|
Introduction
The combined use of exogenous human albumin with total parenteral
nutrition (TPN) for improvement of nutritional status has been reported
in the literature for over one decade and is based on the ability
of modified TPN to quickly normalize the serum levels of albumin,
correcting hypoalbuminemia, maintaining the oncotic pressure and
allowing for a better prognosis ().
Low albumin levels often are observed in malnourished adult and
pediatric hospitalized patients and in ill and preterm newborns,
being associated with several organic dysfunctions (reduction in
oncotic pressure, in resistance to infection and in healing ability,
peripheral and intestinal edema, decrease in gastrointestinal motility
with intolerance to enteral nutrition, pulmonary infiltrate) and
adverse clinical outcome ().
The "restorative" properties of albumin in TPN are believed
to be related to the amino acid profile and to the caloric content
of TPN, resulting in improved protein synthesis or inhibition of
albumin degradation. Thus, albumin supplementation in ill, hypoalbuminemic
patients with poor nutrition may be useful to improve albumin-related
functions. Newborn infants, especially preterm and small ones, have
limited body stores of proteins and energy at birth. Moreover, metabolic
demands, immature or inadequate enzyme systems, gastrointestinal
immaturity, respiratory diseases, poor absorption and/or insufficient
amounts of individual nitrogen and amino acid precursors, may restrict
the appropriate supply of proteins and calories, reducing hepatic
protein synthesis. On the other hand, several studies assert that
supplementation provides no proven clinical benefit, and that it
just elevates serum levels, which does not justify the high cost
of such therapy ().
The use of human albumin to improve the general health and nutritional
statuses of hospitalized newborns remains unclear. Most times, its
use is empirical, as a result of the observed inverse correlation
between serum albumin concentration and morbidity and mortality.
Considering the controversy over the use of human albumin, and using
the available methods for the assessment of nutritional status,
the aim of the present article was to evaluate the effect of human
albumin in the TPN of severely ill newborns on the serum levels
of protein markers such as albumin, total proteins, prealbumin,
retinol-binding protein (RBP), on nitrogen retention, parameters
related to weight outcome, enteral nutrition, length of stay (in
days) in the intensive care unit (ICU) and in the hospital, and
morbidity and mortality.
Patients and methods
A case-control study was performed in preterm newborns weighing
less than 2,500 g, admitted to the Neonatal Intensive Care Unit
(NICU) of Hospital Universitário de Santa Maria (HUSM), between
November 1997 and November 1998. The patients were not on enteral
nutrition as their clinical conditions did not allow so; although
they had severe diseases, their clinical and laboratory conditions
were stable, that is, the amount of fluids or TPN components for
each newborn did not have to be changed during different days within
the study period. The exclusion criteria were: acute renal failure,
cholestasis, protein loss (fistulas), use of medications that could
interfere with water excretion (diuretics, indomethacin) and maternal
use of corticosteroids before delivery. Forty newborns were evaluated
after being randomly placed in two groups. Thirty newborns met the
inclusion criteria at the end of the study, each group consisting
of 15 newborns who were receiving standardized TPN, following the
routine of the hospital, as described next:
- Control group receiving no albumin (C): received regular standardized
TPN;
- Group receiving albumin (A): received standardized TPN combined
with 1g/kg/day of human albumin (Zenalb®20-Human Albumin 20%,
Bio Products Lab.), given daily at every 12 hours, throughout
the study period. After the initial adaptation period, TPN was
gradually introduced according to the routine and tolerance of
the newborn in the seven-day assessment period (starting on the
third or fourth postnatal day and finishing on the 10th or 11th
day). TPN solutions were prepared at the pharmacy of HUSM, according
to the medical prescription. Amino acid solution (starting with
1 g/kg/day with a gradual increase of 0.5 g/kg/day up to 3 g/kg/day)
and lipid solution (1 g/kg/day at the beginning of TPN with an
increase 0.5 g/kg/day up to 3 g/kg/day) were, respectively, PEDIAMINO
PLM 10%, BBraun and INTRALIPID 10%, Darrow. Data on maternal history,
obstetric history, birthweight, gender, gestational age (Capurro's
method), appropriateness of weight for gestational age, confirmed
clinical diagnoses (on admission, at the beginning and at the
end of the study), clinical outcome during the study period, respiratory
acuity score (RAS) ()
at the beginning and at the end of the study and observed complications
were obtained for each newborn. Each newborn received at least
90% of the prescribed TPN volume. Two ml of blood was collected
on the first and seventh days of study. A 24-hour urine sample
was collected on the seventh day using a pediatric urine collection
bag. Flasks were kept in the refrigerator at -4 oC for 24 hours.
At the end of the nitrogen balance study, volumes were measured,
and one sample was frozen at -20 ºC for later analysis and
estimation of nitrogen retention. None of the newborns who participated
in the study had bowel movement on the day close to or during
the balance. A sample of TPN infused on the day of the nitrogen
balance study was collected from each newborn. These samples were
stored in sterile flasks and kept in the refrigerator at -20 ºC
up to the moment of biochemical measurement.
Laboratory measurements
Nitrogen was measured in urine and TPN solutions. Total protein,
albumin, prealbumin and RBP were measured in the plasma. The Kjeldahl
method was used for the determination of nitrogen . Total protein
was determined using the Biuret method, whereas prealbumin and RBP
were measured by nephelometry. All measurements were made in duplicate,
except for prealbumin and RBP.
Statistical analysis
The individual results of each newborn were recorded in a specific
protocol and then stored in a database using Epi-info (version 6.0,
July 1996), and later analyzed in Epi-Info and Stata (1998). The
nonparametric Kruskal-Wallis test and the chi-square test were used,
and a p < 0.05 was considered to be significant. The study protocol
was approved by the Ethics Committee of Hospital de Clínicas,
School of Medicine of Ribeirão Preto-USP-SP, and of Hospital
Universitário de Santa Maria (UFSM-RS). An informed consent
was obtained from parents or legal representatives in all cases.
Results
Thirty newborns (n = 30) were included in the study - 15 received
albumin and 15 did not (control group). No difference was noted
between groups regarding gender, birthweight, gestational age, appropriateness
of weight for gestational age, use of oxygen therapy, use of ventilator
and RAS (Table 1). The study groups showed a similar disease profile
at the beginning of the study, in which there was a predominance
of hyaline membrane disease, hypoxic-ischemic encephalopathy, infection
and surgical complications, such as esophageal atresia, diaphragmatic
hernia, gastroschisis and duodenal atresia, without any significant
differences (Table 2). The values, expressed in median and quartiles,
regarding the volume and nutrients given to the newborns in both
groups during the study period did not show any statistically significant
differences (Table 3).
Table 1 -
Characteristics of the clinical status of the populations studied
Table 2 -
Pathologies of newborns from groups C and A in the beginning of
the study
Table 3 -
Median and quartiles of nutrients given to the newborns from groups
C and A during the study
Serum biochemical markers
There was statistically significant difference between the groups
only at the end of study regarding total protein and albumin (Table
4).
Table 4 -
Median and quartiles of albumin values (g/dl), prealbumin (mg/dl),
RBP (mg/dl) and total proteins (g/dl) in groups C and A, at the
beginning and end of the study
Nitrogen balance
The balance at the end of the study did not reveal statistically
significant differences between the groups in relation to the amount
of nitrogen given, passed in the urine and retained (Table 5).
Table 5 -
Median and quartiles of the values of infused nitrogen in parenteral
nutrition, nitrogen in urine, nitrogen retention and balance at
the end of the study (balance day) in groups C and A
Start of enteral nutrition, time necessary to reach full
enteral volume, length of stay in the ICU, total length of hospital
stay and mortality.
No difference was observed in these variables between the groups
(Table 6).
Table 6 -
Median and quartiles of time (days) of variables associated with
parenteral and enteral nutrition, length of hospital and ICU stay.
Weight outcome
No statistically significant differences were found between groups
with regard to weight gain during the study period (Table 7). However,
in group C only two newborns (13.3%) gained weight, whereas in group
A eight newborns (53.3%) gained some weight by the end of the study
(Table 7).
Table 7 -
Median and quartiles of weight gain (grams) during the study period
Discussion
Proper nutrition is important for the regulation of albumin synthesis,
and serum albumin concentration is still considered to be a good
nutritional indicator by many authors, and despite some limitations,
it is still used on a routine basis for the assessment of nutritional
status in hospitalized patients of all ages ().
In preterm newborns, albumin and total protein levels are usually
lower than in full-term newborns, infants and older children. There
is a positive significant correlation between gestational age and
the concentration of total protein and albumin (),
ranging from 20 g/l at 28 weeks of gestation to 30 g/l at term;
total protein levels rise from 40 g/l at 28 weeks to 60 g/l at birth
().
With regard to albumin, there was statistically significant difference
between groups at the beginning of the study, but a significant
difference was observed between the groups (p < 0.001) at the
end of the study, thus showing that human albumin replacement actually
increases the serum levels of total protein and albumin in the short
run. The long half-life limits the albumin level for the identification
of acute changes in the nutritional status, and low sensitivity
and specificity is a poor parameter for the individual assessment
of nutritional status of patients, being more appropriate for long-term
and epidemiological studies ().
Prealbumin and RBP levels are more sensitive indicators of appropriate
nutrition (),
they seem to correlate best with nitrogen balance during nutritional
therapy, demonstrating earlier response to refeeding than albumin,
total protein and transferrin (),
and their control is useful to identify early changes in protein
and energy intake and to assess the efficacy of nutrition in a short-term
period ().
In the present study, serum levels of prealbumin and RBP are lower
than those described in the literature ()
and did not reveal statistically significant difference between
or within groups. It is likely that, in the present study, the insufficient
protein-energy intake can explain low prealbumin and RBP levels
and the maintenance of these nutritional markers at the same level
after seven days of TPN. A protein intake less than 2 g/kg/day and
an energy intake lower than 100 cal/kg/day result in significantly
lower levels of prealbumin and RBP (),
which occurred in our study (Table 3).
The estimation of protein requirements by nitrogen balance assessments
is based on the observation that requirements are met when maximum
retention is achieved. Protein requirements in newborns obtained
from nitrogen balance studies range from 1.6 to 4.2 g/kg/day, depending
on the type of protein received by the infant, physical conditions
and gestational age ().
In contrast to healthy adults who have a neutral nitrogen balance,
newborns need a positive balance so that they can grow and develop
properly. Nitrogen requirements result mainly from the amount of
metabolized proteins, and are much higher among newborns. In the
first months of life, nitrogen retention corresponds to 140-250
mg/kg/day for full-term breastfed newborns and nearly 350 mg/kg/day
for formula-fed infants ().
Protein requirements, however, vary with age: for low-birthweight
newborns requirements may correspond to 3.5 g/kg/day (in order to
supply around 480 mg/kg/day of nitrogen) and 2 g/kg/day (280 mg/kg/day)
for full-term infants ().
Preterm newborns have an immature amino acid metabolism, therefore,
the amount to be supplied in order to obtain nitrogen retention
similar to that found in utero is not easily reached and depends
on the digestibility and use of the protein received ().
In the present study, patients of both groups received similar amounts
of nitrogen, with no statistically significant difference between
the groups. Such amounts are compatible with those recommended for
newborns, but they may be low for preterm infants. Protein and non-protein
sources were similar in both groups. Nitrogen balance was positive
in both groups, with retention of 65.8% in group C and of 75.5%
in group A of the nitrogen supplied at median levels. This confirms
the urge of newborns to retain nitrogen, even receiving a protein-energy
intake lower than recommended, as occurred in the present study,
which is consistent with the available literature ().
In our study, in one week, 97.7% of the patients were in an anabolic
state, even with a protein-energy intake below that which is recommended
and, probably because of that, there were no significant changes
in prealbumin, RBP and weight.
Newborns in group A lost less weight, regained birthweight apparently
shortly before than did those in the control group, and gained more
weight by the end of the study (Table 7). They also started enteral
nutrition at an earlier stage and the time necessary to reach full
enteral volume was slightly short (2 days before group C), but these
differences were not statistically significant. Energy content was
similar in both groups, as well as amino acid intake (not computing
the protein value of albumin supplementation). Enteral absorption
of fluids partially depends on serum oncotic pressure. Low serum
albumin levels are correlated with reduced fluid intake and increased
intestinal fluid retention ().
This buildup of fluids reduces intestinal motility and nutrient
absorption. Thus, differences in weight gain might be partly explained
by positive effects of a higher serum albumin level and of higher
oncotic pressure, with improved tolerance of the diet by newborns
().
Similarly, studies conducted with adult patients also revealed a
positive correlation between serum albumin levels and better tolerance
of enteral nutrition (),
which could not be statistically confirmed in the present study.
In newborns, the use of albumin in TPN also had some influence on
the implementation of enteral nutrition and showed that the treated
group regained birthweight more quickly ().
Differences in weight gain could also be attributed to improved
intestinal motility and absorption due to the effect of albumin.
In this case, however, improvement in intestinal tolerance should
be expected resulting in earlier implementation of enteral nutrition
and larger weight gain. In the present study, despite a shorter
observation time, differences were not statistically significant
between groups regarding weight, age, implementation of enteral
nutrition and time necessary to reach full enteral volume. Data
available about newborns are scarce and inconclusive when it comes
to show the benefits of such practice; and besides, no literature
studies have been carried out on albumin supplementation in infants
using enteral nutrition. Further studies including a larger number
of infants and longer use of albumin are necessary to determine
to what extent the early recovery of birthweight in infants is due
to some improvement in the tolerance of enteral nutrition or only
to an improvement of the underlying disease.
The available literature describes a significant and inversely
proportional correlation between albumin levels and length of hospital
stay, morbidity and mortality, with evidence that serum albumin
level may be a predictive factor for the risk of death (),
thus justifying the administration of albumin. However, many of
these studies included groups of very old patients with chronic
diseases, in which the disease itself may have caused hypoalbuminemia
and death, instead of the nutritional status. Our study used a homogeneous
population of critically ill newborns, with comparable initial albumin
levels, and no statistically significant difference regarding length
of hospital stay and mortality between the groups. Prospective,
randomized, controlled and double-blind studies did not find significant
differences after albumin supplementation in the total length of
hospital stay, in the necessity for mechanical ventilation, in the
tolerance of enteral nutrition, morbidity, and in the reduction
of mortality, despite a significant increase in serum albumin levels
();
these data are consistent with those obtained in the present study.
In these studies and in the present study as well, there was an
increase in serum albumin levels, but this increase did not reveal
significant differences in relation to biochemical nutritional status
(prealbumin and RBP levels) and did not show any differences as
to the analyzed clinical parameters.
The data obtained in this study do not show major benefits from
the use of exogenous human albumin in the TPN of critically ill
newborns. Therefore, its use is not recommended unless the aim is
to exclusively increase serum albumin levels.
Acknowledgments
Thanks to the Division of Pharmacy, TPN unit of Hospital Universitário
de Santa Maria - HUSM-RS and also thanks to Isabel Machado Souza
(Laboratory of Pediatrics HCRP-USP) for her help and assistance.
|
Introduction
Help menu
