Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Gastrostomy in Infants With Neonatal Pulmonary Disease

Guimber, D.*; Michaud, L.*; Storme, L.; Deschildre, A.; Turck, D.*; Gottrand, F.*

Author Information
Journal of Pediatric Gastroenterology and Nutrition: April 2003 - Volume 36 - Issue 4 - p 459-463
  • Free



Advances in neonatal intensive care using surfactant and new ventilatory strategies (high frequency oscillation ventilation, extracorporeal membrane oxygenation) have led to improved survival among babies with respiratory distress syndrome. However, many of these infants still develop chronic lung disease, whether the bronchopulmonary dysplasia (BPD) in very low birth weight (VLBW) or congenital diaphragmatic hernia (CDH).

Impaired growth and malnutrition is a common problem in patients with BPD (1–3); the reasons for failure of growth in patients with BPD can usually be assigned to one or both of two categories: inadequate nutritional intake and/or inadequate oxygenation. Inadequate nutritional intake may have several explanations: patients with severe BPD may be poor oral feeders, may have a higher metabolic rate, or may exhibit oral aversive behavior. Nutritional status depends on the maintenance of a positive balance between energy intake, and total energy expenditure plus energy losses. Growth-related energy cost, estimated to be 3 to 5 Kcal/gram of newly synthesized tissue, must be added to the latter. Oral caloric intake is often too low to fulfill the energy requirements of these infants, leading to a negative balance and the to a need for supplemental enteral nutrition. Gastrostomy feeding has been shown to be an efficient and easy technique for providing additional caloric intake over the long term.

To the best of our knowledge, there is no publication on the use of gastrostomy feeding in infants with chronic lung disease of neonatal onset (i.e., for infants with VLBW or CDH). The aim of this retrospective study is to report on the efficacy and side effects of gastrostomy feedings in these patients.


The retrospective study population consisted of 13 children born between 1992 and 1998 (female/male = 7/6) with a median age of 13 months (range: 8–35). All children had a history of chronic lung disease of neonatal onset, and presented with severe growth retardation (weight-for-height below 85%) when referred to us for gastrostomy placement. The study population included: four children with CDH at birth who required extracorporeal membrane oxygenation, and nine preterm infants with VLBW (birth weight ≤ 1000g). Children with VLBW had BPD according to following criteria: need for supplemental oxygen, physical findings such as tachypnea, retractions, crackles, or wheezing, abnormal chest radiograph on or after the 28th day of life. All babies required at least 3 days of mechanical ventilation (4). None showed any neurologic abnormalities. Inadequate oxygenation was observed in none of the patients during the entire study period. At the time of referral, respiratory status was stable in all patients. Infants had their oxygen saturation levels checked while awake, during and after feeding, and during sleep. Supplemental oxygen was used at home in 5 children to keep oxygen saturation measured by pulse oximetry in the 95% range (5). Oxygen was administered at median flow rate of 0.5 L/min for three months after gastrostomy placement in four patients, and for 25 months in one patient who had CDH. Eight children did not need any oxygen supplementation during the study period. Intercurrent respiratory illnesses transiently caused arterial desaturation in six children, necessitating the temporary restitution of low-flow oxygen therapy during the study period. All patients were weaned from oxygen supplementation at the completion of study. Inhaled bronchodilators were used in 11 patients. No children received diuretic therapy.

The following items concerning the neonatal period were recorded: birth weight, age at start of enteral feeding, age at end of enteral nutrition, duration of mechanical ventilation, and duration of hospitalization until discharge. The characteristics of all patients are summarized in Table 1.

Characteristics of patients

Anthropometric data and caloric intake, expressed as a percentage of the recommended daily allowance for age (RDA) (6), were recorded at discharge from the neonatal unit (median duration of first hospitalization: 195 days; range: 85–556), at the start of enteral nutrition (EN), and after 4 months of EN. Weight and height were converted into Z-score by relating appropriate reference growth data for decimal age and sex, according to French growth charts (7).

Gastroesophageal reflux (GER), was assessed by 24 hpHmetry (Digitrapper Mark II, Synetics) in all patients. Endoscopy was performed in any child suspected of having esophagitis (n = 8) during percutaneous endoscopic gastrostomy (PEG) placement. The presence or absence of esophagitis was noted during this procedure. In nine patients, foregut motility was also assessed by esophageal manometry before carrying out gastrostomy feeding. After 4 months, if weight gain was not observed, inadequate oxygenation and GER were ruled out before increasing EN.

The feeding history included questions to determine whether oral aversive behavior had occurred previously. Moreover, all infants were observed by an experienced speech pathologist who assessed the acceptance of nipple feeding, quality of sucking and swallowing, oral hypersensitivity, and satiety.

As a result of feeding difficulties and/or failure to thrive, all the infants received EN by means of gastrostomy feeding. Percutaneous endoscopic gastrostomy (PEG) was performed by the pull-through method in 11 patients (CDH n = 4; VLBW n = 7). A button device (Bard, Trappes, France or Mic Key, Draper, U.S.A.) was placed in all children at a median of four months after gastrostomy.

Enteral feeding began six hours after the placement of the PEG and 12 to 24 hours after surgical gastrostomy. Enteral solution was provided using a peristaltic pump. Feeding was started in hospital, and slowly increased over the next five days. Feeding was given by continuous infusion through the gastrostomy tube over 10 to 12 hours during the night. Children were encouraged to receive additional food during the daytime. A polymeric diet especially created for children (Nutrison Paediatric, Nutricia, France) was used. The energy density of the food ranged between 1 and 1.5 Kcal/ml depending on the fluid allowance for the child and the mean osmolarity was 360 mOsm/kg. The fluid allowance was altered appropriately in conditions of vomiting or diarrhea.

Before the child's discharge from the hospital, parents (n = 13) were taught to make up the enteral nutrition, care for the feeding tube, and control the infusion pump. EN could be performed at home for all patients.

Statistical Analysis

Anthropometric data were compared at discharge from the neonatal unit, at the beginning of EN, four months after the start of EN, and at the end or at maximum follow-up of EN by the paired non-parametric test of Wilcoxon. P values < 0.05 were considered statistically significant. All data are expressed as median (and range).


Anthropometric data are presented in Table 2 and Figure 1. The weight-for-height Z-score (Z W/H) decreased significantly from −2 to −3 during the period between discharge from the neonatal unit and the start of enteral nutrition (P = 0.01). After four months of EN, Z W/H, the weight-for-age Z-score (Z W/A), and the height-for-age Z-score (Z H/A) increased significantly from −3 to −1.8 (P = 0.002), from −3.7 to −2.7 (P = 0.002), and from −3 to −2.3 (P = 0.04), respectively. Caloric intake (summarized in Table 2) increased significantly after four months of EN from 100% to 140% of RDA (P < 0.05).

Growth, nutritional data, and caloric intake during follow-up (median, range)
FIG. 1.:
Changes in Weight/height Z scores in children with pulmonary disease of neonatal onset before and after enteral nutrition via gastrostomy.

When the infants were first referred to our nutritional unit, they all had clinical GER. All children in the CDH group and three in the VLBW group had esophagitis. Seven of the 13 children (three children in the CDH group and four in the VLBW group) required fundoplication because of the severity of GER (multiple episodes of aspiration, persistent esophagitis, pneumonia, apnea, poor growth velocity, and failure to thrive), following medical treatment with antisecretory agents. The median age at operation was 12 months (range: 4–35 months). Foregut dysmotility (achalasia (n = 1) or esophageal dyskinesia (n = 5)) occurred in six children (CDH n = 4; VLBW n = 2).


The median follow-up was 30 months (range: 8–54). EN was stopped in seven children (CDH n = 1; VLBW n = 6) at a median age of 43 months (range: 37–64 months) because of improvement in growth after a median duration of 30 months (range: 20–43 months). For these seven children, Z W/A and Z H/A increased significantly from −2.8 to −2.3 and from −2.8 to –2, respectively. For the other six patients (46% of the study population), the median duration of last follow-up was 34 months (range: 8–54 months); ratio CDH/VLBW was 75%. EN had to be maintained because of poor oral intake and oral aversive behavior, rather than pulmonary disease. Indeed none of these children had inadequate oxygenation or major GER at that time.

The main problem occurring during EN was an aggravation of oral aversive behavior in 7 of 13 children (3 CDH, 4 VLBW) especially in those who were hospitalized and ventilated for a long time. Symptoms included a lack of sucking or swallowing (n = 3), early satiety (n = 2), or oral hypersensitivity (n = 2). Three of seven children later developed severe anorexia nervosa.

Late-onset complications included an expelled button that was covered by gastric mucosa (n = 1), gastric metaplasia, granulation tissue around the site of the gastrostomy (n = 3), and a delay in gastrostomy closure after removal of the tube (n = 1). Both gastrostomy and enteral feeding were well tolerated with no adverse clinical events such as diarrhea or chronic vomiting. The rate of GER esophagitis was not increased following the gastrostomy.


Our study shows that infants who survive the sophisticated techniques of neonatal intensive care with severe neonatal chronic lung disease may further develop severe and prolonged nutritional and gastrointestinal complications. Growth failure is a well-known problem in infants with chronic lung disease of neonatal onset but its determinants have not yet been fully elucidated. Disturbances in energy balance have been implicated in growth failure with BPD (1–3,8). The reasons why patients with BPD fail to grow can usually be assigned to one of two categories: inadequate nutritional intake or inadequate oxygenation. Inadequate nutritional intake has several explanations (9): patients with significant BPD tend to be poor oral feeders and their caloric intake may be too low relative to their caloric requirements (which range between 120 and 180 Kcal/Kg/day) because of the anorexia accompanying illness, fatigue due to residual respiratory, or a prescribed fluid restriction. Children with growth failure and BPD have a higher metabolic rate than children with BPD who do not have growth failure (10,11). In our patients, the main mechanism of malnutrition seems to be related to the insufficient energy intake secondary to anorexia and poor tolerance of oral feeding (GER). Z H/A, Z W/A, and Z H/A increased significantly when RDA increased significantly from 100% to 140%.

Reduction in nutrient intake and impairment in growth can be observed early in the course of neonatal chronic lung disease whatever the initial illness. Severity of illness and/or poor gut function appear to limit the quantity of both enteral and parenteral nutrition, as well as delaying the onset of an adequate EN. Our results show that malnutrition can be present during the initial hospitalization (12), after discharge, or during both periods. The nutritional status of these infants worsened after discharge and no spontaneous catch-up growth could be observed during the following months, despite the fact that they were returning home with an improved respiratory status.

Our population is heterogeneous and includes preterm infants with BPD who may have small stature and low weight during most of their childhood as a result of prematurity alone (13). Babies with CDH treated by ECMO are also known to exhibit growth failure (up to 35% in some studies) (14,15). When these patients were first evaluated in our nutritional unit, both populations presented with almost the same nutritional status, as well as with oral and digestive abnormalities. This suggests that impairment in growth and malnutrition are independent of the initial illness and are related more to the severity of the initial respiratory disease and the necessity for aggressive and prolonged intensive neonatal care, and oxygen dependence. In our study population, malnutrition was most likely due to inadequate caloric intake. Before EN, the patients received 100% of RDA, an amount probably insufficient for the needs related to the demands of their chronic pulmonary disease despite the fact that inadequate oxygenation was not observed in any patients.

Enteral nutrition via gastrostomy is efficient and well tolerated, and provides improvement of caloric intake and nutritional status. High efficacy and a low rate of complication suggest that PEG should be considered more frequently for cases of malnutrition related to chronic lung disease (16). Considering that most of our patients experienced a worsening in their nutritional status during their first year of life, we recommend that PEG be considered early in the course of treatment.

In this study, the median follow-up of patients after gastrostomy was 30 months (range 8–54 months) and only seven patients could be definitely weaned from EN. Management of the remaining children seemed to be long-term and their weaning from EN was not possible because of the reappearance or worsening of abnormal feeding behavior. Feeding aversion and feeding difficulties could be explained by the following factors in our patients: hospitalization, lack of feeding experience, or GER. Infants frequently exhibit oral aversive behavior after prolonged periods in neonatal intensive care units, since almost all oral stimuli are noxious in nature (laryngoscopy, positioning and taping of nasogastric tubes, and suctioning) (17). After recovery from an acute illness, infants may continue to interpret oral stimuli as noxious and may not accept nipple feeding. Emesis may occur and the oral aversion behavior may be misdiagnosed as GER. Early non-nutritive sucking and other forms of feeding therapy may possibly be helpful (18). Lack of feeding experience during critical sensitive periods (19) (as with long-term parenteral or enteral feeding), may contribute to the risk of developing anorexia nervosa. Three children in the present study developed anorexia nervosa requiring prolonged psychiatric care. All the children had GER, and 7 of 15 required fundoplication. Fundoplication-gastrostomy has been already reported to be effective in facilitating growth and oral feeding in patients with severe BPD and GER (20,21).

Our study suggests that early gastrostomy should be considered for certain babies requiring neonatal intensive care for severe pulmonary disease. However, EN via gastrostomy is not the only method of improving the poor oral energy intake and growth delay in these infants. Attention must be given to GER and foregut dysmotility, as well as to sucking abnormalities, aversive behavior, and psychologic difficulties. A multidisciplinary approach starting during the neonatal period and including the neonatologist, gastroenterologist, dietitian, respiratory specialist, psychiatrist, surgeon, and speech therapist may prevent the appearance of growth delay and malnutrition caused by feeding difficulties in these infants.


1. Kurzner SI, Garg M, Bautista DB, Bader D, Merritt RJ, Warburton D, et al. Growth failure in infants with bronchopulmonary dysplasia: nutrition and elevated resting metabolic expenditure. Pediatrics 1988; 81:379–84.
2. de Regnier RA, Guilbert TW, Mills MM, Georgieff MK. Growth failure and alterated body composition are established by one month of age in infants with bronchopulmonary dysplasia. J Nutr 1996; 126:168–75.
3. Wilson DC, McClure G, Halliday HL, Reid MM, Dodge JA. Nutrition and bronchopulmonary dysplasia. Arch Dis Child 1991; 66:37–8.
4. Northway WH. Bronchopulmonary dysplasia: twenty-five years later. Pediatrics 1992; 89:969–73.
5. Groothuis JR, Rosenberg AA. Home oxygen promotes weight gain in infants with bronchopulmonary dysplasia. Am J Dis Child 1987; 141:992–5.
6. FAO/WHO/UNU. Energy and protein requirements. Geneva:WHO. WHO technical report Ser. 724. 1985.
7. Sempé M, Pedron G, Roy-Pernot MP. Auxologie: méthodes et séquences. Paris: Théraplix 1979.
8. Markestad T, Fitzhardinge PM. Growth and development in children recovering from bronchopulmonary dysplasia. J Pediatr 1981; 98:597–602.
9. Ryan S. Nutrition in neonatal chronic lung disease. Eur J Pediatr 1998; 157(Suppl 1):S19–S22.
10. de Meer K, Westerterp KR, Houwen RH, Browers HA, Berger R, Okken A. Total energy expenditure in infants with bronchopulmonary dysplasia is associated with respiratory status. Eur J Pediatr 1997; 156:299–304.
11. Davidson S, Schrayer A, Wielunsky E, Krikler R, Lilos P, Reisner S. Energy intake, growth, and development in ventilated very-low-birth-weight infants with and without bronchopulmonary dysplasia. Am J Dis Child 1990; 144:553–9.
12. Berry MA, Abrahamowicz M, Usher RH. Factors associated with growth of extremely premature infants during initial hospitalization. Pediatrics 1997; 100:640–6.
13. Vrlenich LA, Bozynski ME, Shyr Y, Schork A, Roloff DW, McCormick MC. The effect of bronchopulmonary dysplasia on growth at school age. Pediatrics 1995; 95:855–9.
14. Lund DP, Mitchell J, Kharasch V, Quigley S, Kuehn M, Wilson JM. Congenital diaphragmatic hernia: the hidden morbidity. J Pediatr Surg 1994; 29:258–62.
15. Nobuhara KK, Lund DP, Mitchell J, Kharasch V, Wilson JM. Long term outlook for survivors of congenital diaphragmatic hernia. Clin Perinatol 1996; 23:873–87.
16. Behrens R, Lang T, Muschweck H, Richter T, Hofbeck M. Percutaneous endoscopic gastrostomy in children and adolescents. J Pediatr Gastroenterol Nutr 1997; 25:487–91.
17. Bier JA, Ferguson A, Cho C, Oh W, Vohr BR. The oral motor development of low-birth-weight infants who underwent orotracheal intubation during the neonatal period. Am J Dis Child 1993; 147:858–62.
18. Di Pietro JA, Cusson RM, O'Brien-Caughy M, Fox NA. Behavioral and physiologic effects of nonnutritive sucking during gavage feeding in preterm infants. Pediatr Res 1994; 36:207–14.
19. Rudolph CD. Feeding disorders in infants and children. J Pediatr 1994; 125:S116–S124.
20. Giuffre RM, Rubin S, Mitchell I. Antireflux surgery in infants with bronchopulmonary dysplasia. Am J Dis Child 1987; 141:648–51.
21. Hyman PE. Gastroesophageal reflux: one reason why baby won't eat. J Pediatr 1994; 125:S103–S109.

Gastrostomy; Malnutrition; Bronchopulmonary dysplasia; Enteral nutrition

© 2003 Lippincott Williams & Wilkins, Inc.