Cord blood lead tracks the mother's bone lead. Her childhood exposure reaches the fetus decades later.

The isotope studies: the fetus receives the mother's lead, and a rising share of it comes out of her bones.

Lead from different ore bodies has slightly different ratios of its stable isotopes. Brian Gulson's group at Macquarie University used that as a tracer. They recruited women who had grown up in Europe and moved to Australia as adults. The lead in their skeletons had a European signature. The lead in Australian air, dust, water and food had a different one. By measuring isotope ratios in each woman's blood every month through pregnancy and for six months after delivery, the group could separate the lead coming from her current environment from the lead coming out of her own bones.

The interim report came out in 1997. Blood lead rose about 20% over pregnancy, and the rise was measurable even in women whose blood lead was low to begin with (the geometric mean was 3.0 µg/dL). The skeletal share of blood lead increased by a mean of 31% over the course of pregnancy, with a range from 9% to 65% between women. Cord blood lead concentration ran from 0.54 to 1.05 times the mother's. The isotope composition of cord blood matched the mother's blood almost exactly, a ratio of 0.993 to 1.002.

The 1998 follow-up in the same journal, now with 15 migrant women and 7 Australian controls, found the skeletal contribution rose in a roughly straight line across pregnancy, with per-woman increases of 26% to 99%, and that the skeletal contribution after delivery was larger than during the second and third trimesters. The two women in the cohort who took calcium supplements had the lowest skeletal mobilization of anyone in the study. The 2003 summary of the whole program put the geometric mean skeletal contribution to blood lead during pregnancy at about 33%, with a range of 10% to 88%, and reported that the migrant women and the Australian controls alike were eating about 500 mg of calcium a day.

William Manton's group did a smaller isotope study in the United States, 12 women followed from the second trimester until at least eight months after delivery, and got a more detailed picture of which bone gives up lead when. In early pregnancy the body resorbs trabecular bone, the spongy interior, which in their subjects held little lead, and blood lead fell more than dilution alone would explain. In late pregnancy resorption shifts to cortical bone, the dense outer shell with a higher lead content, and blood lead rises. During breastfeeding the whole skeleton resorbs and blood lead keeps climbing, peaking six to eight months after delivery. Blood lead fell from one pregnancy to the next in the two women who were followed through two, which the authors read as the first pregnancy carrying the largest transfer.

Where the bone lead came from.

In an adult, about 94% of the lead in the body is in bone (ATSDR 2020, citing post-mortem measurements). Lead gets there by riding the same pathways as calcium and being laid into the mineral of bone as it forms. Turnover is slow. González-Cossío's group summarized it as lead remaining in bone for years to decades, long after external exposure has stopped. A bone lead measurement is a record of cumulative exposure. A blood lead measurement mostly reflects the last month or two, plus whatever bone is currently releasing.

The exposure that filled those bones was highest in early childhood, when children put their hands in their mouths and absorb a larger fraction of what they swallow, and it was highest in the birth cohorts that grew up while lead was still in gasoline. In our analysis of 14 rounds of the national NHANES survey, US toddlers born 1973 to 1977 averaged 16.4 µg/dL of blood lead. Toddlers born 1986 to 1990 averaged 4.1. Toddlers born 2016 to 2020 averaged 0.63. The longer version of that series, and what it means for the adult skeleton, is on the bone lead by birth year page.

Women born 1973 to 1977 are 49 to 53 in 2026, mostly past childbearing. Women born 1986 to 1990 are 36 to 40, in the middle of it, and their toddler blood lead averaged more than six times what a toddler carries today. Leaded gasoline stayed on sale for on-road use in the United States until the end of 1995. A woman who was a toddler in 1994 is 34 now. The history page covers how much lead that era put into people compared with the pre-industrial baseline. The bone density page covers the same store of bone lead when it comes back out at menopause, which for a woman is the second large mobilization after her pregnancies.

The pathway, in one diagram.

Every figure on this diagram is from a named study in the citations. Where the literature does not give a number, the box does not carry one.

What it does to the baby.

The Boston cohort published by David Bellinger and Herbert Needleman in 1987 is the study the rest of the field is measured against. They enrolled 249 newborns, sorted them by umbilical cord blood lead into three groups (under 3, 6 to 7, and 10 or more µg/dL), and tested each child every six months with the Bayley Scales of Infant Development. At every age the high-cord-lead group scored lowest. Adjusted for confounders, the high group scored 4.8 points below the low group (95% confidence interval 2.3 to 7.3) and 3.8 points below the middle group (1.3 to 6.3). The children's own postnatal blood lead did not predict their scores in this window. The authors wrote that the fetus appeared to be affected at blood lead concentrations well below 25 µg/dL, which was the CDC's ceiling for young children at the time.

Fifteen years later a Harvard and Mexico City group connected it to bone. Gomaa and colleagues measured cord blood lead at birth, then measured the mothers' tibia and patella lead by X-ray fluorescence within four weeks of delivery, and tested 197 children at 24 months. Cord blood lead and maternal patella lead were each independently associated with lower Bayley scores. Compared with the lowest quartile of maternal patella lead, the second, third and fourth quartiles came with 5.4, 7.2 and 6.5 point decrements. A doubling of cord blood lead, for example from 5 to 10 µg/dL, came with a 3.1 point decrement. The mother's bone lead predicted her child's development on top of what the child's own cord blood predicted.

Hu and Schnaas point at different trimesters. Both cohorts are from Mexico City, both are small, and they measured different things (plasma lead in the first trimester in one, whole blood lead in the third in the other). What they agree on is that the maternal lead level during pregnancy predicts the child's cognitive scores years later, independent of the child's own postnatal exposure. Bellinger found the same thing in Boston in 1987 with cord blood. Gomaa found it with the mother's bone. Prenatal exposure is one of the pathways into the population-level IQ loss on the IQ page; that page's estimate is built from childhood blood lead and does not separately count the prenatal dose.

One maternal outcome belongs here too. Rothenberg's Los Angeles cohort of 1,006 pregnant women measured bone lead in the tibia and the heel (calcaneus). For every 10 µg/g of heel bone lead, the odds of hypertension in the third trimester were 1.86 times higher (95% CI 1.04 to 3.32). Tibia lead was not related, and neither bone measure was related to blood pressure after delivery. Blood lead in these women was low, a geometric mean of 1.9 µg/dL during pregnancy. The exposure that showed up in their third-trimester blood pressure was past exposure stored in bone.

The timing inside a pregnancy, and after it.

Rothenberg's earlier Mexico City series, 105 women sampled from week 12 to week 36 and again at delivery, is the clearest description of the shape. Mean blood lead fell 1.1 µg/dL between weeks 12 and 20, then rose 1.6 µg/dL between week 20 and delivery. The fall matches the expansion of blood volume in early pregnancy. The rise does not fit that explanation, since blood volume keeps expanding in the second half of pregnancy and on its own would lower the number. The authors attributed the rise to bone lead mobilization plus increased absorption and retention, and closed with a sentence that is the point of this page: reduction of lifetime lead exposure may be required to decrease risk of fetal exposure.

Téllez-Rojo's 2004 cohort of 193 women measured a urinary marker of bone resorption (N-telopeptides of type I collagen) in each trimester alongside plasma and whole blood lead, then measured bone lead after delivery. Resorption rose across pregnancy. The interaction between a woman's bone lead and her resorption rate predicted her plasma lead, which is the fraction that crosses to the fetus. Dietary calcium intake was inversely associated with plasma lead.

After delivery the release does not stop. In 425 lactating women followed for seven months, those who breastfed exclusively had blood lead 1.4 µg/dL higher than women who had stopped, and mixed feeders 1.0 higher, after adjusting for bone lead and environmental exposure (Téllez-Rojo 2002). Manton's isotope data put the postpartum peak at six to eight months. The papers on this part of the story, including breast milk, are covered at length on the bone lead mobilization page, and the birth-cohort chart of who carries the most is on the remobilization chart page. This page does not repeat them.

The prenatal intake form does not ask.

Routine first-visit prenatal labs are a blood count, blood type and antibody screen, rubella immunity, hepatitis B, HIV, syphilis, a urine culture, and later a glucose test. Blood lead is not on the standard panel. The CDC's 2010 guidelines for lead in pregnancy and the matching ACOG committee opinion both recommend testing only when a risk factor is identified, and the risk factor list is about current exposure: recent immigration from a high-lead country, pica, an occupation or hobby with lead, use of lead-glazed ceramics, living in or renovating pre-1978 housing. None of the questions ask what year the woman was born, where she grew up, or what her blood lead was when she was two. Those are the questions the isotope and bone studies say matter most.

Even if the form asked, the answer would be hard to act on clinically. The measurement that predicted birth weight, head circumference, development scores and third-trimester hypertension in the studies above was bone lead by K-shell X-ray fluorescence. That instrument exists in a handful of research groups. It has never been part of clinical care, and there is no plan for it to be. So the exposure most of these papers say is doing the work is not measured, and the proxy questions that could flag it are not asked.

Calcium: the randomized trial.

Adrienne Ettinger and colleagues randomized 670 women in Mexico City in their first trimester to 1,200 mg of supplemental calcium per day or a matching placebo, double blind, and followed 557 of them through delivery. Adjusting for baseline lead, age, trimester and diet, the calcium group's blood lead ran 11% lower than placebo, about 0.4 µg/dL (p = 0.004). The effect was larger in the second trimester (14%) than the third (8%). Among women who took at least 75% of their pills it was 24%. Among women who had baseline blood lead above 5 µg/dL it was 17%. Among women who used lead-glazed ceramics and had high bone lead it was 31%.

A companion trial in 617 lactating women, also 1,200 mg of calcium carbonate a day, found a small overall reduction that did not reach significance (0.29 µg/dL), and a 1.16 µg/dL reduction, about 16%, in women who took the supplement and had patella lead of 5 µg/g or more (Hernández-Avila 2003). The pattern in both trials is the same. Calcium does the most where there is the most bone lead to hold in place.

The mechanism the trialists proposed is that adequate dietary calcium lowers the parathyroid signal that drives bone resorption, so the skeleton gives up less calcium, and less lead with it. That is a different mechanism from the one on the calcium and lead absorption page, where calcium taken with food competes with lead for uptake in the gut and cuts how much of a new dose gets in. Both apply to a pregnant woman. The recommended dietary allowance for calcium in pregnancy and lactation is 1,000 mg a day for women 19 to 50 and 1,300 mg a day for those 14 to 18 (Institute of Medicine 2011). Gulson's cohort was eating about half of that. Food first, and a supplement to close the gap, was the trialists' advice and is the standard prenatal advice anyway.

The house: the other input you control.

The subgroup with the biggest calcium effect in Ettinger's trial was women who ate off lead-glazed ceramics. That is a source that can be found and taken out of use, and it is a source many families in the United States still have in the cabinet: painted or glazed dishware made before the 1990s, imported pottery, an heirloom set. Every meal off a leaching dish during pregnancy adds to the current blood lead that the fetus shares, and adds to the bone store that the next pregnancy will draw on. The same is true of a painted windowsill in a pre-1978 house, a hand-me-down crib, and old brass plumbing.

If you are pregnant or planning to be, the practical order is the one the papers point to. Ask your provider for a blood lead test with the first-trimester labs; the how to ask page covers the wording and what the number means. Get calcium intake to the recommended level. Then test the surfaces the baby and you will eat off and touch: the dishes, the inherited furniture, the painted trim in the room the baby will sleep in. The prenatal supplement ranking is worth a look too, since some prenatal vitamins carry a measurable lead dose of their own.

None of that changes what is in the bones. It changes how fast it comes out and how much gets added before the next pregnancy. For the mothers born in the last leaded-gasoline cohorts, that is the part of this that is theirs to decide.

Related on this site.

Citations