Blanco, Cynthia L.*; Gong, Alice K.*; Schoolfield, John†; Green, Belinda K.‡; Daniels, Wanda*; Liechty, Edward A.§; Ramamurthy, Rajam*
See “Early High-dose Amino Acids for ELBW Infants: Too Early and Too Much?” by Greer on page 576.
Extremely-low-birth-weight (ELBW) infants are surviving in increasing numbers and their postnatal growth and development remain a challenge. Concerns for their immediate nutritional needs have led neonatologists to advocate for early and aggressive parenteral and enteral nutrition. Parenteral intravenous (IV) amino acid (AA) intake up to 3 g · kg−1 · day−1 initiated soon after birth has rapidly become part of the routine care of prematurely born infants (1,2).
The potential benefits of early protein supplementation in premature infants include positive nitrogen balance, improved glucose use, weight gain, and improved head circumferences (3–5). A serious concern is the ability of ELBW infants to metabolize all of the AAs, leading to high concentrations of some AAs (6,7). Few studies address the long-term effects of early and aggressive protein supplementation based on AA profiles resulting from difficulty in obtaining serum AA concentrations as part of routine care.
One objective of the randomized trial of the metabolic responses of ELBW infants to early and high AA supplementation started shortly after birth (8) was to examine the effects of early and high AA supplementation on growth, overall health, and neurodevelopment of ELBW infants during their first 2 years of life.
Sixty-one ELBW infants admitted to University Hospital, San Antonio, TX, were recruited prospectively between November 2002 and September 2005. Initial enrollment criteria were birth weight (BW) <1000 g and age <12 hours of life. A third inclusion criterion of gestational age (GA) >24 weeks was added by the data safety monitoring board after enrollment of 20 infants because of the high mortality rate of infants <24 weeks gestation. Exclusion criteria were major congenital anomalies and imminent death. The study was approved by the University of Texas Health Science Center San Antonio institutional review board. After informed consent was obtained from parents, infants were randomized by the clinical pharmacist with cards in sealed sequential opaque envelopes to either a standard AA protocol or an early and high AA protocol. The standard AA group received the standard regimen in our neonatal intensive care unit; during the study period, which consisted of 0.5 g · kg−1 · day−1 IV AA (Aminosyn PF; Abbott Laboratories, Chicago, IL, with 40 mg · kg−1 · day−1 of cysteine hydrochloride) in the total parenteral nutrition (TPN) solution starting in the first 24 to 36 hours of life with increases of 0.5 g · kg−1 · day−1 every 24 hours to a maximum of 3.0 g · kg−1 · day−1, and continued until the seventh day of life (DOL). The early and high AA group received 2.0 g · kg−1 · day−1 of IV AA soon after enrollment with increases of 1 g · kg−1 · day−1 every 24 hours up to a maximum of 4 g · kg−1 · day−1, and continued at that level until DOL 7. After the study period, all of the infants were maintained on TPN with AA dosage at 3.5 g · kg−1 · day−1 until sufficient enteral feedings were accomplished and then weaned as TPN volume decreased (approximately at 2 g · kg−1 · day−1 on half of the total fluid intake and then 1 g · kg−1 · day−1 once less than one-third of total fluid intake). All of the infants were prescribed lipids (Intralipid 20%, Fresenius Kabi AG Clayton R&D, Clayton, NC), glucose, minerals, trace elements, and vitamins according to nursery protocol and as tolerated by the infant. Specifically, the protocol for vitamin supplementation is as follows: parenteral pediatric multivitamin injection is added to all of the parenteral nutrition according to weight (<1 kg = 1.5 mL/day, 1–3 kg = 3.25 mL/day, >3 kg = 5 mL/day); enteral multiple vitamin drops are given to all of the infants once full enteral feeds are reached at 0.5 mL twice per day regardless of weight. All of the infants are then discharged on enteral multiple vitamin drops at 1 mL once per day. Initiation and advancement of oral feedings were determined by the attending neonatologist. For the most part, trophic feeds were introduced as soon as the infant was metabolically stable and had evidence of gastric motility.
The AA solution was added via pharmacy per study protocol and the clinical care team was unaware of group allocation. AA solution was discontinued if the plasma ammonia level was >91 μmol/L by DOL 1 or >79 μmol/L by DOL 3. A plasma ammonia level was only obtained if serum blood urea nitrogen (BUN) was >60 mg/dL (Advia 1650 Chemistry System, Siemens Medical Solutions Diagnostics, Tarrytown, NY).
Plasma samples for quantitative AA concentrations were collected a priori on days 1, 3, and 7 of the study protocol. AA concentrations were determined by reverse-phase high-performance liquid chromatography using the Waters PICO-Tag (Waters Corp, Milford, MA) methodology. The high-performance liquid chromatography was recalibrated using standards of known concentration daily, and concentrations in plasma control samples from a single pool were determined in triplicate after each recalibration. The average interassay coefficient of variation was 11% and the intraassay coefficient of variation was 6%. The coefficient of variation values for individual AAs are available on request.
All of the surviving infants were enrolled in the premature infant development premiere program, which evaluates them at 0, 3, 6, 12, and 18 months corrected gestational age (CGA) and at 24 months chronological age. Follow-up occurred from February 2003 through July 2007. At each visit, an interim medical history was obtained and a physical examination was performed. A comprehensive neurological examination was performed by an experienced neonatologist dedicated to the follow-up program. The Bayley Scales of Infant Development II were administered from 6 months of age by a developmental psychologist and a certified examiner. A score of 50 was assigned when the child was too impaired to complete the testing. The Bayley is administered by a bilingual person (English-Spanish) in the language that is used at home. Anthropometric measurements, duration of breast milk feeding, hospitalizations, medications, presence of seizures, vision status, and socioeconomic information were collected at each visit. The criteria used for referral to early childhood intervention services were Bayley scores of mild or significant impairment (categories: normal, mild impairment, significant impairment), failure to attain milestones for a given age (categories: normal, suspect, abnormal), or abnormal neurological examination by the physician. All of the members of the clinic staff were masked to treatment group allocation of each infant.
The sample size calculation based upon a primary outcome variable of reduction of potassium levels has been reported (8). Distributions and means of demographic and clinical variables were compared across standard AA and early and high AA groups using χ2 tests, Student t test, and Fisher exact test. Data from each individual AA and summed concentrations for essential AA, nonessential AA, and total AA were analyzed by repeated measures multivariate analysis of variance, using the randomized nutritional protocol (standard AA vs early and high AA) as the independent variable and the day of sampling (day 1, 3, or 7) as the time-dependent repeated measure. Linear regression and Pearson correlation coefficients were used when indicated.
The weight, length, and head circumference measures were converted to sex-standardized z scores using the Revised Centers for Disease Control and Prevention 2000 sex-specific growth data (9). Estimated means and standard errors for growth z scores were obtained by a general linear model approach using mixed model analysis of variance. A first-order autoregressive structure was used to model covariance for the repeated effect of visit time. The analysis of variances tested the interaction between group membership (standard AA or early and high AA) and visit time. If the F test for interaction was significant (P < 0.10), then group comparisons were performed for each visit time using Student t tests. If the F test for interaction was not significant (P > 0.10), then F tests were performed to assess the main effects of group and visit, with P < 0.05 considered significant. For ease of interpretability, group comparisons were performed for each visit time, even if there was no evidence of significant interaction. Data were collected and encoded for computerized analysis by a statistician with the use of SAS (SAS Institute, Cary, NC) and SPSS 15 (SPSS Inc, Chicago, IL) software.
Initial Course and Outcome at Discharge
Sixty-one infants were enrolled, 31 in the standard AA group and 30 in the early and high AA group. The demographic characteristics were not different between the standard AA and early and high AA at study entry (8). Serum BUN and urine urea were significantly higher in the early and high AA group (8). The demographic characteristics of infants that completed follow-up and those lost to follow-up are shown in Table 1. Total parenteral caloric and enteral feed intakes were similar between groups on DOL 1, 3, and 7 (Table 2). Most infants were still receiving breast milk at discharge, and the time to reach full feeds was similar between groups (Table 2). The number of initial consecutive days and total number of days on TPN were not different between groups (Table 2). No differences were found between the weight (P = 0.4 at birth, P = 0.7 at discharge) and head circumferences (P = 0.4 at birth, P = 0.2 at discharge) or weight gain by DOL 28 (P = 0.8 from all of the survivors at discharge, P = 0.4 from infants who completed follow-up [Table 2]). At discharge, there were no statistical differences between groups in the incidence of intraventricular hemorrhage (IVH), periventricular leukomalacia, retinopathy of prematurity, necrotizing enterocolitis, or in the overall mortality and length of stay for either all of the survivors or those that completed follow-up (Table 1). The incidence of bronchopulmonary dysplasia (BPD [defined as oxygen supplementation by 36 weeks CGA]) was higher in the early and high AA group than in the standard AA group (15/24 vs 8/27 survivors to discharge, respectively, P = 0.01). Most infants were weaned off oxygen a few days after reaching 36 weeks CGA; 5 infants in the early and high AA group and 4 infants in the standard AA group were discharged on oxygen (P = 0.7) and from those, only 2 of each group completed the 2-year follow-up.
Characteristics of the Follow-up Cohort
Infants were examined from February 2003 through June 2007. The number of infants eligible for the study and follow-up is shown in Figure 1. Of the 22 infants in the standard AA, 19 (86%) completed at least 4 visits and of the 21 infants in the early and high AA, 17 (81%) completed at least 4 visits. The demographics and clinical outcomes of those who failed to return or complete the follow-up at 18 to 24 months were comparable to those who did (Table 1). There was a trend toward lower GA on those infants lost to follow-up from the early and high AA group. In the group that completed the 18- to 24-month follow-up, there was a tendency of higher incidence of BPD in the early and high AA group (Table 1). The total number of days of mechanical ventilation was 15 ± 15 in the standard group versus 22 ± 19 in the early and high AA group (P = 0.2).
General Health and Growth Outcomes
The weight, length, and head circumference measures at each visit were converted to sex-standardized z scores. As expected for ELBW infants, all of the z scores were below the mean. The early and high AA infants had z score means for weight that were significantly lower than the standard AA group at 6, 12, and 18 months CGA as well as at 24 months chronological age (Fig. 2). In addition, the z scores for length/height and head circumferences were significantly lower for the early and high AA group at most visits (Fig. 2). The differences in weight and fronto-occipital circumference (FOC) were found mostly in boys at 6-, 12-, 18-, and 24-month visits (P < 0.05 for male differences between groups at each visit). Postdischarge outcomes, including hospital readmissions, use of pulmonary and seizure medications, blindness, and breast-feeding rates, were similar in both groups up to 12 months CGA (Table 2).
The Bayley II Mental Developmental Index (MDI) and Psychomotor Developmental Index (PDI) scores at 6, 12, 18, and 24 months are detailed in Table 2. The MDI and PDI scores were similar between groups at each age tested, with the exception of a lower MDI score at 18 months CGA in the early and high AA group. This difference was not found at 24 months of age. No differences in the rate of cerebral palsy and blindness were seen between groups (Table 2). A total of 3 infants were blind: 1 had bilateral retinal detachment, 1 had bilateral optic nerve atrophy, and 1 had bilateral decreased vision caused by retinopathy of prematurity and laser therapy that qualified for legal blindness. There was no difference in the percentage of infants who received early childhood intervention services between groups at any GA (Table 2) or the number of visits with speech-language, physical pathologytherapy, or occupational therapy (data not shown).
Single and cumulative concentrations of 18 plasma AAs have been published (7). The cumulative concentrations of AAs negatively correlated with MDI at 18 months CGA (Fig. 3). Similarly, cumulative concentrations of essential and nonessential AA had a negative correlation with MDI at 18 months CGA (data not shown). The cumulative AA concentrations showed a negative correlation with FOC and weight z scores at 18 CGA and 24 months of age (Fig. 3).
Four AAs, phenylalanine, isoleucine, valine, and leucine, were associated with poor growth and low MDI and/or PDI scores. Levels of these AAs were divided into quartiles. Those with the highest phenylalanine concentrations had significantly lower MDI score (67 ± 14 high vs 82 ± 12 low phenylalanine, P < 0.05) and lower FOC z score at 12, 18, and 24 months CGA (P < 0.05 at each visit). Infants with the highest isoleucine and valine levels had lower MDI score at 12 and 18 months CGA (high isoleucine 67 ± 16 vs low isoleucine 81 ± 13, P < 0.05; high valine 68 ± 16 vs low valine 81 ± 13, P < 0.05) and lower weight and FOC z scores at 12, 18, and 24 months CGA (P < 0.05 at each visit). High plasma leucine was also associated with lower weight z scores at 6, 12, 18, and 24 months CGA (P < 0.05 at each visit). There were no differences in BW, GA, length of stay, Clinical Risk Index for Babies score, BPD, or IVH between those infants with high cumulative or single AA concentrations (upper quartile) versus those with lower concentrations (lower quartiles) with the exception of isoleucine. Infants with the highest isoleucine concentrations were smaller (BW 706 ± 174 g high quartile vs 846 ± 113 g, P < 0.05) and had a higher incidence of IVH (67% high isoproterenol vs 12% low isoproterenol group). All of the infants in the high quartiles of phenylalanine, isoleucine, valine, and leucine received early and high AA, whereas only 37% of infants in the lower quartiles were from the early and high AA group.
ELBW infants who received early and high IV AA during the first week of life had comparable MDI and PDI scores to infants who received the standard IV AA therapy at 2 years of age; however, those allocated to early and high IV AA were associated with impaired overall growth at 2 years. IV AA are widely used in premature infants and many advocate its administration of up to 4 g · kg−1 · day−1 during the first week of life to mimic fetal protein accretion rates (10); however, there are no prospective data on the long-term effects of this nutritional approach.
The short-term benefits of early provision of AA up to 3 g · kg−1 · day−1 in ELBW infants during the first few days of life have been demonstrated in several studies (2–5). There are reports of an association of early protein and energy intake with improved growth and neurodevelopment in ELBW infants (3,11). In a retrospective review of energy and protein intake during the first week of life, every 10 kcal · kg−1 · day−1 of energy intake was associated with a 4.6-point increase with the MDI and each gram per kilogram per day in protein intake was associated with an 8.2-point increase in the MDI. Their nursery protocol at the time started with parenteral nutrition on DOL 1 with 3.4 kcal/g carbohydrate infused, 1.0 g · kg−1 · day−1 of protein, and advanced by 0.5 g · kg−1 · day−1 up to 2.5 to 3.5 g · kg−1 · day−1(11). This standard protein regimen is similar to our control population. Therefore, the long-term effects of early and high IV AA supplementation (up to 4 g · kg−1 · day−1, reached as early as 48 hours of life) in ELBW infants remain relatively unstudied.
The primary outcome measure for the present study was incidence of hyperkalemia; early and high AA supplementation had no effect on the incidence of hyperkalemia in ELBW infants (8). The majority of ELBW infants with the exception of the most immature neonates appeared to tolerate both regimens. From birth to discharge, growth was similar between groups; however, infants in the early and high AA group had significantly reduced global growth after 3 months CGA, and these differences were maintained up to 2 years of age. Interestingly, those differences were more pronounced in boys; sex differences have been reported in other trials (3).
Several possibilities may explain the differences in growth at 2 years. Disproportional branched-chain AA concentrations have been linked to impaired growth and development in human and animal studies with possible mechanisms related to insulin release, tissue protein synthesis/degradation, catabolism of other branched-chain AAs, and transport of large neutral AAs into tissues, particularly the brain (12). Infants with metabolic disorders involving single high AA concentrations such as phenylketonuria have poor neurodevelopment and growth (13,14). ELBW infants face severe nutritional compromises during critical growth and development that may have lasting consequences (15).
Alternatively, early postnatal diet may have contributed to the differences in growth at 2 years, but enteral intake in the first week of life, time to reach full feeds, and days of TPN were similar between groups. Postdischarge diet would be another factor for the differences in growth at 2 years. The follow-up clinic devotes meticulous attention to caloric intake and calculations up to 12 months, which may be reliable when the infant remains on a predominantly milk diet. Afterwards, tracking the food intake of these infants has been difficult and is based upon parental recall. In the present study, there were no differences in the rates of continued breast-feeding and/or other socioeconomic factors between the 2 groups.
Our data showed a trend of a higher incidence in BPD in the early and high AA group. BPD, known to be associated with poor neurodevelopment and growth, could explain the growth differences (16–18). Our definition of BPD was based on continued oxygen supplementation at 36 weeks CGA. Our unit policy is to wean babies to maintain oxygen saturations in the low 90s. Ten of the 14 infants were weaned off oxygen shortly after the 36th week CGA, and the total days on mechanical ventilation were not different between groups. In addition, the clinical status of the infants at 40 weeks CGA (discharged off oxygen), the rehospitalization rates, and subsequent respiratory medications were similar between groups (Tables 1 and 2). Thus, chronic respiratory compromise does not appear to be the sole explanation for the growth differences between groups. Furthermore, those infants lost to follow-up from the early and high AA group tended to be of a lower GA, which theoretically could contribute to even more growth discrepancies.
We recently reported the plasma AA concentrations from this trial (7). There were major differences in AA concentrations when higher amounts of AA were provided. In the present study, we only found significant differences in MDI scores at 18 months. A larger sample size will be helpful to determine whether the lower MDI scores are truly significant and whether those differences persist beyond 18 months CGA. Of most concern are the negative correlations found between plasma AA concentrations and MDI scores and growth regardless of randomization.
In addition, infants who received early and higher parenteral AA had particularly higher concentrations of isoleucine, leucine, valine, phenylalanine, Lysine, Methionine, and proline (7). These between-group differences resolved by DOL 7 except for leucine, and therefore infants may tolerate this regimen after the first week of life. High leucine increases branched-chain α-keto acid dehydrogenase leading to diminished valine and isoleucine for protein synthesis (19). It is important to note that infants with the highest concentrations from 4 of these 7 AAs were associated with significantly lower MDI scores and worse growth outcomes up to 2 years of age. Demographic characteristics were similar to those with lower AA concentrations with the exception of group allocation and BUN levels. Interestingly, infants with higher isoleucine concentrations were smaller and had higher incidence of IVH, but our numbers are too small to consider any causal relation.
A weakness of the present study is the small sample size, and although follow-up was planned, growth and developmental parameters were not planned outcome measures. It is important to emphasize the strength that most infants had serial, detailed evaluations set a priori during the first 2 years of life by the premature infant development premiere program. This program has reported that infants younger than 27 weeks had decreased weight and FOC that started at 12 months CGA, reached a nadir at 24 months, and persisted. The MDI and PDI scores of those infants paralleled their growth z scores (20). The mean GA for both groups in the present study was <27 weeks gestation, and therefore differences found at 2 years of age may persist.
Because many premature infants have impaired neurological development, long-term follow-up is of utmost importance to evaluate effects of therapies used in the first few weeks of life. Although our study was not powered for long-term follow-up assessments, reporting the 2-year outcome from the present study will provide the basis for power calculations and hypothesis framework for future studies.
In summary, ELBW infants who received early and higher IV AA during the first week of life up to 4 g · kg−1 · day−1 did not have improved global growth at 2 years. Before this therapy becomes standard of care, further larger trials are needed to assess the best strategy of IV AA advancement and dosage to provide the most benefit and long-term safety.
We thank the personnel from the premature infant development premiere program in San Antonio, TX, for their dedication to the children and thorough evaluations.
1. Te Braake FWJM, Van Den Akker CHPM, Wattimena DJL, et al. Amino acid administration to premature infants directly after birth. J Pediatr
2. Thureen PJ, Melara DL, Fennessey AP, et al. Effect of low versus high intravenous amino acid intake on very low birth weight infants in the early neonatal period. Pediatr Res
3. 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
4. Thureen PJ, Anderson AH, Baron KA, et al. Protein balance in the first week of life in ventilated neonates receiving parenteral nutrition. Am J Clin Nutr
5. Van Den Akker CHP, Te Braake FWJ, Wattimena DJL, et al. Effects of early amino acid administration on leucine and glucose kinetics in premature infants. Pediatr Res
6. Clark RH, Chace DH, Spitzer AR, et al. Effects of two different doses of amino acid supplementation on growth and blood amino acid levels in premature neonates admitted to the neonatal intensive care unit: a randomized, controlled trial. Pediatrics
7. Blanco CL, Gong AK, Green BK, et al. Early changes in plasma amino acid concentrations during aggressive nutritional therapy in extremely low birth weight infants. J Pediatr
8. Blanco CL, Falck A, Green BK, et al. Metabolic responses to early and high protein supplementation in a randomized trial evaluating the prevention of hyperkalemia in extremely low birth weight infants. J Pediatr
9. Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC growth charts: United States. Adv Data
10. Adamkin DH. Nutrition management of the very low-birthweight infant: I. total parenteral nutrition and minimal enteral nutrition. NeoReviews
11. 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
12. Harper AE, Miler RH, Block KP, et al. Branch-chain amino acid metabolism. Ann Rev Nutr
13. Enns GM, Koch R, Brumm V, et al. Suboptimal outcomes in patients with PKU treated early with diet alone: revisiting the evidence. Mol Genet Metab
14. Waisbren NK, Fahrbach K, Cella C, et al. Phenylalanine blood levels and clinical outcomes in phenylketonuria: a systematic literature review and meta-analysis. Mol Genet Metab
15. Eriksson JG, Forsn T, Tuomilehto J, et al. Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ
16. Farooqi A, Hagglof B, Sedin G, et al. Growth in 10- to 12-year-old children born at 23 to 25 weeks’ gestation in the 1990s: a Swedish national prospective follow-up study. Pediatrics
17. Niklasson A, Engstrom E, Hard AL, et al. Growth in very preterm children: a longitudinal study. Pediatr Res
18. Vohr BR, Wright LL, Poole WK, et al. Neurodevelopmental outcomes of extremely low birth weight infants <32 weeks’ gestation between 1993 and 1998. Pediatrics
19. Vlaardingerbroek H, Van den Akker CHP, De Groof F, et al. Amino acids for the neonate. NeoReviews
20. Powers GC, Ramamurthy R, Schoolfield J, et al. Postdischarge growth and development in a predominantly hispanic, very low birth weight population. Pediatrics