As early as possible, neonatologists try to identify neonates at risk of unfavorable neurodevelopmental outcomes. They are fairly reliable in predicting very poor outcomes as well as optimal outcomes. However, within these two extremes, the prediction still remains a challenge. Immaturity of the neonatal brain constitutes a limit in itself. During decades with the growing knowledge of brain development, many methods have been developed for neurological assessment of the neonate. Neither of them applied alone was perfect in terms of clinical applicability, sensitivity, reproducibility and specificity.

The motor function is the first to provide the clinician with clues. Higher functions, in particular language and other cognitive functions, will develop later. However, recent researchers give credit to the brainstem for controlling exceedingly rudimentary learning-related cognitive-like activity. At present, the anticipation of late emerging developmental disabilities remains difficult even though early motor dysfunction has repeatedly been associated with a higher risk of intellectual or other learning disabilities. Despite our modest recent contribution to the domain of prediction, further studies on welldefined high risk populations with rigorous methodology that aim to demonstrate these links are still needed. Besides neurological observations, research is in process of including behavioral and stress/ reactivity measures; feasibility and benefits have to be demonstrated.

At present, fetal neurology is supported by neonatal neurology. Obstetricians are wise enough to take from both methods described above the elements they are able to transpose to fetal life. A comparative table of neonatal and fetal assessment is to be found elsewhere. As for neonatal neurology, the future of fetal neurology will have to rely on short- and long-term follow-up studies to define the predictive value of the chosen items. Obstetricians will have to be as patient as pediatricians, to work, step by step, towards defining optimality and impairment. They will have to be very careful when deciding to interrupt pregnancies; at the time being, such decisions are restricted to cases of very severe impairment. In line with the spectrum described above, they can expect to find more cases with moderate to mild abnormalities than cases with severe ones. However the most pleasant aspect for the echographer is to check fetal optimality. Just as a newborn infant categorized as at risk of brain damage is competent enough to demonstrate CNS integrity from birth, a high risk fetus will soon be competent enough to demonstrate CNS integrity before birth.