Magnetic Resonance Spectroscopy (MRS) and Diffusion Tensor Imaging (DTI) Neuroimaging Combined With the Specific Test of Early Infant Motor Performance (STEP) Early Developmental Assessment Predicts One-Year Developmental Outcomes

BACKGROUND: Premature birth adversely affects infant and childhood developmental trajectories. Despite the absence of obvious neurological injury such as intraventricular hemorrhage or periventricular leukomalacia, school-age children born prematurely show microstructural differences in the brain compared with those born at term (Duerden, et. al., 2015). The Specific Test of Early Infant Motor Performance (STEP) is a novel infant motor skills test that has been introduced as an efficient developmental screening assessment for preterm infants that correlates with 12-month Bayley-III outcomes (Coker-Bolt et. al., 2016; Gower, Jenkins, Fraser, Ramakrishnan, Coker-Bolt, 2018). Neuroimaging with MRI provide a non-invasive means to quantify the neurodevelopmental status of infants through metabolic and structural measures. These measures may predict future neurodevelopmental outcomes such as functional motor and/or cognitive deficits. Further validation of the STEP via neuroimaging will therefore provide support that the STEP is assessing functions that are closely related to CNS structures.

PURPOSE: To determine the relationship of the STEP to neuroimaging through Magnetic Resonance Spectroscopy (MRS) metabolite ratios and Fractional Anisotropy (FA).

DESIGN AND METHODS: Prospective cohort study of 16 preterm infants were assessed using STEP outcome measures at 0-3 months corrected gestational age (GA), MRS and Diffusion Tensor Imaging (DTI) at term GA, and Bayley-III at 12 months. The STEP was administered by the same OTt, while a different, blinded OT administered the Bayley-III. Both therapists were blinded to neuroimaging results. Generalized linear models were created using MRS metabolite ratios or white matter (WM) tract FA value averages as covariates to predict STEP. Generalized linear models were created using the metabolite ratios or average FA values and STEP score at term or 3 months to predict Bayley gross motor (GM) and cognitive scores at 12 months.

RESULTS: (n=16); 2 infants with neonatal blood-culture proven sepsis, 2 infants with chorioamnionitis and periventricular leukomalacia, 2 infants had intraventricular hemorrhage, 1 infant with large cerebellar hemorrhage. Ratios of N-acetyl aspartate to creatine (NAA) in both frontal WM and basal ganglia (BG) significantly contribute to a model predicting STEP scores at both term and 3 months. NAA and myoinositol ratios in the BG both significantly predict Bayley cognitive scores at 12 months. Additionally, NAA in the WM and BG interacts with STEP scores to contribute to a model predicting Bayley GM at 12 months. FA values in the posterior thalamic radiations (PTR) and rostral limb of the internal capsule (RLIC) predict scores on STEP at both time points. Multiple, distinct WM tracts also contribute to STEP models predicting Bayley cognitive and gross motor scores at 12 months.

CONCLUSION: Performance on the STEP is related to MRS metabolite NAA and multiple WM tract FA values. Neuroimaging data in a model with STEP scores strongly predict 12-month Bayley outcomes. These findings suggest that performance on the STEP relates to CNS structural and metabolic integrity and that both neuroimaging and early motor performance can be used to better predict later development. Limitations include small sample size, future studies include further validation of STEP in larger cohort.

IMPACT STATEMENT: Neuroimaging combined with early movement assessment via the STEP can predict long-term motor development. Scores on the STEP may relate to metabolic, functional and structural integrity in various left and interhemispheric brain regions. STEP assessment could lead to more accurate detection of infants at high-risk of motor delay and referral to earlier OT services.

References

1. Gower L, Jenkins D, Fraser JL, Ramakrishnan V, Coker-Bolt P. (2018). Early developmental assessment with a short screening test, the STEP, predicts one-year outcomes. Journal of Perinatology. DOI: 10.1038/s41372-018-0234-4

2. Duerden, E., Foong, J., Chau, V., Branson, H., Poskitt, K., Grunau, R., Synnes, A., Zwicker, J.G, & Miller, S.P. (2015). Tract-Based Spatial Statistics in Preterm-Born Neonates Predicts Cognitive and Motor Outcomes at 18 Months. AJNR Am J Neuroradiol 36:1565-1571.

3. Coker-Bolt, P., Barbour, A., Moss, H. Tillman, J., Humphries, E., Ward, E., Brown, T., Jenkins, D. (2016). Correlating early motor skills to white matter abnormalities in preterm infants using diffusion tensor imaging. J Pediatr Rehabil Med 9:185-193.