R: Calculation of Risk for Down
From: Andrea Corda (acord@tin.it)
Mon Feb 26 14:00:08 2001
A couple of years ago I used to repeat the triple test on my patients
whenever the results were somehow "altered" and to my surprise I had to see
that the risk varied from one week to another,in the same patient,from,say
1:249 to 1:2850 and ever since I collected,just for the sake of it,a series
of these "crazy" triple tests.
The fact is that they scare the patient to the bone,and as soon as they get
the result,they want to perform an amnio just to be reassured.
Of course this way of doing things plays in favour of those who want to
perform such procedures and earn in consequence-money and credit-because
they say that the triple test is the "only" valid one.
Ever since,I perform the NT test according to Nicolaides' rules and I do it
between the 12th and the 14th week because at that particular time I'm
allowed to see clearly-transvaginally-the cardiac structures and the great
vessels.(I'm not mentioning the reminder of the fetus' anatomy because cela
va sans dire-the heart is a far more important structure when dealing with
chromosomal anomalies than any other and i don't want to think that seeing a
myelomeningocele or an absent arm or whatever gross anatomy defect is a
difficult job nowadays).I tend to give the due importance to the so-called
mild signs of aneuploidy-EIF,Hyperechogenic bowel,dilated renal pelvis
etc-and double the risk obtained by NT whenever I find one.
This method allowed me to recognize a T21 with normal NT and double EIF at
14 weeks (not to mention the fact that she had-of course-a "normal" result
of her triple test that was performed just before the amnio and just "for
the sake of it"
Regards
Andrea Corda MD
>----- Original Message -----
From: Richard M. Roberts <gene@vol.com>
To: Multiple recipients of list ULTRASOUND
<ultrasound@mail.medispecialty.com>
Sent: Sunday, February 25, 2001 10:04 PM
Subject: Re: Calculation of Risk for Down
>Here is my view on this subject. The biochemical markers are screening
>tests, not diagnostic tests, and the way they are reported in the US is
>atrocious and statistically fallacious. Genetic counseling is also
>corrupted by the way these results are reported, and I have been just as
>guilty as everyone else in utilizing statistically invalid results to
>inappropriately scare the pants off or falsely reassure the patient
>based on the foolish reportage of the laboratory providing the
>biochemical risk. It is high time for us all to realize that the the
>emperor of biochemical screening has no clothes.
>
>Lets analyze this case:
>Biochemical results --a 35 year old...triple screen risk 1:7600, noted
>by Joseph A Worrall, MD RDMS. Does the 1:7600 have any real meaning,
>i.e. her risk is actually 1:7600? NO. What is the real meaning of this
>figure (which is the same as saying, what should be disclosed to the
>patient)? First of all, the lab may not make it clear that this is a
>mid-gestational risk, not the risk AT DELIVERY. They should (but don't)
>provide both. Would anyone care to claim that providing a risk at
>midgeststion and not comparing it to the risk at end of gestation (about
>1/4 less) would be regarded as inappropriate by any sentient pregnant
>woman? This is a 35-year old woman. Women 34 or older with negative
>biochemical screens are at reduced risk, BUT the patient should be
>informed that 10% to 15% of all infants with Down syndrome will be
>detectd AT BIRTH, because the sreening test is not a diagnostic test,
>and has a high rate of falsely negative results. That is the position
>of the American College of Medical Genetics.
>http://www.faseb.org/genetics/acmg/pol-15.htm
>I don't agree with the blanket statement of the ACMG that 2nd trimester
>biochemical screening should not be offered to women older than 34, but
>I most definitely believe that any woman that age MUST have full
>disclosre of the inaccuracy of the testing, so that she may make an
>informed decision. In order for her to make that decision, she must
>have a complete explanation of the natural history of an individual
>affected with Down syndrome (1/3 develop Alzheimers), and she should be
>informed that if she declines amniocentesis, she is going from a 100%
>detection rate to an 85-90% detection rate. If she divorces, I have yet
>to hear of a man who has sued for custody of a severely handicapped
>child. Who should make the informed decision with regard to whether or
>not to have an amnio? THE PATIENT. Does 1:7600 have any other real
>accuracy other than placing her risk under that of a 35-year old? NO.
>There are no published data to say otherwise. I know of no other test
>in clinical medicine which is presented as fact, without confidence
>limits, and thus providing an inappropriate sense of reality as to
>resuts, than biochemical risk evaluation in pregnancy. "Your risk is 1
>in 7600", no more, no less, it is not 1 in 7599 or one in 7601. Very
>reassuring. BUT there is no difference between a biochemical risk
>calculation of 1:7600 or 1:279 at midgestation in their real
>significance. They both have only one known proven result: if a 35-year
>old woman does not have an amniocentesis, then she will go from 100%
>detection rate to 85-90% detection rate. We should not be counseling
>women with such patently imprecise an inaccurate risk figures. As
>Philippe Coquel states, "it is important to know the confidence limits
>of the triple screening". Well, we do not, and therefore we should use
>them appropriately. We should just state whether the woman is at
>"increased risk" (if higher than the usual risk of a 34-year old), or
>"decreased risk" if lower. Why have we been kidding ourselves all these
>years?
>
>Now, and ultrasound finding is different. If you repeat a triple
>screen, you will obtain a result completely unrelated to the first. As
>a screening test, the next try will tend to regress to the norm. An
>ultrasound finding, however, if real, is an objective finding which will
>be able to be seen by any skilled ultrasonographer. If you look again
>in a day, it will still be there (whereas the next day's triple screen
>could be normal). So yes, a reasonable analysis predicts that
>ultrasonographic findings are independent of statistical biochemical
>screening. Why on earth should any of the unusual physical features in
>Down syndrome have any influence on biochemical screening? How would a
>flat midface, incomplete outfolding of the scapha helix, prominent ears,
>inwardly-curing 5th interphalanges, redundant skin over the back of the
>neck, relatively short femora and humeri, even duodenal atresia, have
>any bearing on the levels of biochemical markers? You are comparing
>physical findings to abnormal cellular biochemistry. Can't be done. I
>happen to like Nikolaides' division into "hard" and "soft"
>ultrasonographic findings, based on outcome of, I believe, over 100.000
>successive first and second trimester ultrasounds. He classifies
>echogenic bowel as a "hard" sign--which means offering aminocentesis
>after explaining the increased risk to an 18-year old, let alone a 35
>year old.
>
>AND DONT REPORT AN INCREASED RISK AS ABNORMAL. IT IS NOT ABNORMAL, IT
>IS INCREASED RISK.
>
>Please pardon my capitalization.
>Regards,
>Richard (gene)
>
>>>There is reasonable evidence to indicate that ultrasound and
>>>biochemical markers are largely independent. We can review what data is
>>>available, but I'm sure you'll also see more on this in the future.
>>>Recall that ultrasound and biochemical markers also are independent in
>>>the first trimester so this should not be surprising. Although it would
>>>be preferable to offer a combined ultrasound-biochemical risk (as
>>>Bahado-Singh has does, and as we will present at the AIUM in a few
>>>weeks), a simple method is to base the apriori risk on the triple screen
>>>result (biochemistry plus age). We will need further studies to see how
>>>accurate that approach is, but for now it seems like a reasonable
>>>approach to use in practice.
>>>
>>>David Nyberg
>>>
>>>On Fri, 23 Feb 2001, Ph Coquel wrote:
>>>
>>>> Doc Gyneco said::
>>>>
>>>> > To alter her triple-screen risk you must be sure about statistical
>>>> > independance between the serum markers and the hyperechoic bowel
marker.
>>>> >
>>>> > In my opinion, there is no study in this issue.
>>>> >
>>>> > So, you can't modified the triple screen risk.
>>>> >
>>>> > And the likelihood ratio could be apply to her age related risk.
>>>>
>>>> Yes, but in UOG 2000;16:402-406, Verdin and all analyse the value of
the
>>>> second trimester ultrasound examination among the women whose fetuses
were
>>>> indicated to be at low risk for chromosomal anomalies on the basis of
both
>>>> first trimester nuchal translucency measurement and second trimester
>>>> biochemical screening
>>>>
>>>> LLR for echogenic bowel is 283 (71 -1133) 95% IC with 2 abnormal
karyotype
>>>> for five fetuses at low risk
>>>>
>>>> But the study numbers are relatively small (5500 pregnacies) and 3548
>>>> considered negative after bichemical screening (7 abnormal karyotype in
this
>>>> group)
>>>> ¨
>>>> Ph Coquel,MD
>>>>
>>>> Annecy, France
>>>>
>>>--------------------------------------------------------------------
>>>David A. Nyberg, M.D !!!!
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>> !US!
>>>
>>>--
>>>--------------------------------------------------------------------
>>> --
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>>--------------------------------------------------------------------
>>> / \
>>>--------------------------------------------------------------------
>>> // \\
>>>--------------------------------------------------------------------
>>> /. \ \
>>>--------------------------------------------------------------------
>>> / / \ \\
>>> / . \ .\
>>> /. .\
>>> / // \
>>> / ( ..) \
>>> / o \
>>> / --( )-- \
>>> / ( ) \
>>> / / \ \
>>> _/ /_
>>>
>>> ............................
>>>
>>--
>>art fougner, md
>>
>>A series of 1000 cases begins with but a single anecdote.
>>
>
>--
>Richard M. Roberts, PhD, MD, FACMG
|
|
