It is interesting how much stock people are willing to place in
blood
tests and any other kind of test for that matter. It must be remembered that
computers only do that which humans program them to do. Likewise, the significance
of certain statistics depends on who's interpreting them and the usefulness
of tests also depends upon how they are being interpreted. It is interesting
that we (possibly because of our perception of our medical technology) are easily
convinced by the results of almost any scientific tests. Let me take a moment
now to put tests into a little perspective. Contrary to popular thinking that
tells us that blood tests, x-rays, or other kinds of tests are always conclusive,
they are more frequently than not equivocal and non-conclusive.
Nevertheless, with our increased reliance on technology, frequently
people are led to believe that these tests are extremely conclusive. Somehow
this might be brought about by the notion that if a machine costs $600,000 or
$700,000 and it can measure intricately the vacillations of the electron clouds
of the tissues of the body or something equally amazing and mind boggling, then
the results must necessarily be conclusive and useful in all situations, which
is, of course, not always the case.
One must realize that tests are not 100% accurate. When one
uses a test to determine whether or not someone has a disease, the test sometimes
will be positive and sometimes negative. When a test is positive and the test
is correct, then that is called a true positive. When the test is positive and
the test is incorrect, it is a false positive. When the test is negative and
it is correct, then it is a true negative. When the test is negative and the
test is incorrect, then it is a false negative. One must remember when choosing
a test one is attempting to find a test that has few false positives and few
false negatives. In medicine there isn't anything that is exact or 100%. The
fewer false negatives and the fewer false positives a test has, the more reliable
and the more useful it is.
Normal Range
A few comments should be addressed to the "normal range," since so many people
are often dismissed as having no physical abnormality on the basis of their
blood tests being within the "normal range." When they do have a physical abnormality
that doesn't show up on a particular test looking for a particular problem then
this is what is known as a false negative. It must be remembered that the "normal
range" is an arbitrary setting of upper and lower limits in order to establish
who does and who does not have a problem. As everyone well knows, what is normal
for one person is not necessarily normal for another. Everyone is different.
Some people are low, some people are a little higher, and some are higher still.
If one plots the number of people having a certain value against the possible
values that the tests can give, one finds that in almost any type of medical
test, the distribution will follow what is known as a
"Bell
Curve". It is called this because it is shaped like a bell. Most people will
tend to have similar values (representing the middle of the distribution). With
fewer people having extremely low or extremely high values, the curve takes
on a bell shape.
Now let's consider the next diagram.

Let's suppose there are three groups of people. Group A are people who have
a physical abnormality that causes their values to run below normal, on average.
Let's suppose that Group B is a group of persons who are not sick with an illness
with which this particular test correlates. Group C are people who are sick
with an illness that causes the values of this particular test to be above normal,
on average. By looking at the first shaded area, we can see that some people
who are normal have values that are actually more typical of those who have
an illness that causes low values. In the second shaded area, we can see that
there are some people who are sick that have values that are very much at a
level that would be consistent with a normal person. In section three we can
see that some of the people who are sick with a sickness that can cause high
values also have levels that are more consistent with people who are without
illness. In section four we can see that there are some people who are without
sickness, who have values that are typical for people who have a sickness that
causes elevated levels, on average. So this diagram points out what everyone
already knows, and that is: what is normal for one person is not necessarily
normal for another.
Without getting too deep into statistics, I would like to point
out some common sense reasoning as it relates to "ranges of normal."
The Bell Curve shape has a curved central portion with tails on either side.
These bell curves can be of different shapes for different tests. Some may
be broad and some may be narrow. Broad central portions indicate that the observations
obtained have a greater degree of variation in different people, and narrow
ranges of normal indicate that the test value of each person tends to be much
more similar to the values of other people.
As we discussed earlier, entropy is the nature of all things to go toward disorder.
If there is no influence causing things to be otherwise, things naturally tend
to be
random. Tests that have random results form a curve
shaped like a rectangle as seen at the top of the following diagram. There is
no influence that is causing any one particular value to be found any more commonly
than any other possible value. This causes the curve to take on a flat shape.
But as influences become stronger in affecting the outcome of a certain test,
the shape of the curve becomes less like that of a random test and will take
on more of a bell-shaped curve.

The stronger the influence, the stronger the shaping effect on the curve. It
is the nature of tests to come out with a disordered, random, or flat-shaped
curve. If some values occur much more commonly than others, it is not an accident.
The stronger the tendency for obtaining a certain result over others, the stronger
the likelihood that there must be a very strong influence. So in that sense,
the flatter the curve, the more random its nature, and the less specifically
the values of a particular test are being influenced. The more peaked the curve,
the more likely it is that there is a very specific reason for the peaking,
and the more meaningful the test may be (See above Diagram).
Limitations Of
Thyroid Blood Tests
As it turns out, there is a very good reason that thyroid hormone
blood tests are not always of great predictive value in the resolution of the
symptoms of decreased thyroid system function.
It's because thyroid hormones don't have their action in the
bloodstream,
they have their action at the site of the
nuclear membrane
receptors of the cells. The thyroid hormones, especially T3, interact
with the thyroid hormone receptor much the same way a key interacts with a lock
so that it may be opened.
At this time, in spite of our vast technology, there is
no
way to measure that action. Not for the past forty years, not now, and probably
not for another twenty years, can that be measured
directly. Historically,
doctors have tried to estimate or predict that action based on what floats around
in the bloodstream. But common sense tells us that, at best, these blood tests
are just an indirect measure of what happens at the
cells. And the symptoms
of DTSF are caused specifically by low body temperature pattern abnormalities
resulting from inadequate thyroid hormone stimulation of the thyroid hormone
receptor sites.
So blood tests are very useful in assessing the function of
the
glands of the thyroid system, however, they are frequently hard to
correlate with the onset and resolution of the
symptoms of DTSF. The
symptoms of DTSF ultimately depend on what happens in the
cells of the
body, not the bloodstream. Therefore, the thyroid hormone blood tests are useful
in assessing the function of the glands of the thyroid system, but cannot and
do not directly measure where
the "rubber meets the road" in terms of
the presence or absence of the symptoms of DTSF.
Most cases of DTSF are caused by Wilson's Temperature Syndrome. Since WTS is an impairment in the peripheral conversion of T4 to T3 (outside the glands
of the thyroid system), it is easy to understand why blood tests are not very
predictive in directing treatment of the
symptoms.
Historically, thyroid blood tests have been used in order to
try to
estimate what will happen at the active site, but the
body
temperature can be used to get a better picture of what actually has
happened at the active site. Thus, the monitoring of body temperature patterns
is a more direct reflection of the adequacy of the thyroid hormone/thyroid hormone
receptor interactions of the body.
Since the symptoms are due to abnormal body temperature patterns,
it is also easy to understand why body temperature patterns are so useful in
the treatment of the symptoms of DTSF.
It is the best indicator that we have.
So treatment of DTSF that does not take body temperature patterns into consideration,
isn't being done correctly. If thyroid hormone/thyroid hormone receptor
interactions are ever able to be measured
directly, such a measurement
will add very little to the predictive value already provided by the body temperature
patterns (because the temperature already measures the bottom line: the body
temperature!).
These facts have led to a great deal of confusion and frustration.
I feel that because the limitations of thyroid blood tests are not always borne
in mind, false assumptions are made which is one of the biggest reasons that
DTSF is so frequently overlooked.
Do Blood Tests Or Temperature Better Reflect Thyroid Stimulation Of The Cells?
It is well known that thyroid hormone levels and body temperature
patterns are related. And, it's known that if thyroid hormone levels in the
blood drop to very low levels, then a patient's body temperature can drop well
below normal and the patient can even become comatose. It is also well known
that patients who have thyroid blood tests that are exceedingly high can often
have fevers well over 100 degrees in a condition that is sometimes called "thyroid
storm."
When thyroid blood levels go too high, the temperature can
go too high, and when thyroid blood levels go too low, the temperature can go
too low. In either extreme, severe symptoms can result. So both tests,
thyroid
blood tests, and
temperature tests can both be correlated to conditions
and sickness.
Which test, then, is best able to predict when a person is
suffering from
inadequate thyroid hormone stimulation, or when a person
has excessive thyroid hormone stimulation at the site of the thyroid hormone
receptors? To help us answer this question, let us consider the shapes of the
curves of distribution of values of these tests when they are performed in a
large number of people. To do this, let's again remember the principle of entropy,
the tendency of all things to go toward disorder. Let us remember, also, that
the significance of any test values can be measured, to an extent, by considering
how that value compares to what value one would expect in a random situation.
Considering The Thyroid Blood Tests
Thyroid hormones do not grow on trees and they don't exist
in nature by accident. That can be demonstrated when one considers that without
a thyroid gland there is no thyroid hormone production and, therefore, no thyroid
hormone levels would be found in the body. So if the thyroid gland was not present
or not functioning, then one would expect to find no hormone. So any thyroid
hormone levels detected in the body represent a non-random occurrence.
For our purposes, let us consider four of the more than ten
thyroid hormone blood tests available. Generally speaking, ranges of normal
contain primarily the central portion of a bell curve (see diagram 4-3). Regardless
of how narrow or broad a bell curve, the central portion of the curve contains
a constant percentage of the measurements taken.
For our purposes, let us say that percentage is 80%. So that means that the
central portion of each bell curve contains 80% of the measured values. In medical
tests, normal ranges are often obtained by finding the central portion of the
bell curve. The lower end of the central portion is represented by the "lower
limit" of normal and the upper end of the central portion is represented by
the "upper limit" of normal.
| Test |
Lower Limit |
Upper Limit |
| T4 |
4.4 ug/dl |
13.9 ug/dl |
| TSH |
0.4 uIU/ml |
6.0 uIU/ml |
| RT3 |
100 pcg/ml |
500 pcg/ml |
| T3 |
55 ng/dl |
171 ng/dl |
|
Let us now consider the upper and lower limits of normal for four of the thyroid
hormone blood tests available (Refer to above table). The first test to be considered
is the
T4 test. The lower limit of normal equals 4.5, and the upper limit
of normal equals 13.0 micrograms per deciliter.
The second test is
TSH. Lower and upper limits of normal
are 0.4 and 6.0 microinternational units per milliliter respectively.
The third is Reverse T3 (
RT3). 100 is the lower limit
of normal and 500 picograms per milliliter the upper.
The fourth test is
T3, lower limit of normal is 55,
upper limit, 171 nanogram per deciliter.
Considering that in the random situation one would expect no
thyroid hormone levels to be present, since they are not formed out of the blue
by themselves, then these upper and lower limits of normal can be considered
to represent a certain number of units
above what one would consider
random. Referring back to diagram 4-4 of a random curve, a broad curve, and
a very peaked curve, you can see that the random curve is flat which causes
the two ends of the "central portion of the curve" to be as far as part as possible.
In the broader bell-shaped curve, one can see that the upper and lower ends
of the central portion of the curve are closer together yet still wide apart.
In the extremely peaked bell-shaped curve, the two ends of the central portion
of the curve are extremely close together. So, the closer together the ends
of the central portion of the curve, the more peaked the curve is. The more
peaked the curve, the more dissimilar the curve is to that of a random situation.
Let us now consider, in percentage form, how far apart the
ends of the central portion of the curves are for the four thyroid hormone blood
tests previously discussed. With T4, the upper limit of normal is 2.9 times
higher than the lower limit of normal. For TSH, the upper limit is 15 times
higher than the lower limit of normal. For RT3 the upper limit is 5 times higher
than the lower limit, and with T3 the upper limit of normal is 3.1 times the
lower limit of normal. Therefore, in a T4 test, there is a 190% difference between
the upper and lower limit of normal. For TSH there is a 1400% difference. For
RT3 there is a 400% difference. And with T3, there is a 210% difference between
the upper and lower limits or normal. One can see by these considerations that
the TSH curve is much broader than the T4 and T3 curves. Therefore, the TSH
curve more closely approaches the shape of a curve of a random situation.
Considering The Body Temperature's Range Of Normal
When one considers that it would be the natural tendency of
a body to be the same temperature as its surroundings, then one may consider
the number of degrees a body temperature is above room temperature to be the
number of units away from what one would consider a random situation. We shall
use 75 degrees as room temperature. Fever is considered to be 100 degrees Fahrenheit
and we will consider this to be the upper limits of normal. 97.0 degrees is
probably lower than the lower end of the central portion of the curve for body
temperature, but for our purposes we will consider this to be the lower limit
of normal. 100 degrees minus 75 degrees equals 25 degrees, which we will consider
the upper limit of normal as compared to the random situation (room temperature).
97.0 degrees minus 75 degrees is 22.0 degrees which we will consider the lower
limit of normal above the random situation. 25 degrees is 1.14 times higher
than 22.0 degrees and therefore, the upper limit of normal is 14% greater than
the lower limit of normal. We can see by comparing this percentage difference
to the percentage differences of the thyroid hormone blood tests that the ends
of the central portion of the temperature curve are much closer to each other
than the ends of the central portion of any of the thyroid test curves. This
indicates a much less random situation, therefore, there must be an extremely
strong influence involved. This
14% difference between the upper and
lower limits of normal is especially interesting when compared to the
hundreds
and hundreds of percentage points difference in the upper and lower limits
of the thyroid blood tests.
Extremely peaked bell curves do not happen by accident and
the greater the peaking, the less accidental the situation. Again, they are
about as likely as tossing a deck of cards in the air and having them land as
a card house. The values of the thyroid hormone blood tests and the body temperature
are related to the extent that extremely low blood values can be related to
extremely low temperatures and extremely high blood values can be related to
extremely high temperatures; and both the thyroid hormones and body temperature
are intimately related to the thyroid hormone/thyroid hormone receptor interaction
at the nuclear membrane of the cells.
The question remains, which test better reflects the influence
and purpose of the all important thyroid hormone/thyroid hormone receptor interaction?
Since they are related, it would seem that if it were critical that the body
maintain a particular set of values for the thyroid blood tests, then the body
temperature patterns would vary widely from person to person to accommodate
specific thyroid blood test values. Conversely, if it were critical that the
body maintain a particular body temperature level, then it would make sense
that the thyroid blood tests values would vary widely to maintain a certain
body temperature level. When one considers that the percentage difference between
the upper and lower limits of normal for TSH are
1400%, and the percentage
difference between the upper and lower limits of normal for the body temperature
is
14%, it's easy to see which curve more represents a nonrandom situation
and
which one fluctuates widely to keep the other at a certain level.
It appears that the thyroid hormone blood levels are regulated to fluctuate
widely in order to provide for the temperature rather than the other way around.
If it is the purpose of the thyroid hormones to fluctuate widely to ensure a
certain temperature, then would the thyroid hormone that fluctuates the widest
be any more important than the others in accomplishing that purpose?
It is interesting to see that the TSH (thyroid stimulating
hormone) test has the widest range of normal of the thyroid blood tests, suggesting
that it may be designed, more than any other thyroid system hormone, to ensure
a normal body temperature pattern. It is interesting, because TSH is the
blood test currently regarded as the most sensitive reflection of thyroid gland
function. But if
TSH is regulated to fluctuate widely to help ensure
a normal temperature pattern, then it is easy to see how the best indicator
of the adequacy of the process is the end result, the
body temperature,
itself.
For example, the study habits of students who are trying to
score well on a certain test may vary widely. A more confident student may decide
not to study very hard, while one with less background might choose to be more
diligent. One could try to measure the adequacy of their preparation for the
test by how hard they studied, but a more accurate method would be by their
actual scores.
Some people consider body temperature patterns to be a more
vague and non-conclusive reflection of thyroid system function, because, "everyone's
different, and a lot of people have body temperatures that are lower than the
average." This is an unusual argument, considering that just as many people
have thyroid blood studies that are lower than the average, and their values
vary to a much greater degree than do their body temperature patterns. We must
remember that the "ranges of normal" are arbitrarily set in an attempt to make
the thyroid tests as useful as possible. Thyroid hormone blood tests are invaluable
in helping to evaluate and regulate the thyroid system, especially the portion
in which the function of the glands are important. However, the fact that there
is only a 14% difference between the upper and lower limits of normal for body
temperature readings indicates that the body temperature readings far better
reflect the status of the thyroid hormone stimulation of the thyroid hormone
receptors.
This leads me to believe that the thyroid hormone/ thyroid
hormone receptor interactions are regulated in such a way as to provide such
specific and consistent body temperature patterns for an extremely important
purpose. And I believe that extremely significant purpose is to help provide
the optimal conditions for the enzymatically-catalyzed reactions of the body,
thereby affecting virtually every bodily function. Thus, the reason for the
maintaining of extremely specific and consistent body temperature patterns is
that essentially all the bodily functions depend upon it. Since thyroid blood
tests fluctuate widely, it is easy to see why temperatures end up having much
greater predictive value, in relation to the symptoms of DTSF, than thyroid
hormone blood tests. Just because a test is more expensive or difficult to perform
does not necessarily mean it is better or more useful.