Children who are shorter than expected based on their parents’ heights may have a genetic predisposition for short stature, not an underlying growth disorder.
Vanderbilt University Medical Center researchers report in Genome Medicine that a polygenic score for height — a composite measure of over 1 million common genetic variants that influence height — could help improve diagnosis for children with short stature.
Short stature is defined as having a height that is below the third percentile for the average height of a given age, sex and population group.
“The most common reason for short stature in children is simply having shorter parents, which is captured clinically as the ‘mid-parental height’ — the average of the parents’ heights,” said John Shelley, a student in the Medical Scientist Training Program who led the research with Jonathan Mosley, MD, PhD. “But some children have short stature despite average-height parents, raising concerns about underlying medical conditions. These children are typically referred to pediatric endocrinology specialists for further evaluation.”
Even after comprehensive testing, however, about 30% of these children do not have a definitive diagnosis and are classified as having idiopathic short stature (ISS), Shelley noted.
“Having this diagnostic label can lead to extended surveillance and testing, and to anxiety — for parents and children — about the potential for an undiagnosed disease to be causing the short stature,” said Jill Simmons, MD, professor of Pediatrics and a co-author of the study.
The researchers calculated polygenic scores for height for 534 pediatric patients who had been evaluated for short stature by an endocrinologist and who were participants in BioVU, VUMC’s de-identified DNA biobank and linked electronic health records. Of the 534 patients, 20% had no identifiable cause for short stature and were classified as having ISS.
The investigators found that many children had a stronger genetic predisposition to short stature than predicted by mid-parental height alone. They also discovered that children eventually diagnosed with ISS exhibited a stronger polygenic predisposition for short stature compared to those with an identified medical cause, such as growth hormone deficiency or inflammatory bowel disease.
“Incorporating polygenic scores in the initial evaluation improved the ability to distinguish ISS from short stature caused by underlying medical conditions, even after accounting for parental heights,” Shelley said. “This increased accuracy could lead to fewer unnecessary diagnostic tests, reduced health care utilization, and decreased anxiety for patients and their families.”
The researchers also identified a subgroup of children with ISS who had a family history of short stature but a weaker polygenic predisposition. This difference suggests these patients may carry rare single-gene mutations that are not captured by common-variant polygenic scores. Children with this pattern of genetic predisposition could benefit from advanced genetic testing like whole exome or genome sequencing, the researchers said.
Although more research is needed, the findings suggest that mid-parental height predictions may miss genetic factors that are captured by the polygenic score for height. Adding this score to evaluations of short stature could improve diagnostic accuracy and ultimately, clinical outcomes for children with ISS.
“Using a polygenic score for height to identify patients at lower risk for growth disorders would provide reassurance to clinicians and parents and help reduce unnecessary diagnostic testing,” said Simmons, who holds the Directorship in Pediatric Metabolic Bone Diseases.
Mosley was an associate professor of Medicine and Biomedical Informatics at VUMC; he joined the faculty of UT Southwestern Medical Center in Dallas this month. Other co-authors include Mingjian Shi, PhD, Josh Peterson, MD, MPH, and Sara Van Driest, MD, PhD. The research was supported by the National Institutes of Health (grants R01GM130791, U01HG011181, T32GM007347).
