The relationship of blood lead levels to obstetric outcome

The relationship of blood lead levels to obstetric outcome NORMAN F. ANGELL, PH.D.

J. PATRICK LAVERY, M.D., F.A.C.O.G. Louisr•ill~.

Knttucky

Lead represents a significant environmental hazard to pregnant women and their offspring. Exposure to high environmental levels of lead has been associated with spontaneous abortion, premature rupture of fetal membranes (PROM), and preterm delivery. The relationship between lower exposures and obstetric complications is unknown. The concentratiOn of lead in the blood was measured in 635 specimens of umbilical cord blood collected at delivery. No relationship was found between concentrations of lead in cord blood and the incidence of PROM, preterm delivery, preeclampsia, or meconium staining. Maternal and infant capillary blood wu collected 24 hours post partum from 154 of these deliveries. The concentrations of lead in the blood did not vary significantly among cord, infant, and maternal samples, and the three measurements were highly correlated. Levels of zinc protoporphyrin (ZnP) were increased in cord blood as compared with mothers' blood, but no concentration-response relationships between the ratio of cord ZnP to maternal ZnP and lead were found. (AM. J. OesTET. GYNECOL. 142:40, 1982.)

effects of high concentrations of lead in the blood during pregnancy are well documented.' but the toxic effects of exposure to lower levels have received relatively little attention. Lead may have important toxicity in the absence of clinical symptoms.~ and this toxicity may be manifested years after exposure. Fetal injury may be the consequence of exposure to levels of lead which cause no signs or symptoms in the mother. Lead inhibits several enzymes in the heme biosynthetic pathway. This can result in anemia. One of these enzymes is heme synthetase, which is responsible for the incorporation of iron into the protoporphyrin ring. In the presence of lead, the immediate precursor of hemoglobin, protoporphyrin IX, or zinc protoporTHE ADVERSE

From lhP Departm.ent of Obstetrics and G.ynecology, U nh•enity of Louisville School of Medicine. Support for this project was received from the Jefferson County Department of Public Health, Childhood Lead Poisoning Program, and the William Randolph Hearst Foundation. Receil'ed for publication june 26, 1981. Amptedjuly29, 1981. Reprint requests: J. Patrick Lavery, M.D., DepaTtment Obstetrics and Gyrwcology, 323 East Chestnut St., Louisville, Kentucky 40202.

40

of

phyrin (ZnP), is found in increased concentration. This elevation of the concentration of ZnP in the blood may be the earliest significant hematologic response to exposure to lead, 3 a phenomenon which may be accentuated by the increased demand for hemoglobin duF ing pregnancy. Several other factors increase the danger of exposure to lead during pregnancy. Animal data indicate that diets which are high in fat or low in protein or minerals result in an increased absorption of lead. 4 There is also evidence that deficiencies in metals, such as iron, zinc, or copper, may increase the toxic effects of lead. 4 Thus, poor eating habits during pregnancy mav result in increased absorption of dietary lead.;; Preterm delivery and premature rupture of fetal membranes (PROM) account for approximately 75% of morbidity and mortality in the perinatal period. Clinical and experimental evidence links exposure to high levels of lead to these conditions. Fahim and associates6 found twofold io fourfold higher concentrations of lead in the blood of women with PROM or preterm delivery than in women without these complications. However, the concentrations oflead in the blood found in this study were considerably higher than those found elsewhere in the United States and Europe .7 - 10 The association of lead, PROM, and preterm delivery at levels of exposure more representative of those 0002-9378/82/010040+07$00.70/0 © 1982 The C. V Mosby Co.

Relationship of blood lead to obstetric outcome 41

Volume 142 Number I

found in a population of urban women has not been studied. Animal studies have shown that lead interferes with collagen synthesis, 11 energy metabolism, and membrane structure, 12 all of which may adversely affect the integrity of the chorioamniotic membrane and predispose to PROM. The role of lead in other complications of pregnancy, such as preeclampsia or meconium staining, remains unknown. Metropolitan Louisville, Kentucky, is currently out of compliance with Environmental Protection Agency (EPA) standards for atmospheric lead, 13 thereby placing city residents at risk for increased absorption of lead. The lack of data concerning the relationship between exposure to low levels of lead and PROM, preterm delivery, and other obstetric complications in the high-risk population served by our hospital prompted us to undertake the present study in collaboration with the Jefferson County Department of Public Health. Material and methods The subjects studied were women who underwent delivery at Louisville General Hospital, and their babies. This hospital serves mainly an indigent, urban population. All subjects gave their informed consent according to approved procedures of the Human Studies Committee of the University of Louisville. No selection process was employed. Inclusion in the study was based on the availability of specimens obtained. In 635 deliveries. whole blood was collected at delivery from the umbilical vein by disposable plastic syringe and transferred to vacuum blood collection tubes (Venoject) which contained liquid ethylenediaminetetraacetic acid (EDTA), as an anticoagulant, and potassium sorbate, as an antimycotic agent. In 154 of the deliveries, infant capillary blood was collected 24 hours post partum by finger stick in mothers and by heel stick in neonates in heparinized, lead-free blood collection tubes (Walter Sarstedt) after cleansing of the skin with an alcohol swab. All samples were immediately refrigerated and were analyzed within 3 days of collection. Analysis of whole blood was done by flameless atomic absorption spectrometry with the use of a PerkinElmer Model 5000 instrument equipped with a model HGH-2200 graphite furnace. A modification of the method of Fernandez 14 was employed. Twenty-five microliters of whole blood was diluted with I 00 J.Ll of 0.1% of Triton X-100 solution. The concentration of lead was determined from measurements of the curve peak area. Most results were reported as the mean of two determinations. Zinc protoporphyrin was measured directly by hematofluorometer in 40 J.Ll of whole blood. The instru-

Table I. Characteristics of the study population Motlu!rs: No. of cases Age (Mean :t SD) Range Race(%) Black Caucasian Other Parity: Primiparous Multiparous Hemoglobin concentration (gm/dl) (Mean± SD)

635 21.1 ± 4.9 13-40

Gestational age (wk) (Mean ± SD) Birth weight (gm): Mean± SD Range Cord hemoglobin (gm) (Mean ± SD)

39.6 ± 2.5

Babies:

Incidence of complications studied(%):

46.9 52.4 0.7 42.7 57.3 12.1 ± 1.6

3,211 ± 557 1,304-4.763 15.0 ± 1.6 9.0 (56 cases)

PROM Preterm delivery Preeclampsia Meconium staining

11.8 (74 cases) 5.0 (32 cases) 19.4 (123 cases)

Table II. Concentration ( J.Lg/dl) of lead in cord blood (Mean ± SD) Complicatitm Premature rupture of membranes Preterm delivery Preeclampsia Meconium staining

Affected 9.8 :!: 3.6 ± 4.5 8.7 ± 3.6 9.1:!: 3.9

10.1

I

Nonaffected 9.5

;i:

4.4

9.2 ± 3.9

9.5 ± 4.0 9.4 ± 3.9

Mean concentrations of lead in cord blood in women with each complication compared to those in women who did not exhibit that complication.

ment was calibrated by means of the extraction method of Piomelli. 15 All analyses were performed in the Jefferson County Health Laboratory. Preterm delivery was defined as delivery at less than 38 weeks' gestation. PROM was considered to be rupture that occurred more than 2 hours before the start of labor. Consultants of the Health Sciences Computer Center of the University of Louisville used the Biochemical Co111puter Program (BMD) for statistical analysis. Results Characteristics of the study population. Table I lists characteristics of the mothers and infants in this study, and the incidence rates of the studied complications of pregnancy. Lead in blood. Fig. 1 shows the concentrations of lead in the blood from 154 sets of cord-mother-baby measurements. The correlation coefficients were highly

42 Angell and Lavery

J.muan I. I YH::!

An1. J ObsteL Gvneml.

iS .....

40

j

~ !ii a:

30

w

20

~

z

~

(.)

c < w

10

_,

_,c

0

:t

(.)

0

CD

10

30

20

40

CORD LEAD CONCENTRATION (ug/dl)

40

40

iS ....

01

~

~ !ii

30-

r: zw

20-

10-

30

,

. .~:.:~::

20

~

(.)

..

~

:._·/_:::t .. ·:. . :.

10

w _,

c o;-~~~~,~~-,~~~~,~~~

0

(!)

10

20

30

40

MOTHER LEAD CONCENTRATION (U91dl)

i

0

(.)

0

0

10

20

30

40

MOrtiER LEAD C81111Catl'RA1'JQN (_...)

Fig. l. Concentrations of lead measured in cord, child, and maternal blood. Correlations and regTession equations are: A, Cord-child: r 0.77: y 0.90x + 1.04.B, Mother-cord: r 0.60, y 0.55x + 0.73. y 0.78x + 2.3. ~.:~. C, Mother-child: r

significant (p < 0.01) and indicate high concordance of the three measurements. Mean concentrations (J.tg/dl) (±SD) of lead in the blood were: cord, 9.73 ± 4.1: mother, 9.85 ± 4.4; child, 9.82 ± 4.8. These did not differ significantly (analysis of variance). Relationships of concentration of lead in blood and complications of pregnancy. There were no statistically significant differences in the mean concentrations of lead in cord blood between women with PROM, preterm delivery. preeclampsia, or meconium staining and women who did not suffer these complications (Table II). Cumulative percentage curves were calculated for each complication and are shown in Fig. 2. No significant differences between these curves were found.

Dose-response curves suggest a threshold effect at a concentration of lead of approximately 25 J.tg/dl for meconium and preterm delivery (Fig. 3). Although there were too few cases in this concemration category to allow a conclusive statement to be made. it is intrigu" ing that both women who had concentrations of lead in the blood greater than 2;> J.tg/dl had a complication. Zinc protoporphyrin. The relationship between concentrations of lead and concentrations of Zni> in cord blood and maternal blood is shown in Fig. 4, A and B. The mean concentration of ZnP in cord blood was 41.2 J.tg/dl ± 17.0, whereas the mean concentration of ZnP in maternal blood was 28.6 J.tg/dl.± 12.9. The difference in means was statistically significant (analysis of variance, p < 0.01). However. the mcrease in /nP in

Relationship of blood lead to obstetiic outcome

Volurne i42

43

'\umber I

100

90 80

.... z

I

I

"'a: (.)

•

"'"-

"'>

~..J

;:,

e-PROM

I

I

15

20

25

e-PRETERM

:::E ;:,

8-NO PROM

(.)

8-TERM

5

>25

l

w

70

....z

....~

""

..J

;:,

:::E

•

70

w

(.)

60

a: 60

...

Ill

Q.

w

>25

80

zw

a:

I

25

90

80

(.)

I

20

1001

90

....

I

15

CORD LEAD CONCENTRATION (ug/dl)

CORD LEAD CONCENTRATION (ugtdll

100

I

10

w

50

>

50

.... ""

40

>=

40

e-MECONIUM

;:,

8 - NO MECONIUM

(.)

;:,

:::E

• -NO PREECLAMPSIA

30

30

20

20

10

~

I 5

10

10

15

20

25

e- PREECLAMPSIA

;:,

(.)

~

l 5

>25

10

15

20

25

>25

CORD LEAD CONCENTRATION (ug/dl)

CORD LEAD CONCENTRATION (ug/dl)

Fig. 2. Cumulative percentage plots for PROM (A), preterm

{8),

meconium (C), and preeclampsia

(D). Concentrations of lead in blood were grouped into bins of 5 11-g/dl width, and cumulative

percentages were calculated.

cord blood versus maternal blood was not dose-related to lead. This can be seen in Fig. 4, C. which plots the ratio of cord ZnP/maternal ZnP to cord-lead concentration.

Comment Contamination of the environment by noxious cherniLals is ubiquitous. 1~here is increasing recognition of the adverse effects of agents at levels formerly consid-

cred to be safe. Thus, it is imperative to redefine the limits of safe exposure to various pollutants for pregnant women and their fetuses. Exposure to chemicals in the workplace is becoming important as more women are being employed in industry. Since lead is so \videly used throughout industry, the exclusion of women from areas in which lead is utilized would pose significant social and econotnic probleins. Hi The difficulty encountered in studying exposure in

44

z

0

j::

"'

~

~

:::ll 0

u

J:

u

"'J:w

1-

i

"'z

1-

w

i

u. 0 1-

z w u a: w

0..

Angel! and Lavery

::1

An:

112

Ill

PRETERM

501

J i i i lil 2

I

I

112

761340

·:II II 1 I i

20

I MECONIUM

20

I

PREECLAMPSIA

10

3138

0/12

0 5

10

15

20

25

25

CORD LEAD CONCENTRATION (ug/dl)

Fig. 3. Dose-response curves of lead and obstetric complications. Concentrations of lead in cord blood were grouped into bins of 5 J.,tgidl width. The number of patients with each complication in each bin was then counted and divided by the total number in that bin. The actual numbers of patients are ~huwn .h ;1

ratio in the figure. (Denominators differ because-

some patients could not be classified on the basis of chart rt'1·iew.1 the prenatal period is reHected by the paucity of clinical data concerning lead and pregnancy.' Although many animal studies have clearly shown the adverse effects of exposure to very low leYels of lead during the perinatal period. extrapolation from such data is difficult because of different gestational periods, species differences in metabolism. differential growth of various organ svstems in relation to birth, and the speciesspecific nal\lre of manv teratogens. Exposure to high .levels of lead has been associated with an increased frequencv of abortion and preterm birth.' The present study showed a high correlation between the concentration of lead in the blood of mothers and their infants. a tinding which is consistent with several other studies.'- 9 Despite contrary findings of some previous studies.';· ;. 10 the placenta does not appear to act as a harrier to lead. Concentrations of lead in the blood

.f.

janua1 1 l. l ~lX2 Ohstel. (;nJ<·col

were ele,·ated in women with either preterm deliwn or premature rupture of membranes in a large scrie, reported by Fahim and associates!; The mean concentration of lead iu Iriaternal blood in didt ~tud\ £U11ong 11omen \\·ith pretenn deliveries 11as 29.1 J.Lgidl. \'\'omen \\·ith PRO:'vl had mean concentrations nf lead pf :?:'l.f) J.Lg/dl. Control women had mean comcntrations of lead in the hlood greater than 2:J ,ug!c!l: one of these had a preterrn deli,·erv. and the other had meconiumstained amniotic Huid. :-Jo conclusions can be dr:n1n from such a small sample. However. it is ot interest that both women with lead concentrations greater than 2:) J.Lg/dl had complications, and that this concentration approximates that of affected women in the series of Fahim and associates.'; L'nfortunateiv, Fahim and associates'; did not report dose-response relationships. 11·hich are necessary to eYalnate the sa let,. of exp• Ntn' to lead at levels of less than :\0 J.Lg/dl. rfhc concentration of ZnP is ele\ated in exposure to lead because of inhibition of heme synthetase. This is the earliest sign of a signiticant bio!ogH- response to lead. ll'ith children exhibiting much greater inneases than women in concentrations of /.nP dt a given concentration ot lead in the blood: womt·n. in turn. are more scnsitiYe than men." Increased concentrations of /nP have also been found in fetal membranes \\·hen ;1 comparison is made 11ith maternal levels.' 'Ye attempted to extend the hvpothesis of differential sensitiYitv to maternal-fetal pairs. The mean level <)f /nP 11·as f(nmd to be significantlY greater in cord blood than in maternal blood. If lead was responsible lor thi' increase in ZnP, the expectation \Votdd he that infants with the greatest increase in /nP rclatiH:' to their mother should haYe increased le\·els of lead in the blood. Howe1·er. when the ratio of cot d In P to maternal ZnP was plotted against cord lead, no dosc-elfett relationship was found (Fig. 4, C). This implies that exposure to lead is not the source of the elevation in ZnP in the fetus. Other factors limit the generalizabilitv of this conclusion. The threshold for elevation of ZnP i' at a concentration of lead in the blood of approximate!\ 20 J.Lg/dl. bel60 J.Lg/dl) or children (>40 J.,tg/dl). However, several factors should encourage physicians to be conservative in assessing the risks involved in exposure of the fetus to lead. The levels of lead considered to be

Volume 1-!2 Number I

Relationship of blood lead to obstetric outcome

i....

150

1

~~

45

A 100-

z

i 0

so-

11.

c

N

c

~

0 +...,.....,,......,.._,....-r-'1,-.,......,..-T'""'T,--.-...,....,.......,.-,r-r-r-..,......,-1 40 10 20 30 0 CORD LEAD CONCENTRATION (ug/dl)

;; .....

150

c:

01

.a z Q !;;:

I=

()

a:

w

:I:

b =e

100

.....

N 0

50

8Q

'·.·.

c:

N

2ratio=1

~

a:

UJ

5::li

6-

a. c:

Q.

X

c

N

B

z w () z 0

8~--------------------------~

Q.

a:

0

0

10

20

30

40

MOTHER LEAD CONCENTRATION (ug/dl)

o~~~~~~~~~~TT~~~~~rr~rr,~rl

0

5

10

15

20

25

30

CORD LEAD CONCENTRATION (ug/dl)

Fig. 4. Concentrations of zinc protoporphyrin in cord and maternal blood as a function of concentration of lead in the blood (A and B). C, Ratio of cord ZnP/rnother ZnP a~ a function of concentnltion of lead in cord blood.

safe have been continually lowered in past decades, and, indeed, the suggestion has been made that a threshold for damage due to lead may not exist. w In the case of exposure of the fetus, the goal must be to minimize the possibility of damage to the rapidly developing nervous system. Animal studies have shown subtle behavioral effects of perinatal exposure to very low levels of lead. 17 Exposure to lead has been suggested as a cause of hyperactivity in children. 1H Mild toxicity may cause subtie alterations in behavior, intelligence, and learning ability, 2 which points to the relevance of the animal data to exposure in human beings. The adverse effects in human beings of prenatal exposure to low levels of lead are unknown, since the man-

ifestation of toxic effects may not appear until years after the initial insult. Many of these children will continue to be exposed to lead during their childhood, with possible cumulati\c effects on dcYeloping organ systems, particularly the central nenous system. l-\ en a short period of exposure may be important, in that deficits in performance early in development may affect beha,·ior later in lite. Data from the literature and the present smdy allow us to speculate on safe lew·ls of exposure 10 lead insofar as obstetric complications are concerned. Exposure which results in concentrations of lead in the blood greater than 25 ~-tg/dl may result in an increased incidence of complications. This is of more than academic

46

Angell and Lavery \n!

interest as evidenced by an acceptable blood-lead concentr
~.!!Hi,tn L Jt;"\:_~ Clbstet. { .\ nccol

tors, such as diet, trace mctab, and othn lo:-:tt pulltn· ants, in human beings remains a subjn1 '''r future n:search. We \\ish to exprc>
REFERENCES I. Rom, \\'. \i .: Effects of lead on the female and reproduc· tion: :\ review, Mt. Sinai J. Mecl. 43:.542, I 976. ~. ;\cedletnan, H. L.. Gunnoe, C., Leviton, A'l et al.: Delicits in P'vchologic and classroom performance of children \\'ith de1·ated dentine leacl levels. 1\:. Engl. .J. :Wed. 300:6H~t. i 979. :l. Roek II.. Bruaux, P., Buchet,.J. P., et al.: l•npact ol air polluti
and neonatal blood lead le1·els, Conn. Med. 40:·l5~, 1976. II. Vistica, D. T., Ahrens, F. A .. and Ellison, \\'. R.: The effect; of lead upon collagen synthesis and proline hvdroxvlation in the Swiss mouse 3T61i.broblast, Atch. Binchen;. Biophvs. 177:1.5. 1977. i 'i White . .J. :'\1., and Seih, i l. S.: Lead and the red cdl. Br j. Haemetol. 30: t:l3, 197'i. l'l. Division ot' .\ir Pollution Control, Buil'X:.-:1n1niP nf lt"'~n :~nrl n-->nrnrluctivt· ;tf~cts, Pr~;:~t~d~ 7::~94:·-1976.

-.................... ,. -

-

17. Cory-Slechta. D. A., and Thompson, 1.: Behavi•:>rai toxicil y of post weaning lead exposure in the rat, 'I oxicol. .-\ppl. Pharmacol. 47: I:; I, I 979. !H. David, 0., Clark, J., and Voeller. K. Lead awl lnperactivitv, Lancet 2:900, 1972. 19. Waldron, H. A.: The blood lead threshold. Arch. Lnvi· ron. Health 29:271. llJ71.

:::o.

Lead in the \\'t)rkplace, Lancet 2: iO"iU. i~IK(I.