ORIGINAL ARTICLE

Longer time-to-pregnancy in spontaneously conceived pregnanciesis associated with lower PAPP-A and free β-hCG in first trimesterscreening for Down syndromeI. Kirkegaard1*, N. Uldbjerg1, A. Tabor2,3 and T. B. Henriksen4

1Department of Obstetrics and Gynecology, Aarhus University Hospital, Aarhus, Denmark2Center of Fetal Medicine, Rigshospitalet–Copenhagen University Hospital, Copenhagen, Denmark3Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark4Department of Pediatrics, Aarhus University Hospital, Aarhus, Denmark*Correspondence to: Ida Kirkegaard. E-mail: ida.kirkegaard@ki.au.dk

ABSTRACTObjective The aim of this study was to investigate whether subfertility, measured as longer time-to-pregnancy (TTP)in spontaneously conceived pregnancies, affects the first trimester levels of pregnancy-associated plasma protein-A(PAPP-A) and free beta-human chorionic gonadotrophin (β-hCG) and hence the risk estimates in Down syndromescreening.

Methods The study included a cohort of 10 469 singleton pregnant women who underwent first trimester combinedscreening and responded to a questionnaire regarding TTP. PAPP-A and free β-hCG levels were measured betweengestational week 8 + 0 and 13 + 6 and were related to TTP.

Results The median PAPP-A and free β-hCG MoMs were significantly lower in women with a TTP ≥24monthscompared with the reference group with a TTP <6months (PAPP-A: 0.96 vs 1.06MoM, p = 0.003; free β-hCG: 1.04 vs1.12MoM, p = 0.03). This led to an increased odds for trisomy 21 risk ≥1 : 300 for TTP ≥24months compared withTTP <6months, but when adjusting for potential confounders, the odds ratio (OR) lost significance (OR 1.4, 95%confidence interval; 0.8–2.4).

Conclusion Time-to-pregnancy ≥24months in spontaneously conceived pregnancies is associated with decreasedlevels of PAPP-A and free β-hCG. © 2013 John Wiley & Sons, Ltd.

Funding sources: The Aarhus Birth Cohort has been funded by Tryg Fonden, The Danish National Research Council, and Aarhus University Research Foundation.Conflicts of interest: None declared

INTRODUCTIONMedian levels of the serum markers used in the combinedfirst trimester screening for Down syndrome (DS), that is,pregnancy-associated plasma protein-A (PAPP-A) and freebeta-human chorionic gonadotrophin (β-hCG), have beenshown to be altered in pregnancies conceived after assistedreproduction technologies (ARTs).1 In particular, median levelsof PAPP-A have consistently been reported to be significantlylower in in vitro fertilized (IVF) pregnancies compared withspontaneously conceived pregnancies, varying from 0.78 to0.94MoM.1–6 The results related to levels of free β-hCG aremore conflicting, as they have been reported to vary from 0.94to 1.13MoM, with some studies finding significantly higherβ-hCG in IVF pregnancies compared with spontaneouslyconceived pregnancies, whereas others have found itsignificantly lover.1–6 Nuchal translucency (NT) does not seemto be affected by ART.1These altered levels of the serum

markers result in increased false positive rates, when applyingthe combined DS screening to IVF pregnancies, which is whythe MoM values of PAPP-A and free β-hCG are now routinelycorrected for mode of conception.7

The mechanisms behind this altered level of serum marker inIVF pregnancies are unknown. Thus, whether it is related to theinfertility or subfertility per se, or whether the procedures ormedication used in ART affects the levels is unclear. Time-to-pregnancy (TTP) in spontaneously conceived pregnancies is aclinical way of measuring a couples’ fertility, and longer TTP is anindication of subfertility.8 Studying TTP in spontaneously conceivedpregnancies is a possible way to study the influence of subfertility,without having the effect of the ART procedures interfering. In asmall study, it was suggested that pregnancies conceived after aTTP ofmore than 24months (n=70) had lower PAPP-A comparedwith pregnancies following a TTP of less than 12months,whereas free β-hCG and NT were not associated with TTP.9

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.

DOI: 10.1002/pd.4295

Noother previous studies have investigated the associationbetween TTP and serum marker levels.

Both PAPP-A and free β-hCG are produced by the trophoblastin the placenta, and it has been hypothesized that subfertilitymay affect placental function, perhaps resulting in differentlevels of these placental serum markers. Indeed, pregnancycomplications associated with placental dysfunction, for example,fetal growth restriction, preterm delivery, and preeclampsia, aremore frequent not only in ART-pregnancies10–12 but also inspontaneous pregnancies conceived with a prolonged TTP.12–15

The aim of this study was to explore if longer TTP affectedthe first trimester levels of PAPP-A and free β-hCG and toassess the effect of this potential association on the DSscreening performance, in a large cohort of pregnancies witha known TTP.

METHODSWe studied a cohort of women with singleton pregnancies,registered to give birth at Aarhus University Hospital, Skejby,Denmark. They all attended the prenatal screening programbetween January 2005 and December 2010. The prenatalscreening program has been offered free of charge to allpregnant women since 2004 and consists of a combinedbiochemistry and ultrasound first trimester screening test forchromosomal abnormalities. In the present analyses, onlywomen with complete information on the biomarkers PAPP-A and free β-hCG and the first trimester scanning wereincluded. Pregnancies with abnormal karyotype or majorabnormalities at the second trimester malformation scan wereexcluded from the study.

All pregnant women who accepted the first trimesterscreening test had a blood sample obtained at the first visit atthe general practitioner between gestational weeks 8 + 0 to13 + 6. The serum samples were analyzed at one centrallaboratory, and maternal levels of PAPP-A and free β-hCG weredetermined by Brahm’s Kryptor method and registered in thehospital’s electronic database of biochemical test results by aunique patient identifier. The serum values of PAPP-A and freeβ-hCG were converted to MoMs by expressing the absoluteconcentration relative to the median value of the gestationalage on the day of blood sampling. The MoM values were notcalculated until the gestational age was determined by thecrown-rump length at the first trimester scan. The MoM valueswere corrected for maternal weight.16 During some of the studyperiod, the MoM values were corrected also for mode ofconception, but in this study, we have used the uncorrectedvalues.

An ultrasound examination was performed at gestationalweek 11 + 3 to 13 + 6 where the NT thickness was measured.The fetal size was measured by the crown-rump length, andthe gestational age of the fetus was estimated using theformula of Robinson and Fleming.17 The gestational ageat serum sampling was then determined. All ultrasoundexaminations were performed by experienced sonographers,all certified for NT measurement by the Fetal MedicalFoundation, London, UK. Data regarding the analyses ofPAPP-A and free β-hCG and the information from theultrasound examinations were obtained from the Astraia

Database (http://www.astraia.com), which holds all theultrasound measures and risk estimates for clinical use in theprenatal setting linked to the women’s unique personalidentification number.

Information on TTP was provided from the Aarhus BirthCohort,18 which collects structured information on maternalcharacteristics [age, body mass index (BMI), and parity], lifestylefactors (smoking and alcohol consumption), obstetric andmedical complications during pregnancy and delivery, and stateof the newborn including gestational age and birth weight.Information is obtained by self-administered questionnairescompleted by the pregnant women around week 8 to 14 andthrough structured birth registration forms filled out by theattending midwife immediately after delivery. The completionrate of the questionnaire was 74.1%. In the questionnaire, thewomen were asked if the pregnancy was planned and for howlong the couple had been trying to conceive. They were alsoasked if they had received any fertility treatment and, if so, whatkind. Unplanned pregnancies were excluded. TTP was analyzedas a continuous variable and categorized into <6months(reference), 6 to 11months, 12 to 23months, and ≥24months.IVF pregnancies were used for comparison, whereas pregnanciesas a result of other ART (hormone stimulation, insemination)were excluded.

Information about birth weight and gestational age atdelivery was obtained from the Aarhus Birth Cohort. Wedefined small-for-gestational age (SGA) as birth weight belowthe fifth centile for gestational age and preterm delivery asdelivery before 37 completed weeks’ gestation.

Categorical data were compared using chi-squared test. Themedian MoM levels of PAPP-A and free β-hCG were comparedbetween TTP groups, using nonparametric tests (Mann–Whitney test). Logarithmic transformation of the MoM valueswas applied to achieve a Gaussian distribution, and the log10values were compared using t-test. The odds ratio (OR) forhaving a risk of trisomy 21≥1 : 300 was calculated using logisticregression analyses. We a priori decided to adjust for maternalage, parity, BMI, and smoking status.

Statistical significance was defined as a two-sided p< 0.05.The STATA statistical software package, version IC10, StataCorpLP, Texas, was used for all analyses.

Approval from the scientific ethical committee was notrequired, because a study based on compilation and statisticalhandling of clinical data with no individual identifiers isexempt by the local scientific ethical committee.

RESULTSOf the 18 809 women with singleton pregnancies, 4874 womenfailed to return the questionnaire, whereas 460 women did notprovide information on TTP, 2797 pregnancies were notplanned, and 209 pregnancies were conceived as a result ofhormone stimulation and insemination, and these were allexcluded. The final study population consisted of 10 469pregnancies, of which 962 were IVF pregnancies, and 9507spontaneous, planned pregnancies with information aboutTTP, distributed as shown in Table 1.

Maternal characteristics, lifestyle factors, and pregnancyoutcome in the different groups are shown in Table 2. Women

I. Kirkegaard et al.236

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.

with TTP of 6months or greater were older, had higher BMI,and were more often nulliparous and smokers duringpregnancy compared with the reference TTP. The gestationalage at delivery was lower in women with a longer TTP and asignificantly higher proportion of women delivered pretermcompared with the reference TTP. The birth weight of theinfants was also significantly lower in women with a longerTTP, and the risk of SGA less than fifth centile was increasedin women with TTP at 12 to 23months and TTP ≥24months,compared with the reference pregnancies.

We found that both PAPP-A and free β-hCG were significantlylower in women with a TTP ≥24months compared with thereference women with a TTP <6months (PAPP-A: 0.96 vs1.06MoM, respectively, p=0.003 and free β-hCG: 1.04 vs1.12MoM, p=0.03) (Table 3). There was no significant differencebetween PAPP-A and free β-hCG in women with a TTP at 6 to 11and 12 to 23months compared with the reference group, but atendency towards lower levels with longer TTP was seen forPAPP-A (test for trend, p=0.03) but not for free β-hCG(p=0.19). Significantly lower levels of PAPP-A and free β-hCG

Table 1 The study cohort and distribution of mode of conceptionand time-to-pregnancy, Aarhus Birth Cohort, 2004–2010,Aarhus, Denmark

N

Total cohort 18809

Excluded 8131

Did not complete questionnaire 4 874

Did not answer questions about TTP 460

Not planned pregnancy 2797

Hormone stimulation 209

IVF 962

Spontaneous, planned pregnancy 9507

TTP <6months 7 102 (74.7%)

TTP 6–11months 1 462 (15.4%)

TTP 12–23months 644 (6.8%)

TTP ≥24months 299 (3.1%)

TTP, time-to-pregnancy; IVF, in vitro fertilization.

Table 2 Maternal characteristics and pregnancy outcome in the subgroups of spontaneous pregnancies with different time-to-pregnancyand in in vitro fertilized pregnancies

TTP <6months (reference) TTP 6–11months TTP 12–23months TTP ≥24months IVF

n=7102 n=1462 n=644 n=299 n=962

Age ≥35 years, n (%) 891 (12.6) 241 (16.5)a 132 (20.5)a 78 (26.1)a 303 (31.5)a

BMI ≥30 kg/m2, n (%) 435 (6.1) 119 (8.1)a 71 (11.0)a 44 (14.7)a 92 (9.6)a

Nulliparous, n (%) 3549 (50) 846 (57.9)a 360 (55.9)a 169 (56.5)a 662 (68.9)a

Smoking, n (%) 264 (3.8) 70 (4.8) 54 (8.4)a 32 (10.8)a 39 (4.1)

GA at NT-scan (days) 89 89 89 90 90b

GA at double test (days) 66 67 67 67 67b

GA at delivery (days) 278 277b 276b 277b 277b

Preterm delivery (<37weeks), n (%) 404 (5.7) 93 (6.4) 61 (9.5)a 28 (9.4)a 88 (9.2)a

Birth weight (g) 3550 3530b 3480b 3490b 3460b

SGA less than fifth centile, n (%) 301 (4.2) 74 (5.1) 40 (6.2)a 22 (7.4)a 64 (6.7)a

TTP, time-to-pregnancy; IVF, in vitro fertilization; BMI, body mass index; GA, gestational age; NT, nuchal translucency; SGA, small-for-gestational age.Values are given as either n (%) or medians.ap<0.05 when comparing with the reference group using chi-squared test.bp<0.05 when comparing with the reference group using Mann–Whitney test.

Table 3 Medians and mean log MoM’s of pregnancy-associated plasma protein-A and free beta-human chorionic gonadotrophin inthe subgroups of spontaneous pregnancies with different time-to-pregnancy and in in vitro fertilized pregnancies

TTP <6months (reference) TTP 6–11months TTP 12–23months TTP ≥24months IVF

PAPP-A

Median MoM 1.06 1.06 1.03 0.96a 0.89a

Mean log10 MoM 0.05 0.05 0.04 �0.03b �0.14b

Free β-hCG

Median MoM 1.12 1.13 1.11 1.04a 1.02a

Mean log10 MoM 0.12 0.12 0.12 0.06b 0.04b

TTP, time-to-pregnancy; PAPP-A, pregnancy-associated plasma protein-A; β-hCG, beta-human chorionic gonadotrophin; IVF, in vitro fertilization.ap<0.05 when comparing TTP ≥24months and IVF with the reference group using Mann–Whitney test.bp<0.05 when comparing TTP ≥24months and IVF with the reference group using t - test.

Longer TTP is associated with decreased levels of PAPP-A and free β-hCG 237

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.

were also found in IVF pregnancies compared with referencewomen with a TTP <6months (p< 0.001). PAPP-A wassignificantly lower in IVF pregnancies compared with thepregnancies of women with a TTP ≥24months (0.89 vs0.96MoM, p=0.01), whereas free β-hCG did not significantlydiffer (1.02 vs 1.04MoM, p=0.51). The strength of theassociations studied did not change when adjusting the analysesfor maternal age, parity, and smoking status.

The proportion of women with a risk of trisomy 21 of 1 : 300or greater at the combined first trimester screening test wassignificantly increased in women with a TTP ≥24months(5.4% compared with 3.4% in the reference group, p = 0.05)and in IVF pregnancies (7.8% vs 3.4%, p< 0.001) (Table 4).Thus, the OR of having a trisomy 21 risk ≥1 : 300 was 1.6 [95%confidence interval (CI); 1.0–2.7] for TTP ≥24months, and itwas 2.3 (95% CI; 1.8–3.0) for IVF pregnancies compared withTTP <6months. The ORs decreased to 1.4 (95% CI; 0.8–2.4)for TTP ≥24months and to 2.0 (95% CI; 1.5–2.6) for IVFpregnancies, when adjusting for maternal age, parity, BMI,and smoking status (maternal age was the only factor changingthe estimate).

DISCUSSIONIn this large cohort study, we found that both PAPP-A and freeβ-hCG are significantly decreased in pregnant women with a longTTP (TTP≥24months) comparedwithwomenwith awaiting timeless than 6months. The levels were even lower in IVF pregnancies.Because of the lower levels of PAPP-A and free β-hCG inpregnancies with a prolonged TTP, the odds of an estimatedtrisomy 21 risk ≥1 : 300 was higher, although the increase lostsignificance for TTP ≥24months when adjusting for maternalage, whereas it remained significant for IVF conceptions.

This study has strengths and limitations: Our prenatalscreening program covers the entire population, as there areno other possibilities for DS screening in this area. Theprenatal screening program is free of charge, and the firsttrimester screening test attendance rate in 2007 was 96%.Accordingly, our study population is highly representativewithout oversampling of high-risk pregnancies. All theinformation, both in the Astraia database and in the AarhusBirth Cohort, was prospectively collected without knowledgeof the study in the clinical settings, which means that anymisclassification therefore is nondifferential. The data fromthe Aarhus Birth Cohort made it possible for us to take a widerange of potential explanatory variables into account, whichfurther strengthens our study.

In Denmark, ART is free of charge, and infertility treatment isoffered after trying to become pregnant for 12months withoutsuccess. Because of this, TTP >12months in spontaneouslyconceived pregnancies is relatively rare in Denmark, andtherefore, the large study population is a strength in our study.

Our medians for both PAPP-A and free β-hCG in thereference group are quite high compared with the expectedMoM of 1. A possible explanation could be that during thestudy period, the median used to calculate the MoM valuesin the Astraia System was not a median obtained from ourown population.

A limitation in our study might be that TTP was self-reportedduring early pregnancy and might have been affected by recallbiases. The validity of self-assessed TTP has been studiedpreviously, and it has been shown that self-reported TTP maybe influenced by prior obstetric history, family situation,contraceptive use, and sexual activity, among other variables.19

Other studies have found that validity of self-assessed TTP ishigh, particularly for current or recent pregnancies.20 Weexcluded unplanned pregnancies, as the self-assessed TTP isoften highly unpredictable in these pregnancies. However,excluding these pregnancies may underestimate our results,as unplanned pregnancies probably have a very short TTP.

With this study, we have confirmed the findings of the onlyprevious study investigating first trimester screening parametersin pregnancies with a prolonged TTP but in a population smallerthan ours.9 In agreement with that study, PAPP-A levels weresignificantly decreased in pregnancies with TTP ≥24months,but in contrast to the Finnish study, we also found levels offree β-hCG significantly decreased in pregnancies with TTP≥24months. A possible explanation for this discrepancy couldbe that we determined free β-hCG at a lower gestational age(average 9+4weeks). This timing may influence the associationwith TTP because the level of free β-hCG increases until10weeks’ gestation and then declines to a constant level from20 gestational weeks.21 Our results indicate that measuring freeβ-hCG earlier could result in a decreased level of free β-hCG inpregnancies with TTP ≥24months and in IVF pregnancies thanotherwise found.

We found PAPP-A levels decreased to 0.89MoM in IVFpregnancies, which is slightly higher compared with previousstudies, where the PAPP-A levels on average were around0.80MoM.1–6 We were unable to distinguish between freshand frozen embryo transfer, and previous studies have shownthat PAPP-A levels are only decreased in fresh and not in frozenembryo transfer, which might explain the higher PAPP-A levels

Table 4 Proportion of women with a trisomy 21 risk ≥1 : 300 and the odds ratio of having a trisomy 21 risk ≥1 : 300 in the subgroupsof spontaneous pregnancies with different time-to-pregnancy and in in vitro fertilized pregnancies

TTP <6months (reference) TTP = 6–11months TTP = 12–23months TTP ≥24months IVF

Trisomy 21 risk ≥1 : 300, n (%) 239 (3.4) 54 (3.8) 17 (4.2) 16 (5.4)a 72 (7.5)a

OR (95% CI) Reference 1.1 (0.8–1.5) 1.3 (0.8–1.9) 1.6 (1.0–2.7) 2.3 (1.8–3.0)

ORadjusted (95% CI)b Reference 1.1 (0.8–1.4) 1.2 (0.8–1.7) 1.4 (0.8–2.4) 2.0 (1.5–2.6)

TTP, time-to-pregnancy; IVF, in vitro fertilization; OR, odds ratio; CI, confidence interval.ap<0.05 when comparing TTP ≥24months and IVF with the reference group using chi-squared test.bAdjusted for maternal age, parity, BMI, and smoking status.

I. Kirkegaard et al.238

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.

in our IVF-group, because they consist of a mixture of freshand frozen embryo transfer. We also found free β-hCGsignificantly decreased in IVF pregnancies; in the literature, thisfinding has been controversial.1 Most previous studies foundno association between free β-hCG levels and IVF, whereas afew studies found free β-hCG to be decreased as we did, and afew studies even found free β-hCG increased.4,22 Again, apossible explanation might be the timing of the serum analyses,because the studies who found free β-hCG increased or notaltered had the serum samples drawn later, compared with ours.

The biological basis of altered serum markers in ARTpregnancies is unknown at present and is complex to investigate,because of the variety of etiologies for infertility, the diversity oftreatment methods, and difference in treatment response.Several causes have been suggested, such as hormone treatmentand ovarian hyperstimulation, impairment of early implantation,multiple corpora lutea and multiple implantation sites,vanishing twins and underlying pathology in the cause ofinfertility, and most likely, the alterations may be explained byseveral factors.1 Our results suggest that the mechanism behindalterations in the first trimester screening serum markers couldbe related to low fertility or to the cause of low fertility rather thanto the use of ART.

A few studies have investigated the changes in first trimesterscreening variables according to the underlying cause ofinfertility. Recently, a Turkish study compared first trimesterserum screening levels in IVF pregnancies with polycysticovary syndrome as the underlying infertility factor (IVF-P),with IVF pregnancies with male factor as infertility factor(IVF-H), and spontaneously conceived pregnancies. Theyfound no significant difference in PAPP-A levels but a tendencytowards lower levels in the IVF-H group compared with thetwo other. For free β-hCG, they found higher levels in IVF-H,compared with IVF-P.23 The study was small, but the resultsindicate that the alterations in serum levels may differaccording to the underlying infertility cause. Another smallstudy of first trimester serummarkers in pregnant women with

poor ovarian reserve, compared levels of PAPP-A and freeβ-hCG in normoresponders (less than six oocytes in oocyteretrieval) and poor responders (greater than or equal to sixoocytes at oocyte retrieval) and found no difference in serumlevels of PAPP-A and free β-hCG.24

Another explanation for the lower PAPP-Awith longer TTP couldbe erroneous pregnancy dating at the first trimester scan, as anunderestimation of the gestational age would result in decreasedPAPP-A levels. As there is an association between longer TTP andSGA, the underestimation could be due to the fetuses alreadybeing growth restricted at the time of the first trimester scan.

CONCLUSIONIn conclusion, we found significantly lower PAPP-A and freeβ-hCG levels in spontaneously conceived pregnancies with aprolonged TTP compared with those with a short TTP. Thedecreased serum marker levels however do not affect the oddsof abnormal results at first trimester combined screening fortrisomy 21 once maternal age is taken into account, unlikeIVF conceptions in which the changes in serum markersdouble the risk of abnormal screen results.

WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?

• First trimester pregnancy-associated plasma protein-A levels arelower in pregnancies conceived as a result of assisted reproductiontechnologies (ART), especially in vitro fertilization.

• Levels of free beta-human chorionic gonadotrophin may also bealtered in ART pregnancies.

• The mechanisms behind these alterations are unknown.

WHAT DOES THIS STUDY ADD?

• Prolonged time-to-pregnancy results in decreased levels ofpregnancy-associated plasma protein-A and free beta-humanchorionic gonadotrophin.

• Some of the lower values seen in ART pregnancies could be relatedto subfertility instead of infertility treatment.

REFERENCES1. Gjerris AC, Tabor A, Loft A, et al. First trimester prenatal screening in

women pregnant after IVF/ICSI. Hum Reprod Update 2012;18:350–9.2. Gjerris AC, Loft A, Pinborg A, et al. First-trimester screening markers are

altered in pregnancies conceived after IVF/ICSI. Ultrasound ObstetGynecol 2009;33:8–17.

3. Amor DJ, Xu JX, Halliday JL, 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;24:1330–8.

4. Bender F, Hecken J, Reinsberg J, et al. Altered first-trimester screeningmarkers after IVF/ICSI: no relationship with small-for-gestational-ageand number of embryos transferred. Reprod Biomed Online2010;20:516–22.

5. Engels MA, Kooij M, Schats R, et al. First-trimester serum markerdistribution in singleton pregnancies conceived with assistedreproduction. Prenat Diagn 2010;30:372–7.

6. Lambert-Messerlian G, Dugoff L, Vidaver J, et al. First- and second-trimester Down syndrome screening markers in pregnancies achievedthrough assisted reproductive technologies (ART): a FASTER trial study.Prenat Diagn 2006;26:672–8.

7. Kagan KO, Wright D, Spencer K, et al. First-trimester screening fortrisomy 21 by free beta-human chorionic gonadotropin and pregnancy-associated plasma protein-A: Impact of maternal and pregnancycharacteristics. Ultrasound Obstet Gynecol 2008;31:493–502.

8. Wilcox A. Fertility and Pregnancy: An Epidemiologic Perspective (1st edn).Oxford University Press: Oxford, UK, 2010.

9. Ranta JK, Raatikainen K, Romppanen J, et al. Increased time-to-pregnancyandfirst trimesterDown’s syndrome screening.HumReprod 2010;25: 412–7.

10. Källen B, Finnstrom O, Lindam A, et al. Trends in delivery and neonataloutcome after in vitro fertilization in Sweden: data for 25 years. HumReprod 2010;25:1026–34.

11. Ombelet W, Martens G, De Sutter P, et al. Perinatal outcome of 12,021singleton and 3108 twin births after non-IVF-assisted reproduction: acohort study. Hum Reprod 2006;21:1025–32.

12. Pinborg A, Wennerholm UB, Romundstad LB, et al. Why do singletonsconceived after assisted reproduction technology have adverseperinatal outcome? Systematic review and meta-analysis. Hum ReprodUpdate 2013;19:87–104.

13. Basso O, Baird DD. Infertility and preterm delivery, birthweight, andCaesarean section: a study within the Danish National Birth Cohort.Hum Reprod 2003;18:2478–84.

Longer TTP is associated with decreased levels of PAPP-A and free β-hCG 239

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.

14. Henriksen TB, Baird DD, Olsen J, et al. Time to pregnancy and pretermdelivery. Obstet Gynecol 1997;89:594–9.

15. Thomson F, Shanbhag S, Templeton A, Bhattacharya S. Obstetricwoutcome in women with subfertility. Br J Obstet Gynecol2005;112:632–7.

16. Spencer K, Bindra R, Nicolaides KH. Maternal weight correction ofmaternal serum PAPP-A and free beta-hCG MoM when screening fortrisomy 21 in the first trimester of pregnancy. Prenat Diagn2003;23:851–5.

17. Robinson HP, Fleming JE. A critical evaluation of sonar “crown-rumplength” measurements. Br J Obstet Gynaecol 1975;82:702–10.

18. Henriksen TB, Hedegaard M, Secher NJ. Standing at work and pretermdelivery. Br J Obstet Gynaecol 1995;102:198–206.

19. Olsen J, Juul S, Basso O. Measuring time to pregnancy. Methodologicalissues to consider. Hum Reprod 1998;13:1751–3.

20. Joffe M, Villard L, Li Z, et al. Long-term recall of time-to-pregnancy.Fertil Steril 1993;60:99–104.

21. Cole LA, Kohorn EI, Kim GS. Detecting and monitoring trophoblasticdisease. New perspectives on measuring human chorionicgonadotropin levels. J Reprod Med 1994;39:193–200.

22. Ghisoni L, Ferrazzi E, Castagna C, et al. Prenatal diagnosis after ARTsuccess: the role of early combined screening tests in counselingpregnant patients. Placenta 2003;24(Suppl B):S99–103.

23. Kösüs A, Kösüs N, Duran M, et al. Does assisted reproductivetechnology itself or polycystic ovary syndrome as a cause of infertilityhave any effect on first trimester serum screening results? J Reprod Med2012;57:129–35.

24. Parlagumus HA, Haydardedeoglu B, Simsek E, et al. Are serum markersaltered in first trimester screening in poor ovarian reserve patients? JObstet Gynaecol Res 2011;37:1582–7.

I. Kirkegaard et al.240

Prenatal Diagnosis 2014, 34, 235–240 © 2013 John Wiley & Sons, Ltd.