Author(s): Apolline Wittwer *, Jean-Michel Hascoët
Inappropriate nutrition is an issue for premature infants as it may be responsible for postnatal growth restriction (PGR) and increased morbidity [1-4]. Embleton  demonstrated that 55% of PGR  were related to a cumulative deficit in energy and protein intake. Malnutrition in the first weeks of life is associated with short stature and adverse outcomes in adulthood [6-8]. The goal of nutrition is to allow body composition and outcomes similar to those of infants born at term [3,9]. Optimizing nutrition needs to start from birth as the window for PGR prevention and catch-up growth is rather narrow . The prevention of nutrient deficits may be achieved through the implementation of optimized nutritional policies. More recent studies also demonstrated benefits on growth with the implementation of nutritional strategies [11-13].
We decided therefore to introduce a standardized nutritional protocol in our NICU and study outcomes on premature infants (including growth and morbidity).
In our unit, parenteral and enteral nutrition used to be adjusted according to the 2004 guidelines defining global nutritional intakes [14,15]. From May 1st , 2014, we decided to update our policy according to more recent ESPGHAN and other recommendations [16-19] available at the time of the study, and to follow a strict nutritional protocol rather than only global guidelines.
The objective of this study was to evaluate the longitudinal impact of implementing this strictly-defined nutrition protocol on very premature infants' growth and morbidity using a before/after comparison design, with a 6-months wash-out period. The secondary objective was to evaluate physicians' compliance to the standardized protocol.
Patients and methods
All infants born at the maternity hospital and hospitalized in our NICU were studied retrospectively during two 6-month periods, separated by a 6-month washout period, from May 1 to October 31, 2013 (group 1) then from May 1 to October 31, 2014 (group 2), which occurred before and after the introduction of an optimized nutrition protocol. We performed a longitudinal comparative analysis between these two independent groups. Collection of data from the infants' files was standardized.
This retrospective study was registered by ClinicalTrials.gov under number NCT03217045, and by the French ethic committee "Commission Nationale de l'Informatique et des Libertés" (CNIL) under study number R2015-1 for the Maternity of the CHRU of Nancy. Parents' consent for using the collected clinical data of their infant was obtained and signed at admission.
Infants born between 26 to 32 weeks gestation (GA) and admitted to our NICU were included in the study. Infants who died before discharge or presented with any congenital malformation were excluded. The primary outcome measure was to evaluate the impact of introducing a well-defined nutrition protocol on the longitudinal growth of the infants up to the time of discharge. Weight was assessed by daily measurements, every morning, as defined in our routine policy of care. To account for variations in gestational and postnatal age, body weight was converted into a z-score using the Olsen preterm infants' reference growth chart .