# General Movements Assessment and Amiel-Tison Neurologic Examination in Neonates and Infants: Correlations and Prognostic Values Regarding Neuromotor Outcomes

## TL;DR

This study explores how two neurological exams predict cerebral palsy and motor delays in infants, finding that combining them improves prediction accuracy.

## Contribution

The study demonstrates that combining General Movements Assessment and Amiel-Tison exams improves predictive value for cerebral palsy and motor delays in infants.

## Key findings

- Absence of fidgety movements at 12 weeks CA is the strongest predictor for CP and motor delays.
- Combining GM and AT exam results improves prediction of CP risk and delayed sitting in preterm infants.
- Logistic regression models using combined exam findings reached statistical significance for CP and motor outcomes.

## Abstract

Background: Our study aimed to investigate whether the general movements assessment (GMA) and Amiel-Tison Neurologic Examination performed at term-equivalent age (TEA) and 12 weeks corrected age (CA) could predict the occurrence of cerebral palsy (CP) and delayed gross motor milestones in a sample of term and preterm infants and whether the predictive values could be increased by using the two examinations during the same visit. Methods: A total of 70 infants (62 preterm and 8 at term) were examined at TEA and 12 weeks CA using GMs (General Movements) and Amiel-Tison (AT) examinations. We determined the correlation between the results of the examinations and several selected items (scarf sign, popliteal angle, and axial tone) and neuromotor outcomes (presence of cerebral palsy (CP), independent sitting, and independent walking). We attempted to build binary logistic regression models using items from both examination techniques to assess whether the combined use of the two exams could have a better predictive value than using one technique alone. Results: We analyzed the entire group and, separately, the subgroup of preterm infants. For the whole group, there was a statistically significant correlation between the GM examinations at TEA and 12 weeks CA (p < 0.008) as well as between the results of GM and AT exams performed at TEA (p < 0.001) and 12 weeks CA (p < 0.001). The strongest individual predictor for CP in the whole group was the absence of fidgety movements at 12 weeks CA (AUC (Area Under the Curve) = 0.873; CI (confidence interval) 95%: 0.748–0.998; p < 0.001) and a non-optimal result at the synthesis of findings of AT exam at 12 weeks CA (AUC = 0.755; CI 95%: 0.617–0.892; p < 0.003). In the subgroup of 62 premature neonates, absent fidgety movements at 12 weeks CA (AUC = 0.925; CI 95%: 0.819–1.031; p < 0.001) and a non-optimal result in the synthesis of findings of AT exam at 12 weeks CA (AUC = 0.772; CI 95%: 0.620–0.924; p < 0.005) were statistically significant predictors for the risk of CP. In the case of delayed sitting and delayed/absent walking, absent fidgety movements and non-optimal results in the AT exam at TEA and 12 weeks CA were the strongest individual predictors in the whole group as well as in the subgroup of preterm infants. The following exploratory logistic regression models reached statistical significance: one model combining absent fidgety + abnormal scarf sign + abnormal popliteal angle at 12 weeks CA for CP in the whole group (p < 0.032) and preterm infants (p < 0.034) and for delayed sitting in preterm infants (p < 0.041) and a second model combining absent fidgety + abnormal scarf + abnormal popliteal + abnormal axial + abnormal synthesis for CP risk (p < 0.046) and delayed sitting (p < 0.001) in preterm infants at 12 weeks CA. Conclusions: The absence of fidgety movements at 12 weeks CA represented the strongest individual predictor for the occurrence of CP and delayed gross motor milestones in our sample, both in the whole group and the subgroup of preterm infants. The combination of GM and AT findings reached statistical significance for the detection of CP risk in the whole group and CP risk and delayed sitting in the subgroup of preterm infants. Due to sample size limitations, the results should be regarded with caution, and larger studies are needed before robust conclusions can be drawn.