| a. | the risk of miscarriage is 5%. | False |
| b. | if chromosomes are cultured, the diagnosis is 100% accurate. | False |
| c. | should be offered routinely to women over the age of 37 years. | False |
| d. | is the preferred method of investigating a raised serum AFP result. | False |
| e. | can be used to assess the severity of Rhesus disease. | True |
| f. | has largely been superseded by chorionic villus biopsy. | False |
go to the bottom of the page and other links
This comes up frequently in both MRCOG and DRCOG.
In real life, GPs need to be able to discuss what is done, the risks and the possibility of a false result.
MRCOG candidates will be expected to know more and be able to deal with the topic in a role-play or viva.
return to the list of contents
go to the bottom of the page and other links
FISH: fluorescence in-situ hybridisation
GTG: RCOG's Green-top Guideline
MSAFP: maternal serum alpha-feto protein
SOG: Society of Obstetricians and Gynaecologists of Canada
return to the list of contents
go to the bottom of the page and other links
MRCOG candidates should be know the RCOG's GTG 8.
They would benefit from reading the advice from the Canadian SOG.
return to the list of contents
go to the bottom of the page and other links
Amniocentesis involves sticking a needle through the abdominal and uterine walls.
Local anaesthesia is not used.
Amniotic fluid is aspirated
It contains viable cells from the fetus.
These cells can be cultured.
This relates to the fact that they can be grown, not to them being particularly sophisticated, though the idea appeals.
This enables identification of chromosome problems and some genetic disorders.
Full chromosome culture takes two - three weeks.
A rapid count of specific chromosomes can be done using FISH.
This gives a result within 48 hours.
But results are not always obtained.
Nor are they 100% accurate.
no liquor may be obtained,
cell culture fails in a small % of cases,
there is a tiny risk that maternal cells are grown, not cells from the fetus.
Amniocentesis is sometimes done for evidence of fetal infection.
In rare cases it is used for chemical analysis.
The best known is bilirubin in cases of Rhesus haemolytic disease.
Amniocentesis is done under continuous ultrasound control.
This reduces maternal and fetal risks and improves the success rate of obtaining fluid.
The main disadvantage is the risk of miscarriage.
This is about 1 in 100, the figure you should remember for the exam.
When counselling in real life, you have to point out that the fetus that is lost is almost always normal.
Babies need adequate amounts of liquor to develop.
Chronic loss of liquor can occur after amniocentesis, leading to limb abnormalities.
Fetal lungs need liquor to "breathe" and grow in-utero.
If there is insufficient, lung hypoplasia can result.
Care must be taken to make amniocentesis as sterile as possible to reduce the risk of intra-uterine infection.
Amniocentesis can be used therapeutically:
to remove excess liquor - amnioreduction, e.g., in acute hydramnios in twin-twin transfusion,
to instil liquor when there is insufficient: amnioinfusion.
Anti-D immunoglobulin must be given if appropriate.
Amniocentesis in multiple pregnancy is tricky and should be done by a fetal medicine specialist.
return to the list of contents
go to the bottom of the page and other links
Expanded information for the MRCOG.
immediate
delayed
return to the list of contents
go to the bottom of the page and other links
The great majority are done for chromosome culture, particularly exclusion of T21.
Some will be done to look for evidence of fetal infection.
Some will be done for biochemical tests:
The most common of these is bilirubin assay in cases of Rhesus iso-immunisation.
return to "expanded information for the MRCOG"
Amniocentesis is usually done from 15 weeks.
"Early" amniocenteses are those done before 15 weeks.
CEMAT, the Canadian Early and Mid-trimester Trial reported on this topic.
Women were randomly allocated to early (11 - 12+6 weeks) or mid-trimester (15 - 16+6) weeks) amniocentesis.
2183 women had amniocentesis earlier than 13 weeks,
2185 women had amniocentesis after 15 weeks.
The conclusions were that: "early amniocentesis is associated with an increased risk of fetal loss and talipes equinovarus".
| Early amniocentesis | Mid-trimester amniocentesis | |
| Total miscarriage rate | 7.6% | 5.9% |
| Talipes equinovarus rate | 1.3% | 0.1% |
| Fluid leakage | 3.5% | 1.7% |
The RCOG's advice in the GTG is that early amniocentesis should only be done:
in exceptional circumstances,
and after full discussion of the possible risks.
return to "expanded information for the MRCOG"
The GTG advises:
formal consent before the procedure is done,
following the guidelines on consent from the GMC and RCOG,
using DOH consent form 3.
It states that the following should be included in the information provided:
the results that the test can provide,
this includes what a +ve test means,
what a -ve test means,
possible problems:
failure of the cell culture to grow,
contamination by maternal cells,
information about analysis & storage of the sample in the cytogenetics laboratory,
the success rate of chromosome culture in the local laboratory,
who will do the test,
how their performance is monitored,
the risks of the tests:
miscarriage,
direct fetal and maternal injury,
long-term liquor leakage and its consequences,
with both local and national figures, particularly for miscarriage,
how the test is done,
when the results will be available,
if appropriate, information should be given about FISH,
how the patient will be given the results,
what will be done in the event of a +ve test or one that has not yielded a result.
This is a lot of information to remember for both the person giving it and the person trying to take it in,
It makes good sense to have a patient information leaflet that provides all the detail.
The GTG stresses the need for full documentation in the notes of the counselling and the consent if written materials are not used.
return to "expanded information for the MRCOG"
return to "expanded information for the MRCOG"
return to "expanded information for the MRCOG"
The main immediate risks are:
miscarriage,
intra-uterine infection, which usually causes miscarriage,
direct injury.
return to "expanded information for the MRCOG"
The main delayed risks are:
chronic leakage of liquor
causing
lung hypoplasia: the fetus needs to "breathe" liquor to develop its lungs,
limb deformities: low liquor volumes lead to positional abnormalities,
intra-uterine infection leading to miscarriage,
miscarriage or premature labour.
alloimmunisation:
the mother develops anti-D antibodies,
this is unlikely to be serious in the first affected pregnancy,
and should be prevented by proper use of anti-D.
return to "expanded information for the MRCOG"
In 1964 Geoghan et al reported a maternal death from amniotic embolism after amnioreduction for gross hydramnios.
J Obstet Gynaecol Br Commonw. 1964 Oct;71:673-80.
Bell et al reported one maternal death from amniotic fluid embolism in a series of 2000 amniocenteses.
Bell et al. Med J Aust. 1987 Jan 5;146(1):12-5.
There have been a number of reports of maternal deaths from sepsis.
A case was reported in the "Why Mothers Die", the 5th. Maternal Mortality Report, 1997 - 1999.
Another was reported in "Why Mothers Die", the 6th. Maternal Mortality Report, 2000 - 2002: page112 - 113.
These are thought to be due to bowel perforation, which may be more likely after lower abdominal surgery.
And less likely with continuous ultrasound control.
Thorpe et al recorded a case due to sepsis, but with no evident bowel injury.
Obstet Gynecol. 2005 May;105(5 Pt 2):1213-5.
So, maternal death occurs, but is rare; "extremely rare" according to the 2000 - 2002 Maternal Mortality Report.
There are also loads of psychological aspects, as you might expect.
There are numerous papers reporting anxiety, such as that by Sarkar et al.
Prenat Diagn. 2006 Jun;26(6):505-9.
And the fact that women tend to have an exaggerated impression of the risk of fetal abnormality.
Tercyak et al. Patient Educ Couns. 2001 Apr;43(1):73-84.
return to "expanded information for the MRCOG"
return to "expanded information for the MRCOG"
The RCOG's GTG
return to "expanded information for the MRCOG"
The GTG says that:
Rhesus group should be known in every case,
anti-D "must be offered" .... in line with national recommendations".
The national recommendations reffered to are in Green-top guideline No 22, "Anti-D Prophylaxis" of May 2002.
It states that:
250i.u. is recommended for prophylaxis following sensitising events up to 20 weeks,
at least 500 i.u. anti-D Ig should be given,
followed by a test to identify FMH greater than 4ml red cells, in which case more anti-D is needed.
In practice you are going to give 500 i.u. after 20 weeks while awaiting the result of the Kleihauer test.
Most hospitals use the Kleihauer test.
But the GTG states that flow cytometry is more accurate.
It says that flow cytometry may be best used when the intial Kleihauer test suggests a large FMH.
And that the rosetting technique may also be useful.
The injection is given into the deltoid, not the gluteal muscle.
Injections into fat bottoms may end up in the subcutaneous tissue, not the muscle.
And this delays absorption.
The injection should be given within 72 hours.
But may be effective if given up to 10 days after the potentially sensitising event, so should still be given.
Obviously we are talking about non-sensitised Rhesus negative women.
The GTG also mentions RhDu women.
Du is a variant of "big D".
I first met it as a very junior doctor.
A woman had conflicting results for the Rhesus status on tests taken at different times.
One had made her Rh -ve, another Rh +ve.
It emerged that she had Du.
It is a weak antigen which reacts with some anti-D sera, but not with all.
So the woman could be found to be Rh -ve on one occasion and Rh +ve on another.
The important thing for us is that she is treated as though she had normal "big D".
If she needs blood products, she gets the Rh +ve variety.
From the above, it is clear that she does not develop anti-D when exposed to Rh +ve products.
So she does not need anti-D prophylaxis.
return to "expanded information for the MRCOG"
The GTG stresses the need for audit.
It describes it as the best means of assessing competence.
The need to be able to provide local figures for miscarriage rates etc. is outlined in the "consent" section.
The GTG recommends continuous audit of things like the miscarriage rate and the "need for second insertion".
return to "expanded information for the MRCOG"
return to the list of contents
go to the bottom of the page and other links
MRCOG candidates should be familiar with the RCOG's GTG 8.
They would benefit from also reading the advice from the Canadian SOG.
The RCOG has advice for women on its website.
This lists the risks:
miscarriage: 1%
serious infection: < 1 in 1,000
blood-stained liquor: < 1%
It doesn't explain the significance of a "bloody tap".
It also says that operators doing amniocentesis should do "at least 10 each year".
This is at odds with its own "Greentop" guideline, which suggests at least 50.
Most authorities quote a risk of pregnancy loss of between 1 in 100 and 1 in 200.
The RCOG's GTG 8 from 2005 suggests 1%.
And stresses the need to justify lower figures.
1% is the figure from the only trial in which women were randomised to have or not to have amniocentesis.
This was by Tabor et al in the Lancet in June 1986.
1.7% of those who had amniocentesis miscarried, compared with 0.75 of those who only had ultrasound scanning.
They noted an increased risk of miscarriage with:
raised maternal serum AFP,
perforation of the placenta,
discoloured amniotic fluid
Subsequent papers have tried to deduce the "procedure-associated risk".
But this is more problematic: those who have amniocentesis are not identical to those who do not.
But you are not going to be asked to elucidate the arguments!
Eddleman et al in 2006 suggested the risk is much less.
A remarkable 6 in 10,000 in 3096 tests.
This was pretty sensational, suggesting that the risk was 1 tenth of what had been described.
And it might be quoted to you by patients who are adept at finding information on the Internet.
Eddelman's paper has been criticised for having flawed deductions about "procedure-associated risk" causing miscalculation of the real risk.
And Gaudry et al in 2008 found a loss rate of 1.4% in 4858 tests, an even higher figure than many previous series.
There is a good discussion paper on the website of the Canadian SOG.
Early amniocentesis, before 14 weeks has been tried.
The CEMAT study, published in 1998, randomly allocated women having amniocentesis to:
early procedures at 11- 15 weeks
or routine procedures at 15 - 17 week.
They found an increased risk of miscarriage and talipes.
Philip et al from the NICDH EATA Trial Group compared early amniocentesis and abdominal CVS in a study published in 2004
They found an increased risk of talipes.
And a probable increased risk of miscarriage.
So the early procedures have gone out of fashion.
There has been evidence that operator experience is a big factor in the risk.
Blessed et al found a 6.8-fold difference between occasional and frequent operators.
The generalists had loss rates of 1 in 46 compared with 1 in 312 for the "perinatologists".
Welch et al also found that contamination of the specimen with maternal cells was more likely with the occasional operator.
This compared with 1 in for the more regular operators.
The RCOG's "Greentop" guideline from 2005 doesn't lay down the law.
It "suggests" that those doing amniocentesis:
should perform at least 50 per annum
but says that they must have been formally trained
There is also a tiny risk to the mother.
If the needle goes through the bowel, she could get peritonitis.
A maternal death was noted in the 1997-1999 enquiry into maternal death.
And another in the 2000-2002 report, "Why mothers die".
If chromosomes are cultured, the accuracy is almost 100%, but there is scope for error.
Laboratory error may occur and in one test in about every thousand, the cells come from the mother.
Usually fetal cells are present too, so that two separate cell types grow, but not always.
The last case I had of this problem was in a doctor's wife, needless to say.
There was consternation all round when she delivered a healthy boy after amniocentesis showing normal female chromosomes.
There is also the rare possibility that mosaicism exists.
I.e. the baby has two different cell
types and only one
cell line is identified.
Hospitals should give this general information, but also the local figures.
Ideally the figures should be for the person doing the test.
Abnormal AFP levels are dealt
with in MCQ 1, question 1.
Older women should be told about all screening methods, including the triple test.
They
should not be railroaded into
amniocentesis. (See also MCQ6, question 24.)
Babies affected by Rhesus disease have a haemolytic anaemia.
Bilirubin levels rise and some of this finds its way into the liquor.
So the severity of the process can be gauged by
amniocentesis – see also the answer to question
17 below.
An interesting side issue is middle cerebral artery blood flow.
In severe anaemia, the baby tries to spare its brain from hypoxia.
The blood flow through the middle cerebral arteries increases.
This can be detected by scan.
| Return to MCQ Paper 4 |
| DRCOG Page |
| MRCOG Page |