Growth is traditionally used as a prognostic measure after admission to a neonatal intensive care unit (NICU). Quality of neonatal intensive care is one of the factors determining the health and quality of life for those that survive, with the principal objective of increasing disability-free survival. Nowadays, there is some researches showing us that these preterm babies present a different body composition when they achieve term age compared with babies birth at term and this condition could be putting these babies in risk to metabolic syndrome early in adult age. The majority of infants born between 24 and 29 weeks of gestational age fail to achieve the mean birth weight for fetuses of the same gestational age, and many weigh below the 10th percentile at hospital discharge. The focus of this study is the growth and body composition of very low birth weight infants growing in the extrauterine environment. The central question is, when they reach the corresponding term weight, is the body composition of newborns growing outside of the uterus different from that of infants growing in the intrauterine environment? The investigators are also trying to validate 3 different methods to research body composition: DXA, air-displacement plethysmography and electric bioimpedance.

OBJECTIVES

GENERAL:

Evaluate the growth and body composition of VLBW infants upon reaching the corresponding term gestational age (GA) and weight.

SPECIFIC:

A) Evaluate the longitudinal growth and body composition of VLBW infants from the first week of life and at the corresponding term GA and weight.

B) Relate the weight loss in the first week of life to the weight and percentage of total body water at the corresponding term GA and weight.

C) Compare the body composition of VLBW newborns growing in the extrauterine environment upon reaching the corresponding term GA and weight with the body composition of term newborns, using the 3 methods (DEXA, air-displacement plethysmography, and electric bioimpedance) D) Adjust a predictive model for body weight estimates using the parameters furnished by multi-frequency and single-frequency electric bioimpedance and anthropometry, with the measurements provided by DEXA and plethysmography as the gold standard.

E) Relate body weight and composition at the corresponding term GA to neonatal morbidity.

METHODOLOGY Study characteristics A mixed longitudinal study will be conducted. The newborns will be included at different gestational ages and will remain in the study until reaching the corresponding term gestational age and weight (2,800g to 3,200g).

Study site The study will be performed at the Fernandes Figueira Institute (IFF), a unit of the Oswaldo Cruz Foundation in Rio de Janeiro, Brazil.

Study population The study population will consist of premature newborns with gestational age less than 32 weeks and/or with intrauterine growth restriction. Data will be collected beginning at birth during hospitalization and in the outpatient follow-up after hospital discharge.

Definitions and established criteria

Evaluation of the adequacy of weight in relation to gestational age was based on the curve by Alexander et al. (1996), defining SGA as newborns with birth weight below the 10th percentile in relation to gestational age; AGA was defined as between the 10th and 90th percentiles; and large for gestational age (LGA) as above the 90th percentile.

Intrauterine growth restriction (IUGR) This study defines IUGR as a Z-score for the newborn's birth weight for gestational age less than or equal to -2. The study will use the sex-specific Canadian reference for birth weight for gestational age, which uses the 3rd, 5th, 10th, 50th (mean), 90th, 95th, and 97th percentiles for 22 to 43 complete weeks and standard deviation (Kramer et al, 2001).

Z-score. The measurements to be used with references for obtaining the Z-score will be those of Fenton et al.

Gestational age Gestational age will be based on the mother's date of last menstrual period. In case of doubt, confirmation will be obtained by the date of gestational age estimated by first-trimester ultrasound or, if necessary, by the Ballard method (Ballard et al, 1991), which estimates gestational age based on the newborn's somatic characteristics and neurological evaluation.

Corrected gestational age After obtaining gestational age at birth, age will be corrected over the course of the study by adding the weeks of post-natal life. Each age in weeks will consider the variation from zero to six days.

Morbidities

• Bronchopulmonary dysplasia will be defined as any newborn in use of oxygen therapy at 36 weeks corrected gestational age.
• Late-onset sepsis will be defined as a positive blood or CSF culture in the presence of clinical signs of sepsis after 72 hours of life or clinical infection with a negative culture after 72 hours of life for newborns that have received 5 or more days of antibiotic therapy.
• Necrotizing enterocolitis will be considered when classified as stage II A or higher according to the modified Bell criteria (Walsh & Kleigman, 1986).
• Severe intraventricular hemorrhage will be defined as grade 3 or 4 according to the scale described by Papile et al (1978).

Measurements of body weight, length, foot size, triceps skinfold, mid-arm circumference, and total body water will be taken at 5 moments in the premature infants: in the first 24 hours of life, at the 7th day of life after recovering birth weight, at term (GA ≥ 37 < 42 weeks), and with weight from 2,800 to 3,200g. The two latter measurements may be taken after hospital discharge, during outpatient follow-up. These variables will be evaluated in the first 24 hours of life in the term newborns with weight from 2,800g to 3,200g.

Anthropometric Evaluation Measurements will be performed by the researcher, following previous training, considering the mean of the three measurements. The newborn will be measured unclothed and kept under previously prescribed heating, except during weighing.

Weight Weight will be measured with a Filizola brand digital scale accurate to 5 grams, after taring it before each weighing.

Length Length will be measured using an anthropometric ruler designed and manufactured at the precision workshop of the University of São Paulo in Ribeirão Preto, São Paulo State.

The newborn will be placed in the supine position on a flat surface as stiff as possible, adjusting the child's head in a medial position on the ruler's fixed surface. Holding the child in this position with the knees extended by an assistant, the ruler's mobile extremity will be moved until the ruler is perfectly adjusted to the soles of the newborn's feet.

Head circumference Head circumference will be measured using a non-elastic tape measure marked in millimeters. The tape measure will be adjusted to the newborn's head, passing in front across the supra-orbital region, above the eyebrows, and behind across the occipital bulge, recording the largest circumference obtained.

Foot size The measurement is taken with a caliper marked in millimeters, adjusting it to the largest diameter of the infant's right foot.

Triceps Skinfold (TSF) Triceps skinfold measurements will be taken with a Harpenden type caliper, which measures in millimeters, accurate to 0.1mm.

This measurement will always be taken on the right arm, at the midpoint between the acromion and the olecranon, immediately below the skin hollow, with the arm extended. The measurement recorded will be that obtained after stabilizing the caliper pointers (approximately 60 seconds).

Mid-arm circumference (MAC) This circumference will be obtained, always from the left arm, at the midpoint between the acromion and the olecranon, just below the skin hollow, with the arm extended and the hand in prone position, with a non-elastic millimeter tape measure (Sasanow et al, 1986). The location is the same as that of the triceps skinfold.

Mid-arm muscle circumference (MAMC) This parameter aims to measure the arm's muscle size and thus indirectly indicate the newborn's nutritional status during extrauterine growth. It uses 2 combined measurements, the triceps skinfold and the mid-arm circumference.

Body composition evaluation Body composition will be evaluated using the 3 methods available at the Fernandes Figueira Institute (BIA, plethysmography, and DEXA). The study will measure total body water, lean mass, and body fat.

Bioimpedance measurements will be made with the RJL System (Quantum BIA 101Q equipment) and the Hydra multi-frequency equipment (Xitron technologies) . Both pieces of equipment are inexpensive and the tests are non-invasive and can be performed at bedside. However, neither of them is validated for measuring body fat in newborns and infants (first year of life).

We will use quadripolar electrodes preferably on the right hand and foot. The electrodes are cut in ± 1 cm strips and attached to the hand and foot with a minimum distance of 3 cm between them (wrist and ankle flexure and immediately before the base of the fingers and toes). The anatomical criterion for positioning the electrodes was chosen based on the fact that the study focuses on human beings during their growth process, which can double or triple the birth weight during the study period. The newborns are measured before feeding, in prone position, wearing dry disposable diapers. The arms are kept away from the trunk and the legs should not touch each other. The infant should be comfortable and quiet, and the measurements should be recorded when the variation is no greater than ± 1 point. The difference in the resistance values obtained between the distal positions (right hand and right foot, right hand and left foot, left hand and left foot, and left hand and right foot) is not significant, and the electrodes can thus be used in these positions.

The equations used to calculate total body water (TBW) will be those proposed and validated by Tang Wing et al in 1997, which use 2 anthropometric measurements (weight and foot length) and resistance (R) measured by the electric bioimpedance technique.

Body fat percentage and total body composition will be measured by the air-displacement plethysmography in the PEA POD System and in the iDEXA equipment, using the software for small animals. Based on these measurements of real body fat, predictive models will be adjusted to estimate body fat using the parameters obtained from bioimpedance (resistance and reactance) and anthropometry (Hull et al, 2008, Roggero et al, 2008, Ellis et al, 2008, Ma et al, 2008).

Statistical analysis Data analysis will use Epi Info (Epi - 6.04) from the Centers for Disease Control (CDC - Atlanta, USA).

We will use analysis of variance for continuous variables with normal distribution and non-parametric Kruskal-Wallis tests for the others to analyze differences between the groups. The chi-square test will be used for the categorical variables to calculate the difference in proportions. A frequency table will be used for the categorical variables and the mean or median and standard deviation for the continuous variables.

The models for estimating body fat based on bioimpedance measurements will be fitted using multiple linear regression models.

Inclusion Criteria:

• appropriate for gestational age (AGA) and small for gestational age (SGA) premature newborns with gestational age less than or equal to 32 weeks and less than 1500g at birth
• AGA term newborns with gestational age greater than or equal to 37 weeks and less than 42 weeks and with no history of maternal disease that might compromise intrauterine growth

Exclusion Criteria:

• congenital malformations
• genetic syndromes confirmed by the geneticist
• clinically and laboratory-confirmed congenital infections
• children of mothers with the human immunodeficiency virus (HIV)
• children of diabetic mothers
• newborns that develop severe intraventricular hemorrhage or necrotizing enterocolitis

Preterm infants with birth weight less than 1500g

Term infants without IUGR