CD4
testing
What actually happens to your blood after you go to the clinic
to have your CD4 count taken?
Words Jeffrey Williams
Image Russell Plows
You
have just been told your CD4 or T-cell count is 300 and that this is higher
than the previous result of 280. The doctor says you have no need to worry
as everything is going fine. But what do these numbers mean and how reliable
are they? First, to put things in their true context, the T-cell count is
a measure of the number of a certain type of white blood cells (CD4 cells)
in a minute quantity of your blood.
How much blood?
While it is true that you gave a tube full of blood to the nurse, the final
measurement of 300 corresponds to the number of CD4 cells in one microlitre
(ml) of your blood. That is, in one millionth of a litre, which is about one
ten thousandth of the amount of blood that you gave at the clinic.
The technology for making these measurements is so sophisticated and sensitive
that you only need a tiny percentage of a drop of blood to make the measurement.
So why is so much blood taken? Well, given the size of the tube they stick
in your arm and the speed with which the blood flows, it is often difficult
to collect less than half a tube full. Also, if the equipment used to make
the T-cell measurement in the
laboratory breaks down (which is very unlikely), there is plenty of your blood
left to repeat the measurement, so you don’t have to go back to the
clinic and have your blood taken again.
Leaving the clinic
Once your blood has been put into a tube and labelled appropriately, it goes
to the laboratory for analysis. Most large hospitals have labs on site, so
the blood is analysed relatively quickly, at larger clinics, within a day
or so of leaving your arm. Considering that many blood analysis labs can process
around 500 CD4 count
samples each week, that this is only a small percentage of the type of work
carried out in such labs, and that the number of CD4 count samples is increasing
at about 10 per cent per year, these labs are busy places.
What happens to blood in the lab?
Your tube of blood is put into an automated sampling machine where a quantity
of the ‘full blood’ (the name it is officially given) is removed
by insertion of a micro-pipette through the
septum seal. This precise volume is transferred to an empty tube for further
processing. Full blood contains both red and white blood cells, with more
red than white cells, hence the colour. The first part of the analysis procedure
allows the automated cell counter to discriminate between one type of cell
and another. A variety of chemicals (monoclonal antibodies for those of a
technical bent) are added to the full blood to label the cells of interest,
and the tube is shaken violently to ensure good mixing. When measuring a CD4
count, a specially designed chemical, which absorbs and emits light in a specific
manner, is added to the blood. The chemical attaches itself to the surface
of each CD4 cell in the sample. Before further analysis, the red blood cells,
which are in the majority, are removed. Then the sample is ready to be investigated
by a process called light scattering. A laser beam (usually blue or green)
is shone through the sample. The chemical attached to the CD4 cells absorbs
this laser light and then emits light at a slightly different, but well known
wavelength. As the intensity of the light emitted by the sample is proportional
to the number of absorbing and emitting labels, and assuming that each CD4
cell is labelled, the measured intensity of light scattered from the sample
is related to the number of CD4 cells in the sample. The measured light is
related but not equivalent to the number of CD4 cells in the sample. To give
‘equivalence’, more chemicals are added to the full blood sample
to relate the intensity of scattered laser light to the number of labelled
CD4 cells. The sample of blood and chemicals sits in the laser beam for a
few minutes allowing the electronics to detect and measure the amount of light
being scattered by the sample, which is then recorded. Then the sample moves
out of the laser beam and someone else’s blood is counted.
How accurate is this process?
The equipment used for these measurements is sophisticated and capable of
sensitive measurements. However, the measurements of CD4 counts, like all
experimental recordings, are not without error. If samples of the same blood
were run on the same machine, one after the other, a range of measurements
would be obtained. This means there is a margin of error in the measurement,
and although the first sample might be shown to have 300 CD4 cells per ml
of blood, the second measurement of the same blood could be slightly different.
Experimental errors are present in all measurement, not just in blood analysis.
They arise because of the uncertainty in making laboratory manipulations,
both those made by machines and by technicians. For example, the micro-pipettes
that take a known quantity of full blood and put it into a sample tube are
only accurate to plus or minus three per cent. They are mass produced and
cannot be made more accurate. Other sources of uncertainty arise at different
stages of the measurement of the CD4 cell count. The result of this accumulation
of measurement uncertainty means that the total uncertainty of the final measurement
can be more than plus or minus 10 per cent. So, when you are told that the
number of CD4 cells in your blood on a certain day is 300 per ml, the actual
value is somewhere between 270 and 330 (300 plus or minus 10 per cent) and
cannot be measured more precisely. When two successive measurements of your
CD4 count are compared three months apart and one comes back 300 and the next
280, is there a significant difference between these two numbers? If each
measurement has an uncertainty of plus and minus 10 per cent, we are considering
two measurements which lie between 270-330 and 252-308, and a statistician
would tell you that it is very difficult to see any ‘clear daylight’
between these two measurements with overlapping margins of error. So, it’s
fair to say these measurements are not greatly different.
Using CD4 to decide whether to begin or change therapy
There should be ‘clear daylight’ between successive measurements
before making a firm decision to start therapy; overlapping error bars is
not clear daylight. This problem becomes more important at clinically
relevant points, such as when the CD4 count falls to around 200. An absolute
CD4 count of 400 plus or minus 10 per cent is not a huge problem, but 200
plus or minus 10 per cent is a bigger problem, as at this CD4 level you are
likely to need to start HIV treatment. This is of course why viral load measurements
are also made. If your CD4 count appears relatively stable, (300 one month
and 280 a couple of months later), you need to consider what your viral load
is doing too. The higher it is, the more likely you will need to begin therapy.
If you have two CD4 measures below 350, you should begin to consider treatment.
It is the expertise of the clinician which is crucial in decisions about how
to change treatment; not relying only on a single measurement of your CD4
count. In addition, you should get a resistance test to help identify which
drugs are most likely to work for you.
The basics of bloods
Confused? Here’s a rough guide to blood and CD4 counts. Blood contains
white and red cells. White cells are immune cells and red cells carry oxygen
to the body tissues.The CD4 is a type of white cell which co-ordinates the
immune system’s response to bacteria, infections and viruses; the CD4
count is a blood test which tells you the number of these cells per cubic
ml of blood.
Without HIV treatment, CD4 counts tend to decline over time. A count of 500-1200
cells per cubic ml of blood is the expected range in someone who is HIV negative.
A range of 201-350 cells is the point at which your doctor may recommend you
begin HIV therapy. If you have fewer than 200 cells, you are most likely to
begin developing HIV-related illnesses. Doctors will recommend starting treatment
at this point.
The CD4 percentage is a blood test which tells you what proportion of your
white blood cells are CD4 cells. In someone HIV negative this would be about
40 per cent. As HIV progresses, the percentage tends to decline.
Children have much higher CD4 counts than adults. In a six-month old it would
likely to be over 3,000 per ml of blood. You should have your CD4 count measured
every three months if you are not on HIV treatment, or more often if your
CD4 count is falling rapidly.If you are taking HIV drugs, your CD4 count should
be measured along with your viral load every three months. When you start
therapy your CD4 and viral load should be measured at week four and week 12.
