Presented at the Neonatal Society 2012 Summer Meeting.
Chandler A, Johnson M, Leaf A
University of Southampton, UK
Background: Very preterm infants frequently experience poor growth and it is well documented that protein intake is often below recommended amounts.1 Routine assessment of protein status may allow for better clinical management on neonatal ward rounds. This study aims to assess whether Serum Urea Nitrogen (SUN) and albumin are good reflectors of protein status in preterm infants and whether this can be used to predict growth.
Methods: Nutritional data was taken from a retrospective cohort of preterm infants (<30 weeks gestation at birth or with a birth weight of <1501g) on Southampton Neonatal Unit during 2009. Blood biochemistry data on SUN and albumin for each infant was extracted from the hospital results system and paired with nutrient intake data and weight on a specially designed spreadsheet. Data analysis was carried out using SPSS and relationships were examined between mean protein intake over one week and first blood results and weight gain (g/kg/day) the following week. Correlation and linear regression statistical tests were utilized.
Results: Serum albumin showed a weakly positive correlation with protein intake whilst SUN had a negative relationship with protein intake (Pearson’s correlation coefficient 0.159 and -0.289 respectively, p<0.05 for both). Multiple linear regression using both SUN and albumin correlated with protein intake, R=0.323 (SE=0.815, p≤0.05), accounting for 9.9% of the variation. When immature renal function was taken into account by incorporating creatinine into the regression model the association was stronger at R=0.541 (SE=0.726, p≤0.05) accounting for 28.5% of the variation in protein intake, and the correlation between SUN and protein intake became positive at 0.077 (p≤0.05). Multiple linear regression showed SUN and albumin account for only 0.9% variation in weight gain (R=0.121, SE=18.21, p=0.077). Interestingly, those infants receiving 100% or more of the recommended protein intake (1) gained an average of 10.38 g/kg/day (p≤0.05) more than those infants who did not.
Conclusion: SUN and albumin are related to protein intake in very preterm infants but cannot be used alone as accurate clinical indicators. The impact of creatinine levels on this association suggests that immature renal function has a greater effect than protein intake on SUN levels. In this study the best indicator of weight gain was protein intake, however, this is difficult to calculate accurately in daily clinical practice on infants with complex and changeable dietary regimens. Future priorities would be to simplify cot-side methods for monitoring nutrient intake, or the identification of alternative blood biomarkers of protein status.
Corresponding author: ac1v07@soton.ac.uk
References
1. Tsang C et al.(eds.) Nutrition of the Preterm Infant: Scientific Basis and Practical Guidelines. 2nd ed. Cincinnati, Ohio: Digital Educational Publishing; 2005. p45-80.