MRI scoring systems accurately assess cognitive, motor development in infants

Existing scoring systems for magnetic resonance imaging (MRI) scans are sufficiently predictive of cognitive and motor outcomes in term-born infants, a recent study has found. For language assessment, more detailed scoring inventories seem to perform better.

The study included 90 term-born neonates who showed signs of perinatal asphyxia and underwent MRI within the first 2 weeks of life. A single blinded assessor was asked to score the scans according to three systems: Barkovich, Weeke, and National Institute of Child Health and Human Development (NICHD) Neonatal Research Network (NRN).

Of the 90 enrolled infants, 66 returned for follow-up at 2 years of age, during which their neurodevelopmental outcomes were assessed using the Bayley Scales of Infant and Toddler Development, third edition (Bayley-III).

Bayley-III found nine infants with cognitive impairment (14 percent), four of whom had severe impairment. Language impairment was documented in 10 infants (16 percent) and motor impairments in seven (11 percent).

Of the MRI scores, the Weeke system detected baseline abnormalities with the highest frequency at 77 percent (Barkovich: 38 percent; NICHD NRN: 47 percent). The same remained true at follow-up (Weeke: 77 percent; Barkovich: 40 percent; NICHD NRN: 50 percent).

Notably, Bayley-III cognitive composite score correlated strongly and significantly with the Barkovich (R², 0.1269; p=0.0102), NICHD NRN (R², 0.2236; p=0.0026), and Weeke (R², 0.2645; p<0.0001) scores. Bayley-III motor composite scores were likewise associated with all three MRI systems (Barkovich: R², 0.1053; p=0.0204; NICHD NRN: R², 0.2703; p=0.0006; Weeke: R², 0.3018; p<0.0001).

However, only the Weeke scoring system significantly correlated with Bayley-III language composite scores (R², 0.1747; p=0.0022).