Bacigalupo et al. Total and unconjugated estriol 73

Original articles

J. Perinat. Med.10 (1982)73

Paired measurements of total and unconjugated estriolin matemal plasma during the 2nd and 3rd trimester of pregnancytheir relationship to intrauterine growth retardation

Giovanni Bacigalupo, Erich Z. Saling, Jörg Gesche

Unit of Perinatal Medicine — The Free University of Berlin

Growth retarded fetuses and neonates have highermorbidity and mortality rates than normally devel-oped infants of comparable age [4, 5, 16, 19, 20,24]. During the immediate neonatal period, small-for-date infants run an increased risk of infection,are particularly prone to hypoglycemia and havedifficulties in maintaining normal body tempera-ture. Subsequently some of these infants will notattain normal body size, and a considerable num-ber will suffer from neurologic impairment [8,10,23].Early detection of intrauterine growth retardation(IUGR) and careful monitoring of the affectedfetuses, both by biophysical and biochemicalmeans, are of substantial interest to the obstetri-cian. Once fetal growth retardation is recognized,the obstetrician will attempt to normalize fetalgrowth during the further course of the pregnancyby elimination of maternal causes contributing tothe gestational disorder (nicotine, drug abuse,maternal malnutrition, etc.) or by intensive ther-apy of accompanying maternal complications(toxemia, gestational diabetes, etc?) In some cases,early delivery of a viable fetus will be accom-plished before irreversible damage or death hasoccurred.During the 2nd and 3rd trimester of pregnancy,the major proportion of estriol circulating inmaternal blood and being excreted through mater-nal urine and feces is synthesized in the feto-placental unit [6, 7]. Normal maternal plasmaestriol levels indicate fetal well-being, whereas

abnormal estriol profiles are mostly associatedwith abnormal fetal development or fetal distress[14].The study was designed to examine the possibil-ities of predicting fetal growth retardation äsearly äs during the 2nd trimester of pregnancy bymeans of total and unconjugated estriol assay inmaternal plasma and to compare the usefulness ofthese two feto-placental function tests in the ante-partum Identification of growth retardation.

l Patients and materialsIn 583 pregnancies at 15 to 42 gestational weeks,913 plasma samples of peripheral maternal venousblood were assayed for their concentrations oftotal estriol and unconjugated estriol.Among the pregnancies examined there were 463mothers who were delivered of a healthy singletoninfant of normal birthweight. These women con-tributed 707 plasma samples to this study. Normalbirthweight was defined äs being between the lOthand 90th percentiles for gestational age accordingto the birthweight curves of NICKL [25], whichdescribe the relative frequency of birthweights byweek of gestation for the German population inthe 1960s. Gestational age was determined by theclinical criteria proposed by FARR [9] and PE-TRUSSA [26] which are based upon estimations ofneonatal maturity.

0300-5577/82/0010-0073$02.00©by Walter de Gruyter & Co. · Berlin · New York

o l nl, Total uiul «WKinJugaliHJ wtriol

M )

Hfil O l l i

wrll u«

Wimit'll fXUl ' i l l l i tUl ^IWO W i l l i tt)Ι ι ι Ι Ί ι ΐ ι Ι « wliow lilil l iw^hf fal l hHow Ihe'pnlllo Ι οί tyiiNliiliuiittl tt#c» und wlio w rediwlllwl UN M i l l t M l i i t / I t o iu i n t i i u i t o i l n r

I f M f l l ' d u l i o i l . Ι ' Ι Ο Ι Ι Ι Ulf l l l l l l lKW W i l l ) H(.W'UWnM'y, ,M)f» pI l lMIIM Mllllpltni W IM Γ

nl Im t 'bhlol, (Όιιιριιπιΐΐνρ ( I t u h nl d d t f i I ' IOMI||K< η οιηηιΐ wi'lfjlil Mini 1 1 10 H X i U 11001111(601111

|κ*ΐΊΐΗιιι1ΙΙ<ηι in llumt*filvon in Tttlt , l Mih l l L

I t i l t l ( lliih M! i lo l i i UM f l i t < «'' ) A l i l l l l l ^ l i ^ U l M i t ' M f l M I M l

Huilv

wilhi i i i i i i i l

1l h1»' l ' - Ί Η

wllhIhh i tuUMiiu*

Ν Λ Ι . Ι Ν ΠW ;UI

nf

HU t l . · »

N U , M f 4 M ' l l r i )

ΜΙ,/1* H V l

^ΜΛ A h,l yni,

H.« ' λ

HI.H^ ( Ι ί )

λ/, i A fi.v yi

II^HV« r ml l mttlenit t lh h u u l \VIMO in i i l l i i c ' l y w i l l u l i n w n

Vi ln^M tiiul ^1n(ly Nl ikcn. ΛΓΙοιInhibi t ion Hi^ plrt^niii h iyoiw w^io leiiiov^i! ι ι η ι ΐ«touMl i t l -M) "V i i l l l (h h ριοίΗ^Ιι^4 wli io l ttook ρίΜΐΨ wl i l i i i ) ( i i i0 \vtH*k wIIei1 wl l lu luiwnl ,In mioh plm^nui im^ty, ριιΐκηΐ ^ Μ ο ι ι ι ι ί ι κ ι ΐ ί ο ι ι ^ οΓloli i l n ι ul πικ (οιι|ίΐμ!ΐΚΜΐ oHhiol wnif^ inmlo npplyl i iHllu1 mini^ πκΐί οίηιιηιΐιι οΐο^ΐο pau^ilui'pi Amilyw«ot hoih rMt io l Iov^lr4 ( l o t t i l m\\ \\\\ν\\φ\Μ(ν(\) in!ho m\w »itniplo \vo<o otniUnl ot i l I t t oiui \\\\(\ Ih0

U I A i u n , All οΓ Ihoι HM ίο in UM l In tlnplNu*; H I U M tho wl iu ly

Immnl on JoS?! «Ιΐιμίο

Tt*l), II,MIH t^

φ In- (MI ool-

(A) und ΙΚΗΜΗ*(Γ*|uftm <tollvor)f (H) In tho outt» cxunt

adV, 10 tun! CO* ovmloiu! in urlwlul

^ pH <|u40 * 7/19 In in In in lιο^ίοη ~ *« 6 polnt^ of (Im

|/7|i thc ^orlii i(H0rlu df f iAI, lN(] uroIn llio^uf ΑΓΟΑΙΓ

l»lo<ul.

wllh wlth iniru*utcrln®

Α , ( ί ^ Μ Ι Ι ( ΐ Μ « Ι ( Ι Ι ι ι Ι κ ι Ι ί * «hivniilkTu*i»iiilM

II.

4,1% ( I M ) * 10,0% (12)0 (0) 0,8% ( I )

H.2* (3H)4,3% (20)( , ;«% (6)

Ι ? » ί % (2ΠH,J% (10);i, 1% (4)

Ulol i t l ( ^Nl i lo l i iNNiiy, pfuiniui Hiuupl^N worci

lo on/yinci hydrolytilti, t^or t lni lprcpnrtilln οΓ hrUi Hlnoni'ontdnNti/IVoii i I lol lx ponuitltt 0',^ und Λ π U/ml) WKM( l l ( ) i ' : M i ( i N O i i ' u / M m n i l i o l n i , Ino,) , 0,(),1 ml(),0,( ml rn/yinc tut lul lon luul D- l u l O . I N nootnloInilTor p l l »!,(! wow nilx^d In »loppewd μΐΗ

iul l i idi ihi i led nl fi»! *V Tor 00 mlnt i lo«, CοΓ thd oNl i io l οοιφι^ΐοΝ WIIN

r.hown In lopontod ImUti vyllh puio 0AtrlolIn ικμιοοηκ ruilnllon itml In plminui

wl lh known ^Irlo) oonlont , Totnl oMrlolWMM (hon ixireoltfil hy luKllun ( v ^ Ihd liydrolywKlpro olillknl mniplcn .1 ml nf hlphly purUlcd tllotliyl·olhpr ( ΜΙΐν««οΓ\ I t , M K U I ' K I Ιηο,ι ΟιιηιιβΙικΙΟ iVomiVonlily oiwiwil hot(len. l\v Vortex mlxln f r, < minnl^ mul o^nlill\ipilon al rUHH) μ und u loin^

of i "t1 l\n 10 mlnnl^, Tu cuoh of i UIA, 0,1 ml l lln? »Hinr Inyor oonlHlnlng Iho

Uil i i l Mlrlol wiw imnurprreU, evttporukHlIn H vmnmitt «l ?so M ( \ ιηκΐ llw κιηκίικίβι1 WMB

dl ICH!,irlol wiw oximeted IVom 0,0i ml

wllh »t ml iiher, MubniH|uon(ly ! ml o f l lw(Mlioi ρίπικο \v»s ^Ivon (o m\oh οΓ

RIA lulw»! ovrtponUed HU Ιΐιοωΐφΐιΐχ ilriotl, ωJ,

Bacigalupo et al., Total and unconjugated estriol 75

described. The extraction yield amounted to 98.1± 1.2% äs measured by the recovery of knownquantities of tritium-labelled estriol.Estriol quantitation was carried out by means of aradioimmunoassay which followed in essence theprocedure described by TULCHINSKY and ABRA-HAM [29]. The estriol extracted was incubatedwith estriol antiserum and labelled estriol in0.04 M phosphate buffer pH 7.4 with 0.1% bovineserum albumin added at 0° to 4 °C for 120 min-utes. Unbound estriol was then separated fromantiserum bound estriol by the addition of a char-coal/dextran Suspension (l mg charcoal per tube).After centrifugation the radioactivity of the super-natant was measured in a liquid scintillationcounter. The estriol concentrations of the samplesassayed were calculated using a special Computerprogram.The estriol antiserum used was raised in rabbitsthrough repeated injections of a conjugate ofestriol-6-(0-carboxymethyl)oxim with BSA äsimmunogen (SORIN BIOMEDICA, Saluggia, Italy).This antiserum had a cross-reaction of 0.01% withestradiol and 0.001% with estrone at 50% displace-ment. As tracer 2,4,6,7-3H-(N) estriol was used upto 0.5 (18.5 KBq) for 100 samples.In the ränge of 10 and 100 pg estriol per RIAtube, the intra-assay and inter-assay Variation

coefficients were 5.1% and 6.9%. Using the RIASystem, the lowest detectable amount of estriolwas 6 ± 2 pg. The recovery of known estriolamounts added to plasma samples was at 96%nearly quantitative.

2 Results and comments2.1 Total estriol in pregnancies

with normal outcome

Tab. III and Fig. l show the total estriol concen-trations (conjugated plus unconjugated steroid) inmaternal plasma related to gestational age for the

Fig. 1. Percentile curves of total estriol in maternal plasmarelated to gestational age in pregnancies with healthy, nor-mal weight infants.

Tab. III. Means, medians, 95% and 70% confidence limits, and ranges of total estriol determined in 707 maternalplasma samples from 463 pregnancies with normal weight, healthy neonates. 95% confidence limits = between 2.5thand 97.5th percentiles, 70% confidence limits = between 15th and 85th percentiles. Conversion to Slunits: l ngE3/ml =3.47 nmolE3/l.

Gestationalweeks

15-1617-1819-2021-2223-2425-2627-2829-3031-3233-3435-3637-3839-4041-42

Number of Total estriol (ng/ml)plasmasassayed Mean Mediän

2851604161526330285775955115

7.9411.6718.3027.0734.1441.4152.5357.75 '72.2489.6499.20

127.06163.52144.15

6.8110.9316.7424.6228.4941.1053.3155.4071.2481.0393.98

118.50160.70139.61

95%C.L.

3- 175- 198- 38

11- 5917- 7919- 8021- 9221- 9422-13339-20139-22543-25953-40573-234

70%C.L.

5- 118- 16

12- 2515- 3821- 4927- 5430- 7434- 8540-10645-13151-13664-18784-26096-210

Min.-Max.

2- 253- 306- 408- 59

13- 8217- 9016- 9917- 9617-13637-23730-28236-36148-43264-241

J. Perina t. Med. 10 (1982)

76 Bacigalupo et al., Total and unconjugated estriol

pregnancies with normal weight and healthy out-come. Taken äs a whole, the values were similar tothe reference values recently reported by FRE-SINSKY, CASTANIER, GRENIER and SCHOLLER[11]. Due to the fact that there were asymmetricaldistributions of the hormone concentrations, theestriol figures were represented äs percentiles. Asvalues of comparison to the medians, the appro-priate arithemetical means were given in the estrioltabulation. The median total estriol concentrationswere in general slightly lower than their means,indicating a denser aggregation of the estriol valuesbelow the means than above them. Commencingwith the gestational weeks 35 to 36, the well-known steep rise in total estriol concentrationscould be noticed. This was associated with a widescatter of single values up to and above 400 ng/ml.

15 17 19Gestational weeks

Fig. 2. Percentile curves of unconjugated estriol in mater-nal plasma related to gestational age in pregnancies withhealthy, normal weight infants.

The 15th percentiles for total plasma estriol whichwere used äs thresholds i« the characterization ofabnormally low steroid levels and äs upper screen-ing limits of a fetal danger zone, showed a markedincrease from 5 to 21 ng/ml early in the 2nd tri-mester. They then lay between 25 and 40 ng/ml inthe gestational weeks 25 to 32, and rose to valuesof approximately 45 to 65 ng/ml between weeks33 and 38. Later on, the 15th percentiles of totalplasma estriol varied between 85 and 95 ng/ml.

2.2 Unconjugated estriol in pregnancieswith normal outcome

Tab. IV and Fig. 2 show the unconjugated estriolconcentrations in mäternal plasma related to gesta-tional age. By comparison of Fig. l and 2 one mayeasily recognize that there was a close parallel inthe shape of the curve äscents for unconjugatedand total plasma estriol during the second andthird trimester, with the average proportion of un-conjugated "estriol varying between 6 and 8% oftotal estriol. The above^mentioned steep rise oftotal plasma estriol at gestational weeks 35 to 36with an associated very wide scatter of someindividual values also occurred in the unconjugatedestriol fraction.The 15th percentiles for unconjugated estriol inmäternal plasma rose steeply from 0.3 to 1.7 ng/mlbetween gestational weeks 15 and 24. During the

Tab. IV. Means, medians, 95% and 70% confidence limits, and ränges of uncorijugated estriol in maternal pfasma fromthe pregnancies with normal weight, healthy neonates.

Gestational weeks Number of Unconjugated estriol (ng/ml)plasmasassayed Mean Median 95%C.L.

15-1617-1819-2021-2223-2425-2627-2829-3031-3233-3435-3637-3839-4041-42

2851604161526330285775955115

0.640.881.451.942.403.043.513.975.056.577.008.349.489.17

0.590.801.362.002.132.643.543.735.256.006.027.87

11.079.87

0.2- 1.50.3- 1.60.5- 3.00.8- 3.01.3- 5.21.4- 6.11.6- 5.81.6- 7.21.5- 7.92.0-12.92.4-13.82.7-17.03.0-20.53.0-17.8

70%C.L.

0.3- 1.00.6- 1.20.8- 2.01.3- 2.61.7- 3.12.1- 3.82.4- 4.62.4- 5.63.4- 6.83.8-10.53.6-10.74.5-11.95.3-13.35.6-14.3

Min.-Max.

0.13- 1.640.26- 2.050.46- 3.900.65- 3.380.90- 5.791.32- 9.051.44- 6.521.09- 9.371.01- 8.401.60-14.102.01-23.702.33-20.402.65-25.203.28-18.90

J. Perinat. Med. 10 (1982)

Bacigalupo et al., Total and unconjugated estriol 77

weeks 25 to 32 they were 2.1 to 3.4 ng/ml; in theweeks 33 to 38, 3.8 to 4.5 ng/ml; and 5.3 to5.6 ng/ml during the last four gestational weeks.

2.3 Total estriol in pregnancieswith intrauterine growth retardation

Tab. V shows the total estriol concentrations inmaternal plasma related to gestational age for thepregnancies with small-for-date babies. From com-parison of these values with those in pregnancieswith normal outcome, it becomes clear that IUGRwas associated with low concentrations of mater-nal plasma estriol, both in the 2nd and in the 3rdtrimester. The total estriol medians in the IUGRgroup lay, with the exception of two values duringearly 2nd trimester, below those in the referencegroup. The values strongly diverged especiallyduring the last 6 to 8 weeks of gestation. In thisperiod, the total estriol median curve of the IUGRgroup lacked the characteristic accelerated increaseto be found in the group of pregnancies withnormal outcome (Fig. 3).The absolute and relative distribution of the in-dividual total estriol concentrations measured inthe IUGR group within the above-described estriolpercentiles System from the pregnancies withnormal outcome is represented in Tab. VII. In theIUGR group (gestational weeks 15 to 42) 36.40%(14.56% plus 21.84%) of total estriol values

moved below the 15th total estriol percentile ofthe reference group with normal outcome. Re-markably, during the 3rd trimester (gestationalweeks 29 to 42) the relative frequency of totalestriol values from the IUGR group below the15th reference estriol percentile was 41.27%(16.67% plus 24.60%); whereas, during the 2ndtrimester (gestational weeks 15 to 28), it was only28.75 % (l l .25 % plus 17.50%) (Tab. VII).

Fig. 3. Medians of total estriol levels in maternal plasmafrom 120 pregnancies with intrauterine growth retarda-tion Qower solid line) and 463 pregnancies with normaloutcome (upper broken line). Note the remarkably lowtotal estriol increase in the group of intrauterine growthretardation following gestational week 30.

Tab. V. Means, medians, 95% and 70% confidence limits, and ranges of total estriol determined in 206 maternal plasmasamples from 120 pregnancies with intrauterine growth retarded infants.

Gestationalweeks

15-1617-1819-2021-2223-2425-2627-2829-3031-3233-3435-3637-3839-4041-42

Number of Total estriol (ng/ml)plasmasassayed Mean Median

35

1213101918184

172630238

94512.3911.9819.0327.5043.8251.8147.4849.6760.6378.0485.1395.3877.58

10.3112.2312.1917.9923.9135.6640.5540.1049.9854.63.68.3775.3677.9661.93

95%C.L.

4- 135- 185- 197- 36

14- 5918- 9423- 9915-10122- 7627-12130-15726-17735-20951-131

70%C.L.

5- 128- 169- 159- 29

15- 4128- 6829- 7926- 7230- 6941- 8039-11552-12852-15956- 81

Min.-Max.

3- 144- 194- 176- 37

13- 6412- 9919-10912-10620- 7725-12925-15923-24722-24350-177

J. Perinat. Med. 10 (1982)

78 Bacigalupo et al., Total and unconjugated estriol

Jϊ0

11M£οα)οι!*11·- §Α Χ+2 ο

«1"ο "c gG ^

£"££ 'S||JCO g"o ^

<1> oo— « o*-? ,r32 "

'•B 1U Λcdε »s•2 w^c^3 >ffi•S .s00 -g•p.SP8*c .-,Ό §C grt C|£«4-1 00O O1-81s11Ud O.-P ».55 oo*+3 O

*O ^

i!i!ll

H

gf£ 8

10

ON

Λ

10

*T00

IIc 88 ·4-> ΙΟM g oo

"3 ε *""| §2 ** ^co ^ |

•S, s *°•S" ! ^Hl.Ξ, 28 _* V)

ε ^£ < v

*o•F**O\Λ

Ό:ON|

OO

Dε83 00Ο ι

5 -a -o3 | -·1 i•s'! ·?U ^ |

'S « 10•C3 Λ\ «ΟSS 'S ^«t c1 &ja o

*e3 *°.•M ^ rv]o o °*H JH WVM V

* «R

S* ^ o

* ^ ^»o c^ oO| /-s O\ /~* OO X-Nι-ί S' CO iS VO 2

^ *' ^O /-^ t-H /-v OO /^IO ^t r-· Γ^ OO »H

• CO * Γ** * *H

Tt VO ^^ *O

* * *vo ON oo^ co · cn ^ vo<N <SG <ΝΪ3

^ ^O Tj- *O»o /*^ ^ s-** vo - *^r^ ^" r^ ® i-H ^"t-H Ci' C!' ^H

_ _ _

s. s. ^vo r^ σ>^ C7^ *~i x~s °^ ίί^ίο 3^ cn S* co SS

0 0 0

Ό ο\ r*·C** /* N 1 ) /™\ f^ /^N00 & i-H Ci -^ S'

^ ^ ^o r- Tf /^^O /— ^^ /βχ CO ^r«^^ ^S "53^O Cx ΙΟ ν-ί' »O s-/

O O Tl-l/^ /~s VO ^ OO /~s^ Tf ^ rH ^ 10

^ ϊ£»Ο Γ^ VOCN VO -^^ »Ο /·

00 <N <N

1 1 1*o o\ »o

I I I> 0 S vo S vo

O <N O^J 00 4J *-< ^J <S25 n ^ n *£ κO " β} II ej II0 c c3 c 0 c

2.4 Unconjugated estriol in pregnancieswith intrauterine growth retardation

The concentrations of unconjugated estriol inmaternal plasma from pregnancies with IUGRshowed a rather similar trend towards their totalestriol counterparts. They were substantially lowerin almost all gestational weeks of the 2nd and 3rdtrimester compared with their reference val es ofthe normal outcome group (Tab. VI). The uncon-jugated estriol medians of the IUGR group lay,with the exception of one value in the early 2ndtrimester, always below those f the normal out-come group. As early s the 2nd trimester thedivergence of the two median curves was veryclear, while in the 3rd trimester it did not appearto be s accentuated s it was in the case of totalestriol (Fig. 4).

Unconiugated Estnol ng/ml12 η

11 -

10 -

9 -

8 -

7 -

B -

5 -

4 -

3 -

2 -

1 -

(0 -

1

Λt \1 ^///

ιfi A^/ χΝ»jf

s^"~'^^r't ^Jjf

/ f ^/ /,*-* J

»'&2S~~^^s JG ^^^

5 17 19 21 23 25 27 29 31 33 35 37 30 41Gntational weeks

Fig. 4. Medians of unconjugated estriol levels in maternalplasma. Same patients s represented in Fig. 3. Lowersolid line = IUGR pregnancies, upper broken line = preg-nancies with normal outcome.

As can be seen from Tab. VI, VII, in the IUGRgroup 38.83% of unconjugated estriol val es(1 1.65% plus 27.18%) could be detected below the15th estriol reference percentiles (gestationalweeks 15 to 42), while the corresponding value fortotal estriol was 36.4% (no statistically significantdifference between the two val es).

J. Perinat. Med. 10(1982)

Bacigalupo et al., Total and unconjugatcd estriol 79

During the 2nd trimester the relative frequency ofunconjugated estriol values in the IUGR groupencountered below the 15th estriol referencepercentile was 46.25% (17.50% plus 28.75%).This is in contrast to only 34.13% during the 3rdtrimester (7.94% plus 26.19%); (Tab. VII).

2.5 Total and unconjugated estriol in pregnancieswith intrauterine growth retardation

Fig. 5 demonstrates that in the course of the 2nds well s of the 3rd trimester both total and un-

conjugated plasma estriol levels were significantlylower in the IUGR group compared with theestriol levels in the group with normal outcome.Separate evaluations of total and unconjugatedestriol concentrations in women with severe ormild forms of fetal growth retardation (infantsbelow the 3rd or the lOth birthweight percentiles)did not reveal any clear differences between the

Fig. S. Paircd total estriol and unconjugated estriol levels(mean ± S.E.M.) in maternal plasma determined in thegestational wceks 19 to 20 and 37 to 38 in women with anormal or growth retarded fetus. Pregnancies with agrowth retarded fetus have significantly lower total andunconjugated estriol levels (χ -test) both in the 2nd andin the 3rd trimester, compared with the estriol leveis inwomen with a normal fetus.

Total Estriol ng/ml160 η150 -140 -130 -120 -110-100-90 -80 -70 -60 -60-40 -30-20-10 -

12 η11 -10-9 -8 -7 -6 -5 -4 -3 -2 -1 -

p < 0.005127.0616.59 85.13t8.18

p < 0.0118.30tO.94 11.98H.10

IP rofiro

Κ*Ώ

11s§1Unconjugated Estriol ng/ml

p < 0.0058.34tQ.40 6.16t0.47

T

p < 0.011.45t0.09 0.87*0.12_^«_

κίκίΐ fnftn

ix*^

^vSs1 I

1TB

Normal (n =60) IUGR (n=17) Normal (n -95) IUGR (n=30)Gest. weeks 19-20 Gest. weeks 37-38

Tab. VI. Means, medians, 95% and 70% confidence limits, and ranges of unconjugated estriol determined in 206 mater-nal plasma samples from 120 pregnancies with intrauterine growth retarded infants.

Gestationalweeks

15-1617-1819-2021-2223-2425-2627-2829-3031-3233-3435-3637-3839-4041-42

Number of Unconjugated estriol (ng/ml)plasmasassayed Mean Medi n 95%C. L.

35

1213101918184

172630238

0.400.990.871.192.092.262.883.003.274.825.766.167.247.25

0.331.060.961.351.911.94

. 2.712.703.265.035.456.207.206.86

0.1- 0.70.4- 1.50.3- 1.50.3- 2.20.9- 4.01.0- 4.80.9- 5.11.2- 7.12.7- 3.92.0- 7.61.9-11.82.6-11.62.5-13.74.7-10.4

70%C.L.

0.2- 0.60.6- 1.30.5- 1.30.5- 1.81.2- 3.21.3- 2.81.6- 4.21.8- 3.92.8- 3.83.0- 6.93.1- 8.93.6- 8.04.0-10.05.6- 9.1

Min .-Max.

0.09- 0.750.36- 1.500.28- 1.530.31- 2.360.89- 4.131.07- 5.250.82- 5.300.90- 8.512.64- 3.921.51- 7.611.78-13.502.45-14.672.16-14.094.50-10.65

J. Perinat. Med. 10 (1982)

80 Bacigalupo et al., Total and unconjugated estriol

hormone levels in the two weight categories ofIUGR pregnancies. Nor did the estriol levels frompregnancies with "pure" IUGR or IUGR asso-ciated with toxemia characteristically differ fromeach other.The failure of serial low plasma estriol levels toshow the characteristic rise which is commonlyobserved during the 3rd trimester of pregnancieswith normal outcome was described by KLOPPER,JANDIAL and WILSON [21] äs a reliable sign ofintrauterine growth retardation. Fig. 6 shows thata similar plasma estriol conduct may be seen al-ready during the 2nd trimester öf pregnancies withIUGR.

17 18 19 20 21 22 23 24 25 26 27 28

Fig. 6. Serial low unconjugated plasma estriol levels withreduced rising tendency (solid line) during the 2nd tri-mester in a 42 year-old women with a growth retardedfetus. At gestational week 38 delivery of a male infant of2680 g in good health. - Broken lines = 2.5th and 50thunconjugated estriol percentiles in the group of preg-nancies with normal outcome.

3 Discussion

The paired hormone assays demonstrated thatboth total estriol and unconjugated estriol con-centrations in maternal plasma tended to be low inpregnancies with intrauterine growth retardation.Under the premise that the 15th percentiles ofmaternal plasma estriol concentrations in preg-nancies with normal outcome were used äs screen-ing limits, 36.40% of the blood specimens fromIUGR pregnancies had abnormally low concentra-tions of total estriol, and 38.83% of them hadabnormally low concentrations of unconjugatedestriol in the course of the 2nd and 3rd trimester.In the 2nd trimester the predictive value for IUGRof unconjugated plasma estriol was higher than

that of total plasma estriol. In the 3rd trimesterthe Situation was reversed. A clear superiority inpredicting IUGR by unconjugated plasma estriol(in comparison with total plasma estriol), äs hasbeen claimed on theoretical considerations bysome authors [30], could not be confirmed in thisstudy.The prediction of fetal growth retardation byplasma estriol estimations could be made with asimilar degree of probability during the 2nd tri-mester äs during the 3rd; an observation that mayattract the attention of the obstetrician and per-haps of the perinatologic pathologist, since itsheds some light on the early onset of this gesta-tional complication.Studies of plasma estriol with the aim of predect-ing or monitoring intrauterine growth retardationhave been condücted repeatedly [l, 3,12,15,17,18, 21, 22]. All the$e üivestigations, which dealtalmost entirely with IUGR cases in late pregnancy,led to the unequivocal conclusion that in preg-nancies with retarded fetal growth the maternalestriol values were commonly lower than normaland that the estriol curves obtained by serialplasma estriol measurements were flattened incomparison with those in pregnancies of com-parable gestational age with normal outcome.Less unanimity among the authors existed äs tothe rate of true IUGR prediction by plasma estriol.The quotation of accurate predictions by plasmaestriol assays ranged from 20% reported by Tow-LER et al. [28] to an astonishingly high 86% re-ported by GOEBEL et al. [13] in cases of IUGRwithout hypertension. Evidently some of the dis-crepancies in the assessment of the predictive valüeof plasma estriol measurements were due to appli-cation of different criteria in defining abnormallylow estriol levels or birthweight limits for small-for-dates infants or the use of different hormoneassay procedures.Between the plasma estriol concentrations inpregnancies with normal outcome and those withIUGR there was a large area of overlapping, a factwhich logically limits the diagnostic usefulness ofthe measurements. It presently often remainsobscure why in one individual case of IUGR theclassical low estriol values are found and in anothercase they are not. In this connection one should

J. Perinät. Med. 10 (1982)

Bacigalupo et al., Total and unconjugated estriol 81

keep in mind that the clinical syndrome of intrau-terine growth retardation can result from diseasesof the mother, the placenta and the fetus, or fromspecial circumstances relating to genetics or mater-nal environmental setting.

Taking these various roots of IUGR into considera-tion, it should not be surprising that only a pro-portion of all IUGR cases can be discovered byfeto-placental function tests such äs maternalplasma estriol assay. For example, in a pregnancycomplicated by ample placental infarcts and asubstantial reduction in maternal-placental Sub-strate and gas exchange, the investigator mayeasily find retarded fetal growth associated withabnormally low plasma estriol levels. On the otherhand, a pregnant woman who is a severely un-dernourished drug user with a growth retardedfetus may have normal plasma estriol levels.

In the 2nd and 3rd trimester of gestation maternalplasma estriol is almost entirely a product of theconceptus. Being synthesized by complementarysteroidogenic mechanisms in the fetus and pla-centa, estriol is released from the circulation of theconceptus into the circulation of the pregnant wo-

Summaiy

In 913 plasma samples collected from 583 mothers in the2nd and 3rd trimester of pregnancy (gestational weeks 15to 42), paired determinations of total and unconjugatedestriol levels were performed by radioimmunoassay. 707of the samples were taken from 463 mothers who weredelivered of healthy singleton normal birthweight infants.The remaining 206 plasmas assayed came from 120 preg-nant women who gave birth to singleton small^for-dateinfants whose birthweight feil below the lOth percentilefor gestational age and who were therefore classified ässuffering from intrauterine growth retardation (IUGR).From the levels encounteied in the pregnancies withnormal weight outcome, reference values such äs medians,means, 95% and 70% confidence limits of total and un-conjugated plasma estriol were calculated for the 2nd and3rd trimester. Due to the asymmetrical distribütion of theestriol data, their mathemätical representation was pre-dominantly done in percentiles. Values which feil belowthe 15th plasma estriol percentile for gestational age wereregarded äs abnormally low and indicators of .„Potentialfetal danger.In pregnancies with IUGR, total and unconjugated estriolin maternal plasma tended to be lower than in pregnancieswith normal weight offspring. The medians of total andunconjugated plasma estriol from the IUGR group lay

man. Regulär feto-placental synthesis of estriolrequires a sufficient supply of oxygen and steroidalprecursors; thus, plasma estriol assays may be usedin the assessment of feto-placental function duringthe 2nd and 3rd trimester of pregnancy [6], Nor-mal maternl plasma estriol concentrations with arising tendency in the progression of gestationindicate good outcome.Since plasma estriol measurements can now becarried out readily, precisely, repeatedly and with-out major molestation to the mother, they are avaluable tool in monitoring and managing preg-nancies with chronic complications such äs intrau-terine growth retardation, placental insufficiencyor postmaturity. Assays of total estriol in amnioticfluid seem to be superior to maternal plasmaestriol assays in predicting IUGR in late gestation[2]. Of course, the combined application of differ-ent clinical tests such äs ultrasonography, fetalheart monitoring, HPL assays, total estriol, lecithindeterminations in amniotic fluid and plasma estriolassays will provide the obstetrician engaged in themanagement of an IUGR pregnancy with muchmore useful data than would be obtained with anestriol determination alone.

usually below those from the normal birthweight refer-ence group. During the third trimester the rate of ab-normally low total estriol levels from the IUGR groupamounted to 41%, while that of abnormally low un-conjugated estriol levels was 34%. During the second tri-mester 29% of the total estriol values and 46% of the un-conjugated estriol values from the IUGR group feil belowthe corresponding 15th estriol percentiles. Generally, therate of abnormally low maternal plasma estriol levelsfrom the IUGR group in the 2nd and 3rd trimester wasalmost identical — between 35% and 40% — for the totaland the unconjugated estriol.Three riiain conclusions could be drawn from this study:1. Measurements of total and unconjugated estriol inmaternal venous plasma during the 2nd and 3rd trimesterof pregnancy were of comparable clinical usefulness. Theoveräll effiäency of the two estriol tests in discriminatingpregnancies with IUGR was 36% for total estriol and38% for unconjugated estriol. 2. Predictions of IUGR byabnormally low plasma estriol levels could be made witha similar rate of accuracy both during the 2nd and 3rdtrimester. 3. Plasma estriol determinations appeared to bemost useful when normal estriol results provided somereassurance to the obstetrician that the current Status ofthe fetus was satisfactory.

Keywords: Birthweight, intrauterine growth retardation, maternal plasma, monitoring of pregnancy, total estriol, un-conjugated estriol.

J. Perinat. Med. 10 (1982)

82 Bacigalupo et al., Total and unconjugated estriol

Zusammenfassung

Paarige Bestimmungen des Gesamtöstriols und unkonju-gierten Östriols im mütterlichen Plasma im 2. und 3. Tri-menon - Ihre Beziehung zur intrauterinen WachstumsletardierungIn 913 Plasmaproben von 583 Müttern im 2. und 3. Tri-menon (Schwangerschaftswochen 15 bis 42) wurden paa-rige RIA-Bestimmungen von Gesamt- und unkonjugiertemÖstriol vorgenommen. 707 Proben stammten von 463Müttern, die von gesunden, normal wiegenden Kindernentbunden wurden. Die übrigen 206 untersuchten Plas-men waren von 120 Schwangeren, die small-for-date Kin-der gebaien, deren Geburtsgewicht unterhalb der demSchwangerschaftsalter entsprechenden 10. Perzentile lagund die deshalb als Kinder mit intrauteriner Wachstums-retardierung klassifiziert wurden.Aus den Hormonwerten der Schwangerschaften mit nor-mal wiegenden Kindern wurden Referenzwerte berechnet,d.h. Mediane, Mittelwerte, 95% und 70% Vertrauensbe-reiche von Gesamt- und unkonjugiertem Östriol für das2. und 3. Trimenon. Wegen der asymmetrischen Vertei-lung der Werte erfolgte ihre mathematische Darstellungvorwiegend in Perzentilen. Werte, die sich unterhalb derdem Schwangerschaftsalter entsprechenden 15. Plasma-östriol-Perzentile bewegten, wurden als abnorm niedrigangesehen und als Hinweiszeichen potentieller Gefähr-dung des Feten gewertet.In Schwangerschaften mit intrauteriner Wachstumsretar-dierung tendierten Gesamtöstriol und unkonjugiertesÖstriol im mütterlichen Plasma zu niedrigeren Werten alsin Schwangerschaften mit normal wiegender Nachkom-menschaft. Die Östriol-Mediane der Schwangerschaftenmit intrauteriner Wachstumsretardierung lagen gewöhn-lich unter den Östriol-Medianen der Schwangerschaften

mit normalem Geburtsgewicht der Kinder. Während des3. Trimenons betrug der Anteil der abnorm niedrigen Ge-samtöstriol-Werte bei den Schwangerschaften mit intrau-teriner Wachstumsretardierung 41%, während der Anteilder abnorm niedrigen Werte für unkonjugiertes Östriol34% ausmachte. Während des 2. Trimenons lagen 29% derGesamtöstriol-Werte und 46% der Werte für unkonjugier-tes Östriol in der Gruppe der Schwangerschaften mit in-trauteriner Wachstumsretardierung unter den entsprechen-den 15. Östriol-Perzentilen der Referenzgruppe. Betrach-tete man die Östriol-Werte aus dem 2. und 3. Trimenongemeinsam, dann war der Anteil von abnorm niedrigenÖstriol-Werten in der Gruppe der Schwangerschaften mitintrauteriner Wachstumsretardierung für Gesamtöstriolund unkonjugiertes Östriol nahezu identisch (zwischen35% und 40%).Drei hauptsächliche Schlußfolgerungen wurden aus dervorliegenden Studie gezogen: 1) Bestimmungen vonGesamtöstriol und unkonjugiertem Östriol im mütterli-chen venösen Plasma waren während des 2. und 3. Tri-menons von vergleichbarem klinischem Nutzen. Bei derAbgrenzung von Schwangerschaften mit intrauterinerWachstumsretardierung betrug die Gesamttrefferquoteder zwei Östrioltests 36% für Gesamtöstriol und 38% fürunkonjugiertes Östriol im Plasma. 2) Vorhersagen vonintrauteriner Wachstumsretardierung ließen sich aufgrundvon abnorm niedrigen Östriolkonzentrationen im mütter-lichen Plasma sowohl im 2. wie auch im 3. Trimenon mitähnlicher Trefferquote machen. 3) Östriolbestimmungenim mütterlichen Plasma schienen sehr nützlich zu sein,wenn die normalen Östriol-Werte dem Geburtsmedizinereinige Sicherheit vermittelten, daß der aktuelle Zustanddes Fetens zufriedenstellend war.

i H··

Schlüsselwörter: Geburtsgewicht, Gesamtöstriol, intrauterine Wachstumsretardierung, mütterliches Plasma, Schwanger-schaftsüberwachung, unkonjugiertes Östriol.

Resume

Determinations couplees de Festriol plasmatique materneltotal et non conjugue pendant le deuxieme et le troisiemetrimestre de la grossesse - Relations avec le retard decroissance intra-uterinLes dosages couples de Festriol total et de l'estriol nonconjugu6 ont ete realises par radio-immunologie sur 913plasma de 583 meres au couis du 2eme et 3eme trimestrede la grossesse (15eme a 42eme semaines gestation-nelles) - 707 echantfflons plasmatiques ont ete preleveschez 463 meres ayant accouche d'enfants uniques depoids de naissance normal et en bonne sante. Les 206restants ont ete preleves chez 120 femmes enceintes ayantaccouche d'enfants uniques hypotrophiques avec un poidsde naissance inferieur au dixieme percentile pour Tagegestationnel et ainsi consideres comme ayant subi unretard de croissance intra-uterin (RCIU).

A partir des resultats obtenus pour les grossesses ayantabouti a un nouveau-ne de poids normal, on a calcule desvaleurs de reference telles que les moyennes, la validite,les intervalles de confiance a 95 % et 70%, pour les estriolstotaux et les estriols non conjugues au second et autroisieme trimestre. En raison de la distribution

asymetrique des resultats, leur expression mathematiquea ete effectuee essentiellement en percentiles. Les valeurssituees sous le 15eme percentile de Festriol plasmatiquepour Tage gestationnel ont ete considerees commeanormalement basses et comme indicatrices de dangerpotentiel pour le foetus.Lors des grossesses avec RCIU, l'estriol plasmatiquematernel total et non conjugue tend ä etre plus bas quelors des grossesses ayant abouti a un enfant de poidsnormal. Les moyennes d'estriol plasmatique total et nonconjugue dans le groupe avec RCIU sont en regle inferi-eures a celles du groupe de reference sans hypothrophie.Pendant le troisieme trimestre le taux d'estriol totalanormalement bas est de 41% dans le groupe avec RCIU,alors que le taux d'estriol total non conjugue est de 34%.Pendant le second trimestre, 29% des valeurs de l'estrioltotal et 46% des valeurs de Festriol non conjugues sontatues sous le 15eme percentile dans le groupe avec RCIU.Globalement, la frequence de taux d'estriol anormalementbas dans le groupe avec RCIU pendant le 2eme et le 3emetrimestre est ä peut pres la meme — entre 35 et 40% —pour Festriol plasmatique maternel total et pour Festriolnon conjugue.

J. Perinat. Med. 10 (1982)

Bacigalupo et al., Total and unconjugated estriol 83

On peut tirer trois conclusions principales de ce travail:1. les dosages de Pestriol plasmatique maternel total et del'estriol non conjugue revetent une utilite cliniquecomparable pendant le 2&me et le 3&me trimestre.L'efficacite globale des deux dosages poui determiner lesgrossesses avec RCIU est de 36% pour l'estiiol total et de38% poui l'estriol non conjugue. 2. La prevision du RCIU

par la determination de taux d'estriols plasmatiquesanormalement bas peut etre portee avec une fiabiliteidentique pendant le second et le troistöme trimestre.3. Cest lorsque les resultats sont normaux que lesdosages de l'estriol plasmatique revetent la plus grandeutilite en assuiant quelque peu aux obstetriciens quel'etat actuel du foetus est satisfaisant.

Mots-cles: Estriol non conjugue, estriol total, hypotrophie, plasma maternel, poids de naissance, surveillance de la: grossesse.

Acknowledgement: This study was supported by a grant from the DEUTSCHE FORSCHUNGSGEMEINSCHAFT,Bonn-Bad Godesberg, Federal Republic of Germany.

Bibliography

[1] ALLEN, E. J., G. C. L. LACHELIN: A comparisonof plasma levels of progesterone, oestradiol, un-conjugated oestriol and total oestriol with urinarytotal oestrogen levels in clinical obstetric practice.Brit. J. Obstet. Gynaecol. 85 (1978) 278

[2] BACIGALUPO, G., J. W. DUDENHAUSEN, E. SA-LING: Role of assays of total estriol in amnioticfluid for the diagnosis of fetal hypotrophy. J. Pe-rinat. Med. 5 (1977) 76

[3] BASHORE, R. A., J. R. WESTLAKE: Plasma un-conjugated estriol values in high-risk pregnancy. Am.J. Obstet. Gynecol. 128 (1977) 371

[4] BEARGIE, R. A., V. L. JAMES, J. W. GREENE:Growth and development of small-for-date new-borns. Pediatr. Clin. North Amer. 17 (1970) 159

[5] DEPP, R.: Present Status of the assessment of fetalmaturity. Seminars in Perinat. 4 (1980) 229

[6] DICZFALUSY, E.: Recent progress in the feto-placental metabolism of steroids. In: R. VOKAER,G. de BOCK (eds.): Reproductive endocrinology.Pergamon Press, Oxford 1975

[7] DICZFALUSY, E., S. MANCUSO: Oestrogenmetabolism in pregnancy. In: KLOPPER, A., E.DICZFALUSY (eds.): Fetus and Placenta. Black-well Scientific, Oxford 1969

[8] DRILLIEN, C. M.: The small-for-date infant: Etiol-ogy and prognosis. Pediatr. Clin. North Amer. 17(1970)9

[9] FARR, V., R. G. MITCHELL, G. A. NELLIGAN,J. M. PARKIN: The definition of some externalchaiacteristics used in the assessment of gestationalage of the newborn infant. Develop. Med. Child.Neurol. 8 (1966) 507

[10] FITZHARDINGE, P. M., E. M. STEVEN: The small-for-date infant. II. Neurological and intellectualsequelae. Pediatr. 50 (1972) 50

[11] FRESINSKY, E., M. CASTANIER, J. GRENIER,R. SCHOLLER: L'estriol plasmatique total pendantla grossesse. Path. Biol. 28 (1980) 365

[12] GITSCH, E., H. JANISCH, J. SPONA: HPL andestriol serum levels and placental perfusion äs in-dexes of placental function in EPH gestosis andplacental insufficiency. Gynecol. Obstet. Invest. 11(1980) 102

[13] GOEBEL, R., H. K. RJOSK, E. KUSS: Diagnosti-sche Wertigkeit von Serumoestrogenbestimmungenim Vergleich mit anderen chemischen Überwa-chungsmethoden bei Risikoschwangerschaften. In:DUDENHAUSEN, J. W., E. SALING (eds.): Perina-tale Medizin. Bd. VI. Thieme, Stuttgart 1975

[14] GOEBELSMANN, U.: The uses of oestriol äs amonitoring tool. In: QUILLIGAN, E. J. (ed.): Up-date on fetal monitoring. Clinics in Obstetrics andGynaecology VI/2. W. B. Saunders, London 1979

[15] GORWILL, R. H., J. R. SARDA: Hormonal studiesin pregnancy. II. Unconjugated estriol in maternalperipheral vein, cord vein and cord artery serum atdelivery in pregnancies complicated by intrauterinegrowth retardation. Am. J. Obstet. Gynecol. 127(1977)17

[16] GRUENWALD, P.: Chronic fetal distress and pla-cental insufficiency. Biol. Neonat. 5 (1963) 215

[17] HUGHES, G., B. BISCHOF, G. WlLSON, R. SMITH,A. KLOPPER: Tests of fetal wellbeing in the thirdtrimester of pregnancy. Brit. J. Obstet. Gynaecol. 87(1980)650

[18] ISOUARD, G.: Value of total serum oestriol andhuman placental lactogen in the assessment of foeto-placental function. Aust. N. Z. J. Obstet. Gynaecol.19 (1979) 69

[19] KEUTH, U.: Pränatale Dystrophie. In: OPITZ, H.,F. SCHMID (eds.): Handbuch der Kinderheilkunde1/2. Springer, Berlin 1974

[20] KLINGMÜLLER-AHTING, U., E. SALING, J. GIF-FEI: Frühgeburt und intrauterine Mangelentwick-lung. Gynäkologe 8 (1975) 186

[21] KLOPPER, A., V. JANDIAL, G. WlLSON: Plasmasteroid assay in the assessment of foetoplacentalfunction. J. Steroid Biochm. 6 (1975) 651

[22] KÜNZIG, H. J., W. GEIGER: Estrogens. Contr.Gynec. Obstet. 2 (1976) 2

[23] MAIER, U. E. K.: Prä- und postnataler ftotein-Ka-lorien-Mangel und Gehirnentwicklung. Fortschr.Med. 93 (1975) 726, 771, 821 (3 parts)

[24] MILLER, H. C.: Fetal growth and neonatal mor-tality. Pediatr. 49 (1972) 392

J. Perinat. Med. 10 (1982)

84 Bacigalupo et al., Total and unconjugated estriol

[25] NICKL, R.: Standardkurven der intrauterinen Ent-wicklung von Gewicht, Länge und Kopfumfang.Inaugural-Dissertation, München 1972

[26] PETRUSSA, J.: A scoring System.for the assessmentof gestational age of newborn infants. In: HUN-ilNGFORD, P. J., R. W. BEARD, F. E. HYTTEN,J. W. SCOPES (eds.): Perinatal Medicine. Proc. 2ndEurop. Congr. Perinatal Medicine 1970. Karger, Ba-sel 1971

[27] SALING, E., K. H. WULF: Zustandsdiagnostik beimNeugeborenen. Fortschr. Med. 89 (1971) 12

[28] TOWLER, C. M., C. H. W. HÖRNE, V. JANDIAL,D. M. CAMPBELL,;]. MACGILLIVRAY: Plasmalevels of pregnancy-specific fr-glycoprotein in coiii-plicated pregnancies. Brit. J. Obstet. Gynaecol. 84(1977)258

[29] TULCHINSKY, D., G. E. ABRAHAM: Radio-immunoassay of plasma estriol. J. Clin. Endocrinol.Metab. 33(1971)775

[30] WOLFRUM, R., U. GETHMANN: dU-Gesamtöstro-gene, S-unkonjugiertes Östriol und S-Gesamtöstriolbei Hypotrophie. Z. Geburtsh. Perinat. 184 (1980)413

Received and accepted September 25,1981.

Prof. Dr. med. G. BacigalupoArbeitsgruppe Perinatale MedizinFreie Universität BerlinMariendorfer Weg 28-38D-1000 Berlin 44

J. Perinat. Med. 10 (1982)