A universal measuring stick

International group proposes ideal growth standard for fetuses and newborns

Illustrations: Elisa Carareto over medical photo Body Scans Photo Researchers, Inc. / Latinstock Human babies, under ideal conditions, are almost always born the same size. After spending an average of 40 weeks well protected and fed in the womb, they come into the world approximately 50 cm long. This size may vary two or three centimeters in either direction for this gestational age and seems to represent the optimal growth for the human species today. Parents and pediatricians may even have suspected this. However, obtaining valid measurements from a scientific standpoint required the work of a battalion of people.

Over four years, from 2009 to 2013, about 300 physicians and researchers from 27 institutions weighed and measured 52,171 newborns during their first hours of life in 8 countries. Of these babies, they selected 20,486 that had gestated for 33 to 42 weeks—those that had spent less time in their mother’s womb were smaller, obviously, and those that spent more time were larger. All of the children were healthy, as were their mothers, who belonged to the highest income and highest educational level groups of these communities. None of the women smoked or had diseases that could have reduced the growth of their babies.

There was a reason for the choice of such a select group. Women with more years of schooling and higher socioeconomic status tend to take better care of their health and are less likely to have problems during pregnancy. And the researchers wanted to establish, in detail, the corporal measurements and health of newborns born under the best conditions possible. With this data, they intended to plot curves establishing growth parameters considered ideal during pregnancy and shortly after birth that could be applied to the almost 140 million children born every year worldwide.

José Villar, the Argentine obstetrician who coordinated this extensive task, believes he has finally managed to generate curves that can be used universally, fulfilling a recommendation made by the World Health Organization (WHO) 20 years ago. “Now, all newborns can be measured based on the same reference,” says Villar, a professor at Oxford University, in the UK.

Villar was a member of the WHO committee of specialists that identified the need, in 1994, to establish an international standard in order to determine if babies were born healthy and of the appropriate size, or if they were smaller than they should be and thus ran the risk of becoming ill and dying during the first days of life. In order to achieve this, a growth assessment tool needed to be developed—a sort of universal measuring stick for babies that would serve for different regions of the world.

But a measuring stick would not be enough. They also needed to know if the measurements indicated health or problems.

Since the WHO meeting, many curves had been developed—a recent survey found that 104 had been published since 1990. However, none of them seemed to satisfy the requirements necessary to be considered universally valid. The problem is that the strategies used to construct them imposed limitations.

Developing curves representative of growth standards for all mankind—especially given the number of people on Earth and variations between communities—involves a lot of work and financial resources. It requires the contribution of many professionals and the assessments of a large number of people. Due to these complications, many of the prior curves were developed based on data collected previously, which was generally less reliable and more subject to errors, or without the necessary standardization of measurement strategies. Another reason why these curves often had questionable international validity was that the information on pregnant women and children came from a single region, or at best a single country. This limitation made physicians suspect that curves produced, for example, using data from women and children from the United States would not be a good reference for mothers and babies in Africa or Asia.

The new newborn curves, presented in an article in the September 6, 2014 issue of the journal Lancet, overcome these restrictions, in principle. They were developed by researchers in the International Fetal and Newborn Growth Consortium for the 21st Century (Intergrowth-21st) using the same methodology and the same type of equipment to carry out the measurements and, principally, represent the data from women and babies in eight countries with various levels of social and economic development (United States, Brazil, England, Italy, Kenya, Oman, India and China), spread over four continents.

In addition to collecting information on 20,486 children to produce the newborn growth curves, the researchers also developed new fetal growth curves, which portray the growth standard considered to be desirable for babies during gestation—the two types of curves are available on the Intergrowth site. Unlike the curves for newborns, prepared based on measurements taken soon after delivery, the fetal curves are more difficult to obtain. They require the completion of a series of measurements of the child inside the womb, the first environment of human life.

To that end, the Intergrowth team monitored the pregnancies of another 4,321 women with high social, economic and educational levels in the same eight countries. During pregnancy, these women underwent regular health follow-up visits, during which their babies were assessed and measured through ultrasound exams carried out every five weeks. They will continue to be monitored through physical exams and neurological development tests until 2 years of age. Published in a second article in the same issue of the Lancet, authored by the obstetrician Aris Papageorghiou, also from Oxford, these fetal development curves, according to their developers, are also the first with universal validity.

“We know exactly what happened with each of the women and children,” says Villar, one of the principal authors of the study, together with the Brazilians Fernando Barros and Cesar Victora, of the Federal University of Pelotas (UFPel). “We wanted to develop prescriptive curves,” says Barros, who coordinated data collection in Brazil and Brazil’s participation in the project. “We believe that they describe how a child should grow during a pregnancy that takes place under ideal conditions, with good nutrition, without infections and at altitudes below 1,600 meters, where oxygen availability is greater,” says the researcher from the state of Rio Grande do Sul.

In order to ensure that these curves could serve as a standard for all human populations, the researchers compared the variation in sizes of the babies from the eight populations assessed. The comparison, published in July 2014 in Lancet Diabetes and Endocrinology, showed that the average difference in size was always less than nine millimeters (see figure on facing page).

“This difference is practically negligible,” says endocrinologist Alexander Jorge, of the University of São Paulo (USP) School of Medicine, who investigates the genetic causes of growth disorders. “This data shows that, under ideal conditions, babies are born with very similar sizes, which does not mean that there is no genetic influence on newborn length.”

The genetic characteristics of each population, in fact, seem to have little influence on the variation in size of babies, which was already small. They explain a maximum of 3% or 0.3 mm of the observed difference in length between children from the eight countries. The other 8.7 mm are the result of environmental factors (health, maternal nutrition and quality of life). Within a population, however, genetic variation explained up to 20% of the size difference. The remaining 80% was due to the environment. Aware of the Intergrowth data and the importance of maternal health to a baby’s development, a group of 50 physicians, economists and other international leaders gathered in Oxford earlier this year decided to act. It sent a letter to the governments of 22 emerging economies—including Brazil and India—stressing the need to improve women’s quality of life, “a factor that influences countries’ social and economic development.” “There were few replies,” says Ian Scott, director of the group. “Brazil did not respond.”

There seem to be biological reasons for why babies of the same species grow in a similar manner if gestated under almost identical conditions. Years ago, researcher Jeffrey Baron, of the United States National Institute of Child Health and Human Development, compared the cell multiplication speed of mouse, human and elephant embryos. He found that the rate of cell proliferation in the early stages of embryo development is very similar in the three species. The differences begin to appear at the point in which cell multiplication rates decelerate.  This rate decreases earlier in smaller species, which also have shorter gestation periods—that of the mouse lasts about 20 days, while that of the elephant lasts about 20 months.

Alexander Jorge suspects that, for humans, there could have been a strong selection pressure that led to the genes regulating fetal growth remaining stable in populations in different regions of the world—different growth curves indicate that the length of babies varies little between different groups up to two years of age. “Mutations in these genes may have had a negative influence in terms of perpetuation of the species and were eliminated,” says Jorge.

Today about 180 genes and genetic regions associated with height are known. But, together, they explain only 11% of the height difference between humans. The remainder is believed to be determined by the environment. “A person’s growth potential is determined by his genetic makeup at conception,” explains pediatrician Claudio Leone, of the USP School of Public Health, a researcher in the field of child growth and development. “An unfavorable environment stifles that potential.”

A child that, for genetic or environmental reasons, grows less than desired—especially during gestation and the first years of life, when the rate of development is faster—might not achieve this potential, even if he grows more quickly later. For this reason, parents and physicians are always looking at height. “Growing well is a physical sign of health,” says Leone. And curves such as those produced by Intergrowth are an important screening tool, both during gestation and after birth.

During pregnancy, the ultrasound curves of fetuses small for their age can help identify the most appropriate time to interrupt the pregnancy, explains obstetrician Silvio Martinelli, a physician as the USP Hospital das Clínicas. “We try to deliver the child when the chance of survival and quality of life outside the womb are greater than if the child remains in the womb,” he says. Newborn curves, however, signal to pediatricians the possible problems facing newborns. “They give us an idea of the risks that the children run and the care that they will demand in the first days of life,” says neonatologist Cléa Rodrigues Leone, a researcher at the USP Children’s Institute.

To the eyes of a layperson, the curves resemble a work of minimalist art. In general, they contain five bands of values, or percentiles, for each characteristic—weight or length, for example—measured at different gestational ages. In order to develop the fetal growth curves, researchers recorded the values of five parameters (three head measurements, femur length and abdomen circumference) during every ultrasound examination. Newborn curves were based on weight, head circumference and total length for each gestational age.

The ranges that most concern physicians are the upper and lower extremes. The first two mark the 3rd and 10th percentiles and represent the values that are, respectively, among the lowest 3% and 10% for that characteristic. The last two, the 90th and 97th percentiles, include the values that correspond to the highest 10% and 3%. The intermediate bands contain the remaining 90% of values, those in which parents and physicians hope babies will fall.

The extremes are worrisome because they are signs of a problem. Babies that grow in utero according to the 90th percentile values are, generally, the children of diabetic women who are unable to control their blood sugar (glucose) levels. These children are larger than the others because the mother’s blood, with more glucose than is desirable, stimulates the pancreas to increase insulin production, one of the principal hormones promoting growth in this phase of life. In the first weeks after birth, they need medical monitoring to prevent the glucose levels in their blood—the principal source of energy for the brain—from dropping too much and affecting the development of the central nervous system.

Physicians also pay special attention to babies who grow according to the lower limits of these curves, the 10th percentile. 50% to 60% of these children are healthy babies who are developing according to their genetic makeup. The rest, however, suffer from what doctors call growth restriction. Simply put, they are children who do not grow enough because they do not receive adequate nutrition.

“Growth restriction is the second leading cause of perinatal death,” states Martinelli. It increases the risk of a child dying during gestation sevenfold. A recent study conducted by the Hungarian obstetrician Jason Gardosi, author of an individualized growth curve that uses the mother’s health characteristics as a basis to project the expected development of her baby, evaluated the health of 92,218 children born in England between 2009 and 2011. The death rate among children without growth restriction was 2.4 cases per thousand births, while this index jumped to 16.7 per thousand among those who suffered food deprivation in the womb. According to an article published in 2013 in the British Medical Journal, the proportion of deaths was even higher (19.8 per thousand) when the restriction was not identified early.

“The bodies of children that suffered nutritional deficiencies during intrauterine life have metabolic adaptations that cause them to react differently to environmental stimuli after birth, increasing the risk of cardiovascular disease in the future,” explains Cléa Leone.

The most frequent reason why the fetus fails to receive adequate levels of nutrients during pregnancy is due to changes in the functioning of the placenta, usually associated with maternal hypertension, for which there still is no effective treatment.

Illustrations: Elisa Carareto over medical photo Neil Borden / Photo Researchers, Inc. / LatinstockThe importance of maternal nutrition for the development of babies has been confirmed for more than 60 years. In the winter of 1944, at the end of World War II, the army of Nazi Germany invaded the Netherlands and restricted access to food. There was a large famine, which was recorded in studies published in 1946 and 1947: the Dutch babies who had survived the period of privation were born smaller and thinner than normal, a problem that affected principally boys.

In addition to guiding the work of pediatricians and obstetricians, by providing information on the health of each baby, the growth curves also indicate the health of a region. “The curves are also good indicators of social well-being, better even than child mortality figures,” says Claudio Leone. “If the children of a community, region or country are, on average, growing well, this is a sign that living conditions are improving,” he explains. These averages fall when the group suffers a period of economic hardship.

Leone and the investigators who evaluated the new fetal and newborn curves at the request of Pesquisa FAPESP consider those developed by Intergrowth to be the most rigorous and well-made from a methodological point of view. “Regardless of the discussion on whether or not they will be adopted as a universal standard,” says Leone, “they are the best growth curves we have now.”

Silvio Martinelli still cannot say whether or not the new ultrasound curves will be adopted by the maternity department at the USP Hospital das Clínicas to replace those used today, which were produced in the 1980s and 1990s based on a US population. “We will discuss the new curves with the maternity team,” he says. “We are used to working with curves that include a weight estimate for fetuses, and these do not.” Furthermore, using the Intergrowth fetal curves, a smaller proportion of infants would be classified as suffering from growth restriction. “We need to make sure that we will not run risks and that no baby will fail to be diagnosed,” added the obstetrician.

On the newborn curve, Cléa Leone notes a limitation: there is no data on babies that are born premature, between the 22nd and 32nd weeks of gestation. “If the ideal situation would be to use a single curve in the nursery, it will be difficult to recommend this curve to the risk units, working with more premature children” says the neonatologist, who for many years headed the neonatal care unit of the USP Hospital das Clínicas. “My position at the moment is that we should test the two types of Intergrowth curves to see if they change the mortality and prematurity indicators.”

Upon completion of the neonatal curves, Intergrowth researchers tested them by applying them to the universe of nearly 140 million babies born worldwide each year. Following the standards proposed by Intergrowth, which are stricter and were obtained based on very low-risk pregnancies, about 30 million of the newborns were malnourished, in need of medical support to recover proper growth and development rates. This figure is 2.5 times higher than previously estimated. “Governments must act,” says José Villar. “They need to understand that this problem must be solved in order to improve the human capital of a country. Do they prefer to act now and have a greater chance of obtaining better results at a lower cost, or wait and have to act later, with a lower probability of success?”

Scientific articles
APAGEORGHIOU, A. et al. International standards for fetal growth based on serial ultrasound measurements: the fetal growth longitudinal study of the Intergrowth-21stproject. Lancet. 6 set. 2014.
VILLAR, J. et al. International standards for newborn weight, length, and head circumference by gestational age and sex: the newborn cross-sectional study of the Intergrowth-21st Project. Lancet. 6 set. 2014.
VILLAR, L. O. et al. The likeness of fetal growth and newborn size across non-isolated populations in the Intergrowth-21st project: the fetal growth longitudinal study and newborn cross-sectional study. Lancet Diabetes and Endocrinology. 6 set. 2014.