Home Casteddu Amedeo Prize 2008 Amedeo

Flying Publisher   

HIV Medicine 2007
818 pages
Download PDF, 3.7 MB

Other Languages




Persian (Farsi)
Copyright Removal
Mailing List

Advanced Search

3. HIV Testing

Wolfgang Preiser and Stephen Korsman

Awareness of one's HIV infection has gained enormous therapeutic relevance. Consequently, a shift in attitude towards HIV testing has taken place over the past decade: while an HIV test was previously often regarded as a threat to civil rights, the carer is now, in the age of HAART, obliged to advise - if necessary, emphatically - HIV testing: Only those aware of being infected can increase their life expectancy through HAART. Sometimes, an HIV test may be in the interest of a third person, e.g. testing of an index patient after a needlestick injury or the screening of pregnant women.

Besides individual diagnostic use, HIV tests are used in large numbers in the screening of blood and organ donors and (often in an anonymous way) for epidemiological surveillance (UNAIDS, 1997a and 2001).

More? HIV Medicine 2007, Chapter 3: Download

HIV Medicine
15th edition
818 pages
PDF, 3.7 MB

Besides individual diagnostic use, HIV tests are used in large numbers in the screening of blood donors, blood products, and transplant organs to guarantee their safety, as well as (often in an anonymous way) for epidemiological surveillance (UNAIDS, 1997a and 2001).

How to test

The diagnosis of an HIV infection is normally made indirectly, i.e. through the demonstration of virus-specific antibodies (Gürtler 1996). These are found in virtually 100% of those infected with HIV and constitute a marker of a humoral immune response against the agent. In contrast to many other viruses, the antibodies do not have an immunoprotective effect leading to immunity. Their presence equals the presence of chronic and active HIV infection. Cases in which individuals persistently fail to have detectable antibodies against HIV despite the presence of HIV infection are exceedingly rare and so far play little or no role in clinical practice (Connick 2005). However, this might change in the future (Kassutto 2005).

Besides indirect diagnosis based on detection of antibodies, a direct diagnosis of HIV infection is also possible: either through the demonstration of infectious virus (using cell culture - this is only possible in laboratories of at least biological safety level 3), of viral antigens (p24 antigen ELISA) or of viral nucleic acid (i.e. viral genome); the latter is also termed nucleic acid testing (NAT). Viral genome detection is nowadays most often used, as it does not require a high security laboratory, is more sensitive than antigen detection and allows quantification.

To determine the infection status of a patient, direct virus detection by qualitative tests (providing a "yes/no" answer) is only useful under certain circumstances, such as a suspected primary infection or in the case of babies born to HIV-infected mothers (for details see below). However, quantitative viral genome assays have gained great importance: The determination of the so-called "viral load", i.e. the concentration of viral RNA in plasma, has become an indispensable tool for guiding antiretroviral therapy.

The term "HIV test" (still occasionally but inaccurately referred to as the "AIDS test"), however, almost always refers to testing for HIV-specific antibodies as a marker of infection.

HIV antibody diagnosis

Testing for HIV antibodies invariably necessitates the availability of at least two different assays:

1. A screening test and

2. At least one confirmatory test.

It is important to note that two different specimens from the same patient should be tested before the diagnosis is confirmed (see later).

Most screening tests are based on the ELISA principle (enzyme linked immuno sorbent assay) or other, closely related test formats (UNAIDS, 1997b). Screening tests must be extremely sensitive to minimize the chance of yielding a false-negative result. This means that they have to be able to also detect low-avidity antibodies found e.g. early in the course of a primary infection. They also have to be able to detect antibodies directed against all different HIV types (HIV-1, HIV-2) and subtypes (HIV-1-N, HIV-1-O, HIV-1-M) (UNAIDS/WHO, 1992 and 1997).

If the result of such a screening test is positive, this has to be confirmed by at least one confirmatory assay. For this purpose, some countries such as Germany and the United States, prescribe the use of a so-called Western blot or immunofluorescence assay (IFT or IFA). In others, such as the United Kingdom, confirmation may be achieved through the use of different tests applied in a defined sequence in the form of an algorithm - for example, the use of more specific ELISAs. This latter approach is by no means inferior to confirmation by Western blot and does in fact have several advantages, such as being cheaper and more objective (Tamashiro 1993).

The World Health Organization recommends the following strategy for resource-poor settings (WHO 1992):

  • Diagnosis in an healthy individual when population prevalence < 10 %: three immunoassays in series
  • Diagnosis in a healthy individual when population prevalence > 10 %, or in a symptomatic individual: two immunoassays in series
  • Screening of blood donations: single immunoassay (unless the blood donor is informed of the result).

ELISA screening test

Many commercial ELISA tests are available as 96-well microtitre plates. Although the test may be carried out completely manually, they also allow automatization and thus the safe and economical testing of large numbers of patient samples. Various other test kits based on similar formats are also available, often performed by large, fully automatic pipetting and analyzing machines.

Different ELISA "formats" can be distinguished; they are all based on the principle of a specific antigen-antibody reaction. Initially, HIV "whole virus" antigen obtained from cell cultures was used (1st generation tests); nowadays, a mixture of recombinant virus proteins or synthetic peptides representing immunodominant epitopes is employed (2nd and further generation tests).

In order to avoid missing certain virus strains, it is important to know how well the antigens used are able to detect antibodies directed against the potentially occurring virus types (HIV-1, HIV-2) and subtypes (HIV-1-N, HIV-1-O, HIV-1-M). For example, a patient who was infected with HIV-2 in West Africa but is only tested for HIV-1 antibodies - HIV-2 antibodies would not be detected and the infection therefore not be diagnosed. However, because of more or less marked cross-reactivity, a reliable serological differentiation between infections with HIV-1 and HIV-2 is only possible using special assays and, if necessary, will have to be discussed directly with the laboratory.

In most ELISA tests, viral antigen is bound to the so-called solid phase (e.g. on the bottom of the wells in a microtitre plate). Upon addition of patient serum containing antibodies directed specifically against these antigens, antigen-antibody binding will occur. A washing step ensures that all unbound constituents of the serum, including all antibodies not recognizing the viral antigen, are removed.

If antibody has bound to the viral antigen, it is then detected through addition of an enzyme-labeled "conjugate". This conjugate may be either a second (e.g. goat) antibody directed against human antibody molecules ("antiglobulin" assay) or again a viral antigen (often the same antigen that is coated onto the solid phase: "immunometric" or sandwich assay; 3rd generation tests), coupled with an enzyme. The advantage of "immunometric" assays is that they detect antibodies of all classes (including the "early" IgM antibodies which do not play an important role in HIV infection). Again, a washing step ensures that all unbound conjugate is removed.

Finally a substrate is added. If in the previous step conjugate has been bound, this substrate is converted by the action of the enzyme contained in the conjugate. This causes a change of color; the intensity ("optical density", O.D.) of this color reaction is measured and is proportional to the antibody activity in the sample. Positive and negative control specimens are included in each test run and the O.D. values obtained on them are often used to calculate the test's cut-off (i.e. the O.D. value used to distinguish positive from negative values).

Another commonly used method is the MEIA (microparticle enzyme immunoassay). It is based on the same principle as an ELISA; however, the "solid phase" is in the form of microparticles in liquid suspension. Detection is by means of trapping the particles on a membrane and detecting enzyme activity, as with the ELISA.

A special case is "competitive" assays: Here, enzyme-labeled HIV antibodies are added to the solid phase together with the patient's sample. These antibodies then compete for antigen binding sites with the patient's antibodies. If the patient lacks HIV antibodies, all or most of the enzyme-labeled antibody molecules will bind, causing an intense color reaction after addition of the substrate. And vice versa: the more specific the antibodies that are present in the patient's sample, the weaker the color reaction. The intensity of the color reaction is therefore inversely proportional to the antibody activity in the sample. Such "competitive" assays are normally highly specific.

The different formats have different advantages and disadvantages; it is therefore important to know which format a particular assay is based on. So-called 4th generation antibody tests combine the detection of HIV antibodies with that of viral p24 antigen, in order to detect antigen in the blood sample prior to the formation of antibodies, thereby reducing the "diagnostic window" (see below) (Brust 2000).

The accuracy of a test lies in the combination of two factors: the test's sensitivity and its specificity. Sensitivity denotes the test's ability to correctly identify a positive sample as positive, whereas specificity measures its ability to correctly identify a negative sample as negative.

Screening tests are extremely sensitive (almost 100 %), which means that even very low HIV antibody activities - e.g. early in the course of a primary infection - are detected. High sensitivity reduces the chance of a "false-negative" test result and thus of an erroneous conclusion: "The patient is not HIV-infected", although he in fact is. Provided a suitable screening test is used, a negative result six or more months after a potential infection risk means, due to the test's high sensitivity, that the chance of infection is virtually nil (Preiser 2000).

HIV tests sold for the first time after 7 December 2003 are subject to new European Union legislation on in vitro diagnostic devices and have to carry the CE mark. Amongst the conditions to be fulfilled is that 600 HIV-positive samples, including 200 HIV-2-positive ones, obtained at different stages of HIV infection and disease, all have to be identified correctly as positive.

For screening tests, the emphasis has to be placed on the utmost sensitivity; any failure to identify a positive sample correctly could have grave consequences. This high sensitivity, however, causes a somewhat lower specificity. This means that the test result may occasionally be a "false-positive". The test result then indicates the presence of antibodies against HIV although in fact some substance present in the sample was erroneously misidentified as HIV antibodies. Such false-positive results may be caused by immune stimulation of some sort (acute virus infections, pregnancy, immunizations, autoimmune diseases). Presently available HIV screening tests have a specificity of at least 99.5 %; i.e. among 4,000 HIV-negative samples tested, a maximum of 20 may show a false-reactive test result.

Due to the possibility of non-specific reactivity inherent in any assay, it is preferable to use the term "reactive" - rather than "positive" - screening test result thus avoiding misunderstandings. All reactive screening test results must be confirmed by confirmatory testing in order to exclude the risk of reporting non-specific reactivity as "positive". Only then should one talk of a "positive HIV test"!

Important: a reactive screening test does not mean HIV infection! Only a positive confirmatory test allows the diagnosis of an HIV infection, and normally only such a result should be communicated to the patient! It is also important to send a second specimen, as non-specific reactivity can occur due to, for example, the condition of the blood sample, or the samples may have been incorrectly labeled, or switched before testing.

Other potential causes for false-positive (or also false-negative) test results are errors occurring in the laboratory or in the pre- or postanalytical phase. Besides mistakes caused by confusing samples or contamination with positive sample material through suboptimal pipetting techniques etc., clerical errors (incorrect labeling of sample tube or request form, incorrect data entry into laboratory or clinical software etc.) may occur. Utmost attention must be paid also to seemingly unimportant steps in order to safeguard the quality of laboratory testing!

Western blot confirmatory assay

The Western blot is a methodology commonly used for confirmatory testing of screening test-reactive samples. HIV is propagated in cell cultures, harvested, purified and denatured (i.e. split into its constituents). Then, the viral proteins are separated according to their molecular weight by electrophoresis and blotted onto a nitrocellulose membrane. The membrane is cut into strips. To perform the test, the membrane is incubated with patient serum. If the serum contains antibodies against the various viral proteins, these will bind to the areas on the strip onto which the respective antigens have been blotted. If an antigen-antibody reaction takes place, it is revealed using an enzyme-labeled secondary antibody and matching substrate, causing the so-called "bands" to appear on the test strip.

HIV proteins and corresponding bands on the Western blot are designated "p" (for protein) or "gp" (for glycoprotein), followed by the relative molecular mass in kiloDaltons. They can be divided (here using the example of HIV-1 Western blot) into three groups: the env or envelope glycoproteins (gp41, gp120, gp160), the gag or nuclear proteins (p18, p24/25, p55) and the pol or endonuclease-polymerase proteins (p34, p40, p52, p68).

The Western blot is a confirmatory assay that is only carried out if the sample was reactive in the screening assay. Both HIV-1 and HIV-2 Western blots are available commercially. The result of a Western blot may be either positive or negative or (in case of an incomplete pattern of visible bands) equivocal which may reflect borderline or non-specific reactivity.

Different organizations have developed different sets of criteria for interpretation of HIV Western blot results. In order for a Western blot result to be declared positive, the American Red Cross for instance demands at least three bands, one from each group (i.e. one gag, one pol and one env band). The US-American Food and Drug Administration (FDA) demands the p24, the p34 as well as the gp41 or gp120/160 bands (Centers for Disease Control and Prevention, 1989). According to WHO recommendations, however, a Western blot may be judged positive only if two env bands are found. In Germany, the DIN norm 58969 part 41 applies (Deutsches Institut für Normung, 2000): a serum sample is HIV positive if it reacts with at least one viral glycoprotein and one of the other HIV proteins. All other virus-specific band patterns are regarded as questionable.

Among the disadvantages of Western blot are its relatively high price, the comparatively demanding test procedure and the unavoidable subjectivity when reading and interpreting the result. For these reasons many countries prefer confirmatory tests using suitable testing algorithms, consisting of a combination of different ELISA or rapid tests with well-defined sensitivities and specificities and evaluated in the relevant setting. It should also be noted that in 4th generation assays, where both antigen and antibody are detected, confirmation tests might be non-reactive in the period before antibody production, as they detect only antibody.

In addition to the obligatory safeguarding through confirmatory testing e.g. by Western blot, the serological diagnosis of an HIV infection always requires testing of a second, independently obtained blood sample from the patient. If at all possible, the patient should only then be informed about the diagnosis.

Rapid / simple test devices

Nowadays, a number of rapid HIV tests are available; these are also referred to as "point-of-care", "bedside" tests or "rapid/simple test devices". These tests are based on one of four immunodiagnostic principles: particle agglutination, immunodot (dipstick), immunofiltration or immune chromatography (Giles 1999, Branson 2000). In most cases test results are available within fifteen to thirty minutes; often, whole blood or capillary blood (obtained from the tip of a finger or the lobe of the ear) can be used, thus sparing the centrifugation of a venous blood sample obtained through venepunction.

Many of these rapid tests contain a "built-in" internal control, e.g. as a control band indicating whether the sample material and, if applicable, the reagents were added correctly. If this "built-in" control fails, the test result must not be accepted (important to avoid false-negative results, when e.g. the sample was not added or insufficient time allowed until reading the result).

Such rapid tests may be useful if the result is needed quickly, for instance in emergency rooms, before emergency operations, after needlestick injuries and to minimize the rate of "unclaimed" test results (if the result is only available after a few days, some of those tested will not return to receive it). Rapid tests, which are easy to perform and require little in terms of equipment, are also useful in developing countries (Branson 2003, WHO 2004). Nevertheless, such tests should fulfill the same basic requirements as ELISA screening tests (WHO/UNAIDS 1998). In developed countries, a rapid test should ideally only be used as first guidance, and the patient retested as soon as possible in a regular routine laboratory. Problems commonly encountered with rapid tests - besides the need for adequate training of personnel - are the necessity to counsel the patient before testing and to obtain his consent. Any HIV test which can be performed by laypersons always carries the potential of misuse (such as compulsory testing of prisoners, etc.).

In the meantime, several HIV rapid tests have been licensed by FDA: OraQuick™ (OraSure Technologies, Pennsylvania, USA), Reveal (MedMira Laboratories, Halifax, Nova Scotia), Uni-Gold Recombigen HIV Test (Trinity Biotech, Ireland) and Murex single use diagnostic system (SUDS). After worrying experiences (at least five HIV-infected individuals were found to have been informed that their reactive rapid test result had been false-positive, i.e. that they were HIV-negative!) the Centers for Disease Control and Prevention (CDC) recently emphasized the need for adequate confirmatory testing and, if necessary, follow-up testing after four weeks (CDC 2004).

Sample types

In most cases, serum, EDTA plasma, and also occasionally whole blood are used for HIV antibody testing. If sample processing is delayed, it may be preferable to remove the plasma or serum from the corpuscular constituents of blood, as hemolysis may lead to problems with certain tests.

Immunoglobulins may even be eluted from blood spots that were added onto filter paper and dried (Sherman 2005). Testing of such eluates for HIV antibodies is used for the (anonymous and unlinked) screening of pregnant women, using Guthrie test cards with routinely obtained blood spots from newborn babies (whose antibody prevalence mirrors that of their mothers). In developing countries with insufficient facilities for cold storage and transport, dried blood spots may offer a useful and inexpensive tool for sample storage and transport. Once completely dry, blood even from HIV-infected patients does not constitute an infection risk.

Alternatively, urine or oral fluid (oral transudate, often incorrectly referred to as "saliva") may also be employed for some assays (Tamashiro 1994, King 2000). The FDA licensed a rapid test using oral fluid for the diagnosis of HIV infection in March 2004; the same assay, marketed by OraSure Technologies, has been licensed since November 2002 as a rapid test for the detection of HIV antibodies in whole blood (see above). According to available information this assay allows the detection of antibodies against HIV-1 or HIV-2 with a sensitivity of 99.3 % and a specificity of 99.9 %.

Under certain conditions, such non-blood specimen types make testing possible, as they allow non-invasive sampling. However, their sensitivities and specificities are mostly considerably lower. Therefore, blood remains the preferred type of specimen. Whatever type of sample is used, a reactive test result of course requires confirmatory testing.

Test performance

HIV antibody tests are among the best commercially available immunological assays. Sensitivity (high sensitivity few false-negative results) and specificity (high specificity few false-positive results) are the two most important parameters; they have to be calculated for each assay individually. However, in practice it is not so much the sensitivity and specificity of a test that is of interest but rather its predictive value. This is due to one's lack of knowledge of the real HIV status of the patient tested and the need to deduce his/her status from the test's result. The positive predictive value (PPV) is the probability with which a patient with a positive test result is indeed infected; and vice versa, the negative predictive value (NPV) is the likelihood of a patient who tested negative being truly not infected.

Table 1: Two-by-two table.


Test result:




True patient status
(e.g. as determined by reference test)







Table 1 explains the connection between the parameters

= number true-positive / (number true-positive + number false-negative)
= probability of a positive test result if the patient is infected

= number true-negative / (number true-negative + number false-positive)
= probability of a negative test result if the patient is not infected

Positive predictive value (PPV)
= number true-positive / (number true-positive + number false-positive)
= probability that a patient tested positive is indeed infected

Negative predictive value (NPV)
= number true-negative / (number true-negative + number false-negative)
= probability that a patient tested negative is indeed not infected

Although this may initially not seem plausible, the predictive value of a test not only depends on its sensitivity and specificity, but also on the HIV prevalence (i.e. the pre-test-probability of being positive or negative, respectively) in the population tested.

For example:

1. High HIV prevalence: 10 % (i.e. 10 per 100)

Using a test with a sensitivity of 100 % and a specificity of 99 % (i.e. 1 false positive in 100), and screening 1,000 patients, one would expect to see the following:

  • 100 true positives per 1,000
  • 10 false positives per 1,000
  • Positive predictive value: 100 true positives/110 total positives = 91 %

2. Low HIV prevalence: 0.1 % (i.e. 1 per 1,000)

Using the same test (sensitivity of 100 %, specificity of 99 %) and screening 1,000 patients, one would expect to see the following:

  • 1 true positive per 1,000
  • 10 false positives per 1,000
  • Positive predictive value: 1 true positive/11 total positives = 9.1 %

Figure 1 further illustrates this relation, using fictitious populations with HIV seroprevalence rates between 0.02 % (e.g. European blood donors) and 20 % (e.g. sexually active groups in highly endemic countries). It can be seen that in the former, the vast majority of positive (or more aptly, reactive) test results are indeed false positive: only 4.8 % of those who test positive are truly infected! In contrast, 98 % of the positive test results in the high-risk group with a seroprevalence of 20 % are "true" (which is why, according to WHO, confirmatory testing may exceptionally be omitted here). These examples stress the importance of adequate confirmatory testing strategies for all positive screening test results!

Unfortunately, this statistical phenomenon is frequently, for propaganda purposes, misused: Inevitably, in blood donors e.g. in Germany with a low HIV prevalence, indeed only a small proportion of those with a reactive screening test result are truly infected. However, because any screening test reactivity must be followed up further by confirmatory testing even before the individual concerned is informed, this phenomenon should not have major consequences: for if the Western blot does not confirm the reactive ELISA result, the patient or blood donor is simply not HIV-"positive"! Nevertheless, it is unfortunately often used to "prove" the alleged uselessness of HIV tests.

Figure 1: Dependence of the positive predictive value (PPV) on the seroprevalence rate in the population tested, using an antibody test with a constant specificity of 99.6 % (i.e. 4 false-positive results per 1,000 samples tested).

When to test

After an infection, it normally takes between three to twelve weeks before the antibodies produced through the immune response of the infected individual become detectable. An HIV test should therefore not be performed too early after a potential risk contact, even though, understandably, concerned patients may be anxious and press for it. It is important, however, after an occupational exposure, e.g. a needlestick injury, to test immediately in order to confirm initial seronegativity in the person exposed to HIV! In around 5 % of newly infected individuals it will take more than two months until antibodies are formed. Another test may therefore be indicated later on.

The chance of having acquired HIV infection is increased if "Yes" can be answered to one of the following questions:

  • Have you had sexual intercourse without a condom with someone who is HIV-infected?
  • Have you had sexually transmissible infections such as Chlamydia or gonorrhea lately?
  • Have you used the same syringes or cannulas as other intravenous drug addicts (needle sharing)?
  • Did you receive a blood transfusion or blood clotting factors between 1978 and 1985?
  • Have you ever had sex with someone who would answer one of the above questions with "Yes"?

Increasingly, it is argued that HIV testing "should no longer be accorded any special status" (Manavi 2005, Beckwith 2005). It has been found that, with the epidemiological shift into groups not formerly considered as "high-risk", HIV infections may be missed and that the burden of adequate pre-test counseling may discourage carers to suggest patients to be tested.

Problem: The "diagnostic window"

One important problem of HIV antibody testing is the so-called "diagnostic window". This is the time period that elapses between the time of acquisition of HIV infection until detectable levels of antibodies are present (Busch 1997). The switch from antibody-negative to antibody-positive is called "seroconversion". The screening tests currently used are able to recognize an HIV infection six weeks after primary infection in about 80 % and after the 12th week in almost 100 % of cases; only in very rare cases is an infection recognized after just three or even six months. 4th generation screening assays attempt to shorten the duration of the "diagnostic window" by detecting HIV antibodies and HIV p24 antigen simultaneously (Gürtler 1998, Ly 2001). Although these 4th generation tests become reactive earlier in the course of an acute primary infection, due to methodological reasons (Meier 2001), there may occur a second "diagnostic window" phase later on, during which the tests may again become non-reactive.

Early during seroconversion the antibody screening test will be only borderline or weakly reactive. The Western blot carried out for confirmation may at this stage not show any bands at all or an incomplete band pattern, with the p24 band often the first to become visible. The results obtained in such cases are often indistinguishable from those found in uninfected individuals that display a certain degree of non-specific reactivity; here, too, isolated p24 bands are occasionally seen. This illustrates clearly how important it is to pass important clinical information on to the laboratory carrying out the tests (e.g. "suspected primary infection", "routine screening" etc.)!

Such cases often have to remain unclear for the time being, but are resolved by follow-up testing within a short time. If one is indeed dealing with an early seroconversion, seroreactivity will have increased significantly only a few days later, and within a few weeks a complete band pattern will be found on Western blot. It depends on the individual circumstances whether direct virus detection e.g. by means of PCR is advisable at the outset. Attention: antiretroviral post-exposure prophylaxis may make direct virus detection more difficult and potentially delay seroconversion.

The gradual increase of seroreactivity in the course of seroconversion may be utilized for epidemiological studies to measure HIV incidence (i.e. the rate of new infections - in contrast to standard antibody testing that measures HIV prevalence, i.e. established infections.)There are several methods in use:

Detuned assays - by deliberately combining highly sensitive with less sensitive antibody tests, these strategies are able to estimate which proportion of positive samples have only recently undergone primary infection (Parekh 2001, Constantine 2003).

Avidity assays - by treating specimens with an agent that breaks weaker antigen-antibody bonds, such as those seen in immature antibodies early in infection, a ratio (avidity index) between treated and untreated specimen can indicate high avidity (old infection) or low avidity (recent infection) (Suligoi 2002, Puchhammer-Stöckl 2005).

BED IgG capture ELISA - early in infection, the ratio between HIV-specific antibody and other antibody is low, whereas later it is higher. By capturing all IgG molecules, one can detect the proportion of HIV-specific antibody by detecting the amount of HIV antigen that binds to it. "BED" refers to the use of subtypes B, E, and D in the commercially available assay (Dobbs, 2004).

It should be noted that these assays are NOT intended for the individual diagnosis of a recent or primary HIV infection, but are used for epidemiological studies of HIV incidence in populations.

Direct detection of HIV

An HIV infection may also be diagnosed through the detection of virus, rather than indirectly through the detection of antibodies. Virus detection is only necessary in certain situations and, because of its higher cost, should only be undertaken if indicated.

Virus isolation in cell cultures is reserved for special cases, as it is demanding, carries a certain risk and therefore requires the use of a specialized laboratory.

Alternatively, assays for the detection of HIV-1 p24 antigen are available. Although the p24 antigen ELISA has generally been replaced by the more sensitive nucleic acid detection assays, 4th generation antibody screening tests incorporate p24 antigen detection in addition to HIV antibody detection, to shorten the "diagnostic window" period (see above).

The detection of viral nucleic acid (i.e. of virus genome) may be achieved by different laboratory techniques. These methods may be used to detect either proviral cDNA in leucocytes (which requires EDTA whole blood samples) or viral RNA in the cell-free compartment (which requires EDTA plasma or EDTA whole blood).

Qualitative testing for viral genome serves as a marker of infection. It supplements or substitutes antibody testing for the diagnosis of HIV infection in special situations (such as suspected fresh primary infection: absence of antibodies during the diagnostic window; newborn of infected mother: presence of maternal antibodies - also see below).

The quantitative detection of HIV RNA in plasma is used as a prognostic marker, to monitor therapy and to estimate infectiousness (Berger 2002). The most sensitive tests can detect as little as approximately 50 copies/ml.

Various commercial and "in house" methods are available for quantitative nucleic acid testing. These may be based on different technologies: polymerase chain reaction (PCR), branched DNA (b-DNA), nucleic acid sequence-based amplification (NASBA),ligase chain reaction (LCR), or quantitative detection of reverse transcriptase activity. So-called "viral load" testing has now become an indispensable clinical tool, both as a prognostic and as a therapeutic marker. However, it should be noted that NO viral load test is intended to be used as a diagnostic tool. Recently, it has been suggested that this method could be of benefit as a screening tool, using pooled (i.e. mixing several individual samples together) or unpooled samples, for seronegative patients in high-risk groups (Pilcher 2004, Pilcher 2005).

Test results

False-positive results by appropriate confirmatory testing are very rare. A confirmed positive result therefore confirms the presence of HIV-specific antibodies and thus, HIV infection.

A positive test result (i.e. screening and confirmatory tests positive and mistaken sample identity excluded by testing of a second sample) means that the individual tested

  • is infected with HIV (i.e. carries the virus that causes AIDS) - except in young children (see below);
  • may infect others with HIV unless precautions are taken

A positive test result does NOT mean that the person tested

  • has AIDS;
  • will necessarily develop AIDS.

A negative test result means:

HIV antibodies were not detected in the blood of the individual at the point in time when he or she was tested.

A negative test result does NOT mean that:

  • the individual is not infected with HIV (the test could have been performed during the "diagnostic window" period);
  • the person tested is immune or resistant to HIV;
  • the person tested can have sexual intercourse without taking "safe sex" precautions.

It should be noted that people in the window period who test negative for antibody, may be viremic, and at this stage are at their most infectious!

Beyond the "diagnostic window", meaning later than six months after a possible exposure to HIV, an HIV screening test is rarely "false-negative". Thus, a negative test means that the person is not infected with HIV - always assuming of course that in the meantime no renewed exposure has taken place.

A rare "equivocal" result in the confirmatory assay means:

The test has not given an unequivocal result. As a consequence, follow-up testing after a short while is required. Particularly in the case of clinical symptoms such as fever, lymph node enlargement, a rash or neurological symptoms, there may be the suspicion of an acute HIV infection in which seroconversion has only just begun. First antibody reactivities are found, however the full pattern of Western blot bands is not yet present. Seroconversion tends to follow certain patterns; in Western blot, some bands are positive early on (such as p24 or gp120), others appear later.

In the case of an unequivocal Western blot and if an acute primary infection is suspected for clinical and/or anamnestic reasons, direct detection of virus should be attempted by means of PCR. The aim of this is to detect and possibly treat an acute HIV infection in time (also see chapter "Acute HIV infection"). The earlier, the better!

Caution: If, in the case of a suspected fresh primary infection, a quantitative HIV RNA assay is used for virus detection in plasma (because it is often more easily available than PCRs for the detection of proviral DNA in leucocytes), one has to keep in mind that such tests may occasionally lead to false-positive results (Rich 1999). Such false-positive results - that may be seen typically - indicate low levels of HIV RNA (normally not more than 2,000 copies/ml) that are very unlikely to be found in true acute infection (which normally presents with high "viral load" values). Nevertheless, this problem has sometimes caused confusion and misdiagnoses.

This phenomenon may probably occur with any of the available viral load assays. If it is not recognized, the patient will be given an erroneous diagnosis of "infected" with all its possibly deleterious consequences. To avoid this problem (assuming error-free running of the assay and sufficient quality control practices in the laboratory), a test for proviral cDNA in the leukocyte fraction of the blood should be used; however, this is offered by relatively few laboratories.

Special case: Babies born to HIV-infected mothers

Fortunately, the risk of mother-to-child transmission of HIV (MTCT) (see chapter "HIV and pregnancy") has been extremely reduced in industrialized countries and may be as low as 1 %. Nevertheless, HIV diagnosis is essential in all exposed newborns!

In babies born to HIV-infected mothers, HIV antibodies are normally detectable up to around 12 to 15 months of age, and rarely beyond 18 months. These are passively acquired maternal antibodies transferred transplacentally into the unborn child from around the 30th week of pregnancy onwards. These maternal IgG antibodies confer some physiological immune protection against many infections but in the case of HIV are without protective efficacy. This means, however, that most children of HIV-positive mothers, including those not themselves infected (the majority in any setting), will initially have positive HIV antibody test results, albeit with decreasing reactivities over time, until they become negative after complete elimination of maternal antibodies. The laboratory may be able to indicate that the degree of reactivity is decreasing, but confirmation should still take place at a later date, as this is NOT diagnostic.

Therefore, previously one had to wait - normally for nine months or longer - for a significant fall in the child's antibody level; only testing of repeat blood samples taken at regular intervals could exclude HIV infection of the child with certainty (Newell 1995). If HIV antibodies persist in a vertically exposed child beyond the age of 15 months, the child is usually HIV-infected.

Today, PCR allows a more rapid diagnosis. HIV infection of the child should be diagnosed or, hopefully, excluded directly through detection of the virus. So far it is unclear whether the detection of proviral (intracellular) HIV cDNA (from leucocytes) or of (extracellular) HIV RNA (from blood plasma) is more sensitive. In any case, all positive test results must be confirmed immediately on a second sample.

Important: many methods for the detection of HIV nucleic acid may fail in case of "exotic" (i.e. non-subtype B) HIV-1 subtypes (and with HIV-2) and yield false-negative results (Haas 1996). To exclude this, a maternal sample should also be tested if necessary (e.g. if the mother or her source of infection are from outside Europe) to ensure the test's ability to detect the viral strain in question. If the mother tests PCR-positive with the same assay, a negative test result on the child may be used; otherwise a suitable method must be chosen in a specialized laboratory or one has to resort to antibody testing alone with its limitations (see above). As far as quantitative RNA detection methods are concerned, the problem of falsely low-positive results needs to be recognized (see above)!

In exposed babies, at least two negative HIV PCR results are required in order to exclude HIV infection: the first one between the 1st and the 4th month of life, the second after the 4th month, as only then does it reach its full significance for exclusion of infection (Rossi 1992). In addition, PCR should be performed during the first month of life (however not within the first days after delivery, as contamination with maternal virus may occur), as the earliest possible diagnosis of a neonatal infection is important to allow Pneumocystis prophylaxis and early antiretroviral therapy in the first months of life. If this first sample tests positive (and is confirmed), this points to an intrauterine infection (less frequent); in case of perinatal transmission during birth (most common scenario), virus will only be detectable in the samples obtained later. Attention: breastfeeding carries a significant risk of transmission; whatis stated above is only valid if postnatal acquisition of infection is excluded!

Also, with negative HIV PCR results, the complete elimination of maternal antibodies should be documented at least once in HIV-exposed children.

Special case: Needlestick injury or other occupational HIV exposure

Here, two aspects need to be considered: testing of the index patient (the "donor" from whom the potential risk of infection arose) and testing of the exposed (the "recipient"). The national and local regulations pertaining to this must of course be followed, for legal and other reasons.

If the index patient is known, he should be tested - after relevant counseling and consent; ideally the injured individual's superior should be called in immediately - for HIV antibodies, HBsAg (do not forget immunization against hepatitis B virus if necessary), and HCV antibodies. According to the circumstances (e.g. weekend), the use of a rapid assay should be considered (see above). Often a first, preliminary decision must be made even before the test result of the index patient is available; for any delay in instituting HIV post-exposure prophylaxis (HIV PEP) reduces its chances of success (CDC 2001). Therefore, if there is any doubt, the first one or two doses of HIV PEP (that should be readily available day and night in the form of a so-called "starter pack") should be taken and later discontinued once a negative test result becomes available! It would be a mistake to start HIV PEP only once a positive result has been obtained as this will normally mean delaying the first dose beyond the time period during which its chances of preventing infection are at its highest!

If the index patient is seronegative, the chances of him/her - in the absence of any clinical evidence suggesting an acute retroviral syndrome - currently being in the "diagnostic window" phase are remote. It is therefore normally not advisable to employ a method for direct detection of virus (to exclude a fresh primary infection prior to seroconversion)! In case of an unknown index patient the epidemiological situation needs to be taken into account. Important: used injection needles, etc., should not normally be tested for HIV antibodies or HIV genome; effort, cost and particularly the remaining uncertainty (because of the questionable validity of the result) are out of proportion compared to the extremely remote risk of infection with an agent that fortunately has a low stability and tenacity.

If the index patient is HIV-positive, all available information (including the current "viral load", results of resistance assays, etc.) must be considered in the decision on the type of HIV PEP (see the relevant chapter).

In case of an HIV-positive index patient, the injured individual should at once be tested for HIV antibodies by means of a routine screening assay. Demonstration of initial seronegativity is legally important in order to claim compensation and insurance cover in case of occupational transmission! Further follow-up testing is recommended 6 weeks, 3 months and again 6 months after exposure (Ciesielski 1997). If the index patient is HIV-infected, a further follow-up test of the injured recipient is recommended after 12 months; otherwise this is optional (Ridzon 1997). In addition, an HIV test (and possibly also a test for direct detection of virus) should be done immediately if, at anytime after the incident, the recipient develops illness compatible with an acute retroviral syndrome.

Special case: Screening of blood donations

The risk of HIV transmission by transfusion of blood products (or organ transplants) has been minimized by means of carefully selecting donors - attempting to exclude those with potential risk factors for blood-borne infectious diseases, and by screening all donations. This is done using highly sensitive antibody tests, and in addition, more recently, antigen or nucleic acid detection assays. In this setting, the aim is not to make a diagnosis on a patient, although blood transfusion services usually confirm the reactivity. The aim is to exclude all blood that might be infectious, not only blood confirmed to be infectious.

What is relevant in practice?

With each and every HIV test, different important aspects need to be taken into account. Unfortunately and regrettably, careless and negligent actions still occur sometimes.

It has to be realized:

Despite all therapeutic progress, a positive test result still has massive psychological consequences for the person concerned! The meaning of a positive test result can hardly be overrated.

Every patient therefore must be informed that he/she is to be tested for HIV - in advance! A "routine pre-operative" check, as it is still sometimes carried out, can, particularly if the result is positive, carry not only unpleasant but often considerable legal consequences for the medical staff in charge. Depending on local policy, consent might not have to be obtained in writing, but the patient's consent has to be documented in the patient notes. With children or patients unable to give legally valid consent, parents or legal carers have to be informed.

On the other hand, while obtaining valid consent is important, this does not always require in-depth discussions about risk factors, consequences, etc. Instead, pre-test counseling should be adequate for the individual's situation. One should avoid causing unnecessary anxiety and psychological distress.

HIV test results should only be communicated by doctors who are either themselves knowledgeable about HIV and AIDS or who know where to refer patients with a newly diagnosed HIV infection without delay. Individuals confronted with this diagnosis for the first time are in need of intensive support, and they need this immediately! Sometimes patients are still provided with outdated information, and it may then take them weeks, sometimes months, until they first get into contact with a competent medical person.

A positive HIV test result should never be communicated to the patient over the telephone. Firstly, this does not allow sufficient post-test counseling, and secondly it makes it impossible to judge the patient's reaction - which may even be suicidal Occasionally, the reaction is very irrational and emotional, necessitating an extended counseling session. This individual reaction cannot be anticipated. If at all possible, a repeat appointment should be made at the time of drawing the sample, and it should be stressed that results will be given at this appointment, irrespective of whether they are positive or negative, to allow for further counseling and to avoid assumptions being made about not being contacted by telephone. For the reasons mentioned, a rapid test for self-testing is extremely problematic. So-called "point-of-care" or "bedside" tests may be very useful in certain circumstances, e.g. in emergency rooms, but also harbor the problem of unprofessional utilization without adequate counseling and care of the patient.

A reactive screening test result on its own must never, under any circumstances, be communicated to the patient! As there are numerous factors that may cause a screening test to give a non-specific reactive result, the result of a confirmatory assay always has to be awaited. If the band pattern in Western blot is incomplete, this may represent ongoing seroconversion (i.e. acute infection) or non-specificity. Such tests should always be discussed with the laboratory and with an experienced HIV carer - before the patient is informed. We have seen cases in which the patient was left in the belief to be HIV-infected for days or even years - only because of a cross-reactive ELISA and failure to await the result of confirmatory testing.

But even with HIV seropositivity confirmed by Western blot, the following statement should be added to each first positive result: "The demonstration of HIV-specific antibodies does not mean that the patient is suffering from AIDS. Every positive result must be confirmed on a second, independently obtained blood sample, and the patient should only then be informed!"

With each sample sent for HIV testing, one should be aware of the reason behind its request. AIDS phobia poses a difficult problem and is not so rare - i.e., people who are as firmly as erroneously convinced that they are HIV-infected, almost always without any appreciable risk. Often they want to be tested at short intervals and by different carers, sometimes also requesting expensive PCR tests. Such individuals (who often also suffer from the delusion that positive test results are withheld from them) need psychological and perhaps psychiatric help rather than repeated HIV testing.

Useful Internet sources relating to HIV testing

  • World Health Organization (WHO): http://www.who.int with the following web pages:
  • Department of Essential Health Technologies: http://www.who.int/eht/en/
  • HIV/AIDS Diagnostics: http://www.who.int/hiv/amds/diagnostics/en/index.html
  • Diagnostics and Laboratory Technology: http://www.who.int/diagnostics_laboratory/en/index.html
    with downloadable reports containing the evaluation results of different commercially available HIV tests.
  • The 3 by 5 Initiative: http://www.who.int/3by5/en/
  • Center for Disease Control and Prevention (CDC), USA, Division of HIV/AIDS Prevention:
    http://www.cdc.gov/hiv/testing.htm, http://www.cdc.gov/hiv/rapid_testing/
  • U. S. Food and Drug Administration (FDA), Center for Biologics Evaluation and Research (CBER): Licensed / Approved HIV, HTLV and Hepatitis Tests: http://www.fda.gov/cber/products/testkits.htm
  • The European Commission, Enterprise & Industry, Medical Devices:
    containing information about the In Vitro Diagnostic Directive (IVDD)


  1. Beckwith CG, Flanigan TP, del Rio C, Simmons E, Wing EJ, Carpenter CCJ, Bartlett JG. It is time to implement routine, not risk-based, HIV testing. Clin Infect Dis 2005; 40:1037-40.
  2. Berger A, Preiser W. Viral genome quantification as a tool for improving patient management: the example of HIV, HBV, HCV and CMV. J Antimicrob Chemother 2002; 49:713-21. http://amedeo.com/lit.php?id=12003963
  3. Branson BM. Patientennahe Schnelltests für den Nachweis von HIV-Antikörpern. Point-of-Care Rapid Tests for HIV Antibodies. J Lab Med 2003; 27:288-295.
  4. Branson BM. Rapid Tests for HIV Antibody. AIDS Reviews 2000; 2:76-83.
  5. Brust S, Duttmann H, Feldner J, Gürtler L, Thorstensson R, Simon F. Shortening of the diagnostic window with a new combined HIV p24 antigen and anti-HIV-1/2/O screening test. J Virol Meth 2000; 90:153-65. http://amedeo.com/lit.php?id=11064116
  6. Busch MP, Satten GA. Time course of viremia and antibody seroconversion following human immunodeficiency virus exposure. Am J Med 1997; 102(suppl 5B): 117-24. http://amedeo.com/lit.php?id=9845513
  7. Centers for Disease Control and Prevention. Interpretation and Use of the Western-Blot Assay for Serodiagnosis of Human Immunodeficiency Virus Type 1 Infections. MMWR 1989; 38: 1-7. http://amedeo.com/lit.php?id=2501638
  8. Centers for Disease Control and Prevention. Protocols for Confirmation of Reactive Rapid HIV Tests. MMWR 2004; 53: 221-2. http://www.cdc.gov/mmwr/preview/mmwrhtm l/mm5310a7.htm
  9. Centers for Disease Control and Prevention. Updated U.S. Public Health Service Guidelines for the Management of Occupational Exposures to HBV, HCV, and HIV and Recommendations for Postexposure Prophylaxis. MMWR 2001; 50: 1-54. http://amedeo.com/lit.php?id=11442229
  10. Ciesielski CA, Metler RP. Duration of time between exposure and seroconversion in healthcare workers with occupationally acquired infection with human immunodeficiency virus. Am J Med 1997; 102 (suppl 5B):115-6. http://amedeo.com/lit.php?id=9845512
  11. Connick E. Incomplete antibody evolution and seroreversion after treatment of primary HIV type 1 infection: What is the clinical significance? Clin Infect Dis 2005; 40:874-5. http://amedeo.com/lit.php?id=15736022
  12. Constantine NT, Sill AM, Jack N, Kreisel K, Edwards J, Cafarella T, Smith H, Bartholomew C, Cleghorn FR, Blattner WA. Improved classification of recent HIV-1 infection by employing a two-stage sensitive/less-sensitive test strategy. J Acquir Immune Defic Syndr. 2003 Jan 1;32(1):94-103. http://amedeo.com/lit.php?id=12514420
  13. Deutsches Institut für Normung e.V. (DIN). DIN-Norm 58969-41. DIN-Taschenbuch 222: Medizinische Mikrobiologie und Immunologie, Diagnostische Verfahren, 3. Auflage, Stand 2000. Berlin, Wien, Zürich: Beuth-Verlag.
  14. Dobbs T, Kennedy S, Pau C-P, McDougal JS, Parekh BS. Performance Characteristics of the Immunoglobulin G-Capture BED-Enzyme Immunoassay, an Assay To Detect Recent Human Immunodeficiency Virus Type 1 Seroconversion. J Clin Microbiol 2004; 42:2623-2628
  15. Giles RE, Perry KR, Parry JV. Simple/rapid test devices for anti-HIV screening: Do they come up to the mark? J Med Virol 1999; 59:104-9. http://amedeo.com/lit.php?id=10440816
  16. Gürtler L, Mühlbacher A, Michl U, et al. Reduction of the diagnostic window with a new combined p24 antigen and human immunodeficiency virus antibody screening assay. J Virol Meth 1998; 75:27-38. http://amedeo.com/lit.php?id=9820572
  17. Gürtler L. Difficulties and strategies of HIV diagnosis. Lancet 1996; 348:176-9. http://amedeo.com/lit.php?id=8684160
  18. Haas J, Geiss M, Böhler T. False-negative polymerase chain reaction-based diagnosis of human immunodeficiency virus type 1 in children infected with HIV strains of African origin. J Infect Dis 1996; 174:224-5. http://amedeo.com/lit.php?id=8656008
  19. Kassutto S, Johnston MN, Rosenberg ES. Incomplete HIV type 1 antibody evolution and seroreversion in acutely infected individuals treated with early antiretroviral therapy. Clin Infect Dis 2005; 40:868-73. http://amedeo.com/lit.php?id=15736021
  20. King SD, Wynter SH, Bain BC, Brown WA, Johnston JN, Delk AS. Comparison of testing saliva and serum for detection of antibody to human immunodeficiency virus in Jamaica, West Indies. J Clin Virol 2000; 19:157-61. http://amedeo.com/lit.php?id=11090751
  21. Ly TD, Laperche S, Couroucé AM. Early detection of human immunodeficiency virus infection using third-and fourth-generation screening assays. Eur J Clin Microbiol Infect Dis 2001; 20:104-10. http://amedeo.com/lit.php?id=11305462
  22. Manavi K, Welsby PD. HIV testing: Should no longer be accorded any special status. Brit Med J 2005; 330:492-3. http://amedeo.com/lit.php?id=15746110
  23. Meier T, Knoll E, Henkes M, Enders G, Braun R. Evidence for a diagnostic window in fourth generation assays for HIV. J Clin Virol 2001;23:113-6. http://amedeo.com/lit.php?id=11595590
  24. Newell ML, Loveday C, Dunn D, Kaye S, Tedder R, et al. Use of polymerase chain reaction and quantitative antibody tests in children born to HIV -1-infected mothers. J Med Virol 1995; 47:330-5. http://amedeo.com/lit.php?id=8636699
  25. Parekh BS, McDougal JS . New Approaches For Detecting Recent HIV-1 Infection. AIDS Rev 2001;3:183-93.
  26. Pilcher CD, Fiscus SA, Nguyen TQ, Foust E, Wolf L Williams, D Ashby R, Owen O'Dowd J, McPherson JT, Stalzer B Hightow L, Miller WC, Eron JJ, Cohen MS, Leone PA. Detection of Acute Infections during HIV Testing in North Carolina. N Engl J Med 2005; 352;1873-1883
  27. Pilcher CD, Price MA, Hoffman IF, Galvin S, Martinson FEA, Kazembe PN, Eron JJ, Miller WC, Fiscus SA, Cohen MS. Frequent detection of acute primary HIV infection in men in Malawi. AIDS 2004, 18:517-524
  28. Preiser W, Brink NS, Hayman A, Waite J, Balfe P, Tedder RS. False-negative HIV antibody test results. J Med Virol 2000; 60:43-7. http://amedeo.com/lit.php?id=10568762
  29. Puchhammer-Sto¨ckl E, Schmied B, Rieger A, Sarcletti M, Geit M, Zangerle R, Hofmann H. Low Proportion of Recent Human Immunodeficiency Virus (HIV) Infections among Newly Diagnosed Cases of HIV Infection as Shown by the Presence of HIV-Specific Antibodies of Low Avidity. J Clin Microbiol 2005; 43:497-498
  30. Rich JD, Merriman NA, Mylonakis E, Greenough TC, Flanigan TP, Mady BJ, Carpenter CC. Misdiagnosis of HIV infection by HIV-1 plasma viral load testing: a case series. Ann Intern Med 1999; 130:37-9. http://amedeo.com/lit.php?id=9890848
  31. Ridzon R, Gallagher K, Ciesielski C, et al. Simultaneous transmission of human immunodeficiency virus and hepatitis C virus from a needle-stick injury. N Engl J Med 1997; 336:919-22. http://amedeo.com/lit.php?id=9070472
  32. Rossi. Early diagnosis of HIV infection in infants - Report of a consensus workshop, Siena, Italy, January 17-18, 1992. JAIDS 1992; 5:1168-78. http://amedeo.com/lit.php?id=1403648
  33. Sherman GG, Stevens G, Jones SA, Horsfield P, Stevens WS. Dried Blood Spots Improve Access to HIV Diagnosis and Care for Infants in Low-Resource Settings. J Acquir Immune Defic Syndr 2005; 38:615-617
  34. Suligoi B, Galli C, Massi M, Di Sora F, Sciandra M, Pezzotti P, Recchia O, Montella F, Sinicco A, Rezza G. Precision and Accuracy of a Procedure for Detecting Recent Human Immunodeficiency Virus Infections by Calculating the Antibody Avidity Index by an Automated Immunoassay-Based Method. J Clin Microbiol 2002; 40:4015-4020
  35. Tamashiro H, Constantine NT. Serological diagnosis of HIV infection using oral fluid samples. Bull World Health Organ 1994; 72:135-143. http://amedeo.com/lit.php?id=8131250
  36. Tamashiro H, Maskill W, Emmanuel J, Fauquex A, Sato P, Heymann D. Reducing the cost of HIV antibody testing. Lancet 1993;342:87-90. http://amedeo.com/lit.php?id=8100916
  37. UNAIDS (1997a): Blood safety and HIV. UNAIDS Technical Update (UNAIDS Best Practice Collection: Technical Update). Geneva: UNAIDS, October 1997. WC 503.3 http://hiv.net/link.php?id=232
  38. UNAIDS (1997b): HIV testing methods. UNAIDS Technical Update (UNAIDS Best Practice Collection: Technical Update). Geneva: UNAIDS, November 1997. WC 503.1 http://hiv.net/link.php?id=233
  39. UNAIDS (2001): Guidelines for Using HIV Testing Technologies in Surveillance: Selection, Evaluation, and Implementation. WHO/CDS/CSR/EDC/2001.16. UNAIDS/01.22E. http://hiv.net/link.php?id=234
  40. UNAIDS / WHO. Recommendations for the selection and use of HIV antibody tests. Weekly Epidemiological Record 1992; 67:145-49. http://amedeo.com/lit.php?id=1599842
  41. UNAIDS / WHO. Revised recommendations for the selection and use of HIV antibody tests. Weekly Epidemiological Record 1997;72:81-7. http://amedeo.com/lit.php?id=9238418
  42. Voß L. HIV-Testpolitik und -Testpraxis in Deutschland - eine Bestandsaufnahme. Bundesgesundheitsbl - Gesundheitsforsch -Gesundheitsschutz 2000; 43:683-93.
  43. WHO / UNAIDS. The importance of simple/rapid assays in HIV testing - WHO/UNAIDS recommendations. Weekly Epidemiological Record 1998;73:321-6.
  44. WHO. Rapid HIV tests: Guidelines for use in HIV testing and counseling services in resource-constrained settings. Geneva 2004. http://www.who.int/hiv/pub/vct/rapidhivtests/en/
  45. WHO. WHO recommendation for HIV antibody testing. Wkly Epidemiol Rec 1992; 67: 145-52.




General Disclaimer | Mailing List

The editors and the authors of HIV Medicine agree - under certain conditions - to remove the copyright on their book for all languages except English and German.

Please see the conditions under which you may benefit from this offer.