatal screening for hepatitis B surface antigen (HBsAg) restricted to women with defined risk factors for chronic hepatitis B virus (HBV) infection fails to identify many carriers. A centralized program of routine HBsAg screening for all pregnant women in Alberta was introduced in 1985. We collected and analysed data for the first 2 years of the program in Edmonton to determine the frequency of risk factors for HBsAg positivity, the proportion of multiparous HBsAg-positive women not identified in previous pregnancies, the efficiency and cost-effectiveness of providing immunoprophylaxis to infants at risk of HBV infection and the degree of success in inducing adequate protection. A total of 149 women (158 pregnancies) were found to be HBsAg positive. Risk factors were readily ascertainable for 85% of the women; the remaining 15% would not have been identified through risk-selective screening. The most common risk factors were Oriental ethnic origin, history of hepatitis, jaundice or multiple transfusions of blood or blood products, and occupational exposure to blood. Although 86% of the multiparous HBsAg-positive women had risk factors, only 7% had been identified in previous pregnancies. The Alberta program appears to be cost-effective. We conclude that only routine prenatal screening will identify all infants at risk of perinatal HBV infection and that a comprehensive public health program involving central laboratories, private physicians and public health staff can be highly effective and efficient in protecting infants against hepatitis B.     

Autologous versus allogeneic transfusion: patients' perceptions and experiences

BACKGROUND: Preoperative autologous donation is one way to decrease a patient''s exposure to allogeneic blood transfusion. This study was designed to determine patients'' perceptions about the autologous blood donation process and their experiences with transfusion. METHODS: To assess patient perception, a questionnaire was administered a few days before surgery to patients undergoing elective cardiac and orthopedic surgery in a Canadian teaching hospital. All patients attending the preoperative autologous donation clinic during a 10-month period were eligible. A convenience sample of patients undergoing the same types of surgery who had not predonated blood were selected from preadmission clinics. Patient charts were reviewed retrospectively to assess actual transfusion practice in all cases. RESULTS: A total of 80 patients underwent cardiac surgery (40 autologous donors, 40 nondonors) and 73 underwent orthopedic surgery (38 autologous donors, 35 nondonors). Of the autologous donors, 75 (96%) attended all scheduled donation appointments, 73 (93%) said that they were "very likely" or "likely" to predonate again, and 75 (96%) said that they would recommend autologous donation to others. There was little difference in preoperative symptoms between the autologous donors and the nondonors, although the former were more likely than the latter to report that their overall health had remained the same during the month before surgery (30 [75%] v. 21 [52%] for the cardiac surgery patients and 30 [79%] v. 18 [51%] for the orthopedic surgery patients). When the autologous donors were asked what they felt their chances would have been of receiving at least one allogeneic blood transfusion had they not predonated, the median response was 80%. When they were asked what their chances were after predonating their own blood, the median response was 0%. The autologous donors were significantly less likely to receive allogeneic blood transfusions (6 [15%] for cardiac surgery and 3 [8%] for orthopedic surgery) than were the nondonors (14 [35%] for cardiac surgery and 16 [46%] for orthopaedic surgery). They were, however, more likely to receive any transfusion (autologous or allogeneic) than were the nondonors (25 [63%] v. 14 [35%] for cardiac surgery and 31 [81%] v. 16 [46%] for orthopedic surgery). INTERPRETATION: Patients who underwent preoperative autologous blood donation were positive about the experience and did not report more symptoms than patients who did not donate blood preoperatively. Autologous donors overestimated their chances of receiving allogeneic blood transfusions had they not predonated and underestimated their chances after they had predonated. They were less likely to receive allogeneic transfusions, but more likely to receive any type of transfusion, than were patients who did not predonate.     

Genotyping of the Duffy Blood Group among Plasmodium knowlesi-Infected Patients in Malaysia

The Duffy blood group is of major interest in clinical medicine as it plays an important role in Plasmodium knowlesi and Plasmodium vivax infection. In the present study, the distribution of Duffy blood group genotypes and allelic frequencies among P. knowlesi infected patients as well as healthy individuals in Peninsular Malaysia were determined. The blood group of 60 healthy blood donors and 51 P. knowlesi malaria patients were genotyped using allele specific polymerase chain reaction (ASP-PCR). The data was analyzed using Fisher''s exact test in order to assess the significance of the variables. Our results show a high proportion of the FY*A/FY*A genotype (>85% for both groups) and a high frequency of the FY*A allele (>90% for both groups). The FY*A/FY*A genotype was the most predominant genotype in both infected and healthy blood samples. The genotype frequency did not differ significantly between the donor blood and the malaria patient groups. Also, there was no significant correlation between susceptibility to P. knowlesi infection with any Duffy blood genotype.     

Serological detection of Plasmodium vivax malaria using recombinant proteins corresponding to the 19-kDa C-terminal region of the merozoite surface protein-1

BACKGROUND: Serological tests to detect antibodies specific to Plasmodium vivax could be a valuable tool for epidemiological studies, for screening blood donors in areas where the malaria is not endemic and for diagnosis of infected individuals. Because P. vivax cannot be easily obtained in vitro, ELISA assays using total or semi-purified antigens are rarely used. Based on this limitation, we tested whether recombinant proteins representing the 19 kDa C-terminal region of the merozoite surface protein-1 of P. vivax (MSP119) could be useful for serological detection of malaria infection. METHODS: Three purified recombinant proteins produced in Escherichia coli (GST-MSP119, His6-MSP119 and His6-MSP119-PADRE) and one in Pichia pastoris (yMSP119-PADRE) were compared for their ability to bind to IgG antibodies of individuals with patent P. vivax infection. The method was tested with 200 serum samples collected from individuals living in the north of Brazil in areas endemic for malaria, 53 serum samples from individuals exposed to Plasmodium falciparum infection and 177 serum samples from individuals never exposed to malaria. RESULTS: Overall, the sensitivity of the ELISA assessed with sera from naturally infected individuals was 95%. The proportion of serum samples that reacted with recombinant proteins GST-MSP119, His6-MSP119, His6-MSP119-PADRE and yMSP119-PADRE was 90%, 93.5%, 93.5% and 93.5%, respectively. The specificity values of the ELISA determined with sera from healthy individuals and from individuals with other infectious diseases were 98.3% (GST-MSP119), 97.7% (His6-MSP119 and His6-MSP119-PADRE) or 100% (yMSP119-PADRE). CONCLUSIONS: Our study demonstrated that for the Brazilian population, an ELISA using a recombinant protein of the MSP119 can be used as the basis for the development of a valuable serological assay for the detection of P. vivax malaria.     

Physiopathology of cytomegalovirus infections

Since more than fifteen years ago, the occurrence of Cytomegalovirus (CMV) infections in transfused patients has incriminated blood as a potential vehicle for CMV transmission, resulting in post-transfusion mononucleosis (PTM). Although only Diosi and al. (1969) reported viraemia in two of 35 asymptomatic blood donors, CMV was not recovered from any leukocyte specimen of numerous healthy blood bank donors, assayed by cocultivation of undisrupted buffy coat with a virus sensitive monolayer of human fibroblasts. Nevertheless numerous experimental studies have shown that CMV might persist in very low titer or in a latent state in mononuclear leukocytes in monocytes and macrophages and in lymphocytes. A better knowledge of the site an mechanisms of latency should help to control CMV transmission by fresh and bank blood or transfusion of viable leukocytes.     

Prevalence of hepatitis-B surface antigen among blood donors and human immunodeficiency virus-infected patients in Jos, Nigeria

Information is very scarce on the prevalence of hepatitis-B virus (HBV) infection among blood donors and patients with human immunodeficiency virus (HIV) infection in Nigeria. Hepatitis-B surface antigen (HBsAg) ELISA was used to determined the prevalence of HBsAg among 175 blood donors (aged 20-40 years) and 490 HIV-infected patients (aged 17-60 years) in Jos, Nigeria. Twenty-five (14.3%) of the blood donors and 127 (25.9%) of the HIV-infected individuals were HBsAg seropositive, indicating a higher HBV infection among HIV-infected persons than among healthy blood donors. A slightly higher HBsAg seroprevalence was recorded in the males (14.6%) than females (12.9%) of the blood donors. Among the HIV-infected patients, the males had considerably higher HBsAg seroprevalence than the females (31.8 vs 22.1%) with the highest prevalence of HBsAg occurring in the 51-60 years age group (44%), followed by those of 31-40 years (28.2%). Results confirmed the high endemicity of HBV infection in Jos, Nigeria and the significantly greater prevalence of HBV infection among HIV-infected patients than among blood donors.     

The serologic screening for celiac disease in the general population (blood donors) and in some high-risk groups of adults (patients with autoimmune diseases,osteoporosis and infertility) in the Czech republic

The prevalence of celiac disease (CD) was determined in healthy blood donors and in high-risk groups of adults (a total of 1835 adults—randomly selected 1312 healthy blood donors, 102 patients with primary osteoporosis, 58 patients with autoimmune diseases and 365 infertile women). It was calculated on the basis of a two-step serologic screening method—in the first step IgA and IgG antigliadin antibodies (AGA) and IgA anti-γ-glutamyltransferase (‘transglutaminase’) antibodies (ATG) were estimated, in the second step sera positive for IgA AGA and/or IgA ATG were examined for antiendomysial IgA (AEA) antibodies. Immunoenzymic assay (ELISA) was used for determining of AGA and ATG antibodies; immunoflurescence method, performed on human umbilical cord tissue, was used for assaying of AEA antibodies. Total serum IgA level in only IgG AGA positive subjects was measured by routine turbidimetric method. 0.45% of healthy blood donors, 0.98% of osteoporotic patients, 2.7% of patients suffering from autoimmune disease and 1.13% of women with infertility considered as immunologically mediated were found to be positive in both steps of serologic screening (AGA and/or ATG and antiendomysium positive). The presumed high prevalence of seropositivity for CD in apparently healthy Czech adult population was confirmed. In the high-risk groups, the prevalence of seropositivity for CD was approximately 2–4 times higher than in healthy blood donors. The real prevalence of CD in the tested groups, however, can be estimated after performing small intestinal biopsy in the seropositive patients.     

Effect of a flow chart on use of blood transfusions in primary total hip and knee replacement: prospective before and after study

Problem A suspected high proportion of unnecessary blood transfusions occur in patients undergoing total joint replacement.Design Prospective before and after study evaluating the impact of a one page flow chart aimed at reducing the use of blood transfusions.Setting Orthopaedic tertiary care centre in Winterthur, Switzerland. 208 patients underwent primary total joint replacement of hips or knees during the control period (October 1998 to September 1999) and 217 during the intervention period (October 1999 to September 2000).Key measures for improvement Proportion of patients receiving allogeneic blood transfusions.Strategies for change A simple one page flow chart, which summarised graphically the perioperative decision pathways for anaemic patients, was placed in all charts of patients undergoing total joint replacement and handed out to medical staff from 4 October 1999 onwards. The implementation of the flow chart focused on its endorsement by chief physicians and the development of a sense of “ownership” among physicians and nurses.Effects of change The proportion of patients receiving allogeneic blood decreased from 35.0% to 19.8% (absolute difference -15.2%, 95% confidence interval -23.3 to -7.0%). The percentage of patients donating and receiving autologous blood also decreased. This led to overall savings of about £23 000 ($42 470; €34 441) (£103.50 per patient undergoing total joint replacement). Differences became more pronounced after adjustment for confounding factors.Lessons learnt Allogeneic blood transfusions in primary hip and knee replacement surgery may be reduced cost effectively by implementing a one page flow chart. Five key elements may have contributed: simplicity; wide distribution; no requirement for major changes; endorsement by local opinion leaders; and development of a sense of ownership. These elements may be used in other contexts to achieve sustained change of clinical practice.