European Journal of Obstetrics & Gynecology and Reproductive Biology 157 (2011) 48–52

Decreased PAPP-A is associated with preeclampsia, premature delivery andsmall for gestational age infants but not with placental abruption

Jenni K. Ranta a, Kaisa Raatikainen b, Jarkko Romppanen c, Kari Pulkki d, Seppo Heinonen e,*a Department of Clinical Chemistry, School of Medicine, Faculty of Health Sciences, University of Eastern Finland and Eastern Finland Laboratory Centre, P.O. Box 1700,

70211 Kuopio, Finlandb Department of Obstetrics and Gynaecology and Faculty of Health Sciences, University of Eastern Finland, P.O. Box 1777, Kuopio University Hospital, 70211 Kuopio, Finlandc Eastern Finland Laboratory Centre and University of Eastern Finland, P.O. Box 1700, 70211 Kuopio, Finlandd Department of Clinical Chemistry, School of Medicine, Faculty of Health Sciences, University of Eastern Finland and Eastern Finland Laboratory Centre, P.O. Box 1700, 70211 Kuopio, Finlande Department of Obstetrics and Gynaecology and Faculty of Health Sciences, University of Eastern Finland, P.O. Box 1777, Kuopio University Hospital, 70211 Kuopio, Finland

A R T I C L E I N F O

Article history:

Received 2 September 2010

Received in revised form 9 February 2011

Accepted 5 March 2011

Keywords:

First trimester

Screening

Adverse outcomes

Pre-eclampsia

Placental abruption

A B S T R A C T

Objective: To investigate links between first trimester Down’s syndrome screening markers and adverse

pregnancy outcomes; preeclampsia (PE), small for gestational age (SGA), preterm delivery (PD) and

placental abruption (PA) in spontaneous, chromosomally normal pregnancies.

Study design: Cohort study in a university hospital. Data during pregnancy were routinely collected from

a total study population of 2844 pregnant women between 2005 and 2007. Four study groups were

pregnancies with PE (N = 175), PA (N = 17), PD (N = 213) and SGA (N = 275) plus a reference group with

normal outcome (N = 2164). The median MOMs of maternal serum concentrations of pregnancy

associated plasma protein A (PAPP-A) and free beta human chorionic gonadotropin (fb-hCG) were

compared using two-tailed pooled t-tests, continuous variables were compared using Student’s two-way

t-tests, and Chi-square tests were used to analyse dichotomous variables. Fisher’s exact test was used

when there were fewer than five units in any of the classes.

Results: The median MOM of maternal serum PAPP-A was significantly lower in women with PE, PD and

SGA (0.79, 0.80 and 0.79 MOM, respectively) than in the reference group (0.99 MOM) (p < 0.01). The

median MOM of maternal serum fb-hCG was also significantly lower in the SGA group (0.90 MOM) and

in the PE and PD groups (0.86 and 0.92 MOM) than in the reference group (0.99 MOM, p = 0.02). There

was no detectable difference between the biochemical markers in the PA group and the reference group.

No statistical difference was found between NT MOMs in the reference and study groups.

Conclusion: The concentrations of first trimester screening (FTS) serum markers were lower in

pregnancies where PE, PD and SGA occurred. In the latter two cases, there was an inverse association

between incidence and PAPP-A and fb-hCG values. However, the development of PA during pregnancy

could not be predicted from biochemical marker concentrations. The mechanism behind PA is probably

less dependent on the placenta than on the decidua.

� 2011 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology andReproductive Biology

journa l homepage: www.e lsev ier .com/ locate /e jogrb

1. Introduction

Adverse pregnancy outcomes such as pre-eclampsia (PE), smallfor gestational age (SGA) infants, preterm delivery (PD) andplacental abruption (PA) are associated with higher mortality andsubsequent need for intensive care, in addition to affecting thechild adversely for many years [1–3]. Whilst it is known thatchronic maternal diseases, such as hypertension, lead to an

Abbreviations: fb-hCG, human chorionic gonadotropin, free beta subunit; FTS, first

trimester screening; MOM, multiples of medians; NT, nuchal translucency; PA,

placental abruption; PAPP-A, pregnancy associated plasma protein A; PE,

preeclampsia; PD, preterm delivery; SGA, small for gestational age.

* Corresponding author. Tel.: +358 17 172325; fax: +358 17 172685.

E-mail address: seppo.heinonen@kuh.fi (S. Heinonen).

0301-2115/$ – see front matter � 2011 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.ejogrb.2011.03.004

increased risk of adverse pregnancy outcomes [4,5], the vastmajority of such outcomes happen to healthy pregnant womenwith no known chronic diseases [6]. PE has been shown to beassociated strongly with alterations in angiogenetic and anti-angiogenetic factors in the first and second trimesters [7–9]. Levelsof serum markers during first trimester screening (FTS) for Down’ssyndrome have been consistently reported abnormal in chro-mosomally normal pregnancies affected by PE, PD and SGA [10–15], whereas in pregnancies affected by PA the findings are morecontroversial [16,17]. Some degree of familial clustering is shownwith PA. Additionally, the risk of recurrence of PA is known to berelatively high; of the order of 12% [18], whereas PE usually occursin only one pregnancy.

The mechanisms underlying these adverse outcomes are likelyto be initiated early in pregnancy, although the clinical endpoint

J.K. Ranta et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 157 (2011) 48–52 49

usually does not manifest itself until the second half of pregnancy.So far, clinically effective predictive and preventive strategies havebeen unavailable [19], but recently evidence has been found forantithrombotic medication in prevention of PE [20].

The objective of this study was to evaluate the relationshipbetween FTS markers and pregnancy outcomes in a cohort of 2844pregnant women. In particular, the performance of PAPP-A inscreening for specific pregnancy complications was explored, as itis known to play a role in placentation and early fetal growth.Outcomes investigated included PA, a disease of the decidua ratherthan the placenta. The hypothesis was that alterations in maternalserum PAPP-A concentrations would be associated with PE andfetal growth restriction but not with PA.

2. Materials and methods

The study was a retrospective cohort study. The data werecollected from the Department of Gynaecology and Obstetrics ofKuopio University Hospital and the Eastern Finland LaboratoryCentre. The study setting was a university hospital.

The women included in the study had spontaneously conceivedand had structurally normal singleton pregnancies. They allunderwent FTS, lived in the Kuopio catchment area, and gavebirth at Kuopio University Hospital between January 2005 andDecember 2007. Information was collected in maternity care unitsof the Kuopio University Hospital district and at the maternityclinic of Kuopio University Hospital. Patient-reported informationincluded the weight and height of the mother, chronic diseases(hypertension and diabetes), smoking habits before the pregnancy(yes/no), possible prior pregnancy complications and infertilitytreatment. Systematically gathered information included theresults of the FTS (the age of the mother, maternal serum fb-hCG and PAPP-A concentrations and the nuchal translucency (NT)of the fetus), maternal body mass index (BMI) before and at the endof the index pregnancy, pregnancy duration, possible pregnancycomplications, pregnancy outcome, gender of the fetus and birthweight. The following criteria were used to assess the pregnancyoutcomes: PE (repeated blood pressure measurements >140/90 mm Hg with proteinuria >0.5 g/day), PD (birth before 37completed pregnancy weeks), SGA (sex- and age-adjusted birthweight below the normal 10th percentile according to our records)and abruption of the placenta. Reference pregnancies were takento have a normal outcome when the mother and newborn did notrequire any prenatal, perinatal or postnatal follow-up, care orinterventions over and above what was considered to be routine.

The exclusion criteria for the study population were pregnan-cies with more than one fetus and those which showed major fetalstructural anomalies, since such pregnancies carry a higher risk ofadverse outcomes. The register of births included only pregnanciesextending to at least the 22nd gestational week, thereforemiscarriages and induced abortions were excluded, as were caseswith information missing (N = 259). Of the total study population,individuals were assigned to study groups as follows: pregnancieswith PE (N = 175), pregnancies with PD (N = 213), pregnancies withSGA (N = 275), and pregnancies with PA (N = 17). The referencegroup included 2164 women. The numbers of cases with missinginformation in one or more information classes (maternal age, BMI,weight gain, primigravida, smoking before the pregnancy, chronicand gestational illnesses, placenta praevia, birth weight, placentalweight, caesarean section, vacuum delivery, asphyxia, Apgarscores, intensive care, stillbirth, neonatal death, low birth weight,fetal gender, and FTS information) in each of the groups were: 11 inthe PE group (6.29%), 20 in PD group (9.39%), 23 in the SGA group(8.36%), 0 in PA group, and 156 in the reference group (7.21%).

The FTS was performed according to the recommendations ofthe Finnish Ministry of Social Affairs and Health. The maternal

serum samples were collected in maternity care units duringweeks 9 + 0 to 13 + 0. The NT and crown-rump length measure-ments were performed at healthcare centres and the KuopioUniversity Hospital maternity clinic by ultrasound-trained mid-wives and gynaecologists between weeks 11 + 0 and 13 + 6. Serumsamples were analysed in the Eastern Finland Laboratory Centre inKuopio. Concentrations of maternal serum fb-hCG and PAPP-Awere measured by time-resolved fluoroimmunoassay using anAuto-DELFIA kit (PerkinElmer Wallac, Turku, Finland). The riskfigures for biochemical markers and the NT scan were calculatedusing LifeCycle software version 2.0 (PerkinElmer LifeSciences,Wallac, Turku, Finland). The risk figure program compares apatient’s results with a population model described by a set ofmultivariate Gaussian distributions. The results were given asmultiples of medians (MOMs) relative to values recorded fornormal pregnancies at specific weeks. We used a cut-off limit of1:250, for a fixed false-positive rate of 5%. Both the fb-hCG andPAPP-A results were corrected for maternal weight and diabetes,but only fb-hCG measurements were corrected for smoking. Todetermine the effects of smoking on screening with PAPP-A, aseparate analysis was run to compare median PAPP-A MOMsamong the smoking and non-smoking women. Within- andbetween-assay variations were both <3.4% in the detection rangeof 4–157 ng/mL for fb-hCG and<2.4% and<4.0%, respectively, andin the detection range of 44–7300 mUI/L for PAPP-A. The limit ofdetection for fb-hCG and PAPP-A was 0.2 ng/mL and 5 mUI/L,respectively. Quality assurance was supervised by an internationalquality assurance company (UK NEQAS, Edinburgh, Great Britain).

Parameters of variables were tabulated and the differencesbetween the subject and reference groups were tested forstatistical significance. The median MOMs of maternal serumconcentrations of PAPP-A and fb-hCG were compared using two-tailed pooled t-tests, since the MOMs fitted Gaussian distributions.Continuous variables such as mean maternal age, birth weight andthe pregnancy duration were compared using Student’s two-way t-tests, and Chi-square tests were used to analyse dichotomousvariables. Fisher’s exact test was used when there were fewer thanfive units in any of the classes. When p values were less than 0.05differences were considered statistically significant. The oddsratios for adverse outcomes at different MOM levels were alsocalculated. Data were analysed using SAS software (SAS InstituteInc., Cary, NC, USA).

This study was approved by the Ethnical Research Committee ofKuopio University Hospital and the Institutional Review Board. TheCommittee has given permission for the results to be published.

3. Results

Table 1 shows the characteristics of the mothers and pregnancies.The proportion of overweight women was highest in the PE group(48.0%, p < 0.001). As expected, the highest proportions ofprimigravidas were in the PE and SGA groups (52.0% and 58.2%,respectively), and the differences were significant in all study groupsas compared to the reference group (36.3%, p < 0.001). Theproportion of mothers with hypertension was significantly higherin all study groups as compared to the reference group (p < 0.001).Diabetes was most common amongst mothers with PE and PD, andsignificantly higher as compared to the reference group (p < 0.001),whereas gestational diabetes was most common in the PE group(p < 0.01). There was no statistical difference between the groups inoccurrence of anaemia or gestational hepatosis. The highestoccurrence of placenta praevia was in the PD group (p < 0.001).

The mean maternal age in the whole study population was29.9 years. The proportion of mothers aged �36 years was lowestin the reference group and highest in the PA group, but differencesbetween the reference and study groups were not statistically

Table 1Maternal and pregnancy characteristics in the study groups (N = 2844).

Characteristic Reference

group

(N = 2164)

Preeclampsia

(N = 175)

p Premature

delivery

(N = 213)

p Small for

gestational age

infants (N = 275)

p Placental

abruption

(N = 17)

p

Mean maternal age (years) 29.8 30.0 0.66 30.3 0.24 29.7 0.76 31.6 0.24

BMI�25 before pregnancy 609a (28.1%) 84a (48.0%) <0.001 63a (29.6%) 0.92 66a (24.0%) 0.15 6 (35.3%) 0.65

Mean BMI prior to

pregnancy (kg/m2)

24.3 27.1 <0.001 24.5 0.62 23.5 0.01 23.8 0.65

Mean BMI at the end of

pregnancy (kg/m2)

29.7 32.6 <0.001 28.3 0.01 28.1 <0.001 28.4 0.41

Mean weight gain during

pregnancy (kg)

14.3 13.8 0.49 11.0 <0.001 12.0 <0.001 12.0 0.21

Primigravida 784 (36.3%) 91 (52.0%) <0.001 89 (41.8%) 0.11 160 (58.2%) <0.001 7 (41.2%) 0.67

Smoking before

pregnancy (yes/no)

582a (26.9%) 42a (24.0%) 0.27 60a (28.2%) 0.54 79a (28.7%) 0.44 5 (29.4%) 1.00

Chronic illness 353 (16.3%) 49 (28.0%) <0.001 52 (24.4%) 0.003 45 (16.4%) 0.98 5 (29.4%) 0.15

Hypertonia arterialis 1 (0.1%) 68 (38.9%) <0.001 28 (13.1%) <0.001 13 (4.7%) <0.001 2 (11.8%) <0.001*

Diabetes 18 (0.8%) 11 (6.3%) <0.001 20 (9.4%) <0.001 3 (1.1%) 0.72* 0 1.00*

Gestational diabetes 390 (18.0%) 47 (26.9%) <0.01 25 (11.7%) 0.02 39 (14.2%) 0.12 0 0.76*

Gestational hepatosis 26 (1.2%) 6 (3.4%) 0.01 7 (3.3%) 0.01 3 (1.1%) 1.00* 0 1.00*

Placenta praevia 23 (1.1%) 1 (0.6%) 1.0* 8 (3.8%) <0.001 1 (0.4%) 0.51* 1 (5.9%) 0.17*

a Data missing in less than 10% of cases.* When a class included less than 5 variables, Fisher’s exact test was used.

J.K. Ranta et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 157 (2011) 48–5250

significant. When previous infertility treatment, miscarriages,induced abortions, intrauterine deaths, or surgically scarred uteriwere compared, no significant differences were found between thestudy groups. Across all groups approximately 28% of the pregnantwomen were smokers before the pregnancy, and 15% during thecurrent pregnancy.

Mean birth weights were 16.3–43.1% lower in all the studygroups as compared to the reference group (p < 0.001), beinglowest in the PA group. In addition, the mean placental weight/birth weight ratio expressed as a percentage was higher in all thestudy groups as compared to the reference group (p < 0.001). Themean pregnancy duration was shorter in all study groups ascompared to the reference group (p < 0.001) and was shortest inthe PD group (17.8% shorter than in the reference group). Asanticipated, caesarean sections, abnormal Apgar scores, the needfor intensive care and low birth weight infants were more commonin all the study groups and their occurrence in these groups wassignificantly greater as compared to the reference group (p < 0.01).The outcomes of the pregnancies are presented in Table 2.

Table 2Pregnancy outcomes in the study groups (N = 2844).

Characteristic Reference

group

(N = 2164)

Preeclampsia

(N = 175)

p

Mean birth weight (study group

percentage of the reference group)

3660.2 g 3064.2 g

(83.7%)

<0.001

Mean placental weight/birth weight

ratio (%)

17.2 19.0 <0.001

Mean duration of pregnancy

(study group percentage of the

reference group)

280.3 days 265.2 days

(94.6%)

<0.001

Caesarean section 346 (16.0%) 68 (38.9%) <0.001

Vacuum delivery 154a (7.1%) 11a (6.3%) 0.46

Asphyxia 26 (1.2%) 4 (2.2%) 0.22*

Abnormal Apgar (<7) at 1 min 98 (4.5%) 32 (18.3%) <0.001

Abnormal Apgar (<7) at 5 min 36 (1.7%) 13 (7.4%) <0.001

Intensive care 99 (4.6%) 43 (24.6%) <0.001

Stillbirth 0 0 1

Neonatal death 0 0 1

Low birth weight 0 39 (22.3%) <0.001

Fetal gender male 1156 (53. 4%) 95 (54. 3%) 0.83

a Data missing in less than 10% of cases.* When a class included less than 5 variables, Fisher’s exact test was used.

The results of FTS screening are presented in Table 3. The mostsignificant finding was the lower PAPP-A median MOMs in the PE,PD and SGA groups as compared to the reference group (p < 0.001).The fb-hCG median MOM was also lower in these groups (p = 0.02in the PE and PD groups, and p < 0.01 in the SGA group) ascompared to the reference group. However, the median MOMs ofPAPP-A and fb-hCG of the PA and the reference group appeared tobe statistically similar. The median MOM of NT was similar inreference, PE, PD and SGA pregnancies, whereas in the PA group itwas higher, but the difference was statistically insignificant. Theproportion of screening positives was lowest in the referencegroup (2.6%) and PA group (0%), whilst the highest rate of positivescreening results was in the PD group (6.6%), and the differencewas statistically significant (p = 0.001). Excluding women (ap-proximately 15% in each group) who continued smoking duringpregnancy (>5 cigarettes/day) did not change the results (data notshown).

It was estimated that the odds ratio of fb-hCG MOM<1.0 for anadverse pregnancy outcome was 1.32 for PE, 1.16 for PD, 1.22 for

Premature

delivery

(N = 213)

p Small for

gestational

age infants

(N = 275)

p Placental

abruption

(N = 17)

p

2082.3 g (56.9%) <0.001 2714.5 g

(74.2%)

<0.001 2605.0 g

(71.2%)

<0.001

23.8 <0.001 18.1 <0.001 21.8 <0.001

230.4 days

(82.2%)

<0.001 275.8 days

(98.4%)

<0.001 253.8 days

(90.5%)

<0.001

106 (49.8%) <0.001 65 (23.6%) <0.01 8 (47.1%) <0.001

7a (3.3%) 0.01 25a (9.1%) 0.47 1a (5.9%) 1.00*

4 (1.9%) 0.34* 4 (1.5%) 0.77* 2 (11.8%) 0.02*

68 (31.9%) <0.001 27 (9.8%) <0.001 7 (41.2%) <0.001

48 (22.5%) <0.001 18 (6.5%) <0.001 4 (23.5%) <0.001*

136 (63.8%) <0.001 37 (13.5%) <0.001 8 (47.1%) <0.001

9a (4.2%) <0.001 1 (0.4%) 0.11* 1 (5.9%) <0.001*

16 (7.5%) <0.001 1 (0.4%) 0.11* 1 (5.9%) <0.001*

141 (66.2%) <0.001 48 (17.5%) <0.001 7 (41.2%) <0.001*

123a (57.7%) <0.01 144 (52.4%) 0.74 8 (47.1%) 0.63*

Table 3First trimester Down’s syndrome screening in the study groups (N = 2844).

Characteristic Study groups

Reference

group

(N = 2164)

Preeclampsia

(N = 175)

p Premature

delivery

(N = 213)

p Small for gestational

age infants (N = 275)

p Placental

abruption

(N = 17)

p

fb-hCG median MOM 0.99 0.86 0.02 0.92 0.02 0.90 <0.01 0.92 0.62

PAPP-A median MOM 0.99 0.79 <0.001 0.80 <0.001 0.79 <0.001 1.00 0.76

NT median MOM 1.05 1.06 0.63 1.06 0.52 1.05 0.75 1.22 0.59

Screening positive 57 (2.6%) 9 (5.1%) 0.06 14 (6.6%) 0.001 13 (4.7%) 0.05 0 1.00*

* When a class included less than 5 variables, Fisher’s exact test was used.

Table 4Odds ratios (95% confidence interval) for adverse pregnancy outcomes relative to MOM values in the study groups (N = 2844).

Biochemical

marker

Reference

group

(N = 2164)

Preeclampsia

(N = 175)

OR (CI) Premature

delivery

(N = 213)

OR (CI) Small for

gestational

age infants

(N = 275)

OR (CI) Placental

abruption

(N = 17)

OR (CI)

PAPP-A MOM

<1.0 1097 (50.7%) 121 (69.1%) 2.06

(1.51�2.81)

138 (64.8%) 1.70

(1.30�2.23)

177 (64.4%) 1.65

(1.31�2.09)

9 (52.9%) 1.09

(0.42�2.82)

<0.8 768 (35.5%) 92 (52.6%) 1.91

(1.43�2.53)

107 (50.2%) 1.73

(1.34�2.24)

140 (50.9%) 1.75

(1.40�2.18)

5 (29.4%) 0.76

(0.27�2.15)

<0.6 445 (20.6%) 61 (34.9%) 1.94

(1.44�2.60)

71 (33.3%) 1.80

(1.38�2.36)

99 (36.0%) 1.96

(1.56�2.46

2 (11.8%) 0.52

(0.12�2.25)

<0.4 154 (7.1%) 27 (15.4%) 2.18

(1.49�3.18)

29 (13.6%) 1.89

(1.32�2.71)

54 (19.6%) 2.62

(2.02�3.40)

2 (11.8%) 1.73

(0.40�7.50)

fb-hCG MOM

<1.0 1101 (50.9%) 102 (58.3%) 1.32

(0.99�1.76)

117 (54.9%) 1.16

(0.90�1.50)

155 (56.6%) 1.22

(0.97�1.52)

11 (64.7%) 1.76

(0.65�4.75)

<0.8 780 (36.0%) 80 (45.7%) 1.45

(1.09�1.93)

85 (39.9%) 1.16

(0.89�1.51)

122 (44.4%) 1.36

(1.09–1.70)

7 (41.2%) 1.24

(0.47�3.24)

<0.6 429 (19.8%) 49 (28.0%) 1.51

(1.11�2.07)

57 (26.8%) 1.42

(1.07�1.89)

74 (26.9%) 1.42

(1.11�1.82)

4 (23.5%) 1.24

(0.41�3.79)

<0.4 143 (6.6%) 13 (7.4%) 1.12

(0.65�1.93)

20 (9.4%) 1.41

(0.91�2.17)

32 (11.6%) 1.70

(1.22�2.38)

3 (17.7%) 2.97

(0.87�10.28)

J.K. Ranta et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 157 (2011) 48–52 51

SGA, and 1.76 for PA. The odds ratio of PAPP-A MOM<1.0 was 2.06for PE, 1.70 for PD, 1.65 for SGA, and 1.09 for PA. In pregnancieswith PD and SGA, the odds ratio with decreasing PAPP-A MOMincreased linearly, whereas in pregnancies with PE the odds ratioswere variably higher with low PAPP-A MOMs. In PA pregnancies,the odds ratios were most variable; 0.76 for PAPP-A MOM <0.8,0.52 for PAPP-A MOM <0.6, and 1.73 for PAPP-A MOM <0.4. Theodds ratios of developing an adverse outcome at different MOMlevels are presented in Table 4, including the sizes of the subgroupswith different MOM levels.

4. Comment

Many previous studies have examined the usefulness of firsttrimester maternal biochemistry in predicting adverse pregnancyoutcomes and have shown that, in particular, low PAPP-A isassociated with preterm birth, PE and impaired fetal growth. Ourdata are consistent with previously reported results [10,12,21].Low levels of fb-hCG and PAPP-A probably reflect impairedplacentation, and in addition, pregnancy-associated plasmaprotein A (PAPP-A) is known to be an insulin-like growth factorbinding protein (IGFBP) protease and to increase the bioavailabilityof IGF. In turn, IGF is believed to play a role in fetal growth [22] bymediating trophoblast invasion to the decidua, and to regulatesteroidogenesis and glucose and amino acid transport in thechorionic villi [23]. The mechanism involved with most adverseoutcomes such as PD, PE, and SGA may be related to the proteaseactivity of PAPP-A affecting free IGF concentrations in earlypregnancy. In our study, the more the PAPP-A and fb-hCG

concentrations were below the normal median value, the morelikely was the occurrence of PD, PE and SGA.

Our main interest, however, was in pregnancies with PA, sinceprediction of this outcome has clear clinical relevance, due tofamilial clustering and a substantial risk of recurrence. It wassignificant that no changes were seen in first trimester maternalserum PAPP-A and fb-hCG concentrations in patients whosubsequently developed PA. It is also interesting that during thesecond trimester, no difference between normal pregnancies andPA pregnancies has been found in fb-hCG levels, whereas alphafetoprotein (AFP) has been found normal [24] or significantlyhigher in PA pregnancies [25]. Alpha fetoprotein is a glycoproteinsecreted by fetal tissues alone. PA has been considered to be adisease of the decidua rather the placenta. Furthermore, it has beenestablished that in pregnancies with PA fetal growth is usually notcompromised [6] and, therefore, the lack of signs of early placentaldysfunction was to be expected. One multi-centre study into theassociation between low PAPP-A and PA, where diagnosis of thecondition was based on bleeding in late pregnancy, found asignificant association in cases with a very low first trimesterPAPP- A (<5th percentile, �0.42 MoM); but using this cut-off lessthan 10% of all PAs were recognized [16].

The association between serum markers and adverse outcomeswas relatively weak in the present study, which, therefore, limitsthe clinical usefulness of the screening data. On the other hand,clinical risk factors such as smoking during pregnancy, whichincrease the risk of preterm birth [26], are considered significant ata much lower risk level. Our findings, as well as earlier reports,imply that pregnancies with moderate or high risk for developingPD, PE or SGA can be recognised during the first trimester, and

J.K. Ranta et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 157 (2011) 48–5252

guided for further testing. Since the number of pregnancies with PAis very low at population level, our estimation of odds ratios on lowPAPP-A is only directional. Unfortunately, cases of PA, even wherepatient and familial history may both identify patients at risk,appeared to be unpredictable on the basis of the biochemicalscreening used here. Furthermore, there is no treatment for PA,whereas for PE evidence has been found that antithromboticmedication is efficient in prevention [20].

In PD, PE, and SGA, placental dysfunction may be the primarydisturbance underlying the mechanisms that give rise to theseoutcomes. Assessment of placentation and placental function in thefirst trimester may offer new insights into the etiological aspects ofadverse outcomes of pregnancy. The overall number of complicatedpregnancies was low in the present study, and it is obvious thatfurther studies are needed to fully evaluate the potential of firsttrimester screening markers to predict adverse outcomes.

References

[1] Yanney M, Marlow N. Paediatric consequences of fetal growth restriction.Semin Fetal Neonatal Med 2004;9(October (5)):411–8.

[2] Marlow N, Wolke D, Bracewell MA, Samara M, EPICure Study Group. Neuro-logic and developmental disability at six years of age after extremely pretermbirth. N Engl J Med 2005;352(January (1)):9–19.

[3] Duley L. The global impact of pre-eclampsia and eclampsia. Semin Perinatol2009;33(June (3)):130–7.

[4] Mavalankar DV, Gray RH, Trivedi CR. Risk factors for preterm and term lowbirthweight in Ahmedabad, India. Int J Epidemiol 1992;21(April (2)):263–72.

[5] Catov JM, Nohr EA, Olsen J, Ness RB. Chronic hypertension related to risk forpreterm and term small for gestational age births. Obstet Gynecol2008;112(August (2 Pt 1)):290–6.

[6] Toivonen S, Heinonen S, Anttila M, Kosma V-M, Saarikoski S. Reproductive riskfactors, Doppler findings and outcome of affected births in placental abrup-tion—a population-based analysis. Am J Perinatol 2002;19:451–9.

[7] Catarino C, Rebelo I, Belo L, et al. A prospective cohort study of the value ofmaternal plasma concentrations of angiogenic and anti-angiogenic factors inearly pregnancy and midtrimester in the identification of patients destined todevelop preeclampsia. Growth Factors 2009;27(December (6)):345–51.

[8] Kusanovic JP, Romero R, Chaiworapongsa T, et al. Soluble and membranousvascular endothelial growth factor receptor-2 in pregnancies complicated bypre-eclampsia. J Matern Fetal Neonatal Med 2009;22(November (11)):1021–38.

[9] Tripathi R, Rath G, Ralhan R, Saxena S, Salhan S. Soluble and membranousvascular endothelial growth factor receptor-2 in pregnancies complicated bypre-eclampsia. Yonsei Med J 2009;50(October (5)):656–66.

[10] Smith GC, Stenhouse EJ, Crossley JA, Aitken DA, Cameron AD, Connor JM. Earlypregnancy levels of pregnancy-associated plasma protein a and the risk ofintrauterine growth restriction, premature birth, preeclampsia, and stillbirth. JClin Endocrinol Metab 2002;87(April (4)):1762–7.

[11] Tul N, Pusenjak S, Osredkar J, Spencer K, Novak-Antolic Z. Predicting compli-cations of pregnancy with first-trimester maternal serum free-betahCG, PAPP-A and inhibin A. Prenat Diagn 2003;23(December (12)):990–6.

[12] Liu SS, Lee FK, Lee JL, et al. Pregnancy outcomes in unselected singletonpregnant women with an increased risk of first-trimester Down’s syndrome.Acta Obstet Gynecol Scand 2004;83(December (12)):1130–4.

[13] Brameld KJ, Dickinson JE, O’Leary P, et al. First trimester predictors of adversepregnancy outcomes. Aust N Z J Obstet Gynaecol 2008;48(December (6)):529–35.

[14] Spencer K, Cowans NJ, Avgidou K, Molina F, Nicolaides KH. First-trimesterbiochemical markers of aneuploidy and the prediction of small-for-gestationalage fetuses. Ultrasound Obstet Gynecol 2008;31(January (1)):15–9.

[15] Poon LC, Kametas NA, Maiz N, Akolekar R, Nicolaides KH. First-trimesterprediction of hypertensive disorders in pregnancy. Hypertension2009;53(May (5)):812–8 (Epub 2009 Mar 9).

[16] Dugoff L, Hobbins JC, Malone FD, et al. First-trimester maternal serum PAPP-Aand free-beta subunit human chorionic gonadotropin concentrations andnuchal translucency are associated with obstetric complications: a popula-tion-based screening study (the FASTER Trial). Am J Obstet Gynecol2004;191(October (4)):1446–51.

[17] Smith GC, Crossley JA, Aitken DA, et al. First-trimester placentation and therisk of antepartum stillbirth. JAMA 2004;292(November (18)):2249–54.

[18] Toivonen S, Keski-Nisula L, Saarikoski S, Heinonen S. Risk of placental abrup-tion in first-degree relatives of index patients. Clin Genet 2004;66(September(3)):244–6.

[19] Pasupathy D, Smith GC. The analysis of factors predicting antepartum still-birth. Minerva Ginecol 2005;57(August (4)):397–410.

[20] Bujold E, Morency AM, Roberge S, Lacasse Y, Forest JC, Giguere Y. Acetylsa-licylic acid for the prevention of preeclampsia and intra-uterine growthrestriction in women with abnormal uterine artery Doppler: a systematicreview and meta-analysis. J Obstet Gynaecol Can 2009;31(September(9)):818–26.

[21] Kwik M, Morris J. Association between first trimester maternal serum preg-nancy associated plasma protein-A and adverse pregnancy outcome. Aust N Z JObstet Gynaecol 2003;43(December (6)):438–42.

[22] Bale LK, Conover CA. Disruption of insulin-like growth factor-II imprintingduring embryonic development rescues the dwarf phenotype of mice null forpregnancy-associated plasma protein-A. J Endocrinol 2005;186(August(2)):325–31.

[23] Sun IY, Overgaard MT, Oxvig C, Giudice LC. Pregnancy-associated plasmaprotein A proteolytic activity is associated with the human placental tropho-blast cell membrane. J Clin Endocrinol Metab 2002;87(November (11)):5235–40.

[24] Tikkanen M, Hamalainen E, Nuutila M, Paavonen J, Ylikorkala O, Hiilesmaa V.Elevated maternal second-trimester serum alpha-fetoprotein as a risk factorfor placental abruption. Prenat Diagn 2007;27:240–3.

[25] Chandra S, Scott H, Dodds L, Watts C, Blight C, Van Den Hof M. Unexplainedelevated maternal serum alpha-fetoprotein and/or human chorionic gonado-tropin and the risk of adverse outcomes. Am J Obstet Gynecol 2003;189(Sep-tember (3)):775–81.

[26] Raatikainen K, Huurinainen P, Heinonen S. Smoking in early gestation orthrough pregnancy: a decision crucial to pregnancy outcome. Prev Med2007;44(January (1)):59–63 (Epub 2006 Sep 7).