COLUMBUS, Ohio, April 7 (UPI) -- Using data to determine abnormalities in gray matter in premature babies' brains is a better method for predicting later developmental delays or impairments, according to researchers in Ohio.
Researchers at Nationwide Children's Hospital found in a recent study that gyral maturational delay and diffuse cystic abnormality gray matter of the brain were better predictors for neurodevelopmental impairments, cognitive delays and cerebral palsy than opinion-based scoring systems with extremely-low-birth-weight premature infants.
While MRI tests are often used to evaluate brain development in premature babies, the researchers say the scoring system is heavily based on expert opinion.
Unlike an opinion-based system of diagnosis, which suggests an expert has certain expectations of what they'll see, the researchers say looking at their study data for indicators of delays or impairment allows for a more objective interpretation.
While the two predictors were much more accurate at prediction than using opinion, the researchers caution the absence of either predictor didn't necessarily mean there were not impairments in infants in the study.
"We can't predict with certainty that these babies are going to do well just because the MRIs looked good," Dr. Laura Slaughter, an assistant professor of clinical pediatrics at Ohio State University, said in a press release. "These are still significantly premature babies that we need to monitor."
For the study, published in the journal Neonatology, researchers performed MRIs on 122 extremely-low-birth-weight infants -- weighing under 2.2 pounds -- at the age they would have been at a 40-week full-term pregnancy.
The researchers measured the MRIs by conventional algorithms, and using a multivariable regression model to predict cerebral palsy, bilateral deafness, bilateral deafness requiring amplification, cognitive or language delays or a combination by the age of 18- to 24-months-old, comparing the results of each evaluation.
The predictors were highly specific -- when either was detected, impairments were found 95 to 99 percent of the time -- but their absence did not predict normal development, only being correct 30 to 67 percent of the time.
Slaughter said most previous studies have focused on white matter, as do doctors counseling families, but said the study shows gray matter may be a superior measure for predicting delays or impairments in order to start therapies earlier and attempt to prevent them.
"We let the data drive our model," Slaughter said. "We measured numerous individual imaging factors and their correlation to outcomes, instead of deciding ahead of time what we believed would be important. There is still some subjectivity, and neuroradiologists are going to have slightly different readings or interpretations. But our model is more objective."