Window periods

The window period refers to the time after infection and before seroconversion, during which markers of infection (HIV-specific antigen and antibodies) are still absent or too scarce to be detectable. Standard screening tests cannot reliably detect HIV infection until after the window period has passed.

Testing guidelines therefore recommend that a person who may have been recently infected should wait for several weeks before being tested. However, there is a natural variation in the time it takes different individuals to produce detectable antigen and antibodies, making definitive statements challenging.

Older recommendations were to defer testing until as much as three months after exposure.¹ However, the effective window period has grown shorter with more sensitive, newer-generation assays. Current UK testing guidelines² (2008) state that the time between infection and testing positive is typically one month. Many sources agree that, in many cases, the effective window period is probably shorter still: as little as one to three weeks.

However, messages around window periods may not always reflect these newer realities, and may not be consistent or clearly communicated. Some experts have expressed concerns that people may unnecessarily defer or avoid testing due to concerns or confusion about the window period.^{3 4} (The window period also depends on the type of assay used, which may also add some confusion: the figures stated here refer to the fourth-generation, antibody/antigen assays in standard use, while community-based, point-of-care tests used outside the medical clinic setting still have a suggested window period of twelve weeks.)

While testing during the window period should not necessarily be discouraged, it should be made clear that a negative result is not necessarily reliable (see False negatives below), and that the person should return to the clinic to be retested.^{5 6 7}

Advice on how soon to retest varies. Some recommend retesting after six weeks, whilst it is often suggested that a test three months after a possible exposure is needed to definitively exclude infection.⁸ If someone has taken post-exposure prophylaxis, this period needs to be extended to six months.⁹ 

Precise figures for the duration of the window period are difficult to come by, for a number of reasons:

• There are individual variations in its duration.
• People infrequently present to healthcare and have multiple plasma samples taken during this period, making this a difficult topic to investigate.
• A single, precise date of exposure is rarely known.

To be calculated accurately, researchers would have to know the precise date that a person was exposed to HIV, and then have multiple plasma samples to test with different assays (for RNA, antigen or antibodies). From these results, it would be possible to give an average number of days during which tests were not able to detect the infection.

It is more common to be able to identify the first date on which HIV RNA was detectable, and then to calculate the number of days before which other assays are reactive. It is therefore possible to say HIV RNA becomes detectable approximately eleven days before antibodies (or that use of an HIV RNA test reduces the window period by eleven days). It is more challenging to say how many days after exposure HIV RNA is detectable, or what the total length of the window period is.

One way of calculating window periods¹⁰ therefore uses the first detection of HIV RNA as ‘day zero’:

• First detection of p24 approximately five days later (minimum three, maximum eight days).
• First detection of antibodies approximately ten days after detection of RNA (minimum seven, maximum 13 days).

Nonetheless, there is also a gap between exposure and the first detection of HIV RNA. This is sometimes referred to as the eclipse phase, and refers to the time during which there is viral replication principally at the site of infection, before widespread dissemination of virus in the body (as observed in animal models). This time period has been thought by some to vary between four and eleven days,^11,12 or by others to be between one and two weeks, but occasionally longer.^{13 14}

If we use ten days as a figure for this period, the previously given figures would work out as follows:

• First detection of HIV RNA: approximately ten days after exposure.
• First detection of p24: approximately 15 days after exposure.
• First detection of antibodies: approximately 20 days after exposure.

However, these are averages, and if the period between exposure and detectable viraemia is as variable as some authors suggest, these periods will occasionally be several weeks longer.