Children with neurological impairments frequently experience feeding difficulties, which can lead to malnutrition and growth failure. Gastrostomy feeding is now the preferred method of providing nutritional support to children with neurological impairments who are unable to feed adequately by mouth. Complications may arise as a result of gastrostomy placement, and the development or worsening of gastro-oesophageal reflux (GOR) has been widely reported. This has led to the frequent use of surgical antireflux treatment in the form of a fundoplication, or other antireflux procedures. Fundoplication is associated with a high recurrence rate, surgical failure, and significant morbidity and mortality.
Since proton pump inhibitors (PPIs) were introduced in the 1990s, they have come to play a larger part in the medical management of GOR in children with neurological impairments. Uncontrolled studies suggest that PPIs may be a safe, appropriate treatment for GOR. Other agents currently used include milk thickeners, acid suppression drugs, acid buffering agents, gut motility stimulants and sodium alginate preparations.
There are risks and benefits associated with both surgical and medical interventions and further comparison is necessary to determine the optimal treatment choice.
To compare the effectiveness of antireflux surgery and antireflux medications for children with neurological impairments and GOR who are undergoing placement of a gastrostomy feeding tube.
We searched the following databases on 23 March 2012: the Cochrane Central Register of Controlled Trials (CENTRAL), Ovid MEDLINE, EMBASE, CINAHL, LILACS and ISI Web of Science. Previously, we searched the Child Health Library in June 2009. We also performed online searches of trial registries, medical journals, conference proceedings, dissertations and theses. We contacted specialists in the medical and industry setting for knowledge of completed or ongoing trials.
We sought to include randomised controlled trials that recruited children up to the age of 18 years with neurological impairments and GOR who were undergoing gastrostomy tube insertion.
Data collection and analysis
The review authors worked independently to select trials; none were identified.
We identified no trials that satisfied the criteria for this review.
There remains considerable uncertainty regarding the optimal treatment when faced with the decision of fundoplication surgery versus antireflux medications for children with GOR and neurological impairment who are undergoing gastrostomy insertion. There is a need for robust scientific evidence in order to provide data on the comparable risks or benefits of the two interventions.
Comparaison de la chirurgie de fundoplicature et des médicaments postopératoires contre le reflux gastro-œsophagien chez les enfants présentant une atteinte neurologique faisant l'objet d'une gastrostomie
Les enfants présentant une atteinte neurologique rencontrent souvent des difficultés pour manger, ce qui peut entraîner une malnutrition et un retard de croissance. L'alimentation par une sonde d'alimentation gastrique (gastrostomie) est à l'heure actuelle la méthode privilégiée pour apporter une assistance nutritionnelle aux enfants présentant une atteinte neurologique qui sont dans l'incapacité de se nourrir eux-mêmes normalement par la bouche. Des complications peuvent survenir du fait de la pose d'une sonde d'alimentation gastrique (gastrostomie), et le développement ou l'aggravation du reflux gastro-œsophagien (RGO) ont été largement rapportés dans les résultats. Cela a conduit à l'utilisation fréquente du traitement anti-reflux chirurgical sous la forme d'une fundoplicature, ou d'autres procédures anti-reflux. La fundoplicature est associée à un taux élevé de récidive, d'échec chirurgical, ainsi qu'à une morbidité et une mortalité significatives.
Depuis que les inhibiteurs de la pompe à protons (IPP) ont été introduits dans les années 1990, ils jouent désormais un rôle plus important dans la prise en charge médicale du RGO chez les enfants présentant une atteinte neurologique. Les études non contrôlées suggèrent que les IPP pourraient représenter un traitement sûr et approprié du RGO. D'autres agents actuellement utilisés comprennent les épaississants du lait, les médicaments de suppression des acides, les agents tampons des acides, les stimulants de la motilité de l'intestin et les préparations d'alginate de sodium.
Il existe des risques et des bénéfices associés aux interventions chirurgicales ainsi que médicales et une comparaison plus approfondie est nécessaire pour déterminer le choix du traitement optimal.
Comparer l'efficacité de la chirurgie anti-reflux et des médicaments anti-reflux chez les enfants présentant une atteinte neurologique et un RGO qui font l'objet d'une pose de sonde d'alimentation gastrique (gastrostomie).
Stratégie de recherche documentaire
Nous avons effectué une recherche dans les bases de données suivantes le vendredi 23 mars 2012 : le registre Cochrane des essais contrôlés (CENTRAL), Ovid MEDLINE, EMBASE, CINAHL, LILACS et ISI Web of Science. Auparavant, nous avions effectué une recherche dans la bibliothèque de la santé de l'enfant (Child Health Library) en juin 2009. Nous avons également effectué une recherche dans les registres d'essais en ligne, les revues médicales, les actes de conférences, les mémoires d'université et les thèses de doctorat. Nous avons contacté des spécialistes dans le secteur médical et industriel pour obtenir des renseignements sur les essais achevés et en cours.
Critères de sélection
Nous souhaitions inclure les essais contrôlés randomisés ayant recruté des enfants âgés d'au plus 18 ans présentant une atteinte neurologique et un RGO qui faisaient l'objet d'une pose de sonde d'alimentation gastrique (gastrostomie).
Recueil et analyse des données
Les auteurs de la revue ont travaillé indépendamment pour sélectionner les essais et n'en ont identifié aucun.
Nous n'avons identifié aucun essai répondant aux critères d'inclusion de cette revue.
Conclusions des auteurs
Il subsiste une forte incertitude quant au traitement optimal lorsqu'il s'agit de prendre la décision de la chirurgie de fundoplicature par rapport aux médicaments anti-reflux chez les enfants présentant une atteinte neurologique et un RGO qui font l'objet d'une pose de sonde d'alimentation gastrique (gastrostomie). Il est nécessaire de recueillir des données scientifiques solides pour produire des preuves sur les risques ou les bénéfices comparables des deux interventions.
Surgery versus medication for treating acid reflux in brain-damaged children having a feeding tube inserted
Children with cerebral palsy often have oral motor impairment and need help with eating and drinking. Frequently this entails surgery to place a feeding tube (gastrostomy) directly into their stomach. They may also be found to have gastro-oesophageal reflux (where stomach acid flows back up into the feeding tube (oesophagus)), which can be made worse by gastrostomy surgery. Reflux can be treated either with additional surgery at the same time as the gastrostomy (a fundoplication) or with antireflux medications. We carried out this review to determine which was the safest and most effective form of treatment. We found no randomised controlled trials that provided scientific evidence on which to base a conclusion, highlighting the need for a trial comparing the two interventions.
Comparaison de la chirurgie et des médicaments pour le traitement du reflux acide chez les enfants atteints de lésions cérébrales faisant l'objet d'une pose de sonde d'alimentation gastrique
Les enfants atteints d'une infirmité motrice cérébrale présentent souvent des déficiences maxillo-faciales et motrices et ont besoin d'aide pour manger et boire. Cela implique fréquemment la réalisation d'une intervention chirurgicale pour la pose d'une sonde d'alimentation gastrique (gastrostomie) directement dans leur estomac. Il est aussi possible qu'ils souffrent de reflux gastro-œsophagien (quand l'acide gastrique remonte dans la sonde d'alimentation (œsophage)), qui peut être aggravé par la chirurgie de la gastrostomie. Le reflux peut être traité soit par une autre chirurgie en même temps que la gastrostomie (fundoplicature) soit par des médicaments anti-reflux. Nous avons mené cette revue afin de déterminer laquelle, parmi ces deux options de traitement, offrait la plus grande sécurité et la plus grande efficacité. Nous n'avons trouvé aucun essai contrôlé randomisé qui avait recueilli des preuves scientifiques pour fonder notre conclusion, ce qui souligne le besoin d'essais comparant les deux interventions.
Notes de traduction
Traduit par: French Cochrane Centre 16th October, 2013 Traduction financée par: Pour la France : Minist�re de la Sant�. Pour le Canada : Instituts de recherche en sant� du Canada, minist�re de la Sant� du Qu�bec, Fonds de recherche de Qu�bec-Sant� et Institut national d'excellence en sant� et en services sociaux.
Children with neurological impairments frequently experience profound feeding difficulties, which can lead to malnutrition and growth failure. These are often seen as inevitable consequences of many neurological disorders due to their associated physical abnormalities, communication difficulties and motor impairments (Samson-Fang 2003).
The causes of malnutrition in children with neurological impairments are multifactorial. Physiological abnormalities may manifest as oro-motor dysfunction and the associated inco-ordination of oropharyngeal reflexes (Puntis 2000; Wadie 2002). Children may also be severely compromised in their inability to feed independently and subsequently experience frustration and behavioural problems (Puntis 2000). Many of the factors that lead to malnutrition and failure to thrive dictate a need to find alternative measures of providing adequate nutrition, such as an enteral feeding tube (Wadie 2002).
Enteral feeding is considered to be justified to prevent or reverse growth failure when other efforts to increase nutritional intake, such as dietary supplementation, have failed (Puntis 2000). The placement of a gastrostomy feeding tube is now the preferred method of providing long-term nutritional support to children with neurological impairments. A gastrostomy is considered when enteral feeding is required beyond the short-term period (that is, greater than six weeks), when there are prolonged feeding times, inadequate weight gain and an unsafe swallow (Sullivan 1997; Samson-Fang 2003). The use of gastrostomy tube feeding has previously been shown to increase weight, improve overall health and decrease feeding times for children with neurological impairment (Sullivan 2005). It has also demonstrated a significant, measurable improvement in the quality of life of carers (Sullivan 2004). The literature on prevalence of gastrostomy insertion in this population is limited but a study carried out in the UK by Sullivan 2000 showed that 8% of children with neurological impairments were gastrostomy fed (Sullivan 2000).
Gastrostomy tube insertion can be performed by laparotomy, laparoscopy or endoscopy, and the most favoured procedure for tube placement has become percutaneous endoscopic gastrostomy (PEG), which was introduced by Gauderer in 1981 (Gauderer 1981). PEG is associated with decreased morbidity in comparison with other techniques for gastrostomy insertion and has been assessed to be both efficient and cost-effective (Byrne 1990; Samuel 2002). Complications may arise as a result of gastrostomy placement including infection, perforation, bleeding, tube migration/dislocation, intestinal fistulas and obstruction (Razeghi 2002). In addition, the development or worsening of gastro-oesophageal reflux (GOR) has been widely reported and is considered to be one of the long-term sequelae of gastrostomy placement (Mollitt 1985; Razeghi 2002).
GOR in children with neurological impairments is a well-described phenomenon. It is attributed to a motility disturbance affecting the oesophagus and the lower oesophageal sphincter (LOS) mechanism, and leads to retrograde, involuntary, effortless regurgitation of gastric contents into the oesophagus (Rosen 2000; Vandenplas 2000; Richards 2001). The prevalence of GOR among children with neurological impairments has been reported to range from 14% to 75% depending on the diagnostic criteria of GOR used, highlighting the fact that the role of reflux in this group is not well understood and may have a wide range of manifestations and symptoms (Sullivan 1992; Reyes 1993; Gangil 2001). The pathophysiological mechanisms of GOR are multifarious. The underlying neurological damage may cause delayed gastric emptying and oesophageal dysmotility, while scoliosis, seizures, spasticity of abdominal musculature or constipation can all cause increased abdominal pressure. Due to their often profound physical disabilities, many children spend long periods in the supine position, thus minimising the effect of gravity to aid oesophageal clearance (Halpern 1991; Spitz 1993; Vandenplas 2000).
Description of the intervention
For children with neurological impairments who have pre-existing GOR and are referred for the insertion of a gastrostomy, the placement itself can lead to a worsening of symptoms with severe vomiting and aspiration (Mollitt 1985; Wheatley 1991). Historically, medical management of GOR in children with neurological impairments has had a high failure rate (Cheung 2001; Gold 2002). This has led to the frequent use of surgical treatment in the form of an antireflux procedure (ARP) such as a Nissens, Thal, Toupet or Belsey fundoplication (Ostlie 2002; Hassall 2005). Fundoplication involves strengthening the barrier to acid reflux by wrapping the fundus of the stomach around the oesophagus at the gastro-oesophageal junction. Fundoplication procedures vary according to the degree of fundal wrap that is performed at the distal oesophagus and can range from a full 360° wrap to a partial 180° wrap depending on the operation performed (Hassall 2005). The fundoplication is designed to prevent GOR by correcting hiatal herniation, lengthening the intra-abdominal portion of the oesophagus, tightening the crura and increasing the pressure of the LOS (Di-Lorenzo 2002).
Despite its value in preventing GOR, fundoplication has other consequences. It permanently alters the gastro-oesophageal anatomy and function, and, as has been shown in animal work, may lead to vagal nerve damage that sensitises the emetic reflex (Richards 2000). Complications may ensue from these changes (Di-Lorenzo 2002). These may be related to the underlying condition of the child; for example, seizure disorders or pulmonary disease may lead to increased abdominal pressure, or may arise from the effect of the surgery itself (Kimber 1998). Complications directly related to the surgery may include gas bloat syndrome, impaired gastric accommodation, gastric hypersensitivity, rapid gastric emptying (or 'dumping syndrome'), retching or dysphagia (Di-Lorenzo 2002; Connor 2005). The challenging anatomy and physiology of this group of children mean they suffer from many conditions seemingly unrelated to GOR, such as scoliosis and epilepsy, which subsequently make it difficult for any antireflux intervention to be a guaranteed success. The ARP (fundoplication) is associated with a high recurrence rate and significant morbidity and mortality in this group of children, with a 40% surgical failure rate (Isch 1997; Kimber 1998). About 12% to 30% of children with neurological impairments experience recurrent reflux post fundoplication, while 59% of children experience postoperative complications with a 1% to 3% mortality rate (Byrne 1990; Sullivan 1999).
Traditional medical therapies for GOR in this group of children include milk thickeners, acid suppression drugs (histamine 2 receptor antagonists (H2RAs)), acid buffering agents (antacids) and gut motility stimulants (prokinetics). These therapies have poor response rates with studies showing that only 13% of neurologically impaired children respond completely to medical management, the remainder experiencing persistence of symptoms (Wilkinson 1981; Cheung 2001; Gold 2002). The prokinetic drug, cisapride, was withdrawn from the market in many countries following concerns over safety and a lack of robust evidence of efficacy (Gold 2002), and there is now no effective prokinetic drug available.
Since the 1990s, proton pump inhibitors (PPIs), for example omeprazole, have played a larger part with the medical management of GOR in children with neurological impairment. Uncontrolled studies of omeprazole have reported high levels of tolerability and efficacy suggesting high rates of healing, symptom relief and reduction of vomiting in up to 90% of participants (Cheung 2001; Hassall 2005). PPIs are reported to work by decreasing the acidity of the refluxate and decreasing gastric acid secretion volume, thereby improving gastric emptying (Hassall 2005). The long-term effects of PPI use are not known at this point, but after a decade of use in the adult population, this group of medications have been found to have a good safety profile (Gold 2002).
Why it is important to do this review
This review aims to assess the efficacy of fundoplication for neurologically impaired children who are at a high risk of ARP failure compared with medical antireflux treatment. This is clinically important as there is some evidence of poor outcomes from surgery and the uncontrolled studies suggest that PPIs may be a safe, appropriate, cost-effective alternative (Heudebert 1997; Hassall 2000; Cheung 2001).
To compare the effectiveness of antireflux surgery and antireflux medications for children with neurological impairment and GOR who are undergoing placement of a gastrostomy feeding tube.
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCTs).
Types of participants
Children up to 18 years of age with neurological impairments (as defined by trialists) and GOR who are undergoing insertion of a gastrostomy feeding tube.
We excluded studies if there were neurologically normal children included in the trial (unless studied as a separate subgroup) or if the majority of children participating had neurological impairments that were caused by degenerative or metabolic conditions.
Types of interventions
Any type of surgical fundoplication ARP compared with the use of antireflux medications.
Types of outcome measures
Resolution of clinically measured, identifiable symptoms of GOR following gastrostomy surgery.
Adverse events: morbidity - including wrap failure, retching, dumping, dysphagia, drug side effects and interactions.
Adverse events: mortality.
Pain: measured by parental reporting, behavioural observations.
Health-related quality of life for child or carer, or both: measured using validated scales.
Search methods for identification of studies
The searches for previous versions of this review were run in June 2006 and June 2009. We performed the searches for this update on 23 March 2012.
We searched the following databases on 23 March 2012.
The Cochrane Central Register of Controlled Trials (CENTRAL) 2012 (Issue 1)
Ovid MEDLINE, 1966 to March 2012
EMBASE (Ovid), 1980 to week 12, 2012
CINAHL (EBSCO),1982 to March 2012
ISI Web of Science, 1970 to March 2012
LILACS, 1982 to March 2012
Current Controlled Trials
UK Clinical Research Network
The following online resources were searched for previous versions, but were not available to us in 2012.
Child Health Library, last searched June 2009
National Research Register Archive (contains records up to 2007), last searched in June 2009
ClinicalStudyResults.org, last searched June 2009
TrialsCentral, last searched June 2006
Pharmaceutical Industry Clinical Trials Database, last searched June 2009
International Federation of Pharmaceutical Manufacturers and Associations, last searched June 2009
Please see Appendix 1 for the most recent search strategies and Appendix 2 for the original and 2009 update search strategies. We used subject headings as well as free-text terms where available.
We used RCT filters, where appropriate, with the search strategies. We searched bibliographies of identified articles for additional studies when the original article was directly on the topic, even if not reporting an RCT itself. We did not apply any language or date restrictions.
Searching other resources
We searched the following online journals using their web-based search engines.
American Journal of Surgery (1926 to 2012)
Archives of Disease in Childhood (1973 to 2012)
British Medical Journal (1994 to 2012)
Child Care Health and Development (1975 to 2012)
Developmental Medicine and Child Neurology (1960 to 2012)
Gastroenterology (1965 to 2012)
Gastrointestinal Endoscopy (1995 to 2012)
Journal of Paediatric Gastroenterology and Nutrition (1982 to 2012)
Journal of Pediatric Surgery (1966 to 2012)
Journal of Pediatrics (1932 to 2012)
Pediatrics (1948 to 2012)
Surgery (1995 to 2012)
Surgical Endoscopy (1986 to 2012)
Pharmaceutical companies and equipment manufacturers
The following databases/websites were searched in an attempt to identify grey literature in the form of theses, abstracts or conference proceedings: Index to Theses (searched March 2012), ProQuest Digital Dissertations (searched June 2012), ISI Proceedings (1990 to 2012), Cambridge Scientific Abstracts (CSA), Illumina (searched March 2012) and the System for Information on Grey Literature in Europe (SIGLE) (searched March 2012).
Please see Table 1 for the methods that we would have used if there had been any included studies in our review.
Table 1. Methods to be used in future updates
Selection of studies
Titles and abstracts of studies will be identified from the search and read independently by two review authors (AVR and PBS). Those that do not meet the inclusion criteria will be discarded and any disagreements over suitability for inclusion will be resolved by discussion or adjudication, or both by editors of the Cochrane Developmental, Psychosocial and Learning Problems Group. If further information is sought from trial authors, the study will be categorised as awaiting assessment.
Data extraction and management
Data extraction forms will be designed to include study design, participant characteristics, study setting, intervention type and outcomes. Data will be extracted independently by two review authors (AVR and PBS) and disagreements will be resolved through discussion or by contacting the study author for further information. Data will be organised using Review Manager 5.2.
Assessment of risk of bias in included studies
Selected studies will be independently evaluated by two review authors (AVR and PBS) using The Cochrane Collaboration's tool for assessing the risk of bias (Higgins 2011). Disagreements over suitability for inclusion will be resolved by discussion or adjudication, or both, by editors of the Cochrane Developmental, Psychosocial and Learning Problems Group. The tool will assess the following items: sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other sources of bias. Studies will be assigned a quality category (high, low and unclear risk of bias) and supporting evidence will be presented in a 'Risk of bias' table. We will make our judgements based on the following criteria.
Adequate sequence generation
'low' indicates that adequate methods of concealment were used for random allocation. This would include computer-generated random numbers, a random numbers table, coin-tossing or dice throwing if the procedure was performed by someone now otherwise involved in participant recruitment.
'unclear' indicates uncertainty as to whether allocation was adequately concealed from trialist or participant. This would include methodology where the method of concealment is not known. If the method of concealment and allocation is unclear the study authors will be contacted to obtain precise information.
'high' indicates that although the methods of randomisation were described, the method was inadequate to guarantee allocation concealment. This would include alternation, dates of birth or any procedure that is transparent before allocation such as an open list of random numbers.
'low' indicates the use of centralised or pharmacy-controlled randomisation, telephone randomisation or sequentially numbered, sealed, opaque envelopes.
'unclear' indicates that allocation concealment was not clearly stated or the methods were not described.
'high' indicates that the allocation sequence was not concealed from investigators or participants.
Blinding of participants and personnel
Blinding to the study intervention following allocation would not be possible either for trialists or participants in these studies. The intervention compares a surgical procedure to a medication regimen following insertion of a gastrostomy feeding tube. This means it would be very difficult to eliminate detection and performance bias.
Incomplete outcome data
In the event that data are missing or unclear in any aspect of the trial (from methodological aspects to trial results) primary investigators will be contacted. Data will, if possible, be analysed on an intention-to-treat basis.
'low' indicates that losses to follow-up were equally distributed between treatment and comparison groups, and the reasons for dropout or withdrawal were adequately reported.
'unclear' indicates that information about losses to follow-up was unavailable.
'high' indicates that losses to follow-up were in excess of 30% or unevenly (more than 10% difference) distributed between treatment and comparison groups.
Intention to treat
'low' indicates that intention-to-treat analyses was performed or could be performed using available data.
'unclear' indicates that information about whether intention-to-treat analyses were performed was not available and could not be acquired by contacting the researchers of the study.
'high' indicates that intention-to-treat analyses were not performed and could not be done using available data.
'low' indicates that the prespecified primary outcomes have been reported and all collected data has been reported.
'unclear' indicates that it is insufficient information to allow a judgement to be made.
'high' indicates that the primary outcome data have not been reported or have been reported incompletely.
Other sources of bias
We will attempt to avoid publication bias by seeking data from all sources including conference proceedings and unpublished data from pharmaceutical companies (see search strategy). Should sufficient data be identified, funnel plots will be drawn to investigate any relationship between effect size and study precision (closely related to sample size). Such a relationship could be due to publication or related biases, systematic differences between small and large studies, or poor methodology. If a relationship is found, clinical diversity of the studies will be further examined as a possible explanation.
Measures of treatment effect
For dichotomous data, the risk ratio and its 95% confidence interval will be calculated.
For studies using standardised assessment tools that have generated a scored outcome measure, we will calculate a mean difference. For studies using different scales, we will calculate the standardised difference in means with a 95% confidence interval.
Assessment of heterogeneity
Result consistency will be examined graphically to determine overlap of confidence intervals, poor overlap signifying heterogeneity. We will also consider the I2 descriptive statistic available in Review Manager 5.1 to calculate the proportion of variation that is due to heterogeneity rather than chance. Both fixed-effect and random-effects analyses will be performed.
Where meta-analysis is feasible (that is, if two or more included studies are identified that are considered sufficiently homogeneous), we will perform a meta-analysis.
Subgroup analysis and investigation of heterogeneity
If sufficient data are identified, we anticipate undertaking subgroup analyses as follows:
degree of clinically measured symptomatic gastro-oesophageal reflux prior to gastrostomy surgery;
level of disability;
presence of underlying seizure disorders or pulmonary disease.
We will conduct a sensitivity analysis to assess the impact of study quality on the outcome of the meta-analysis. This will determine whether, for example, studies with unclear allocation concealment or high rates of loss to follow-up are more likely to show positive outcomes.
Description of studies
Results of the search
We identified 1033 reports from the original electronic searches. Papers identified varied from reviews and editorials, to different study designs, both retrospective and prospective. The search update from June 2006 to June 2009 produced a further 476 reports but identified no trials meeting the inclusion. We revised the search strategies in 2012 and ran searches on 23 March 2012, which produced 385 reports, none of which met the inclusion criteria.
Figure 1 shows a study flow diagram including the number of identified reports, duplicates and discarded studies.
We identified no trials meeting the inclusion criteria.
Through our search of the grey literature, we identified one thesis that was an RCT of surgical versus medical management for clinically significant GOR, but this was in adults (Whelan 1979).
We identified one ongoing study in our search of the online trials registries that matched the inclusion criteria (Lloyd 2000); however, we contacted the lead investigator who subsequently explained that the trial had been registered but not started due to methodological problems with the randomisation of children following previous failed medical management.
The other papers we discarded involved, for example, alternative options for reflux treatment, effectiveness studies of different surgical antireflux techniques, and risk and benefit studies of gastrostomy surgery with fundoplication. The studies identified did not meet inclusion criteria either because they involved children who were neurologically normal or they were not RCTs or they did not compare medical versus surgical treatments.
Risk of bias in included studies
We identified no trials meeting the inclusion criteria.
Effects of interventions
We identified no trials meeting the inclusion criteria.
The objective of this review was to compare the effectiveness of antireflux fundoplication surgery with the use of antireflux medications in children with neurological impairments undergoing gastrostomy feeding tube insertion. We identified no relevant RCTs. This review has demonstrated that there is no evidence available from reliable sources on which to draw any conclusions. We cannot provide data on the comparable risks or benefits of either treatment and are subsequently unable to provide recommendations for the best approach in this group of children.
Implications for practice
There continues to be considerable uncertainty regarding optimal treatment option when faced with the decision of performing surgery or prescribing medications for gastro-oesophageal reflux (GOR) in children with neurological impairment who are undergoing a gastrostomy. As surgeons and physicians should ensure that parents are fully informed about the risks and benefits of both treatments, the fact there is a lack of high-quality evidence regarding the relative merits and drawbacks of each option should be shared with families.
It is important to highlight the fact that, although this review deals specifically with the comparison of fundoplication versus medication, other surgical and endoscopic treatment options are becoming available for managing GOR in this group of children. The durability and long-term outcomes are not yet known within the paediatric population and further research is required before they are included as part of clinical practice.
Implications for research
A randomised controlled trial of fundoplication versus postoperative medication for children with neurological impairment and severe GOR undergoing gastrostomy should be undertaken to resolve the current uncertainties about the optimal management strategy, for preference, a large multicentre trial. Due to the nature of the two treatment options, it would not be possible to blind trialists or participants to allocation, but assessor blinding may be possible when measuring pre- and postoperative GOR. The trial would have to be analysed on an intention-to-treat basis as those randomised to the medication arm of the trial could subsequently suffer worsening reflux and require fundoplication, but those having surgery could not revert to the medication arm post procedure. There would need to be strict inclusion and exclusion criteria to identify those children with very severe GOR, such as those experiencing life-threatening events, in whom a fundoplication is strongly indicated.
Through personal communication with Professor David Lloyd (Lloyd 2006 [pers comm]) and Professor Jacob Langer (Langer 2006 [pers comm]), an ethical consideration was brought to the review authors' attention with regards to the process of randomisation. Children with neurological impairments who are referred for gastrostomy insertion and have pre-existing reflux may have previously been treated with medications that have not resolved their symptoms. It would not be ethical in this instance to allocate them randomly to a treatment group knowing that one of the two options had previously failed to alleviate symptoms. This would not provide a fair chance of symptom relief from either therapy. In order to avoid this issue when performing a trial, children would need to be randomised to a treatment arm as soon as the diagnosis of GOR was made, and the gastrostomy surgery, with or without fundoplication, should proceed as soon as possible.
While it is important to acknowledge that any research performed with this group of children will be difficult due to their anatomical and physiological problems, this is a population with ever increasing health needs and this is an area that is critically in need of research.
This update was produced within the Cochrane Developmental, Psychosocial and Learning Problems Group (CDPLPG). We would like to thank Jane Dennis (previous Review Group Co-ordinator) and Jo Abbott (previous Trial Search Co-ordinator) for the original review and first update.
Angharad Vernon-Roberts was funded by Cerebra (Charity for Brain Injured Children and Young People).
We thank the following colleagues who replied to our request for knowledge of any published or unpublished RCTs: Frederic Gottrand; Lewis Spitz; David Lloyd; Carlo Di-Lorenzo; Ciro Esposito; Benjamin Gold; Hisayoshi Kawahara; Yvan Vandenplas; David Gremse; Eric Hassall; Susan Orenstein; Vasundhara Tolia; Jacob Langer; Ralf Heine.
Our thanks also to manufacturer contacts including: Alistair MacDonald (AstraZeneca); Sanjay Kapoor (Altana Pharma); Linda Armstrong (Merck Pharma); Derek Edwards (Vygon UK Ltd).
We also wish to thank Mrs Mary Hines, who gave comments based on her experience of looking after children with neurological impairment.
Cochrane Central Register of Controlled Trials (CENTRAL), 2012, Issue 1, part of The Cochrane Library Searched 23 March 2012
#1 MeSH descriptor Gastrostomy explode all trees #2 (gastrostom*) #3 (#1 OR #2) #4 MeSH descriptor Fundoplication explode all trees #5 (fundoplicat*) #6 (nissen) and (operation*) #7 (thal) #8 (toupet) #9 (belsey) #10 (anti-reflux) #11 (anti) and (reflux) #12 (antireflux) #13 (#4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12) #14 MeSH descriptor Child explode all trees #15 MeSH descriptor Infant explode all trees #16 MeSH descriptor Adolescent explode all trees #17 (baby) or (babies) #18 (infant*) #19 toddler* #20 child* #21 girl* #22 boy* #23 pre-school* #24 preschool #25 teen* #26 adolescen* #27 (#14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26) #28 MeSH descriptor Cerebral Palsy explode all trees #29 (cerebral) and (palsy) #30 little* and disease* #31 spastic near/6 diplegia #32 spastic near/6 quadraplegia #33 neuro* near/6 disab* #34 neuro* near/6 impair* #35 (#28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34) #36 (#3 AND #13 AND #27 AND #35)
Ovid MEDLINE, 1966 to March 2012 Last searched 23 March 2012
1. gastrostomy/ 2. gastrostom$.tw. 3. or/1-2 4. fundoplication/ 5. fundoplicat$.tw. 6. nissen operation$.tw. 7. thal.tw. 8. toupet.tw. 9. belsey.tw. 10. anti-reflux.tw. 11. anti reflux.tw. 12. antireflux.tw. 13. or/4-12 14. exp child/ 15. infant/ 16. adolescent/ 17. (baby or babies).tw. 18. infant$.tw. 19. toddler$.tw. 20. child$.tw. 21. girl$.tw. 22. boy$.tw. 23. pre-school$.tw. 24. preschool$.tw. 25. teen$.tw. 26. adolescen$.tw. 27. or/14-26 28. cerebral palsy/ 29. cerebral palsy.tw. 30. little$disease.tw. 31. (spastic adj3 diplegia).tw. 32. (spastic adj3 quadriplegia).tw. 33. (neuro$ adj3 disab$).tw. 34. (neuro$ adj3 impair$).tw. 35. or/28-34 36. 3 and 13 and 27 and 35 37. randomized controlled trial.pt. 38. controlled clinical trial.pt. 39. randomized controlled trials.sh. 40. random allocation.sh. 41. double blind method.sh. 42. single-blind method.sh. 43. or/37-42 44. (animals not human).sh. 45. 43 not 44 46. clinical trial.pt. 47. exp clinical trials as topic/ 48. (clin$ adj25 trial$).ti,ab. 49. ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. 50. placebos.sh. 51. placebo$.ti,ab. 52. random$.ti,ab. 53. research design.sh. 54. or/46-53 55. 54 not 44 56. 55 not 45 57. comparative study.sh. 58. exp evaluation studies/ 59. follow up studies.sh. 60. prospective studies.sh. 61. (control$ or prospectiv$ or volunteer$).ti,ab. 62. or/57-61 63. 62 not 44 64. 63 not (45 or 56) 65. 45 or 56 or 64 66. 36 and 65
EMBASE (Ovid), 1980 to week 12 2012 Last searched 23 March 2012
1. gastrostomy/ 2. gastrostom$.tw. 3. or/1-2 4. fundoplication/ 5. fundoplicat$.tw. 6. nissen operation$.tw. 7. thal.tw. 8. toupet.tw. 9. belsey.tw. 10. anti-reflux.tw. 11. anti reflux.tw. 12. antireflux.tw. 13. or/4-12 14. exp child/ 15. infant/ 16. adolescent/ 17. (baby or babies).tw. 18. infant$.tw. 19. toddler$.tw. 20. child$.tw. 21. girl$.tw. 22. boy$.tw. 23. pre-school$.tw. 24. preschool$.tw. 25. teen$.tw. 26. adolescen$.tw. 27. or/14-26 28. cerebral palsy/ 29. cerebral palsy.tw. 30. little$ disease.tw. 31. (spastic adj3 diplegia).tw. 32. (spastic adj3 quadraplegia).tw. 33. (neuro$ adj3 disab$).tw. 34. (neuro$ adj3 impair$).tw. 35. or/28-34 36. 3 and 13 and 27 and 35 37. clin$.tw. 38. trial$.tw. 39. (clin$ adj3 trial$).tw. 40. singl$.tw. 41. doubl$.tw. 42. trebl$.tw. 43. tripl$.tw. 44. blind$.tw. 45. mask$.tw. 46. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw. 47. randomi$.tw. 48. random$.tw. 49. allocat$.tw. 50. assign$.tw. 51. (random$ adj3 (allocat$ or assign$)).tw. 52. crossover.tw. 53. 52 or 51 or 47 or 46 or 39 54. exp randomized controlled trial/ 55. exp double blind procedure/ 56. exp crossover procedure/ 57. exp single blind procedure/ 58. exp randomization/ 59. 54 or 55 or 56 or 57 or 58 or 53 60. 36 and 59 61. 36 and 59
CINAHL (EBSCO), 1982 to March 2012 Last searched 23 March 2012
S34 S24 and S25 and S32 and S33 S33 S3 or S4 or S5 or S6 or S7 or S8 or S9 or S10 S32 S26 or S27 or S28 or S29 or S30 or S31 S31 neuro* AND impair* S30 neuro* AND disab* S29 spastic AND quadriplegia S28 spastic AND diplegia S27 little* disease S26 cerebral palsy S25 S11 or S12 or S13 or S14 or S15 or S16 or S17 or S18 or S19 or S20 or S21 or S22 or SS23 S24 S1 or S2 S23 toddler* S22 adolescen* S21 teen* S20 preschool* S19 pre-school* S18 boy* S17 girl* S16 child* S15 infant* S14 baby or babies S13 adolescent S12 infant S11 child Expanders - Apply related words S10 antireflux S9 anti reflux S8 anti-reflux S7 belsey S6 toupet S5 thal S4 nissen operation S3 fundoplicat* S2 gastrostom* S1 gastrostomy/
ISI Web of Science
ISI Web of Science was searched via Web of Knowledge (1970 to March 2012)
reflux reflux AND child* reflux AND child* AND gastrostom* reflux AND child* AND gastrostom* AND fundoplicat* reflux AND child* AND gastrostom* AND fundoplicat* AND neuro*
LILACS searched 1982 to March 2012
Tw gastrostomy AND child* AND reflux
Appendix 2. Search strategies for the original review and the 2009 update
Issue 2, 2009 of Cochrane Central Register of Controlled Trials (CENTRAL) was searched via The Cochrane Library
#1 MeSH descriptor gastrostomy explode all trees #2 gastrostom* in All Text #3 (#1 or #2) #4 MeSH descriptor fundoplication explode all trees #5 fundoplicat* in All Text #6 (nissen in All Text and operation* in All Text) #7 thal in All Text #8 toupet in All Text #9 belsey in All Text #10 anti-reflux in All Text #11 (anti in All Text and reflux in All Text) #12 antireflux in All Text #13 (#4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12) #14 MeSH descriptor child explode all trees #15 MeSH descriptor infant explode all trees #16 MeSH descriptor adolescent explode all trees #17 (baby in All Text or babies in All Text) #18 infant* in All Text #19 toddler* in All Text #20 child* in All Text #21 girl* in All Text #22 boy* in All Text #23 pre-school* in All Text #24 preschool* in All Text #25 teen* in All Text #26 adolescen* in All Text #27 (#14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26) #28 MeSH descriptor cerebral palsy explode all trees #29 (cerebral in All Text and palsy in All Text) #30 (little* in All Text and disease in All Text) #31 (spastic in All Text near/6 diplegia in All Text) #32 (spastic in All Text near/6 quadriplegia in All Text) #33 (neuro* in All Text near/6 disab* in All Text) #34 (neuro* in All Text near/6 impair* in All Text) #35 (#28 or #29 or #30 or #31 or #32 or #33 or #34) #36 (#3 and #13 and #27 and #35
MEDLINE was searched through Ovid (1966 to June 2009)
1 Gastrostomy/ 2 gastrostom$.tw. 3 or/1-2 4 Fundoplication/ 5 fundoplicat$.tw. 6 nissen operation$.tw. 7 thal.tw. 8 toupet.tw. 9 belsey.tw. 10 anti-reflux.tw. 11 anti reflux.tw. 12 antireflux.tw. 13 or/4-12 14 exp Child/ 15 Infant/ 16 Adolescent/ 17 (baby or babies).tw. 18 infant$.tw. 19 toddler$.tw. 20 child$.tw. 21 girl$.tw. 22 boy$.tw. 23 pre-school$.tw. 24 preschool$.tw. 25 teen$.tw. 26 adolescen$.tw. 27 or/14-26 28 Cerebral Palsy/ 29 cerebral palsy.tw. 30 little$ disease.tw. 31 (spastic adj3 diplegia).tw. 32 (spastic adj3 quadriplegia).tw. 33 (neuro$ adj3 disab$).tw. 34 (neuro$ adj3 impair$).tw. 35 or/28-34 36 3 and 13 and 27 and 35 37 randomized controlled trial.pt. 38 controlled clinical trial.pt. 39 randomized controlled trials.sh. 40 random allocation.sh. 41 double blind method.sh. 42 single-blind method.sh. 43 or/37-42 44 (animals not human).sh. 45 43 not 44 46 clinical trial.pt. 47 exp Clinical Trials/ 48 (clin$ adj25 trial$).ti,ab. 49 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab. 50 placebos.sh. 51 placebo$.ti,ab. 52 random$.ti,ab. 53 research design.sh. 54 or/46-53 55 54 not 44 56 55 not 45 57 comparative study.sh. 58 exp Evaluation Studies/ 59 follow up studies.sh. 60 prospective studies.sh. 61 (control$ or prospectiv$ or volunteer$).ti,ab. 62 or/57-61 63 62 not 44 64 63 not (45 or 56) 65 45 or 56 or 64 66 36 and 65
EMBASE was searched via Ovid (1980 to week 23 2009)
1 Gastrostomy/ 2 gastrostom$.tw. 3 or/1-2 4 Fundoplication/ 5 fundoplicat$.tw. 6 nissen operation$.tw. 7 thal.tw. 8 toupet.tw. 9 belsey.tw. 10 anti-reflux.tw. 11 anti reflux.tw. 12 antireflux.tw. 13 or/4-12 14 exp Child/ 15 Infant/ 16 Adolescent/ 17 (baby or babies).tw. 18 infant$.tw. 19 toddler$.tw. 20 child$.tw. 21 girl$.tw. 22 boy$.tw. 23 pre-school$.tw. 24 preschool$.tw. 25 teen$.tw. 26 adolescen$.tw. 27 or/14-26 28 Cerebral Palsy/ 29 cerebral palsy.tw. 30 little$ disease.tw. 31 (spastic adj3 diplegia).tw. 32 (spastic adj3 quadriplegia).tw. 33 (neuro$ adj3 disab$).tw. 34 (neuro$ adj3 impair$).tw. 35 or/28-34 36 3 and 13 and 27 and 35 37 clin$.tw 38 trial$.tw. 39 (clin$ adj3 trial$).tw. 40 singl$.tw. 41 doubl$.tw. 42 trebl$.tw. 43 tripl$.tw. 44 blind$.tw. 45 mask$.tw. 46 ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw. 47 randomi$.tw. 48 random$.tw. 49 allocat$.tw. 50 assign$.tw. 51 (random$ adj3 (allocat$ or assign$)).tw. 52 crossover.tw. 53 52 or 51 or 47 or 46 or 39 54 exp Randomized Controlled Trial/ 55 exp Double Blind Procedure/ 56 exp Crossover Procedure/ 57 exp Single Blind Procedure/ 58 exp RANDOMIZATION/ 59 54 or 55 or 56 or 57 or 58 or 53 60 36 and 59
CINAHL was searched via Ovid (1982 to December 2008)
1 Gastrostomy/ 2 gastrostom$.tw. 3 or/1-2 4 fundoplicat$.tw. 5 nissen operation$.tw. 6 thal.tw. 7 toupet.tw. 8 belsey.tw. 9 anti-reflux.tw. 10 anti reflux.tw. 11 antireflux.tw. 12 exp Child/ 13 Infant/ 14 Adolescent/ 15 (baby or babies).tw. 16 infant$.tw. 17 toddler$.tw. 18 child$.tw. 19 girl$.tw. 20 boy$.tw. 21 pre-school$.tw. 22 preschool$.tw. 23 teen$.tw. 24 adolescen$.tw. 25 or/12-24 26 Cerebral Palsy/ 27 cerebral palsy.tw. 28 little$ disease.tw. 29 (spastic adj3 diplegia).tw. 30 (spastic adj3 quadriplegia).tw. 31 (neuro$ adj3 disab$).tw. 32 (neuro$ adj3 impair$).tw. 33 or/26-32 34 or/4-11 35 3 and 25 and 33 and 34 36 randomi$.mp. [mp=title, subject heading word, abstract, instrumentation] 37 clin$.mp. [mp=title, subject heading word, abstract, instrumentation] 38 trial$.mp. [mp=title, subject heading word, abstract, instrumentation] 39 (clin$ adj3 trial$).mp. [mp=title, subject heading word, abstract, instrumentation] 40 singl$.mp. [mp=title, subject heading word, abstract, instrumentation] 41 doubl$.mp. [mp=title, subject heading word, abstract, instrumentation] 42 tripl$.mp. [mp=title, subject heading word, abstract, instrumentation] 43 trebl$.mp. [mp=title, subject heading word, abstract, instrumentation] 44 mask$.mp. [mp=title, subject heading word, abstract, instrumentation] 45 blind$.mp. [mp=title, subject heading word, abstract, instrumentation] 46 (40 or 41 or 42 or 43) and (44 or 45) 47 crossover.mp. [mp=title, subject heading word, abstract, instrumentation] 48 random$.mp. [mp=title, subject heading word, abstract, instrumentation] 49 allocate$.mp. [mp=title, subject heading word, abstract, instrumentation] 50 assign$.mp. [mp=title, subject heading word, abstract, instrumentation] 51 (random$ adj3 (allocate$ or assign$)).mp. 52 Random Assignment/ 53 exp Clinical Trials/ 54 exp Meta Analysis/ 55 51 or 47 or 46 or 39 or 36 or 52 or 53 or 54 56 55 and 35
ISI Web of Science
ISI Web of Science was searched via Web of Knowledge (1970 to June 2009)
reflux reflux AND child* reflux AND child* AND gastrostom* reflux AND child* AND gastrostom* AND fundoplicat* reflux AND child* AND gastrostom* AND fundoplicat* AND neuro*
The Child Health Library
The Child Health Library (searched in June 2009)
reflux gastrostomy* AND reflux
National Research Register
National Research Register (to the end of 2007) (archived)
#1 GASTROSTOMY single term (MeSH) #2 gastrostom* #3 (#1 or #2) #4 FUNDOPLICATION single term (MeSH) #5 fundoplicat* #6 (nissen next operation*) #7 thal #8 toupet #9 belsey #10 anti-reflux #11 (anti next reflux) #12 antireflux #13 (#4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12) (#3 and #13) (1)
LILACS (1982 to June 2009)
Tw gastrostomy and (Tw fundoplicat$ or Tw nissen operation$ or Tw thal or Tw toupet or Tw belsey or Tw anti-reflux or Tw anti reflux or Tw antireflux)
11 September 2012
New citation required but conclusions have not changed
No new studies suitable for inclusion found.
23 March 2012
New search has been performed
Updated searches run.
Protocol first published: Issue 3, 2006 Review first published: Issue 1, 2007
11 August 2009
New search has been performed
Searches repeated and review updated
15 September 2008
Converted to new review format.
14 November 2006
New citation required and conclusions have changed
Contributions of authors
Both review authors (AVR and PB) independently selected abstracts and screened potential trials. Both review authors (AVR and PB) contributed to writing the text of this review. Joanne Abbott (previous TSC of the Cochrane Developmental, Psychosocial and Learning Problems Group) developed the search strategy.
Declarations of interest
Angharad Vernon-Roberts - research nurse salary funded by Cerebra (Charity for Brain Injured Children and Young People). Peter B Sullivan - has received grant funding, speakers fees and consultancy fees from Nutricia Ltd. Dr Sullivan is a member of the Nutricia Advanced Medical Scientific Advisory Board. Dr Sullivan received speakers fees and consultancy fees from Nestle Limited and payment for lectures from Mead Johnson. Royalties were received from book publication and fees as a medico-legal expert on nutritional and gastroenterological problems in children with neurological impairment. Grant support was received for an RCT on optimising nutrition in children at risk of brain damage from SPARKS, The Castang Foundation and Nutricia.
Sources of support
No sources of support supplied
Cerebra (Charity for Brain Injured Children and Young People), UK.
Differences between protocol and review
The protocol for this review was published in The Cochrane Library in Issue 3, 2006. The Methods section of the original protocol has been reorganised and some parts of it re-labelled to meet the standards of the latest version of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). The search strategy was updated for this version.
This is an update of:
Vernon-Roberts A, Sullivan PB. Fundoplication versus post-operative medication for gastro-oesophageal reflux in children with neurological impairment undergoing gastrostomy. Cochrane Database of Systematic Reviews 2007, Issue 1. Art. No.: CD006151. DOI: 10.1002/14651858.CD006151.pub2