Western Blots are a Confirmation Test, Not a Screen for Human Immunodeficiency Virus (HIV) Infection

2008, Volume 19, Number 1


David F. Keren, M.D.

Download PDF

In the past few months, we have seen an increase in the number of serum samples sent to Warde Medical Laboratory that end up having a Negative or Indeterminate Western blot for HIV.

This raises a concern about the type of specimens sent for confirmation. We are concerned that laboratories may be sending unscreened samples, or ones that have had only a single screen for an HIV Western Blot.

The Centers for Disease Control (CDC) does NOT recommend screening with the Western blot. It is a confirmatory test.

By screening with a Western blot, an individual who would be screen Negative could be classified as Indeterminate.

Following a positive screening result (whether by enzyme immunoassay or rapid screening test) the assay must be repeated twice. At least 2 of 3 assays need to be positive to proceed to confirmatory testing. Because the current screening tests are of high quality, I recommend that the laboratory require that the repeats have values that are within 30% of the original EIA OD units. If not, I recommend a fresh second sample to rule out possible contamination of the specimen.

The most common confirmatory testing for repeatedly positive HIV screens is the Western blot. For this, a viral lysate is electrophoresed to separate the molecules by molecular weight. The electrophoretic gel is blotted onto nitrocellulose paper which is cut into strips for immunologic assay. The patient’s serum is incubated with the strip of nitrocellulose, which is then washed and incubated with enzyme-conjugated anti-human immunoglobulin. After washing again, the substrate is added and bands are visualized.

Interpretation of Western blot. Most FDA-approved procedures require two of the following: 120/160, 41, and 24 (where 120/160 envelope proteins are considered as one band—either or both may be present, but even when both are present they only count as one band). The presence of any band on the Western blot, even one that does not correspond to known viral proteins is considered to be indeterminate. Indeterminate sera from low risk populations almost never indicate an early HIV infection. Usually, the cause of the indeterminate is not known.

Indeterminate Western blots are difficult to deal with. They have been associated with heat-related inactivation of serum samples, rheumatoid factor, polyclonal gammopathies, systemic lupus erythematosus, in vitro hemolysis and antibodies against DR human lymphocyte antigens. Only those patients who have repeated reactivity to p24 are likely to seroconvert to positive within the next 6 months. As many as 20% of HIV negative individuals may have antibodies in their sera that will give an Indeterminate pattern when tested by the Western blot. This is not useful clinical information and is an important reason why the CDC does not recommend performing Western blots on HIV screen negative samples.

When you submit serum for HIV Western blot confirmatory testing to a reference laboratory be sure that the sample has been repeatedly positive by an FDA approved HIV screen first.