US Public Health Service Guidelines for Testing and Counseling Blood and Plasma Donors for HIV-1 Antigen

In the United States, the implementation of antibody testing in1985 of all donated blood for human immunodeficiency virus type1 (HIV-1) resulted in a substantial decrease in the transmissionof HIV through blood transfusions. To further decrease the riskfor transmission of HIV by transfusion, the testing of all blooddonations with a combination antibody test for HIV-1 and HIV type-2(HIV-2) was implemented in June 1992.

The risk for HIV transmission by transfusion of screened bloodis minimal. Nearly all cases of transfusion-associated HIV transmissionare now caused by blood donated during the infectious window period(ie, when recently infected donors are infectious but have notyet developed detectable levels of HIV antibody). When whole-virus-lysateenzyme immunosorbent assays (EIAs) were used to screen blood donationsfrom 1985 through 1990, the average length of the window periodwas 45 days (95% confidence interval [CI] = 34 to 55 days). Theaverage window period of the most sensitive contemporary recombinantprotein-based EIA for HIV-1 and HIV-2 antibodies is now 20 daysless, yielding an average infectious window period of 25 days(95% CI = 9 to 41 days).

The increased sensitivity of contemporary HIV-antibody EIAs, improveddonor interviewing about behaviors associated with risk for HIVinfection, and deferral of donors who test positive for HIV, hepatitis,human T-cell leukemia virus type 1 (HTLV-I), or syphilis haveconsiderably improved the safety of the US blood supply. In 1993,only approximately 6 per 100,000 blood donations collected bythe American Red Cross tested positive for HIV antibody. In addition,only an estimated 1 in 450,000 to 1 in 660,000 donations per year(ie, 18 to 27 donations) were infectious for HIV but were notdetected by current screening tests.

During the acute period of infection, tests for p24 antigen candetect HIV infection earlier than antibody tests. P24 antigen,the core structural protein of HIV, is detectable 2 to 3 weeksafter HIV infection during the initial burst of virus replicationassociated with high levels of viremia. During this time, theblood of infected persons is highly infectious, and tests forp24 antigen are usually positive. On average, p24 antigen is detectedan estimated 6 days before antibody tests become positive. Whenantibodies to HIV become detectable, p24 antigen is often no longerdetectable because of antigen-antibody complexing and viral clearance.

In August 1995, the FDA recommended that all blood and plasmadonations be screened for p24 antigen, effective within 3 monthsof licensure of a test labeled for such use. The FDA recommendedp24 screening as an additional safety measure because (1) recentstudies indicated that p24 screening reduces the infectious windowperiod; (2) implementation of p24-antigen testing had become logisticallyfeasible for mass screening; and (3) such testing would reducethe risk for HIV infection for persons who receive donated bloodor blood products. Among the 12 million annual blood donationsin the United States, p24-antigen screening is expected to detect4 to 6 infectious donations that would not be identified by otherscreening tests. If each of these units were divided into an averageof 1.8 blood components, antigen testing would result in removalof an estimated 7 to 11 infectious components each year that wouldotherwise be available for transfusion. The FDA regards donorscreening for p24 antigen as an interim measure pending the availabilityof technology that would further reduce the risk for HIV transmissionfrom blood donated during the infectious window period.

This report provides guidelines for (1) interpreting p24-antigen-assayresults, (2) counseling and follow-up of blood and plasma donorswho have positive or indeterminate antigen-test results, and (3)using p24-antigen testing in settings other than blood banks.These guidelines may be modified when additional information concerningantigen testing under mass screening conditions is collected andanalyzed.

p24-Antigen Test Algorithm and Interpretation of Results

The p24-antigen screening assay is an EIA performed on serum orplasma. If the first screening test is nonreactive, the test resultis reported as negative (see Table 1). If the first screeningtest is reactive, the p24 EIA is repeated in duplicate. If bothduplicate tests are nonreactive, the test result is reported asnegative. If at least one of the repeated p24 EIA tests is reactive,the test is considered repeatedly reactive; the donation is thendiscarded, the donor is deferred from donating blood, and a morespecific assay (the neutralization assay) is performed to verifythe presence of p24 antigen. The neutralization assay should beperformed before informing donors of test results.

As specified by the FDA, donations collected within 3 months ofa repeatedly reactive p24-antigen test (regardless of neutralization-assayresults) should be quarantined pending results of repeat donortesting for antigen and antibody to HIV. Sample storage requirementsand time restrictions specified in the test kit package insertshould be closely followed to prevent sample deterioration, andthus, invalid test results. The FDA recommends that units of wholeblood, blood components, source leukocytes, and source plasmaobtained from donors whose blood samples are repeatedly reactiveon p24-antigen screening tests be destroyed or quarantined andnot used for transfusion or for manufacturing into injectableproducts.

Available data indicate that the p24-antigen assay is sensitiveand specific. The specificity of the p24-antigen test was calculatedby two test-kit manufacturers to be 99% to 95% and 99.93% (Table2). In addition, in one study of 514,000 donations, 225 were repeatedlyreactive on the screening test. Of these donations, neutralizationtests were negative for 220 (98%). Five (2%) donations were negativeon neutralization tests and had detectable HIV antibodies. Testingby the polymerase chain reaction (PCR) for HIV DNA and RNA wasperformed on 120 of these nonneutralizing blood donations, allof which were negative for HIV. Follow-up samples were obtainedfrom 79 of these donors, all of which were negative for HIV-1antibody.

In a prospective study conducted from September 1993 through September1995, a total of 305,989 donations were tested for p24 antigen;3 donors had both repeatedly reactive p24-antigen EIA screening-testresults and positive neutralization results (2 of whom were alsoHIV-antibody positive), and 223 donors had repeatedly reactivep24-antigen EIA screening-test results and negative neutralizationresults. Of those donors who had negative neutralization results,81 later returned to donate blood again. Of these donors, 65 hadnegative test results for HIV-1/HIV-2 antibody and for antigenEIA and neutralization. However, 16 donors who were HIV-1/HIV-2antibody negative on subsequent donations continued to have repeatedlyreactive p24-antigen EIA screening tests that did not neutralize.

A recent study of 51 seroconversion panels has yielded an estimateof the clinical sensitivity of the p24-antigen screening testin detecting blood donated during the infectious window period.An analysis of 69 preseroconversion samples that were positivefor HIV-1 RNA by PCR demonstrated that the antigen test was reactivefor 51 (74%) of those samples. To assess the sensitivity of theneutralization test, two antigen-test-kit manufacturers also performedneutralization testing on samples of blood from persons seroconvertingto HIV. One manufacturer tested 102 repeatedly reactive specimensfrom 30 seroconverting plasma donors; 100% were positive on theneutralization assay. Similarly, a second manufacturer found thatall 52 repeatedly reactive specimens from 25 seroconverting plasmadonors were positive on the neutralization test.

Donor Counseling, Follow-up, and Deferral

Counseling blood and plasma donors who have positive or indeterminateHIV-test results is an essential adjunct to HIV testing. Counselingin the blood-bank setting (1) provides information about follow-updiagnostic evaluation and available medical, preventive, and psychosocialservices and (2) assists infected persons in preventing transmissionto others. HIV counseling should be conducted in accordance withPHS standards and guidelines.

PHS guidelines for notification and counseling of donors who haverepeatedly reactive antigen-test results are based on availabledata. These guidelines may be modified after the collection andanalysis of additional information concerning antigen testingunder mass screening conditions.

Donors with Positive P24-Antigen Results--Donors whoseHIV-antibody-test results are negative but whose screening-testresults for

HIV antigen are repeatedly reactive and neutralization-assay resultsare positive, should be counseled that they are probably infectedwith HIV (Figure 1). Donors who have such test results shouldbe notified promptly after a positive neutralization test. Promptnotification is important because persons who are newly infectedwith HIV and do not have HIV antibodies often have high viraltiters and may be at high risk for transmitting HIV infection.According to FDA recommendations, donors who have repeatedly reactiveand neutralizing p24-antigen tests should be advised that theyare permanently deferred from future blood and plasma donation.

Donors who have repeatedly reactive EIA and neutralizing HIV antigentests should have their results confirmed by follow-up antibodytesting; diagnosis of HIV infection should not be made on thebasis of p24-antigen test results alone. Arrangements for follow-upantibody testing should be incorporated into routine counseling.Because the time between detection of antigen and antibody isestimated to be an average of 6 days, donors who have positivep24-antigen test results can be offered repeat antigen and antibodytesting at any follow-up visit. If repeat antigen tests remainreactive and antibody tests remain negative, antibody testingshould be repeated after a minimum of 8 weeks to allow time forantibodies to develop. Pending repeat testing to confirm the initialpositive antigen test result, strategies to reduce transmissionshould be implemented immediately by the donor (eg, abstainingfrom sexual intercourse and using condoms consistently and correctly).If follow-up antibody tests are positive--thus confirming HIVinfection--infected persons should be referred for medical care;sex and needle-sharing partners of such persons should be advisedto seek HIV testing at clinical sites.

Some donors repeatedly have positive p24-antigen and negativeHIV-antibody-test results, although such an occurrence is unusual.If after 8 weeks such persons still have negative antibody-testresults, they should be referred for further medical evaluation,including determining CD4+ T-lymphocyte cell count or percentage.Testing for HIV by PCR or culture also may be helpful in determiningHIV status; however, neither test is licensed for diagnosis ofHIV infection.

Donors with Repeatedly Reactive P24 Antigen Results but Negativeor Invalid Neutralization Results--In the blood- and plasma-donorpopulation, which has a low prevalence of HIV infection, mostrepeatedly reactive p24-antigen screening test results are expectedto be false-positive reactions. Donors who have repeatedly reactivep24-antigen screening tests but have negative neutralization andHIV antibody tests are probably not infected with HIV. Donorswho have negative neutralization results should be counseled thattheir antigen screening tests were reactive but that their supplementarytests were negative, which likely represents a false-positivetest. Although donors who have negative neutralization resultsshould be reassured that their test result probably does not representinfection, they should be counseled to take HIV risk-reductionprecautions until repeat testing has confirmed their HIV infectionstatus.

An invalid neutralization result occurs when a sample is repeatedlyreactive in the initial screening test, but in the neutralizationassay, the neutralized sample and unneutralized control both fallbelow the cut-off level. Most frequently, invalid results occurwhen a screening test value is low or borderline reactive in anuninfected person; however, invalid results can also occur whena screening test value is low or borderline reactive in an infectedperson. An invalid result may also be the result of sample deteriorationor antigen-antibody complex formation during storage.

Donors who have invalid neutralization results should be counseledthat their antigen screening tests were reactive but their supplementarytests were inconclusive because the neutralization test resultswere invalid. These donors are probably not infected with HIV,but infection cannot be excluded. Donors who have invalid neutralizationtest results should be counseled to take HIV risk-reduction precautionsuntil their HIV infection status is confirmed. Retesting a freshsample may clarify the result.

To exclude HIV infection, donors who have repeatedly reactivep24-antigen screening test results, invalid or negative neutralizationtest results, and negative HIV antibody test results can be offeredrepeat antigen and antibody testing at any follow-up visit. Negativeantigen and antibody test results at the time of follow-up indicatethat the donor was not infected with HIV at the time of the initialtest. For donor reentry purposes, or for diagnostic purposes amongdonors who continue to have reactive antigen screening test resultsand negative antibody test results, repeat antigen and antibodytesting should be performed a minimum of 8 weeks after the initialrepeatedly test.

Negative antigen and antibody tests performed > 8 weeks afterthe initial repeatedly reactive antigen screening tests indicatethat the donor was not infected with HIV at the time of the initialtest. If the donor's screening antigen test remains repeatedlyreactive, neutralization test remains negative, and HIV antibodytests are negative or indeterminate, the donor is probably notinfected with HIV. Further follow-up and additional tests (ie,CD4+ T-lymphocyte cell count or percentage, HIV PCR, and culture)may clarify infection status, although these tests are not licensedfor the diagnosis of HIV infection.

FDA recommendations indicate that donors whose blood samples arerepeatedly reactive by the p24 screening test and negative orinvalid on the neutralization test should be temporarily deferredfrom donating blood or plasma for a minimum of 8 weeks. Thesepersons should be counseled that donor deferral does not indicateinfection, because the screening test on which the deferral wasbased was most likely a false-positive result. If, after 8 weeks,samples from such donors test negative by screening tests forp24 antigen and antibodies to HIV, the donors are not infectedand can be reinstated as blood or plasma donors. However, if,after 8 weeks, samples test repeatedly reactive on the screeningtest for HIV antigen or are positive for antibody, the donorsshould be permanently deferred from donating blood or plasma--regardlessof HIV antigen neutralization-test results.

Implications for Other HIV Test Sites

Initiation of p24-antigen screening of the blood supply may motivatesome persons who are at high risk for HIV infection to donateblood to determine HIV status. Such an unintended inducement couldoffset the benefits of p24-antigen screening. Since 1983, PHShas discouraged persons at high-risk for HIV infection from donatingblood. Such persons should be discouraged from donating bloodand plasma and encouraged to be tested for HIV antibody at othersites.

Initiation of routine testing for p24 antigen in publicly fundedHIV counseling and testing sites, physicians' offices, or othernon-blood-bank settings is not recommended. Few additional infectedpersons would be identified by routine antigen testing in suchsettings, because the estimated average time from detection ofp24 antigen to detection of antibody is only 6 days. In addition,recently infected persons may not have detectable levels of p24antigen. However, antigen testing may be appropriate in certaincircumstances, such as diagnosis of perinatally exposed children.Diagnostic testing for HIV infection in children > 18 monthsof age and adults should routinely consist of an HIV-1-antibodyscreening test and Western blot or immunofluorescent assay forconfirmation of antibodies to HIV-1.

Conclusions

PHS is committed to maintaining a safe blood supply. To furtherpromote this goal, the FDA has recommended that all blood andplasma donations be screened with tests for HIV-1 p24 antigenbecause these test results frequently become positive before assaysfor HIV antibodies. Screening blood donors for p24 antigen isexpected to remove 4 to 6 infectious donations from the bloodsupply each year that would not be removed by other screeningtests.

After implementation of p24-antigen screening, CDC will collaboratewith blood- and plasma-collection agencies to evaluate the sensitivity,specificity, and positive predictive value of the p24-antigentest. As a result, the p24-antigen testing algorithm and counselingguidelines may be modified after additional data are collectedand analyzed.

Adapted from Morbidity and Mortality Weekly Report, vol.45, no. RR-2, March 1, 1996.