effect of early feeding on maturation of the preterm infant's small intestine

Effect of early feeding on maturation of the preterm infant's small intestine

C. L. Berseth, MD

From the Department of Pediatrics and the Gastroenterology Research Unit, Mayo Clinic, Rochester, Minnesota

To determine the response of the preterm infant's intestine to entire feedings at different postnatal ages, we recorded results of manometry of the gastroduo- denum and determined fasting plasma concentrations of gastrin, gastric inhibitory peptide, neurotensin, and peptide YY three times in each of two groups: 27 preterm infants were randomly assigned to receive hypocaloric enteral nutrition on postnatal days 3 to 5 (early feeding) or on days I0 to 14 (late feeding). Initial observations (study I) were performed by the fifth postnatal day; study 2 was performed on days 10 to 14, and study 3 on days 24 to 28. Early-fed infants received hypocaloric feedings immediately after study 1; late-fed infants did not receive enteral feedings until the complet ion of study 2. Although motor activity and fasting gastrointestinal peptide concentrations did not differ between groups at study 1, at study 2 early-fed infants had significantly more mature motor patterns than did babies not being fed. Early-fed infants also had significantly higher plasma concentrations of gastrin and gastric inhibitory peptide than did late-fed infants; neurotensin and peptide YY values were similar in both groups. By the time of study 3, when late-fed infants had also received enteral feedings, gut development was not different in the two groups. However, early-fed infants were able to tolerate full oral nutrition sooner, had fewer days of feeding intolerance, and had shorter hospital stays. Thus the provision of early hypocalor ic nutrition was associated with earlier nutrition of preterm infants' intestinal function and resulted in improved feeding tolerance. These findings support the use of early feedings in preterm infants. (J PEDIATR 1992;120:947-53)

Tolerance of adequate enteral nutrition in preterm infants is determined by the functional immaturity of the liver, pancreas, and gastrointestinal tract. Moreover, complica- tions attributed to methods of nutrient delivery are thought to contribute to morbidity and death in these infants. Studies investigating the use of various regimens have been limited to indirect measurements of intestinal function, such as somatic growth or days of hospitalization. There is a need for more direct measurements of intestinal function

Supported by grant Nos. HD24558 and DK34988 from the National Institutes of Health, Bethesda, Md. Submitted for publication Sept..13, 1991; accepted Jan. 3, 1992. Reprint requests: C. L. Berseth, MD, Department of Pediatrics, Newborn Section, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030. 9/23/36178

to act as guides to the management of enteral nutrition in preterm infants.

Intestinal motor activity contributes to the aboral movement of foods through the gastrointestinal tract. Intestinal motor activity is commonly monitored manometrically in adults by recording intraluminal pressures; pressure rises when gut muscle contracts and falls when it relaxes, l Low- compliance, continuous-perfusion manometry has been adapted for use in infants. 2-5 The motor patterns differ in preterm and term infants. 2, 4, s At term, the fasting intestine

GIP Gastric inhibitory peptide NT Neurotensin PYY Peptide YY

exhibits all three forms of motor activity of the interdiges- tire cycle typically seen in adults: quiescence (phase I) is

947

948 Berseth The Journal of Pediatrics June 1992

T a b l e I. P a t i e n t c h a r a c t e r i s t i c s

Early-fed infants Late- fed infants (n = 14) (n = 13) p

Gestational age (wk) Birth weight (gm) Postnatal age at study 1 (days) Feeding experience at study 2 (days) Feeding experience at study 3 (days)

Values are expressed as mean + SEM. NS, Not significant.

32 + 1 32 _+ 1 NS 1322 + 81 1430 + 120 NS

3.3 -+ 0.3 4.2 _+ 0.7 NS 8 _ + 1 - - - -

22+ 1 14+_ 1 - -

followed by nonmigrating activity (phase II), and then by a migrating motor complex (phase 111). 2, 4 On the other hand, the preterm infant frequently does not have normal, fasting phase I I I activity; moreover, the characteristics of phase II activity change with increasing gestational age. 2' 4, 6 Thus serial manometry, which times the appearance o f phase I I I and tracks changes in phase II motor activity, provides a direct monitor of maturation of intestinal function in preterm infants. 2, 7

Fasting plasma concentrations of intestinal peptides also differ among preterm and term infants and among adults.S-l~ Peptide concentrations change with increasing gestational and postnatal age. 9' 10 Moreover, small intestinal motor ac-

tivity and fasting intestinal regulatory peptides both change postnatally in preterm infants who receive enteral nutrition. 7 Plasma peptide concentrations are predictive of motor activity patterns in preterm infants H and, if manometry is not available, could serve as markers for maturation of motor activity patterns.

Some neonatologists delay the use of enteral nutrition for fear of precipitating necrotizing enterocolitis. 12 On the other hand, the lack of enteral feeding may result in an absence of the natural stimulus for growth of the intestinal mucosa; moreover, these stimuli are thought to be provided by enteral feeding. TM 14 Indeed, infants who are fed early have better somatic growth and have a lower incidence of hepatic dysfunction than do those who are fed later in life.iS, 16 The purpose of this study was to use two techniques--manometry and determination of the levels of gastrointestinal peptides--to assess the effect on maturation of the intestinal responses to early or late feeding.

M E T H O D S

Subjects. We studied 27 preterm infants admitted to St. Mary's Hospital Newborn Intensive Care Unit in Roches- ter, Minn., at 28 to 32 weeks of gestational age who required ventilator support for the respiratory distress syndrome (Table I). They were clinically stable at the time of study, a n d required one or two ventilator changes per day; all had a fractional inspired oxygen value <0.40. No infant had

been fed before the study, but all were receiving parenteral nutrition. There were no differences between the two groups with respect to clinical status, as assessed by the need for ventilator support, the presence of intraventricular hemor- rhage or a patent ductus arteriosus, or the presence of metabolic derangements (culture-proven sepsis, electrolyte ab- normality, hypoglycemia, partial pressure of oxygen <50 mm Hg, or hypercapnia with partial pressure of carbon di- oxide >55 mm Hg on the day of the study). No infant had a congenital anomaly. This protocol was approved by the Mayo Clinic Institutional Review Board, and all parents gave informed consent for their infant's participation.

S t u d y des ig n . According to the routine of our neonatal intensive care unit, all babies supported by mechanical ventilation were given parenteral nutrition supplemented by orogastric tube feedings. On entry, babies were randomly assigned to begin enteral supplements by postnatal day 3 to 5 (early) or postnatal day 10 to 14 (late). Babies were given no enteral feedings before study 1, which was performed by postnatal day 5. After the completion of study 1, "early-fed babies" were fed Similac 20 by orogastric tube in 4-hour cycles. A 2-hour infusion of 4 ml/kg (16 kcal/kg per day) was followed by 2 hours of fasting; parenteral nutrition provided an additional 100 kcal/kg per day. "Late-fed babies" received 120 kcal/kg per day by parenteral nutrition, without enteral feedings. Fluid intake in both groups of infants was limited to 150 ml/kg per day.

Study 2 was performed after approximately 10 days on these regimens. After study 2 was completed, babies in the late-fed group were given enteral nutrition to provide 16 kcal/kg per day. Thereafter, enteral feedings were advanced in volume, as determined by the attending neonatologist, in both groups. Study 3 was performed l0 to 14 days after study 2. On each of the 3 study days, manometry was performed and2 ml of blood were withdrawn after 4 hours of fasting for determination of plasma gastrin, gastric inhibitory peptide, neurotensin, and peptide YY concentrations.

Although physician staff could not be unaware of when an infant had been fed, frequent physician turnover often

Volume 120 Preterm intestinal motility and hormones 9 4 9 Number 6

T a b l e II. Characteristics of motor activity during fasting in early- and late-fed infants (study 2)

E a r l y - f e d in fants L a t e - f e d in fants p (n = t 4 ) (n = t 3 ) - -

Phase I (% total recording time) 41.0 _+ 4.0 25.0 • 2.6 <0.005 Phase II

Irregular activity (%) 10.1 _+ 1.4 NS Clusters (%) 62.1 _+ 3.5 <0.0005

Phase III (%) 4.3 _+ 1.6 NS Babies with MMCs 6 NS Cluster occurrence (No./hr/lead) 13.7 _ 0.6 <0.005 Cluster duration (min/hr/lead) 35.7 + 2.1 <0.0005

11.6 + 1.3 42.1 + 3.7

5.7 --- 1.3 11

11.1 + 0.7 23.1 + 2.2

Values (except number of babies with MMCs) are expressed as mean +_ SEM. MMC, Migrating motor complex, defined as phasic activity that migrates over three or more intestinal manometric ports; NS, not significant.

ensured that a given baby's primary care physician near the time of discharge was unaware of the baby's original feeding regimen. Feeding between studies 2 and 3 was uniformly managed: volumes of enteral feedings were begun at 1 ml/hr and advanced daily by 1 to 2 ml/hr until a caloric intake of 120 kcal/kg per day was achieved. Furthermore, standard criteria for discharge were uniformly applied to both sets of infants. To qualify for discharge, a baby was required to maintain normal body temperature in an open crib, no longer to require oxygen, to take all feedings orally, and to sustain a mean 15 gm daily weight gain. Most babies achieved these goals when their body weights were 1800 to 2000 gm.

Maaometry. All studies 1 were performed in the fasting

state. On the days of studies 2 and 3, feedings were withheld for 2 hours or more. A 3.5 mm manometric-feeding tube was manufactured from polyvinyl extrusion tubing; it contained four manometric ports, spaced 2.5 cm apart, and a distal feeding port. The tube was placed by nurses in the neonatal intensive care unit and perfused by a manometric system previously validated and used safely in preterm infants. 4 The ports were perfused continuously with sterile water at 0.6 ml/hr per port, and connected, in turn, by transducers to a Grass recorder. The system had a response rate of 57 mm Hg/sec at 10 psi. 4

The manometric tube was positioned so that three or more ports were located in the duodenum. Placement of the tube was verified by the presence of motor activity charac- teristic of the duodenum; that is, pressure waves were occurring at 9 to 11 times per minute. Our previous studies verified fluoroscopically that the presence of this pattern of motor activity identified predictably that the tube was in the duodenum, 2-4

Manometric recordings were obtained during fasting for 4 hours thereafter. A 2 ml blood sample was drawn from a subset of 17 infants (7 early-fed, 10 late-fed) from central indwelling catheters or by venipuncture after 4 or more

hours of fasting. All blood samples were collected in 2 ml tubes containing 1.5 mg ethylenediaminetetraacetic acid and 500 kallikrein inhibitory units of aprotinin per milliliter of whole blood. Plasma concentrations of gastrin, GIP, NT, and PYY wer e determined on duplicate samples by established radioimmunoassays. 17-a~

Data analysis . All manometric tracings were analyzed in 30-minute segments by a technician unaware of the identity of the patients. During fasting, the presence or absence of three types of motor activity was noted. As previously defined for preterm infants, 2, 4 phase I was one of motor quiescence; phase III was characterized by phasic activity for 2 minutes or more, with migration over three or more ports (equivalent to the adult migrating motor complex).1 Phase II was irregular activity or nonmigrating phasic bursts ex- ceeding 0.4 minute in duration; the latter were defined as "clusters." The duration, amplitude, and frequencies of phase I, II, and III activity were calculated.

All plasma concentrations of gastrointestinal hormones and peptides were expressed as picograms per milliliter. Characteristics of motor activity and plasma concentrations of intestinal regulatory peptides were compared between early- and late-fed infants by unpaired t testing.

Clinical assessment of these babies included their weight gain, the number of days required to establish oral feedings, and the presence of feeding intolerance, which was defined as any day in which feedings were withheld by the attending neonatologist because of the presence of increased ab- dominal girth or vomiting, or when milk was retained in the stomach for 3 hours after completion of a feeding. The establishment of enteral nutrition was defined as the hospital

day when 120 kcal/kg per day intake was provided by the orogastric or oroduodenal route. The establishment of oral nutrition was defined as the day when tube feedings were no longer required. These characteristics were compared between early- and late-fed infants by unpaired t testing. Se- rial body weights were compared with birth weights for each

9 5 0 Berseth The Journal ofPediatrics June 1992

Proximal duodenum

Mid duodenum

Distal duodenum . . , x . . , 1 ~ 1 1 ~

m m Hg P r o x i m a l j e j u n u m

A 1 rain

Proximal duodenum

|

Mid duodenum

* * t * ~25 Distal duodenum mm Hg

P r o x i m a l j e j u n u m

.~ :~ ~ * "~ ~ 1 min /

B Figure. Small intestinal motor activity in preterm infants. A, Representative tracing from infant who received early hypocaloric enteral feedings. B, Representative tracing from infant who received no hypocaloric enteral feedings. Motor activity from proximal duodenal lead is shown at top of each set (lead 1) and from distal lead at bottom of each tracing (lead 4). Clusters are designated by asterisk. Note that in tracing from early-fed infant, clusters occur infrequently in leads 3 and 4 and that many are sustained for 3 to 4 minutes. Quiescence is well defined and sustained for approximately 7 to 8 minutes in leads 3 and 4. Note that, in contrast to the early-fed infant, the late-fed infant displays clusters that occur with greater frequency in all four leads. Although one cluster appears to migrate across all four leads, designated by arrows, the remaining clusters occur randomly and are sustained for only 1 to 2 minutes. Quiescence is ill defined compared with that in early-fed infant.

group of infants by paired t testing. The Bonferroni rule was used to correct for multiple comparisons, and a value of p <0.01 was considered to be significant for weight changes. A p value of <0.05 was considered to be significant for comparisons of motor activity, peptide concentration, and historical characteristics.

R E S U L T S

Motor activity Study 1. There were no significant differences between

the initial motor activities of early- and late-fed babies. Approximately half of the babies in each group had complete interdigestive cycles including phases I, II, and III.


There were no differences in the duration or periodicity of phase I I I activity. Clusters of motor activity occupied half of the total recording time and 80% of phase II activity. Quiescence was observed for 40% of the total recording time. The occurrence and duration of phases 1 and II were similar in both groups of babies.

Study 2. At the time of study 2, early-fed infants had received feedings for 8 +_ 1 days. Motor activity was significantly different in early- and late-fed infants (Table II). By visual inspection, early-fed babies had motor activity that was better organized; periods of quiescence were longer and individual clusters were sustained for longer periods (Fig- ure). Nevertheless, the majority of infants in both groups had phase III activity; moreover, the amplitude, periodicity, and mean duration of the phase III activity were similar in both groups. Clustered motor activity occupied approximately three fourths of all of phase II in both groups of infants; however, clusters occurred less frequently and con- stituted less of the motor activity in early-fed infants than they did in late-fed infants (Table II; p <0.005). Quiescence occupied a significantly greater portion of total recording time in tracings from early-fed infants than in those from late-fed infants (Table II; p <0.005).

Study 3. By study 3, all characteristics of fasting motor activity were identical in the two groups. Three fourths of all babies had migrating motor complexes, of which 80% migrated distally. Clusters represented two thirds of all phase II activity, and quiescence occupied approximately 50% of all recording time.

Plasma gastrointestinal hormones and peptides. There were no differences in fasting plasma levels of gastrin, GIP, ST, or PYY between early- and late-fed infants at the time of study 1. By study 2, the fasting plasma gastrin level was significantly higher in in early-fed infants (Table III; p <0.03), and the plasma GIP level was more than twofold higher in early-fed infants than in late-fed infants (Table III; p <0.001). Compared with study 1, neurotensin doubled and PYY increased by 30% by study 2, but there was no difference in plasma PYY levels between early- and late-fed infants. By the time of study 3, the GIP level was almost 50% higher in early-fed infants than in late-fed infants (1305 + 140 vs 782 + 103;p <0.02). Gastrin had doubled in both groups, but there was no difference between the two groups. The NT level was fourfold higher at study 3 than at study 11 and PYY increased 40% above the concentration at study 1. However, there were no differences in NT or PYY levels between the two groups.

Nutritional data. Early-fed infants had more rapid establishment of complete nipple feedings. Early-fed infants received full enteral nutrition by day 17, but late-fed infants did not achieve this goal until day 31 (Table IV; p <0.01). Early-fed babies established complete nipple feedings on

Table IIh Plasma peptide concentrations during fasting in early- and late-fed infants (study 2)

Early-fed Late-fed infants infants p

Gastrin 93 • 6 64 • 7 0.03 GIP 1505 -+ 65 781 • 75 0.00l NT 69 + 24 53 • 21 NS PYY 1121 • 149 799 • 109 NS

Values are expressed as mean _+ SEM. NS, Not significant.

day 35 and were discharged by day 38, whereas late-fed infants did not achieve full oral nutrition until day 47 and were discharged on day 55 (Table IV; both p <0.05). Late-fed infants had three times more days of feeding intolerance than did early-fed infants (Table IV; p <0.01). Neither group of infants gained weight the first 2 weeks of life. By the third postnatal week, early-fed infants had significant weight gain; late-fed infants did not establish significant weight gain until the fifth postnatal week.

D I S C U S S I O N

These experiments show that early feeding of preterm infants enhanced the maturation of small intestinal motor activity and peptide responses to food or, conversely, that a delay of enteral feedings prevents normal maturation. A delay of 8 days in the provision of enteral nutrition resulted in significant differences in clinical outcome for up to 5 weeks.

Motor activity changes rapidly postnatally in preterm infants who receive routine feedings; occurrence and over- all duration of cluster activity decrease as quiescence increases. 7 In the current study, babies who received early enteral nutrition had a significant decrease in cluster activity and an increase in quiescence by study 2. These changes were comparable to those in preterm babies who receive normal feedings] On the other hand, babies who had received no feedings by study 2 did not display these changes. Hence the use of minimal feedings resulted in the same maturation patterns of motor activity as those seen in babies who had received full oral feedings.

Early feeding also enhanced the postnatal increase in fasting levels of certain gastrointestinal hormones, in par- ticular, those regulatory peptides that influence and modulate intestinal motor activity. In infants who receive enteral feedings, plasma levels of gastrin and GIP are higher than in infants given only intravenous infusions of dextrose} l The use of "minimal feedings" of 1 ml/hr has been shown to result in higher fasting gastrin concentration in preterm infants. 22 In the present study, NT and PYY levels were%lso elevated by enteral nutrition. In animals, gastrin and neu-

9 5 2 Berseth The Journal of Pediatrics June 1992

Table IV. Feeding characteristics in early-and late-fed babies

Early-fed infants Late-fed infants p

Time to establish complete enteral nutrition (days) Time to establish complete oral nutrition (days) Hospitalization (days) Infants with feeding intolerance (No.) Duration of feeding intolerance (days)

17.8 + 2.6 31.2 _+ 4.0 <0.01 35.3 _+ 3.1 47.2 _+ 2.9 <0.05 38.6 _+ 3.0 55.2 _+ 5.7 <0.05

5 10 <0.0l 1.8 _+ 0.8 5.2 _+ 1.2 <0.01

Values (except number of infants with feeding intolerance) are expressed as mean _+ SEM.

rotensin increase the amplitude and/or occurrence of motor activity; GIP and PYY inhibit it. 23, 24 Thus the changes

i'n peptide concentrations in early fed infants may have contributed to the changes in motor activity that we observed. For example, the plasma gastrin concentration has been shown to be predictive of cluster activity and the duration of quiescence in preterm infants.12 However, these

peptides also contribute to other aspects of gut function and maturation. Gastrin is an intestinal trophic agent, 25, 26 and

GIP plays a significant role in the regulation of the entero- insular axis; insulin, in turn, influences growth of the fetus and newborn infantS, 28 Thus the elevation of gastrin and GIP levels by early feedings may have additional value to the preterm infant beyond the localized effect on motor activity in the intestine. Although the peptides measured in

this study modulate motor activity, postnatal changes in their concentrations parallel those in motor activity. Hence they may be important markers for maturation of motor activity. Thus if the capability to perform manometrics is lacking in a neonatal intensive care unit, plasma peptide concentrations may be usable as a marker of intestinal

maturation. After the initiation of feedings, motor activity patterns

changed rapidly in the late-fed infants; all infants had mature motor activity by the time of study 3. These findings indicate that inhibitory mechanisms may be important to the developing gut and that these influences, in turn, may well regulate the organization of fasting motor activity in the infant.

Because fasting motor activity governs the movement of nutrients through the gastrointestinal tract, we speculate that earlier maturation of motor activity may contribute to improved absorption coefficients of nutrients and, hence,

better somatic growth. Although the focus of this study was muscle function, enteral nutrition may also have matured mucosal receptors, thereby contributing to improved mucosal growth.

Previous studies evaluating feeding regimens in preterm infants have required the entry of many subjects from sev- eral centers because indirect measurements such as weight gain and days of feeding intolerance were compared. Such factors are influenced by other clinical events that cannot be

tightly controlled and therefore are less sensitive indexes for monitoring effects of feeding than are the two direct measurements used in this study. Because these techniques provide a direct, sensitive measure of gastrointestinal maturation, they may permit the design of well-controlled studies in small numbers of babies to assess "feeding readiness ''29 and to test various feeding regimens for preterm infants. 3~ Such studies may permit the tailoring of specific nutrients and feeding regimens for preterm infants.

I thank Cindy Nordyke for her dedication, tenacity, and compassion in providing technical assistance in the performance of these studies; Drs. Fredric Kleinberg, Robert Johnson, and Douglas Derleth for their encouragement and support in conducting this research; and Dr. Sidney F. Phillips for his editorial assistance and his many hours of consultation in establishing manometry in newborn infants.

REFERENCES

1. Malagelada J-R, Camilleri M, Stanghellini V. Manometric diagnosis of gastrointestinal motility disorders. New York: Thieme, 1986.

2. Berseth CL. Gestational evolution of small intestinal motility in preterm and term infants. J PEDIATR 1989;115:646-51.

3. Berseth CL. Neonatal small intestinal motility: the motor responses to feeding in term and preterm infants. J PEDIATR 1990;117:777-82.

4. Amarnath RP, Berseth CL, Malagelada J-R, et al. Postnatal maturation of small intestinal motility in preterm and term infants. Journal of Gastrointestinal Motility 1989;1:138-43.

5. Tomomasa T, Itoh Z, Koizumi T, et al. Nonmigrating rhyth- mic activity in the stomach and duodenum of neonates. Biol Neonate 1985;48:1-9.

6. Morriss FH, Moore M, Weisbrodt NW, et al. Ontogenic development of gastrointestinal motility. IV. Duodenal contrac- tions in preterm infants. Pediatrics 1986;28:1106-13.

7. Berseth CL. Postnatal maturation of motor activity patterns and gastrointestinal hormones and peptides in preterm infants [Abstract]. Gastroenterology 1990;99:1204.

8. Lucas A, Bloom SR, Aynsley-Green A. Metabolic and endo- crine events at the time of the first feed of human milk in preterm and term infants. Arch Dis Child 1978;53:731-6.

9. Lucas A, Bloom SR, Aynsley-Green A. Postnatal surge in plasma gut hormones in term and preterm infants. Biol Neo- nate 1982;41:63-7.

10. Valdez MG, Go VLW, Berset h CL. Gestational maturation of


regionally distributed gastrointestinal (GI) hormones in preterm and term infants [Abstract]. Pediatr Res 1990;27:55A.

11. Berseth CL, Go VLW. Correlation of gastrointestinal hormones with maturation of neonatal small intestinal motility [Abstract]. Gastroenterology 1990;97:A178.

12. LaGamma EF, Ostertag SG, Birenbaum H. Failure of delayed oral feedings to prevent necrotizing enterocolitis. Am J Dis Child 1985;139:385-9.

13. Dworkin LD, Levin GM, Farber N J, Spector MH. Small intestinal mass of the rat is partially determined by indirect effects of intraluminal nutrition. Gastroenterology 1976;71:626- 30.

14. Feldman E, Dowling R, McNaughton J, Peters T. Effects of oral versus intravenous nutrition on intestinal adaptation after small bowel resection in the dog. Gastroenterology 1974;70: 712-9.

l 5. Brosivs KK, Ritter DA, Kenny JD. Postnatal growth curve of the infant with extremely low birth weight who was fed enter- ally. Pediatrics 1984;74:778-82.

16. Moyer-Mileur L, Chan GM. Nutritional support of very-low- birth-weight infants requiring prolonged assisted ventilation. Am J Dis Child 1986;140:929-32.

17. Kuzio M, Dryburgh JR, Mallay KM, et al. Radioimmunoas- say for gastric inhibitory peptide. Gastroenterology 1974;66: 357-64.

18. Sizemore GW, Go VLW, Kaplan EL, et ai. Relations of caI- citonin and gastrin in the Zollinger-Ellison syndrome and medullary carcinoma of the thyroid. N Engl J Med 1973; 288:641-4.

19. Yaksh TL, Michener SR, Bailey JE, et al. Survey of distribution of substance P, vasoactive intestinal peptide, cholecysto- kinin, neurotensin, met-enkephalin, bombesin, and PHI in the

spinal cord of cat, dog, sloth, and monkey. Peptides 1988;9:35% 72.

20. Roddy DR, Koch TR, Reilly WM, et al. Identification and distribution of immunoreactive peptide YY in the human ca- nine and murine gastrointestinal tracts. Regul Pept 1987;18: 201-12.

21. Lucas A, Bloom SR, Anysley-Green A. Metabolic and endo- crine consequences of depriving preterm infants of enteral nutrition. Acta Paediatr Scand 1983;72:245-9.

22. Govnaris A, Anatolitov F, Costalas C, et al. Minimal enteral feeding, nasojejunal feeding and gastrin levels in premature infants. Acta Paediatr Scand 1990;79:226-7.

23. Morgan K, Schmalz P, Go VLW, et al. Effects of pentagastrin G17 and G34 on the electrical and mechanical activities of ca- nine smooth muscle. Gastroenterology 1978;75:405-12.

24. Alien J, Fitzpatrick M, Yeats J, Darcy K, et al. Effects of peptide YY and neuropeptide Y on gastric emptying in man. Di- gestion 1984;30:255-62.

25. Johnson LR. The trophic action of gastrointestinal hormones. Gastroenterology 1976;70:278-88.

26. Johnson LR. New aspects of the trophic action of gastrointestinal hormones. Gastroenterology 1977;72:788-92.

27. Hill DE. Effect of insulin on fetal growth. Semin Perinatol 1978;2:319-28.

28. Axelsson IE, Ivarsson SA, Rfiih/i NCR. Protein intake in early infancy: effects on plasma amino acid concentrations, insulin, metabolism and growth. Pediatr Res 1989;26:614-7.

29. Berseth CL. Small intestinal motor activity predicts feeding intolerance in preterm infants [Abstract]. Journal of Gas- trointestinal Motility 1991;3:172.

30. Koenig W, Berseth CL. Manometrics for preterm infants: a new tool for old questions. Am J Perinatol (in press).

BOUND VOLUMES AVAILABLE TO SUBSCRIBERS

Bound volumes of the 1992 issues of THE JOURNAL OF PEDIATRICS are available to subscribers (only) from the Publisher, at a cost of $58.00 for domestic, $78.06 for Canadian, and $74.00 for international subscribers, for Vol. 120 (January-June) and Vol. 121 (July-December), shipping charges included. Each bound volume contains subject and author indexes, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram, with the Journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Subscrip- tion Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318, USA/800-325-4177, ext. 4351, or 314-453-4351.

Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.

effect of early feeding on maturation of the preterm infant's small intestine

Documents