Nutrient Metabolism

6 downloads 0 Views 232KB Size Report
d gestation) and fed total parenteral nutrition (TPN) or enteral sow's milk (ENT) for 6 d after .... dorsal aorta catheter as above, and an orogastric feeding tube (6F;.
Nutrient Metabolism

Preterm Birth Affects the Intestinal Response to Parenteral and Enteral Nutrition in Newborn Pigs1,2 Per T. Sangild,*3 Yvette M. Petersen,* Mette Schmidt,* Jan Elnif,* Thomas K. Petersen,† Randal K. Buddington,** Gorm Greisen,‡ Kim F. Michaelsen†† and Douglas G. Burrin‡‡ *Divisions of Nutrition and Reproduction, Royal Veterinary and Agricultural University, Frederiksberg DK-1870, Denmark; †Leo Pharmaceutical Products, Department of Pharmacology, Ballerup DK-2750, Denmark; **Department of Biological Sciences and the College of Veterinary Medicine, Mississippi State University, Starkville, MS 39762; ‡Neonatology Clinic, State University Hospital (Rigshospitalet), Copenhagen DK-2100, Denmark; ††Research Department of Human Nutrition, Royal Veterinary and Agricultural University, Frederiksberg, Denmark; and the ‡‡U.S. Department of Agriculture/Agricultural Research Service Children’s Nutrition Research Center, Houston, TX 77030 ABSTRACT Maturation of gastrointestinal (GI) function in neonates is stimulated by enteral nutrition, whereas parenteral nutrition induces GI atrophy and malfunction. We investigated whether preterm birth alters the GI responses to parenteral and enteral nutrition. Pigs were delivered either preterm (107 d gestation) or at term (115 d gestation) and fed total parenteral nutrition (TPN) or enteral sow’s milk (ENT) for 6 d after birth. Immaturity of the preterm pigs was documented by reduced blood pH, oxygen saturation and neutrophil granulocyte function, impaired intestinal immunoglobulin G uptake from colostrum, and altered relative weights of visceral organs (small intestine, liver, spleen, pancreas, and adrenals). For both ages at delivery, increases occurred in pancreatic weight (30 –75%) and amylase activity (0.5- to 13-fold) after birth, but much more in ENT than in TPN pigs (P ⬍ 0.05). Six days of TPN feeding was associated with reduced intestinal weight for both delivery groups (60% of values in ENT, P ⬍ 0.001), but only in term TPN pigs was the weight lower than at birth (⫺20%, P ⬍ 0.05). Likewise, it was only in term TPN pigs that intestinal maltase activity increased, compared with ENT, and the absorption of glucose and proline decreased. Only in preterm pigs did TPN feeding increase lactase activity (⫹50% compared with ENT, P ⬍ 0.05). For both delivery ages, the mRNA of lactase-phloridzin hydrolase and sodium-coupled glucose transporter 1 were increased in TPN, compared with ENT. In conclusion, the trophic effect of enteral vs. parenteral nutrition on the GI tract is also present after preterm birth, but the postnatal maturation of many GI functions is modified, compared with term birth. The effects of nutritional regimen on the maturation of the gut epithelium in neonates depend on gestational age at birth. J. Nutr. 132: 2673–2681, 2002. KEY WORDS:



ontogeny



birth



nutrient absorption

In many species, including humans, the gastrointestinal tract (GIT)4 undergoes marked structural and functional maturation in the periods immediately before and after birth (1– 8). Correspondingly, preterm birth is associated with a variable degree of GIT immaturity and intolerance to oral



brush-border enzymes



development

feeding. Preterm neonates, thus, may require a period of total parenteral nutrition (TPN) before enteral nutrition is administered (9 –11). The intensity of care is inversely related to the length of gestation and degree of organ immaturity at birth. The onset of enteral nutrition at birth is a key signal for the marked maturation of the GIT in term newborn animals (4,7,8,12). Conversely, administering TPN to neonates induces intestinal atrophy and may reduce the digestive and absorptive capacity (13–16). It is not known whether maturity at birth affects the responsiveness of the intestine to nutrition. In guinea pigs, intestinal growth after 24 h of enteral feeding is greater in prematurely delivered animals (93% gestation) than in those delivered at term (17). These results suggest that the intestinal trophic response to enteral feeding may be influenced by gestational age at birth. One of the limitations in our understanding of prematurity and intestinal function is the lack of an appropriate animal model. Whereas primates and guinea pigs are viable from ⬃70% gestation, species such as pigs, dogs and rats are not

1 Supported by the Danish Agricultural and Veterinary Research Council, Programme 9702803. 2 A preliminary account of the data has been published: Sangild, Petersen, Y. M., Elnif, J., Schmidt, M., Buddington, R. K. & Burrin, D. G. (2000) Premature and term newborn pigs differ in their intestinal response to parenteral and enteral nutrition. Gastroenterology 118: A76. 3 To whom correspondence should be addressed. E-mail: [email protected]. 4 Abbreviations used: ApA, aminopeptidase A; ApN, aminopeptidase N; DPP IV, dipeptidyl peptidase IV; ENT, enteral nutrition; GI, gastrointestinal; GIT, gastrointestinal tract; IgG, immunoglobulin G; LPH, lactase-phloridzin hydrolase; LS, least square; O2ct, oxygen content; O2sat, oxygen saturation; pCO2, partial carbon dioxide pressure; PMN, polymorphonuclear neutrophil granulocytes; pO2; partial oxygen pressure; RT-PCR, reverse transcription polymerase chain reaction; SGLT-1, sodium-coupled glucose transporter 1; TPN, total parenteral nutrition.

0022-3166/02 $3.00 © 2002 American Society for Nutritional Sciences. Manuscript received 20 March 2002. Initial review completed 9 April 2002. Revision accepted 18 June 2002. 2673

SANGILD ET AL.

2674

viable until after 90% gestation, even with extensive neonatal care. However, the similarities in body composition and the stage of metabolic and intestinal maturation make the neonatal pig a good nutritional model for premature infants. Despite their potential utility, we are unaware of any studies that have used pigs to investigate whether prematurity affects the gastrointestinal response to enteral nutrition. The aims of this study were to investigate the degree of prematurity in preterm pigs (93% gestation) based on blood chemistry data and GIT function as measured in vitro (brush-border enzyme activities and nutrient absorption), and to clarify whether the gastrointestinal response to enteral and parenteral nutrition is affected by gestational age at birth. MATERIALS AND METHODS Delivery of piglets. Fifteen full-term pigs were obtained from two litters delivered by cesarean section at 114 –115 d of gestation (Large White ⫻ Danish Landrace, term ⫽ 115 ⫾ 2 d). Cesarean section was performed 20 h after induction of parturition by prostaglandin F2␣ (175 ␮g, i.m., cloprostenol, Estrumate; Pitman-Moore, Harefield, UK) to mimic the normal metabolic and endocrine changes in the fetus and pregnant sow associated with spontaneous delivery at term (18,19). Before cesarean section, the sows were sedated with azaperone (0.05 mL/kg, i.m.; Janssen, Beerse, Belgium). Anesthesia was induced with thiopental sodium (5 mg/kg, i.v.; Abbott, North Chicago, IL) and maintained with isoflurane inhalation (1–2% in oxygen; Abbott) after endotracheal intubation. An incision was made in the uterus and each fetus was removed after transecting and ligating the umbilical cord. The sows were killed immediately after removal of all piglets (sodium pentobarbitone, 60 mg/kg, i.v.). The piglets were randomly selected from each of the two litters to receive either parenteral nutrition (TPN; n ⫽ 8) or enteral sow’s milk (ENT; n ⫽ 7) for 6 d after delivery. Seventeen preterm pigs were removed from three litters after delivery by cesarean section (as above) at 107 ⫾ 1 d of gestation, but without previous induction of parturition (elective caesarean section). Previous studies have indicated that few pigs survive if delivery occurs before 90% gestation, even with extensive neonatal care (18,19). Preterm pigs were randomly selected and assigned to the TPN (n ⫽ 9) or ENT (n ⫽ 8) group. Finally, to evaluate the direction of change in physiological variables over the 6-d experiment, a number of newborn premature (n ⫽ 9 from three litters, cesarean section at 107 ⫾ 1 d gestation with previous induction of parturition) and newborn term pigs (n ⫽ 10 from three litters, spontaneous birth at 114 –115 d) were killed for tissue collection immediately after birth. All procedures were approved by the National Committee on Animal Experimentation (Denmark). Catheterizations for blood sampling and feeding. Immediately after delivery, all piglets were placed individually in infant incubators (Air-Shields, Hatboro, PA) with regulated temperature (30 –37°C, as required to maintain normal body temperature) and 80 –100% moisture and extra oxygen supply (only premature pigs, 0.5–2 L/min as required to achieve ⬎92% arterial blood oxygen saturation). Piglets assigned to the TPN groups were fitted with a vascular catheter (infant feeding tube 4F; Portex, Kent, UK) inserted into the dorsal aorta via the transected umbilical cord while they were still anesthetized. After the cord was ligated with a soft cotton thread close to the skin to prevent bleeding, the catheters were secured by sutures to the cord and skin. Pigs assigned to the ENT groups were fitted with a dorsal aorta catheter as above, and an orogastric feeding tube (6F; Portex), which was passed through the cheek and secured to prevent damage by chewing. Finally, an elastic body suit (Danagrib, Copenhagen, Denmark) was fitted onto all pigs to protect the catheters. Nutrient solutions and feeding protocol. The parenteral nutrition solution used for both groups of TPN pigs was prepared aseptically and consisted of a series of commercial products mixed into 3-L TPN bags (Nutrimix; Braun, Melsungen, Germany). The products, nutrient composition and concentrations of macrominerals are given in Table 1. The composition was based on that used previously for 3-

TABLE 1 Nutrient and macromineral content of the TPN solution1 Component

Content unit/L

Total energy, kJ Nonprotein energy, kJ Water, g Glucose, g Lipids, g Nitrogen, g Aminoacids, g Sodium, mmol Potassium, mmol Magnesium, mmol Calcium, mmol Phosphate, mmol Osmolality, mosm

3210 2444 881 71.9 31.1 7.1 45.1 9 24.1 3.2 4.5 14.3 940

1 All concentrations kindly calculated by Fresenius-Pharmacia KABI using standard values. Each 1011 mL of the solution contained 152 mL Intralipid 20%, 400 mL Vamin 18 EF, 6.0 mL Addiphos, 5.7 mL Peditrace, 10.0 mL Vitalipid Infant, 1 vial Soluvit N (all supplied by FreseniusPharmacia KABI), 20.0 mL Calcium-Sandoz (Novartis, Copenhagen, Denmark) and 400 mL glucose (200 g/L), 15.4 mL KCl (1 mmol/L), 1.6 mL MgSO4 (2 mol/L), 0.5 mL ZnSO4 (150 mmol/L) (all supplied by SAD, Copenhagen, Denmark).

to 10-d-old term pigs (20,21) but with reduced concentrations of protein and macrominerals. The nutrient concentrations were determined from preliminary results, which showed that the growth rate and the maximum energy and protein intakes tolerated by premature TPN-fed pigs were considerably less than those tolerated by term TPN-fed pigs, as assessed by blood urea and glucose levels. In the preliminary studies, excessive nutrient intakes were associated with urea levels of 8 –12 mmol/L and glucose levels of 6 – 8 mmol/L (normal ranges: 3– 6 mmol/L for both metabolites). Sow’s colostrum was collected from a number of different sows within 6 h of parturition. Likewise, sow’s milk was collected at 4 –10 d after parturition, and the pooled sow’s milk contained 4950 kJ/L and 51 g protein/L. To adjust the energy and protein concentrations of the sow’s milk match to those of the TPN fluid, it was mixed (50/50, v/v) with skimmed cow’s milk, resulting in 3350 kJ/L and 44 g protein/L in the final modified sow’s milk solution. The TPN solution was infused continuously via the arterial catheter and the milk via the orogastric tube, using automatic infusion pumps (Infusomat Secura; Braun). Feeding began 10 h after birth and the piglets were weighed daily to adjust their nutrient infusion rates. The premature TPN pigs received 580 kJ/kg per day, 8.1 g amino acid kg/d and a fluid intake of 180 mL/kg per day during the 2- to 6-d period. During the initial 2 d of the experiment, they were fed only 50% of this rate. For term pigs, the rate of TPN fluid input was equivalent to 730 kJ/kg per day, 10.3 g amino acid kg/d and a fluid intake of 230 mL/kg per day with an initial adaptation period as above. Both the ENT groups received a continuous orogastric infusion of sow’s colostrum (for 1 d) or the modified sow’s milk (for 5 d) at an hourly rate identical to that of the TPN solution. The nutritional goal was to provide the pigs with sufficient energy and protein to allow for a slightly positive energy balance (e.g., growth rate) that was similar in preterm and term pigs, and in TPN and ENT pigs. No transplacental transport of immunoglobulins occurs in pigs (in contrast to humans) and neonatal pigs depend on colostral immunoglobulins for passive immunization (22). Although the ENT pigs received this immunological protection via colostrum, the TPN-fed piglets were immunized with serum from their mother. Via the arterial catheter, a dose of 4 mL of maternal serum per kilogram of body was injected at 6, 12 and 24 h after birth. The serum was separated aseptically from maternal blood (4000 ⫻ g, 4°C, 10 min) that was collected from a maternal uterine vein at the time of surgery. To further reduce the risk of infections, the pigs were given antibiotics each day (2 mg/kg, enrofloxacin, Baytril; Bayer, Leverkusen,

TPN AND ENTERAL NUTRITION AFTER PREMATURE BIRTH

Germany). Iron dextran was injected subcutaneously on d 1 (80 mg, Ferridex; Rosco, Copenhagen, Denmark) to prevent anemia. Blood sampling and analyses. Arterial blood samples (1.5 mL) were collected from the cord at delivery, 10 h postpartum (just before the beginning of feeding) and each morning thereafter. Blood pH, partial pressures of oxygen and carbon dioxide (pO2, pCO2), hematocrit, glucose and ion levels were all analyzed using an automatic blood gas analyzer (NOVA Stat 5, Waltham, MA). Oxygen saturation (O2sat) and total oxygen content (O2ct) were measured by a hemoximeter (OSM3; Radiometer, Copenhagen, Denmark) and blood urea by reflectance photometry (Reflotron; Roche, Mannheim, Germany). For immunoglobulin G (IgG) and cortisol determinations, blood was collected in EDTA-containing tubes, centrifuged (4000 ⫻ g, 4°C, 10 min) and the plasma used for analysis of IgG by immunoelectrophoresis (22) and of cortisol by ELISA (Biomar Diagnostics, Marburg, Germany). Blood hematology and chemotaxis of polymorphonuclear neutrophil granulocytes. Standard microscopic visualization of blood cells was carried out using Eosin dye on glass slides. The maturity and function of the blood polymorphonuclear neutrophil granulocytes (PMN) was evaluated using blood collected at birth and by the end of the experiment. The chemotactic and spontaneous migratory patterns of the PMN were determined by the under-agarose method (23), using a final agarose concentration of 12 g/L (Litex HSA agarose; Sigma, St. Louis, MO) with gelatin (2.5 g/L; Difco, Detroit, MI) in HEPES buffer (0.1 mmol/L, pH 7.4) and 1X RPMI (no. 22511; Grand Island Biological, Grand Island, NY). Briefly, series of three wells were cut in the gel using a template and a cell suspension of ⬃5 ⫻ 105 cells (10 ␮L) was applied to the middle well. Zymosanactivated serum (10 g/L) was applied to the chemoattractant well and buffer alone was added to the control well. After a 135-min incubation at 37°C, the migration of PMN cells was determined as the mean linear distance from the margin of the sample well toward the chemoattractant and control wells, respectively. Tissue collection. At the end of the feeding period, the pigs were killed (sodium pentobarbitone, 200 mg/kg, i.v.) for organ collection. The small intestine, from the pyloric sphincter to the ileocolonic junction, was rapidly removed by cutting along the mesenteric border and weighed. Intestinal length was measured in a relaxed state on a table top and divided into three segments of equal length, designated proximal, middle and distal small intestine. From the middle of each region, a 10-cm piece was removed for measuring rates of nutrient uptake (see below). A second piece of each intestinal segment and a sample of the pancreas were immediately frozen in liquid nitrogen and kept at ⫺80°C for later determination of enzyme activity and mRNA (see below). A third 10-cm intestinal segment was removed and opened along its length for measurements of intestinal circumference, wet mass and the percentage of mucosa that could be removed by gentle scraping with a plastic slide. The proportion of mucosa was determined on a dry matter basis after drying both the mucosa and the underlying tissues (50°C for 72 h). The nominal surface area of the small intestine (not taking into account changes in villus surface area) was calculated as intestinal length multiplied by mean intestinal circumference. Finally, the wet mass was recorded for the lungs, liver, spleen, heart, adrenals, kidneys, stomach and pancreas. Enzyme activity analyses. The three frozen intestinal samples from each pig were homogenized in 1.0% Triton X-100 (6 mL/g tissue) and the homogenates were assayed for disaccharidase (lactase, maltase and sucrase) and peptidase (aminopeptidases N, aminopeptidase A and dipeptidylpeptidase IV) activities, as described previously (8,24). Frozen pancreas was homogenized in Tris-HCl buffer (100 mmol/L containing 20 mmol/L CaCl2, pH 7.9, 2 min, 0°C), centrifuged (20000 ⫻ g, 45 min, 4°C) and the supernatant used for analysis of amylase, trypsin and chymotrypsin activity. Ethylidene-pnitrophenyl, D-maltoheptaside was used as a substrate for amylase and the liberated glucose determined spectrophotometrically at 405 nm (577-50P; Sigma). Benzoyl-arginine-p-nitroanilide was used as the substrate for trypsin after activation of the trypsinogen with enterokinase (B 4875 and E 0632; Sigma), and chymotrypsin was measured using Succ-AAPF-p-nitroanilide (S 7388; Sigma) as a substrate. For all enzymes, a hydrolytic rate of 1 ␮mol substrate released/min at

2675

37°C was considered to represent 1 U of enzyme activity. Enzyme activities were expressed per gram of wet intestine (proximal, middle and distal) or pancreas. Intestinal nutrient uptake. Three intestinal segments were kept in ice-cold, aerated Ringer’s solutions until 1-cm everted sleeves were mounted on stainless steel rods and incubated for 2 min at 37°C in solutions containing 50 mmol/L nutrient concentrations (12,25,26). The accumulation of nutrients by the mucosa was quantified by adding trace amounts of either 14C-D glucose (ICN Biomedicals, Irvine, CA) or 3H-labeled proline or leucine (New England Nuclear, Boston, MA) as described in detail previously (25). For glucose, uptake rates represent active transport mediated by the apical sodium-dependent glucose transporter 1 (SGLT-1). For proline and leucine, the uptakes include passive influx of amino acids as well as transport mediated by the imino (proline) and neutral (leucine) carrier systems, respectively (25). The results were expressed as uptake rates per mg wet intestinal tissue (nmol/mg per min). mRNA measurements for selected genes. To investigate in more detail the molecular basis for treatment differences in lactase activity and glucose absorption, we quantified the mRNA of lactasephloridzin hydrolase (LPH) and SGLT-1 by reverse transcription polymerase chain reaction (RT-PCR) using pig ␤-actin as a reference. Only tissues from 6-d-old pigs were analyzed. Oligonucleotide primers were designed to identify the pig LPH (sense, 5⬘-CTC AGG TGT ACA AGT TCT-3⬘; annealing temperature, 51 oC) and pig SGLT-1 (sense, 5⬘-CGA AGT ATG GTG TGG TGG CC-3⬘; annealing temperature, 55°C). Individual intestinal tissue samples were then RNA extracted, reverse transcribed into cDNA, and subjected to PCR amplification, as described in detail elsewhere (27). The PCR products were visualized after electrophoresis in agarose gels and their identity was confirmed by entering the sequenced cDNA (TACG, Copenhagen) into BLAST (National Center for Biotechnology Information). The abundance of LPH and SGLT-1 mRNA, relative to ␤-actin mRNA, was determined by optical densitometry reading of the PCR bands on digitalized pictures (BioCapt 97 software; Vilber Lourmat, Cedex, France). Data analyses and calculations. The effects of treatment (TPN and ENT), delivery age (preterm and term), and the interactions were evaluated by ANOVA using a general linear model with sow (litter) as a covariate (28). The analyses of blood data also included sample time (d 0 – 6) as a main effect, whereas for the intestinal results, region (proximal, middle and distal) was added to the model as a main effect. Because piglets from each of the five litters could not be equally distributed across both preterm (three litters) and term (two litters) age at delivery, the effects of gestational age may be confounded with litter effects in the full statistical model. Values presented in tables and figures are presented as means [or least square (LS) means] and SEM as analyzed separately for each gestational age. Significant differences (P ⬍ 0.05) between treatments (within each gestational age at delivery) were identified by the LSD test comparing adjusted LS means. Student’s t test was used for pairwise comparisons between the means in two groups of 6-d-old pigs with different gestational age at delivery, and between 6-d-old pigs and newborn pigs. All organ weights were expressed relative to body weight. In the figures, the postnatal changes in tissue variables, relative to the time of birth, are shown. These ratios were calculated by dividing values in 6-d-old pigs with the mean of newborn pigs (premature or term, respectively). Unless otherwise noted, the intestinal data represent mean values for the proximal, middle and distal small intestine. The total enzymatic or absorptive capacity of the entire small intestine was estimated by multiplying these means (tissue-specific enzyme activity or rates of nutrient uptake) by total intestinal mass normalized to body weight.

RESULTS Clinical observations and blood chemistry values. Within 3– 6 h after placement in the incubators, the term pigs had open eyes, moved normally and exhibited sucking behavior. None of the term pigs exhibited adverse clinical signs during

SANGILD ET AL.

2676

the experiment. One term TPN pig died on d 4 of the experiment because of a catheter leak. The eyes of preterm pigs were closed at delivery and the pigs moved very little for the first 24 –36 h after delivery. Despite transferring the premature pigs to the heated incubators (35–37°C) within 15 min of birth, they exhibited considerable hypothermia (rectal temperatures, 33–36°C) during the first 6 – 8 h after birth, whereafter rectal temperature reached the normal range (38 –39°C). Four of the 17 premature pigs died within 24 h of birth. Symptoms indicative of feed intolerance and intestinal dysmotility (distended bowel, hyperventilation, and no feces after 2–3 d of feeding) occurred in 4 of the 6 ENT premature pigs. The problem and the associated distress either resolved itself spontaneously, or after applying lubricant to the hind gut via the rectum, leading to normal defecation after a meconium plug was voided. No diarrhea was observed over the 6-d experimental period and there were no signs of sepsis. Newborn surviving premature piglets showed variable degrees of acidosis, hypoxia and hypercapnia, compared with newborn term pigs, as indicated by significantly lower arterial blood pH and oxygen saturation and elevated pCO2 values at 10 h after delivery (Table 2, all P ⬍ 0.05). With the extra oxygen supply, these blood gas values were normalized after 1–2 d in the surviving premature pigs. Analyzed across the 2to 6-d period, blood gas values in premature pigs (both TPN and ENT) did not differ from those of term pigs (Table 2). During the 2- to 6-d period, blood sodium and potassium concentrations were significantly lower in premature pigs (141.3 ⫾ 1.1 vs. 146.1 ⫾ 1.1 mmol/L and 4.5 ⫾ 0.1 vs. 4.8 ⫾ 0.1 mmol/L, respectively), although calcium concentrations did not differ (1.43 ⫾ 0.02 vs. 1.43 ⫾ 0.02 mmol/L). Both delivery groups of ENT pigs had higher hematocrits and tended to have lower plasma cortisol concentrations than did their corresponding TPN-fed pigs (P ⬍ 0.10), during the 2- to 6-d period (Table 2). Urea levels were elevated in premature ENT

pigs, and lowered in term ENT pigs, relative to TPN pigs, although blood glucose levels did not differ among the four groups. Both groups of ENT pigs, but particularly the term ENT pigs, had higher circulating IgG levels than the TPN pigs (Table 2). Blood leukocytes from newborn premature pigs were immature as assessed by an elevated frequency of PMN with a hyposegmented nucleus, compared with those in older premature pigs or newborn term pigs. This hematological immaturity was confirmed by a severely reduced chemotactic PMN migration in newborn premature pigs, compared with newborn term pigs, or compared with 6-d-old pigs, both preterm and term (Fig. 1, all P ⬍ 0.001). Differences between delivery ages in the spontaneous migration patterns were similar to the differences in chemotactic migration (Fig. 1). Organ weights and intestinal dimensions. All surviving pigs gained weight with slightly higher gains in both groups of term pigs, compared with the preterm groups (P ⬍ 0.05, Table 3). Premature newborn pigs had greater relative lung, spleen and pancreas weights, and lower liver, heart, adrenal gland and intestinal weights than newborn term pigs (Table 3). The mucosa proportion, circumference and nominal surface area of the small intestine was also lower in premature newborn pigs. At 6 d of age, the organ weight differences between delivery ages had disappeared and there were no consistent differences between nutritional regimens, except for pancreas and intestinal data (Table 3). For both delivery ages, the ENT pigs had markedly greater relative pancreas weight and a greater relative weight, proportion mucosa, circumference and surface area of the small intestine, compared with TPN pigs (Table 3, Fig. 2). Compared with newborn pigs, the term TPN-fed pigs had lower intestinal relative weight, length and surface area, whereas in premature TPN pigs they were not less than those in newborn pigs (Table 3, Fig. 2). Intestinal and pancreatic enzyme activities. At birth, the tissue-specific disaccharidase and peptidase activities did not differ between premature and term pigs. The activities in

TABLE 2 Blood chemistry in newborn and 2– 6 d-old preterm and term piglets given total parenteral nutrition (TPN) or enteral nutrition (ENT)1 Preterm pigs TPN pH

Newborn 2–6 d

7.451 ⫾ 0.00

pCO2, mmHg

Newborn 2–6 d

38.4 ⫾ 0.8

O2sat, %

Newborn 2–6 d

Hematocrit

Newborn 2–6 d

19.9 ⫾ 0.8

Glucose, mmol/L

Newborn 2–6 d

5.2 ⫾ 0.3

Urea, mmol/L

Newborn 2–6 d

4.4 ⫾ 1.5

IgG, g/L

Newborn 2–6 d

1.5 ⫾ 0.3

Cortisol, ␮g/L

Newborn 2–6 d

99.6 ⫾ 0.5

41.9 ⫾ 5.9

Term pigs ENT

7.362 ⫾ 0.0092 7.469 ⫾ 0.012 56.5 ⫾ 1.32 87.5 ⫾ 2.22 0.264 ⫾ 0.012 3.3 ⫾ 0.2 n.d. 0.03 ⫾ 0.02 54.6 ⫾ 5.8

TPN

ENT

7.473 ⫾ 0.008 7.469 ⫾ 0.008 7.445 ⫾ 0.0103 39.1 ⫾ 1.2

35.5 ⫾ 1.03

35.4 ⫾ 1.1

96.8 ⫾ 0.63

98.4 ⫾ 0.3

24.8 ⫾ 1.03

0.256 ⫾ 0.013 21.7 ⫾ 0.5 23.9 ⫾ 0.63

5.2 ⫾ 0.3

5.4 ⫾ 0.2

11.5 ⫾ 1.83

6.6 ⫾ 0.5

4.5 ⫾ 0.73

1.6 ⫾ 0.3

32.8 ⫾ 7.5

48.9 ⫾ 3.8

97.9 ⫾ 2.0

3.7 ⫾ 0.3 n.d. 0.03 ⫾ 0.02 67.7 ⫾ 5.2

41.8 ⫾ 1.33 98.1 ⫾ 0.3

4.2 ⫾ 0.33 4.3 ⫾ 0.63 16.5 ⫾ 2.03 35.8 ⫾ 4.13

1 Values are least square means ⫾ SEM, n ⫽ 6 –14. n.d., not determined. IgG and cortisol values in newborn pigs were measured at delivery, while other newborn values were measured at 10 h postpartum. 2 Newborn premature pigs differ from newborn term pigs (P ⬍ 0.05). 3 2– 6 d-old ENT pigs differ from the corresponding TPN pigs (P ⬍ 0.05).

TPN AND ENTERAL NUTRITION AFTER PREMATURE BIRTH

2677

FIGURE 2 Pancreas and intestinal weights in 6-d-old preterm and term piglets administered total parenteral nutrition (TPN) or enteral nutrition (ENT). Values are expressed relative to the corresponding mean weight measured for newborn pigs (⫽1.0) and are least square means ⫾ SEM, n ⫽ 7–9. *Greater than TPN, P ⬍ 0.05.

IV). For lactase, the activity was reduced in premature 6-d-old pigs, relative to values at birth, and most dramatically in the premature ENT pigs (21.7 ⫾ 2.4 vs. 34.3 ⫾ 4.0 U/g in TPN pigs, P ⬍ 0.05, Fig. 3). The most marked difference was present in the proximal intestine (23.4 vs. 54.0 U/g, P ⬍ 0.001). Activities of maltase (Fig. 3) and sucrase (data not shown) were unchanged relative to birth values in both groups of 6-d-old premature pigs, although they increased sharply in both groups of term pigs (100 –200% increases, P ⬍ 0.05). For maltase this increase was less pronounced for ENT than for TPN pigs (Fig. 3). Postnatal development of activities for the three peptidases was dependent on the combined effects of nutritional regimen (TPN vs. ENT) and gestational age at delivery (preterm or term). Intestinal ApA and DPPIV activities (Fig. 3) were higher after ENT than after TPN feeding, and these differences were most pronounced in the distal small intestine (80 –160% increases in ENT vs. TPN, P ⬍ 0.01). Analyzed across the nutritional treatments, the preterm 6-d-old pigs had higher ApA and lower DPP IV activities, than 6-d-old term pigs (30 – 40% differences, P ⬍ 0.05). After birth, the intes-

FIGURE 1 Chemotactic and spontaneous migration of polymorphonuclear neutrophil (PMN) blood granulocytes in newborn (0 d) and 6-d-old preterm and term pigs (pooled data for pigs administered total parenteral nutrition and enteral nutrition, respectively). Values are means ⫾ SEM, n ⫽ 7–13. *Lower than preterm 6-d-old pigs, and lower than both groups of term pigs, P ⬍ 0.01.

6-d-old pigs, therefore, were related to a pooled mean for tissue-specific activity in newborn pigs (both preterm and term), which were (U/g intestine, n ⫽ 19): 44.1 ⫾ 2.5 (lactase), 1.58 ⫾ 0.09 (maltase), 0.29 ⫾ 0.02 (sucrase), 8.21 ⫾ 0.47 (ApN), 2.30 ⫾ 0.15 (ApA), and 3.12 ⫾ 0.20 (DPP

TABLE 3 Body weight and relative organ weights in preterm and term piglets administered total parenteral nutrition (TPN) or enteral nutrition (ENT) for 6 d after birth1 Preterm pigs Treatment Body weight, kg Daily gain, g/d Lung weight, g/kg Liver weight, g/kg Spleen weight, g/kg Heart weight, g/kg Adrenal weight, mg/kg Kidney weight, g/kg Stomach weight, g/kg Intestinal length, cm/kg Proportion mucosa, % Circumference, mm Surface area, cm2

Term pigs

Newborn

6 d, TPN

6 d, ENT

Newborn

6 d, TPN

6 d, ENT

1183 ⫾ 35 — 20.5 ⫾ 0.92 21.7 ⫾ 1.02 1.52 ⫾ 0.082 6.76 ⫾ 0.162 113 ⫾ 182 8.28 ⫾ 0.67 4.18 ⫾ 0.14 242 ⫾ 21 67.2 ⫾ 1.52 7.2 ⫾ 0.22 216 ⫾ 202

1537 ⫾ 93 30 ⫾ 6 17.3 ⫾ 0.9 33.5 ⫾ 1.8 1.87 ⫾ 0.22 7.66 ⫾ 0.23 139 ⫾ 15 8.89 ⫾ 0.56 4.86 ⫾ 0.25 226 ⫾ 12 53.5 ⫾ 1.3 8.7 ⫾ 0.2 285 ⫾ 9

1463 ⫾ 40 26 ⫾ 6 16.4 ⫾ 0.9 30.1 ⫾ 2.1 1.66 ⫾ 0.15 7.44 ⫾ 0.28 156 ⫾ 12 10.13 ⫾ 0.79 5.01 ⫾ 0.23 263 ⫾ 113 62.5 ⫾ 1.23 10.9 ⫾ 0.33 417 ⫾ 103

1282 ⫾ 43 — 15.3 ⫾ 0.9 30.2 ⫾ 0.8 1.11 ⫾ 0.06 7.56 ⫾ 0.29 199 ⫾ 7 7.75 ⫾ 0.25 4.11 ⫾ 0.13 271 ⫾ 11 71.9 ⫾ 1.2 8.8 ⫾ 0.2 309 ⫾ 25

1641 ⫾ 30 48 ⫾ 4 15.1 ⫾ 0.5 31.0 ⫾ 1.3 1.47 ⫾ 0.09 6.60 ⫾ 0.22 151 ⫾ 8 8.26 ⫾ 0.33 4.33 ⫾ 0.15 225 ⫾ 14 62.8 ⫾ 1.2 8.8 ⫾ 0.3 242 ⫾ 9

1699 ⫾ 33 60 ⫾ 4 16.5 ⫾ 0.5 27.2 ⫾ 1.03 1.68 ⫾ 0.11 7.35 ⫾ 0.54 158 ⫾ 14 7.53 ⫾ 0.31 4.76 ⫾ 0.20 254 ⫾ 16 67.9 ⫾ 1.13 12.1 ⫾ 0.43 395 ⫾ 133

1 Values are means ⫾ SEM, n ⫽ 6 –10. 2 Newborn premature pigs differ from newborn term pigs (P ⬍ 0.05). 3 Six d-old ENT pigs differ from the corresponding TPN pigs (P ⬍ 0.05).

2678

SANGILD ET AL.

FIGURE 4 Pancreatic amylase and trypsin activities in 6-d-old preterm and term piglets administered total parenteral nutrition (TPN) or enteral nutrition (ENT). Values are expressed relative to the corresponding mean activity measured for newborn pigs (⫽1.0) and are least square means ⫾ SEM, n ⫽ 7–9. *Different from TPN, P ⬍ 0.05.

FIGURE 3 Intestinal lactase, maltase, aminopeptidase A (ApA) and dipeptidylpeptidase IV (DPP IV) activities in 6-d-old preterm and term piglets administered total parenteral nutrition (TPN) or enteral nutrition (ENT). Values are expressed relative to the corresponding mean activity measured for newborn pigs (⫽1.0) and are least square means ⫾ SEM, n ⫽ 7–9. *Different from TPN, P ⬍ 0.05.

tinal ApN activity (data not shown) decreased in both groups of premature pigs (⫺25%, P ⬍ 0.05) and term pigs (⫺60%, P ⬍ 0.05), with no effect of nutritional treatment. When the mean tissue-specific activity for each of the three peptidases was multiplied by total intestinal weight to estimate the total enzymatic capacity of the small intestine, ENT pigs had much higher total activity than TPN pigs, and the differences between treatments were higher for premature pigs (100 –350% differences, P ⬍ 0.0001) than for term pigs (50 –150% differences, P ⬍ 0.001). The amylase and trypsin activities were lower in newborn preterm pigs compared with newborn term pigs (11.8 ⫾ 1.4 vs. 25.6 ⫾ 5.1 and 1.8 ⫾ 0.2 vs. 3.3 ⫾ 0.5 U/g, respectively, P ⬍ 0.05). The tissue-specific activity of both enzymes increased after birth (Fig. 4), whereas the chymotrypsin activity decreased to 15– 40% of values at birth (P ⬍ 0.05, data not shown). For both delivery ages, ENT feeding stimulated amylase activity (150 –350% increase relative to TPN, P ⬍ 0.05) and reduced the trypsin and chymotrypsin activities (P ⬍ 0.05). For trypsin, this reduction was most pronounced in term pigs (20% in ENT vs. TPN, P ⬍ 0.001, Fig. 4). Intestinal nutrient absorption. At delivery, the total intestinal glucose uptake capacity in premature pigs was only 30% of that in term pigs as a result of the combined effects of lowered intestinal mass (Table 3) and immature tissue-specific glucose uptake (2.13 ⫾ 0.24 vs. 4.88 ⫾ 0.51 nmol mg/min, P ⬍ 0.001). In premature pigs, glucose uptake did not differ between birth and 6 d, while it decreased postnatally in term pigs, most markedly in the TPN group (2.77 ⫾ 0.20 vs. 3.82 ⫾ 0.19 nmol/mg per min, TPN vs. ENT, P ⬍ 0.001, Fig. 5). Tissue-specific amino acid uptake did not differ in preterm and term pigs at birth (leucine: 2.58 ⫾ 0.09 and proline: 3.49

⫾ 0.16 nmol/mg per min, n ⫽ 19), and increased in both after birth (Fig. 5, only proline data shown). Similar to the results for glucose uptake, TPN-fed term (but not preterm) pigs had reduced uptake of proline (P ⬍ 0.05; Fig. 5) and tended to have lower leucine uptake (3.13 ⫾ 0.13 vs. 3.45 ⫾ 0.13 nmol/mg per min, P ⫽ 0.10), relative to ENT pigs. These results, coupled with the TPN-induced reduction in intestinal tissue mass (Fig. 2, Table 3), resulted in markedly lower total intestinal uptakes of glucose and proline in term TPN pigs (46 –52% of values in ENT pigs, P ⬍ 0.001). In premature pigs, TPN was associated with lower total proline uptake (P ⬍ 0.05) and tended to show lowered glucose uptake (P ⫽ 0.20), compared with ENT pigs, because of the decrease in intestinal mass. LPH and SGLT-1 mRNA expression. In both preterm and term 6-d-old pigs, LPH and SGLT-1 mRNA levels were reduced in ENT pigs relative to TPN (P ⬍ 0.05). The effects were most pronounced for term pigs in the distal small intestine (Fig. 6). DISCUSSION The changes in intestinal function and development during the perinatal period have a critical influence on the clinical care and outcome of infants, especially those born prematurely. However, the study of how prematurity affects the

FIGURE 5 Intestinal uptake rates for glucose and proline in 6-dold preterm and term piglets administered total parenteral nutrition (TPN) or enteral nutrition (ENT). Values are expressed relative to the corresponding mean uptake rate measured for newborn pigs (⫽1.0) and are least square means ⫾ SEM, n ⫽ 7–9. *Different from TPN, P ⬍ 0.05.

TPN AND ENTERAL NUTRITION AFTER PREMATURE BIRTH

FIGURE 6 The relative mRNA abundance for lactase-phloridzin hydrolase (LPH) and sodium-coupled glucose transporter (SGLT-1) in the distal small intestine of term 6-d-old pigs administered total parenteral nutrition (TPN) or enteral nutrition (ENT). Values are expressed relative to ␤-actin mRNA and are means ⫾ SEM, n ⫽ 7. *Lower than TPN, P ⬍ 0.05.

responsiveness of the perinatal intestine has been limited by the availability of appropriate animal models. Enteral nutrition is arguably the most potent stimulus for postnatal growth and maturation of the gut. Moreover, the tolerance to enteral nutrition is a major determinant of the clinical care of premature infants. Therefore, the first aim of this study was to establish the viability and extent of intestinal development in piglets delivered prematurely. Our results indicate that the premature piglet at 93% of gestation is viable with appropriate clinical care and that there are demonstrable signs of prematurity in some organ functions, including the small intestine. Our second aim was to determine whether the stage of prematurity affects the trophic and maturational response of the GIT to enteral feeding. In this regard, our findings suggest that the intestinal and pancreatic trophic response is not affected by prematurity; however, the maturation of some intestinal functions is altered by prematurity. The signs of immature organ function in premature cesarean-delivered pigs were most evident and mortality highest during the first 2 d after premature birth. During this period, the blood chemistry indicated variable degrees of acidosis, hypoxia and hypercapnia probably related to immature lung function. The pigs also exhibited a variable degree of hypothermia during the first 6 – 8 h after the premature delivery by cesarean section. We observed no signs of sepsis although the immature neutrophil function would predispose preterm pigs to infection, similar to preterm infants (29). Probably, the provision of passive immunity (IgG) to both parenterally and enterally fed premature pigs minimized this risk. The lower blood IgG levels in colostrum-fed premature pigs compared with term pigs are consistent with earlier reports (22,30) and reflect immature endocytotic function of intestinal enterocytes (31). Many of the measured organ variables in premature newborn pigs were significantly different from those in term pigs (e.g., relative organ weights, enzyme activities and nutrient absorption) reflecting disproportionate organ growth and a very rapid maturation of organ structure and function in the

2679

prenatal period. In this study, the body weight of premature pigs at birth was only slightly reduced, compared with that of term pigs, although the organ to body weight ratios were significantly lower for liver, heart, adrenals and small intestine, and elevated for lungs, spleen and pancreas. Consistent with earlier studies (18,24,25,30 –33), many of the functional variables of the GIT measured at birth were affected by premature delivery. Unlike this study, however, lower body weight and reduced activities of lactase and aminopeptidases have also been reported for pigs delivered at 93% gestation (24). Hence, both body weight and GIT maturation vary considerably at the same postconceptual age during late gestation in pigs. The fact that values measured in 6-d-old premature pigs generally differed from values measured in newborn term pigs (similar postconceptual ages) suggest that factors other than postconceptual age exert major effects on organ growth and function in the perinatal period. In many species, including pigs, a rapid rise in fetal adrenal growth and cortisol production stimulate prepartum maturational events (24,31–33). Because the prepartum cortisol rise in pigs does not begin until d 100 of gestation and is accelerated by parturition (34), premature cesarean-delivered pigs would be exposed to high cortisol levels for a shorter period than those delivered at term. A reduced period of cortisol stimulation of fetal organ development may, therefore, be a key element in the impaired organ function in premature newborns, at least in pigs. In the postnatal period, TPN-fed pigs had slightly higher cortisol concentrations than ENT pigs and we can, therefore, not exclude that cortisol effects contributed to the differences in pancreatic and intestinal function between 6-d-old TPN and ENT pigs. The nutrient composition of the TPN solution was formulated to meet the basic needs of premature pigs. More studies are clearly required to test whether postnatal growth of premature pigs can be enhanced by further optimization of the nutritional regimen. We aimed to achieve only a moderate body growth (50 – 80 g/d) after the initial days of adaptation to the TPN regimen. This may better reflect the common clinical situation for premature infants than the normal growth rate of sow-reared term pigs during the first postnatal week (100 –150 g/d). In our preliminary studies, elevated levels of blood glucose and urea frequently occurred in TPN-fed premature pigs, while term pigs tolerated a higher nutrient and fluid intakes. Nutrient intake was restricted for term pigs (⫹25% relative to that for preterm pigs) so that term pigs would not gain much more weight than preterm pigs. In agreement with Wykes et al. (20), we observed a tendency for the liver to be enlarged in response to TPN-feeding (probably due to cholestasis), but the relative weights of all other organs, apart from the small intestine and pancreas, did not differ between TPN- and enterally fed pigs. For the two enteral feeding groups, a modified sow’s milk was used because artificial formulas or TPN fuel mixes differ from natural milk in the pigs’ metabolic and intestinal responses after enteral administration (4,8,35,36). The developing intestine is very sensitive to the nature of the nutritional stimuli, and we can, therefore, not exclude that the observed postnatal changes in organ function and the differences between treatments (TPN and ENT) are largely dependent on the chosen nutritional regimen. The premature intestine may also be more sensitive to the nature of the enteral diet than the mature intestine. We have recently observed that if newborn pigs are fed a milk replacer in stead of sow’s colostrum, only premature pigs develop severe mucosal atrophy and dysfunction in the immediate postnatal period (8,37). This syndrome appears similar to necrotizing enterocolitis in premature infants.

2680

SANGILD ET AL.

The premature pigs were less sensitive to the TPN-induced intestinal atrophy than term pigs, in which TPN caused significant losses of intestinal mass and surface area (⫺20%). However, enteral milk nutrition induced large increases in intestinal growth, relative to TPN, for both premature and term pigs. Consequently, the total intestinal enzymatic and absorptive capacities (intestinal mass multiplied with tissue specific function) were significantly elevated for most investigated variables in enterally fed pigs, compared with TPN-fed pigs (50 –200% increases). Also in pig fetuses, enteral infusion of milk diets exerts trophic effects on the developing GIT (38), and fetuses prevented from fetal swallowing for 2 wk during late gestation show significantly reduced intestinal mass and digestive capacity (31). The marked trophic response of the newborn GIT to enteral nutrition is, therefore, not entirely birth-dependent, although parturition and birth per se stimulate the maturation of GI function in newborn pigs (30,32). Enteral nutrition induced both increases and decreases in the tissue-specific activity of enzymes, relative to TPN, and the effects were dependent on gestational age at birth. For intestinal nutrient uptakes, enteral nutrition generally stimulated tissue-specific uptake rates, but only in term pigs. These effects indicate that enteral nutrition induces selective effects on gene expression and/or the post-translational processing and activity of individual functional proteins. The complexity of the regulatory pathways is emphasized by our finding that enteral nutrition decreased LPH and SGLT-1 mRNA without corresponding decreases in lactase activity or glucose absorption rates, at least not in term pigs. Enteral nutrition appears to be able to maintain, or even increase, the biological activity, or the insertion of certain functional proteins into the brush-border membrane (LPH and SGLT-1), despite the fact that it down-regulates the intracellular transcription or decreases the stability of the mRNA for the corresponding proteins. Additional mRNA analyses have shown that such discrepancies between biological activity and mRNA levels in response to enteral nutrients are present for enzymes other than LPH (e.g., maltase-glucoamylase and peptidases) (39). The signals for the up- or down-regulation of each mRNA and enzyme activity in response to enteral food may include humoral factors because enteral nutrition has pronounced effects on organs distant to the gut (e.g., the pancreas) and on GIT regions with limited exposure to enteral food components (e.g., the distal intestine). The pancreas appeared to be less sensitive than the small intestine to TPN-induced growth retardation. Nevertheless, enteral nutrition was associated with significantly higher relative pancreas weights than TPN. Enteral food intake is associated with changes in the storage and synthesis of individual enzymes compared with TPN (40,41). The present study shows that such differential regulation of pancreatic enzymes (e.g., amylase and trypsin/chymotrypsin) by enteral nutrients relative to TPN is present in both premature and term newborn pigs. Lactase activity and glucose absorption were two variables that differentially responded to enteral and parental nutrition according to gestational age at delivery. In pigs, these two variables show more pronounced age- and region-specific variation in the perinatal period than most other intestinal parameters (8,12,24,25,42). Consistent with this, the capacities to hydrolyze lactose and absorb glucose increase significantly with gestational age at birth for infants (5,6). Possibly, intestinal functions, which show the most marked developmental changes during the perinatal period, are those for which the responses to enteral nutrition are most dependent on gesta-

tional age at delivery. Such selective regulation of functional GIT proteins by luminal nutrients may involve both differential modulation of gene expression and post-translational influences on the individual proteins. Despite the physiological differences between preterm- and term-delivered pigs, the GI response to parenteral vs. enteral nutrition generally showed the same direction and magnitude. Even very immature newborns may therefore benefit from the tropic and maturational effects of enteral feeding on the GIT. Further studies on preterm animals may help to clarify whether the optimal adaptation of individual organ systems to extrauterine life requires clinical interventions that should vary according to gestational age at birth. ACKNOWLEDGMENTS The expert technical assistance of Bente Synnetsvedt, Inger Heintze, Anny Pedersen and Anna Sierkierska is gratefully acknowledged. The Department of Pharmacy at Righospitalet, Copenhagen (Susanne Gerlac) is gratefully acknowledged for help in the preparation of the TPN solutions. Finally, we thank Joe Vestergaard, Braun, Copenhagen, for his generous help in maintaining the TPN infusion equipment.

LITERATURE CITED 1. Grand, R. J., Watkins, J. B. & Torti, F. M. (1976) Development of the human gastrointestinal tract: a review. Gastroenterology 70: 790 – 810. 2. Henning, S. J., Rubin D. C. & Shulman, R. J. (1994) Ontogeny of the intestinal mucosa. In: Physiology of the Gastrointestinal Tract. 3rd ed. (Johnson, L. R., ed.), pp. 571– 610. Raven Press, New York, NY. 3. Montgomery, R. K., Mulberg, A. E. & Grand, R. J. (1999) Development of the human gastrointestinal tract: twenty years of progress. Gastroenterology 116: 702–731. 4. Schwarz, S. M. & Heird, W. C. (1994) Effects of feeding on the small intestinal mucosa of beagle pups during the first 5 d of life. Am. J. Clin. Nutr. 60: 879 – 886. 5. Shulman, R. J., Schandler, R. J., Lau, C., Heitkemper, M., Ou, C. N. & Smith, E. O. (1998) Early feeding, feeding tolerance and lactase activity in preterm infants. J. Pediatr. 133: 645– 649. 6. Shulman, R. J. (1999) In vivo measurements of glucose absorption in preterm infants. Biol. Neonate 76: 10 –18. 7. Widdowson, E. M., Colombo, V. E. & Artavanis, C. A. (1976) Changes in the organs of pigs in response to feeding for the first 24 h after birth: the digestive tract. Biol. Neonate 28: 272–281. 8. Jensen, A. R., Elnif, J., Burrin, D. G. & Sangild, P. T. (2001) Intestinal immunoglobulin absorption and enzyme activity in neonatal pigs fed colostrum, plasma or milk replacer. J. Nutr. 131: 3259 –3265. 9. Berseth, C. L. (1985) Minimal enteral feeding. Clin. Perinatol. 22: 195– 205. 10. Heird, W. C. & Gomez, M. R. (1994) Total parenteral nutrition in necrotizing enterocolitis. Clin. Perinatol. 21: 389 – 409. 11. Heird, W. C. & Gomez, M. R. (1996) Parenteral nutrition in low-birthweight infants. Annu. Rev. Nutr. 16: 471– 499. 12. Zhang, H., Malo, C., Boyle, C. R. & Buddington, R. K. (1998) Diet influences development of the pig (Sus scrofa) intestine during the first 6 hours after birth. J. Nutr. 128: 1302–1310. 13. Bertolo, R. F., Pencharz, P. B. & Ball, R. O. (1999) A comparison of parenteral and enteral feeding in neonatal piglets, including an assessment of the utilization of a glutamine-rich, pediatric elemental diet. J. Parenter. Enteral Nutr. 23: 47–55. 14. Burrin, D. G., Stoll, B., Jiang, R., Chang, X., Hartmann, B., Holst, J. J., Greeley, G. H. & Reeds, P. J. (2000) Minimal enteral nutrient requirements for neonatal intestinal growth in piglets: how much is enough? Am. J. Clin. Nutr. 71: 1603–1610. 15. Dudley, M. A., Wykes, L. J., Dudley, A. W., Jr., Burrin, D. G., Nichols, B. L., Rosenberger, J., Jahoor, F., Heird, W. C. & Reeds, P. J. (1998) Parenteral nutrition selectively decreases protein synthesis in the small intestine. Am. J. Physiol. 274: G131–G137. 16. Burrin, D. G., Stoll, B., Jiang, R., Petersen, Y., Elnif, J., Buddington, R. K., Schmidt, M., Holst, J. J., Hartmann, B. & Sangild, P. T. (2000) GLP-2 stimulates intestinal growth in premature TPN-fed pigs by suppressing proteolysis and apoptosis. Am. J. Physiol. 279: G1249 –G1256. 17. Thompson, J. F., Singh, M., Wang, Y., Zucker, C. & Heird, W. C. (1986) Developmental differences in the effect of natural feeding on early enteric mucosal growth of guinea pigs. J. Pediatr. Gastroenterol. Nutr. 5: 643– 647. 18. Sangild, P. T., Silver, M., Schmidt, M. & Fowden, A. L. (1996) The perinatal pig in pediatric gastroenterology. In: Advances in Swine in Biomedical

TPN AND ENTERAL NUTRITION AFTER PREMATURE BIRTH Research (Tumbleson, M. E. & Schnook, L., eds.), pp. 745–756. Plenum Press, New York, NY. 19. Silver, M., Comline, R. S. & Fowden, A. L. (1983) Fetal and maternal endocrine changes during the induction of parturition with the PGF analogue, cloprostenol, in chronically catheterized sows and fetuses. J. Dev. Physiol. 5: 307–321. 20. Wykes, L. J., Ball, R. O. & Pencharz, P. B. (1993) Development and validation of a total parenteral nutrition model in the neonatal piglet. J. Nutr. 123: 1248 –1259. 21. Park, Y. K., Monaco, M. M. & Donovan, S. M. (1998) Delivery of total parenteral nutrition (TPN) via umbilical catheterization: development of a piglet model to investigate therapies to improve gastrointestinal structure and enzyme activity during TPN. Biol. Neonate 73: 295–305. 22. Sangild, P. T., Trahair, J. F., Loftager, M. K. & Fowden, A. L. (1999) Intestinal macromolecule absorption in the fetal pig after infusion of colostrum in utero. Pediatr. Res. 45: 595– 602. 23. Petersen, T. K., Jensen, A. L. & Aaes, H. (1998) Chemotaxis of canine polymorphonuclear neutrophil granulocytes using the under-agarose method applied to glass microscope slides. Comp. Haematol. Int. 8: 31–36. 24. Sangild, P. T., Fowden, A. L., Sjo¨ stro¨ m, H., Nore´ n, O. & Silver, M. (1995) The prenatal development and glucocorticoid control of brush-border hydrolases in the pig small intestine. Pediatr. Res. 37: 207–212. 25. Buddington, R. K., Elnif, J., Puchal-Gardiner, A. A. & Sangild, P. T. (2001) Intestinal apical amino acid absorption during development of the pig. Am. J. Physiol. 280: R241–R247. 26. Karasov, W. H. & Diamond, J. M. (1983) A simple method for measuring solute uptake by intestine in vitro. J. Comp. Physiol. 152: 105–116. 27. Petersen, Y. M., Burrin, D. G. & Sangild, P. T. (2001) GLP-2 has differential effects on small intestine growth and function in fetal and neonatal pigs. Am. J. Physiol. 281: R1986 –R1993. 28. SAS Institute Inc. (1993) SAS/STAT Users’ Guide: Statistics. SAS Institute, Cary, NC. 29. Bektas, S., Goetze, B. & Speer, C. P. (1990) Decreased adherence, chemotaxis and phagocytic activities of neutrophils from preterm neonates. Acta Paediatr. Scand. 79: 1031–1038. 30. Sangild, P. T., Holtug, K., Diernæs, L., Schmidt, M. & Skadhauge, E. (1997) Birth and prematurity influence intestinal function in the newborn pig. Comp. Biochem. Physiol. 118A: 359 –362.

2681

31. Sangild, P. T., Schmidt, M., Elnif, J., Bjo¨ rnvad, C. R., Westro¨ m, B. R. & Buddington, R. K. (2002) Prenatal development of gastrointestinal function in the pig and the effects of fetal esophageal obstruction. Pediatr. Res. 52: (in press). 32. Sangild, P. T., Hilsted, L., Nexø, E., Fowden, A. L. & Silver, M. (1995) Vaginal birth versus elective caesarean section: effects on gastric function in the neonate. Exp. Physiol. 80: 147–157. 33. Sangild, P. T., Westro¨ m, B. R., Fowden, A. L. & Silver, M. (1994) Developmental regulation of the porcine exocrine pancreas by glucocorticoids. J. Pediatr. Gastroenterol. Nutr. 19: 204 –212. 34. Silver, M. (1990) Prenatal maturation, the timing of birth and how it may be regulated in domestic animals. Exp. Physiol. 75: 285–307. 35. Burrin, D. G., Davis, T. A., Ebner, S., Schoknecht, P. A., Fiorotto, M. L., Reeds, P. J. & McAvoy, S. (1995) Nutrient-independent and nutrient-dependent factors stimulate protein synthesis in colostrum-fed newborn pigs. Pediatr. Res. 37: 593–599. 36. Burrin, D. G., Davis, T. A., Ebner, S., Schoknecht, P. A., Fiorotto, M. L. & Reeds, P. J. (1997) Colostrum enhances the nutritional stimulation of vital organ protein synthesis in neonatal pigs. J. Nutr. 127: 1284 –1289. 37. Sangild, P. T., Bjo¨ rnvad, C. R., Petersen, Y. M., Elnif, J. & Burrin, D. G. (2002) Spontaneous NEC-like symptoms in formula-fed premature pigs. Gastroenterology 122: A341 (abs.). 38. Bjo¨ rnvad, C. R., Schmidt, M. & Sangild, P. T. (2002) Premature birth makes the immature small intestine sensitive to diet-induced mucosal atrophy and malfunction. Gastroenterology 122: A784 (abs.). 39. Petersen, Y. M., Schmidt, M., Elnif, E. & Sangild, P. T. (2002) Glucagon like peptide 2 enhances malatase-glucoamylase and sucrase-isomaltase gene expression and activity in parenterally fed premature neonatal pigs. Pediatr. Res. 52: (in press). 40. Rossi, T. M., Lee, P. C. & Lebenthal, E. (1987) Nutrition and exocrine pancreatic function in infancy: effects of total parenteral nutrition and first degree malnutrition. Pancreas 2: 463– 469. 41. Fan, B. G., Salehi, A., Sternby, B., Axelson, J., Lundquist, I., AndrenSandberg, A. & Ekelund, M. (1997) Total parenteral nutrition influences both endocrine and exocrine function of rat pancreas. Pancreas 15: 147–153. 42. Sangild, P. T., Diernæs, L., Christiansen, I. J. & Skadhauge, E. (1993) Intestinal transport of sodium, glucose and immunoglobulin in neonatal pigs: effect of glucocorticoids. Exp. Physiol. 78: 485– 497.