Article Text

Download PDFPDF

Rapid home-based weaning of small children with feeding tube dependency: positive effects on feeding behaviour without deceleration of growth
  1. Markus Wilken1,
  2. Vanessa Cremer1,
  3. Jennifer Berry2,
  4. Peter Bartmann3
  1. 1Institute for Pediatric Feeding Tube Management and Weaning, Siegburg, Germany
  2. 2Spectrum Pediatrics, Alexandria, Virginia, USA
  3. 3Department of Neonatology, University of Bonn, Bonn, Germany
  1. Correspondence to Dr Markus Wilken, Institute for Pediatric Feeding Tube Management and Weaning, Hohlweg 4, Siegburg D-53721, Germany; mail{at}


Background Oral aversion and feeding disorders occur frequently after long-term feeding via a tube. A rapid home-based tube-weaning programme that provides feeding disorder treatment under medical supervision to overcome oral aversion has been developed.

Aims To describe the results of the 4–10-day rapid home-based feeding tube-weaning programme and to determine the prevalence of food aversion and the overall longitudinal effects on feeding behaviour and growth after treatment.

Subjects and Methods The parents of children who participated in the programme between 2005 and 2007 were followed up via questionnaire in 2008. The questionnaire included inquiries about the characteristics of tube feeding, feeding behaviour and growth parameters.

Results The study group comprised of 18 boys and 21 girls who, at entry, had a median age of 16 months (range, 5−57) and a median duration of tube feeding of 15 months (range, 3−56). The children were followed up for a median of 2 years after treatment (range, 1−3 years). The weaning programme established oral feeding in 89.7% of the children. In all cases where oral feeding was achieved, the feeding behaviour improved after treatment. The growth velocity of all children, as indicated by weight, length and body mass index (BMI), remained constant from the initial evaluation (BMI z-score M=−1.1 ± 1.7) to follow-up (BMI z-score M=−1.2 ± 1.1).

Conclusions This rapid weaning programme for children with feeding tube dependency (FTD) improved the feeding behaviour and was associated with constant growth velocity.

  • Nutrition
  • Child Psychology
  • Infant Feeding
  • Growth
  • Gastroenterology

Statistics from

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.

What is already known on this topic

  • Tube-dependent children are assumed to be at risk of oral aversion.

  • Treatment programmes provide controlled conditions for feeding tube withdrawal.

  • The data regarding feeding behaviour and growth velocity after withdrawal are limited.

What this study adds

  • Evidence showing that tube-dependent children exhibit marked oral aversion that had largely resolved at follow-up.

  • The majority of children selected for tube weaning over a 10-day period tolerated this without major ill effects and remain tube free at follow-up.

  • Tube withdrawal had no negative effect on long-term growth velocity.


Enteral feeding is a vital support for medically fragile infants unable to meet their nutritional requirements orally.1 ,2 It has been suggested that long-term tube feeding results in oral aversion, which in turn results in feeding tube dependency (FTD); this makes the tube withdrawal challenging.3–10 To establish an eating pattern where children can meet their nutritional and fluid needs orally, it will be necessary to address the oral aversion and food refusal.6 Otherwise, enteral feeding must be resumed to prevent excessive weight loss.3

Several inpatient and outpatient programmes that provide controlled conditions for feeding tube withdrawal have been published over the past two decades.3–9 However, to obtain these treatments, parents often have to travel for long distances, in some cases even fly overseas.5 Therefore, a new treatment modality was established by the Institute for Pediatric Feeding Tube Management and Weaning in 2005. This modality means that instead of the children having to come to the institute, the tube-weaning therapist travels to the child to implement a rapid tube-weaning programme in cooperation with local services. This treatment modality allows the children to stay in their home and establish appropriate feeding behaviour in a natural environment, thus reducing the need to travel and exposure to nosocomial infection.8

There is only limited evidence regarding the long-term outcomes of most tube-weaning programmes in terms of feeding behaviour6 and growth velocity.3 ,4 In particular, rapid tube-weaning programmes have been subjected to criticism because of the lack of evidence.3 This study followed up all families involved in the rapid home-based weaning programme.

The primary objective of the present study was to assess the efficacy of the rapid home-based weaning programme in this cohort of children with FTD. The secondary objectives were (1) to determine the prevalence of feeding disorder symptoms, (2) to measure the overall longitudinal effects of home-based rapid weaning on feeding behaviour and (3) to determine the impact of the tube-weaning programme on growth.

Subjects and methods

Parents or local services in Germany referred patients with FTD to the programme. Of the 55 patients referred to the programme, 45 were eligible and participated in the rapid weaning programme between 1 January 2005 and 30 December 2007 (figure 1 and table 1). At entry to the programme, the parents completed the Assessment Questionnaire for Feeding Disorder and Tube Weaning (AFT).11 In May 2008, the parents again received the AFT along with a parental informed consent form. The parents who did not respond within 2 weeks were telephoned to request their participation. Children were enrolled after parental informed consent was obtained. Follow-up data were obtained from 39 of the 45 patients; for eight patients, feeding data were retrieved from telephone interview; and for 31 patients, data were retrieved via the questionnaire (table 1). The median follow-up period was 2 years (range, 1−3 years). From 39 patients at entry, 18 were 15 months old or younger, 12 were between the ages of 16 and 30 months and nine were between the ages of 31 and 60 months. All children received outpatient feeding therapy. Seven also received inpatient intensive feeding therapy before being involved in this programme.

Figure 1

Flow chart of patients’ progress through the process.

Table 1

Medical diagnosis according to the classification system for complex feeding disorders of the 45 children who received the rapid tube-withdrawal programme between 2005 and 2007

Inclusion criteria were as follows: Patients who had a FTD, received at least 50% of their nutrition via feeding tube for a minimum of 3 months and were younger than 60 months.

The treatment exclusion criteria included the following: Unsafe swallowing ability, inability to meet the required oral intake and certain congenital metabolic abnormalities (see further details in the Results section).2

The Research Ethics Board of Bonn University Hospital provided written approval of the protocol.


Medical conditions: Medical reports were reviewed to identify the medical conditions of the patients and the patients were grouped according to the classification system for complex paediatric feeding disorders.12 The hunger induction plan and weight loss during the intensive wean phase was determined from the treatment reports. A paediatrician measured height and weight; body mass index (BMI) was calculated using the determined height and weight measurements and then matched to standardised German growth charts.13

The AFT-questionnaire comprises the following aspects:

Tube feeding data: Data was gathered regarding the patients’ tube feeding characteristics. These included feeding tube type; date of tube placement, daily nutritional, fluid and medication percentages, presented orally and via tube.

Feeding Adversity Scale (FAS): The FAS is a 4-point, Likert-scaled parental report questionnaire used to assess feeding. Ten items refer to the child's behaviour and four to parental behaviour. The validity of the scale has been estimated with r=0.72.14

Frequency of occurrence of feeding disorder symptoms: Parents reported the frequency of food refusal, vomiting, gagging, force-feeding, bizarre eating habits and oral motor dysfunction. The total sum of the symptoms was calculated as mean sum per day. If the symptoms occurred weekly or monthly, the total sum of the symptoms was converted to a daily occurrence.

Treatment programme

The programme was designed to establish sufficient oral intake by moving from food aversion to food acceptance and enjoyment. One psychologist worked full time for the programme (see box 1 for the programme details). All patients underwent multimodal assessment 1−4 months prior to treatment. During the assessment phase, the psychologist and a paediatrician from the local paediatric practice discussed eligibility and medical management. After finding that a child was eligible, the 5-day hunger induction phase was started 1−4 months after the assessment. This phase was then followed by the 4−10-day intensive phase, during which time the psychologist stayed closed to the patient's home and performed 3−6 home visits of 1−4 h duration at mealtimes. Medical supervision took place in the paediatrician's office. Depending on the child’s oral capacity, speech-language therapists were also involved. The psychologist monitored the children for 4 weeks after discharge and on an as-needed basis for the next 5 months.

Box 1

Phases of assessment and treatment in the rapid tube-withdrawal programme

ASSESSMENT (1−4 months prior to hunger induction)

Home visit consists of:

  • Parental interview

  • Play intervention

  • Feeding observation

Swallowing evaluation:

  • Clinical swallowing evaluation and videofluoroscopic study when clinical swallowing evaluation suggests aspiration risk

Review of the medical reports for medical conditions that meet exclusion criteria such as:

  • Condition impacting oral intake: untreated gastroesophageal reflux disease, dumping syndrome and re-feeding syndrome

  • Cardiorespiratory instability: uncorrected life-threatening heart or lung conditions

  • Metabolic diseases such as: glycogen storage diseases or ketolytic defects

  • Growth and weight: children below the third percentile for body mass index were excluded

In children who fulfilled the inclusion criteria, but not the exclusion criteria, the intervention programme was initiated 2 to 8 weeks later.


Total original fluids and nutrition amounts were reduced by 10% over the course of 5 days until 50% of the prewean amount was achieved. Oral feeding as well as daily contact with the feeding therapist was encouraged.


Nutrition intervention:

  • Intensive phase started on the last day of hunger induction.

  • Tube feeds were completely stopped or reduced to the medically indicated minimal amount.

Daily paediatric monitoring:

  • Monitoring of the child's overall health condition included: weight check, hydration and symptoms of hypoglycaemia.

  • Urinalysis was performed when symptoms of dehydration were observed. Fluids were supplemented via the tube until sufficient oral intake of fluids was achieved.

Feeding intervention:

  • The feeding situation was rearranged to avoid refusal triggers and to make the feeding environment more comfortable.

  • The children were fed according to their hunger and thirst signals.

  • Parents were assisted to comfort their children when they displayed stress.

  • Parents were assisted to feed slowly and in a non-intrusive manner.

  • Food refusal was accepted during mealtime and modified by using play intervention.

  • Oral desensitisation and swallowing treatment.

  • Strategies were introduced to increase child's swallowing ability and reduce oral hyper-sensitivity and aversion.

Parental intervention:

Parents were given psychological support during phase food refusal, weight loss or when traumatic memories of the child's medical history were triggered.

FOLLOW-UP (6 months)

  • Follow-up once a week for 4 weeks after the intensive phase of treatment.

  • Ongoing follow-up occurred for the next 5 months in cases of food refusal, infections or insufficient oral intake, as needed.

Treatment goal: Establish an eating pattern where children meet their nutritional and fluid needs orally.

Abort criteria: These were defined as a weight loss exceeding 10% of the child's starting weight, severe infection, cardiorespiratory instability or no improvement in oral intake during treatment.


The data were analysed using SPSS V.13.0 for Windows (SPSS, Chicago, Illinois, USA). Intergroup differences were evaluated using the t test for independent samples. Paired t tests were calculated for differences before treatment and at the time of follow-up. Pearson's correlation coefficient was calculated for inter-relationships between variables. χ2 test was performed frequency distributions. A p value of <0.05 was considered to be significant.


Baseline characteristics at programme entry:

Tube feeding: Twenty patients had a nasogastric tube, 16 had a gastrostromy and three had a jejunostomy tube for a median duration of 15 months (range, 3−56 months). Of these, 26 children had been tube fed since birth. Twenty-nine patients received 100% of their medication, fluids and nutrition via a feeding tube at entry.

Feeding behaviour: All patients exhibited symptoms of a feeding disorder that manifested itself on a daily basis and had been present for 4 months or longer. Daily food refusal and daily vomiting were reported in 38 and 23 patients, respectively. The FAS before treatment correlated with food refusal (r=0.39, p<0.05), vomiting (r=0.42, p<0.05), bizarre eating habits (r=0.46, p<0.05) and the total frequency of feeding disorder symptoms (r=0.45, p<0.05).

Exceptional enteral feeding during the hunger induction and intensive weaning phase: One patient was given an enteral feed of 10 grammes of maltodextrin plus an overnight glucose solution every 2 h due to metabolic disease. The nasogastric tube was replaced in two patients for rehydration purposes when urinalysis indicated an early state of dehydration. In one of these cases, treatment failed and the tube could not be removed during or post-treatment (see further details below).

Outcome measurement

From the 39 patients, 35 patients (89.7%) successfully transitioned from tube feeding to oral feeding (figure 1). One patient continued to receive her medication, and one patient continued to receive a glucose solution comprising 25% of his fluids via a tube due to a metabolic condition. Treatment was unsuccessful in four patients (10.3%). In one patient, the oral intake could not be improved (weight loss, 9%). While three patients showed improved feeding behaviour with weight loss between 7.7 and 8.1%, weight gain was not achieved and tube feeding had to be restarted 4–12 weeks post-treatment.

Feeding behaviour: Parents reported a significant decrease in feeding disorder symptoms at follow-up (table 2). After treatment, 42% of the patients displayed no symptoms, 30% displayed occasional symptoms and 28% continued to regularly display symptoms of a feeding disorder. The mean mealtime duration decreased from 41 min per meal to 20 min (p<0.004). The most frequent symptoms at the conclusion of treatment were inability to chew (n=4) and bizarre eating habits (n=2).

Table 2

Symptoms of feeding disorders before treatment and at follow-up 1 to 3 years after treatment

Growth outcome: Post-treatment weight loss data were available for 35 cases (90%) and indicated that the average weight loss was 6.4 ± 2.9% (range, 0−11.6%). In two cases, weight loss exceeded the 10% level. Treatment was not aborted as the children had already shown major improvement in feeding behaviour and gained back lost weight rapidly. There was no significant association of weight loss and growth, with feeding behaviour, gender or age. Differences for weight, height and BMI before treatment and at follow-up did not reach significance for the entire sample population or for the successfully treated patients (table 3).

Table 3

Tube feeding and growth characteristics before treatment and at the time of follow-up 1–3 years after treatment


This report describes a prospective study on the effects of a rapid feeding tube-weaning programme on the successful weaning, feeding behaviour and growth of patients with FTD. This cohort contained a broad range of medical conditions that are known to be indicators for tube feeding. Three aspects are of particular interest.

Feeding behaviour

Food refusal and food aversion have been described as the key symptoms of FTD.3–9 To our knowledge, the frequency of FTD symptom occurrence has not been measured previously. This is the first study which shows the occurrence of feeding disorder symptoms in patients with FTD, such as daily food refusal, vomiting and frequent gagging, detailing that these symptoms can persist for months to years. Our study also showed that the rapid weaning programme significantly improved these target symptoms. Indeed, over a relatively short treatment period of 5–10 days, oral feeding was established in 90% of the patients. The small percentage of patients in whom rapid weaning failed was similar to or smaller than the failure rates of other treatment modalities.3 ,4 ,6 ,7 Over the long term, the programme was associated with a marked decrease in the frequency of the predominant symptoms of FTP. In particular, food refusal decreased from occurring on a daily basis to occurring on a monthly basis.

Growth and body composition

There have been concerns about the possibility of weight loss during rapid weaning.3 ,6 However, in this cohort, the short-term weight loss was moderate when compared to 3.1–9.2% in other studies.3–5 In 23% of the patients, low-level supplemental tube feeding was provided for the first 5 days. This may prevent aversive patients from losing weight. The present study suggested that rapid weaning did not have a long-term negative impact on growth, weight gain or BMI. The long-term growth pattern in our study in terms of z-score is very similar to the growth pattern of children who are weaned gradually.3 Nevertheless, due to the possibility of short-term weight loss, it may be necessary to follow-up the patients intensively to ensure that the child's feeding behaviour is progressing positively and that the growth rate does not drop.

Treatment modalities

While the findings in this study are encouraging, the appropriateness of generalising the results must be evaluated carefully. A control group was not included in this study because finding matching controls in a heterogeneous sample of children might not be possible.15 To control the sample heterogeneity in our data analysis, it was decided to let the children in the study serve as their own control. It should be noted that the study included young patients, whom have been suggested to be easier to wean.3 ,5 ,6 Two studies show that the success rate in younger patients is comparable or lower with 47–80% to the success rate of older patients.4 ,9 The rigorous selection process may have influenced the outcome but a recent follow-up study with less rigorous selection methods did not observe different outcomes.16 While it has been questioned whether weaning programmes have to be intense,9 studies show that a high treatment intensity is associated with a higher success rate,5 ,8 less weight loss5 and more rapid tube withdrawal5 ,8 ,16 than programmes with a lower treatment intensity.3 ,6 ,7 ,9

This programme is well established in Germany, where treatment costs are mostly covered by the health system. A home-based tube-weaning programme was also established in the USA in 2011, while home-based weaning are applied in private practice in other countries.16 The treatment costs are relatively low ($9000−$25 000) compared to inpatient programmes ($17 000−$130 000)7 or continued tube feeding ($16 320–54 000 per year).7 Moreover, it could be argued that the comfortable home environment may have positively influenced the outcome, especially when one considers the high rate of traumatised parents and children in tube-feeding cohorts.17 In conclusion, the present findings suggest that a rapid enteral withdrawal programme is effective and that its long-term outcome is encouraging.


We are grateful to the participating families and to the paediatricians who carefully guided the parents through the tube-withdrawal procedure.



  • Contributors MW conceived and designed the study and acts as guarantor. VC collected and calculated the data and wrote the first manuscript draft. JB undertook the initial literature review. PB and MW advised and trained VC in measurement techniques and helped in patient recruitment and study execution. All authors contributed to the writing and revision of the manuscript.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval Ethics Committee at Rheinischen Friedrich-Wilhelms-Universität Bonn.

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

  • Data sharing statement Data regarding psychosocial stress are available from Dr Martina Jotzo (Child Hospital Darmstadt) as an SPSS file.

Linked Articles

  • Editorial
    Charlotte M Wright
  • Atoms
    R Mark Beattie