A Physiological Study to Determine the Enteral Threonine Requirements in Infants Aged 1 to 6 Months

INTRODUCTION

Threonine is an indispensable AA that must come from dietary sources. It is critical in the production of mucins in the gut (Law, Bertolo et al. 2007) and contributes significantly to collagen, elastin and tooth enamel formation (Kramsch, Franzblau et al. 1971, Robinson, Lowe et al. 1975). The current adequate intake (AI) for threonine in infants proposed by the Institute of Medicine, Food and Nutrition Board, in the 2005 Dietary Reference Intakes (DRIs) for macronutrients is based on a factorial approach due to the lack of conclusive empirical data. Studies that lead to the determination of amino acid (AA) requirements for infants used the nitrogen analysis technique and were performed over 20 years ago (Snyderman, Norton et al. 1959). Methodological advances have made it possible to determine AA requirements in humans with a more precise technique (Zello, Wykes et al. 1995). However, no studies of threonine requirement in infants (1- 6mo of age) with the use of the newer stable-isotope methods have been reported.

To accurately determine amino acid requirements, it is necessary to provide a range of amino acid intakes. For ethical reasons, infants cannot be fed very low or high test amino acid diets for prolonged periods of time. The minimally invasive IAAO method, initially developed in adults in our laboratory (Zello, Pencharz et al. 1993) and now used internationally (Huang, Hogewind-Schoonenboom et al. 2011) to determine amino acid requirements, overcomes this problem. The IAAO method takes less than 24 hours to conduct and has been used safely in parenterally fed neonates, enterally fed neonates, children and adults.

Our laboratory determined the enteral requirement of threonine in piglets (Bertolo, Chen et al. 1998) which was 0.42g/kg/d. We are able to extrapolate from this data, and propose that the enteral requirement for neonates would be 84 mg/kg/d. In this study we will be determining the enteral requirement for the 1 - 6 months old infant and predict that since growth slows and protein requirement decreases after the first month of life, the enteral requirement of infants will be similar to the DRI's AI of 73 mg/kg/d. Thus, the objective of this study is to determine the enteral Threonine requirement in infants from 1 to 6 months of age with the IAAO method.

SUBJECTS AND METHODS

Subjects: A total of 18 studies will be performed where participants will be randomized to 1 of 18 test threonine intakes. Infants between the ages of 1 and 6 months will be recruited from the in-patient post-surgical population of the Hospital for Sick Children (SickKids), Toronto, Canada. Participants may be eligible to participate in more than one study intake following enrolment, therefore being randomized to a second threonine intake. All study procedures and consent forms will have the approval of the research ethics board at SickKids. Permission to enroll the infants in the study will be obtained from the attending surgeon and written informed consent will be obtained from one or both parents/guardians. Recruitment of patients is expected to begin in December 2015. The study is registered with clinicaltrials.gov # NCT02364843.

Infants will be included in the study if they are between the ages of 1 to 6 months, were born between 32 weeks and 43 weeks gestational age, are clinically stable as determined by normal blood values and vital signs, and are in-patients who are being fed via nasogastric (NG-), jejunal (NJ- or J-), or gastric (G-) feeding tube. Infants will be excluded if they are receiving supplemental oxygen, are mechanically ventilated, have any endocrine or genetic anomalies which would influence protein and AA metabolism, or are receiving medication which would influence protein and AA metabolism (eg. Corticosteroid therapy).

Experimental design and study diet

The study design is the minimally invasive IAAO design, which is based on the concept that, because AA's are not stored in the body, the intake of dietary essential AAs relative to their use for protein synthesis determines whether they are oxidized or incorporated into protein. When the intake of one indispensable AA is restricted, the other indispensable AAs are in excess and are oxidized because they cannot be incorporated into protein as previously described (Zello, Pencharz et al. 1993). Each study will take place over an 18 - 24 hour period following confirmation from the infant's clinical dietitian that an appropriate enteral intake of protein and non-protein energy from the infant's feeds has been delivered over 24 to 48 hours preceding the study day. On the one study day (the duration which will be less than 24 hours; see Fig. 1), the infant will receive a formula made from a commercial protein free enteral formula (Pro-Phree) which supplies a known amount of carbohydrate, fat, minerals, trace elements and vitamins appropriate for an infant and to which an AA mixture will be added. Amino acids will be added in the pattern of breast milk as follows in mg·kg-1·d-1: Lys 207, Met 48, Leu 288, Ile 165, Val 165, Phe 126, Try 51, His 63, Tyr 156, Arg 69, Pro 240, Cys 51, Gly 69, Ala 114, Ser 150, Asp 270, Glu 534 (WHO 2007). Each infant will receive a different intake of Threonine per study and the 18 intakes for the 18 infants will fall on a continuum between 15 and 130 mg·kg-1·d-1. As the intake of threonine is varied, the formulas will be maintained as isonitrogenous by varying intakes of alanine which is a dispensable AA. The amino acid mixture will be put through a 0.22 µ filter to ensure it is sterile and pathogen free. This is an added safety step that is not done with commercial formulas but was done with our parenteral solutions when studying Threonine and Lysine requirements in parenterally fed infants (Chapman 2009, 2010). Combining the AA mixture and the commercially prepared Pro-Phree formulation will be done in the hospital's Formula Room under their standard operating procedure (SOP) guidelines for clinical formula preparation. The infants will receive the study formula via their feeding tube, or a combination of tube feeds and oral feeds, over the 18 - 24 hour study period as ordered by the clinical dietician.

A research-grade stable isotope tracer, manufactured by Cambridge Isotope Laboratories (CIL), Maine, USA, will be used to measure oxidation or protein deposition in response to the varying intakes of threonine. The stable isotope is an amino acid is L-[1-13C]Phenylalanine, and is administered with the infant's feeds at a dose of 15 µ·kg-1/h-1. This stable isotope is found in all humans as 1% of the total percentage of the amino acid phenylalanine and the stable isotope itself has been used successfully and without harm when given intravenously to infants in 4 previous studies conducted at The Hospital for Sick Children, Toronto Canada (Roberts 2001; Courtney Martin 2008; Chapman 2009, 2010).

We are using a study protocol that provides sufficient time for the infant to reach a steady state with the stable 1-13CPhenylalanine isotope above baseline values as determined by our infant parenteral studies (Chapman 2009, 2010). We will conduct the first two studies using only the phenylalanine stable isotope. If on analysis, we find that we do not see sufficient difference between the baseline and steady state levels of the 1-13CPhenylalanine stable isotope, then we will introduce a one-time dose of 13C-sodium bicarbonate (NaH13CO2) at 14 μmol kg-1 administered by NG/G tube to permit us to see a differentiation between the infant's naturally occurring 1-13CPhenylalanine and the amount we administer with the formula. We do not believe we will need to use the bicarbonate isotope but have added to our protocol for approval should its use become necessary. It is used routinely by a group of scientists in the Netherlands who also study infant amino acid requirements and who have not reported any issues with its administration (Hogewind-Schoonenboom 2014)

The stable isotope tracers are labelled "For Research Use Only" and are acceptable for use in humans of all ages which will be confirmed by the Isotope manufacturing company who will provide a Certificate of Analysis with testing for chemical purity, presence, if any, of heavy metals (in parts per million) and microbiological testing including pyrogens. Both the isotopes will be prepared in the Diet Kitchen of the hospital at the time of preparation of the study formula under the same SOP guidelines as clinical diets.

Sample collection and analysis

Breath samples and urine will be collected to determine changes in the infants' response to the study diet. Three baseline breath and urine samples will be collected prior to the introduction of the study diet. Four plateau breath and urine samples will be collected every hour beginning 12-18 hours after the introduction of the Research grade stable isotope-labeled tracers. Breath samples will be collected using a Carbon Dioxide Analyzer Cart which is similar in design and use as a clinical test device called a Calorimeter. Both devices use a vented hood system under which the infants sleep or rest and VCO2 is measured. Each breath sample will be collected over a 10 minute period by bubbling the breath effluent into a reflux condenser and trapping the breath in a test tube. The breath samples will be measured by a continuous flow isotope ratio mass spectrometer. Urine samples will be collected by placing cotton balls in the infant's diapers and analyzed on a tandem mass spectrometer.

Statistical analysis

The effect of threonine intake on phenylalanine flux, oxidation, and F13CO2 will be tested by using analysis of variance with the PROC GLM procedure (SAS version 9.4). Estimates of the mean threonine intake will be derived by breakpoint analysis of the rate of release of 13CO2 (F13CO2) data with the use of a 2-phase linear regression crossover model. The breakpoint will be calculated using the mixed models and regression procedure of SAS with the slope of the line not being significantly different from zero. Statistical significance will be established at P< 0.05.

Regression analysis variables are threonine as the independent variable and F13CO2 and phenylalanine oxidation as the dependent variables. Selection of the best model will be determined by factors relating to fit (significance of the model and r2) and estimates of variation about the model (CV and SE of the estimate). The population Recommended Dietary Allowance will be estimated by determining the upper 95% confidence limits of the breakpoint estimate.