| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Clinical Research Centers |
Divisions of Endocrinology, Departments of Pediatrics and Medicine (W.J.S., L.E.U., D.R.C.), Design and Statistics Unit (L.K.), Frank Porter Child Development Center, University of North Carolina, Chapel Hill, North Carolina 27599
Address all correspondence and requests for reprints to: David R. Clemmons, M.D., CB #7170, University of North Carolina, Chapel Hill, North Carolina 27599.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Serum IGF-I correlated strongly with length of gestation, increasing 4.03 ± 0.95 ng/mL for each additional week of gestation (P < 0.0001) and 0.36 ± 0.07 ng/mL·day each day since birth (P < 0.0001). A higher intake of calories increased IGF-I by 0.07 ± 0.01 ng/mL for each calorie per kg ingested over the previous 3 days (P < 0.0001). IGF-I increased quadratically as protein intake increased. For each change of 1% in calories as protein squared, IGF-I increased 0.36 ± 0.11 ng/mL (P < 0.0001).
Serum IGFBP-3 concentrations also correlated with length of gestation, increasing 25.06 ± 11.83 µg/L·wk (P = 0.035) and 4.14 ± 1.33 µg/·day since birth (P = 0.003). Unlike IGF-I, variation in the amount of protein supplied did not change IGFBP-3. As calorie intake increased, IGFBP-3 increased by 0.54 ± 0.17 µg/L for each calorie per kg consumed over the previous 3 days (P = 0.0015).
In contrast to IGF-I and IGFBP-3, IGFBP-2 declined as the length of gestation increased (56.12 ± 16.92 ng/mL·week; P = 0.001) and with each additional day of life (7.57 ± 2.44 ng/mL·day; P = 0.003). Dietary protein, the predominant regulator of IGFBP-2, caused a decrease of 33.22 ± 9.00 ng/mL with each percent increase in dietary calories as protein (P < 0.0003). Calorie intake had less effect on IGFBP-2 than protein intake.
These results indicate that each of the three peptides studied is regulated in premature infants by nutritional intake, and that their regulatory patterns are qualitatively similar to those observed in older individuals. Measurements of these peptides in premature infants may be useful indicators of nutritional status and adequacy of nutrient intake.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Insulin-like growth factor I (IGF-I) is essential for normal growth. Synthesis, secretion, and blood concentrations of IGF-I are regulated by food intake and change rapidly as a function of changes in nutrient intake. IGF-I is reduced in normal adults by fasting (1, 2) and in adults and children by short term decrements in dietary protein or calories (3, 4). Concentrations in blood are restored toward normal by refeeding. These diet-related changes are detectable within 1–2 days after the diet is altered (1, 3), taking place more rapidly than changes in nutrient-dependent serum proteins such as transferrin, prealbumin, or retinol-binding protein (5).
The bioavailability of IGF-I is regulated by IGF-binding proteins (IGFBPs). Six of these high affinity carriers of IGF-I are present in the circulation and, like IGF-I, are regulated in children and adults by intake of nutrients (3). Serum concentrations of IGFBP-3, the principal carrier of IGF-I, are reduced when the intake of dietary protein or calories is reduced (3, 6) and are restored by refeeding (3, 6). Serum concentrations of IGFBP-2 are also nutritionally controlled; they are elevated during fasting (7) or when dietary protein is restricted (3).
We carried out the study reported here to determine whether peptides of the IGF system provide reliable indicators of nutritional status in premature infants. We studied 50 premature infants prospectively to define the relationship between nutrient intake and serum concentrations of IGF-I, IGFBP-2, and IGFBP-3.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Over a 12-month period, we attempted to recruit all medically stable premature infants weighing 1000–2000 g who were hospitalized in the Newborn Critical Care Center of the North Carolina Children’s Hospital. Because some of these infants had been critically ill before entry into the study, the age at entry varied between 1–77 days. For inclusion, each infant had to be free of any chromosomal, cardiovascular, or gastrointestinal disorder; receiving no ventilatory support; have weight appropriate for gestational age; and not have documented sepsis during the period of observation. We entered 83 infants into the study and followed each until he/she was discharged home (n = 5), transferred back to the referring hospital (n = 27), or became medically unstable and required ventilatory support (n = 1). We report on the 50 infants who successfully completed at least 2 weeks of continuous study.
We determined each infant’s gestational age from the date of the mother’s last menstrual period (n = 12), from early gestation ultrasound examinations (n = 4), by dates of last menstrual period plus ultrasound and/or newborn examination (n = 31), or by postnatal maturity ratings (n = 3). The study was approved by the University of North Carolina institutional committee for the protection of the rights of human subjects, and written informed consent was obtained from the parent(s).
Study design
The care of each infant and his/her feeding regimen were directed by the attending neonatologists. Feeding regimens conformed to established routines for premature infants and included iv hyperalimentation, continuous gastric gavage, or bolus feedings by gastric tube or mouth. Enterally fed infants received infant formula containing 20–30 Cal/30 mL given as continuous gavage feeding, as bolus feeding by orogastric tube, or as oral feedings every 2–3 h. Eighteen infants received iv feedings during a portion of the study and were gradually advanced to enteral feedings as they matured.
Nutritional intake was quantified from the daily log maintained in each infant’s medical record. These records contain the type of enteral and parental formula(e) received and the amount taken at each feeding or per day when given continuously. The supplies of both enteral and parental formula(e) administered were used to calculate energy and protein intake. Loss of food by emesis was subtracted from intake by weighing napkins on which emesis was absorbed. These data were used to calculate the total daily intake of calories and protein, which are expressed in kilocalories per kg/day and as proportion of calories derived from protein. A value of 4 Cal/g protein was used to calculate the amount of energy derived from protein. The weight of each study subject was obtained daily by the nursing staff. Blood for measurement of IGF-I, IGFBP-2, and IGFBP-3 were obtained twice weekly by heel stick unless the infant required phlebotomy for other laboratory studies. In infants who were not receiving continuous feedings, blood was drawn before feeding (usually 2–4 h after the previous feeding).
Laboratory methods
IGF-I was determined by RIA after extraction from serum using C18 Sep-Pak (Waters Associates, Milford, MA) (8). The sensitivity of this IGF-I assay is 5.0 ng/mL, and the intra- and interassay coefficients of variation (CVs) are 3.3% and 5.5%, respectively. IGFBP-2 was determined by RIA (9). The sensitivity of this assay is 0.2 ng/mL, and intra- and interassay CVs are 5.6% and 5.9%, respectively. The IGFBP-3 RIA (10) has a sensitivity of 1.0 µg/L, and intra- and interassay CVs are 4.9% and 5.9%, respectively.
Statistical analyses
Serum levels of IGF-I, IGFBP-2, and IGFBP-3 were modeled as quadratic functions of calorie and protein intakes. Calorie intake was computed by calculating the total calories consumed per kg BW over 3 days before each serum protein measurement. Protein intake was computed as the percentage of the total calories ingested over the prior 3 days. Gestational age, gestational age squared, chronological age, and nutrient phase (fixed or variable nutrient intake) were examined for their possible predictive value as control variables. The squared values of calorie and protein intake were included to allow for the possibility of a quadratic relationship in a linear statistical model. The age squared term was not used because it was redundant as a predictor (age and age squared were collinear). Two-way interactions among calorie intake, protein intake, and each control variable also were tested.
A model suitable for the analysis was found using a backward elimination strategy, working from the largest number of prediction variables to the smallest. All model calculations were performed using SAS PROC MIXED 211. The first step evaluated the interaction terms in the largest model. Any nonsignificant interaction was deleted. An exception to this rule was made for IGFBP-3, where an interaction with a P value of 0.056 was retained. After assessing the importance of the interactions, predictors by themselves were tested. A single variable was retained, whether significant or not, if it was present in a significant interaction with another variable. Similarly, linear terms were always included when the quadratic term was significant. This strategy was applied individually to the response variables, and therefore, the models for the various response variables may differ.
Data from two subgroups based on age at entry into the study were examined for differences. One subgroup was composed of neonates who entered before 12 days of age, and the other was composed of infants beginning the study after 14 days of age. It was found that variables representing differences between the two groups were redundant (collinear) with variables already in the model. Therefore, age at entry does not provide information above and beyond the predictors in the model.
A separate analysis compared subjects who were ingesting progressively more calories and protein with those receiving constant intake. This indicated that for IGF-I and IGFBP-2, a change in food intake had no effect on the form of the model, and the data could be pooled. This was assessed by tests of interactions. There was a small effect on IGFBP-3 with change in food intake, but this did not alter the validity of the analysis of the pooled data set.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Of the 50 study subjects (Table 1
),
21 were enrolled on the first day of life, an additional 15 were
enrolled by the 22nd day after birth, and the remaining 14 subjects
were enrolled between days 22–78 (Fig. 1). The mean duration of study for
individual subjects was 30 days, and only 3 subjects were observed for
the 14-day minimum. One subject was observed for 72 days. The median
gestational age was 31.2 weeks (range, 25–34 weeks).
|
|
Effects of feeding on serum concentrations of IGF-I
Serum concentrations of IGF-I correlated strongly with length of
gestation and days since birth (Table 2).
We observed a significant interaction between protein intake and
gestational age (P < 0.05; Table 2). At greater
gestational ages, the magnitude of the rise in serum IGF-I that
occurred with increasing protein intake was larger (Fig. 2B). Likewise, there was a significant
interaction between the dietary protein intake and the days since birth
(P < 0.001; Table 2); with each additional day after
birth, the mean IGF-I increased.
|
|
). Among the older infants,
the magnitude of the rise in serum IGF-I that occurred with increasing
protein intake was markedly greater than that observed among younger
infants (Fig. 2C). If gestational age, days since birth, and calorie
intake were held constant, IGF-I increased quadratically as the
percentage of intake due to protein increased (Fig. 2A). This
relationship between IGF-I and protein intake was not affected by the
calorie intake (Fig. 2A).
Increases in calorie intake also resulted in increased IGF-I. A mean
increase in IGF-I of 0.07 ± 0.01 ng/mL was observed for each
additional calorie per kg ingested over the previous 3-day period
(P < 0.0001; Table 2). In contrast to protein intake,
the relationship between calorie intake and change in IGF-I was linear
(Fig. 2A).
Effects of feeding on serum concentrations of IGFBP-3
As with IGF-I, serum concentrations of IGFBP-3 correlated with the
length of gestation and days since birth (Table 3). IGFBP-3 exhibited a mean increase of
25.06 ± 11.83 µg/L·week between weeks 24 and 34 of gestation
(P = 0.035; Fig. 3B). Likewise, for each additional day
after birth, the mean increase in IGFBP-3 was 4.14 ± 1.33 µg/L
(P = 0.003; Table 3 and Fig. 3C). Unlike the dramatic effect of
dietary protein on IGF-I, increasing the amount of protein supplied to
the infant increased serum IGFBP-3 modestly (Fig. 3A). The small
P value for the interaction between calorie and protein
intake (P = 0.059; Fig. 3A) suggested that this
increase may have been modified somewhat by total calorie intake (Fig. 3A). Note that the lowest and highest IGFBP-3 values were observed at
the extremes of both calorie and protein intake.
|
|
The regulation of IGFBP-2 in the premature infant contrasted
sharply with the regulation of IGF-I and IGFBP-3. As the length of
gestation increased, serum IGFBP-2 concentrations declined, such that
there was a linear decrease of 56.12 ± 16.92 ng/mL with each
additional week of gestation (P = 0.001; Table 4 and Fig. 4B). Likewise, with each additional day
of age, IGFBP-2 declined 7.57 ± 2.44 ng/mL in a linear fashion
(P = 0.003; Table 4 and Fig. 4C). There was also a
linear decrease in IGFBP-2 as protein intake increased, with each 1%
increase in dietary protein causing IGFBP-2 to decrease by 33.22
± 9.00 ng/mL (P = 0.0003; Fig. 4A). Increasing the
intake of calories also caused a linear decline in IGFBP-2, but to a
lesser degree (decrease of 0.61 ± 0.23 ng/mL for each calorie per
kg increase; P = 0.008; Table 4 and Fig 4A). As in
older individuals (see below), the effect of dietary protein may have
been dominant over that of calories, and no interactions between these
dietary components were observed.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Serum concentrations of IGF-I increase progressively throughout childhood, reaching maximal values around midpuberty (12, 13). The present study suggests that this rise in IGF-I with age is a continuation of events beginning before birth at term, with significantly lower levels being measured in infants born after short gestations than in those born after longer gestations. This observation confirms a previous report (14). IGF-I values also increase in our subjects as a function of age after birth, findings consistent with cross-sectional observations reported previously (15).
In children and adults, IGF-I concentrations in serum are regulated by intake of both calories and protein (3, 16). IGF-I values are low in patients with protein-calorie malnutrition (17, 18), and refeeding, particularly with protein-rich foods, raises the concentration of this peptide (18). The serum concentration of IGF-I in adults declines to approximately 20% of basal values after 10 days of fasting (1). When the calorie intake of children aged 8–11 yr is reduced from 70 to 35 Cal/kg ideal BW·day for 6 days (protein held constant), serum IGF-I concentrations decline by slightly more than 30% (3). Likewise, a reduction in protein from 1 to 0.66 g/kg results in a 17% mean decline in IGF-I in four of six children (3).
Our results show that, as in older individuals, IGF-I values in the serum of premature infants are dependent on the intake of both calories and protein. IGF-I values are more than 50% lower when calorie intake is low (over the range of 230–437 Cal/kg·3 days; 77–146 Cal/kg·day), and this difference is independent of protein intake. Likewise, as the protein in the diet is increased, IGF-I increases quadratically, with a most pronounced effect occurring when 12–15% of dietary calories are derived from protein.
The capacity of dietary protein to increase IGF-I is more attenuated in the most premature infants and in those with the youngest postnatal ages. This effect of dietary intake on IGF-I corresponds temporally to the lag period for growth in newborn premature infants. We speculate that shortened gestation is accompanied by protracted (and perhaps increased) resistance to the action of GH. One of the manifestations of this GH resistance may be attenuation of the diet-induced rise in IGF-I.
Like IGF-I, the concentrations of IGFBP-3 in serum of premature infants are low compared to those in children and adults (14, 19). This is not surprising because the overall age-related pattern of IGFBP-3 in the serum of children is similar to that of IGF-I (20). The effects of diet on IGFBP-3 are also similar to those on IGF-I. IGFBP-3 concentrations in serum are reduced by protein-calorie malnutrition (21) and rise with refeeding, particularly when a protein-rich diet is used (21). IGFBP-3 concentrations in serum are reduced in 8- to 11-yr-old normal children who are subjected to 6 days of dietary caloric reduction (70 Cal/kg reduced to 35 Cal/kg), but no reduction is observed when protein is restricted by 33% (3). Likewise, IGFBP-3 in serum is reduced in malnourished neonatal rats (22), and IGFBP-3 messenger ribonucleic acid (mRNA) is reduced in the liver of dietary protein-restricted rats (23).
The effects of calorie and protein intake on IGFBP-3 in our premature infants are similar in direction and magnitude to those on IGF-I. The interactions of dietary protein with gestational age and postnatal age, however, are much stronger with IGF-I than with IGFBP-3. This suggests that the protein content of the diet is more important in regulating IGF-I. This conclusion is in keeping with the lack of effect on IGFBP-3 observed in children who are subjected to dietary protein restriction while the supply of calories is maintained (3). In view of the decrease in IGFBP-3 mRNA in protein-deprived rats (23), it seems likely that the low serum concentrations in premature infants with lower protein intake is due in part to decreased synthesis of IGFBP-3. Additionally, however, it is possible that IGFBP-3 in the serum of premature infants is subject to proteolysis, as observed in undernourished children (21), children with Laron syndrome (24), severely ill patients (25), patients after surgery (26), and pregnant women (27, 28). The possible involvement of proteolysis in determining the concentrations of IGFBP-3 in our subjects was not pursued in this study.
Unlike other peptides in the IGF system, the level of serum IGFBP-2 is high in infancy and declines with age (10). In our infants, the highest IGFBP-2 values were observed in 24-week gestation premature infants, and there was a linear decrement in values to 34-week gestation infants. Likewise, there was a steady decline in IGFBP-2 with increasing postnatal age. Protein is the crucial nutritional regulator of IGFBP-2 concentrations in serum. This peptide is significantly increased in young children with protein-calorie malnutrition (21) and declines markedly when they are refed a protein-enriched diet (15% of calories is protein), but not when they are refed a diet with only 6% of calories as protein (21). Similarly, in normal children and in adults subjected to dietary protein restriction for 6 days, IGFBP-2 values increased by approximately 40% (3). With caloric restriction, however, IGFBP-2 did not change (3). Our current findings are consistent with those observed previously. Minimal differences in IGFBP-2 are observed in premature infants over a wide range of calorie intake, but there is a greater than 50% decrease in IGFBP-2 as the protein content of the diet increases from 6% to 15%. This effect was similar regardless of gestational age, suggesting that unlike the effect of protein on IGFBP-3 and IGF-I, age is not a determinant of this response. It seems likely that the mechanism for the increased IGFBP-2 in serum that occurs with a reduction in dietary protein is increased synthesis of the peptide. Dietary protein restriction in rats causes marked increases in liver IGFBP-2 mRNA (23).
Our results provide normative data for IGF-I, IGFBP-3, and IGFBP-2 in premature infants and confirm the relationship of calorie and protein intake to the maintenance of serum levels of these peptides. The results also suggest that, as in older individuals (3, 29, 30), measurements of these peptides in premature infants have the potential to be useful indicators of nutritional status and adequacy of nutrient intake. We do not define absolute values that indicate whether nutrient intake is adequate. However, a single measurement of these peptides that deviates greatly below (for IGF-I or IGFBP-3) or above (for IGFBP-2) the mean may be taken as evidence that the diet is likely to be suboptimal. When serial measurements of these peptides do not show improvement once nutritional support has been provided, it seems likely that either the support is not adequate or the patient has excessive utilization/catabolism of food stuffs. Further studies are needed to determine the precise values of IGF-I, IGFBP-3, or IGFBP-2 that predict the sufficiency of the nutrient supply.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Pediatrics, University of Kentucky
Medical Center, Lexington, Kentucky 40536.
Received February 28, 1997.
Revised August 22, 1997.
Accepted September 2, 1997.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. C. R. Prickett, B. Dixon, C. Frampton, T. G. Yandle, A. M. Richards, E. A. Espiner, and B. A. Darlow Plasma Amino-Terminal Pro C-Type Natriuretic Peptide in the Neonate: Relation to Gestational Age and Postnatal Linear Growth J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 225 - 232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Chellakooty, A. Juul, K. A. Boisen, I. N. Damgaard, C. M. Kai, I. M. Schmidt, J. H. Petersen, N. E. Skakkebaek, and K. M. Main A Prospective Study of Serum Insulin-Like Growth Factor I (IGF-I) and IGF-Binding Protein-3 in 942 Healthy Infants: Associations with Birth Weight, Gender, Growth Velocity, and Breastfeeding J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 820 - 826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Balogh, A. Treszl, A. Vannay, and B. Vasarhelyi A Prevalent Functional Polymorphism of Insulin-Like Growth Factor System Is Not Associated With Perinatal Complications in Preterm Infants Pediatrics, February 1, 2006; 117(2): 591 - 592. [Full Text] [PDF]
|
|
|
|
 |
|
 V. Roelfsema, A. J. Gunn, B. H. Breier, J. S. Quaedackers, and L. Bennet The Effect of Mild Hypothermia on Insulin-like Growth Factors After Severe Asphyxia in the Preterm Fetal Sheep Reproductive Sciences, May 1, 2005; 12(4): 232 - 237. [Abstract] [PDF]
|
|
|
|
 |
|
 H.-C. Lo, L.-Y. Tsao, W.-Y. Hsu, C.-Y. Chi, and F.-A. Tsai Changes in Serum Insulin-Like Growth Factors, Not Leptin, Are Associated With Postnatal Weight Gain in Preterm Neonates JPEN J Parenter Enteral Nutr, March 1, 2005; 29(2): 87 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hellstrom, E. Engstrom, A.-L. Hard, K. Albertsson-Wikland, B. Carlsson, A. Niklasson, C. Lofqvist, E. Svensson, S. Holm, U. Ewald, et al. Postnatal Serum Insulin-Like Growth Factor I Deficiency Is Associated With Retinopathy of Prematurity and Other Complications of Premature Birth Pediatrics, November 1, 2003; 112(5): 1016 - 1020. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kajantie, L. Dunkel, E.-M. Rutanen, M. Seppala, R. Koistinen, A. Sarnesto, and S. Andersson IGF-I, IGF Binding Protein (IGFBP)-3, Phosphoisoforms of IGFBP-1, and Postnatal Growth in Very Low Birth Weight Infants J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2171 - 2179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hellstrom, C. Perruzzi, M. Ju, E. Engström, A.-L. Hård, J.-L. Liu, K. Albertsson-Wikland, B. Carlsson, A. Niklasson, L. Sjödell, et al. Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: Direct correlation with clinical retinopathy of prematurity PNAS, May 1, 2001; (2001) 101113998. [Abstract] [Full Text]
|
|
|
|
|
|
A. Hellstrom, C. Perruzzi, M. Ju, E. Engstrom, A.-L. Hard, J.-L. Liu, K. Albertsson-Wikland, B. Carlsson, A. Niklasson, L. Sjodell, et al. Low IGF-I suppresses VEGF-survival signaling in retinal endothelial cells: Direct correlation with clinical retinopathy of prematurity PNAS, May 8, 2001; 98(10): 5804 - 5808. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
