Original Research: Nasogastric Tube Position and Intragastric Air Collection in a Neonatal Intensive Care Population 
Jacoba (Coby) de Boer MA, RN 
Bert J. Smit MD, PhD 
Rosalie O. Mainous PhD, ARNP, NNP-BC 

Advances in Neonatal Care
December 2009 
Volume 9 Number 6
Pages 293 - 298


PURPOSE: For neonates receiving intensive care, nasogastric tube feeding is essential. Since nasogastric tube placement techniques are not well standardized and common verification methods can be unreliable, placement errors may lead to unsafe situations. In mechanically ventilated neonates and neonates on continuous positive airway pressure, malpositioning of the nasogastric tube may prevent excess air within the stomach to escape. In this study, we aimed to relate tube position to amount of air. The hypothesis was: the better the position of the tube, the smaller the amount of air in the stomach.

SUBJECTS: A 1-year cohort of neonates in a level IIIc neonatal intensive care unit with a nasogastric tube.

DESIGN AND METHODS: We retrospectively reviewed 326 radiographs and classified nasogastric tube position and gastric air. Descriptive statistics were used to describe demographic data. Kendal's [tau] statistic was applied to explore the relationship between nasogastric tube position and amount of gastric air. A Mann-Whitney U test was performed to confirm the differences in gastric air in neonates with Ch5 and Ch6 gastric tubes and neonates with Ch8 gastric tubes.

RESULTS: One or both orifices of nasogastric tubes were in the esophagus in 7.1% of cases, tubes were curled up in the stomach in 35.3% of cases, and tube tips were beyond the pyloric sphincter in 5.5% of cases. Substantial or excessive air was found in 37.7% of cases. Kendal's [tau] value indicated that there was no significant correlation between nasogastric tube position and gastric air. The Mann-Whitney U value indicated that children with Ch5 and Ch6 gastric tubes had significantly more gastric air than children with Ch8 gastric tubes.

CONCLUSION: Nasogastric tubes were malpositioned in nearly half of cases, and substantial or excessive air was found in more than one-third of cases. The hypothesis-the better the position of the tube, the smaller the amount of gastric air-was not confirmed by the data. However, a significant relationship was found between tube size and gastric air.


Enteral feeding is an important component of neonatal care and is preferred over total parenteral nutrition. Enteral feeding supports development of the gastrointestinal system1,2 and secretion of gastrointestinal hormones,3 and enteral feeding of breast milk may decrease the incidence of neonatal sepsis.4 Until neonates are able to breastfeed or nipple all feeds, gavage feeding is the method of providing enteral nutrition.

Placement of an oral-gastric or nasal-gastric tube is considered to be relatively safe and easy. Common measurement techniques used to place a nasogastric tube include measuring from the tip of the nose to the earlobe and then to the xyphoid process, (NEX-method) from the tip of the nose to the earlobe to midway the xyphoid process and umbilicus (NEM-method),5 and from the tip of the nose to the earlobe to the umbilicus method (NEU-method).6 One study7 determined insertion lengths for neonates weighing up to 1500 g (13-17 cm for different weight groups). In daily practice, nurses may use many slightly different techniques that could broadly be described as from the nose to the ear to the stomach.

The abdominal radiograph is the only certain way to determine the correct position of the nasogastric tube. Radiation risk however precludes its regular use. Of the available means to ensure placement within the stomach, the auscultatory method (insufflation of air while auscultating for a gurgling sound over the epigastrium) is commonly used in neonatal care, despite its proven unreliability.8 In adult patients, injection of air into the tracheobronchial tree or into the pleural space could produce a sound indistinguishable from that produced by injecting air into the gastrointestinal tract.9-11 The other widespread method, that of examining the visual aspects of aspirates, was found inadequate as well, when applied as a single method.8

Because tube placement techniques are not well standardized and frequently applied verification methods are unreliable, placement errors are common. In adult patients, radiographic evaluation revealed tube tip positioning in the intestines, the esophagus, or the respiratory tract.10,12,13 Incidental pleural perforations have been described in the literature.10,14,15 In 3 pediatric studies, errors in placement ranged from 20.9% to 50.0%,16-18 and in 2 NICU studies,5,6 errors in placement ranged from 39.3% to 55.6%. Premature infants have also suffered from perforations by the feeding tubes. In a study of 75 children weighing less than 750 g, 3 pharyngoesophageal perforations were observed.19

Various problems can arise from inappropriate positioning that remains unnoticed. Feedings erroneously inserted into the esophagus can increase the risk of apnea, desaturation, bradycardia, and aspiration, with resulting morbidity.20,21 If the feeding tube bypasses the pylorus, transpyloric feedings may inadvertently be given. Transpyloric feeding in preterm neonates is associated with greater incidence of gastrointestinal disturbance than is gastric feeding, and there is some evidence of increased mortality.22 A tube curled up inside the stomach may falsely suggest no retention of feedings whereas in fact the orifices are above fluid level. When the tube orifices are in the fluid pool, excess air within the stomach cannot easily escape. Distension of the stomach and air inflated into the intestine, with abdominal distension, are frequently observed in ventilated neonates.

Patient care and patient safety would greatly benefit from refined, standardized insertion techniques and new (combinations of) verification methods that are as reliable as radiography and applicable to even our smallest patients.


In this study, we aimed to determine the rate of improper positions of nasogastric tubes in a NICU. Other goals of the study included the introduction of a classification system for gastric air and to seek evidence for the following hypothesis: the better the position of the tube, the smaller the amount of air in the stomach.


Setting and Sample

The setting was a level IIIc23 NICU in a large pediatric teaching hospital in the Netherlands, with approximately 620 admissions yearly. Total capacity of its 3 units is 24 predominantly ventilator-dependent newborns, mainly transitioning to high-care centers. We retrospectively reviewed abdominal or chest radiographs of children admitted and discharged in 2005 (N = 597). Radiographs of neonates with gastrointestinal anomaly, radiographs without a clearly visible nasogastric tube tip, and radiographs with tubes positioned in the stomach but without a visible greater curvature were excluded. The latter exclusion criterion was added because we defined an accurate position as "not bending along the greater curvature."


Eligible radiographs in the Picture Archiving and Communications System (PACS/Spectra 10.2) were reviewed according to previously set standards. In total, 326 radiographs were found suitable to be classified. If additional radiographs of the same neonate were available, only the first one that satisfied the criteria was used. Proper position was defined as tube tip 17 mm or more below the gastroesophageal sphincter (ie, both side orifices within the stomach), provided the tube was bending neither along the greater curvature nor in the intestine. From these rules, 5 categories were derived; only category 3 was considered appropriate (Table 1).

Table 1 - Click to enlarge in new window TABLE 1. Nasogastric Tube Position

Gastric air was classified according to exemplar radiographs (Figure 1). Table 2 presents an overview of the 5 different categories that represent the different amounts of air.

Figure 1 - Click to enlarge in new window FIGURE 1. Amount of air within the stomach with varied gastric tube placement.

Table 2 - Click to enlarge in new window TABLE 2. Intragastric Air


SPSS, Version 10.1 (SPSS Inc, Chicago, Illinois) for Windows was used for data analysis. Interobserver reliability for ratings of gastric tube position and amount of air was determined by the Kappa statistic. When raters disagreed initially, they jointly reached a final conclusion. Proportions of the 5 different tube tip positions and overall correct versus incorrect positions, as well as those of the different amounts of air, were calculated. Finally, as the data were ordinal and the distribution of tube positions was severely skewed (2.104; SE = 0.166), the degree of correspondence between gastric air and tube tip position was calculated with Kendall's [tau].

Of the 597 eligible neonates, 452 had at least 1 abdominal or chest radiograph. For 123 children, however, there was either no visible tube tip or no visible greater curvature and 3 children had gastrointestinal malformation. Consequently, 326 radiographs were assessed. Mean gestational age was 33 weeks + 1 day (range: 24-42 weeks/SD = 34 d). Mean birth weight was 1945 g (range: 450-5000 g/SD = 1039 g).

For classification of tube tip position, interobserver reliability achieved substantial agreement ([kappa] = 0.66).24 For classification of air, interobserver agreement was substantial as well ([kappa] = 0.69).24

Tube tip position was found to be improper in almost half of the cases (47.5%). Table 1 provides an overview of tube tips in the different positions.

Substantial or excessive air in the stomach was found for 123 neonates (37.7%).

An overview of gastric air classification for all tube positions and for accurate tube positions (N = 169) is shown in Table 2. In the latter subgroup, substantial or excessive of air was noted in 40.4% of cases.

To relate tube position to gastric air, tube positions were ranked regarding their expected ability to let escape excess air in descending order: (1) proper position within the stomach, (2) proximal orifice in esophagus and distal orifice in stomach, (3) both orifices in esophagus, and (4) tube tip beyond pyloric sphincter. Tubes bending along the greater curvature or curled up in the stomach were excluded, as the radiographs did not allow for distinguishing between tube tip position in the fluid pool or in air. No evidence of correlation between the 2 measures was found ([tau] = -0.021; P = .726).


Retrospective evaluation of the position of nasogastric tubes in a population of neonates in a level IIIc NICU23 revealed that nearly half of the tubes (47.5%) were in an inaccurate position. This figure implies that our methods, both for tube insertion and for confirming the expected position (injecting air and examining the visual aspect of aspirate), often failed. The tube tips were either too deep (40.5%) or too superficial (7.1%). Intestinal position was confirmed in 5.2% with a risk of gastrointestinal disturbance. In 7.1%, 1 or 2 orifices were in the esophagus, with risk of apnea, desaturation, bradycardia, and aspiration. In terms of patient safety, these are unacceptable proportions. Although the applied insertion technique could only roughly be described as "from the nose to the ear to the stomach," the findings are in line with outcomes of previous studies in comparable populations.5,6 One study5 compared 2 placement methods, the nose to ear to xyphoid process method and the nose to ear to midway between the xyphoid process and umbilicus method. Both procedures resulted in high proportions of incorrect placements (55.6% and 39.3%, respectively). All tube tips were too high; however, none of the tubes were even bending along the greater curvature. A remarkable detail that probably could explain the high positions was that measurement did not start at the nose tip but at the nostril. One study,6 applying the nose to ear to midway between the xyphoid process and umbilicus method, found that most of the improperly placed tubes were too deep (34.9%) and a remarkably smaller proportion was too superficial (11.6%), as was the case in our study.

Classification of gastric air showed substantial or excessive air in 124 neonates (37.7%). Such amounts are unpleasant for the neonate and increase the likelihood of air accumulating in the intestine as well. Little research has been done on the effects of gastrointestinal distension, but in a recent publication, the amount of gas in the hepatic portal vein on postmortem computed tomographic scan was significantly related to the amount of gastrointestinal distension after cardiopulmonary resuscitation.25

Large amounts of gastric air imply that for some reason, excess air did not sufficiently escape through the tube. Tube malposition, with orifices outside the stomach or into the fluid pool in the stomach, might possibly explain this phenomenon. Our hypothesis, "the better the position of the tube, the smaller the volume of air within the stomach," however, was not confirmed by the data. This outcome suggests that tube position was not the hallmark factor in air collection. Even for tube tips in a proper position in the stomach, the proportions for the different amounts of air collection were roughly comparable with those of the whole sample (Table 2).

Another explanation for air collection may be the very small diameter of the tubes for children weighing 3000 g or less Ch5 or Ch6 (for children weighing >3000 g Ch8). To check for this possibility, additional analysis was performed. The proportions for the different amounts of air were calculated separately for children weighing 3000 g or less and greater than 3000 g. In the latter group (N = 61), only 18.0% had substantial or excessive air collection, and 65.6% had no air to some air in the stomach. For children weighing 3000 g or less (N = 241), these proportions were 42.8% and 25.9%, respectively. To test whether these differences were statistically significant, a Mann-Whitney U test was performed, with "amount of air" as the test variable and "weight" (2 weight groups as representatives for the expected tube size) as the grouping variable. It confirmed that amount of air was significantly lower for children weighing more than 3000 g than for children weighing 3000 g or less (U = 3875.5; P = .000). To ascertain whether mechanical ventilation or nasal continuous positive airway pressure (both related to air collection in the stomach) and expected tube size (Ch5/Ch6 for children weighing <=3000 g, and Ch8 for children weighing > 3000 g) were independent, a [chi]2 test was performed. The result ([chi]2 = 1.40; P = .236) suggests that tube size is an important factor and that Ch5 and Ch6 tubes may simply provide too much resistance to let excess air escape spontaneously. Consequently, the air migrates into the intestines. Another possibility might be that the smaller nasogastric tubes (Ch5 and Ch6) are more easily blocked by feedings that stay in the tube are than the larger ones (Ch8).

Application of nasogastric tubes with cm-markings could help. Nurses can note the inserted length and when a radiograph (for another purpose) is made they can check the external tube position and adjust if necessary. Keep in mind however, that a proper position on the outside does not guarantee a correct tip position in the stomach.27


Our study is limited by the retrospective correlational design. As a consequence, we could not be entirely sure of the size of the tube. Generally, for children weighing 3000 g or less, a Ch5 or Ch6 gastric tube is applied, and for children weighing more than 3000 g, a Ch8 gastric tube is applied. Occasionally, however, a child weighing 3000 g or less may have had a Ch8 tube, or a child weighing more than 3000 g, may have had a Ch6 tube. A prospective study could give complete certainty with respect to tube size. Disadvantage of the correlational design is that it limits causal inferences.

The retrospective design, however, is an effective and efficient means of collecting data for explorative reasons. The findings with respect to tube positions, showing that nasogastric tubes were malpositioned in nearly half of cases, are in line with outcomes of previous studies in comparable populations.5,6 Another limitation is the 2-dimensional, 1-directional classification of air. In a prospective design, 3-dimensional or bidirectional classification could improve accuracy with respect to this measure.


As our method(s), both for tube insertion and for confirming the expected position (injecting air and examining the visual aspect of aspirate), often failed, leading to unsafe situations, new methods need to be developed and tested to determine gastric tube placement. Positive findings came from one study7 that compared fixed insertion lengths (13 cm/<750 g, 15 cm/750-999 g, 16 cm/1000-1249 g, and 17 cm/ 1250-1499 g) against the tip of the nose to earlobe to xyphoid process method. Accurate orogastric placements in children weighing less than 1500 g increased from 61% with the measurement method to 86% with the fixed insertion lengths.

With respect to gastric air, a Ch8 tube for the smallest neonates may not be an option, as this larger tube may compromise the small airway.26 A recommendation for daily practice however could be frequent manual removal of air in these small neonates with nasogastric tubes Ch5 and Ch6. Another recommendation could be that nurses frequently check for tube patency.


Because minimal insertion lengths for orogastric tube placement resulted in accurate tube placements in neonates weighing less than 1500 g of 86%, it is worth testing this method for nasogastric placement in children weighing less than 1500 g and to possibly expand it for application in the entire NICU population.

Because applied verification methods failed to detect tube malposition, more reliable (combinations of) methods suitable for neonates should be developed. With respect to gastric air, in a prospective study, frequent manual removal of air could be tested for its effectiveness.


Nasogastric tubes were malpositioned in nearly half of cases. Accurate placement of nasogastric tubes is not as easy as it seems. There is a definite need for an insertion procedure that improves the rate of accurate tube position. Despite the presence of an open nasogastric tube, more than one-third of neonates showed substantial or excessive gastric air. Not the position but the size of the tube (Ch5 or Ch6) influenced the amount of air that could escape from the stomach. Manual removal of air in the smaller neonates seems desirable.


The authors are grateful to Prof Dr Meradji for his advice on the classification of tube position and intragastric air. In addition, our appreciation goes to Harmke Boerman, Robin Tjon Pian Gi, and Stefanie van der Velden for systematically reviewing the radiographs.


1. Berseth C. Early feeding enhances maturation of the preterm small intestine. J Pediatr. 1992;120:947-953. [Context Link]

2. Berseth C, Nordyke C. Enteral nutrients promote postnatal maturation of intestinal motor activity. Am J Physiol. 1993;264:G1046-G1051. [Context Link]

3. Ainsly-Green A. Hormones and postnatal adaptation to enteral nutrition. J Pediatr Gastroenterol Nutr. 1983;2:418-427. [Context Link]

4. Fidel-Rimon O, Friedman S, Lev E, Juster-Reicher A, Amitay M, Shinwell E. Early enteral feeding and nosocomial sepsis in very low birth weight infants. Arch Dis Child fetal Neonatal Ed. 2004;89:F289-F292. [Context Link]

5. Weibley TT, Adamson M, Clinkscales N, Curran J, Bramson R. Gavage tube insertion in the premature infant. MCN Am J Matern Child Nurs. 1987;12(1):24-27. [Context Link]

6. Tedeschi L, Altimier L, Warner B. Improving the accuracy of indwelling gastric feeding tube placement in the neonatal population. Neonatal Intensive Care. 2004;16(1):16-18. [Context Link]

7. Gallaher KJ, Cashwell S, Hall V, Lowe W, Ciszek T. Orogastric tube insertion length in very low birth weight infants. J Perinatol. 1993;13:128-131. [Context Link]

8. Metheny NA, Dettenmeier P, Hampton K, Wiersema L, Williams P. Detection of inadvertent respiratory placement of small-bore feeding tubes: a report of 10 cases. Heart Lung. 1990;19:631-638. [Context Link]

9. Aronchick JM, Epstein DM, Gefter WB, Miller WT. Pneumothorax as a complication of placement of a nasoenteric tube. JAMA. 1984;252:3287-3288. [Context Link]

10. El-Gamel A, Watson DTC. Transbronchial intubation of the right intrapleural space: a rare complication of nasogastric intubation with a polyvinylchloride tube-a case study. Heart Lung, 1993;22:224-225. [Context Link]

11. Neumann MJ, Meyer CT, Dutton JL, Smith R. Hold that x-ray: aspirate pH and auscultation prove enteral tube placement. J Clin Gastroenterol. 1995;20:293-295. [Context Link]

12. Hendry PJ, Akyurekli Y, McIntyre R. Bronchopleural complications of nasogastric feeding tubes. Crit Care Med. 1986;14:892-894. [Context Link]

13. Sweeney J. How do I verify NG tube placement? Nursing. 2005;35(8):25. [Context Link]

14. Roubenoff R, Ravich WJ. Pneumothorax due to nasogastric feeding tubes. Report of four cases, review of the literature, and recommendations for prevention. Arch Intern Med. 1998;149(1):184-188. [Context Link]

15. Wendell GD, Lenchner GS, Promisloff RA. Pneumothorax complicating small-bore feeding tube placement. Arch Intern Med. 1991;151(1):599-602. [Context Link]

16. Ellett MLC, Beckstrand J. Examination of gavage tube placement in children. J Soc Pediatr Nurses. 1999;4(2):51-60. [Context Link]

17. Ellett ML, Maahs J, Forsee S. Prevalence of feeding tube placement errors & associated risk factors in children. MCN Am J Matern Child Nurs. 1998;23:234-239. [Context Link]

18. Scalzo AJ, Tominack RL, Thompson MW. Malposition of pediatric gastric lavage tubes demonstrated radiographically. J Emerg Med. 1992;10:581-586. [Context Link]

19. Filippi L, Pezzatti M, Poggi C. Use of polyvinyl feeding tubes and iatrogenic pharyngo-oesophageal perforation in very-low-birth weight infants. Acta Pediatr. 2005;94:1825-1828. [Context Link]

20. Giotoerrez F, Baffe D. Fluoroscopically guided nasogastric feeding tube placement: results of a one-year study. Radiology. 1991;178:759-762. [Context Link]

21. Metheny N. Minimizing respiratory complications of nasoenteric tube feedings: state of the science. Heart Lung. 1993;22:213-223. [Context Link]

22. McGuire W, McEwan P. Transpyloric versus gastric tube feeding for preterm infants. The Cochrane Database Syst Rev. 2005;(3):CD003487. [Context Link]

23. American Academy of Pediatrics, Committee on Fetus and Newborn. Levels of neonatal care. Pediatrics. 2004;114:1341-1347. [Context Link]

24. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33:159-174. [Context Link]

25. Shiotani S, Kohno M, Ohashi N, Yamazaki K, Nakayama H, Watanabe K. Postmortem computed tomographic (PMCT) demonstration of the relation between gastrointestinal (GI) distension and hepatic portal venous gas (HPVG). Radiat Med. 2004;22(1):25-29. [Context Link]

26. Sporik R. Why block a small hole? The adverse effects of nasogastric tubes. [Including commentary by A. D. Milner]. Arch Dis Child 1994;71:393-394. [Context Link]

27. Metheny NA, Titler MG. Assessing placement of feeding tubes. Am J Nurs. 2001;101(5):36-45. [Context Link]