Transmission of HIV and other blood-borne viruses can occur during transfusion of blood components (ie, whole blood, packed red cells, fresh-frozen plasma, cryoprecipitate, and platelets) derived from the blood of an infected individual.(1,2) Depending on the production process used, blood products derived from pooled plasma can also transmit HIV and other viruses, but recombinant clotting factors cannot.(3,4,5) This chapter discusses the history of HIV transmission through blood products; incidence; control measures; and the current estimated safety of U.S. blood components, blood products, tissues used for transplantation, and sperm used for artificial insemination.
| HIV Infection Transmitted During Blood Transfusion|
| Single Donor Components Before 1985|
HIV infection resulting from blood transfusion has been documented repeatedly since the first case report in late 1982.(6-8,9) In the United States, almost all cases are due to blood transfused before March 1985, when HIV antibody testing became available to screen donated blood. As of December 2001, an estimated 14,262 persons have been diagnosed with AIDS as a result of transfusing contaminated blood or blood products.(10)
In some resource-rich countries, testing of donated blood for HIV antibodies was not immediately initiated for a variety of reasons. France began HIV antibody testing in June 1985, Canada began testing in November 1985, and Switzerland began testing in May 1986. Germany inconsistently tested plasma products between 1987 and 1993, as did Japan in 1985 and 1986. These delays led to criminal investigations in France, Germany, Switzerland, and Japan, which in some cases led to criminal conviction of those persons found to be responsible.(11) At least 20 countries initiated compensation programs for at least some individuals infected by transfusion of HIV-contaminated blood and blood products.(11,12)
The risk of HIV transfusion through infected blood products exceeds that of any other risk exposure. Ninety percent of recipients transfused with HIV antibody-positive blood are found to be HIV infected at follow-up.(1) No HIV-infected but persistently seronegative transfusion recipients have been identified. The 90% probability of seroconversion is independent of the age or sex of the recipient, the reason for transfusion, and the type of component transfused (excluding washed red blood cells, which transmit HIV at a lower rate).(13)
HIV infectivity of red blood cell components that were not washed before transfusion decreases as storage time increases. HIV-contaminated red blood cells stored for <8 days are 96% infectious, whereas those stored for >3 weeks are 50% infectious.(1) The level of a donor's viremia at the time of donation is also an important determinant of HIV transmission risk, but no other donor characteristics have been found to affect transmission.(14) Of all transfused patients, half die within 6 months after transfusion from the underlying disease that necessitated the transfusion. Currently, cases involving transfusion of HIV-positive blood do not increase the overall 1-year posttransfusion mortality rate of recipients in the United States. In Zaire, however, patients transfused with HIV-positive blood are 31% more likely to be dead 1 year after transfusion than are patients transfused with HIV-negative blood.(15) This difference is unexplained but emphasizes the importance of screening blood for HIV in developing countries.
HIV disease due to transfusion progresses in the recipient at rates comparable to those in individuals infected for similar duration but by other routes.(16,17) One report found that a transfusion recipient may develop AIDS more rapidly if the infected blood component comes from a blood donor who develops AIDS soon after the time of the blood donation.(18) Other analyses, however, do not confirm this finding.(19) It is more likely that host factors, particularly the recipient's age and immune status, and perhaps other as-yet-undefined cofactors influence the progression to AIDS.(20,21) The mean time of progression to AIDS is estimated to be 8.2 years for adult transfusion recipients who receive no antiretroviral therapy, with a cumulative prevalence of 20% having AIDS 5 years after infection.(22) This progression rate may be overestimated, and the mean time to AIDS development underestimated, because these values are based primarily on data from recipients identified because they developed AIDS or because they received blood from donors who subsequently developed AIDS. The data exclude many donors and recipients who have not been identified because they remain asymptomatic.
| Single Donor Components Since 1985|
Transmission of HIV by transfusion has become rare in developed countries since the initiation of voluntary deferral of donors at risk for HIV infection and routine HIV antibody testing of all donations. Continued improvement in donor recruitment practices, donor education, donor screening, and blood testing has resulted in continued decreases in the risk of transfusion transmission of HIV. In 1995, the risk in the United States of HIV-1 transmission per unit transfused was estimated to be between 1 in 450,000 and 1 in 660,000.(23,24) By 2003, this estimated risk had decreased to between 1 in 1.4 million and 1 in 1.8 million units.(25,26)
HIV antibody tests fail to identify HIV-infected blood donated by HIV-infected persons who have not yet seroconverted. Exclusion of donors is voluntary. Interviews with HIV antibody-positive donors reveal that most recognize their risk but fail to exclude themselves.(27) As a result, laboratory efforts to eliminate HIV-infected donors have continued and testing has improved. Currently, HIV antibody tests detect both HIV-1 and HIV-2 and detect antibody approximately 22 days (the "window period") after the viremic phase of HIV infection begins. Antigen testing for p24, mandated by the U.S. Food and Drug Administration (FDA) in 1996, shortened the window period to approximately 16 days. The nucleic acid amplification test (NAT), which detects HIV-1 RNA in minipools (16-24 donation samples/pool), was introduced in the United States in 1999 and further reduces the window period of potential HIV transmission to 11 days.(25,26) As of early 2003, three transfusion recipients are known to have become HIV infected by transfusion of HIV antibody-negative, p24 antigen-negative, and HIV NAT-negative blood from two different blood donors (among 25 million donations).(28)
The global perspective is not so bright as that described for resource-rich countries. Worldwide, 75 million units of blood are estimated to be donated annually, compared with 13 million donations in the United States. Of the 191 WHO member states, only 43% test blood for HIV, hepatitis C, and hepatitis B viruses. Transfusion-transmitted HIV infection is thought to account for 80,000-160,000 infections annually, contributing 2-4% of all cases of HIV transmission.(25,29,30) Only 20% of the world's supply of safe blood is available to countries with 80% of the world's population.
Inadequate funding for HIV testing is only part of the problem. Specific issues that urgently need to be addressed include the lack of a sufficient volunteer blood donor pool, and inadequate blood donor screening, information, counseling, and confidentiality. Implementation of standardized and monitored test manufacturing practices, inclusion of test validation procedures, ongoing staff training, and continuous internal and external quality assessment programs are all necessary components of an effective program to prevent transmission. Moreover, transfusion practices must be monitored locally so that HIV transmission from unnecessary transfusions does not occur.
No transfusion-associated HIV-2 cases have been reported in the United States. HIV-2 infection is endemic to western Africa and rarely is reported in the United States, with 24 of the 62 reported cases identified in the northeastern United States.(31) Occasional cases have occurred in Europe.(32,33) Forty-eight of the 62 persons in the United States known to be infected with HIV-2 were born in, had traveled to, or had a sex partner from western Africa. Since June 1, 1992, United States blood banks screen donations for HIV-2. HIV-1 antibody ELISA tests detect 8-92% of HIV-2 infections, depending on the geographic source of sera and the test used. As of 1998, screening has identified four blood or plasma donors with HIV-2 infection.(31,34)
| Idiopathic CD4 Lymphocytopenia|
The term idiopathic CD4 T lymphocytopenia (ICL) describes the condition of patients who develop HIV-associated diseases and have CD4 lymphocyte counts of <300 cells/µL, but are not HIV infected. To date, follow-up studies of blood donors and transfusion recipients with ICL provide no evidence that ICL is caused by a blood-borne agent.(35-37) ICL appears to be caused by a heterogeneous number of immunosuppressive conditions rather than by a single virus.
| HIV Transmission from Plasma-Derived Blood Products|
To produce plasma-derived products, plasma from 2,000 to 30,000 donors is pooled and processed into a single batch (lot). One HIV-infected donor can contaminate an entire lot of product and consequently infect each of the recipients if HIV is not neutralized by sufficient heat, cold ethanol, or other treatments during production. A variety of different blood products can be manufactured by successive precipitation with increasing concentrations of cold ethanol. Individual fractions are then further processed, during which time partially concentrated fractions from as many as 100,000 donors may be combined.(38)
| Albumin and Immune Globulin Products|
Albumin and plasma protein (Cohn fractions IV and V) are extracted with the maximum concentration of cold ethanol and are then pasteurized. They do not transmit HIV. Cohn fraction II products (ie, immune globulins such as Rh immune globulin, gamma globulin, and hepatitis B immune globulin) are treated with somewhat lower concentrations of cold ethanol and cannot be pasteurized without loss of activity. Nevertheless, HIV has not been cultured from lots of Cohn fraction II products known to be positive for HIV antibody. The presence of high-titer antibody to HIV in some lots of hepatitis B immune globulin has resulted in transient (<6 months' duration), low-titer antibody to HIV in recipients and has raised questions about the safety of these products. There have been, however, no documented cases of HIV disease as a result of their use. Over 4.5 million doses of Rh immune globulin have been given since 1968, with no reported cases of HIV disease in recipients. Thus, although recipients of hepatitis B immunoglobulin may become transiently HIV antibody positive by passive acquisition of antibodies from the immunoglobulin preparation, there is no evidence that these individuals are actually infected.(39,40)
| Clotting Factor Concentrates|
Pooled plasma is also precipitated and processed into the factor VIII concentrates used to treat hemophilia A and into factor IX concentrates used to treat hemophilia B. Before 1984, factor concentrates were not heat treated, because heat treatment causes a loss of hemostatic activity. As a result, HIV was not inactivated, and roughly 80% of treated hemophilia A patients and 50% of treated hemophilia B patients were infected with HIV-1.(41-43) The severity of hemophilia, and thus the amount of factor concentrate received, correlated directly with the probability of becoming HIV seropositive. Lower rates of seroprevalence in hemophilia B patients compared to hemophilia A patients appear to be related to the use of higher concentrations of ethanol in the manufacture of factor IX concentrate.
Since 1984, multiple methods for inactivating HIV have been developed and applied.(44,38) Methods vary, but all use both heat treatment and at least one other viral inactivation process. No HIV antibody seroconversions have yet occurred among uninfected persons using factor products now on the market. Improved safety and purity of plasma-derived concentrates does, however, result in a sixfold increase in the annual cost of clotting factor replacement.(45)
Cloning of the factor VIII gene in the late 1980s allowed for the development of recombinant factor products, and preparations purified by monoclonal antibody affinity techniques are now available. Some hemophilia clinicians use these "ultrapure" and "high-purity" products for factor replacement in HIV-infected hemophiliacs because this method decreases exposure to foreign antigen, which evidence suggests may hasten progression of HIV disease.(46-48)
| Hemophilia and Progression to AIDS|
The rate of progression to AIDS in HIV-positive hemophiliacs is directly related to age at the time of HIV infection. The incidence of AIDS in older adult hemophiliacs infected with HIV is comparable to that of HIV-infected homosexual men and transfusion recipients in the same age group.(49,50) Over an 8-year period, older HIV-infected hemophiliacs are more likely than younger HIV-infected hemophiliacs to develop AIDS (13.3% for ages 1-17 years, 26.8% for ages 19-34 years, and 43.7% for ages 35-70 years).(50) Severity of hemophilia and amount of factor concentrate received have not been shown to influence the rate of progression to AIDS in HIV-infected hemophiliacs.(51)
| Lack of Blood Product-Associated Kaposi Sarcoma or Lymphoma|
There is a striking and unexplained absence of lymphoma and Kaposi sarcoma (KS) in hemophiliacs with advanced HIV disease, two thirds of whom present with Pneumocystis jiroveci pneumonia. The next most common presenting diagnoses are esophageal candidiasis (9%) and extrapulmonary cryptococcosis (6%).(52) The absence of KS suggests that human herpesvirus 8 (HHV-8), a gamma herpesvirus detected in 100% of KS lesions, is not readily transmitted by blood products.
| HIV Transmission in Transplant Recipients|
HIV has been transmitted through transplantation of kidney, liver, heart, pancreas, bone, and skin--all of which are blood-containing organs or highly vascular tissues. There are no reports of HIV tissue transmission from HIV-seropositive donors of cornea, ethanol-treated and lyophilized bone, fresh-frozen bone without marrow, lyophilized tendon or fascia, or lyophilized and irradiated dura mater.(53)
Although more than 100,000 organ transplants and 1 million tissue allografts have been performed since 1980, one review described only 32 reports of 75 cases of HIV transmission.(53) The majority of cases occurred before the availability of HIV antibody testing in 1985. Of the 11 reported recipients infected with HIV via transplant after 1985, one received an urgent liver transplant before HIV test results from the donor were available,(54) and one received a kidney from a donor who had been HIV tested 8 months before and then had subsequently seroconverted.(55) Two recipients received an organ from a donor whose plasma had been diluted by emergency transfusion before HIV testing.(56) Notably, however, one HIV-seronegative donor without known risk factors for HIV donated tissues and organs to 48 recipients, of whom 41 were tested for HIV and seven were found to be infected.(57) Four of the seven infected recipients received organ transplants, and three received unprocessed fresh-frozen bone. Of the 34 HIV-tested, seronegative recipients, 25 received ethanol-treated bone, corneas, soft tissue (including fascia lata, tendons, ligaments, arteries, veins), or dura mater.
The low incidence of transplant-related HIV infection is due to adequate donor screening procedures, sensitive and accurate HIV testing, and, when possible, virucidal processing techniques. As with blood transfusion, tissue from the recently HIV-infected donor who has not yet seroconverted must be specifically targeted for exclusion. This process is a greater challenge in the setting of cadaveric transplants, where donor selection cannot incorporate the direct questioning and confidential exclusion techniques used for blood collection.
| HIV Transmission by Artificial Insemination|
| Unprocessed Donor Semen|
Currently, 15 women are known to have been infected with HIV via artificial insemination using sperm from anonymous donors: one in Germany,(58) two in Italy,(59) four in Australia,(60) two in Canada,(61) and six in the United States.(61,62) All but one of these cases of insemination-related infection occurred before the availability of HIV antibody testing. Another woman was infected after insemination with processed sperm from her HIV-seropositive hemophiliac husband.(63) These reports demonstrate that HIV transmission can occur after the use of fresh, cryopreserved, or processed sperm. Both intrauterine insemination and cervical insemination can result in HIV transmission. Other generalizations about the risk of HIV transmission via this route are limited by the retrospective nature of the follow-up, the limited number of infected recipients, and the lack of information regarding HIV viral load of the seropositive donors or other host factors in either donor or recipient at the time of insemination. However, given that approximately 75,000 women are artificially inseminated annually in the United States, it can be concluded that HIV transmission from unrelated semen donors was an infrequent event prior to the availability of HIV testing.(64)
| Donor Screening|
Current guidelines of the U.S. Centers for Disease Control and Prevention (CDC) recommend screening semen donors for HIV antibody on the day of semen donation, freezing the semen, and not using it until the donor is tested again 6 months later and found to be HIV antibody negative.(65) Comparisons of fresh versus frozen sperm show that frozen sperm is 50% less efficacious in fertilization because of decreased sperm motility and viability. The increased number of inseminations required to offset the use of frozen sperm are thought to cost millions of dollars a year in the United States. Proponents for less restrictive recommendations that would not require freezing of sperm argue that an updated review of the risks, benefits, and costs of donor insemination indicate that the risk of HIV infection would be similar to that of blood transfusion (see above).(66) Advances in donor screening, education, and testing are such that the number of newly HIV-infected semen donors within the "window period" prior to the development of detectable antibody is small. Use of HIV NAT tests of donor serum may further limit the number of HIV-transmitting donors.
| Processed Semen from HIV-Positive Males|
The CDC recommends against insemination from HIV-positive men, but some HIV-discordant couples are highly motivated to conceive genetically related children and have used processed semen to reduce the risk of HIV transmission. In a case report from the United States, processing fresh ejaculate by centrifugation to remove cells failed to prevent HIV transmission to women following artificial insemination.(63) A modification of this technique in Italy, however, appears successful in preventing transmission; 29 women were inseminated, and none seroconverted.(67) All 10 babies born to these women have remained HIV negative. The processing method involved centrifugation, repeated washing, and an incubation period that allowed spermatozoa to swim up to the upper layer of the culture medium. Several centers in Europe have now successfully inseminated several hundred HIV-negative women by HIV-positive partners without HIV transmission using this type of protocol.(68) Although this procedure involves some risk of HIV transmission, it may offer reduced risk for highly motivated HIV-discordant couples seeking artificial insemination. These results also provide evidence against the controversial claim that HIV is transmitted by spermatozoa as opposed to infected leukocytes and free virus in seminal fluid.(69,70)
|| ||CDC HIV Statistics. www.cdc.gov/hiv/stats.htm#exposure(accessed March 23, 2003)|
|| ||Donegan E, Perkins H, Vyas G, et al. Mortality in the recipients of blood in the Transfusion Safety Study. Blood 1986;68:296A.|
|| ||CBS News. HIV-tainted blood infects two in Florida. http://www.cbsnews.com/stories/2002/07/19/health/main515694.shtml (accessed March 23, 2003)|
|| ||Noel L. Safe blood starts with me, blood saves lives. Transcripts of World Health Day 2000 April 7, 2000. http://www.who.int/multimedia/whd2000/# (accessed March 23, 2003).|
|| ||Rekart M. HIV transmission by artificial insemination [Abstract]. Program and Abstracts from The IVth International Conference on AIDS, Stockholm, 1988:4026.|