How does oesophageal atresia/trache-oesophageal fistula affect feeding and swallowing in the first year of life?

Approximately 150 babies a year are born with oesophageal atresia and/or trache-oesophageal atresia. Oesophageal atresia occurs when the oesophagus (food pipe) fails to join up during early foetal development. Trache-oesophageal fistula describes an abnormal connection that forms between the oesophagus and trachea (wind pipe). When the baby feeds, milk cannot pass into the stomach but can pass into the lungs.

Surgery is needed within the first few days of life and is extremely successful, with 90-95% of babies surviving. However, approximately 50-80% of babies will have ongoing feeding or swallowing difficulties resulting in choking, chest infections and pneumonia. They can also lead to food refusal, distress at mealtimes and parental anxiety.

Currently there is limited understanding of:

• the exact nature of the feeding and swallowing difficulties
• the risk factors for developing difficulties
• the impact on parents.

This study will address these issues through two complimentary projects:

Project 1: A study of swallow physiology Babies with oesophageal atresia and/or trache-oesophageal fistula who have surgery at Great Ormond Street Hospital will be invited to have a detailed swallow assessment using videofluoroscopy (video x-ray) and high resolution impedance manometry (a pressure monitor) at 2-4 months of age and at 8-12 months of age. Results will be compared to parent report of feeding at 1 year old.

Project 2: A study of the impact on parental well-being This study will be carried out in conjunction with the 'TOFS' support group. A Facebook page will be used to carry out an online forum. Parents of any child who has had TOF/OA will be invited to share their experiences of establishing feeding. This information will be used to develop a questionnaire to look at how wide-ranging these experiences are.

This study is made up of two standalone projects (described as work packages), the results of which will be combined to describe feeding and swallowing difficulties seen in children born with OA/TOF.

Work package one: Prospective cohort study This study will follow a group of children (a cohort) born with OA/TOF through their first year of life. At Great Ormond Street Hospital all children with OA/TOF have swallow assessments but not all have the same assessments at the same time. Therefore, using information from hospital records is not a reliable or useful way of learning about swallowing in this group. Children taking part in the cohort study will have two assessments; one at 2-4 months and one at 8-10 months of age. They will be followed up when they are 12 months old to see what and how they are eating and drinking. This will provide detailed information about how swallowing works and how it relates to what a child eats and drinks.

Although the planned cohort is small, the data collected will be the first of its kind. The investigators will use a new type of swallow assessment, called high resolution impedance manometry (HRIM). The study will also allow us to explore how useful and feasible this test is. Results of this study can be used to help design future, larger studies of swallowing in TOF/OA.

What will participating involve? For each swallow assessment children will have a half day hospital visit. The assessment will involve our current 'gold standard' test called a videofluoroscopy/barium swallow and the new test, HRIM. There will be a telephone review at one year of age.

High resolution impedance manometry will compare the swallowing of the OA/TOF cohort with data that has already been collected from children with 'normal' swallowing. This will give new information about:

• Differences in swallow muscle pressures.
• Differences in the coordination of muscle movements.
• How these differences effect the movement of food and fluid through the throat and food pipe.

HRIM will be carried out alongside a videofluoroscopy/barium swallow. Videofluoroscopy/barium swallow is the most commonly used swallow assessment in children. It will be used to:

• Assess the timing and coordination of swallowing.
• Identify any anatomical differences that may be causing swallowing difficulties.
• Give information from our current 'gold standard' swallow assessment to compare with the new information available from HRIM. This will help us understand whether both assessments are needed in the future.

No previous research has investigated whether swallowing changes over time in children born with OA/TOF. By carrying out assessments at 2-4 months and 8-10 months the investigators will:

• Be able to assess milk drinking and weaning foods.
• Investigate whether swallow pressures change over time.
• Determine whether it is important to assess children more than once. The final part of this study will involve the family completing a food diary at home for three days when their child is one year old.
• This is a simple paper diary on which the parent notes what their child eats and drinks, how long the meal took and whether they had any difficulties.
• For example, breakfast: 1 weetabix with mashed banana and milk, ½ cup of water, took 25 mins. Ate easily.
• The food diary is returned by post or through the MyGOSH app (parent portal of the hospital notes system) to the research team.

Once completed a telephone call will be used to:

• Clarify the food diary entries to ensure they are understood by the research team.
• Get the most recent weight and height.
• Ask whether the child has had any chest infections since the last assessment.
• Ask whether the child has been admitted to hospital since the last assessment. The telephone call is expected to take up to 15-30 minutes. If the parent is in agreement, results of all the assessments will be shared with the appropriate hospital and community healthcare teams to inform clinical care and optimise the support given around eating and drinking.

Analysis of the swallow assessments Videofluoroscopy data will be analysed using a protocol that has been used in previous research (McGrattan et al, 2018; McGrattan et al, 2019). It involves detailed analysis of 11 components of the child's swallow. High resolution impedance manometry data will be analysed using online software called 'swallow gateway'. The data are uploaded anonymously. The software is free to use. This will provide information on 17 components of the upper part of the swallow (mouth and throat) and 14 components of the lower part of the swallow (food pipe).

All the swallowing data will be summarised using averages, counts and percentages. Given the number of children in the study the investigators will not carry out statistical analysis that would infer the results are applicable to all children with OA/TOF.

Work package two: a mixed methods study of parent experience This study will investigate the impact that establishing feeding has on parents. The first part of the study will use an online forum (via Facebook) to understand the issues parents face when they are helping their child learn to feed. This will include breast or bottle feeding, moving on to pureed or weaning foods and then moving on to family foods. The information gained from the forum will be used to design a questionnaire. The questionnaire will give a deeper understanding of how common these issues are, whether there are factors that make the impact on parents more or less significant and whether there are time points that are particularly challenging.

The online forum

In recent years online social networks have become increasingly important sources of support for parents of children with and without health needs. They provide a quick and easy means of asking questions and sharing opinions and ideas. Previous research has found that carrying out an online forum is a successful way of understanding parents experiences (Wray et al, 2018). It is good for parents of children with health needs for a number of reasons:

• Parents can participate at a time and in a place convenient to them.
• Participants do not have to answer every question. Therefore, parents can give as much or as little time as they wish or are able.
• Parents do not have to travel anywhere to take part. This means that people in geographically remote places can participate.
• Parents do not need additional childcare to take part.
• It is a familiar format. Many parents are used to asking or responding to questions in this way. Previous research has found participants find an online forum a comfortable place to share information.
• Participants to influence the direction of discussion, arguably ensuring that the issues explored are those most important to those taking part, reducing the risk of researcher bias.

The online forum will be conducted using a method described by Wray et al (2018). It will be carried out in conjunction with the 'TOFS', the UK support group for OA/TOF via their existing Facebook page. A separate 'closed' group will be created specifically for the purposes of the research project. The forum will be moderated by the 'TOFS' administration team for an agreed fee of £750. This administrator will be responsible for all day-to-day running and moderation of the forum in line with a standard operating procedure developed in collaboration with the research team. This will include processes for managing inappropriate or offensive messaging and distressed users as well as procedures for running the forum.

Questions and a probing guide will be generated from available literature, topics raised by public involvement activities and with the study steering group, comprising three parents, a Dietitian, a Speech and Language Therapist, a Surgeon and a specialist Nurse.

Questions will be put to the forum one at a time with the order and probing determined by the responses received.

Data analysis Each week a full transcript will be sent by the forum moderator to the research team. They will ensure the responses are anonymous by replacing names with a study number.

Once the forum is complete the data will be analysed using a process called thematic analysis. This involves:

1. Data familiarisation: identification of interesting or significant points
2. Data coding: a summary of the point being made
3. Integration of similar codes into categories
4. Integration of categories into over-arching themes
5. Consensus discussion to reach agreed themes This process will be conducted by three members of the research team and a parent representative to ensure the influence of researcher bias is minimised. These themes will be used to develop a questionnaire to explore the prevalence of the issues raised.

The questionnaire A questionnaire will be developed from the themes identified in the online forum to understand how common these issues are in parents of children aged 0-5 years and whether there are critical time points. It will target a larger number of parents than the online forum. Questionnaire development requires results from the online forum, therefore an amendment will be submitted in the future for approval.

The investigators aim to use questionnaires that capture the themes within the online forum that have already been developed and validated. For example, questionnaires about parental anxiety or mealtime behaviours. The project steering group will review the possible questionnaires. Discussion of relevance and importance will be combined with ease of completion to inform final questionnaire selection. In the absence of a suitable questionnaire to capture an important theme, specific questions will be developed. The questionnaires will be piloted to make sure they are readable, not too lengthy and as easy to complete as possible. Following advice from parents questionnaires will be online and mobile optimised. A paper version will be made available if needed.

Information regarding the child's age, medical conditions, feeding method (oral or tube), household members and access to formal and informal support services accessed will be collected. Names will not be collected.

How will the information be analysed? Analysis will be finalised once the questionnaires are selected. Statistical analysis to understand the importance of a child's age, diagnosis and access to support is expected.

Inclusion Criteria:

1. Diagnosis of OA and/or TOF
2. Consent obtained from an adult with parental responsibility
3. Under 5 months of age
4. Undergone surgical repair
5. Orally feeding at least 30mls at any one time
6. Under the care of a Great Ormond Street Hospital Consultant

Exclusion Criteria:

1. Unrepaired OA and/or TOF
2. Repair requiring gastric transposition/jejunal transposition or other non-primary repair
3. Fully tube fed
4. Requiring non-invasive ventilation
5. Under a child protection order

Babies undergoing primary repair of OA/TOF and are under the care of a GOSH Consultant will be invited to participate.

Approximately 90% of children have both OA and TOF. Participants will have undergone primary surgical repair and commenced oral feeding before they are 4 months of age. Participants may have OA/TOF as part of a syndromic diagnosis, such as CHARGE syndrome or VACTERL association. Children with non-syndromic OA/TOF may have comorbidities, most commonly cardiac abnormalities.