**1. Introduction**

Despite improved perinatal medical care and increased focus on enhanced nutritional support to premature infants, pre- and postnatal growth-restriction still occurs in 60%–100% of infants with a very low birth weight (<1500 g) [1,2]. Premature infants with growth restriction are at risk of impaired cognitive function and adverse metabolic and cardiovascular outcomes later in life [3,4]. Metabolic changes occurring *in utero*, during birth and the postnatal weaning period, seem to be of particular importance for future health [5–7]. Nutritional alterations during these periods are associated with a predisposition to obesity, cardiovascular diseases and associated co-morbidities later in life [5,8]. However, the time frame for these programming effects on long-term disease risk is controversial. Present evidence favors proactive nutritional support in premature infants to promote growth similar to the intrauterine growth rate and to support cognitive development [1,3,4]. This is in contrast with the potentially advantageous effects of relative undernutrition and slower growth on long-term cardiovascular health [8].

Recently, we published results from a randomized, controlled trial comparing the effect of enhanced nutritional supply (intervention) as opposed to a standard (control) diet, on postnatal growth in premature infants with a birth weight < 1500 g [9,10]. The infants in the intervention group, with median nutrient supplies in the upper range of current recommendations (Table 1) [9,11,12], exhibited postnatal growth along their birth percentiles for both weight and head circumference, whereas the control infants fell off their expected growth trajectories from birth to 36 weeks postmenstrual age (PMA). However, a preplanned safety analysis after the enrolment of 50 infants revealed an increased occurrence of late onset septicemia without increased mortality in the intervention group, and it was decided to halt further recruitment [10].

**Table 1.** Daily nutrient supply up to four weeks after birth.


Detailed records of actual nutrient supply was available for 44 infants [9]. Data are presented as medians (interquartile ranges) and compared using the Mann-Whitney U test.

To assess the metabolic status of these premature infants and to explore potentially different responses to the two diets, we used state-of-the-art nuclear magnetic resonance (NMR)-based metabolomics to analyze urine samples obtained during the postnatal period. Metabolomics is recognized as a powerful top-down systems biology approach that explores the geneticenvironment-health interaction [13,14]. The approach is to obtain broad snapshots of the metabolism by detecting and quantifying hundreds of small-molecular substances (molecular mass < 1000 Da) in tissues or body fluids, and then link them to disease or development states using multivariate statistical methods such as principal component analysis (PCA) and partial least squares (PLS) regression that handle and integrate large datasets [15,16]. Metabolomics of biofluids is thought to be a promising new tool in neonatology, especially in premature infants, due to its comprehensive and usually non-invasive nature [17]. Metabolomic analysis of urine from the neonatal period may be used to understand metabolic processes linked to early nutrition. It may also be used to identify biomarkers for diagnosis, prognosis and risk prediction of different diseases [5,13,17–19].

The main objectives of the present study were to analyze urine samples from the premature infants of the previous trial in relation to the two different nutritional exposures and to assess the infants' postnatal metabolic maturation. Secondary objectives were to explore potential differences related to age, sex, infections as well as pre- and postnatal growth.

#### **2. Materials and Methods**

#### *2.1. Study Design and Population*

The study was part of an open, randomized, controlled clinical trial [9,10], approved by the Regional Committee for Medical and Health Research Ethics and in accordance with the principles of the Helsinki Declaration. Fifty premature infants with birth weight < 1500 g were recruited from the neonatal intensive care units at Oslo University Hospital and Akershus University Hospital, Norway, from 17 August to 21 December 2010; 24 in the intervention group and 26 in the control group. Exclusion criteria were congenital malformations, chromosomal abnormalities, critical illnesses with short life expectancy and clinical syndromes known to affect growth and development. Morbidities were registered according to routine clinical practice and standard definitions [20–23]. Infants were classified as small for gestational age (SGA) if their birth weight was below the 10th percentile of a reference population [24], or as appropriate for gestational age (AGA) otherwise. Growth velocity was calculated by the exponential equation described by Patel *et al.* [25].

Two infants in the control group died during the first week of life, leaving 48 infants for the analysis [10]. Demographic and clinical characteristics are presented in Table 2. The significantly higher occurrence of septicemia and electrolyte deviations observed in the intervention group have been reported previously [10].


**Table 2.** Baseline characteristics and clinical outcomes.

Student *t*-test or Fisher's exact test was applied as appropriate. NEC: necrotizing enterocolitis; IVH: intraventricular hemorrhage; PVL: periventricular leukomalacia; ROP: retinopathy of prematurity; PDA: persistent ductus arteriosus; PMA: post-menstrual age; Hypophosphatemia < 1.4 mmol/L; Hypokalemia < 3.5 mmol/L. <sup>a</sup> Two more infants died during hospitalization, leaving 46 infants in the analyses of ROP and O2-dependency at 36 weeks PMA; <sup>b</sup> Only 45 infants had their phosphate concentrations measured during the first week of life.

The nutritional intervention was started on the first day of life, after informed consent was obtained [9,10]. The intervention group started with 3.5 g/kg/day of amino acids and 2.0 g/kg/day of intravenous lipids, whereas the control group started with 2.0 g/kg/day of amino acids and 0.5 g/kg/day of lipids. To improve the parenteral supply of the long chain polyunsaturated fatty acids (PUFAs), the intervention group received a lipid emulsion containing marine omega 3 fatty acids (SMOF®, Fresenius Kabi Norge AS, Oslo, Norway), whereas the control group received the lipid emulsion used in our units (Clinoleic®, Baxter AS, Oslo, Norway). The supply of human milk was increased equally in both groups, and standard fortification with 4.2 g Nutriprem® (Nutricia Norge AS, Oslo, Norway)/100 mL human milk was initiated when the infants tolerated 110 mL/kg/day as enteral supply. In addition to standard fortification, the intervention group was given 0.6 g Complete Amino Acid Mix® (Nutricia Norge AS, Oslo, Norway)/100 mL human milk, 60 mg/kg/day of docosahexaenoic acid (DHA; 22:6, *n*-3) as well as arachidonic acid (AA; 20:4, *n*-6), and 1500 ȝg/kg/day vitamin A (Ås Laboratory, Ås, Norway). On average the energy supply was approximately 10% higher and the protein supply 25% higher in the intervention group as compared to the control group [9].
