Growth reference standards and growth curves

Growth charts based on reference standards are used to assess adequacy of growth in children. In the UK since 2009, the growth reference used for preterm and unwell neonates is the UK-WHO Neonatal and Infant Close Monitoring growth chart, which incorporates longitudinal WHO growth data for children from 2 weeks to 4 years of age, but uses older, cross-sectional birthweight data from the UK 1990 dataset for preterm infants from 23 weeks gestation until 42 weeks. There is debate about the utility of cross-sectional birthweight data to approximate intrauterine growth; the birth weights of preterm infants may not represent ‘normal’ intrauterine growth and intrauterine growth may not be an appropriate postnatal target for preterm infants given the differences between the in-utero and ex-utero environments. A true ‘growth chart’ using longitudinal data from preterm infants followed up and measured over time may be preferable, but such a standard would depict ‘actual’ rather than ‘ideal’ growth, especially in the context of preterm infants receiving suboptimal nutrition. Cole et al7 recently published weight centiles for preterm infants born below 32 weeks gestation based on longitudinal UK data, which demonstrated this, with infants’ mean weight falling at least two marked centile lines in the first 2 weeks of life.

While what constitutes ‘ideal’ growth for preterm infants remains unclear, regardless of the growth standard used, the utility of growth charts is to monitor an infant's growth over time in relation to marked centiles. Although immediately after birth growth may be slow, with infants crossing down 2–3 marked centile lines, from the second or third week of life infants should be growing consistently along, or parallel to, a centile line, as demonstrated by the data of Cole et al.7 Consistent downward crossing of centiles after the third week of life should be considered growth failure, while crossing up centiles may represent catch up growth (providing it occurs proportionally across all parameters). Increases in weight in the absence of gains in length and head circumference may represent excess relative adiposity due to a failure to gain lean mass.8

Recommended nutrient intakes

Recommendations for the nutrient intakes of term infants have been derived based on the assumption that breast milk was the ideal feed for term infants. Using the composition and volume of breast milk taken by term infants growing in accordance with contemporary reference standards, it was possible to calculate reference nutrient intakes (RNIs) (see table 1 and figure 1) for term infants (table 2).9 It is not possible to use the same method to derive preterm infant requirements, as such assumptions cannot be made and ‘ideal’ growth has yet to be adequately defined. Nonetheless, published recommendations for the nutrient intakes of preterm infants are available, with the most widely used being those in ‘Nutrition of the Preterm Infant: Scientific Basis and Practical Guidelines’ originally by Tsang et al,10 and recently updated by Koletzko et al.11 This provides recommendations for infants <1500 g (table 2). Similar recommendations were produced in 2010 by the ESPGHAN for infants <1800 g at birth12 and by the WHO for infants weighing <2500 g (table 2).13 Such recommendations have been derived from fetal composition studies, nutrient balance studies, clinical trials, umbilical cord blood nutrient levels and extrapolated term infant data. Consensus was used in the absence of robust evidence and recommendations for each nutrient given as a ‘reasonable range of intake’ (table 1), rather than RNI, in recognition of this. While there is debate as to the validity of these recommendations, there is evidence that achieving them can improve growth.14

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Estimated average requirement for intake of a nutrient and reference nutrient intakes displayed in relation to the requirement of a given population.

A—anthropometry

Anthropometry, specifically regular measurement of weight, length and head circumference, is one of the mainstays of nutritional assessment. Most infants on the NICU will be weighed at least twice weekly, with head circumferences and length ideally measured weekly. This is essential for the correct interpretation of trends in weight in the context of overall growth. Measuring rods built into ward scales (figure 2A) allow length to be conveniently measured alongside weight, negating the need to use more cumbersome ‘infantometers’ (specific devices for measuring the length of supine infants, figure 2B). Newer ‘incubator measures’ (figure 2C) allow length measurement of preterm infants while in their incubator. It is important to consider both the quantity and quality of growth, and plotting all measures on an appropriate growth chart allows assessment of whether an infant is growing adequately and in proportion. Quality of growth can also be assessed by measuring an infant's body composition (their relative proportions of fat and lean tissue), and there is evidence that the pattern of body composition achieved by preterm infants by term equivalent age is different from that of term born infants.8 There are several methods which are valid and feasible for use in neonates and these are covered in detail elsewhere.15 However, all require specialist equipment and are subject to limitations, with no practical and reliable method of assessing body composition during the NICU stay currently available. There is a need to develop such techniques to help guide nutritional care.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Measuring devices for neonates ((A) scales with built in measuring rod, (B) infantometer, (C) incubator measure).

B—biochemistry

Certain biochemical parameters can provide a useful insight into nutritional status while others are less useful, as detailed below.

• Glucose and lipids: Hyperglycaemia (blood glucose >10 mmol/L16) is common in preterm infants after birth and can necessitate a reduction in the glucose content of parenteral nutrition (PN) or the commencement of insulin. Similarly, serum triglyceride levels measured during infusion will indicate tolerance of intravenous lipid, with a level >2.8 mmol/L suggesting a need to reduce PN lipid content.17 Both hyperglycaemia and hyperlipidaemia are important; subsequent reduction in the carbohydrate and lipid content of PN will limit the ability to provide adequate protein, as an energy to protein ratio of around 20–25 non-protein kcal per gram of protein11 should be maintained in order to provide enough energy to enable protein to be used for tissue accretion and growth. Inadequate energy to protein ratio or increased catabolism (e.g. sepsis or surgery) preventing growth results in unused excess nutrients that must be excreted, adding to renal solute load and increasing metabolic demand. It is therefore important to ensure adequate energy to protein ratios and consider avoiding excessive nutrient intakes during periods of catabolism.

• Markers of protein status: Serum total protein levels are not related to intake, as the levels of many proteins, particularly acute phase proteins such as C reactive protein and ferritin, are independent of nutritional status. Albumin has a long half-life (around 20 days) and so cannot be used to monitor protein status in the short term. While serum urea cannot accurately predict protein intake, levels of <1.6 mmol/L suggest a protein intake of <3 g/kg/day.18 Protein intakes exceeding an infant's metabolic capacity (likely >5 g/kg/day) could potentially lead to high plasma amino acids levels (including tyrosine and phenylalanine, which may affect mental development).19

• Electrolytes: Serum sodium and potassium are poor indicators of total body stores due to homeostatic mechanisms acting to maintain normal serum levels despite changes in supply and demand. However, excessively low or high levels can be useful in determining the requirement for these electrolytes in feeds and fluids. Preterm infants have limited ability to concentrate their urine, and so high urinary sodium and potassium levels (>30 mmol/l) can be normal. In this context, a urinary sodium <10 mmol/L may indicate insufficient intake, especially in the presence of relevant pathology such as high ileostomy losses.10

• Markers of bone status: Like sodium and potassium, serum calcium is a poor marker of body stores. Serum phosphate does fluctuate in response to intake, with low levels indicating the need for supplementation. Chronic phosphate deficiency increases the risk of metabolic bone disease by upregulating 1,25(OH)₂ vitamin D production, increasing osteoclast activity and liberating calcium from bone, raising alkaline phosphatase (ALP) levels. As discussed in a previous ‘Interpretation’ by Tinnion and Embleton, high (>900 IU/L) ALP levels combined with a phosphate <1.8 mmol/L can indicate metabolic bone disease.20

• Vitamin and trace element status: Serum levels of trace elements (zinc, copper and selenium) and fat-soluble vitamins (A, D and E) may help determine insufficiency where there is poor growth despite adequate macronutrient intake, relevant clinical signs (such as poor wound healing in zinc deficiency) or in infants receiving long term (>28 days) PN.

C—clinical assessment

As most preterm infants appear thin with low muscle bulk and scarce subcutaneous fat, differentiating between babies from a nutritional perspective clinically is not possible. Clinical assessment should be guided by information from anthropometry and biochemistry. Assessment of hydration status using capillary refill, urine output and examination for oedema is useful in determining the volume of fluid available for nutrition, and therefore vital when working out how to optimise intake. Oedema (and its subsequent resolution) will impact on weight and so should be taken into account when assessing growth. Concurrent illnesses that might affect nutritional needs should be considered, such as chronic lung disease, which may increase energy requirements. Nutrient absorption and losses (such as stoma output or nasogastric aspirates) as well as nutrient tolerance (such as hyperlipidaemia or hyperglycaemia) should also be considered here.

D—dietary assessment

This element of nutritional assessment is the most challenging but is essential in understanding the extent of shortfalls in nutrient delivery, providing an opportunity to modify and improve nutritional intake. The key here is to move away from considering the feeds and fluids an infant is receiving in terms of their volume, and instead consider their nutritional value. The nutrients to focus on are energy and protein, although consideration needs to be given to the adequacy of micronutrients, particularly those known to be an issue for preterm infants such as sodium, phosphate and fat soluble vitamins (A, D and E). For proprietary ‘stock’ PN solutions, formula milks and fortifier, manufacturers’ datasheets provide the amounts of energy (in kcal) and protein (in grams) in a given volume of fluid (see table 3 for the nutritional content of common feeds). While breast milk will vary widely between individuals, there are published reference values which can enable an estimate of its nutrient content. It is also worth considering at this stage the availability of breast milk, breastfeeding support and alternative feed options where breast milk is unavailable. While alternative formula feeds may provide higher nutrient content, this is far outweighed by the overall nutritional benefits of breast milk, which should be used wherever possible. For pharmacy manufactured individualised PN, pharmacy support is essential, although such PN is provided with a printout providing information on the energy content (usually in kcal/kg) and nitrogen content (usually in grams of nitrogen per kilogram: this can be converted to protein multiplying by a factor of 6.25). It is also worth noting that carbohydrates (including glucose) and protein both provide energy at approximately 4 kcal/g, while fat provides energy at approximately 10 kcal/g.

E—evaluation

This final stage in nutritional assessment is to use the above information to make a judgement on an infant's nutritional status. The Academy of Royal Medical Colleges Intercollegiate Course on Human Nutrition uses the framework of ‘what they are’ (their growth, body size and composition), ‘what they eat’ (their dietary intake) and ‘what they can do’ (their functional activity, including physical activity, metabolic capacity, biochemistry and the impact of disease) to consider an individual's nutritional status, which is a helpful approach.21 It is important to consider adequacy of growth and nutrient intakes, and then make a management plan to address any issues identified.

Topics for future research

It is still not clear what constitutes optimal growth for preterm infants, and an appropriate growth standard needs to be determined for this group. Such a standard would need to be based on the patterns of growth which lead to optimal neurodevelopmental and metabolic outcomes in later life. As such, there is also a need to establish improved methods of assessing growth on the NICU, particularly with regard to body composition. As discussed above, recommendations for nutrient intakes for preterm infants are also an area where more research is needed, and better understanding of optimal growth in preterm infants will allow more robust recommendations for their optimum intake of key nutrients at critical periods during early life.