at SciVerse ScienceDirect

Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47

Contents lists available

Placenta

journal homepage: www.elsevier .com/locate/placenta

Review: Biochemical markers to predict preeclampsia

U.D. Anderson a, M.G. Olsson b, K.H. Kristensen c, B. Åkerströmb, S.R. Hansson c,*

aDepartment of Obstetrics and Gynecology, Department of Clinical Sciences, Lund University Hospital, Lund University, Tornavägen 10, 22184 Lund, SwedenbDivision of Infection Medicine, Department of Clinical Sciences, Lund University, SwedencDepartment of Obstetrics and gynecology, Department of Clinical Sciences, Malmö University Hospital, Lund University, Sweden

a r t i c l e i n f o

Article history:Accepted 23 November 2011

Keywords:Biochemical markersPredictionPreeclampsiasflt-1PlGFPAPP-AFetal hemoglobinHbFAlpha-1-microglobulinA1MPP13

* Corresponding author. Tel.: þ46 462223011.E-mail address: stefan.hansson@med.lu.se (S.R. Ha

0143-4004/$ e see front matter � 2012 Published bydoi:10.1016/j.placenta.2011.11.021

a b s t r a c t

Worldwide the prevalence of preeclampsia (PE) ranges from 3 to 8% of pregnancies. 8.5 million cases arereported yearly, but this is probably an underestimate due to the lack of proper diagnosis. PE is the mostcommon cause of fetal and maternal death and yet no specific treatment is available. Reliablebiochemical markers for prediction and diagnosis of PE would have a great impact on maternal healthand several have been suggested. This review describes PE biochemical markers in general and firsttrimester PE biochemical markers specifically. The main categories described are angiogenic/anti-angiogenic factors, placental proteins, free fetal hemoglobin (HbF), kidney markers, ultrasound andmaternal risk factors. The specific biochemical markers discussed are: PAPP-A, s-Flt-1/PlGF, s-Endoglin,PP13, cystatin-C, HbF, and a1-microglobulin (A1M). PAPP-A and HbF both show potential as predictivebiochemical markers in the first trimester with 70% sensitivity at 95% specificity. However, PAPP-A is notPE-specific and needs to be combined with Doppler ultrasound to obtain the same sensitivity as HbF/A1M. Soluble Flt -1 and PlGF are promising biochemical markers that together show high sensitivity fromthe mid-second trimester. PlGF is somewhat useful from the end of the first trimester. Screening preg-nant women with biochemical markers for PE can reduce unnecessary suffering and health care costs byearly detection of mothers at increased risk for PE, thus avoiding unnecessary hospitalization of pregnantwomen with suspect or mild PE and enabling monitoring of the progression of the disease therebyoptimizing time for delivery and hopefully reducing the number of premature births.

� 2012 Published by IFPA and Elsevier Ltd.

1. Introduction

Preeclampsia (PE) is one of the most serious pregnancycomplications. The worldwide prevalence of PE ranges from 3 to 8%of pregnancies, affecting a total of 8.5 million women worldwide.PE is responsible for about 18% of maternal deaths and up to 40% offetal mortality. At this time, PE still lacks a safe and effectivetherapy, as well as a reliable, early means of diagnosis or prediction.

PE has been named the "disease of theories" [1] and wasdescribed as a pregnancy-related disease as early as 3000 yearsago by the ancient Egyptians [2]. The disease evolves in twostages. The first stage is characterized by an altered formation ofthe placenta [3]. During placentation a defective invasion of theextravillous trophoblast cells into the muscle layers of the spiralarteries has been shown [4]. This contributes to a reduced utero-placental blood flow that can result in fetal intrauterine growthrestriction (IUGR), seen in one of four women with PE. A growing

nsson).

IFPA and Elsevier Ltd.

body of evidence suggests that oxidative stress further aggravatesvascular function in the placenta [5], which in turn [6] gives rise toinsufficient blood perfusion [7], inflammation, apoptosis andstructural damage [8e10]. The second stage, the clinical mani-festations, i.e. hypertension and proteinuria, appears from 20weeks of gestation onwards. As the disease progresses, angio-spasms in the brain and brain edema may cause severe epilepticseizures e eclampsia [11].

According to the International Society for the Study of Hyper-tension in Pregnancy (ISSHP), PE can be defined as de novo hyper-tension occurring after 20 weeks of pregnancy together withproteinurea. Hypertension is defined as a systolic blood pressure �140 and/or a diastolic blood pressure � 90 mmHg measured at twooccasions with at least 4 h in between. Proteinurea is defined as �300 mg per day [12]. Proteinurea is questionable as a marker for PEsince it lacks predictive value and does not correlate with severityof the disease.

A severe form of PE is the Hemolysis, Elevated Liver enzymesand Low Platelets syndrome (the HELLP-syndrome). It is defined bythe laboratory findings of hemolysis, elevated liver enzymes andlow platelet count [13]. Altogether, the wide range of clinical

U.D. Anderson et al. / Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47 S43

manifestations makes PE more syndrome-like than a defineddisease, which complicates the clinical diagnosis [14]. Lately, thetime of onset of the clinical manifestations, early onset PE (<34weeks of gestation) and late onset PE (>34weeks of gestation), havebeen used to further characterize PE, but the overall classificationstill lacks stringency.

In the last decade, methodologies such as genomics, proteomicsand metabolomics have been made more widely available forclinical research. In search of the etiology of PE, several newpathways and factors have been described using these techniques.Many of the described etiological factors are measurable in thematernal blood and have therefore been evaluated as biochemicalmarkers for prediction and diagnosis of PE. These include serum/plasma markers for renal dysfunction, endothelial dysfunction,metabolic status, oxidative stress, placenta-derived factors, hemo-lysis and inflammatory markers.

Several maternal clinical characteristics have been identified asrisk factors for developing PE. The most important of these areethnicity, age, parity, multiple pregnancy and a history of PE inearlier pregnancies. In addition, some systemic disorders alsoincrease the risk: obesity, diabetes mellitus, essential hyperten-sion, renal disease, antiphospholipid syndrome and certain auto-immune diseases [15]. However, none of these alone or incombination predict PE sufficiently, as they are reported to havea combined prediction rate of about 30%, early onset PE betterthan late onset PE [16].

Few biochemical markers have been proven specific and sensi-tive as single markers to predict and/or diagnose PE. Algorithmsalso include clinical measurements such as Doppler ultrasound andclinical risk factors, to further enhance the prediction rate at a lowfalse positive rate. In this review we describe the most promisingindividual biochemical markers suggested for both prediction anddiagnosis of PE. The biochemical markers are presented in the orderthey are shown to appear in pregnancy, i.e. first, second or thirdtrimester.

Table 1Biochemical predictionmarkers of preeclampsia. Shownwith function, gestational age at sof preeclampsia.

Biochemical marker Function of protein

Free fetal hemoglobin (HbF)and alpha-1-microglobulin(A1M)

Oxygen transportation protein in thefetus and A1M is an antioxidant anda heme scavenger

Pregnancy-associated Protein A(PAPP-A)

In fetuses with normal chromosomesin probably influences placentaldevelopment.

Placental Protein 13 (PP13) A member of the Galectin family.Function(s) are unclear e but itprobably induces apoptosis insome immune cells.

Soluble Endoglin Anti-angiogenetic protein

Metabolomics Metabolites measured in maternalblood as a final down stream productof gene expression.

Angiogenesisfactors(SFlt-1/PlGF)

PlGF alone

PlGF and sFlt-1 regulate angiogenesisof the placenta.

a Combined with maternal risk factors.b Early onset PE only, and combined with maternal risk factors.

2. Predictive biochemical markers

PAPP-A, PlGF and the combination HbF/A1M all show potentialas predictive biochemical markers in the first trimester (Table 1).

2.1. Pregnancy-associated protein A (PAPP-A)

PAPP-A is a glycoprotein synthesized in the placenta and thestudy of it as a biochemical marker in pregnancy has been pursuedfor almost 30 years [17]. The maternal plasma concentrationincreases throughout pregnancy. PAPP-A has been used in combi-nation with b-human chorionic gonadotropin (b-hCG) and nucaltranslucency thickness, to screen for trisomy 21, 13 and 18 at 11 þ 0to 13 þ 6 weeks of gestation [18]. In fetuses with normal chro-mosomes, decreased levels of PAPP-A in the first trimester havebeen associated with increased risk for PE, IUGR, fetuses small forgestational age (SGA) and preterm delivery [19e22].

PAPP-A has been evaluated as a predictive and diagnosticbiochemical marker for PE, but the screening performance, whenused as a single biochemical marker, is only about 10e20 %[23e25]. Combined with Doppler ultrasound, PAPP-A is a powerfulpredictive biochemical marker of PE with prediction rates of 70% atfalse positive rates of 5%. At term, plasma PAPP-A concentrationshave been shown to increase in pregnancies complicated by PE andHELLP, but its concentration is still not predictive for the severity ofthe disease [26].

2.2. Fetal hemoglobin and a1-microglobulin

Recent reports suggest that free, extracellular fetal hemoglobin(HbF) is involved in the pathogenesis of PE. Furthermore the heme-and radical scavenger a1-microglobulin (A1M) is involved in thephysiological defence against HbF. Their concentrations inmaternalserum or plasma can be used as early predictive biochemicalmarkers. Increased mRNA levels of HbF in the placental tissue and

ampling and prediction rate/false positive rate. Prediction rates are given for all kinds

Gestational ageat prediction

Prediction rate/false positives

10 þ 0e16 þ 0 69%/5% [8]Optimal 90%/23% [8]

11 þ 0e13 þ 6 9.6e20.1%/5% [20, 24]20.5e50%/10% [20, 24]In combination with Dopplerultrasound - 25e64.4%/5%[20, 24, 38]

11 þ 0e13 þ 6 37.5%a/5% [38]52.1ae80b%/10% [38, 39]43%/20% [40]In combination with Dopplerultrasound e 66.7%a/5% [38] or71.4e77.1%/10% [38, 39]

11 þ 0e13 þ 6 30%/5% [51]In combination with Dopplerultrasound e 66.7/5% [50]

14 þ 0e16 þ 0 77%/10% [55]

Second trimester:24 þ 0 and throughoutpregnancy.

82%/5% [53]88.5%/10% [53]88.5%/11.5% [54]

11 þ 0e13 þ 6 53.5%/5% [16]65%/10% [16]

U.D. Anderson et al. / Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47S44

free HbF protein in the placental vascular lumen were described inwomenwith PE [27]. Hemoglobin is a highly reactive molecule thatis capable of damaging and disrupting cell membranes [28], andbinds and inactivates nitric oxide (NO) with vasoconstriction asa consequence [29]. Its metabolites, heme and iron, damage lipids,protein and DNA through direct oxidation and/or generation ofreactive oxygen species (ROS). Heme is also a pro-inflammatorymolecule that activates neutrophils [30]. Several Hb- and heme-detoxification systems have been described in humans. Recently,the plasma and tissue protein A1Mwas shown to bind and degradeheme [31], have radical-scavenger properties [32], and protect cellsand tissues against extracellular Hb, heme and ROS [33,34]. A1M-expression in liver and placental cells has been shown to be up-regulated by Hb, heme and ROS [33,35]. A pathogenic role of Hband protective role of A1M in PE is supported by ex vivo placentaperfusion experiments [9].

Studies evaluating maternal serum/plasma concentrations ofHbF and A1M, as predictive and diagnostic markers for PE, haveshown promising results [8,33]. In a cohort of 96 patients (60subsequently developed PE) the serum concentrations of HbF andA1M were significantly increased at 10e16 weeks’ gestation inwomen who subsequently developed PE. The AUC for the ROC-curve was 0.89 for the two biochemical markers in combination.At a false positive rate of 5% the prediction rate was 69%. Optimalprediction rate was 90% for a false positive rate of 23% [8]. Theplasma concentrations of HbF and adult hemoglobin (HbA) werealso significantly correlated to maternal blood pressure in patientswith established PE [33]. These markers still need to be validated inlarger cohorts.

2.3. Placental protein 13 (PP13)

PP13 is a member of the galectin family and is produced by theplacental trophoblast cells [36,37]. The function(s) of PP13 is stillnot clearly understood, but it is involved in normal placentation[36]. In normal pregnancies, serum levels of PP13 slowly rise withgestational age. Several studies have shown lowered serum levelsin the first trimester in pregnancies that subsequently developedPE. As a first trimester screening marker for PE, PP13 showsdifferent prediction rates in different studies. In two differentcohort studies, PP13 levels were determined at 11 þ 0 to 13 þ 6weeks of gestation [38,39]. Both studies showed significantly lowerfirst trimester levels of PP13 in women who later developed PE. Atfalse positive rates of 5% and 10% they found detection rates of 37.5%and 69% respectively, using PP13 as a single biochemical marker.When combining serum screening with Doppler ultrasound pul-satility index (PI), the prediction rate increased to 71% at a falsepositive rate of 10% [39].

Romero et al. [40] studied a cohort of 300 patients 50 of whichdeveloped PE. At a false positive rate of 20% the detection rate was36% for all types of PE. For early onset PE it was 100% (n¼ 6) and forpreterm PE 85% (n ¼ 44). Preterm was defined as onset before37 þ 0 weeks. The prediction rate for severe PE at term was 24%(n ¼ 21). Based on these findings, PP13 was concluded to bea reasonable biochemical marker for early onset and preterm PE buta weak marker for PE at term [40].

Nicolaides et al. [41] studied PP13 as a biochemical marker forearly onset PE at 11 þ 0 to 13 þ 6 weeks of gestation. At a falsepositive rate of 10%, PP13 showed a prediction rate of 80% as a singlebiochemical marker. In combination with Doppler ultrasound PI,the prediction rate increased to 90% [41].

PP13 has also been used in combinationwith PAPP-A. In a cohortof 499 patients (47 PE cases) Stamatopoulou et al. [42] studied PP13and PAPP-A at 11 þ 0 to 13 þ 6 weeks of gestation in hypertensivepregnancies and pregnancies complicated by SGA. They found that

PAPP-A was significantly lower in SGA and in hypertensive disor-ders but interestingly, the levels of PP13 did not differ between thecases and the controls [42].

2.4. Soluble fms-like tyrosine kinase 1(sFlt-1) and soluble endoglin(sEng)

Two angiogenesis-related factors are particularly well studied:soluble fms-like tyrosine kinase (sFlt-1), a soluble VEGF receptor,and soluble endoglin (s-Eng), a co-receptor for TGF-beta. Both areelevated in maternal plasma in patients with PE compared tonormal pregnancies [43e46]. Elevated levels of sFlt-1 occur beforethe clinical symptoms. The levels correlate with the time of onset ofclinically manifest PE and partly with disease severity. Early-onsetPE exhibits higher levels of sFlt-1 [43,44,47,48]. Moreover, in animalexperiments, proteinuria and hypertension, as well as a HELLP-likesyndrome, were induced by infusion of high levels of sFlt andendoglin [49].

As a first trimester screening marker, s-Eng shows conflictingresults [50]. Used in combination with Doppler ultrasound (PI) andPlGF, the prediction rate for early onset PE was 77.8% at a falsepositive rate of 5%. AUC of the ROC curve was 0.95 [51].

2.5. Placental growth factor (PlGF) and sFlt-1

The ratio of the PlGF/sFlt-1 is well described and a promising setof biochemical markers for prediction of PE [52]. Automated fastanalysis methods have been developed for these proteins [53], buttheir role as first trimester markers is not clear [52].

Several studies have shown the predictive power of PlGF/sFlt-1ratio from the second trimester. The prediction rate is about 89%[54]. In a recent multicenter study by Verlohren et al. [53],including 351 patients (71 with PE), the sFlt-1/PlGF ratio wasmeasured longitudinally throughout pregnancy. The AUC of theROC-curve for detecting PE was 0.95. At a false positive rate of 5%the detection rate was 82% for all PE. For early onset PE, at a falsepositive rate of 3%, the detection rate was 89% [53]. Hence, the sFlt-1/PlGF ratio has no predictive value in the first trimester. As a singlebiochemical marker, PlGF has been shown to predict 53.5% of earlyonset PE at a false positive rate of 5% and 65% at a false positive rateof 10% [16] in late first trimester.

2.6. Metabolomics

Metabolic profiling is a powerful strategy to investigate themetabolites that a specific cellular event leaves behind. Metabolicprofiling can be used to reveal the pathophysiological mechanismsin a disease such as PE [55]. Recently, in a study of 60 patients whosubsequently developed PE and 60 normal pregnancies, 45metabolites were shown to be significantly altered in the firsttrimester in pregnancies that later developed PE. A predictivemodel, based on 14metabolites showed AUC of the ROC-curve to be0.94. For early and late onset PE, the prediction ratewas between 73and 77% at a 10% false positive rate [55]. The findingswere validatedin a cohort of 39 patients with subsequent PE matched with 40normal pregnancies. Interestingly, 3 out of the 40 up-regulatedmetabolites were shown to be hemoglobin metabolites.

2.7. Cystatin C

Cystatin C is a protease inhibitor widely used by clinicians asa sensitive marker for renal function and for estimation ofglomerular filtration rate. The maternal plasma level of cystatin C isincreased in women with PE and studies have demonstrated thatthe level of cystatin C is a reliable diagnostic marker for PE [56,57].

U.D. Anderson et al. / Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47 S45

Increased levels of cystatin C are suggested to be caused by eitherimpaired renal function and/or by increased placental synthesis[58]. Cystatin C has recently been suggested as a predictive firsttrimester marker for PE [59]. However, given the low screeningperformance of the study, cystatin C is probably not clinically usefulas a single marker but could be useful in combination with otherpredictive markers [60].

2.8. Other biochemical markers

As genomics, proteomics and metabolomics are being devel-oped and made more available, the number of potential biochem-ical markers will increase. Ideally, the biochemical markers willgive us new hints as to the pathogenesis behind PE. These newtechniques have revealed many of the above mentioned biochem-ical markers, and worth mentioning are free mRNAs [61] andmiRNAs [62] inmaternal blood. Both types of RNAs are expressed inthe placenta and can be found in the maternal circulation. Furtherinvestigation is needed but profiling of these RNAs might showpotential in predicting pregnancy outcomes.

2.9. New algorithms

The lack of a specific and sensitive biochemical marker has led tothe development of mathematical models that combine severalfactors in order to predict PE (Table 2). Akolekar et al. [16] combinedmaternal characteristics, PI and mean arterial pressure (MAP) withserum levels of PAPP-A, PlGF, PP13, inhibin-A, activin-A, sEng,pentraxin-3 and p-selectin in a large study (n¼ 33,602) at 11þ0 to13 þ 6 weeks of gestation. The prediction rates, at a false positiverate of 5%, were 91% for early onset, 79.4% for intermediate onset

Table 2Suggested algorithms of predictional biochemical/biophysical markers of PE. Shownwith gestational age at prediction, prediction rates and false positives. Predictionrates are for all PE e unless else is noted.

Combination ofbiochemical/biophysicalmarkers

Gestational ageat prediction

Prediction rate/falsepositives

Model 1 [16]Doppler ultrasound PIMean arterial blood

pressurePAPP-APlGFPP13Inhibin-AActivin-AsEngPentraxin-3p-Selectin

11þ0e13þ6 91% for early onset PE79% for intermediateonset PE61% for late onset PE/5%

Model 2 [63]PAPP-ABeta-hCGPlGFDesintegrinADAM12

First trimester 44%/5%

Model 3 [64]Doppler Ultrasound L-PIPAPP-AInhibin-APlGF

11þ0e13þ6 40%/10%100%/10% for earlyonset PE

Model 4 [65]PP13PAPP-ADoppler ultrasound

11þ0e14þ0 68%/5%

and 60.9% for late onset PE [16] (intermediate onset PE was definedas PE that led to delivery between 34 þ 0 and 36 þ 6 weeks ofgestation). Wortelboer et al. [63] developed a model based on thefirst trimester biochemical markers, PAPP-A, beta-hCG, PlGF,desintegrin and ADAM metallopeptidase domain 12 (ADAM12).Their prediction of all PE was only 44% at a 5% false positive rate[63]. Another first trimester model based on maternal character-istics, PI and the biochemical markers PAPP-A, inhibin-A, PP13,ADAM12, free beta-hCG and PlGF was developed by Audibert et al.[64]. In a large cohort (n ¼ 893) the model showed a 100%prediction rate for early onset PE at a false positive rate of 10% [64].It is worth noting that PP13 and ADAM12 levels did not improve theprediction rates. In a very recent study by Odibo et al. [65] maternalcharacteristics were combined with serum PP13, PAPP-A and PI inthe first trimester. In a cohort of 450 patients, the prediction ratewas 68% at a false positive rate of 5%. Interestingly, PI measure-ments did not increase the prediction rate in this study [65].

3. Discussion

The ideal biochemical marker for PE should exhibit thefollowing characteristics:

1) Play a central role in the pathogenesis and be specific for thecondition.

2) Appear early or before the clinical manifestations. Placentalfactors that can be detected early in pregnancy are likely to begood biochemical marker candidates for PE prediction.However, placental disorders can cause IUGR without PE andvice versa, which makes the clinical evaluation of newmarkersparticularly hard.

3) Be easy and cheap to measure in maternal blood or urine. Fewof the described factors are easy to measure; most of themrequire advanced laboratory systems.

4) Show a high sensitivity and specificity. A small number of thedescribed biochemical markers fulfill this requirement andstrategies to use them in combination with other markers and/or, with PI measurements and other clinical parameters arebeing investigated.

5) Correlate with the severity of the condition. As the diseaseprogresses, several organ systems are affected, which causesthe number of factors to increase throughout pregnancy. Agood candidate marker ought to appear early in pregnancy andcontinue to rise as the disease progresses.

6) Be non-detected or expressed at very low levels in normalpregnancies. Again, a placental factor is favored since theclinical symptoms disappear after removal of the placenta.

Screening pregnant womenwith an effective diagnostic markerfor PE� IUGR could reduce unnecessary suffering andmajor healthcare costs [66]. PE is still a dominant problem in the Third World,where it is often first diagnosed when the women present witheclamptic seizures. Basic equipment for blood pressure monitoringis often lacking, which requires clinicians to make careful clinicalobservations and basic examinations. Fetal monitoring withDoppler ultrasound and ECG is rarely available. Therefore, algo-rithms that summarize maternal risk factors are valuable and it ismost important to develop them further. In order for a PEbiochemical marker to be useful in the Third World, the need forsimplicity cannot be emphasized enough. Furthermore, thebiochemical marker must be detectable before the disease prog-resses into a dangerous stage, so that remote health care centres canrefer their pregnant women to larger hospitals in timely manner.

In the Western world, the trend goes towards general and earlyscreening for defects and diseases. Screening for Downs’s

U.D. Anderson et al. / Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47S46

syndrome in the first trimester is a good example where a combi-nation of ultrasound scanning and biochemical markers are used.This could be a suitable time also to screen for PE [67]. It is generallyaccepted, and a part of the WHO criteria, that when setting upa screening model for diseases like PE, the accepted level of falsepositives is set to 5%. However, as described above, publishedresults sometimes describe prediction rates at false positive rates of10%, 20% or higher. Potential first trimester biochemical markersare PAPP-A [18], HbF and A1M [8]. Both HbF and A1M play a role inthe pathophysiology of PE [9,27,33]. The biochemical markersappear as early as 10 weeks of gestation [8]. Furthermore, they canbe measured with basic ELISA techniques and show a highprediction rate at a low false positive level. Maternal plasmaconcentrations of free HbF have also been shown to correlate wellwith severity, i.e. blood pressure, in term PE pregnancies [33].

Angiogenic and anti-angiogenic factors are also very promisingbiochemical markers. Although the combination sFlt-1/PlGF mightnot be useful in the first trimester, they are definitelywell evaluatedin the second trimester. Alterations of sFlt-1 and PlGF levels occurabout 6 weeks before the onset of clinical symptoms and correlatewith the severity of the disease. PlGF could be a promisingbiochemical marker even in the first trimester particularly ifcombined with HbF and A1M.

PP13 has shownpotential as a biochemical marker of early onsetPE. Especially if combined with Doppler ultrasound uterine arteryPI [41]. However, as a general screening marker for all types of PE,the data is conflicting and needs further investigation.

Metabolomics is a promising technique to reveal early patho-physiological mechanisms. The predictive model, based on 14metabolites measured at 1416weeks of gestation, is interesting as itmeets the first criterion of a good predictive biochemical markerand appears before onset of the maternal symptoms [55]. Themodel has a high sensitivity, and might be used as a screeningmodel. However, due to the complex requirements regardingsample collection and analysis, the method may be difficult toimplement into clinical praxis, particularly in the Third World.

New factors should not be viewed solely as competingbiochemical markers for prediction and diagnosis of PE. Instead,each new factor ought to be welcomed as a new important puzzle-piece that contributes to illuminating the etiology of PE. In the endthese advances will hopefully lead to better prophylactic treat-ments reducing maternal and fetal morbidity.

Conflict of interest

The authors Stefan R. Hansson, Magnus G. Olsson and Bo Åker-ström hold patents related to prediction, diagnosis and treatmentof preeclampsia.

References

[1] Roberts JM, Cooper DW. Pathogenesis and genetics of pre-eclampsia. Lancet2001 Jan 6;357(9249):53e6.

[2] Stevens JM. Gynaecology from ancient Egypt: The Papyrus Kahun: a trans-lation of the oldest treatise on gynaecology that has survived from the ancientworld. Med J Aust 1975 Dec 20e27;2(25e26):949e52.

[3] Roberts JM, Hubel CA. The two stage model of preeclampsia: variations on thetheme. Placenta 2009 Mar;30(Suppl. A):S32e7.

[4] Brosens JJ, Pijnenborg R, Brosens IA. The myometrial junctional zone spiralarteries in normal and abnormal pregnancies: a review of the literature. Am JObstet Gynecol 2002 Nov;187(5):1416e23.

[5] Roberts JM, Hubel CA. Is oxidative stress the link in the two-stage model ofpre-eclampsia? Lancet 1999 Sep 4;354(9181):788e9.

[6] Shennan AH, Poston L, Chappell LC, Seed PT. Prevention of pre-eclampsia.Lancet 2001 May 12;357(9267):1534.

[7] Hung TH, Skepper JN, Charnock-Jones DS, Burton GJ. Hypoxia-reoxygenation:a potent inducer of apoptotic changes in the human placenta and possibleetiological factor in preeclampsia. Circ Res 2002 Jun 28;90(12):1274e81.

[8] Anderson UD, Olsson MG, Rutardottir S, Centlow M, Kristensen KH, Isberg PE,et al. Fetal hemoglobin and alpha(1)-microglobulin as first- and early second-trimester predictive biomarkers for preeclampsia. Am J Obstet Gynecol 2011Jun;204(6)(520):e1e5.

[9] May K, Rosenlof L, Olsson MG, Centlow M, Morgelin M, Larsson I, et al.Perfusion of human placenta with hemoglobin introduces preeclampsia-likeinjuries that are prevented by alpha1-microglobulin. Placenta 2011 Apr;32(4):323e32.

[10] Burton GJ, Jauniaux E. Oxidative stress. Best Pract Res Clin Obstet Gynaecol2011 Jun;25(3):287e99.

[11] Lipstein H, Lee CC, Crupi RS. A current concept of eclampsia. Am J Emerg Med2003 May;21(3):223e6.

[12] Brown MA, Lindheimer MD, de Swiet M, Van Assche A, Moutquin JM. Theclassification and diagnosis of the hypertensive disorders of pregnancy:statement from the International Society for the Study of Hypertension inPregnancy (ISSHP). Hypertens Pregnancy 2001;20(1):IXeXIV.

[13] Schroeder BM. ACOG practice bulletin on diagnosing and managingpreeclampsia and eclampsia. American College of Obstetricians and Gyne-cologists. Am Fam Physician 2002 Jul 15;66(2):330e1.

[14] Redman CW, Sargent IL. Latest advances in understanding preeclampsia.Science 2005 Jun 10;308(5728):1592e4.

[15] Duckitt K, Harrington D. Risk factors for pre-eclampsia at antenatal booking:systematic review of controlled studies. BMJ 2005 Mar 12;330(7491):565.

[16] Akolekar R, Syngelaki A, Sarquis R, Zvanca M, Nicolaides KH. Prediction ofearly, intermediate and late pre-eclampsia from maternal factors, biophysicaland biochemical markers at 11-13 weeks. Prenat Diagn 2011 Jan;31(1):66e74.

[17] Westergaard JG, Teisner B, Grudzinskas JG. Serum PAPP-A in normal preg-nancy: relationship to fetal and maternal characteristics. Arch Gynecol 1983;233(3):211e5.

[18] Kagan KO, Wright D, Valencia C, Maiz N, Nicolaides KH. Screening for triso-mies 21, 18 and 13 by maternal age, fetal nuchal translucency, fetal heart rate,free beta-hCG and pregnancy-associated plasma protein-A. Hum Reprod 2008Sep;23(9):1968e75.

[19] Ranta JK, Raatikainen K, Romppanen J, Pulkki K, Heinonen S. Decreased PAPP-A is associated with preeclampsia, premature delivery and small for gesta-tional age infants but not with placental abruption. Eur J Obstet GynecolReprod Biol 2011 Jul;157(1):48e52.

[20] Odibo AO, Zhong Y, Longtine M, Tuuli M, Odibo L, Cahill AG, et al. First-trimester serum analytes, biophysical tests and the association with patho-logical morphometry in the placenta of pregnancies with preeclampsia andfetal growth restriction. Placenta 2011 Apr;32(4):333e8.

[21] Spencer K, Yu CK, Cowans NJ, Otigbah C, Nicolaides KH. Prediction of preg-nancy complications by first-trimester maternal serum PAPP-A and free beta-hCG and with second-trimester uterine artery Doppler. Prenat Diagn 2005Oct;25(10):949e53.

[22] 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.J Clin Endocrinol Metab 2002 Apr;87(4):1762e7.

[23] Goetzinger KR, Singla A, Gerkowicz S, Dicke JM, Gray DL, Odibo AO. Predictingthe risk of pre-eclampsia between 11 and 13 weeks’ gestation by combiningmaternal characteristics and serum analytes, PAPP-A and free beta-hCG.Prenat Diagn 2010 Dec;30(12e13):1138e42.

[24] Poon LC, Maiz N, Valencia C, Plasencia W, Nicolaides KH. First-trimestermaternal serum pregnancy-associated plasma protein-A and pre-eclampsia.Ultrasound Obstet Gynecol 2009 Jan;33(1):23e33.

[25] Spencer K, Cowans NJ, Nicolaides KH. Low levels of maternal serum PAPP-A inthe first trimester and the risk of pre-eclampsia. Prenat Diagn 2008 Jan;28(1):7e10.

[26] Atis A, Tandogan T, Aydin Y, Sen C, Turgay F, Eren N, et al. Late pregnancyassociated plasma protein A levels decrease in preterm labor. J Matern FetalNeonatal Med 2011 Jul;24(7):923e7.

[27] Centlow M, Carninci P, Nemeth K, Mezey E, Brownstein M, Hansson SR.Placental expression profiling in preeclampsia: local overproduction ofhemoglobin may drive pathological changes. Fertil Steril 2008 Nov;90(5):1834e43.

[28] Tsemakhovich VA, Bamm VV, Shaklai M, Shaklai N. Vascular damage byunstable hemoglobins: the role of heme-depleted globin. Arch Biochem Bio-phys 2005 Apr 15;436(2):307e15.

[29] Kim-Shapiro DB, Schechter AN, Gladwin MT. Unraveling the reactions of nitricoxide, nitrite, and hemoglobin in physiology and therapeutics. ArteriosclerThromb Vasc Biol 2006 Apr;26(4):697e705.

[30] Kumar S, Bandyopadhyay U. Free heme toxicity and its detoxification systemsin human. Toxicol Lett 2005 Jul 4;157(3):175e88.

[31] Allhorn M, Berggard T, Nordberg J, Olsson ML, Akerstrom B. Processing of thelipocalin alpha(1)-microglobulin by hemoglobin induces heme-binding andheme-degradation properties. Blood 2002 Mar 15;99(6):1894e901.

[32] Akerstrom B, Maghzal GJ, Winterbourn CC, Kettle AJ. The lipocalin alpha1-microglobulin has radical scavenging activity. J Biol Chem 2007 Oct 26;282(43):31493e503.

[33] Olsson MG, Centlow M, Rutardottir S, Stenfors I, Larsson J, Hosseini-Maaf B,et al. Increased levels of cell-free hemoglobin, oxidation markers, and theantioxidative heme scavenger alpha(1)-microglobulin in preeclampsia. FreeRadic Biol Med 2010 Jan 15;48(2):284e91.

U.D. Anderson et al. / Placenta 33, Supplement A, Trophoblast Research, Vol. 26 (2012) S42eS47 S47

[34] Olsson MG, Olofsson T, Tapper H, Akerstrom B. The lipocalin alpha1-microglobulin protects erythroid K562 cells against oxidative damageinduced by heme and reactive oxygen species. Free Radic Res 2008 Aug;42(8):725e36.

[35] Olsson MG, Allhorn M, Olofsson T, Akerstrom B. Up-regulation of alpha1-microglobulin by hemoglobin and reactive oxygen species in hepatomaand blood cell lines. Free Radic Biol Med 2007 Mar 15;42(6):842e51.

[36] Than NG, Pick E, Bellyei S, Szigeti A, Burger O, Berente Z, et al. Functionalanalyses of placental protein 13/galectin-13. Eur J Biochem 2004 Mar;271(6):1065e78.

[37] Visegrady B, Than NG, Kilar F, Sumegi B, Than GN, Bohn H. Homologymodelling and molecular dynamics studies of human placental tissue protein13 (galectin-13). Protein Eng 2001 Nov;14(11):875e80.

[38] Akolekar R, Syngelaki A, Beta J, Kocylowski R, Nicolaides KH. Maternal serumplacental protein 13 at 11-13 weeks of gestation in preeclampsia. PrenatDiagn 2009 Dec;29(12):1103e8.

[39] Khalil A, Cowans NJ, Spencer K, Goichman S, Meiri H, Harrington K. First-trimester markers for the prediction of pre-eclampsia in women with a-priorihigh risk. Ultrasound Obstet Gynecol 2010 Jun;35(6):671e9.

[40] Romero R, Kusanovic JP, Than NG, Erez O, Gotsch F, Espinoza J, et al. First-trimester maternal serum PP13 in the risk assessment for preeclampsia. Am JObstet Gynecol 2008 Aug;199(2):122e1e122e11.

[41] Nicolaides KH, Bindra R, Turan OM, Chefetz I, Sammar M, Meiri H, et al.A novel approach to first-trimester screening for early pre-eclampsiacombining serum PP-13 and Doppler ultrasound. Ultrasound Obstet Gynecol2006 Jan;27(1):13e7.

[42] Stamatopoulou A, Cowans NJ, Matwejew E, von Kaisenberg C, Spencer K.Placental Protein-13 and pregnancy-associated plasma Protein-A as firsttrimester screening markers for hypertensive disorders and small for gesta-tional age outcomes. Hypertens Pregnancy; 2010 Aug 11.

[43] Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al. Circulatingangiogenic factors and the risk of preeclampsia. N Engl J Med 2004 Feb 12;350(7):672e83.

[44] Levine RJ, Lam C, Qian C, Yu KF, Maynard SE, Sachs BP, et al. Soluble endoglinand other circulating antiangiogenic factors in preeclampsia. N Engl J Med2006 Sep 7;355(10):992e1005.

[45] Maynard S, Epstein FH, Karumanchi SA. Preeclampsia and angiogenic imbal-ance. Annu Rev Med 2008;59:61e78.

[46] Reddy A, Suri S, Sargent IL, Redman CW, Muttukrishna S. Maternal circulatinglevels of activin A, inhibin A, sFlt-1 and endoglin at parturition in normalpregnancy and pre-eclampsia. PLoS One 2009;4(2):e4453.

[47] Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placentalsoluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelialdysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003Mar;111(5):649e58.

[48] Wikstrom AK, Larsson A, Eriksson UJ, Nash P, Norden-Lindeberg S,Olovsson M. Placental growth factor and soluble FMS-like tyrosine kinase-1 inearly-onset and late-onset preeclampsia. Obstet Gynecol 2007 Jun;109(6):1368e74.

[49] Li Z, Zhang Y, Ying Ma J, Kapoun AM, Shao Q, Kerr I, et al. Recombinantvascular endothelial growth factor 121 attenuates hypertension and improveskidney damage in a rat model of preeclampsia. Hypertension 2007 Oct;50(4):686e92.

[50] Erez O, Romero R, Espinoza J, Fu W, Todem D, Kusanovic JP, et al.The change in concentrations of angiogenic and anti-angiogenic factorsin maternal plasma between the first and second trimesters in riskassessment for the subsequent development of preeclampsia and small-for-gestational age. J Matern Fetal Neonatal Med 2008 May;21(5):279e87.

[51] Foidart JM, Munaut C, Chantraine F, Akolekar R, Nicolaides KH. Maternalplasma soluble endoglin at 11-13 weeks’ gestation in pre-eclampsia. Ultra-sound Obstet Gynecol 2010 Jun;35(6):680e7.

[52] Akolekar R, de Cruz J, Foidart JM, Munaut C, Nicolaides KH. Maternal plasmasoluble fms-like tyrosine kinase-1 and free vascular endothelial growth factorat 11 to 13 weeks of gestation in preeclampsia. Prenat Diagn 2010 Mar;30(3):191e7.

[53] Verlohren S, Galindo A, Schlembach D, Zeisler H, Herraiz I, Moertl MG, et al.An automated method for the determination of the sFlt-1/PIGF ratio in theassessment of preeclampsia. Am J Obstet Gynecol 2010 Feb;202(2):161e1e161e11.

[54] De Vivo A, Baviera G, Giordano D, Todarello G, Corrado F, D’Anna R. Endoglin,PlGF and sFlt-1 as markers for predicting pre-eclampsia. Acta Obstet GynecolScand 2008;87(8):837e42.

[55] Kenny LC, Broadhurst DI, Dunn W, Brown M, North RA, McCowan L, et al.Robust early pregnancy prediction of later preeclampsia using metabolomicbiomarkers. Hypertension 2010 Oct;56(4):741e9.

[56] Kristensen K, Wide-Swensson D, Schmidt C, Blirup-Jensen S, Lindstrom V,Strevens H, et al. Cystatin C, beta-2-microglobulin and beta-trace protein inpre-eclampsia. Acta Obstet Gynecol Scand 2007;86(8):921e6.

[57] Strevens H, Wide-Swensson D, Grubb A. Serum cystatin C is a better markerfor preeclampsia than serum creatinine or serum urate. Scand J Clin Lab Invest2001;61(7):575e80.

[58] Kristensen K, Lindstrom V, Schmidt C, Blirup-Jensen S, Grubb A, Wide-Swensson D, et al. Temporal changes of the plasma levels of cystatin C, beta-trace protein, beta2-microglobulin, urate and creatinine during pregnancyindicate continuous alterations in the renal filtration process. Scand J Clin LabInvest 2007;67(6):612e8.

[59] Thilaganathan B, Ralph E, Papageorghiou AT, Melchiorre K, Sheldon J. Raisedmaternal serum cystatin C: an early pregnancy marker for preeclampsia.Reprod Sci 2009 Aug;16(8):788e93.

[60] Thilaganathan B, Wormald B, Zanardini C, Sheldon J, Ralph E,Papageorghiou AT. Early-pregnancy multiple serum markers and second-trimester uterine artery Doppler in predicting preeclampsia. Obstet Gynecol2010 Jun;115(6):1233e8.

[61] Farina A, Zucchini C, Sekizawa A, Purwosunu Y, de Sanctis P, Santarsiero G,et al. Performance of messenger RNAs circulating in maternal blood in theprediction of preeclampsia at 10-14 weeks. Am J Obstet Gynecol 2010 Dec;203(6):575e1e7.

[62] Mouillet JF, Chu T, Sadovsky Y. Expression patterns of placental microRNAs.Birth Defects Res A Clin Mol Teratol 2011 Aug;91(8):737e43.

[63] Wortelboer EJ, Koster MP, Cuckle HS, Stoutenbeek PH, Schielen PC, Visser GH.First-trimester placental protein 13 and placental growth factor: markers foridentification of women destined to develop early-onset pre-eclampsia. BJOG2010 Oct;117(11):1384e9.

[64] Audibert F, Boucoiran I, An N, Aleksandrov N, Delvin E, Bujold E, et al.Screening for preeclampsia using first-trimester serum markers and uterineartery Doppler in nulliparous women. Am J Obstet Gynecol 2010 Oct;203(4)(383):e1e8.

[65] Odibo AO, Zhong Y, Goetzinger KR, Odibo L, Bick JL, Bower CR, et al. First-trimester placental protein 13, PAPP-A, uterine artery Doppler and maternalcharacteristics in the prediction of pre-eclampsia. Placenta 2011 Aug;32(8):598e602.

[66] Hadker N, Garg S, Costanzo C, Miller JD, Foster T, van der Helm W, et al.Financial impact of a novel pre-eclampsia diagnostic test versus standardpractice: a decision-analytic modeling analysis from a UK healthcare payerperspective. J Med Econ 2010;13(4):728e37.

[67] Cuckle HS. Screening for pre-eclampsiaelessons from aneuploidy screening.Placenta 2011 Feb;32(Suppl):S42e8.