Medicine laboratory - blood cholesterol

Author

blood cholesterol

Allan Adler

2005-07-12, 11:10 pm

The main number I'm interested in is blood cholesterol. There are
various devices made by Accu-Chek that measure blood sugar level;
one of them also measures blood cholesterol. There are walk-in clinics
that will charge a few dollars for a blood cholesterol test performed by
a technician with such a device. Consumers can also purchase the device
directly from Accu-Chek. I've never needed to fast before going to this
clinic. I just go and get the blood test. When I go to a doctor, he usually
asks me to fast before the test.

Questions:
(1) To the extent that the test measures cholesterol, is there any difference
in accuracy between a test that uses the Accu-Chek device and one that is
done by a lab such as Quest, assuming that one fasts for both tests?
(2) If cholesterol is measured in the blood of someone who has not fasted,
will the result generally be higher than it would be if the person
has fasted?
(3) If the answer to (1) is yes, which test generally leads to a higher
measure of blood cholesterol, assuming that the patient fasts before
the blood is taken?
--
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions and
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.

John Gentile

2005-07-12, 11:10 pm

in article y93hdf07qvg.fsf@nestle.csail.mit.edu, Allan Adler at
ara@nestle.csail.mit.edu wrote on 7/11/05 9:30 PM:

> Questions:
> (1) To the extent that the test measures cholesterol, is there any difference
> in accuracy between a test that uses the Accu-Chek device and one that is
> done by a lab such as Quest, assuming that one fasts for both tests?

All "point of care testing" or home testing instruments are not as accurate
as fully controlled and calibrated laboratory instruments. They are pretty
good as long as proper care, controls and calibrations are carried out by
the user, and the absolute proper technique is used for testing. There are
many factors in the collection of the sample that can affect the result of
the test.

> (2) If cholesterol is measured in the blood of someone who has not fasted,
> will the result generally be higher than it would be if the person
> has fasted?

Cholesterol testing does not require any fasting since it does not measure
fat as it comes from your diet. The fasting requirement is for other parts
of the lipid profile - the Triglycerides. This test is extremely sensitive
to diet and we will cancel the test if the patient tells us he did not fast.

> (3) If the answer to (1) is yes, which test generally leads to a higher
> measure of blood cholesterol, assuming that the patient fasts before
> the blood is taken?

Since fasting is not required - the lab test would be more accurate. However
you need to look into the full picture of lipids and heart health. There is
a lot more to the picture than just the cholesterol number. Some one with a
low total cholesterol could still be at risk if their LDL is high and the
HDL is low. Likewize someone with a high total cholesterol would be a low
risk if the HDL is high and the LDL was low.

--
John Gentile MS M(ASCP) yjgent@cox.net
Laboratory Information, QA Manager
VA Medical Center
Providence, RI

The contents of this message are mine personally and do not reflect any
position of the Government or VA.

Allan Adler

2005-07-12, 11:10 pm

In a very helpful reply to my questions, John Gentile <yjgent@cox.net> writes:

>
> Since fasting is not required - the lab test would be more accurate. However
> you need to look into the full picture of lipids and heart health. There is
> a lot more to the picture than just the cholesterol number. Some one with a
> low total cholesterol could still be at risk if their LDL is high and the
> HDL is low. Likewize someone with a high total cholesterol would be a low
> risk if the HDL is high and the LDL was low.

I agree. More on my reasons below. As regards (3), it's possible that when
I wrote "higher", you might have thought I meant "better". I was actually
asking which of the two tests tends to give a larger number as its measure
of blood cholesterol. Suppose patient X tends to be at the high end of
the normal blood cholesterol range. Suppose test T tends to overestimate
the blood cholesterol. Then if test T says that patient X is within the
normal range, then it is probably true that patient X is within the
normal range. So, it is useful to know whether the Accu-Chek tends to
overestimate.

Regarding my reasons for being interested in this: I don't really know
what direct effect the food I eat has on my blood cholesterol, HDL, LDL,
triglycerides. I can avoid foods that tend to have what appears subjectively
to me to have high levels of saturated fat and I can emphasize foods that
fall into certain categories (fruits, vegetables,...) that are supposed
to be healthy. But I have no rational way to determine what the effect
that particular foods have on my numbers. If the doctor tells me that
some number is high, I have no way of knowing, beyond generalities,
what eating habits might be at fault. If I change something in my diet,
I have to wait months before seeing the doctor again and, in any case,
it is expensive to have all the tests done. I see the Accu-Chek as giving
me a way to get frequent and measurable information about the direct effects
of certain foods without having to pay a lot for the information. That is
no substitute for what a doctor does, but I think it can still be useful.

Your answer to my second question suggests that I don't really understand
what the blood cholesterol test measures. Maybe the effects of my diet on
blood cholesterol can't be measured in the direct way that I have in mind.

I'd be glad to know of some technical literature to read on this topic.
I'm a mathematician and I'm used to doing a lot of work to understand
something. In this case, I'm prepared to read medical books if I know
which ones to read.
--
Ignorantly,
Allan Adler <ara@zurich.csail.mit.edu>
* Disclaimer: I am a guest and *not* a member of the MIT CSAIL. My actions and
* comments do not reflect in any way on MIT. Also, I am nowhere near Boston.

Robert

2005-07-12, 11:10 pm

"Allan Adler" <ara@nestle.csail.mit.edu> wrote in message
news:y93d5pob6gn.fsf@nestle.csail.mit.edu...
> In a very helpful reply to my questions, John Gentile <yjgent@cox.net>
writes:
>
higher[vbcol=seagreen]
However[vbcol=seagreen]
is[vbcol=seagreen]
with a[vbcol=seagreen]
the[vbcol=seagreen]
low[vbcol=seagreen]
>
> I agree. More on my reasons below. As regards (3), it's possible that when
> I wrote "higher", you might have thought I meant "better". I was actually
> asking which of the two tests tends to give a larger number as its measure
> of blood cholesterol. Suppose patient X tends to be at the high end of
> the normal blood cholesterol range. Suppose test T tends to overestimate
> the blood cholesterol. Then if test T says that patient X is within the
> normal range, then it is probably true that patient X is within the
> normal range. So, it is useful to know whether the Accu-Chek tends to
> overestimate.

Suggest you read his answer more carefully and his bottom line.
"All "point of care testing" or home testing instruments are not as accurate
as fully controlled and calibrated laboratory instruments."

Reagent stability, instrument stability, sample variability as tissue juices
are involved all add up in terms of "probability" that it will agree with
calibrated controlled instruments. It is an approximation that may be
accurate one day and not the next.

If you take fish oils for example your LDL might go up just a little along
with your HDL with a net increase in your total cholesterol.
The overall bottom line is you need to have an LDL done sooner or later.

>
> Regarding my reasons for being interested in this: I don't really know
> what direct effect the food I eat has on my blood cholesterol, HDL, LDL,
> triglycerides. I can avoid foods that tend to have what appears
subjectively
> to me to have high levels of saturated fat and I can emphasize foods that
> fall into certain categories (fruits, vegetables,...) that are supposed
> to be healthy.

There is dispute on whether saturated or unsaturated fat is bad for you. The
other factor is oxidized cholesterol.

But I have no rational way to determine what the effect
> that particular foods have on my numbers. If the doctor tells me that
> some number is high, I have no way of knowing, beyond generalities,
> what eating habits might be at fault.

Keep a record of what you eat.

If I change something in my diet,
> I have to wait months before seeing the doctor again and, in any case,
> it is expensive to have all the tests done. I see the Accu-Chek as giving
> me a way to get frequent and measurable information about the direct
effects
> of certain foods without having to pay a lot for the information. That is
> no substitute for what a doctor does, but I think it can still be useful.

Most people use it for that.

Prev Med. 2001 Jul;33(1):1-6. Related Articles, Books, LinkOut

Cholesterol screening among children and their parents.

Muratova VN, Islam SS, Demerath EW, Minor VE, Neal WA.

Department of Pediatrics, West Virginia university School of Medicine,
Morgantown 26506, USA. vmuratova@hsc.wvu.edu

BACKGROUND: The Coronary Artery Risk Detection in Appalachian Communities
(CARDIAC) project is designed to test the hypothesis that universal
cholesterol screening of prepubertal schoolchildren is effective in
identifying children and their parents at risk of developing premature
coronary heart disease (CHD) in a high-risk rural population. METHODS: Seven
hundred nine fifth-grade schoolchildren from seven rural Appalachian
counties participated in a school-based cholesterol screening program.
Family history of premature CHD, anthropometric and blood pressure
measurement, tobacco smoke exposure, dietary history, and physical activity
levels were collected. RESULTS: One-fourth (174) of the children were
"presumptively" dyslipidemic upon measurement of nonfasting finger-stick
blood cholesterol (FSC). Subsequent fasting lipid profile obtained for 63 of
these children and 79 of their parents confirmed the presence of
dyslipidemia in 37 children (59%) and 52 parents (66%). Among confirmed
dyslipidemic children, family history was not a good predictor of
dyslipidemia (sensitivity 21.6%). FSC levels were significantly correlated
with fasting total cholesterol of children and their parents. CONCLUSIONS:
Universal nonfasting FSC screening of prepubertal schoolchildren is
effective in identifying dyslipidemic children and their parents, whereas
family history has low sensitivity in predicting children with elevated
blood cholesterol concentrations. Copyright 2001 American Health Foundation
and Academic Press.

PMID: 11482989 [PubMed - indexed for MEDLINE]