Updated December 23, 2009
"House of Numbers" offers new information about
the late Christine Maggiore's experience with HIV testing. In the movie,
her oral narrative and the dated lab reports on screen simply don't correspond.
By including the lab documents,
the film provides evidence that Maggiore was indeed HIV infected, and the
earliest reports suggest that her immune system was controlling the virus well
initially. What follows is a
verbatim transcription of the history she gave in House of Numbers, with a
description of what was shown on screen and what it means medically.
Christine Maggiore: “In 1992, I was encouraged by a doctor to take what’s
called an HIV test as a matter of social responsibility, and I was shocked and
devastated and horrified when the results came back positive. It was one
of those moments that everyone fears their whole life. A week later, I
take the same test to an AIDS specialist. He looks and says, this isn’t a
positive test. I don’t know what this test means.”
The screen shows a lab report from Patricia O’Connell, NP,
for Christine Maggiore, dated 02/24/92. The ELISA (EIA) screening test for HIV
antibodies result is positive, and the Western blot shows two reactive bands,
p24 and gp120/160, but no others. The Western blot is reported as “positive,”
with the additional comment running out of shot, saying, “If the test results
are not consistent with the clinical…”
A WB showing both p24 and gp120/160 bands but not Pol bands
such as p31 is quite likely to be from a person truly infected with HIV but in
the earliest stages before the full range of antibodies have developed.
However, there have been rare cases where such patterns have later proven to be
false positives. Kleinman et al, for example, found 20 such false positives
(without p31) in a series of over 5 million tests from low risk blood donors (a
0.0004% false positive rate). Particularly given that at this stage Maggiore
had no known risk for HIV infection, the specialist was quite correct to be
cautious and read the test results as “indeterminate” – that is, requiring
further testing before making a firm diagnosis.
See, for example: Kleinman et al: False-positive HIV-1 test
results in a low-risk screening setting of voluntary blood donation. Retrovirus
Epidemiology Donor Study. JAMA. 1998 Sep 23-30;280(12):1080-5. http://www.ncbi.nlm.nih.gov/pubmed/9757856
Maggiore: “So I take the test again, and this time my results come
back from the lab marked “’indeterminate.”
The screen shows part of an undated report with GAG and ENV
bands reactive but no reactivity to POL proteins. This is essentially the same
result as the first Western Blot, but is more cautiously interpreted as
“indeterminate” (requiring further testing) rather than “positive”.
Indeterminate results can be from cross-reacting antibodies, or they can be
because the test was taken before the full range of antibodies to different HIV
proteins has developed, often several months after the initial infection. While
a p24 plus gp120/160 pattern is more likely to be the latter, you cannot be
absolutely certain without doing follow-up testing over the ensuing weeks.
Maggiore: “I’m faced with a decision: do I want to wait six weeks to
test again, or do it right away? I opted for right away. My results that time
come back positive.”
At this stage she is still with the first specialist, in
early 1992. But we don’t see the positive result slip for the test Maggiore is
talking about on the screen: instead we are shown a result from 23rd September 1993 – a year and a half
later. The 23rd September 1993 result shows all eight
bands on the Western blot – an unequivocally positive result.
Maggiore: “Took it again, came back negative.”
We now fast forward from early 1992 to a result slip dated 9th August 1993. Only a fragment is
visible. It appears to be a test related in some way to HIV, but it is not
clear what kind of test it is, whether an antibody test or perhaps a p24
antigen test. HIV-1 p24 antigen tests in 1992 were very insensitive and usually
only reacted positive during seroconversion and later in the course of the
disease when viral load levels were high. P24antigen tests cannot, by
themselves, be used to exclude HIV infection.
The shot has been cropped so that the text immediately
before the “HIV” is out of vision, but there is a fragment of a character at
the edge of the frame that looks like a “4.” Immediately after
and below the “HIV” text are illegible printer marks that don’t occur elsewhere
on this image of the slip, suggesting the possibility some text may have faded
or been incompletely erased.
The test result slip as shown is unhelpful in determining whether
Christine Maggiore was ever given a negative diagnosis for HIV infection.
Maggiore: “I took it again—positive”
Presumably the test she is referring to here is the one
shown earlier, dated 23rd September 1993, and showing all eight bands on the Western
blot.
The screen shows a result from 29th September. It is normal practice to
immediately repeat a positive HIV test to exclude handling or laboratory
errors. The result shown appears to be the check test for the positive result
from the specimen taken a week earlier.That result is:
HIV-1 IgG CONFIRMATION IB Positive
(IB stands for “immunoblot” or “Western blot”)
HIV-1 Ab IgG EIA 9.9
(An EIA or ELISA reading of 9.9 is very high, well beyond
the range of most false positive readings which are usually between about 1.0
and 2.0)
Maggiore: I’m very much opposed to the concept of mandatory testing
of any population, because the tests are scientifically shown to be unreliable
and inaccurate.
With the exception of the August 1993 test “result” which
lacks any information even about what type of test it was, the series of
results shown on HoN shows beyond any reasonable doubt that Maggiore had
HIV. The first tests from February 1992-on show reactive EIA with initially
indeterminate WBs which, according to Maggiore, eventually resolved to a clear
positive (but we don’t see this first clear positive result on screen). This is
a typical series of results for someone who starts testing a short time after
being infected with HIV.
The second series of diagnostic tests in September 1993
includes two WBs a week apart, one of which is reported simply as positive, and
one of which individually reports all eight bands as reactive – an
unequivocally positive result. There is also at least one EIA done in this
series, and it reacts at 9.9 – a very clear positive.
Far from showing that HIV testing is
“scientifically shown to be unreliable and inaccurate,” the series of tests
shown on HoN demonstrates that Maggiore’s specialists were diligent and
cautious in interpreting her test results, and only gave her a positive
diagnosis once the evidence proved beyond any reasonable doubt that the
results were true positives.
HouseofNumbers.org thanks the specialists who provided a careful analysis of Maggiore's test results.
