ORIGINAL ARTICLE

Uptake of invasive prenatal tests in pregnancies conceived via assistedreproductive technologies: the experience in Queensland, AustraliaLauren Hunt1,2*, Madelyn Peterson2,3, Stephen Sinnott3, Bridget Sutton3, Robert Cincotta4, Gregory Duncombe4,5, Jackie Chua4 andAideen McInerney-Leo6

1Genetic Health Queensland, Royal Brisbane and Women’s Hospital, Brisbane, Australia2School of Biomolecular and Physical Sciences, Griffith University, Brisbane, Australia3Specialised Obstetric & Gynaecological Imaging (so + gi), South Bank, Brisbane, Australia4Queensland Ultrasound for Women, Brisbane, Australia5Department of Obstetrics and Gynaecology, University of Queensland, Brisbane, Australia6University of Queensland Diamantina Institute, Brisbane, Australia*Correspondence to: Lauren Hunt. E-mail: lauren.hunt@health.qld.gov.au

ABSTRACTObjective Prenatal diagnosis of a chromosomal abnormality currently involves the use of an invasive procedure,which has a risk of fetal loss. The aim of this study was to identify whether pregnancies conceived via assistedreproductive technologies were more or less likely to be subjected to an invasive procedure compared withpregnancies that were conceived spontaneously.

Method Population data were collated from three private ultrasound clinics across southeast Queensland, Australia.

Results Of the 15032 spontaneously conceived pregnancies, 775 (5.2%) had invasive testing, while 95 (6.0%) of the 1581pregnancies conceived through assisted reproductive technologies had invasive testing. When the uptake of testing isadjusted by the maternal age the assisted reproductive population was significantly less likely to pursue invasive testing(p=0.003). Similarly when adjusted for the combined first trimester screen risk estimate, the assisted reproductionpopulation is significantly less likely to undergo invasive testing than the spontaneous population (p=0.005).

Conclusion Pregnancies conceived using assisted reproductive technologies are significantly less likely to be subjectedto invasive testing than pregnancies conceived spontaneously in women of the same age and combined first trimesterscreen risk. © 2012 John Wiley & Sons, Ltd.

Funding sources: NoneConflicts of interest: None declared

INTRODUCTIONDiagnostic testing via an invasive procedure, such as chorionicvillus sampling (CVS) or amniocentesis, is currently the onlyway a chromosomal abnormality can be confirmed prenatally.Both procedures are quoted (in Australasia) as having amiscarriage risk of approximately 1%.1 Because of the associatedrisk of miscarriage, these tests are only offered to women whosepregnancies are considered to be at high risk for aneuploidy. Thecombined first trimester screen (cFTS) is the most sensitivenoninvasive clinical method of identifying a fetus as high riskfor chromosomal abnormality.2 At the time of this study, a highrisk (screen-positive) cFTS result indicated a risk between 1 in 2and 1 in 300. The consequences of a screen-positive resultinclude the offer of an invasive test, with risk of fetal loss becauseof the procedure, and psychological distress. With the cFTS thereis an expected screen-positive rate of 5%, which means that 1 in20 pregnancies screened will be placed in the high risk category.3

Pregnancies conceived via assisted reproductive technologies(ART) are sometimes considered ‘precious’ because of relateddifficulty in conception.4,5 The rate of achieving a pregnancyfrom one assisted reproductive cycle is approximately 23%,and the live birth rate per assisted reproductive cycle isestimated to be 17%.6,7 Because of the modest success rate,many patients require multiple cycles before having a healthybaby, and some will never have a successful outcome.8 Thenumber of clinical pregnancies resulting from assistedconception in Australia and New Zealand has increasedsignificantly in recent years. At present, 1% to 3% of infantsborn in all developed countries are conceived throughART.9,10 The types of assisted reproductive technologiesthat are included in this study are in vitro fertilisation,intrauterine insemination, intracytoplasmic sperm injection(ICSI), gamete intrafallopian transfer and medications thatstimulate ovulation.

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

DOI: 10.1002/pd.3953

We hypothesized that women carrying pregnancies conceivedvia ART would be less likely to proceed to a diagnostic testbecause of the ‘precious’ nature of these pregnancies and therisk of miscarriage associated with an invasive procedure.This study aimed to identify if there was a difference in thefrequency of diagnostic testing between those who conceivedspontaneously and those who conceived via ART.

METHODOLOGYData related to prenatal screening and diagnosis were collatedfrom three private ultrasound clinics in southeast Queensland,Australia. All centres use a commercial database, Viewpoint,(General Electric Healthcare, Waukesha, United States ofAmerica) to generate risks for chromosomal abnormality afterfirst trimester screening. The databases were searched toidentify pregnancies that had had first trimester screening, todetail method of conception, risks generated throughscreening and uptake of invasive testing. Pregnancies withmultiple fetuses were excluded because of the additionalcomplications associated with risk assessment and prenataldiagnosis. Any fetus identified with chromosomal anomalieswas removed from the data set because other risk factorsmay have been identified on ultrasound, which could haveaffected the decision to test.

The data were analysed using Predictive Analytics SoftwareStatistics (PASW) GradPack 17.0, part of Statistical Package forthe Social Sciences (SPSS) Inc. (Chicago, USA). The findingswere considered significant when the p value was less than 0.05(p< 0.05). A chi-square test was used to compare the uptake ofinvasive tests in the ART and spontaneous conceptionpopulations. Binary logistic regression analyses calculated theodds ratio (OR) that a pregnancy conceived through ART wouldhave invasive testing, using the spontaneous conceptionpopulation as a reference. The ORs were adjusted by two variables

independently of each other. Both maternal age and the cFTS riskindependently influence the decision to undergo an invasive test.

Ethics approvalApproval was granted from the Griffith University HumanResearch Ethics Committee, protocol number BPS/01/09/HREC.

RESULTSA total of 17 889 women had combined first trimester screeningat one of three centres between January 2004 and September2009. Of these, 1157 multiple pregnancies and 119 fetuses withchromosomal anomalies were excluded. The final populationof 16 613 singleton fetuses included 1581 (9.5%) assistedconceptions.

Significant demographic difference when compared withspontaneous conceptions was an increase in maternal age(Table 1).

Eight hundred and seventy (5.2%) women had an invasiveprenatal test, 775 (5.2%) of those conceived spontaneouslyand 95 (6.0%) of those conceived through ART. Multivariatelogistic regression that accounts for maternal age and the riskgenerated through combined first trimester screening showsthat women who have had ART are significantly less likely tochoose to proceed with an invasive test (Table 2, Figure 1).

DISCUSSIONOn face value, invasive testing rates appear to be higher in theART population (6.0% vs 5.2%) but adjustment for maternalage and the risk generated through combined first trimesterscreening shows that women who have conceived throughART are significantly less likely to choose to proceed withinvasive testing. This supports previous studies that found thatthe personalised risk from combined first trimester screening isthe primary consideration in deciding upon invasive testing.11

These findings support the hypothesis that women who haveconceived through ART are less likely to proceed with adiagnostic test after combined first trimester screening. Thisis likely due to the clients’ heightened perception of the riskof miscarriage associated with an invasive procedure.4,5 It isunsurprising that the ART group would be more reluctant torisk miscarriage from an invasive procedure given this group’sunderlying difficulty with conceiving. In some cases thepregnancy is highly probably, the pregnant woman’s onlychance to have a child of her own and therefore particularly‘precious’.5 Long-standing infertility or difficulty conceiving

Table 1 Population demographics

Spontaneous ART

N 15032 1581 (9.5%)

Maternal age (years) 32 (15–50) 34 (20–56) (p<0.001)

Maternal weight (kilograms) 64.5 (36–162) 65 (40–130)

Ethnicity 87% Caucasian 91% Caucasian

Smoking history 3% smokers 1% smokers

Table 2 Analysis of uptake of an invasive test with multivariate logistic regression

Multivariate logistic regression analysis adjusting for

CVS or amniocentesis Maternal Age cFTS Risk

n n % AdjOR 95% CI p-value AdjOR 95% CI p-value

All conceptions

Spontaneous 15032 775 5.16 Reference Reference

ART 1581 95 6.01 0.701 0.55–0.89 0.003 0.718 0.57–0.91 0.005

L. Hunt et al.

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

can evoke intense emotional responses4 and so consequentlyexperiencing pregnancy loss could potentially lead to severepsychiatric responses.12 Hence, they may not be willing totake any additional risks, such as invasive testing duringthe pregnancy.

This study has the third largest cohort known to date.Amor et al.13 in Victoria, Australia similarly found a higherproportion of women who conceived using ART (10.6%) hadinvasive testing compared with those who conceivedspontaneously (5.3%). Amor et al.13 had a population including1793 women who conceived via ART. When Amor et al.13

adjusted their data to account for maternal age the resultsdiffered from ours, with the ART population 1.24 times morelikely to undergo an invasive procedure.13 They went on tosimultaneously adjust their data for maternal age and thecFTS result which showed the ART population was 0.78times less likely to undergo invasive testing than thespontaneous population.

A Danish study of 8531 women who conceived via ARTfound that the invasive testing rate in this cohort wassignificantly lower than the general Danish population.14

Meschede et al.15 studied a group of 107 women whoconceived via ICSI and found that this group was less willingto undergo an invasive procedure than women who hadconceived spontaneously. Geipel et al.16 confirmed thehypothesis that women who conceived via ICSI favoured anoninvasive prenatal testing approach in a cohort of 153women. Schover et al.17 found in a sample of 55 couples thatthe risk of miscarriage associated with invasive testing stronglyinfluenced the decision to undergo prenatal invasive testing.One study however, found no difference in the acceptance rateof invasive testing among women who conceived through ARTcompared with the general population.18

The data collated for this study was collected fromthree private ultrasound clinics and the population may

therefore not reflect the general population seen in a publichealthcare system. Most women undergoing assistedreproduction have private care in Queensland, given thatin vitro fertilisation, intrauterine insemination, ICSI andgamete intrafallopian transfer are not available in theQueensland public healthcare system. Therefore, thiscontributes to the high prevalence of assisted reproductiveconceptions in this cohort.

The Royal Australian and New Zealand College ofObstetricians and Gynaecologists recommend that allwomen, regardless of their age, are offered the cFTS, whichmay lead to an offer of an invasive test.3 All patients in thisstudy received individual comprehensive counselling byeither a genetic counsellor, fetal medicine specialist orspecialised radiologist prior to any prenatal screeningor invasive test. This counselling enables patients tounderstand the benefits, risks and limitations of bothscreening and invasive tests so an informed decision couldbe made. All the clinics reported that the vast majority ofpatients will undergo the cFTS initially and make decisionsregarding an invasive test based on that result. Furtheranalysis of this cohort found that the cFTS false-positiverate, when adjusted for maternal age, was not significantlydifferent between the assisted and spontaneously conceivedpopulations.19 There was one exception, the false positiverate was significantly increased (p = 0.012) when conceptionincluded ICSI and the maternal blood serum was analysedby the KRYPTOR platform.19

Pregnancies conceived via ART are not homogeneous – but itwas not possible to examine subgroups because of the lownumbers involved – which would have reduced the power ofthe study. It would be interesting to continue to investigatewhether the type of treatment, number of cycles necessary toconceive and/or the reason for using ART had a directrelationship to the probability of pursuing invasive testing.

Figure 1 The likelihood that spontaneously conceived pregnancies and pregnancies conceived via ART undergo an invasive procedure. Thefirst pair of columns represents the odds ratio of the percentages calculated from the raw data. The second pair of columns represents the oddsratio when adjusted for maternal age. The third pair of columns represents the odds ratio when the combined first trimester screen risk isadjusted for. When the odds ratio is adjusted for maternal age or the combined first trimester screen risk, the likelihood that the ART populationwill undergo invasive testing is significantly reduced in comparison with the spontaneous conception population

Uptake of invasive testing by women who conceived via ART

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.

Unfortunately, some demographic variables such as educationlevels and parity were not examinable from the data collected.

Women that have had ART are less likely to proceed toinvasive testing after combined first trimester screening. Thisis consistent with our clinical impression of these beingparticularly ‘precious’ pregnancies and this study reminds usthat risk alone does not dictate uptake of invasive testing.Rather, there are a myriad of factors that weigh into suchdecisions and it is our roles as clinicians to help the clientidentify and reconcile these factors so that they can make thebest decision possible for them.

ACKNOWLEDGEMENTThank you to Dr Paula Sivyer from Diagnostic Imagining forWomen, Brisbane, Australia for donating data to this study.

WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?

• Prenatal diagnosis of chromosome abnormalities currently involvesan invasive procedure, which has risk of fetal loss.

• Mixed results have been seen for the likelihood that women whohave conceived via assisted reductive technologies elect forinvasive testing.

WHAT DOES THIS STUDY ADD?

• This study provides further evidence that pregnancies conceivedusing assisted reproductive technologies are significantly lesslikely to be subjected to invasive testing than pregnanciesconceived spontaneously.

• Third largest cohort known to date.

REFERENCES1. Royal Australian and New Zealand College of Obstetricians and

Gynaecologists. A Decision Aid: Testing in Pregnancy for fetalabnormalities 1 ed. [Brochure]. Victoria: Mi-tec Medic Pty. Ltd. Fundedby National Health and Medical research Council; 2008.

2. Nicolaides K, Spencer K, Avgidou K, et al. Multicenter study of first-trimester screening for trisomy 21 in 75 821 pregnancies: results andestimation of the potential impact of individual risk-orientated two-stage first-trimester screening. Ultrasound Obstet Gynecol 2005;25(3):221–6.

3. Joint Committee of the Royal Australian and New Zealand College ofObstetricians and Gynaecologists and Human Genetics Society ofAustralasia. Prenatal screening tests for trisomy 21 (Down syndrome),trisomy 18 (Edwards syndrome) and neural tube defects. Victoria: RoyalAustralian and New Zealand College of Obstetricians andGynaecologists, 2007. College Statement C-Obs4.

4. Shalev J, Meizner I, Rabinerson D, et al. Elective CytogeneticAmniocentesis in the Third Trimester for Pregnancies with High RiskFactors. Prenat Diagn 1999;19(8):749–52.

5. Harris J. Precious Fertility and Third-trimester Tests. Prenat Diagn1999;19(8):753–4.

6. Gleicher N, Weghofer A, Barad D. A formal comparison of the practice ofassisted reproductive technologies between Europe and the USA. HumReprod 2006;21(8):1945–50.

7. Yli-Kuha A, Gissler M, Luoto R, Hemminki E. Success of infertilitytreatments in Finland in the period 1992–2005. Eur J Obstet GynecolReprod Biol 2009;144(1):54–8.

8. Malizia B, Hacker M, Penzias A. Cumulative Live-Birth Rates after InVitro Fertilization. N Engl J Med 2009;360(3):236–43.

9. Klemetti R, Gissler M, Hemminki E. Comparison of perinatal health ofchildren born from IVF in Finland in the early and late 1990s. HumReprod 2002;17(8):2192–8.

10. Wright VC, Chang J, Jeng G, Macaluso M. Assisted reproductivetechnology surveillance – United States, 2005. MMWR: Surveill Summ.2008;57(5):1–23.

11. Nicolaides K, Chervenak F, McCullough L, et al. Evidence-basedobstetric ethics and informed decision-making by pregnant womenabout invasive diagnosis after first-trimester assessment of risk fortrisomy 21. Am J Obstet Gynecol 2005;193(2):322–6.

12. Lazarus RS, Folkman S. Stress Appraisal and Coping. New York: SpingerPublishers, 1984.

13. Amor D, Xu J, Halliday J, et al. Pregnancies conceived using assistedreproductive technologies (ART) have low levels of pregnancy-associated plasma protein-A (PAPP-A) leading to a high rate of false-positive results in first trimester screening for Down syndrome. HumReprod 2009;24(6):1330–8.

14. Gjerris A, Loft A, Pinborg A, et al. Prenatal testing among womenpregnant after assisted reproductive techniques in Denmark 1995–2000:a national cohort study. Hum Reprod 2008;23(7):1545–52.

15. Meschede D, Lemcke B, Stussel J, et al. Strong preference for non-invasive prenatal diagnosis in women pregnant throughintracytoplasmic sperm injection (ICSI). Prenat Diagn 1998;18(7):700–5.

16. Geipel A, GembruchU, LudwigM, et al. Genetic sonography as the preferredoption of prenatal diagnosis in patients with pregnancies followingintracytoplasmic sperm injection. Hum Reprod 1999 Oct;14(10):2629–34.

17. Schover L, Thomas A, Falcone T, et al. Attitudes about genetic risk ofcouples undergoing in-vitro fertilization. Hum Reprod 1998;13(4):862–6.

18. Elimian A, Demsky M, Figueroa R, et al. The influence of IVF, multiplegestation and miscarriage on the acceptance of genetic amniocentesis.Prenat Diagn 2003;23(6):501–3.

19. Hunt L, McInerney-Leo A, Sinnott S et al. First-trimester ScreenBiochemistry: Comparison of pregnancies conceived spontaneouslyand via assisted reproduction in Queensland, Australia. (Unpublished).

L. Hunt et al.

Prenatal Diagnosis 2012, 32, 1–4 © 2012 John Wiley & Sons, Ltd.