Correlation Between Cerebral Oxygenation and Neurodevelopment in VLBW Preterm Infants.

The improvements in neonatal intensive care occurred over the last decades have led to a substantial increase in the survival rate of preterm infants, especially of extremely low birth weight. Nevertheless, the delicate maturation of central nervous system (CNS) in the extrauterine environment, together with the possible development of prematurity-related clinical complications that could result in brain injury, place this population at high risk for long-term neurodevelopmental disabilities.

Significant anatomical changes in brain structures have been associated with preterm birth; among these, enlarged ventricles and subarachnoid spaces, reduced subcortical white matter and decreased cortical and deep nuclear grey matter are the most common. Additionally, there is increasing evidence of abnormal patterns of neuronal myelination and brain connectivity in preterm infants, with possible concerning long-term effects on language, motor and social skills. As for cognitive abilities, a direct correlation with gestational age (GA) has been previously shown in a large cohort of preterm born children at school age.

In its early phases, the developmental of CNS is characterized by an intrinsic neuroplasticity, according which repeated experiences could influence the arrangement of synaptic connections and neural circuitries, thus resulting in structural and functional changes. By exploiting this feature, an early establishment of supportive interventions¬ might aid to compensate the anatomical and functional constraints connected to preterm birth and to improve preterm infants' neurodevelopment.

The validation of clinical tools able to predict neurodevelopmental outcomes in the preterm population might help at identifying infants at greatest risk of impairment, who would benefit most from early rehabilitative interventions. Near infrared reflected spectroscopy (NIRS) provides a bedside, non-invasive, continuous monitoring of cerebral oxygen saturation (CrSO2) and has been largely adopted to assess neonatal brain oxygenation and perfusion in intensive care settings. Perioperative CrSO2 has been recently proposed as a possible predictor of later neurodevelopmental outcomes in neonates undergone cardiac surgery; data evaluating the correlation between CrSO2 and later neurodevelopment in preterm infants, however, are very scarce and limited to the first days of life; at this time, CrSO2 can be widely influenced by the occurrence of systemic haemodynamic instability and is blind to several clinical complications, known as potential causes of brain damage, that can occur later on during hospital stay and thus are not necessarily directly related to changes in cerebral oxygenation within the first days of life (e.g. necrotizing enterocolitis [NEC], late onset sepsis).

Hence, the aim of this study was to evaluate whether CrSO2 monitoring, performed before NICU discharge in clinically stable very low birth weight (VLBW) preterm infants, can predict psychomotor outcomes during the first 24 months of corrected age (ca).

Infants meeting eligibility criteria (for details see specific section) undergo a 3-hour continuous monitoring of CrSO2 by means of the INVOS 5100 oximeter when satisfying the following condition: stable clinical condition, no need for oxygen supplementation, full enteral feeding (defined as 160 ml/kg/day) After discharge, the enrolled infants are included in a long-term neurodevelopmental follow-up, which is part of the normal routine care for preterm-born infants and aims at the early identification and treatment of neurocognitive sequelae possibly related to preterm birth.

In this context, psychomotor outcome (PSO) is assessed at 6, 12, 18 and 24 months ca by means of the Griffiths Mental Development Scales 0-2 years [26]. These scales investigate five main areas (locomotor, personal and social skills, hearing and language, eye and hand coordination, performance), providing a general developmental quotient (DQ) of infant's abilities and five sub-scale quotients (SQ) for each of the developmental areas.

The Griffiths Scales are administered individually, over a 45-min period, by the same professional trained psychologist for the whole study period and are implemented in the presence of the infant's parent.

A correlation analysis between mean CrSO2 values and GQ/SQ scores at each follow-up appointment will be performed. In order to evaluate the effect of possible confounding variables on the observed results, a mixed model including a number of risk factors known to affect preterm infants' neurodevelopment will be built for hierarchical multiple regression analysis. Significance level will be set at p<0.05.