Clinical Education Center
Strategies for a Challenging Field
Among the specific challenges of caring for pediatric patients are those related to diagnostic testing, in particular establishing accurate reference ranges (reference intervals) and the need to work with small sample volumes.1, 2
Dr Jon M. Nakamoto, Laboratory Medical Director, Quest Diagnostics Nichols Institute, San Juan Capistrano, CA and Associate Professor (Voluntary) of Pediatrics and Endocrinology, University of California, San Diego, discusses these challenges and outlines strategies to help both clinicians and laboratory professionals in managing the pediatric population.
With experience of working both in pediatric endocrinology clinical practice and in a laboratory, Dr. Nakamoto has been on “both sides of the fence.” He understands the needs of clinicians, while seeing the requirements, barriers and limitations from the laboratory’s perspective.
Even as regulatory agencies and clinical chemistry groups demand increasingly stringent testing accuracy, there’s an equal need for accurate reference ranges as well. “What value is there if I can measure down to 0.0001, if I can’t tell you what that number means?” asks Dr Nakamoto. But reference range accuracy remains a significant challenge in pediatric diagnostics, according to Dr. Nakamoto.. “We have a crisis today,” he says. “Many of the reference intervals, particularly for younger children, are twenty years old and don’t apply to current methods of testing, so in many cases they’re an approximation, based on old data.”
Providing Guidance to Clinicians
The problem is that establishing reliable reference ranges is not as clean, as objective, or as easy as is often thought, he explains. Defining who qualifies as healthy and normal, obtaining blood samples, and running sizable clinical studies are particularly difficult in pediatrics. “Even if kids and their parents are willing to have their blood drawn, in many cases, institutional review boards will block drawing blood on healthy kids because they consider it unethical.” This barrier is particularly high in the infant and young child population, the group where accurate reference ranges are most lacking.
From his work as a laboratory director, Dr. Nakamoto recognizes the need for statistical rigor and careful conduct of reference range studies. But because such studies may take years to complete, what is to be done in the meantime? As a clinician, Dr. Nakamoto is more concerned with gaining some information to help direct patient care, rather than waiting forever for precise data that are not available. “While we should always strive for the best possible reference intervals, we shouldn’t let the desire for perfection prevent us from implementing the good. Sometimes general guidance, even if not exact, is better than no guidance at all.”
A Call for Consensus
Recognizing both the shortcomings of current reference ranges and the needs of clinicians, Dr Nakamoto urges a pragmatic approach based on consensus between clinical groups, regulatory agencies, and the clinical chemistry community. “Without a dialog between all stakeholders,” he warns, “you might have a regulatory agency or a clinical chemistry group setting the bar too high to allow timely and feasible generation of the data clinicians require for guidance.
Dr Nakamoto strongly believes the clinical chemistry side needs to reach out to the clinical sub-specialties who depend on this information to understand where precision is critical and where more general guidance is adequate. “You need to find the right balance of what’s truly needed” he says. Conversely, clinicians need to understand better the proper use and limitations of reference ranges. “And that comes back to the need for a collaborative effort. Otherwise any of us who are trying to do this individually - a laboratory, a university, any single organization – will find it expensive, time-consuming, inefficient and incomplete.”
Initiatives to Address the Gaps
A number of initiatives have been previously undertaken or are underway to address the lack of up-to-date data in pediatrics. These include the CALIPER initiative in Canada3 and the National Children’s Study, led by the National Institutes of Health4 with consultation from the American Association of Clinical Chemistry (AACC). The NIH study will study children from pre-birth up to the age of twenty-one to examine changes throughout childhood. “The only problem,’’ notes Dr Nakamoto “is that it’s a very large, ambitious study, and it’s going to take 21 years to get all those data. But at least people are realizing it has to be done in a systematic fashion and that’s fantastic.”
Data Set Analysis
As we await the findings from initiatives such as the NIH study, Dr. Nakamoto proposes another, pragmatic approach. “What some experts are suggesting,” he says, “is that we widen our perspective beyond the classic paradigm of taking healthy individuals, measuring their levels and creating reference ranges from that.”
“What we need is to take large data sets, for example from all one year olds we have tested over the past three years and actually look at the distribution of results. With appropriate statistical analysis to remove the outliers that represent individuals with disease, you can often see what is a reasonable approximation to what ‘the normal, healthy range’ might be. And that might be the only way we can get some guidance while we’re working towards better data.”
Another approach would be to capitalize on any opportunity, where blood samples are obtained from healthy children: lead screening, allergy testing, cord blood, even through the IV line when a child has a tonsillectomy. “We need to look for everything we can.” Says Dr. Nakamoto,” We can protect the child’s privacy and comply with legal requirements but what a waste not to collect the information when there’s the opportunity. We need to be as creative as possible.”
Enabling Smaller Sample Volumes
Another area that has traditionally presented difficulties in pediatric testing is the small sample volumes that are typically obtained. New laboratory technology is helping to address this. It’s now possible, for example, to perform tests on smaller and smaller volumes of sample. “Before, it use to be a problem,” notes Dr Nakamoto, “because drawing even one more milliliter of blood from a 600g infant could mean yet another transfusion for that patient. But, thanks to advances such as the use of tandem mass spectrometry, we now have the ability to do a battery of tests on a single draw, for example, an entire panel of 13 adrenal steroids on 100 micro liters – two big drops -- of serum. Certainly a lot of things that just weren’t feasible to do before now are becoming possible. When you add to this some innovative process changes in how you handle pediatric samples, you may be able to cut your “Quantity Not Sufficient” rate by half or more. And that provides another ray of hope.”
- Schnablet al. Closing the Gaps in Paediatric Reference Intervals: The CALIPER Initiative. Clin Biochem Rev Vol 29 November 2008 89-96 http://www.sickkids.ca/Caliperproject/Laboratory-professionals/Study-design-paediatric-reference-intervals/index.html
- Coffin. Pediatric Laboratory Medicine Current Challenges and Future Opportunities; Am J Clin Pathol 2002;117:683-690