Article Text

Improved radiological assessment of neonatal feeding tubes
  1. Daniel Quandt1,
  2. Egil Brøns1,
  3. Philipp Meyer Schiffer1,
  4. Thomas Schraner2,
  5. Hans Ulrich Bucher1,
  6. Romaine Arlettaz Mieth1
  1. 1Division of Neonatology, University Hospital Zurich, Switzerland
  2. 2Department of Diagnostic Imaging, Children's Hospital Zurich, Switzerland
  1. Correspondence to Daniel Quandt, Division of Neonatology, Department of Obstetrics & Gynecology, University Hospital Zurich, Frauenklinikstrasse 10, 8091 Zurich, Switzerland; daniel.quandt{at}


Background In about one-fifth of radiographs performed in neonates, no exact gastric feeding tube position can be defined.

Objectives To determine whether injection of air via feeding tube before taking radiographs improves radiological assessment of its position.

Methods In the study group (n=153), air was injected via gastric feeding tube before taking a radiograph. The tube position on radiographs was compared with a blinded control group (n=381) with no injection of air.

Results The definition of exact gastric tube position was possible in 95% of the study group compared with 78% in the control group (p<0.001).

Conclusion Injection of air before taking a radiograph significantly improves the definition of exact gastric feeding tube position in neonates.

Statistics from

What is already known on this topic

  • Radiographical control is considered to be ‘the gold standard’ for checking the gastric feeding tube position in neonates, especially in doubtful situations.

  • Even though radiographical control is considered the ‘gold standard’, it is not possible to indicate with certainty the position of the gastric tube in about 20% of the cases.

What this study adds

  • Standardised injection of air before taking the radiograph allows better stomach visibility and thereby significantly improves the definition of the exact gastric feeding tube position.


Accurate positioning of orogastric and nasogastric feeding tubes is important to ensure safe and effective enteral feeding in neonates. Radiographs, though not routinely performed to check tube placement, are actually considered to be the gold standard.1 In a previous study, we analysed 381 radiographs retrospectively.2 We found that in one-fifth of the radiographs, no exact statement regarding tube position was possible, because the gas bubble of the stomach was not always visible and radiographs in neonatal intensive care units are usually performed only in an anterior–posterior plane. The aim of this study was to examine if injecting air through the feeding tube into the stomach immediately before taking the radiograph, a method already used in older patients to verify placement of percutaneous endoscopic gastrostomy tubes,3 improves its visibility on radiographs. We hypothesised that injection of air improves the visibility of the stomach and thereby allows better definition of exact gastric tube placement in neonates.


This is a prospective observer-interventional study. The study group consisted of 153 consecutive radiographs of 105 neonates. In this cohort, 5 ml of air in preterm infants and 10 ml in term infants were injected by a neonatologist via lying gastric feeding tube (metric/x-ray manufactured by Unomedical, usually CH 04 to CH 06) within 5 s before taking a radiograph. Tubes had a radio-opaque marker with a line running along the entire length with an interruption at the side hole. Feeding tube positions on radiographs were then assessed and compared with a control group (n=381 radiographs of 173 neonates) having no injection of air. A minimal sample size of 148 for each group was calculated on the hypothesis that the rate of 21% of ‘undefinable’ feeding tube positions from a previous study2 could be reduced by 50% (one-sided α 0.05, β 0.2).

All radiographs were standardised antero-posterior planes comprising chest and abdomen. They were performed for other medical reasons (eg, respiratory distress syndrome and tracheal intubations) not to check gastric feeding tube placement alone. Definition of tube position was performed as described in a previous study and as illustrated in figure 12. If the tip or orifice of the gastric feeding tube was not identifiable, the tube position was defined as being ‘undefinable’. The radiological assessment of both groups (study and control group) was performed by two observers (QD, BE) after all radiographs were blinded, meaning that the observers did not know which group the radiographs belonged to. To test retest variability, all radiographs were scored twice by the same rater.

Figure 1

Definition and illustration of different feeding tube tip positions on radiographs. The dotted lines mark the alignments of zoning structures visible on radiographs. Position 1: tube tip or side orifice above diaphragm and gastroesophageal junction. Position 2: tube tip or side orifice under diaphragm or gastroesophageal junction, not in the body of the stomach. Position 3: tube tip or side orifice in the body of the stomach (correct placement) Position 4: Tube tip bending along or touching greater curvature, or tip at pyloric sphincter region. If the tip or orifice of the gastric feeding tube was not at all identifiable, the tube position was defined as being ‘undefinable’. Adapted from2.

All neonates were hospitalised in the Division of Neonatology at the University Hospital Zurich. Insertion depth of the feeding tubes was estimated by the Nose-Ear-Xiphoid (NEX) method,4 measuring the distance from the tip of the nose to the earlobe and then to the lower end of the xiphoid process. All data were processed anonymously and parental consent was not required. The study was approved by the Ethical Commission of the Children's Hospital in Zurich.

Statistical analysis was performed using the Cohen's κ test to calculate inter-rater and intrarater agreement and nominal logistic regression analysis for group comparison.


A total of 534 radiographs from 278 infants (study group: median gestational age 30 5/7 weeks, range 25–42 weeks, median birth weight 1460 g, range 530–3980 g) were analysed, 153 in the study group and 381 in the control group. Intrarater agreement of the scoring system for definition of gastric tube tip positions was 0.96 and inter-rater agreement was 0.95.

Definition of exact gastric tube position was possible in 95% in the study group (145/153) compared with 78% in the control group (297/381), leading to a significant reduction of indefinable gastric tube positions (p=0.001). Distribution of other predefined tube positions did not show significant differences between the two groups. All results are illustrated in figure 2.

Figure 2

Bar diagram illustrating percentages of different gastric tube positions (as defined in figure 1) in comparison between the two samples. Significant difference between the two samples in position 0=indefinable tube tip position.

In the study group (n=153), a total of 77 radiographs demonstrated incorrect placement. Of those, 65% (50/77) had either the tube tip touching or bending along the greater curvature of the stomach or were located at the pyloric sphincter region (position 4). Eight per cent (6/77) of the malpositioned feeding tube tips were located in the upper oesophagus (position 1, in projection above the diaphragm on radiographs). No displacement into the trachea was observed.

There was no statistically significant association with tube position for the factors gender, gestational age, birth weight, route for application of the tube (nasal or oral) nor mechanical ventilation.

No side effects possibly related to the injection of air like pain, vomiting, ventilation problems or gastric perforation was observed during the study period.


In this study, we demonstrated that a standardised procedure with the injection of air before taking a radiograph improves the visibility of the stomach's gas bubble and reduces the number of indefinable gastric tube positions on radiographs (5% vs 22%). This procedure is easy to perform. Further, no side effects possibly related to this procedure occurred during the study period. To our knowledge, there are no descriptions of complications related to injection of air via lying gastric feeding tube in the literature. We therefore suggest this effective procedure as a clinical tool to improve stomach visibility on radiographs.

One limitation of this study is the small sample of the study group. Thus, it is not possible to demonstrate that the intervention is definitively safe. Nevertheless, precise documentation of adverse events did not show any complications related to the procedure.

In the future, sonography might become an alternative method to identify the position of the tip of gastric feeding tubes. Advantages like lack of radiation and bedside availability in some units are promising, but before becoming a standard assessment test for neonatal feeding tube placements, its reliability and practicability in neonatology remain to be determined.5


Insufflating air before taking radiographs in neonates improves the visibility of the stomach's gas bubble and thereby the definition of exact gastric feeding tube positions on radiographs. This method is easy to perform, and its practicability in clinical routine is warranted.


The authors thank Bill Hesselink for his valuable comments and corrections of this manuscript.


View Abstract


  • Competing interests None.

  • Ethics approval Ethical Commission of the Children's Hospital in Zurich, Switzerland.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.