Br. J. Surg. 1988, Vol. 75, June, 533-535

D. Kumar, R. J. Brereton, L. Spitz and Christine M. Hall

Departments of Surgery and Radiology. The Hospitals for Sick Children at Great Ormond Street and Queen Elizabeth Hospital, London, UK

Correspondence to: Mr R. J. Brereton, Department of Surgery, Institute of Child Health, Guilford Street, London WC1 N 1 EH, UK

Gastro-oesophageal reflux and intestinal malrotation in children

In a series of 74 children undergoing Nissen fundoplication for persistent gastro-oesophageal rejlux unresponsive to intensive medical treatment, there was an unexpectedly high incidence (54 per cent, n =40) of intestinal malrotation. The series was not homogeneous, 25 of the children having reflux as a complication of a serious congenital abnormality (48 per cent incidence of intestinal malrotation), and 49 presenting with ‘idiopathic’ rejlux (57 per cent incidence of intestinal malrotation). Intestinal malrotation is best detected pre-operatively by careful barium radiology, requiring clinicians to be aware of the association. I n our experience, in children over the age of 3 months, both an antirejlux operation and Ladd’s procedure are often necessary to stop gastro-oesophageal reflux when an intestinal malrotation is present. Keywords: Gastro-oesophageal reflux, intestinal malrotation, pyloric stenosis, oesophageal atresia, abdominal wall defects

There has been growing appreciation of the extent of gastro- oesophageal reflux in infants and children, and of reflux as a cause of various complications, especially of the respiratory tract. In 1947, Neuhauser and Berenberg’ described the condition ‘chalasia’ in 12 infants with relaxation of the cardia as the cause of vomiting. The management was confined to thickening the feeds and attempting to maintain an upright posture. In 1952, Carre et al. described hiatus hernia as a cause of unexplained vomiting in children’ and in 1959, he documented the natural history of ‘partial thoracic s t ~ m a c h ’ ~ . Only in the last decade has it been realized that gastro-oesophageal reflux may be an important cause of failure to thrive, oesophagitis, oesophageal stricture, recurrent pneumonitis, asthma and ‘near miss sudden infant death ~yndrome’~. The purpose of our study is to report the poorly documented association of gastro- oesophageal reflux of such severity as to necessitate antireflux surgery with intestinal malrotation.

Patients and methods During a 4-year period, all infants and children with reflux of such severity as to necessitate antireflux surgery were analysed. The main symptoms of the patients were persistent vomiting, failure to thrive, haematemesis, dysphagia or recurrent respiratory infections which were unresponsive to more conservative forms of management. Patients with ‘idiopathic’ gastro-oesophageal reflux were included as well as those with reflux associated with congenital anomalies such as intestinal malrotation, oesophageal atresia, diaphragmatic hernia, and exomphalos. Children with reflux associated with neurological diseases were excluded, as our experience in this group has been reported elsewhere5. A very large group of patients responding to conservative management either as out-patients, or after a short period in hospital, were excluded. Patients who had intestinal malrotation without demonstrable reflux (n = 21) have been excluded from this analysis as they had no antireflux operation.

The principal diagnostic investigation was the upper gastrointestinal barium contrast study, which was performed in all patients. During this examination, the anatomy and peristaltic activity of the oesphagus were assessed. The presence and degree of gastro-oesophageal reflux were noted,as was theability of theoesophagus toclear the refluxed material. The presence ofcontrast within the tracheobronchial tree was recorded. The anatomy of the gastric outlet was assessed and the contrast followed through the duodenum into the upper small intestine in order to document the position of the duodenojejunal flexure. This was defined as the first visible kink in the distal part of the duodenum.

Malrotation was diagnosed when there was an abnormal configuration of the duodenal C-loop or when the duodenojejunal flexure was located to the right of the vertebral bodies (Figure I ) . In doubtful cases, contrast was followed through the small intestine in an

attempt to define the position of the caecum. In such cases, malrotation was diagnosed if the caecum was found lying high in the right hypochondrium instead of in the right iliac fossa. Upper gastrointestinal endoscopy was carried out in 69 of the 74 patients. The presence and grade of oesophagitis were noted. The tone of the gastro-oesophageal junction was assessed and the presence or absence of reflux documented. If possible, the stomach, pylorus and proximal duodenum were examined. Technetium-99 sulphur colloid ‘milk scans” were per- formed in six patients. Oesophageal pH and manometric studies were performed in only two patients because the majority had obvious radiological confirmation of severe reflux.

A detailed examination of the position of the caecum, its mesenteric attachments and the site of the duodenojejunal flexure was made in all patients at laparotomy before proceeding to fundoplication. Whenever diagnosed, intestinal malrotation was treated by Ladd’s procedure, which consisted of division of Ladd’s bands between the duodenum and the caecum and right colon, the gallbladder, and the lateral abdominal

Figure 1 Radiograph of an upper gastrointestinal contrast study demonstrating midgut malrotation. Note the position of the duodenojejunal flexure to the right of the midline

0007-1323/88/060533-03$3.00 0 1988 Butterworth & Co (Publishers) Ltd 533

Gastro-oesophageal reflux and intestinal malrotation: D. Kumar et at.

wall. The mobilized colon was placed in the left side of the abdominal cavity with the caecum adjacent to the spleen.

Results A total of 74 infants and children had reflux demonstrated by barium meal or pH monitoring and subsequently underwent antireflux surgery in the form of a Nissen fundoplication (Table 1). The age of the patients at operation ranged from 10 weeks to 13 years (mean age 35.6 months). All patients undergoing surgery during the first year of life had a hiatus hernia. Of the 21 patients undergoing surgery for idiopathic reflux unresponsive to medical treatment, 8 had an established oesophageal stricture. An additional associated congenital anomaly was present in the remaining 53 patients.

Midgut malrotation was found at the barium study, and confirmed at laparotomy, in 32 patients with persistent vomiting. A standard Ladd's procedure was carried out in 18, and a further 14 had a Nissen fundoplication in addition during the same operation because endoscopy performed immediately before surgery had demonstrated marked oesophagitis, confirming the pre-operative radiological findings. All of the latter group were cured of their vomiting but only four infants undergoing Ladd's procedure alone were cured of reflux. Vomiting persisted in 14 of the older children undergoing Ladd's procedure alone and all 14 subsequently required an antireflux operation.

In addition to the 28 patients with isolated malrotation (Table I), 12 children underwent Nissen fundoplication for reflux associated with intestinal malrotation combined with other anomalies, namely pyloric stenosis (n = l), oesophageal atresia ( n = 4), umbilical defects (n = 1) and diaphragmatic hernia (n = 6). In 33 patients, the malrotation took the form of a classic incomplete midgut rotation with a narrow-based mesentery and subhepatic caecum. In the remaining seven cases, the malrotation was confined to the duodenum, the duodeno- jejunal flexure being to the right of the midline, the proximal duodenum being dilated (both on barium meal and at laparotomy) indicating a degree of obstruction. Intrinsic duodenal obstruction was excluded by endoscopy in all of these patients.

In the immediate postoperative period, only one infant developed complications. He presented at the age of 11 weeks with a completely intrathoracic stomach. Postoperatively he developed pneumococcal bronchopneumonia resulting in meningitis and necrotizing enterocolitis. He recovered on medical treatment and, at the age of 16 months, was developing normally. Late complications occurred in seven patients, four of whom developed 'gas bloat syndrome'. The other three developed adhesion obstruction, two of whom recovered after laparotomy. One child died suddenly at home without being referred for medical attention; post-mortem examination revealed gangrenous bowel caused by volvulus about a band adhesion.

Of the 73 survivors, 4 still have mild intermittent gas bloat; 4 children with oesophageal atresia have had a fair result; 12 children were reluctant to feed initially but eventually had a

Table 1 Analysis of 74 patients undergoing Nissen fundoplication ,for gastrointestinal reflux unresponsive to medical treatment

Idiopathic G O R Without intestinal malrotation With intestinal malrotation

GOR associated with serious congenital anomalies Midgut malrotation and pyloric stenosis Oesophageal atresia (previous primary repair) Oesophageal atresia and malrotation Oesophageal atresia and pyloric stenosis Exomphalos/gastroschisis and malrotation Diaphragmatic hernia and malrotation

Associated malrotation: 40 (54 %)

21 28

1 12 4 1 1 6

GOR, gastrointestinal reflux

534

good result, and the antireflux procedure had an immediate and excellent result in the remaining 53 infants and children.

Discussion Gastro-oesophageal reflux in infants and children is associated with significant morbidity and with a low, but definite, mortality rate4. Only a few reported series have mentioned the association of gastro-oesophageal reflux with congenital anomalies such as intestinal malrotation, pyloric stenosis and oesophageal atresia'-". One-third of our patients were in the group of children with serious congenital abnormalities other than intestinal malrotation, and 17 of these had oesophageal atresia with reflux"*I2; the remainder of the infants with serious abnormalities had abdominal wall defects including diaphragmatic hernia, exomphalos and gastroschisis with intestinal malrotation as a component of the basic embryological defect. The incidence of intestinal malrotation in infants with oesophageal atresia is unknown, but at least 3percent of patients are affectedI3. The role of reflux in pyloric stenosis, and the effect of reflux oesophagitis on the incidence and severity of postoperative vomiting, have been well doc~mented '~ .

Most children with gastro-oesophageal reflux do not have obvious associated congenital anomalies and the magnitude of the problem of idiopathic reflux is far greater than that first defined by Carre et al.'. Bettex and Kuffer'' published the results of fundoplication for uncomplicated hiatus hernia in 194 children. As we do not have detailed figures for the larger number of infants treated conservatively in our out-patient clinics, or by our medical colleagues, we accept that the patients in this series may not be representative of those presenting in other centres. Many of the patients had been investigated and treated elsewhere before being referred to our surgical unit. Two-thirds of our patients were without apparent congenital abnormality, but 57 per cent of these were found to have midgut malrotation, which has emerged as the most common associated anomaly. This high incidence is all the more surprising because it exceeds that of 48 per cent for the group with other serious congenital anomalies. This is largely attributable to careful attention to the position of the duodenojejunal flexure during the upper gastrointestinal radiological study and at laparotomy before fundoplication. Obviously, the incidence is biased as all children with severe reflux and diagnosed intestinal malrotation underwent operation, whereas numerous children with reflux without a proven malrotation were successfully treated without operation. Similarly, bias may have been introduced by the exclusion of the 21 patients with malrotation in whom reflux could not be demonstrated. However, we are concerned only with the incidence of intestinal malrotation in those with reflux of such severity as to necessitate an antireflux operation.

Intestinal malrotation may be easily overlooked on the barium meal. Before presenting here with persistent vomiting and failure to thrive, two patients had undergone unsuccessful antireflux procedures elsewhere. On careful review of the radiographs at this hospital, the duodenojejunal flexure was noted to be on the right side of the midline in each case and a diagnosis of intestinal malrotation was confirmed at laparotomy. Both were relieved of their symptoms by repeat fundoplication accompanied by Ladd's procedure.

Gastric emptying is said to be slow in patients with idiopathic refluxI6 and more research is required to ascertain the role of delayed gastric emptying as an aetiological factor in reflux associated with anomalies such as malrotation. It is reported that the symptoms of reflux resolve after the problem of outlet obstruction has been corrected'.*, but this is contrary to our experience. Of 18 infants and children over the age of 3 months who underwent Ladd's procedure alone for intestlnal malrotation, vomiting persisted in 14 and an antireflux procedure was required. Outside the neonatal period, it is now our policy to consider treating children with significant reflux and intestinal malrotation with a combined antireflux operation and Ladd's procedure, as the latter operation does not always

Br. J. Surg., Vol. 75, No. 6, June 1988

Gastro-oesophageal reflux and intestinal malrotation: D. Kumar et al.

cure the reflux and the Nissen operation is dangerous and prone to failure in the presence of incomplete gastric outlet obstruction. 9.

10.

References 1.

2.

3.

4.

5.

6.

7.

8.

Neuhauser EBD, Berenberg W. Cardio-oesophageal relaxation as a cause of vomiting in infants. Radiology 1947; 48: 48&3. Carre IJ, Astley R, Smellie JM. Minor degrees of partial thoracic stomach in childhood. Lancet 1952; ii: 1150-3. Carre I J . The natural history of the partial thoracic stomach (“hiatus hernia”) in children. Arch Dis Child 1959; 34: 344-53. Randolf JG, Lilly JR, Anderson UD. Surgical treatment of 13. gastro-oesophageal reflux in infants. Ann Surg 1974; 180: 479-86. Spitz L. Surgical treatment of gastro-oesophageal reflux in 14. severely mentally retarded children. J Roy Soc Med 1982; 75: 525-9. 15. Jona JZ, Sty JR, Glickchick M. Simplified radio-isotope technique for assessing gastro-oesophageal reflux in children. J 16. Pediatr Surg 1981; 16: 114-17. Johnson DG, Jolley SG. Gastro-oesophageal reflux in infants and

11.

12.

children. Surg CIin North Am 1981;b1:~1101-15. Fonkalsrud EW, Ament ME, Byrne WJ, Rachelefsky GS. Gastro- Paper accepted 17 January 1988

oesophageal fundoplication for the management of reflux in infants and children. J Thorac Cardiovusc Surg 1978; 76: 655-64. Kim SH, Hendren WM, Donahoe PK. Gastro-oesophageal reflux and hiatus hernia in infants and children: Experience with 70 cases. J Pediatr Surg 1980; 15: 43-51. Arasu TS, Wyllie R, Fitzgerald JF et a/. Gastro-oesophageal reflux in infants and children: Comparative accuracy of diagnostic methods. J Pedintr 1980; %: 798-803. Fonkalsrud EW. Gastro-oesophageal fundoplication for reflux following repair of oesophageal atresia. Arch Surg 1979; 114:

Myers NA. Oesophageal atresia with distal tracheo-oesophageal fistula-a long term follow-up. Progr Pediatr Surg 1977; 10: 5-19. Brereton RJ, Zachary RB, Spitz L. Preventable death in oesophageal atresia. Arch Dis Child 1978; 53: 276-83. Spitz L. Vomiting after pyloromyotorny for infantile hypertrophic pyloric stenosis. Arch Dis Child 1979; 54: 8869. Bettex M, Kuffer F. Fundoplication in hiatai hernia-Results after 10 years. Progr Pediatr Surg 1977; 10: 25-33. Milleemeier AC, Lange R, McCallum R, Seashore J, Gryboski J. Delayed gastric emptying in infants with gastro-oesophageal reflux. J Pediatr 1981; 98: 190-3.

48-5 1.

Change of address As a result of editorial re-organization, all correspondence should now be sent directly to:

Ms Lynne Ponjee, BJS Editorial Secretary, Butterworth Scientific Ltd., PO Box 63, Westbury House, Bury Street, Guildford, Surrey G U 2 5BH, UK Tel: 0483 300966

No further correspondence should be sent to the Middlesex Hospital address.

Br. J. Surg., Vol. 75, No. 6, June1988 535