Methods

Consecutive preterm infants were recruited and studied on the regional neonatal intensive care unit, St James’s University Hospital, Leeds. The routine policy on this unit was for all infants with respiratory distress syndrome of prematurity to receive total parenteral nutrition until ventilatory support was stopped, at which time milk feeding was introduced.

On day 3 of life, infants who satisfied all the following criteria were recruited into both parts of the study: gestation less than 37 completed weeks; birthweight less than 1750 g; respiratory distress syndrome requiring ventilatory support beyond the second day; and in receipt of no enteral feeding. Infants were excluded if they had clinically significant congenital gastrointestinal pathology, a lethal congenital malformation, or if they were receiving drug treatment known to interfere directly with gastric or intestinal motility. Treatment with caffeine was not an exclusion criterion as its use was routine to prevent apnoea of prematurity. Infants left the study if they were transferred to another hospital for further medical care. Data obtained before discharge were included for analysis.

Using data from previous relevant studies,13-15 the number of infants required to demonstrate a 10 minute difference in gastric emptying with 90% power was estimated at 40. Similarly, using previous study data,16-18 30 infants was estimated as being required to demonstrate a 5 hour difference in the whole gut transit time (WGTT).

Informed parental consent was obtained before each infant was recrited to the study. The study was approved by the local research ethics committee.

RANDOMISATION

The following protocol was applied to both parts of the study. To prevent chance sampling inequality, infants were stratified by gestation and birthweight ratio (defined as birthweight divided by the relevant 50th centile value for weight for the gestational age at birth) before randomisation. Randomisation was then into one of two groups: TF (to receive trophic feeding as well as parenteral nutrition), or C (control group to receive parenteral nutrition alone). Randomisation was performed using computer software: minimisation program for allocating patients to treatments in clinical trials, Department of Clinical Epidemiology, The London Hospital Medical College.

FEEDING SCHEDULE

Trophic feeding in group TF was started on day 3 of life at the rate of 0.5 ml/hour if birthweight was less than 1 kg, or 1 ml/hour if birthweight was equal or greater than 1 kg. Expressed breast milk (EBM) or preterm formula milk (Nutriprem; Cow & Gate Nutricia) was given according to maternal choice. If EBM was unavailable preterm formula was given. All milk was administered as hourly bolus feeds through a 5 or 6 French gauge silastic nasogastric tube (Vygon (UK) Ltd.). Trophic feeding was continued at the same rate until intermittent positive pressure ventilatory support stopped. Feeding was omitted while infants received continuous positive airway pressure (CPAP) support because of the possible increased risk of milk aspiration. Once ventilatory support was no longer required nutritive milk feeds were introduced and increased until full enteral feeding was established. The rate of increase was determined by feed tolerance and was at the discretion of the attending nurse, usually volume was increased by 1 ml every 12 hours. EBM was fortified with a commercial preparation (Breast Milk Fortifier, Cow & Gate Nutricia) once an infant was receiving 90 ml/kg/day of milk.

Group C received parenteral nutrition alone until ventilatory support, including CPAP, was no longer required. Milk feeding was then introduced, increased, and fortified as for group TF.

The regimen of parenteral nutrition used for both groups was the same and consisted of a solution of amino acids (Vaminolact, Pharmacia-Upjohn), 10% dextrose, electrolytes, trace minerals and water soluble vitamins, plus a fat emulsion Intralipid 20% (Pharmacia-Upjohn) with added fat soluble vitamins. Parenteral nutrition was started on day 2 of life and the content was incrementally increased to a maximum on day 7. The rate of increase and total fluid intake for both groups were matched and included that contributed by trophic feeding.

No attempt was made to blind carers to infant group allocation. It was considered both impractical and unethical to pass nasogastric feeding tubes and give sham milk feeds to infants in group C.

PART 1: ASSESSMENT OF GASTRIC EMPTYING

Gastric emptying was assessed within 24 hours of the first occasion that each infant received a milk intake equal to or exceeding 90 ml/kg/day (at least a 4 ml/kg hourly bolus feed). Gastric emptying was assessed ultrasonically by serially measuring the change in the gastric antral cross sectional area (ACSA) following a feed.13 ACSA was measured by obtaining real time, ultrasound, cross sectional images of the gastric antrum with the infant lying in the right lateral position. Vertebral bodies, the aorta, and the superior mesenteric artery were used as constant anatomical landmarks to ensure the same cross sectional plane of the gastric antrum was measured on each occasion. Images were obtained using the Hewlett Packard Ultrasound Imaging System, model HP Sonos 100, with a 7.5 MHz probe. Software available on the machine allowed images to be frozen on the screen, the gastric antrum cross sectional circumference to be marked with a steerable cursor, and the calculation of the ACSA.

Before each study the stomach contents were aspirated up the nasogastric tube. The ACSA was measured before the test feed. A bolus feed of 4 ml/kg of whichever milk the infant was receiving was given over a period of 10 minutes. Immediately the feed was completed the ACSA was remeasured. ACSA was subsequently measured at 10 minute intervals until 30 minutes had elapsed, and then at 15 minute intervals until 2 hours after the feed, or the pre-feed value was reached. Fifty per cent ΔACSA, defined as the time taken for the ACSA to decrease by half of the maximum change seen during gastric filling, was calculated. This value provides a simple, single measurement of gastric emptying allowing subsequent infant comparison.13

PART 2: ASSESSMENT OF WHOLE GUT TRANSIT TIME

WGTT was measured by timing the transit of the marker carmine red through the gut. Carmine red is an inert non-absorbed dye, which is bright red in colour and readily identifiable in neonatal stools.16-18 WGTT was assessed in each infant on two occasions, at 3 weeks and at 6 weeks of postnatal age. Carmine red (125 mg) was syringed over 1 minute down a 5 or 6 French gauge nasogastric feeding tube and the time recorded. At each subsequent nappy change the nurse inspected any stool for the first presence of carmine red. Once detected, its presence was confirmed by an investigator (RM) and the time elapsed since administration recorded. WGTT was defined as the time between carmine red administration and its first appearance in the stool.

Carmine red was always given at about 3 pm to minimise any effect of circadian rhythm. In infants receiving enteral feeds carmine red was given at the mid point between feeds. Nappy change frequency was at the discretion of the attending nurse, but was generally 3–4 hourly. Stool inspection was continued for a maximum of three weeks. If no carmine red was seen by this time then it was felt likely that passage had been missed and no value for WGTT was recorded.

DATA ANALYSIS

Statistical analysis was performed using the computer software program Excel 5.0 (Microsoft Corporation). The two tailed Student’s t test was used for comparison of means of variables approximating a normal distribution. For non-normal variables, the Mann-Whitney U test was used. Fisher’s exact test was used for categorical variables. A p value of less than 0.05 was considered significant. Where appropriate the mean difference or median difference, with 95% confidence interval, was also calculated using the software Confidence Interval Analysis (BMJ Publishing Group).