Skip Navigation LinksHome > April 2014 - Volume 58 - Issue 4 > Higher Protein Intake Improves Length, Not Weight, z Scores...
Journal of Pediatric Gastroenterology & Nutrition:
doi: 10.1097/MPG.0000000000000237
Original Articles: Hepatology and Nutrition

Higher Protein Intake Improves Length, Not Weight, z Scores in Preterm Infants

Olsen, Irene E.*; Harris, Cheryl L.; Lawson, M. Louise; Berseth, Carol L.

Free Access
Article Outline
Collapse Box

Author Information

*School of Nursing, University of Pennsylvania, Philadelphia, PA

Department of Medical Affairs, Mead Johnson Nutrition, Evansville, IN

Department of Mathematics and Statistics, College of Science and Mathematics, Kennesaw State University, Kennesaw, GA.

Address correspondence and reprint requests to Carol L. Berseth, MD, Department of Medical Affairs, Mead Johnson Nutrition, 2400 W Lloyd Expressway, Evansville, IN 47721 (e-mail: ieolsen@yahoo.com).

Received 20 June, 2013

Accepted 5 November, 2013

This study was funded by the study sponsor, Mead Johnson Nutrition (MJN; Evansville, IN). C.L.H. and C.L.B. are employees of MJN. MJN provided a consulting fee to I.E.O. to interpret data and prepare the article. The other authors report no conflicts of interest.

Collapse Box

Abstract

Objective: The aim of the study was to evaluate the relation between nutritional intake (kilocalories, protein) and weight and length growth in preterm infants, and to describe their metabolic tolerance with a focus on those with high protein intake (≥4.6 g · kg−1 · day−1).

Methods: Secondary analysis of data from appropriate-for-gestational age preterm infants in a 28-day randomized clinical trial that evaluated growth, tolerance, and safety of a new ultraconcentrated liquid human milk fortifier (original study n = 150). This subset of 56 infants had complete growth and nutrition data and met criteria for the original study's “efficacy analysis” (eg, >80% of kilocalorie intake from study diet). Nutritional intake was estimated, not actual. Regressions were used to test cumulative kilocalories and protein as the predictors of 28-day change in weight and length z scores (growth status), and to evaluate protein tolerance.

Results: Average intake was 118 ± 8 kcal · kg−1 · day−1 and 4.3 ± 0.4 g protein · kg−1 · day−1, with 16 ± 3 g · kg−1 · day−1 and 1.1 ± 0.2 cm/week growth for 28 days. Cumulative total kilocalories and protein were significant predictors of improved length z score (P = 0.0054, 0.0005) but not weight z score change. Regression models indicated that protein not kilocalories explained the improvement in length z score, with protein explaining 19% of the variability. The high protein group averaged 4.6 to 5.5 g · kg−1 · day−1 (n = 16). Protein tolerance was adequate for all of the study infants based on metabolic measures (blood urea nitrogen, serum carbon dioxide, pH).

Conclusions: Higher cumulative protein intake was tolerated and overall lessened the commonly occurring decline in the length but not weight growth status in a 28-day study of preterm infants.

The relation between inadequate nutritional intake and poor growth is well documented (1–10). Newer aggressive nutrition practices, including earlier and higher protein intakes, have been successful in achieving growth at intrauterine rates or more in preterm infants, especially for weight (11–13). These advancements are reflected in the lower rates of extrauterine growth failure [usually defined as weight-for-gestational age below the 10th percentile (14)] in the neonatal intensive care unit (NICU) (8,13,15) than in many earlier studies (2,14,16–19). Nevertheless, the prevalence of inadequate nutrition and extrauterine growth failure in the NICU remains unacceptably high especially given the association with impaired neurocognitive development (10,20–25).

Should the growth assessment of preterm infants focus on weight? There is a growing concern over the focus on weight in preterm infants (26,27), given the higher rates of postnatal fat accretion (28,29) with unknown long-term sequelae and recent evidence of a relation between poor linear growth and poor cognitive outcomes (30). The problem is that there is little research available on the relation between length growth and health outcomes in preterm infants. Often research on preterm infant growth does not address linear growth (1–5,7,8,13,15,28,29), sometimes owing to the lack of confidence in its accuracy (15), or length growth is improved but is not statistically significantly better in the infants who were fed more aggressively (11,26,31–34). This lack of improved length growth may be the result of suboptimal study diets (26,32,33) that result in the linear growth that fails to keep up with weight growth, or this could be because it is difficult to detect differences in the length growth among study groups owing to small sample sizes (12,14,26,31–33) and wide variability or error in the data collected in the clinical setting (15,35). Regardless of the reason, growing evidence supports the need for routine measurement and assessment of length of preterm infants in the clinical setting (30).

The research findings indicate that protein, not kilocalories, is the rate-limiting nutrient to weight growth in preterm infants (1,3,32,36), but the optimal nutrition to support length growth is unknown. Further, the ideal protein intake for preterm infants is a topic of ongoing research. Several studies have tested protein intakes at or slightly above the present recommendations of 3.5 to 4.5 g · kg−1 · day−1(37) and found promising results of improved weight growth and adequate gastrointestinal and metabolic tolerance: Cooke et al (5) compared an average intake of 3.8 to 4.6 g · kg−1 · day−1 protein, Fanaro et al (11) compared an average intake of 3.9 to 4.6 g · kg−1 · day−1 protein, and Miller et al (5,11,26) compared a median intake of 3.6 to 4.2 g · kg−1 · day−1 protein. In contrast, our original study (38) of infants receiving predominantly fortified human milk with more protein (compared an average protein intake of 3.9 versus 4.3 g · kg−1 · day−1) showed improved growth, for weight and also for length. The purpose of the present study was to further evaluate the relation between nutritional intake (kilocalories and protein) and weight and length growth in a subset of infants from our original 28-day study and describe their metabolic tolerance, with a focus on those who had the highest levels of protein intake (≥ 4.6 g · kg−1 · day−1).

Back to Top | Article Outline

METHODS

This study was a secondary analysis of data collected in a multicenter, third party–blinded, randomized, controlled, 28-day prospective study of premature infants (38). The primary objective of the original study was to evaluate the growth, tolerance, and safety of using a new ultraconcentrated liquid human milk fortifier (HMF) to fortify human milk compared with using a powder HMF (control group) in preterm infants. The new ultraconcentrated liquid HMF was designed with 20% more protein in a sterile liquid that minimized dilution of the breast milk when mixed. Docosahexaenoic and arachidonic acids were also added to the liquid HMF and some vitamin and mineral levels were modified (38). The study results included improved infant growth on the new liquid HMF compared with the powder HMF and no clinically significant differences in safety and tolerance.

Premature infants were eligible for the original study if they were ≤1250 g and ≤ 30 3/7 weeks at birth (38), appropriate weight-for-gestational age (based on the Lubchenco curves (39)) at birth, exclusively fed human milk (mother's milk or donor milk), and on ≥ 80 mL · kg−1 · day−1 of unfortified human milk at the time of enrollment. The exclusion criteria were described in detail in the original study by Moya et al (38); briefly, study infants were healthy, not fluid restricted to <120 mL · kg−1 · day−1, and tolerating unfortified human milk. Of the 150 infants enrolled (Fig. 1 “Flow of study participants” (38)) and randomized (75 infants per HMF group), 4 infants did not consume any HMF (thus, eliminated from analysis) and 40 infants were discontinued—8 for gastrointestinal intolerance and the rest for a variety of nonstudy related reasons (eg, breast milk not available for >96 consecutive hours, hospital transfer/discharge before the study day 28, death resulting from cause not related to HMF according to site investigators). The dropout rates were not statistically different between HMF study groups (24% control group vs 31% liquid HMF group, P = 0.3561). Although 106 infants completed the 28-day study, 109 were included in the original study's primary growth analysis on study day 28 because data analysis allowed for ± 3 days around the designated study day. The study was approved by institutional review boards, monitored by a data monitoring board, and a parent or guardian consented to each infant's participation.

Figure 1
Figure 1
Image Tools
Back to Top | Article Outline
Sample Selection

The present study included the 56 infants who met the more strict criteria defined for the original study's efficacy analysis on study day 28 (38). Specifically, these infants had complete kilocalorie and protein intake data and weight and length measurements (per protocol) for the 28-day study to evaluate the effect of cumulative nutritional intake on the change in size during the study period. In addition, the infants were fed ≥ 80% of their energy intake from human milk fortified with the study fortifier for at least the first 2 weeks of the study; feedings were held (ie, nil per os) for ≤ 2 days during this period; infants did not receive glucocorticoids and were free of any disease known to negatively affect ingestion of food or growth (eg, inborn metabolic errors, congenital malformation). From the 109 study infants (58 controls, 51 liquid HMF) in the primary analysis, 53 infants were excluded based on these criteria (26 controls and 27 liquid HMF, P = 0.4455), resulting in our subset of 56 infants.

Back to Top | Article Outline
Nutritional Intake

All of the parenteral and enteral nutrition intake data were collected from the nursing flow sheets. Source and volume of nutrition (eg, 100 mL human milk) consumed and amounts of supplements ordered were obtained, converted into nutrient data (based on the published data and manufacturer's product information), and summed for cumulative total kilocalorie and protein intake during the study period. The kilocalorie and protein intakes from human milk were estimated based on postpartum week 3 preterm human milk (66.4 kcal/dL and 1.62 g/dL, respectively) (40). Thus, the kilocalorie and protein intake data used in this study were estimated (ie, calculated based on the published nutrient intake data in the literature and from manufacturer's product information), not actual intake (ie, based on analysis of individual diets). The estimated (also called assumed) intake has been shown to overestimate the actual protein intake by Arslanoglu et al (36) owing to the variability of human milk protein content. Each study day for this 28-day study was based on a 24-hour interval (eg, study day 1 beginning at 7 AM and ending the following day at 6:59 AM). Because growth measurements and blood samples could be collected at any time during study day 28 (eg, study day 28 at 9 AM), cumulative intake was based on the study day 1 through 27.

Back to Top | Article Outline
Growth Measurements

Weights were measured daily without clothing or diaper using the same electronically calibrated scale throughout the study and recorded to the nearest gram. Lengths were measured weekly by 2 trained individuals using standardized technique on a preterm infant length board, and recorded to the nearest 0.5 cm (38).

Back to Top | Article Outline
Outcomes
Growth

Growth velocity was calculated via linear regression assuming an exponential model. For analyses, weight and length growth during the study period were expressed as changes in the z scores (eg, study day 28 weight z score minus study day 1 weight z score). The z score changes were used to account for initial size, postmenstrual age, and sex of each infant. The z scores were calculated based on the Lambda Mu Sigma method (41) using data from a published set of intrauterine growth curves by Olsen et al (42). The following formula was used: z = [(X/M)L – 1]/LS, where X is the measured value of weight (in kilograms) or length (in centimeters), M is the median, L is the Box-Cox power transformation of skewness, and S is the coefficient of variation.

Throughout this article, the terms growth status and z score and percentile are used to describe the size of an infant compared with other infants of the same age and sex. A z score of 0 is equivalent to the 50th percentile on a growth curve, for example; negative z scores are less than the 50th percentile and positive z scores are greater than the 50th percentile. A change in z score can be negative, positive, or zero. A negative z score change is a decline in growth status (eg, a decline from a z score of −1.0 to −1.5), a positive z score change is an improvement in growth status (eg, an increase from a z score of −1.0 to −0.5), or a z score change of zero is a stable or unchanged growth status.

Back to Top | Article Outline
Protein Tolerance and Safety

Blood urea nitrogen (BUN) and serum carbon dioxide (CO2) values collected on study day 28 and blood gas pH collected on study day 14 were used to evaluate protein tolerance and safety for the 56 infants in this study, which included the subset of infants with high protein intake. Of note, a more comprehensive assessment of protein tolerance and safety for all infants who received HMF in our original study (ie, an intent-to-treat analysis), are presented by Moya et al in Table 4 (38) and have been redone using the efficacy analysis sample (unpublished data); both sets of analyses concluded adequate metabolic tolerance of protein provided. Per protocol (38), blood gases were not drawn past study day 14 to minimize infant blood loss. Laboratory data were compared with average protein intake (in grams per kilogram per day) based on twenty-seven 24-hour intervals of nutritional intake (as described above). High protein intake was defined as a mean intake of ≥ 4.6 g · kg−1 · day−1 of protein because this was above the most recent set of recommendations of 3.5 to 4.5 g · kg−1 · day−1(37).

Back to Top | Article Outline
Statistical Methods

One- and 2-variable regression models were used to test cumulative intake (total kilocalories and protein) as the predictors of 28-day change in z score (weight and length). The slope of the regression line and the adjusted coefficient of determination (R2) for each model were used to evaluate the ability of nutritional intake to predict z score change. An adjusted R2 value of >10% was considered to be significant for a biological measure. Because the sample for this study included 2 fortifier groups (ie, liquid HMF group and powdered HMF group) from the original study, the regression models were also run with the HMF group in the model. These led to similar conclusions regarding the relation between nutritional intake and growth and are not discussed further in this article.

Model fit was examined initially using scatterplots, which showed some lack of fit with the linear model. In particular, the 6 infants with the lowest protein intake fell below the length regression line, although all were within 0.3 standard deviations (SDs) of the line. Outliers and/or influential observations (based on leverage and residuals) were removed from the models individually to test their effect on the overall conclusions, which were minimal.

One-variable regression models and laboratory values (study day 28 BUN and CO2, study day 14 pH) were used to evaluate protein tolerance and safety. In addition, summary statistics were used to describe protein intake and tolerance for infants with the highest protein intake (n = 16) and included average total protein intake for the study period, maximum daily protein intake, and timing and duration of high protein intake. Postmenstrual age and postnatal age were also summarized as protein tolerance and laboratory values may vary with age (27,43–45).

Back to Top | Article Outline

RESULTS

Study Population

There were 56 infants (46% boys) who met the entry criteria for this secondary analysis. Table 1 summarizes infant characteristics at the beginning (study day 1) and end (study day 28) of the study (38). On average, the infants consumed 118 ± 8 kcal · kg−1 · day−1 and 4.3 ± 0.4 g protein · kg−1 · day−1 (estimated intake) and grew 16 ± 3 g · kg−1 · day−1 in weight and 1.1 ± 0.2 cm/week in length during the 28-day study period.

Table 1
Table 1
Image Tools

The percentage of infants classified as <10th percentile (or small-for-gestational age [SGA]) based on the Olsen weight-for-age curves (42), which were used to calculate age and sex-specific z scores, was 38% at study entry and 54% on the study day 28. This was consistent with the decreases in weight and length z scores during the 28-day study period in our sample overall (Fig. 1).

Back to Top | Article Outline
Growth Outcomes
Weight
Back to Top | Article Outline
28-Day Change in Weight z Scores

The weight z scores of the group decreased from −1.03 ± 0.58 to −1.25 ± 0.59 (mean ± SD) during the study period, which is equivalent to moving from the 15th to the 11th percentile (Fig. 1). These results indicate a small decline in the growth status or weight growth that was slightly less than the rate needed to track along the 15th percentile. The number of infants classified as SGA on study day 1 increased by the study's end (from 38% to 54%, as above).

Back to Top | Article Outline
Effect of Nutritional Intake on 28-Day Change in Weight z Scores

As seen in Figure 2A and B, our regression models showed that the cumulative total kilocalorie and protein intake were not significant predictors of the change in weight growth status (P = 0.1531 and P = 0.0853). Overall, increasing the cumulative kilocalorie or protein intake did not lessen the decline in weight z score during the 28-day study. The total kilocalorie and protein 1-variable linear regression models accounted for only 2% and 4%, respectively, of the variability in the change in weight z score data.

Figure 2
Figure 2
Image Tools
Back to Top | Article Outline
Length
Back to Top | Article Outline
28-Day Change in Length z Scores

The length z scores of the group decreased from −0.78 ± 0.57 to −1.21 ± 0.52 (mean ± SD) during the study period, which is equivalent to moving from the 22nd to the 11th percentile (Fig. 1). These results indicate a decline in length growth status or linear growth that was less than the rate needed to track along the 22nd percentile. The number of infants classified as <10th percentile length-for-age on the study day 1 increased by the study end (from 21% to 39%).

Back to Top | Article Outline
Effect of Nutritional Intake on 28-Day Change in Length z Scores

Cumulative total kilocalorie and protein intake were significant predictors of change in length z score (P = 0.0054 and P = 0.0005, Fig. 3A and B). Overall, the decline in length growth status lessened as cumulative kilocalorie or protein intake increased. For example, the protein model predicted that feeding an infant an additional 100 g of protein during the study period would lead to a 0.65 increase in length z score change. In most infants, higher cumulative intakes during the 28-day study led to smaller declines in length z score (or length percentile) than lower intakes. For some infants, higher kilocalorie or protein intakes led to a stable length growth status during the 28-day study (ie, z score change ≈ 0) and even led to improved length growth status for a few infants (z score change > 0). The total kilocalorie and protein 1-variable linear regression models accounted for 12% and 19%, respectively, of the variability in the change in length z score data.

Figure 3
Figure 3
Image Tools

When cumulative total kilocalorie and protein intake were included in a regression model together to determine their joint effect on linear growth status, the effect of nonprotein calorie intake became essentially zero (eg, a predicted change in the 28-day length z score change of −0.03 with an additional 400 kcal, P = 0.6774) but protein intake showed a stronger predictive value (eg, a predicted improvement in the 28-day length z score change of 0.77 with an additional 100 g protein, P = 0.0322). Using a conversion of 4 kcal/g of protein, 100 g protein is equivalent to 400 kcal of protein. These results demonstrate that protein rather than nonprotein calories may be primarily promoting length growth.

Back to Top | Article Outline
Protein Tolerance and Safety
BUN: Study Day 28

Overall, there was a statistically significant and positive relation between average protein intake (grams per kilogram per day) and BUN on day 28 (P = 0.0047). The regression model predicted that adding 1 g · kg−1 · day−1 protein to the mean intake would increase BUN by approximately 3.5 mg/dL. BUN values ranged from 1 to 19 mg/dL. Twenty-seven percent (n = 15) were low at <5 mg/dL despite a range of average protein intake of 3.5 to 5.5 g · kg−1 · day−1. In fact, the infant with the highest average protein intake of 5.5 g · kg−1 · day−1 had a BUN of <5 mg/dL. Of note, because protein intake was estimated not actual, some of these may be examples of overestimated protein intake and emphasize the importance of closely monitoring tolerance to adjust diets accordingly.

Back to Top | Article Outline
Serum CO2: Study Day 28

Although individual serum CO2 values trended down as average protein intake rose overall, the relation was not statistically significant (P = 0.0699); this was likely as a result of the variability in these data. Serum CO2 values ranged from 20 to 34 mEq/L over the range of protein intake. The infant with the highest average protein intake of 5.5 g · kg−1 · day−1 had a normal CO2 of 25 mEq/L, and no infants were treated for acidosis during the study.

Back to Top | Article Outline
Arterial or Venous pH: Study Day 14

Overall, there was a statistically significant negative relation between average protein intake and pH (P = 0.0138). The model predicted that adding 1 g · kg−1 · day−1 protein to the mean intake would decrease pH by approximately 0.04 units. The pH values ranged from 7.28 to 7.49, except for 1 infant with a pH of 7.6 on a lower average protein intake of 3.7 g · kg−1 · day−1. The infant with the lowest pH (7.28) had an intake of 7.6 g · kg−1 · day−1 protein on that day (a 1-day peak in protein intake) and had the study's highest average protein intake of 5.5 g · kg−1 · day−1. (Of note, this infant's BUN was 10 on that day, which was not consistent with a high protein intake and low pH.) No infants were treated for acidosis during the study.

Back to Top | Article Outline
Infants With the Highest Average Protein Intake

There were 16 infants who had an average protein intake for the study period of ≥ 4.6 g · kg−1 · day−1 (range 4.6–5.5 g · kg−1 · day−1). These 16 infants were similar in size and age to the whole sample (n = 56) on study day 1 and day 28 (Table 1). The highest daily protein intake was 6.0 ± 0.9 g · kg−1 · day−1 (mean ± SD, range 5.1–7.8 g · kg−1 · day−1) on average for these 16 infants, and occurred on or before study day 14 for half of the sample. On the day of highest protein intake, all of the infants were at least 30 weeks’ postmenstrual age and median postnatal age of 31 days (interquartile range 23.5–38.5). The majority (n = 13) achieved a protein intake of ≥ 4.6 g · kg−1 · day−1 for approximately two-thirds of the study period (ie, ≥ 18 days).

Most of the higher protein intakes were achieved with increased volumes of human milk fortified with a study HMF to provide 24 kcal/oz. Starting after study day 8, 6 of the 16 infants received human milk fortified to 26 or 27 kcal/oz using study HMF; starting on day 18, 1 infant received a combination of study HMF and postdischarge formula. Four of the 16 infants received donor milk. Three of the 16 infants with high protein intake were fed human milk fortified with the lower protein HMF (or in the control group of the original study). Modular protein supplementation was not used in this subset of study infants.

Despite the higher protein intake in this group, the laboratory values in Table 2 indicated adequate and similar protein tolerance to that of the study sample as a whole (n = 56). Specifically, BUN values on study day 28 were between 6 and 19 mg/dL, except for 3 infants, who had values <5 mg/dL despite mean protein intakes of 4.9 to 5.5 g · kg−1 · day−1 (which suggests possible overestimates of protein intake on these infants as discussed above). All CO2 values were between 20 and 30 mEq/L on study day 28. For the 14 infants with pH data on day 14, values ranged between 7.28 and 7.49; the average protein intake for the week before this pH measurement (ie, study days 7 to 13) was 4.8 ± 0.3 g · kg−1 · day−1 (mean ± SD, range 4.4–5.5 g · kg−1 · day−1).

Table 2
Table 2
Image Tools
Back to Top | Article Outline

DISCUSSION

When the growth assessment of preterm infants is focused on weight alone, clinically important changes in the growth status may fail to be detected, according to the results of this study. Despite achieving more than 15 g · kg−1 · day−1 in weight growth and 1 cm/week in length growth, on average, infant growth status decreased during the 28-day study as indicated by declines in weight and length z scores based on new standards (42). The decline in the length growth status was slightly larger and the rate was twice as fast as for weight—a common clinical observation that is not well documented in the literature; however, higher cumulative nutritional intake, in particular of protein, lessened the decline in length growth status but did not affect the change in weight growth status during the 28-day period for the study group overall. The range of protein intake used in this study was tolerated by all of the study infants, including those who had the highest protein intake—consistent with results from the more comprehensive evaluation of protein tolerance in the original study (38).

Historically, weight has been the primary focus of growth assessment for preterm infants in the clinical setting and as an outcome in nutrition research. Many conclusions drawn in the literature about growth are based on weight growth alone (1–3,5,7,8). The postnatal decline in the weight growth status in this study was consistent with previous studies (2,8,13,18,26,30,31,34). Although linear growth is recommended as part of preterm infant growth assessment along with the weight and head circumference, often it is not routinely measured and monitored in the clinical or research settings. In the studies that present NICU weight and length data (13,18,26,30,31,34), our results of a larger decline in length than weight growth status were consistent with a few (13,18,26,30) but not all studies (31,34).

The most notable findings to date that link linear growth to an important health outcome are from Ramel et al (30); these researchers studied the relation between growth and 2-year neurodevelopment in preterm infants and found a significant association between poor linear growth and lower neurocognitive scores. Although the relation of neurodevelopmental outcomes with weight and head growth are well documented (10,23,25), the study by Ramel et al (30) was the first to show a link between this potentially devastating health outcome with poor linear growth in preterm infants.

Several recent studies have examined the positive effect of aggressive nutrition practices on growth (5,12,13,31,34), of those that included length, our recent clinical trial (the original study described in this study (38)) was the only study to find significantly greater linear growth in the higher protein group along with the higher achieved weight. In a subanalysis of only the infants included in the present study, the higher protein group achieved increased size in weight, length, and head circumference by study end (38).

The findings of this secondary analysis demonstrate that higher cumulative nutritional intake, in particular of protein, during the 28-day study had a positive and significant effect on length but not weight growth status. This translated into a range of smaller negative changes to small positive changes in length z scores during the study period, which lessened the decline in length growth status during the 28 days overall; however, the change in weight z scores was not affected by increasing nutritional intake within the ranges provided in this study. This may be a favorable clinical situation for some infants in the NICU. Given the higher rates of body fat accretion associated with rapid weight gain and with higher caloric intakes in preterm infants (33,46) and their unknown long-term health effect, the avoidance of excessive catch-up weight growth may be beneficial, as others have suggested (47). Nevertheless, more research is needed to better understand the balance between adequate and excessive caloric intake in preterm infants and its effect on the quantity and quality of weight growth.

Our results were consistent with a randomized clinical trial conducted by Miller et al (26) that tested 2 levels of protein intake (medians of 3.6 vs 4.2 g · kg−1 · day−1) using length gain as the primary outcome. Similar to our results, length z score declined less in the high protein group (P = 0.087), and fewer infants were <10th percentile length-for-age (P = 0.047) by the end of study. Although the study found that the higher protein group had a statistically higher achieved weight by study end, no significant change in weight growth status was reported based on percentage of infants <10th percentile weight-for-age. Other studies of more aggressive nutrition on growth (that included weight and length) documented significant improvements in the weight but not linear growth status (using z scores) in the higher protein groups (31,34).

The explanations for the variation in results between ours and other studies require further research. Interstudy variation in the level of protein intake tested, choice of primary outcomes, and/or sample size likely played a role in the differences. Roggero et al (34) used z scores and had larger samples (n = 102 treatment, n = 69 controls), but the mean protein intake by group was 3.5 versus 2.2 g · kg−1 · day−1, which may not have been enough to promote significantly faster growth in length. Costa-Orvay et al (31) used z scores and protein intake up to 4.7 g · kg−1 · day−1, but samples were small (8–12 in each treatment group). The clinical trial conducted by Miller et al (26) had a smaller sample size and a moderately high protein intake (4.2 g · kg−1 · day−1), which the authors commented may not have been high enough to have a positive effect on length that was statistically significant.

Infant growth in this study achieved more than 15 g · kg−1 · day−1 in weight and 1 cm/week in length, which approximates the widely used intrauterine growth goals for preterm infants (48–50); however, the weight and length z scores of infants in our study declined during the 28-day period. This suggests that higher rates of growth are required to track along percentiles of the new intrauterine growth curves (42), and the growth goals for preterm infants in the NICU may need to be updated.

The optimal protein intake for preterm infants and the ideal markers and reference values to assess protein tolerance remain unknown. Clinicians often use measures of uremia and metabolic acidosis to guide their assessment of protein tolerance as best they can and adjust diets accordingly. Protein recommendations for preterm infants have increased during the years to help overcome the substantial protein deficits accumulated during early postnatal life (37). Several recent studies of preterm infants have tested higher protein intakes using fortified human milk (26,38) and/or formula products (5,11,31) and demonstrated improved growth and adequate metabolic tolerance. Consistent with these results, infants in the present study tolerated a range of protein intake from 3.5 to 5.5 g · kg−1 · day−1 (mean of 4.3 g · kg−1 · day−1) based on BUN, CO2, and pH values that were within clinically acceptable limits for preterm infants. In part, this could be the result of the older postmenstrual and postnatal age of our participants; in particular, serum BUN decreases with age (43,44) and has been shown to provide a reasonable assessment of protein tolerance (12,27,45). The long-term sequelae of higher protein intake in preterm infants are unknown and also should be tested in future research.

Although the relatively small size of our sample (n = 56), in particular for the high protein group (n = 16), warrants caution when interpreting our results, our data have shown consistently adequate tolerance of protein intake in all of the samples of infants tested; our original study's primary analysis sample (Table 4) (38), the efficacy analysis sample—which includes our study's 56 infants (Table 4 analyses redone, unpublished data) and our high protein group. We also found similar and adequate metabolic tolerance for the 53 infants excluded from the efficacy analysis (unpublished data). Further, there were no statistically significant differences between the number of lower protein HMF controls and higher protein HMF infants in our samples. Combined with the metabolic tolerance data, these data help to rule out the presence of selection bias in this sample.

This study had several other potential limitations. Comparable to most nutritional intake data collected in NICUs for clinical and/or research purposes, our measures of protein and kilocalorie intake were based on published data (estimated intakes) rather than analysis of human milk samples (actual intakes), which may have made our intake estimates less precise; for example, Arslanoglu et al found that the estimated intake of protein was 0.5 to 0.8 g · kg−1 · day−1 higher than the actual intake (for which human milk samples were analyzed) in a study of fortified human milk–fed preterm infants (36). Because not knowing the actual protein intake in the NICU is common, careful and frequent monitoring of protein tolerance remains essential to make informed decisions about the diet. Despite the error that may have been added to our data by using estimated protein intake, we found a significant relation between protein and linear growth status that we suspect would only be stronger with more precise intake data. Also related to these data, we found that the fit of the regression models was imperfect. Several potentially influential observations (leverage greater than 0.10 or df β > 0.26 from single models) were removed from the models, which resulted in a slight change in adjusted R2 and no change in the overall conclusions or the plotted regression lines. There were other nutrient differences between the high protein and lower protein HMF, such as long-chain polyunsaturated fatty acids; however, the regression models run with the HMF group in the model led to similar conclusions regarding the relation between nutritional intake and growth. Finally, the growth assessment of the preterm infants at a minimum should include the evaluation of weight, length, and head circumference measurements; however, this study focused on weight and length.

Back to Top | Article Outline

CONCLUSIONS

This study provides evidence of the positive effect of protein on length growth; higher cumulative protein intake during our 28-day study was tolerated and improved length growth substantially enough to lessen the seemingly inevitable decline in length growth status seen in most preterm infants. Further, this improved change in growth status in length occurred without a concomitant effect on weight growth status. Given the potential adverse health consequences of the early rapid growth and fat accretion in these small infants, as well as the new evidence of a relation between poor linear growth and poor cognitive outcomes, the picture of growth illustrated by this study may be favorable for some infants in the NICU. More research into these findings is needed; however, it seems clear at this time that the length must be routinely included in the assessment and monitoring of preterm infant growth.

Back to Top | Article Outline
Prolapsus Ani Redux

Why shou’d I name how the Posterior Pipe

Is apt the Bounds in weakly Babes slip?

The Muscles, moistn’d when the Belly's loose,

Their nat’ral Duty to discharge, refuse;

And out the Anus hangs, a grievous Pain;

Nor is it easily got in again.

The Body bind, foment it when ’tis out,

And gently with thy Hand replace the Gut.

A French lawyer, Scaevola de Sainte Marthe (1536–1623), wrote a pediatric book in Latin hexameter—Paedotrophia—consisting of 3 books totaling 1726 lines dedicated to King Henry III. His son became ill, and frustrated by physicians’ failure to cure the boy, de Sainte Marthe consulted texts on the healing arts in order to help his child. He became quite knowledgeable, and, in his book, he covered a broad range of subjects that included prenatal care, labor, postnatal care, feeding, and weaning, and, in the final section, diseases of children. Although there was nothing original in its contents and the Latin was ponderous, the Paedotrophia became a popular textbook. Between 1584 and 1742, 20 editions of the work were published. Medical poems were popular for some 5 centuries because much learning was by rote and the memorization of a poem more facile.

—Contributed by Angel R. Colón, MD

Back to Top | Article Outline

REFERENCES

1. Carlson SJ, Ziegler EE. Nutrient intakes and growth of very low birth weight infants. J Perinatol 1998; 18:252–258.

2. Embleton NE, Pang N, Cooke RJ. Postnatal malnutrition and growth retardation: an inevitable consequence of current recommendations in preterm infants? Pediatrics 2001; 107:270–273.

3. Olsen IE, Richardson DK, Schmid CH, et al. Intersite differences in weight growth velocity of extremely premature infants. Pediatrics 2002; 110:1125–1132.

4. Bloom BT, Mulligan J, Arnold C, et al. Improving growth of very low birth weight infants in the first 28 days. Pediatrics 2003; 112:8–14.

5. Cooke R, Embleton N, Rigo J, et al. High protein pre-term infant formula: effect on nutrient balance, metabolic status and growth. Pediatr Res 2006; 59:265–270.

6. Poindexter BB, Langer JC, Dusick AM, et al. Early provision of parenteral amino acids in extremely low birth weight infants: relation to growth and neurodevelopmental outcome. J Pediatr 2006; 148:300–305.

7. Senterre T, Rigo J. Reduction in postnatal cumulative nutritional deficit and improvement of growth in extremely preterm infants. Acta Paediatr 2012; 101:e64–e70.

8. Colaizy TT, Carlson S, Saftlas AF, et al. Growth in VLBW infants fed predominantly fortified maternal and donor human milk diets: a retrospective cohort study. BMC Pediatr 2012; 12:124.

9. Clark RH, Wagner CL, Merritt RJ, et al. Nutrition in the neonatal intensive care unit: how do we reduce the incidence of extrauterine growth restriction? J Perinatol 2003; 23:337–344.

10. Ziegler EE. Meeting the nutritional needs of the low-birth-weight infant. Ann Nutr Metab 2011; 58 (Suppl 1):8–18.

11. Fanaro S, Ballardini E, Vigi V. Different pre-term formulas for different pre-term infants. Early Hum Dev 2010; 86 (Suppl 1):27–31.

12. Arslanoglu S, Moro GE, Ziegler EE. Adjustable fortification of human milk fed to preterm infants: does it make a difference? J Perinatol 2006; 26:614–621.

13. Senterre T, Rigo J. Optimizing early nutritional support based on recent recommendations in VLBW infants and postnatal growth restriction. J Pediatr Gastroenterol Nutr 2011; 53:536–542.

14. Clark RH, Thomas P, Peabody J. Extrauterine growth restriction remains a serious problem in prematurely born neonates. Pediatrics 2003; 111:986–990.

15. Hanson C, Sundermeier J, Dugick L, et al. Implementation, process, and outcomes of nutrition best practices for infants <1500 g. Nutr Clin Pract 2011; 26:614–624.

16. Ehrenkranz RA, Younes N, Lemons JA, et al. Longitudinal growth of hospitalized very low birth weight infants. Pediatrics 1999; 104 (2 pt 1):280–289.

17. Henriksen C, Westerberg AC, Ronnestad A, et al. Growth and nutrient intake among very-low-birth-weight infants fed fortified human milk during hospitalisation. Br J Nutr 2009; 102:1179–1186.

18. Hulst J, Joosten K, Zimmermann L, et al. Malnutrition in critically ill children: from admission to 6 months after discharge. Clin Nutr 2004; 23:223–232.

19. Lemons JA, Bauer CR, Oh W, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development neonatal research network, January 1995 through December 1996. NICHD Neonatal Research Network. Pediatrics 2001; 107:E1.

20. Isaacs EB, Morley R, Lucas A. Early diet and general cognitive outcome at adolescence in children born at or below 30 weeks gestation. J Pediatr 2009; 155:229–234.

21. Lucas A, Morley R, Cole TJ. Randomised trial of early diet in preterm babies and later intelligence quotient. BMJ 1998; 317:1481–1487.

22. Stephens BE, Walden RV, Gargus RA, et al. First-week protein and energy intakes are associated with 18-month developmental outcomes in extremely low birth weight infants. Pediatrics 2009; 123:1337–1343.

23. Ehrenkranz RA, Dusick AM, Vohr BR, et al. Growth in the neonatal intensive care unit influences neurodevelopmental and growth outcomes of extremely low birth weight infants. Pediatrics 2006; 117:1253–1261.

24. Weisglas-Kuperus N, Hille ET, Duivenvoorden HJ, et al. Intelligence of very preterm or very low birthweight infants in young adulthood. Arch Dis Child Fetal Neonatal Ed 2009; 94:F196–200.

25. Belfort MB, Rifas-Shiman SL, Sullivan T, et al. Infant growth before and after term: effects on neurodevelopment in preterm infants. Pediatrics 2011; 128:e899–e906.

26. Miller J, Makrides M, Gibson RA, et al. Effect of increasing protein content of human milk fortifier on growth in preterm infants born at <31 wk gestation: a randomized controlled trial. Am J Clin Nutr 2012; 95:648–655.

27. Hay WW, Thureen P. Protein for preterm infants: how much is needed? How much is enough? How much is too much? Pediatr Neonatol 2010; 51:198–207.

28. Griffin IJ, Cooke RJ. Development of whole body adiposity in preterm infants. Early Hum Dev 2012; 88 (Suppl 1):S19–24.

29. Johnson MJ, Wootton SA, Leaf AA, et al. Preterm birth and body composition at term equivalent age: a systematic review and meta-analysis. Pediatrics 2012; 130:e640–e649.

30. Ramel SE, Demerath EW, Gray HL, et al. The relationship of poor linear growth velocity with neonatal illness and two-year neurodevelopment in preterm infants. Neonatology 2012; 102:19–24.

31. Costa-Orvay JA, Figueras-Aloy J, Romera G, et al. The effects of varying protein and energy intakes on the growth and body composition of very low birth weight infants. Nutr J 2011; 10:140.

32. Kashyap S, Schulze KF, Forsyth M, et al. Growth, nutrient retention, and metabolic response of low-birth-weight infants fed supplemented and unsupplemented preterm human milk. Am J Clin Nutr 1990; 52:254–262.

33. Kashyap S, Schulze KF, Forsyth M, et al. Growth, nutrient retention, and metabolic response in low birth weight infants fed varying intakes of protein and energy. J Pediatr 1988; 113:713–721.

34. Roggero P, Gianni ML, Orsi A, et al. Implementation of nutritional strategies decreases postnatal growth restriction in preterm infants. PLoS One 2012; 7:e51166.

35. Kuschel CA, Harding JE. Protein supplementation of human milk for promoting growth in preterm infants. Cochrane Database Syst Rev 2009; 1–15.

36. Arslanoglu S, Moro GE, Ziegler EE. Preterm infants fed fortified human milk receive less protein than they need. J Perinatol 2009; 29:489–492.

37. Agostoni C, Buonocore G, Carnielli VP, et al. Enteral nutrient supply for preterm infants: commentary from the European Society of Paediatric Gastroenterology, Hepatology and Nutrition Committee on Nutrition. J Pediatr Gastroenterol Nutr 2010; 50:85–91.

38. Moya F, Sisk PM, Walsh KR, et al. A new liquid human milk fortifier and linear growth in preterm infants. Pediatrics 2012; 130:e928–e935.

39. Battaglia FC, Lubchenco LO. A practical classification of newborn infants by weight and gestational age. J Pediatr 1967; 71:159–163.

40. Gross SJ. Growth and biochemical response of preterm infants fed human milk or modified infant formula. N Engl J Med 1983; 308:237–241.

41. Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method and penalized likelihood. Stat Med 1992; 11:1305–1319.

42. Olsen IE, Groveman SA, Lawson ML, et al. New intrauterine growth curves based on United States data. Pediatrics 2010; 125:e214–e224.

43. De Curtis M, Rigo J. Nutrition and kidney in preterm infant. J Matern Fetal Neonatal Med 2012; 25 (Suppl 1):55–59.

44. Ridout E, Melara D, Rottinghaus S, et al. Blood urea nitrogen concentration as a marker of amino-acid intolerance in neonates with birthweight less than 1250 g. J Perinatol 2005; 25:130–133.

45. Roggero P, Gianni ML, Morlacchi L, et al. Blood urea nitrogen concentrations in low-birth-weight preterm infants during parenteral and enteral nutrition. J Pediatr Gastroenterol Nutr 2010; 51:213–215.

46. Schulze KF, Stefanski M, Masterson J, et al. Energy expenditure, energy balance, and composition of weight gain in low birth weight infants fed diets of different protein and energy content. J Pediatr 1987; 110:753–759.

47. Simmer K. Aggressive nutrition for preterm infants—benefits and risks. Early Hum Dev 2007; 83:631–634.

48. American Academy of Pediatrics, Committee on NutritionNutritional needs of low-birth-weight infants. Pediatrics 1977; 60:519–530.

49. Lubchenco LO, Hansman C, Boyd E. Intrauterine growth in length and head circumference as estimated from live births at gestational ages from 26 to 42 weeks. Pediatrics 1966; 37:403–408.

50. Lubchenco LO, Hansman C, Dressler M, et al. Intrauterine growth as estimated from liveborn birth-weight data at 24 to 42 weeks of gestation. Pediatrics 1963; 32:793–800.

Keywords:

growth status; kilocalorie intake; linear growth; neonatal nutrition; preterm infants; protein intake; weight growth; z scores

© 2014 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,

Login

Article Tools

Images

Share

Search for Similar Articles
You may search for similar articles that contain these same keywords or you may modify the keyword list to augment your search.

Connect With Us

 

 

Twitter

twitter.com/JPGNonline

 

Visit JPGN.org on your smartphone. Scan this code (QR reader app required) with your phone and be taken directly to the site.