September 2009: Volume 35, Number 9
Big Study on Small Patients
National Children’s Study Could Provide Key to Standardized Pediatric Reference Intervals
By Deborah Levenson
Laboratorians have long struggled with defining reference intervals for infants and children. Not only do changing metabolic conditions related to development present a challenge, but some analytes—including electrolytes and blood glucose—have subtle differences in values between children and adults. Others pose stark contrasts. These include enzymes of the bone and liver and steroid hormone values, for which labs simply cannot rely on adult reference intervals. Lacking a gold standard to which results for these analytes can be traced, many hospital labs have devised their own pediatric reference intervals (PRIs). But because laboratorians and researchers have difficulty identifying enough children of a given age who represent a normal population, such PRIs and research that establishes them generally rely on populations that are far from ideal for setting standards. These populations tend to be fairly small and comprised primarily of sick children who reflect the local population’s racial and ethnic mix. The use of resulting reference intervals has the potential to subject other children to inaccurate diagnoses, unnecessary diagnostic procedures, and inappropriate treatment.
AACC’s Pediatric Reference Range Committee is now taking a big step toward improving this situation. The committee is preparing a proposal to conduct research that would establish age-related, standard PRIs for steroid hormones and amino acids based on dried blood spot samples analyzed by tandem mass spectrometry. The AACC study will be proposed as an adjunct to the huge National Children’s Study (NCS), which could provide access to more than 20 years worth of samples from children in 100,000 households across the nation. NCS will track children from before birth to age 21 (See Sidebar, below).
The National Children’s Study
A Focus on Environment
The AACC Pediatric Reference Range Committee’s proposed project to determine standard pediatric reference intervals would use samples from what federal agencies say will be the largest, most comprehensive study of how genes and the environment interact to affect children’s health.
Led by the National Institute of Child Health and Human Development (NICHD) in collaboration with National Institute of Environmental Health Sciences (NIEHS), CDC, and the U.S. Environmental Protection Agency (EPA), the National Children’s Study (NCS) will track the health and development of more than 100,000 children from before birth to age 21. Congress authorized the planning and implementation of the study with the Children’s Health Act of 2000.
NCS aims to provide researchers, healthcare providers, and public health officials with information to inform prevention strategies, health and safety guidelines, and possibly new treatments and cures for disease.
To establish links between children’s environments and their health, researchers will analyze how various factors—including chemical, physical, biologic, geographic, and social and behavioral elements—interact with each other and what helpful or harmful effects they might have on children’s health. Researchers will also look for links between these factors and health issues such as birth defects and pregnancy-related problems, injuries, asthma, obesity, diabetes, and behavior, learning, and mental health disorders. By tracking children’s development through infancy, childhood, and early adulthood, researchers hope to pinpoint the root causes of these and many other childhood and adult diseases.
“The principal benefit of a large scale, long-term study like the National Children’s Study is that it will uncover important health information at virtually every phase of life,” said Duane Alexander, MD, Director of NIH’s Eunice Kennedy Shriver NICHD, one of the consortium of federal agencies implementing the study. “Initially, it will provide major insights into disorders of birth and infancy, such as preterm birth and its health consequences. Ultimately it will lead to a greater understanding of adult disorders, many of which are thought to be heavily influenced by early life exposures and events.”
For more information is available online.
The committee soon will submit its proposal to the National Institute of Child Health and Development, one of the government agencies involved in NCS. The proposed study could be of use to NCS because no one really knows what the true reference intervals are for most analytes involved in the study, an NCS investigator notes. “Now there’s lots of uncertainty, especially about indeterminate values in the tail ends of the range, so currently we have to approximate. The availability of a huge number of samples would enable establishment of those tail ends with greater precision,” he explained. Although NCS has not begun accepting adjunct study proposals, once it does, AACC’s proposal will be considered with others, NCS officials noted.
Just the Beginning
The adjunct study proposal the Pediatric Reference Range Committee will submit to NCS is essentially a starting point for a more ambitious agenda encompassing multiple analytes, according to Patricia M. Jones, PhD, a member of the committee and proposal author. Jones is clinical director, chemistry, in the Department of Pathology at Children’s Medical Center in Dallas.
Jones listed a few reasons why the committee wants to start with amino acids and steroid hormones. A major advantage is that these analytes can be measured in the dried blood spots NCS plans to collect, and the samples are easily stored. But also driving this choice is the urgent need for good, standardized PRIs for these analytes. Jones mentioned other factors. “They are also analytes that have a definite method for analysis, which is tandem mass spectrometry. Eventually we would like to have access to NCS samples for many other analytes over the course of the study.”
Bone markers are one such group of analytes the committee hopes to study in the future, according to Jones and Michael Bennett, PhD, committee chair and professor of pathology and laboratory medicine at University of Pennsylvania and director of metabolic disease laboratory at Children’s Hospital of Philadelphia. “There are big differences in bone markers as kids grow. Bone markers are hugely elevated in puberty, especially in boys,” noted Jones. Bennett added that standard intervals would eliminate much guesswork for labs.As for amino acids and steroid hormones, the AACC-proposed adjunct study could give the committee as much as 21 years of data, Bennett noted. “At the start of the study, the kids will be small babies, and there will be blood draws throughout these kids’ childhoods,” he explained. “This kind of material has thus far has been unavailable to children’s hospitals,” he added. However, smaller studies will yield additional PRI information before the NCS’s cohort’s follow-up is expected to end in 2032.
NCS could benefit from using standardized or traceable assays, AACC committee members maintain. That’s because NCS is likely to see changes in methodology arise over the course of its very long follow-up, explained Vijay Grey, PhD, professor in the department of pathology and molecular medicine at McMaster University Medical Center in Hamilton, Ontario. In addition to serving on the AACC committee, Grey is also one of the principal investigators on CALIPER, a large Canadian study that also aims to establish PRIs (See “Targeting Kids,” below).
“The technology we’re using today won’t be the technology we’ll use in 20 years,” NCS researchers agreed, adding that AACC’s proposed adjunct study could prove valuable. “Having PRI standards will enable researchers to translate to new platforms without re-establishing a reference interval.”
More Pediatric Reference Interval Projects
While the federal-funded National Children’s Study gets underway, other groups in both the U.S. and Canada already have established research projects that aim to define pediatric reference ranges.
In the U.S., the Children’s Health Improvement through Laboratory Diagnostics (CHILDx), sponsored by ARUP Laboratories and the University of Utah Department of Pathology, is working on three studies related to pediatric reference intervals. One intends to establish reference intervals for 70 serum, 20 plasma, and 12 urine analytes in children age 7 to 17. A second study focuses on 13 citrated plasma analytes in children age 7 to 17 years, and a third focuses on serum/chemistry tests in children age 6 months to 7 years, said William R. Roberts, professor of pathology, University of Utah and medical director, Automated Core Lab, ARUP, both in Salt Lake City. The sizes of these studies vary in size from 480 to 2,640 subjects.
Roberts and his colleagues published the first paper based on the study in 2005 (Clin Chem 2005; 51:1738–1742), which described seven common coagulation assays. A second paper (J Pediatr 2006; 149: 275–277) reported findings on coagulation assays.
Laboratorians can review published intervals and see if they can verify them for use in their hospitals, said Roberts. He cautioned, however, that subjects for these studies are local to Utah, and primarily Caucasian, so labs would need to use the same methods in a small number of local subjects to carefully verify that his reference interval would work in a given hospital’s population.
Roberts says the proposed AACC study will be valuable because of its size and the geographic, ethnic, and racial diversity of the children involved. “Right now we have no idea of how important a role race and geographic factors play in kids’ reference intervals,” he commented.
In Canada, Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) intends to establish a comprehensive database of reference intervals for both traditional and emerging biomarkers of pediatric disease in Canadian children from birth to 18 years, said principal investigator, Vijay Grey, PhD, associate professor, department of pathology and molecular medicine at McMaster University Medical Center, Hamilton, Ontario. She is also a member of AACC’s Pediatric Reference Range Committee. CALIPER began in 2004, but so far only local pilot projects have been completed. No date has been set for the end of the study, according to Grey.
The proposed AACC study and CALIPER have similar endpoints, but the approaches are slightly different, said Grey. For example, NCS will track children prenatally and includes blood samples from parents. And while the AACC proposal will ask NCS to supply AACC with bloodspots, CALIPER researchers are actually drawing blood because Canadian regulations prohibit researchers from establishing reference intervals with leftover blood samples, said Grey. Canadian researchers are recruiting healthy children from hospital nurseries, day cares, schools and families of hospital volunteers, according to a website about the study.
Financial support and participation are major problems for the study, said Grey. Recruiting healthy individuals is costly and there’s a big ethics approval process. “Reference intervals aren’t considered a sexy research project,” she said, noting she got only 300 samples from one site after a year. “But I’ve gotten some local support because the study is viewed as valuable to quality assurance,” she noted.
A Healthy Population
Besides the sheer enormity of the NCS study sample, the AACC committee is excited at the prospect of accessing an overwhelmingly healthy population, Jones and Bennett emphasized. Healthy populations have been difficult to come by when establishing reference intervals because of funding and procedural issues. “It’s very costly to recruit healthy individuals because there’s a big ethics approval process,” Bennett explained. For this reason, most children’s hospitals have established problematic reference intervals using their own pediatric patients, who are usually sick.
PRIs are in even worse shape at hospitals that have few pediatric patients. Some institutions don’t even have PRIs, according to Jones, who noted that labs at such hospitals sometimes ask about using PRIs her lab has established for particular analytes. This situation is problematic. “I don’t know if our reference intervals apply to another institution’s platforms and patient population,” she explained, adding that establishing some standards would alleviate the problem.
Because the NCS population will be overwhelmingly healthy, the committee plans to decide between two methods of establishing PRIs, either standard deviation or the percentile approach based on the distribution of data. “We have to look at the data before deciding upon a method,” Jones said. “If there’s a normal distribution with a bell-shaped curve, probably we’ll use standard deviation. If there’s a skewed distribution, we might try to normalize data using the percentile approach.”
Another future decision will be the particular age groups of a particular reference interval target, according to Grey. She pointed out that distribution of data would affect whether intervals will, for example, apply to ages 9 to 10, 10 to 11, or 9 to 12.
Tiny Kids, Huge Problems
While the AACC committee’s proposal would provide it with access to a huge, mostly healthy population, its proposed study could also help NCS researchers make sense of values for analytes in samples from the smallest children, for whom reference intervals are especially problematic. Labs find values for children under age 3 particularly troublesome, Jones and Bennett agreed, because these children have a limited volume of blood. Furthermore, human research restrictions and operational procedures that limit how much blood can be drawn mean that it’s difficult to get samples from very young children, especially considering that several hundred samples are needed for even a small study. Finally, the whole process of establishing PRIs for children is tricky because differences in analytes commonly measured in children are often functions of children’s different stages of growth. Changes are especially rapid and frequent in infants and toddlers, the committee members noted.
Bennett sees the AACC proposal as a win-win proposition with huge benefits, but emphasizes that even if NCS were to accept the proposal, both sides need to take several steps—including finalizing details, setting a budget, and securing funding—before the study starts. If AACC’s proposal is accepted, its adjunct study could start sometime after the launch of the main study, currently anticipated in 2011.
Bennett is both hopeful and wistful about the project. “If the proposal is accepted, it will provide a starting point to getting standards reference intervals for analytes in a pediatric population, for all ethnic groups, in both rural and urban areas,” Bennett explained. “Standard pediatric reference intervals will be universally available. It will take a long time, so I hope I’m still around when the committee finishes its work.”
Deborah Levenson is a freelance writer based in College Park, Md.