1. Objectives

To compare the effect of nasal versus oral placement of enteral feeding tubes in preterm or low birth weight infants on feed tolerance, growth and development, and adverse events

2. How studies were identified

The following databases were searched in September 2012:

• CENTRAL (The Cochrane Library 2012, Issue 10)
• MEDLINE
• EMBASE
• ClinicalTrials.gov
• Current Controlled Trials
• WHO Clinical Trials Registry

Reference lists and conference proceedings were also searched

3. Criteria for including studies in the review

3.1 Study type

Randomized controlled trials, including quasi-randomized controlled trials

3.2 Study participants

Preterm (<37 weeks’ gestation) or low birth weight (<2500 g) infants receiving enteral tube feeding

3.3 Interventions

Nasal versus oral placement of enteral feeding tubes, placed continuously (indwelling) or removed between feeds provided the same strategy was used in both treatment groups

3.4 Primary outcomes

• Time to establish full enteral tube feeds independent of parenteral fluids or nutrition

Secondary outcomes included growth (time to regain birth weight; rates of weight gain, length gain, head circumference growth, or change in skinfold thickness), adverse events (non-intentional removal/displacement of feeding tube; apnoea, defined as breathing pauses for >20 seconds or >10 seconds if associated with bradycardia or oxygen desaturation; aspiration pneumonia/pneumonitis, with clinical or radiological evidence of lower respiratory tract compromise attributable to aspiration of gastric contents), time to independence from supplemental oxygen, and time to establish full oral feeds (≥120 mL/kg/day)

4. Main results

4.1 Included studies

Three randomized trials, enrolling 123 children, were included in this review

• Two trials were parallel randomized controlled trials and one trial was a randomized crossover trial
• One trial randomized 42 preterm infants (30 to 34 weeks’ gestation) to naso-enteric or oro-enteric feeding between one and 12 days of age, with infants remaining in the trial until breast or bottle-feeding had been established. Some infants in each group also received transpyloric feeds
• One trial randomized 46 very preterm or very low birth weight infants (<30 weeks’ gestation or birth weight <1200 g) to gavage feeding via a nasogastric tube or an orogastric tube until 32 weeks’ post-menstrual age, with all infants subsequently fed via orogastric tube
• One trial randomized 35 very preterm infants (<32 weeks’ gestation) with evidence of apnoea or bradycardia of prematurity to 12 hours of nasogastric feeding followed by 12 hours of orogastric feeding, or vice versa

4.2 Study settings

• Germany, Sweden, and the United Kingdom of Great Britain and Northern Ireland
• All trials were conducted in neonatal units
• One trial was published in 1984, and the other two trials were conducted between 1998 to 2001 and 2008 to 2009

4.3 Study settings

How the data were analysed
Nasal placement of enteral feeding tubes was compared with oral placement of enteral feeding tubes. Data were not pooled and were presented as mean differences (MD) with corresponding 95% confidence intervals (CI). In the case that data were sufficient for meta-analysis, sensitivity analyses by study design, risk of bias, and participant characteristics were planned where substantial heterogeneity was detected (I²>50%). To further explore potential sources of heterogeneity, subgroup analysis including only very low birth weight (<1500 g) and very preterm (<32 weeks’ gestation) infants was planned.

Results
Time to establish full enteral tube feeds
In one randomized controlled trial involving 46 very preterm or very low birth weight infants, no statistically significant difference in the time to establish full enteral feeds was found between nasogastric and orogastric groups (MD -2.7 days, 95% CI [-11.9 to 6.5]). No other trials reported on this outcome.

Growth
Time to regain birth weight was not different between groups in one trial of 46 very preterm or very low birth weight infants (MD 0.90 days, 95% CI [-1.3 to 3.1]). In the parallel randomized trial enrolling 42 preterm infants, 37 were remaining at two weeks post-randomization. Among these infants, those randomized to receive naso-enteric feeding tubes had a statistically significantly lower rate of weight gain in the first week (0.6 versus 8.3 g/kg/day), a difference that disappeared in the second week (13.6 versus 12.3 g/kg/day). The mean postnatal age at study entry was two days in the naso-enteric group versus six days in the oral group, which may explain the different rates of growth over the first week of the intervention.

Adverse events
In one trial enrolling 42 preterm infants, apnoea on day three of the intervention period was not statistically significantly different between treatment groups; however, on day seven, the naso-enteric group had significantly more recorded apnoea episodes. In this study, apnoea was defined as cessation of breathing for ≥ five seconds rather than the more conventional definition of ≥20 seconds, and, as detailed above, there was an age difference between the two groups. In the crossover trial, no significant difference between treatments was found for the outcomes bradycardia/oxygen desaturation (nasogastric: 1.6 episodes/hour, 95% CI [0.8 to 1.9]; orogastric: 1.0 episodes/hour, 95% CI [0.9 to 1.6]) or apnoea (nasogastric: 0.8 episodes/hour, 95% CI [0.7 to 1.2]; orogastric: 0.8 episodes/hour, 95% CI [0.5 to 1.2]). No other adverse events were reported on in the included trials.

Additional outcomes
Time to independence from supplemental oxygen was not statistically significantly different between groups in one trial of 46 very preterm or very low birth weight infants (MD 7.6 days, 95% CI [-15.2 to 30.4]). No other pre-specified outcomes were reported on.

5. Additional author observations*

Few trials were identified for inclusion in the review and sample sizes in all included studies were small. The overall methodological quality of the included trials was poor, with adequate allocation concealment methods described in only one trial. In addition, the parallel randomized controlled trial conducted in the early 1980’s had a baseline difference in age between treatment groups of four days, an important difference in the first week of life during which time postnatal weight loss occurs.

Current evidence on feeding tube placement in preterm or low birth weight infants is limited, but is not suggestive of any effect on time to establish full enteral feeding, time to regain birth weight, or on the frequency of apnoea, bradycardia or oxygen desaturation episodes. However, the trials contributing data for these outcomes were likely underpowered to detect an effect.

As there is currently insufficient evidence to inform practice or policy regarding the placement of enteral feeding tubes for preterm or low birth weight infants, further large randomized controlled trials are needed. A focus on the recruitment of very low birth weight and very preterm infants in future trials would be prudent, as these infants are most at risk of the adverse events associated with feeding tube placement.