Estimating the transfusion transmission risk of Q fever.

BACKGROUND: The Q fever outbreaks in the Netherlands in 2007 to 2009 initiated discussion on the necessity of measures to prevent transmission through blood products. Risk assessments help transfusion regulators decide when and where measures are required. This study assesses the transfusion transmission (TT) risk of Q fever using the European Up-Front Risk Assessment Tool (EUFRAT) model. STUDY DESIGN AND METHODS: We estimated the number of Q fever infections in recipients during the 2007 to 2009 outbreaks' peaks using selected notification data; estimates are calculated from the probability of a donor being infected. We compared this probability to the prevalence of infection estimated from an independent donation testing study and using the Biggerstaff model. We also quantified the risk reduction by implementing measures such as donation testing and donor deferral. RESULTS: At the peak of the 2007, 2008, and 2009 outbreaks, there were an estimated 0.21, 0.96, and 1.59 recipients infected with Q fever, respectively. Between June 1, 2009, and January 31, 2010, the probability of a donor being infected with Q fever in the high-incidence areas was estimated at 260 (95% confidence interval, 192-340) per 100,000 donors, consistent with results from the donation testing study. The EUFRAT estimates were also consistent with estimates from the Biggerstaff model. Scenario analyses showed that donation testing provided the largest risk reduction of various risk reduction strategies. CONCLUSION: The TT risk of Q fever during the 2007 to 2009 outbreaks was small, a result that is consistent with results of other studies. EUFRAT can be applied successfully to support decision making during outbreaks.

Modeling the transmission risk of emerging infectious diseases through blood transfusion.

BACKGROUND: A timely risk assessment is desired to guide decisions on preventive transfusion safety measures during emerging infectious disease (EID) outbreaks. The European Up-Front Risk Assessment Tool (EUFRAT) model was developed to provide quantitative transmission risk estimates of EIDs through blood transfusion. STUDY DESIGN AND METHODS: The generic model comprises five sequential steps to estimate the infection risks in the blood transfusion chain: 1) the prevalence of infection in the donor population, 2) the risk of obtaining infected donations, 3) infected components, 4) infected blood products, and 5) the risk of transmitting the infection to recipients. The model uses inputs from epidemiologic characteristics of an EID and transfusion practice. The model was applied to data from a recent chikungunya outbreak in Italy. RESULTS: Based on data from the outbreak peak, an estimated prevalence of 1.07 (95% confidence interval [CI], 0.38-2.03) per 100,000 donors would lead to 0.04 infected donations (95% CI, 0.01-0.10), 0.13 infected blood components, 0.13 infected end products, and 0.0001 severe infections in recipients. This estimated risk can be reduced by increasing the duration of quarantine of the donated blood and becomes zero after 7 or more days of quarantine. The model also estimated the probability of a donor returning from the outbreak area and subsequently donating infected blood in his home country to be 0.30 (95% CI, 0.01-0.65) per 100,000. CONCLUSION: The model can be used to quantify EID outbreak risks to blood transfusion recipients and the effect of targeted safety interventions and as such support public health decision-making.

Cost-effectiveness of additional blood screening tests in the Netherlands.

BACKGROUND: During the past decade, blood screening tests such as triplex nucleic acid amplification testing (NAT) and human T-cell lymphotropic virus type I or I (HTLV-I/II) antibody testing were added to existing serologic testing for hepatitis B virus (HBV), human immunodeficiency virus (HIV), and hepatitis C virus (HCV). In some low-prevalence regions these additional tests yielded disputable benefits that can be valuated by cost-effectiveness analyses (CEAs). CEAs are used to support decision making on implementation of medical technology. We present CEAs of selected additional screening tests that are not uniformly implemented in the EU. STUDY DESIGN AND METHODS: Cost-effectiveness was analyzed of: 1) HBV, HCV, and HIV triplex NAT in addition to serologic testing; 2) HTLV-I/II antibody test for all donors, for first-time donors only, and for pediatric recipients only; and 3) hepatitis A virus (HAV) for all donations. Disease progression of the studied viral infections was described in five Markov models. RESULTS: In the Netherlands, the incremental cost-effectiveness ratio (ICER) of triplex NAT is €5.20 million per quality-adjusted life-year (QALY) for testing minipools of six donation samples and €4.65 million/QALY for individual donation testing. The ICER for anti-HTLV-I/II is €45.2 million/QALY if testing all donations, €2.23 million/QALY if testing new donors only, and €27.0 million/QALY if testing blood products for pediatric patients only. The ICER of HAV NAT is €18.6 million/QALY. CONCLUSION: The resulting ICERs are very high, especially when compared to other health care interventions. Nevertheless, these screening tests are implemented in the Netherlands and elsewhere. Policy makers should reflect more explicit on the acceptability of costs and effects whenever additional blood screening tests are implemented.

Demographic changes and predicting blood supply and demand in the Netherlands.

BACKGROUND: Concerns have been raised that aging of the general population will increase the demand for blood products. Modeling can be applied to assess trends in blood demand and supply and predict how these will develop over time. STUDY DESIGN AND METHODS: We developed mathematical models to describe and predict the national demand of red blood cell (RBC) products. The first demand model assumes that the mean numbers of transfusions per inhabitant per age and sex are constant. A second demand model incorporates observed changes in clinical blood use over time. Further, a donation model is developed to predict future RBC supply. To estimate the supply of whole blood donations, we used annual donor retention rates, donor recruitment rates, and mean numbers of donations per donor year. RESULTS: The model based on demography only predicts an increase of 23% in RBC demand over 2008 to 2015. The second model, incorporating both demographic changes and trends in clinical RBC use, predicts a decrease of RBC demand by 8% over the same period. The predicted RBC supply closely follows the demand as predicted by the second model. CONCLUSIONS: Despite an aging population, RBC demand may not increase as much as predicted in other studies. This depends on the extent to which other effects, like that of optimal blood use, will neutralize the effects of aging of the transfusion recipient population. Still, the observed downward trend in donor recruitment in the Netherlands must be stopped to maintain a sufficient RBC supply.

A probabilistic model for analyzing viral risks of plasma-derived medicinal products.

BACKGROUND: The prevention of transmission of viral infections by plasma-derived medicinal products is of concern to manufacturers, legislators, and patient representative groups. Recent European legislation requires a viral risk assessment for all new marketing applications of such products. STUDY DESIGN AND METHODS: A discrete event Monte Carlo model was developed to determine the viral transmission risks of the plasma-derived medicinal products. The model incorporates donor epidemiology, donation intervals, efficiency of screening tests for viral markers, inventory hold period, size and composition of the manufacturing pool, production time, process virus reduction capacity, and product yield. With the model, the human immunodeficiency virus (HIV) and hepatitis C virus (HCV) contamination risks of a typical hypothetical plasma product were calculated, and the sensitivity of the risk to various model variables was analyzed. RESULTS: The residual HIV and HCV risks of the finished products are linear in change with viral incidence rate and inversely linear with product yield and process virus reduction capacity. For the product analyzed in this article, the residual risk is less sensitive to changes in screening test pool size, donation frequency, and inventory hold period. There is only a limited dependency on the donation type (apheresis or whole-blood donations) and a negligible dependency on the manufacturing pool size. CONCLUSIONS: The use of probabilistic model simulation techniques is indispensable when estimating realistic residual viral risks of plasma-derived medicinal products. In contrast to conventional deterministic residual risk estimations, the probabilistic approach allows incorporation of specific manufacturing decisions and therefore provides the only feasible alternative for a correct assessment of residual risks.

Surveillance of risk profiles among new and repeat blood donors with transfusion-transmissible infections from 1995 through 2003 in the Netherlands.

BACKGROUND: To evaluate the effectiveness of blood donor selection, this study reports risk profiles of donors with transfusion-transmissible infections as obtained by ongoing surveillance, 1995 through 2003, in the Netherlands. STUDY DESIGN AND METHODS: A surveillance program was installed to monitor risk profiles among new and repeat donors infected with human immunodeficiency virus (HIV), hepatitis C virus (HCV), hepatitis B virus (HBV), or human T-lymphotropic virus (HTLV), or positive for the presence of syphilis antibodies. At posttest counseling, a physician interviewed donors to clarify possible sources of infection. RESULTS: A total of 167 repeat donors and 404 new donors were interviewed: 33 with HIV, 123 with HCV, 279 with HBV, 21 with HTLV, and 112 with syphilis antibodies. Most HBV, HCV, and HTLV infections were among new donors (80, 85, and 67%), whereas most HIV infections were among repeat donors (79%). Nearly 25 percent of the donors did not report factors at screening that would have deferred them from donating blood. At posttest interviews, new donors with HCV often reported injecting drug use (19%). Repeat donors with HIV often reported male-to-male sex (8/26, 31%). CONCLUSION: A significant level of deferrable behavioral risks was found among donors with confirmed transfusion-transmissible infections that persist despite current donor selection. Reporting such behavior at initial donor selection would have eliminated a substantial part of the infections found. This study argues against relaxing the existing donor deferral of persons practicing male-to-male sex, given their significant proportion of HIV infections among repeat donors. Systematic surveillance of risk factors among infected blood donors provides ongoing information about the effectivity of donor selection and is recommended to evaluate and optimize blood policies.

Costs and benefits of bacterial culturing and pathogen reduction in the Netherlands.

BACKGROUND: Bacterial contamination is a life-threatening risk of blood transfusion, especially with platelet (PLT) transfusions. Bacterial culturing (BCU) of PLTs as well as pathogen reduction (PRT) reduce the likelihood of such contamination. The cost-effectiveness (CE) of these interventions was analyzed after the introduction of the diversion pouch during blood collection. STUDY DESIGN AND METHODS: The balance between costs and benefits of preventing adverse events due to PLT transfusion was assessed with a mathematical decision model and Monte Carlo simulations. Model parameters were obtained from the literature and from Dutch Sanquin blood banks. The balance between costs and benefits is assessed in terms of costs per quality-adjusted life-year (QALY). RESULTS: The costs per 100,000 PLT concentrates in the Netherlands are estimated at $3,277,032 (euro2,520,794) for BCU and at $18,582,844 (euro14,294,495) for PRT. In comparison to the situation without BCU and PRT, costs per QALY are estimated at $90,697 (euro69,767) for BCU (95% confidence interval [CI], $18,149-$2,088,854) and at $496,674 (euro382,057) for PRT (95% CI, $143,950-$8,171,133). The ratio of differences in costs and QALYs between BCU and PRT (the relative CE) is estimated at $3,596,256 (euro2,766,351; 95% CI, $1,100,630-$24,756,615). Large uncertainty in sepsis complication rates and PLT recipient survival exist, causing large uncertainties in the absolute CE for both interventions. CONCLUSIONS: As a result of the unknown probability of sepsis complications and PLT recipient survival, the CE ratios of BCU and PRT in the Dutch setting are highly uncertain. Despite these large uncertainties, it can be concluded that BCU is without doubt more cost-effective than PRT.

Effects of skin disinfection method, deviation bag, and bacterial screening on clinical safety of platelet transfusions in the Netherlands.

BACKGROUND: Bacterial contamination of blood products is a great hazard for development of fatal transfusion reactions. Bacterial screening of platelet concentrates (PC) by aerobic and anaerobic culturing (BacT/ALERT, bioMérieux) was introduced in the Netherlands in October 2001. STUDY DESIGN AND METHODS: In November 2002, a nationwide, uniform skin cleansing method was introduced with a double-swab disinfection with 70 percent isopropyl alcohol. One location routinely used an integrated diversion bag to collect the first 20 to 30 mL. RESULTS: Over the calendar years 2002 and 2003, in total 113,093 PCs derived from pooled buffy coats were screened. After introduction of the new disinfection method, 0.85 percent were initially positive. This was a small reduction compared to the previous disinfection methods under which 0.95 percent were initially positive. The location with use of the diversion bag showed a significantly lower frequency of bacterial contamination, with 0.50 percent before and 0.37 percent after introduction of 70 percent isopropyl alcohol. In addition 8000 apheresis PCs were also screened, showing 24 initially positive samples (0.30%). CONCLUSION: The use of the diversion bag and, to a lesser extent, the use of double swabs with 70 percent isopropyl alcohol, led to a reduction of contamination. As expected, predominant contamination with resident skin bacteria was reduced. The combination of diversion bag and new disinfection led to a frequency of initial positive results for pooled five-donor PCs, which is similar to that of single-donor apheresis PCs. Furthermore, the bacterial detection system and associated product recall procedures have been shown to be effective in preventing transfusion of contaminated PCs and/or related red cells, especially for rapidly growing bacteria.

Diversity and origin of hepatitis C virus infection among unpaid blood donors in the Netherlands.

BACKGROUND: To improve transfusion policy and to increase understanding of the spread of hepatitis C virus (HCV) in the general population, HCV infections among voluntary Dutch blood donors were examined with molecular epidemiologic techniques. STUDY DESIGN AND METHODS: During 6 years, 1997 through 2002, confirmed anti-HCV-positive donors were interviewed on HCV-associated risk behavior with a standardized questionnaire. Additionally, HCV isolates were genotyped, partially sequenced, and compared to sequences obtained from Dutch injecting drug users (IDUs). RESULTS: HCV prevalence and incidence rates among Dutch donors were extremely low; the residual risk of transmitting HCV was calculated to be 1 in 30 million donations. Former IDUs (21%), transfusion recipients (30%), and immigrants (>12%) were identified as major HCV risk groups. Cryptogenic transmission caused 18 percent of infections among new donors and all infections among repeat donors. Compared to IDUs, genotype distribution among donors was highly diverse; major subtypes were 3a (27%), 1a (24%), 1b (24%), 2a/b (10%), and 4 (9%). Half of the donors were infected with IDU-related subtypes 1a and 3a, whereas subtype 1b mainly spread via blood transfusion and various other nosocomial modes of transmission in the past. HCV infections acquired in endemic countries could be clearly identified based on genotype. CONCLUSION: Different modes of transmission are linked to infections with certain HCV subtypes, suggesting separate HCV epidemics, but spillover between different risk groups underlines the value of molecular epidemiologic techniques to gain insight into the origin and dynamics of HCV infections on a population level.

Validation of the NucliSens Extractor combined with the AmpliScreen HIV version 1.5 and HCV version 2.0 test for application in NAT minipool screening.

BACKGROUND: Routine HCV NAT minipool screening (48 donations) of all blood donations was implemented in July 1999 and was combined with HIV NAT in November 2000. This report describes the validation of the NAT methods and the results of quality control testing. STUDY DESIGN AND METHODS: Nucleic acid was extracted from 2-mL plasma samples by using an automated silica-based extraction method (NucliSens Extractor, Organon Teknika). Eluates were tested with RT-PCR (AmpliScreen HIV-1 version 1.5 and AmpliScreen HCV version 2.0 test, Roche Diagnostic Systems). HIV-1 and HCV RNA reference panels and run controls (PeliCheck and PeliSpy, respectively, Sanquin-CLB) and human plasma minipools were used for NAT validation. RESULTS: The 95-percent detection limit (and 95% CI) for HIV-1 RNA genotype B, HIV-1 RNA genotype E, and HCV RNA genotype 1 was 32 (19-76), 30 (17-72), and 21 (13-44) genome equivalents (geq) per mL, respectively. During initial validation, 2332 samples for HIV-1 RNA and 2644 samples for HCV RNA were analyzed, with 13 (0.56%) and 12 (0.45%) invalid test results, respectively. Thereafter, over 19,600 samples (minipools and run controls) were analyzed during the first 11 months of routine screening. Invalid test results for HIV-1 RNA and HCV RNA were found in 1.1 and 1.07 percent of the samples tested, respectively. HIV-1 RNA minipool testing resulted in 27 (0.16%) initial false-positive results and 3 (0.02%) confirmed positive results. HCV RNA minipool testing resulted in four (0.02%) initial false-positive results and five (0.02%) confirmed positive results. CONCLUSION: Routine HIV and HCV NAT minipool screening using the NucliSens Extractor, AmpliScreen HIV-1 version 1.5, and AmpliScreen HCV version 2.0 meets the sensitivity criteria set by the regulatory bodies and provides sufficient specificity and robustness for timely release of blood donations.