AG Catto-Smith

Professor Lau, Members of the Hong Kong Paediatric Society, ladies and gentlemen, I am honoured to have been invited to present this, the eighth James Hutchison Memorial Lecture. Professor Hutchison is rightly remembered as having had a seminal influence on the development of the speciality of paediatrics in Hong Kong. The Memorial Lecture is a right and fitting tribute both to him and to the impressive achievements in paediatrics in Hong Kong that have occurred after him.

James Holmes Hutchison was born in Rangoon, but subsequently educated in Glasgow, Scotland. He had a distinguished early career, graduating in medicine in 1934. Two years later he received a scholarship for research into childhood tuberculosis which led to the award of an MD with honours in 1939. After the Second World War he became a consultant paediatrician at both the Royal Hospital for Sick Children and the Royal Maternity Hospital in Glasgow. In 1961, he was appointed to the Samson Gemmel chair in Paediatrics at the University of Glasgow, a post which he held until 1977. He had an ongoing interest in pulmonary tuberculosis, but also made valuable contributions to our understanding of inherited thyroid disease, neonatal respiratory distress and leukaemia. Medical education was of particular interest and his textbook Practical Paediatric Problems was of course enormously popular. Professor Hutchison regarded his appointment to the chair of paediatrics in Hong Kong as the crowning point of his career.

Professor Hutchison had wide-ranging interests in general paediatrics. I have been asked to talk tonight on a subject that also touches on many aspects of child health - gastro-oesophageal reflux, particularly in relationship to children with chronic disease.

We first need to develop an understanding of the pathophysiology behind reflux disease.

Reflux of gastric contents into the oesophagus is in itself a normal physiological event. Regurgitation of small amounts of gastric contents is common during infancy and is regarded as a physiological process. Reflux is pathological only if it results in harm. In adults, this usually correlates with acid exposure. In children, injury can result from either acid exposure, nutrient loss or respiratory complications. Recognised morbidities include heartburn, oesophagitis, dysphagia, haematemesis, stricture, growth failure, aspiration, apnoea and pneumonia.

The lower oesophageal sphincter functions as a valve connecting the oesophagus and stomach. It is made up of two functional components, the crura of the diaphragm and the lowest few centimetres of the oesophagus. The basal pressure of this sphincter is about 10 to 25 mmHg. The sphincter opens to allow swallowed food to reach the stomach, and for vomitus and gas to be passed. The lower oesophageal sphincter, however, in common with many other body sphincters, also undergoes transient spontaneous relaxations. These were first defined in the 1980's by John Dent. They appear to be important in venting the stomach, particularly in response to fundal distension. Most reflux episodes occur after transient relaxations and not because of a "weak" sphincter. It was subsequently recognised that adults and children who had excessive oesophageal acidification and oesophagitis also had an excessively increased number of transient relaxations of the lower oesophageal sphincter.

The number of reflux episodes in the first 9-12 months of life is markedly increased compared to adult life. Infants have up to 40-50 episodes of reflux per day, with the oesophagus being acidified for up to about 10% of each 24 hours. In adult life these figures are about 25 episodes per day and 5% fractional reflux time. The vast majority of these reflux episodes are in infancy, the lower oesophageal sphincter appears to open much freqently. Clearance of acid from the oesophagus may be much slower, and the upper oesophageal sphincter opens to allow regurgitation much more readily.

A number of processes appear to be important in expelling or neutralizing refluxed gastric content. An upright posture and gravity undoubtedly help - these are obviously not able to be brought into play in a recumbent infant. The refluxed fluid also initiates a reflex peristaltic wave and sometimes a swallow which aids clearance. This may be defective in a child with a dysmotile oesophagus as in repaired oesophageal atresia. The importance of neutralization of refluxed acid has only recently been recognized. Oesophageal mucosal glands are able to secrete small amounts of bicarbonate. Swallowed saliva is of crucial importance in acid neutralization. This is of particular relevance to children with cerebral palsy and excessive loss of saliva through drooling. An elegant study by Helm, et al¹ determined the importance of saliva. Subjects were given an oesophageal bolus of radioactively labelled acid and asked to repeatedly swallow. The residual volume of the bolus, and oesophageal acidification were measured. Repeated swallowing was able to effectively remove the bolus volume within two swallows, and oesphageal pH returned to normal in a stepwise manner coincident with subsequent swallows. However, after diversion of the salivary flow, oesophageal pH remained unchanged for an extended period despite repeated swallowing, and no stepwise change occurred.

The importance of acid injury to the oesophagus in childhood is only just becoming appreciated. Children with cerebral palsy are at significant risk of development of oesophagitis and stricture. Philosophical, ethical and communication issues have in the past limited recognition of these problems and also hampered effective therapy. Dysplastic change, Barret's metaplasia and oesophageal adenocarcinoma appear distant issues in paediatric practice, but there is considerable current concern about the long term effect of gastro-oesophageal reflux in young adults after childhood repair of oesophageal atresia. Reflux and impaired clearance of oesophageal acid is common in this group, but fundoplication carries the risk of inducing significant dysphagia. Careful follow-up of these patients, even as adults, is early required.

Accurate evaluation of oesophagitis is aided by two main diagnostic techniques:- prolonged oesophageal pH recording, and endoscopic biopsy. Oesophageal pH recording is usually well accepted, but must be used appropriately. It is of most use in defining the total duration of oesophageal acidification to determine whether a patient is at risk of developing oesophagitis. It is also of use to link specific clinical events such as irritability, cough or apnoea with a reflux episode. Flexible upper endoscopy with biopsy is of use in evaluating the severity of oesophagitis.

Other investigative techniques are also available. Each has their advantages but also disadvantages. Barium swallows and meals are very helpful in defining the anatomy of the upper gastrointestinal tract. However, the demonstration of a reflux episode during a barium study means little. Radioisotopically labelled feeds may be of use in determining the rate of gastric emptying and also in defining episodes of milk aspiration. However, they are limited in being able to study the patient for only a short period.

Reflux may result in respiratory disease through a number of different mechanisms, both regurgitant and non-regurgitant. This is an important issue because the complexity of interactions makes accurate diagnosis difficult. The aspiration of a large amount of gastric fluid into the lungs may for instance occur in the context of repeated and frequent gastro-oesophageal reflux, but it may also occur as a "once-only" event. Oesophageal pH monitoring and oesophageal endoscopy after the event may be of limited value. Repeated reflux and microaspiration of small amounts of gastric fluid can lead to a chemical pneumonitis which may exacerbate preexisting lung disease. Bronchoscopic lavage with demonstration of lipid laden macrophages has been suggested as a reliable method of diagnosis of reflux-aspiration, but there are critics of this technique. Dr Cohn Sherrington, who is currently undertaking a PhD with me, is researching new techniques for the more accurate diagnosis of aspiration lung disease using both bronchoscopic and serological methods.

Recent studies by my colleagues² and myself³ have established that in a small proportion of children with cystic fibrosis, postural drainage chest physiotherapy causes reflux which can then profoundly exacerbate existing lung disease. This was demonstrated both by oesophageal PH, and by symptom scores. Cessation of postural drainage and a change in chest physiotherapy technique resulted in significant clinical improvement.

Non-regurgitant reflux can also produce significant sequelae through reflexes induced by oesophageal afferent neurones. These can induce bronchospasm, bradycardia or even apnoea. Accurate delineation of the cause and effect relationships between these events requires multimodal monitoring of oesophageal pH, electrocardiogram, respiration, oxygenation and occasionally even electroencephalography.

In some situations, therapies can cause or worsen reflux disease. One example is in children who have a gastrostomy placed for feeding, particularly in those who are neurologically disabled. This has been recognised for some time as a complication of formally-placed surgical gastrostomies such as the Stamm procedure. A recent surge in placement of gastrostomies has occurred with the development of the percutaneous endoscopically-placed gastrostomy (PEG). We found that the PEG procedure also resulted in a significant increase in reflux, but that pre-operative oesophageal pH was unable to define those patients in whom increased reflux would develop.⁴

Gastro-oesophageal reflux has recently become a focus of interest in otherwise well children with "colic".

Infantile colic is a condition that we as medical practitioners find difficult to treat effectively. Sudden, apparently inexplicable episodes of paroxysmal crying can occur in otherwise normal children. This pattern of crying may have its onset in the first few weeks of life, peaking between 6 weeks and 3 months. Typically the episodes are more frequent in the late afternoon or early evening. They can result in enormous psychological stress to the family. The cause of infantile colic is unclear, but certainly in Australia, there has been a recent trend to treat these infants for gastro-oesophageal reflux on the basis that their irritability is due to heartburn. This has been even in the face of an absence of a history of regurgitation. Evidence from our own hospital⁵ suggests that, in a minority of infants, reflux-induced irritability does occur. Importantly, however, it was only observed where there was a definite history of frequent regurgitation. "Silent" reflux as a cause of infant irritability was not observed in our study. Oesophageal pH monitoring was extremely useful in establishing a linkage between episodes of irritability and reflux.

Gastro-oesophageal reflux can present in a number of different ways in early childhood. Symptoms and signs that should raise the possibility of gastro-oesophageal reflux include haematemesis or melaena, recurrent pneumonia, recurrent apnoea, failure to thrive and feeding difficulties.

In older children, rumination, recurrent vomiting, dysphagia and Sandifer's syndrome should also trigger the consideration of reflux as a diagnostic possibility.

The therapy of gastro-oesophageal reflux in childhood is usually approached in a graded stepwise manner depending on severity. Historically, the first approach has been to use postural therapies and to thicken feeds. Infants have been nursed angled at 30-40° in a bid to reduce regurgitation. The prone position was shown to be far preferable to the supine position in terms of reducing reflux episodes. Unfortunately, the use of the prone position can no longer be recommended because of its association with an increased risk of sudden infant death syndrome. Some recent evidence suggests that nursing infants on their sides may be as effective in reducing reflux episodes.

Thickening of feeds can certainly reduce the frequency and volume of reflux episodes, but neither postural therapy or thickening appear to be very effective in the presence of severe gastro-oesophageal reflux.

The availability of prokinetic agents such as cisapride, in combination with effective acid-suppressive drugs such as ranitidine or omeprazole has profoundly changed the medical treatment of severe reflux. It has also lead to a reduction in the number of children who have surgical antireflux procedures. For many infants with significant gastro-oesophageal reflux, the long-term natural history is for spontaneous improvement. These medications often provide a very useful bridge. Spontaneous resolution of gastro-oesophageal reflux is uncommon after the age of two years. The role of surgery in this situation must be examined carefully. Procedures such as the Nissen fundoplication are generally safe and very effective. It is important, however, to recognise that they are associated with a small but definite morbidity. Postoperative dysphagia, bloating and dumping syndrome are well recognised as well as inability to vomit.

In concluding my talk, I wish to thank the Hong Kong Paediatric Society for inviting me to deliver this oration and also the sponsors of the meeting. Gastro-oesophageal reflux has an important role to play in many different conditions in childhood. I hope that in my talk I have been able to provide an indication of the care which must be exercised in its evaluation and management.

References

1. Helm JF, Dodds WJ, Pelc LR, et al. Effect of esophageal emptying and saliva on clearance of acid from the esophagus. N Engl J Med 1984;310:284-8.

2. Button AG, Heine RG, Catto-Smith AG, Phelan PD, Qlinsky AO. Postural drainage and gastro-oesophageal reflux in infants with cystic fibrosis. Arch Dis Child 1997;76:1-4.

3. Heine RG, Button B, Catto-Smith AG, Olinsky A, Phelan PD. Gastro-oesophageal reflux in infants under six months with cystic fibrosis. Arch Dis Child 1998;78:44-8.

4. Heine R, Reddihough D, Catto-Smith AG. Gastro-oesophageal reflux after gastrostomy in children with severe neurological impairment. Dev Med Child Neurol 1995;37:320-9.

5. Heine RG, Jaquiery A, Lubitz L, Cameron DJS, Catto-Smith AG. Role of gastro-oesophageal reflux in infant irritability. Arch Dis Child 1995;73:121-5.