| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
Department of Neonatology, Lis Maternity Hospital, Tel Aviv (S.D., F.B.M.), ISRAEL
Department of Neonatology, Sheba Medical Center, Tel Hashomer (J.K., R.M.), ISRAEL
Address reprint requests to: Francis B. Mimouni, MD, Department of Neonatology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman Street, Tel Aviv, 64239, Israel. E-mail: mimouni{at}tasmc.health.gov.il
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Design: Two-center, prospective, randomized, unmasked clinical trial.
Patients: 28 VLBW infants (birth weight <1250 g). A strict feeding protocol was followed.
Intervention: Patients were randomized to IGB or CGI.
Main outcome measures: Time to reach full feeds (160 cc/kg/d)(by design and real), daily weight, caloric intake, residual gastric volume and type of feeding (formula vs. human milk vs. both).
Results: Five infants failed to complete the study because of death (n = 4) or protocol violation (n = 1). The two groups did not differ by birth weight or gestational age; infants fed via IGB reached full feeds earlier (p = 0.03) and had less delay in reaching full feeds than infants fed via CGI.
Conclusion: Contrary to our hypothesis, gravity IGB is more effective than CGI in improving feeding tolerance in small VLBW infants.
Key words: prematurity, infant nutrition, weight gain
Abbreviations: CGI = continuous gastric infusion • IGB = intermittent gastric bolus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The sample size was established to be initially 25 to 30 infants, as a pilot study that would enable us to calculate a final sample size (if necessary) or to stop the study at this point.
The study was approved by our local Institutional Review Boards for Human Subject Research. Informed written consent was obtained from both parents, except in two cases of single mothers.
Study Population
A total of 28 premature infants from the nurseries at the Lis Maternity Hospital, Tel Aviv, and at the Sheba Medical Center, Tel Hashomer, were enrolled between January 1998 and September 1998 based on the following criteria: birth weight 501–1250 g, appropriate for gestational age (between the 10th and the 90th percentiles of the Lubchenco intrauterine growth charts [2]), postnatal age <48 hours, absence of major congenital malformations and written consent form signed by the parents.
Feeding Protocol
A feeding protocol was approved by the neonatologists at both institutions and strictly adhered to. In practice, a feeding schedule was written for each child and kept at his bedside. The schedule clearly indicated when feeds should be started, the rate of daily increments, feeding composition (pumped human milk whenever available, diluted or non-diluted formula) and time to reach full-feeds (160 mL/kg/d).
In all patients, nasogastric tubes were inserted. Bolus feedings were given by gravity drainage after priming of the tubing, every two hours in infants in the lowest birth-weight group and every three hours in the two other groups. Continuous feeds were given using an automatic syringe-pump. Gastric residuals were checked for every two hours in infants in the lowest birth-weight group and every three hours in the two other groups, regardless of assignment.
Feeds were started at 20 mL/kg/day in all three birth-weight groups, with small equal daily increments in such a manner that full volume feeds (160 cc/kg/day) were reached within eight days in the largest weight group, within ten days in the intermediate group and within 14 days in the smallest weight group. They were started on day 2 at the earliest, but were postponed by 72 hours in case of asphyxia or confirmed sepsis or by 24 hours following a stable blood pressure within the normal values established by Kitterman [3]. These targets were based on the general principle of careful, slower progression of feeds in the smallest and sickest infants to reduce the potential risk of necrotizing enterocolitis.
Expressed human milk, when available, was the nutrition of choice and was used undiluted. When unavailable, it was replaced by Similac Special Care Ready to Feed formula (Ross Laboratories, Columbus, OH), which was used by the decreasing birth-weight group (1001–1250 g, 751–1000 g, 501–750 g), initially diluted to 12 kcal/oz until day 3, 5 or 6 of feeds (depending upon birth-weight subgroup), then replaced by 18 kcal/oz until day 4, 6 or 8, by 20 kcal/oz by day 6, 8 or 11, then by 24 kcal/oz by day 8, 10 or 14, respectively. The rationale for starting feeds with diluted formula was to prevent exposure of an immature gut to excess nutrients [4,5].
Supplemental fluids were given as a glucose solution on day 1, then as a standardized parenteral nutrition, and volumes were adjusted to allow a weight loss of no more than 10% to 15% of birth weight.
Outcome Measures
Body weight was measured every morning to the closest gram, using an electronic scale (Model T-15-s, Shekel Scales, Tel Aviv, Israel).
Feeding tolerance (gastric residuals measurement) in both groups was assessed every two hours in infants whose birth weight ranged between 501 and 750 grams and every three hours in all other infants. In infants who were randomized to bolus feeds, gastric residuals were checked for just prior to the next scheduled meal. The following were recorded: residual gastric volume, the presence of vomiting, abdominal distention and the stool number. Whenever the residual gastric volume was between 0% and 20% of the feeding volume of the previous two hours (lowest birth-weight group) or three hours (two highest birth-weight groups), the residual was re-fed and the feeding schedule was resumed as planned. When it was between 21% and 40%, without any ominous abdominal signs, two bolus feeds were held (or their volume-equivalent in the CGI group); the feeding schedule was then resumed as planned. If it was >40%, or repeatedly between 21% and 40%, feeds were held for 12 hours and resumed according to the feeding schedule of the previous day. In all infants, the absence of stools for more than 24 hours prompted the rectal insertion of a small glycerin chip. Necrotizing enterocolitis was defined as the presence of feeding intolerance with or without abdominal distention together with the presence of intramural gas on abdominal radiograph [6].
Data Analyses
Minitab Statistical Software was used for analyses (Minitab Inc., State College, PA). All randomized infants were included in the statistical analysis (by intention to treat). The Kruskal-Wallis test was used for assessing differences between continuous variables. Chi square and Fisher exact tests were used to assess differences between discrete variables. Backward stepwise regression analysis was used to test the influence selected independent variables such as birth weight (or gestational age), mode of feeding (CGI versus IGB), type of feeding or ventilatory status or study site had on major dependent variables such as the time to reach full feeds. Results are expressed as mean ± SE, or n (%). Statistical significance was set at 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
In backward stepwise multiple regression analysis taking into account birth weight, mode of feeding (CGI versus IGB), type of feeding (human milk versus formula versus both), birth center and ventilatory status at entry (ventilated versus non-ventilated) in the study, only the mode of feeding affected feeding tolerance (R2 = 20.3, p < 0.03).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
To the best of our knowledge, there have been nine studies comparing IGB to CGI in VLBW infants [7–14]. Two of them [7,9] have been published only in an abstracted form, which does not allow for a complete, critical assessment.
Not all studies examined feeding tolerance as an outcome variable. Blondheim et al. [11] examined mostly pulmonary mechanics in a study of 24 VLBW infants and concluded that IGB negatively affects pulmonary function, while feeding with CGI does not alter baseline pulmonary mechanics. That specific study related to infants who already tolerated feeds, regardless of the method used. Aynsley-Green et al. [8] focused on entero-insular hormones and concluded that IGB elicits cyclical surges of these hormones not found with CGI. Since several of these hormones have a trophic effect on the intestinal mucosa [15], it has been speculated that IGB may be a preferable method of feeding in VLBW infants, because of its potential for inducing faster gut maturation. In contrast, Parker et al. [16] found that in infants with intestinal disease (protracted diarrhea or short bowel), CGI improves enteral balance, leading to higher nutrient retention than IGB. The latter study prompted the examination of the hypothesis that in VLBW infants, CGI would improve nutrient balance when compared with IGB. This hypothesis has been tested in six studies [4,9,10,12–14]; in all of them, CGI was not found to improve nutrient balance, whether this was tested indirectly through the examination of weight gain [4,10,12,14] or time to regain birth weight [9,13,14] or directly through balance studies [4].
Six studies out of the mentioned above have compared IGB to CGI in terms of feeding tolerance [4,7,10,12–14]. Only one [7] reports improved feeding tolerance in the CGI group. This must be taken with caution, as this study is a retrospective one, and is available only in an abstracted form. Three additional studies [4,12,13] found no significant differences between the two methods. However, that of Macdonald et al. [12] was limited by the facts that infants were distributed in three small groups (an additional group of intra-pylorically-fed infants was also studied) and that feeding intolerance was not precisely defined. Although the investigators were blinded, the study by Silvestre et al. [4] was plagued by the intervention of the attending physicians, who switched many infants from the IGB to the CGI group because of what they felt were feeding-related events such as apneas and bradycardias. Such infants switched to CGI tolerated feeds only after a prolonged period, which may indicate that in these specific infants, feeding intolerance may not have been related to the method of feeding. The third study to find no significant differences between the methods is that by Akintorin et al. [13]. The latter study is, in our opinion, very limited by the fact that the definitions of feeding tolerance in the CGI and the IGB groups were not identical, which may well have introduced a systematic bias [13].
We are left with two studies, which like ours found a significant advantage of IGB over CGI in terms of feeding tolerance [10,14]. Toce et al. [10] found this advantage to exist only in the 1000 to 1250 g birth-weight stratum of their patients. In the large, elegant study by Schanler et al. [14], which also contained balance data, the advantage of IGB over CGI was mostly present in the 26 to 27 weeks gestational age group, while no significant differences were found in the 28 to 30 weeks gestational age group. Importantly, in Schanler’s study, more protocol violations (switches from one method to the other because of protracted failure to tolerate feeds) were observed in the CGI group. In our study, only one infant was dropped for a similar reason (she did not tolerate feeds after 42 days), and she belonged to the CGI group as well. When switched to boluses, she tolerated feeds almost immediately.
The time to regain birth weight was very similar in both of our study groups. We believe this likely to be due to the fact that patients received complementary parenteral nutrition, which allowed for appropriate growth.
We conclude that when feeds are initiated in a VLBW infant, the IGB method is probably preferable to CGI and should decrease the incidence of episodes of feeding intolerance. It will not eradicate them, as such episodes were seen in both feeding groups. Moreover, our findings do not necessarily apply to those infants who already tolerate feeds and who, under specific circumstances (such as poor pulmonary function and deterioration during IGB), may still benefit from CGI. We can only speculate about the mechanism, which may explain our results: one possible explanation is that there might be a need for stomach distention by a minimum volume of feeds in order to optimize gut peristaltism. This speculation might appear to be probable with the use of modern techniques such as electrogastrography.
Received March 24, 2000. Accepted October 1, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S S Premji Standardised feeding regimens: hope for reducing the risk of necrotising enterocolitis Arch. Dis. Child. Fetal Neonatal Ed., May 1, 2005; 90(3): F192 - f193. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Caple, D. Armentrout, V. Huseby, B. Halbardier, J. Garcia, J. W. Sparks, and F. R. Moya Randomized, Controlled Trial of Slow Versus Rapid Feeding Volume Advancement in Preterm Infants Pediatrics, December 1, 2004; 114(6): 1597 - 1600. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. B. Kenton, C. J. Fernandes, C. L. Berseth, B. A. Cobb, W. A. Carlo, and N. Ambalavanan Gastric Residuals in Prediction of Necrotizing Enterocolitis in Very Low Birth Weight Infants Pediatrics, June 1, 2004; 113(6): 1848 - 1849. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
