Blood Safety Decision Making

Most consider the blood supply in the developed world to be at its highest historical safety level. This reflects incremental improvements in donor selection and history screening, blood testing, and process control that span four decades. For years, blood collection professionals and government regulators formulated blood safety policy decisions on the premise that a zero-risk blood supply was achievable. In part, this reflects the perceived tardy transfusion medicine community’s response in the early 1980s to the emergence of human immunodeficiency virus (HIV) in the blood supply and the recognition (of) the scope and severity of posttransfusion non-A, non-B hepatitis (subsequently hepatitis C [HCV]) following that. It reflects also the “dread fear” associated with transfusion-associated HIV. This visceral reaction occurs when devastating, unpredictable, and stigmatizing events threaten potential victims who have minimal avoidance discretion. This fear was validated by numerous transfusion-related HIV cases. It was amplified by widely publicized lawsuits, indictments, and criminal convictions of health ministers and policy makers in the 1980s and 1990s (l’ affaire du sang contaminé in France addressing HIV and Canada’s Royal Commission of Inquiry on the Blood System into blood collection agencies’ response to non-A, non-B hepatitis risk).

Not surprisingly, from the 1980s until now, donor deferrals and blood-testing interventions have been rapidly, successively, and additively implemented for emerging and theoretical risks. Collection facilities introduced anti-HBc and alanine aminotransferase (ALT) testing as surrogates for non-A, non-B hepatitis, HIV p24 antigen testing, then nucleic acid testing (NAT) for hepatitis C, HIV, extensive deferrals for the risks attending transmissible spongiform encephalopathies (TSEs,) NAT for West Nile virus (WNV) and hepatitis B, and antibody testing for Trypanosoma cruzi. The cost-benefit estimates for some of these interventions exceeded by orders of magnitude generally accepted thresholds but did not deter their adoption. For example, the costs per quality-adjusted life-year proximate to implementation include HIV NAT, $1,966,000; HCV NAT, $1,830,000; WNV NAT, $520,000 to $897,000; human T-lymphotropic virus (HTLV) antibody testing, $63,000,000; and T. cruzi antibody testing, $2,123,000 (Fig. 54-1). It is unlikely that this reactive approach can be sustained in the current health care–reform environment.

Application, after HIV entered the blood supply, of a stringent form of the precautionary principle (originally promulgated for environmental protection, not transfusion safety) pushed decision making toward avoidance of all risks. The precautionary principle promotes implementation of measures to mitigate risk even if evidence of a risk is incomplete. It is supposed to be tempered by proportionality; that is, any measures adopted are to be proportional to the risk and with those used in similar circumstances, but some have argued that this has not been the case with blood safety measures, at least by the metric of cost-benefit. Nevertheless, although in potential conflict with evidence-based decision making, this approach resonated with policy advocates charged with transfusion safety. The impact on transfusion safety is mixed. In retrospect, the implementation of HIV p24 antigen testing in the mid-1990s despite enormous studies demonstrating its lack of utility was extreme. Likewise, the recent decision to defer donors with chronic fatigue syndrome, based on a single study, to prevent xenotropic murine leukemia virus–related virus (XMRV) transmission was not necessary. Continued lifetime deferral of men having sex with men even once since 1977 is seen by many to be discriminatory in light of increasingly sensitive in vitro tests and alternative approaches to other risk behaviors. In contrast, when the risk for transfusion transmission of vCJD was theoretical, modeling was used to balance any such danger against the impact of broad donor deferrals on the adequacy of the blood supply and to arrive at a policy decision. Some argue that the magnitude of risk does not justify the stringency of donor criteria and number of resulting deferrals given the tiny risk in a country that is not bovine spongiform encephalopathy (BSE) endemic, but the process was rational and should perhaps be seen as a precedent that “an acceptable risk” is estimable.

Hemovigilance programs, such as the Serious Hazards of Transfusion (SHOT) in the United Kingdom and others in Canada, France, and a fledgling program in the United States, have emerged, supplying evidence about a much broader range of transfusion hazards than just infections. For example, a data-driven decision to minimize plasma transfusions from potentially alloimmunized female donors resulted in a dramatic reduction in transfusion-related acute lung injury in the United Kingdom, and early studies in the United States are consistent with this effect.

These systems provide an opportunity for monitoring the risks and benefits of new initiatives, (e.g., proactive pathogen reduction). Pathogen-reduction processes offer the opportunity to abrogate most of the residual risk for all of the historically important transfusion-transmitted infections, bacterial contamination, babesiosis, malaria, WNV, and Chagas disease. Pathogen reduction could eliminate the often lengthy, reactive, iterative paradigm of emergence of a new pathogen in the population, recognition of a material threat to transfusion recipients, development of donor-deferral strategies followed by development and refinement of test systems that has characterized our historical approach. Critically, broadly active pathogen-reduction processes offer a layer of protection against unsuspected emergence of new agents. If already in use, they would need only to be validated as active against a new agent or appropriate model agents. The challenge to precautionism is balancing the impact of pathogen reduction on product quality and the potential long-term toxicities that may not be apparent in premarketing clinical trials against the unquantified probability that new agents will emerge and threaten enough morbidity to make pathogen reduction clinically and economically feasible.¹

More recently, as cost pressures for health care increase, transfusion professionals are questioning whether the zero-risk paradigm remains relevant. A consensus conference held in Toronto in October 2010 addressed concerns about “safety at any cost” and inconsistent decision-making practices affecting the blood supply.

The conference’s consensus statement potentially portends a paradigm change toward risk-based decision making: risk identification, risk assessment, risk management, and risk communication.² Risk will never be zero, and there is now a realization that cost considerations,³ politics, ideology, and public opinion cannot be ignored. Further policy making considerations will likely take these factors into account, balancing risk, safety, stakeholder interest, and overall impact on public health.

Figure 54-1 VARIOUS SAFETY MEASURES INTRODUCED DURING THE PAST THREE DECADES ENHANCED TRANSFUSION SAFETY.

The x-axis displays costs associated with these advances per quality-adjusted life-year (QALY). Hepatitis C virus (HCV) seroscreening saves costs. Estimates for platelet bacterial culturing and human immunodeficiency virus (HIV) p24 antigen testing are not available. ALT, Alanine aminotransferase; anti-HBc, antibodies against HBcAg; HBV, hepatitis B virus; HLTV, human T-lymphotropic virus; NAT, nucleic acid testing; T. cruzi, Trypanosoma cruzi.

Table 54-1 Patients Benefiting From Cytomegalovirus Risk–Reduced Blood Components*

CMV, Cytomegalovirus; HIV, human immunodeficiency virus.

^*CMV risk–reduced components include CMV-seronegative and leukocyte-reduced components.

Pathogen Reduction

Our approach to new transfusion-transmitted infections is intrinsically reactive. We await disease emergence and identification of the new agent, develop an understanding of the infection’s risk factors, devise donor deferrals, and then implement screening assays. Before deploying a sensitive and specific test, morbidity and mortality will accumulate. This reactive strategy will continue unless more broad-reaching interventions are brought forward. Pathogen-reduction technology (PRT) offers a proactive strategy to address new threats; these technologies involve physical, chemical, and photochemical treatments of blood components to inactivate or decrease viral, bacterial, and parasite infectivity.⁴

Beginning in the 1980s, heat treatment, nanofiltration, and solvent/detergent treatment eliminated or reduced viral transmission in products derived from large-scale plasma fractionation such as albumin, immunoglobulin, and hemostatic factor concentrates.

In the past decade, attention turned to whole blood–derived components: frozen plasma, platelets, and red cells. For these, PRT utilizes methylene blue and visible light treatment for plasma; amotosalen (psoralen) and ultraviolet A (UVA) light, and riboflavin (vitamin B₂) and UVB and UVA light for plasma and platelets; and riboflavin/UV light and S303 (a labile alkylating compound) for red cells. Many European countries use PRT for platelets and plasma. None is implemented in North America at this time.

Amotosalen/UVA and riboflavin/UV have advanced furthest in investigation in North America. They provide significant antiviral activity against all agents for which tests are performed currently, human immunodeficiency virus (HIV), hepatitis B and C, human T-lymphotropic virus (HTLV), West Nile virus, Trypanosoma cruzi, and cytomegalovirus. PRT also inactivates agents causing bacterial contamination of platelets; inactivates white blood cells to prevent transfusion-associated graft-versus-host disease; decreases formation and release of cytokines during storage, reducing febrile, nonhemolytic transfusion reactions; and abrogates white blood cell–induced alloantibody (e.g., human leukocyte antigen (HLA) antibody) formation mitigating alloimmune platelet refractoriness.

Adverse events linked to PRT relate to toxicity and cost. Toxicity affects cell function and recipient safety. Solvent/detergent-treated frozen plasma products prepared commercially in pools of 500 to 2500 donations distributed in the 1990s had reduced α₂-antiplasmin, antitrypsin, and protein S levels. They were associated with deep venous thrombosis in patients with liver disease and, in 2002, withdrawn by the manufacturer. A reformulated product used in Europe has not been associated with these events but has not been submitted for regulatory approval in the United States. Solvent/detergent is not active against hepatitis A or parvovirus. Because solvent/detergent disrupts lipid membranes, it cannot be used with cellular components.

Methylene blue and visible light inactivates pathogens in plasma by targeting nucleic acids. However, it alters fibrinogen structure. Because it damages cell membranes, methylene blue is not used for platelets or red cells. It is not effective against hepatitis A or parvovirus and is not recommend for treatment of thrombotic thrombocytopenic purpura.

Amotosalen/UVA and riboflavin/UVB also target nucleic acids. Treated plasma retains 72% to 77% of fibrinogen and factor VIII levels, greater than 80% of protein S and α₂-antiplasmin, and 96% of ADAMTS13 (a disintegrin and metalloproteinase with thrombospondin motifs 13). Platelets treated with these technologies have approximately 30% lower 1-hour posttransfusion corrected-count–increments than control platelets.⁵ In clinical trials, mild and moderate bleeding frequency is increased, but not severe bleeding complications; and the time between transfusions and total platelet transfusions has not generally been different. It is unclear whether a proportion of platelets is impaired (in which case, platelet dose escalation would ameliorate concerns about lower corrected count increment) or all platelets are affected (increasing the dose would be insufficient). Pulmonary toxicity similar to transfusion-related acute lung injury (TRALI) has been reported in clinical trials and in animal model experiments in which UV light has been implicated. Previous clinical trials in red blood cells (RBCs) were halted because of nonsymptomatic immunoreactivity against S303–induced red cell neoantigens and are being resumed with a reformulated process. Preliminary reports suggest riboflavin/UV causes functional impairment in red cells stored nearest the 42-day expiration date.

Toxicity relates also to residual chemical contamination. However, amotosalen and its products are removed before infusion, and chemical removal is unnecessary for riboflavin-treated blood components.

Cost-effectiveness studies project $1.3 million per quality-adjusted life-year (QALY) for PRT whole blood and $1.4 million per QALY for PRT platelets and plasma. This is costly, but consistent with NAT for HIV, and these analyses do not integrate the risk- or cost-benefit of mitigating an unanticipated emerging pathogen.

The toxicity issues likely will be resolved through technical adjustments. Cost, presumably, would decrease following large-scale implementation.

Interest in PRT remains high because it reduces sepsis-related platelet transfusion complications; inactivates parasites such as Babesia microti, and Plasmodium falciparum; mitigates risks associated with recognized emerging pathogens such as dengue and chikungunya viruses; and proactively decreases threats from unknown, emerging pathogens.

Emerging Infections

The experiences with human immunodeficiency virus (HIV), new variant Creutzfeldt-Jakob disease (vCJD), and West Nile virus (WNV) have made it clear that planning for the emergence of new pathogens that may threaten the blood supply is critical in order to shorten the interval from emergence to mitigation. To that end, the characteristics of pathogens likely to enter the blood supply must be understood. These include the ability of the agent to establish an asymptomatic blood-borne phase, to survive under contemporary processing and storage conditions, to establish infection by the intravenous route, and to cause significant morbidity in a transfusion recipient.

Using these characteristics and a review of contemporary literature, the Transfusion Transmitted Diseases Committee of the American Association of Blood Banks (AABB), from 2005 to 2009 produced a compendium of infectious agents that might become relevant to transfusion medicine and attempted to prioritize those that were identified.¹⁹ A series of 68 fact sheets was developed. The fact sheets provide transfusion medicine professionals and clinicians with an overview of the agents’ phylogeny, epidemiology, clinical characteristics, and interventions that might be useful to protect the safety of the blood supply. Three agents were included in the highest priority stratum: Babesia sp., dengue viruses, and the vCJD prion; a section on each is included in this chapter. The fact sheets are freely available online, cover many of the agents discussed in this chapter, and are to be updated when required.

Since the original publication, several new monographs have been added, reflecting an ongoing “horizon-scanning” initiative designed to alert the medical community to potential threats. These include yellow fever and the yellow fever vaccine viruses after the latter was transmitted to blood recipients when recent vaccine recipients failed to divulge that information during blood donation, and xenotropic murine leukemia virus–related virus after alleged associations with prostate cancer or chronic fatigue syndrome/myalgic encephalitis were published. Other fact sheets have been updated and rereleased to capture relevant new information. The availability of these materials to the transfusion medicine and general medical communities is meant to increase awareness that unexpected infections in transfusion recipients should trigger consideration of that source and to provide background allowing the rational consideration of approaches to avert and minimize their spread.