Comparison of oesophagogastroduodenoscopy findings and diagnostic value in neurologically impaired children and 'normal' children.

OBJECTIVES To review the oesophagogastroduodenoscopy findings in children with severe neurological impairment and 'normal' children, over a 7-year period from 2000 to 2007. DESIGN Retrospective study. SETTING Paediatric Unit of Caritas Medical Centre, Hong Kong. MAIN OUTCOME MEASURES The frequencies of Helicobacter pylori status, peptic ulceration, and oesophagitis were compared. The diagnostic value of oesophagogastroduodenoscopy in these two groups of children was also examined. PATIENTS Patient data were retrieved from the Hospital Authority Clinical Management system, excluding those under surgical care. The children were divided into two groups: 'normal' and neurologically impaired. Their demographic data, indications for oesophagogastroduodenoscopy, endoscopy diagnoses, and Helicobacter pylori status were compared, as was the diagnostic value of oesophagogastroduodenoscopy. RESULTS From 2000 to 2007, 223 oesophagogastroduodenoscopies were performed in 176 patients aged 3 to 22 years; 134 were performed in 'normal' children (median age, 14; range, 3-22 years) and 89 in neurologically impaired children (median age, 12; range, 3-20 years). The three most common indications in 'normal' children were: epigastric pain (60%), gastro-intestinal bleeding (13%), and vomiting (7%). In neurologically impaired children, they were gastro-intestinal bleeding (51%), assessment for percutaneous endoscopic gastrostomy (27%), and follow-up for previous lesions (9%). Among 'normal' children, 14 had duodenal ulcers (associated with Helicobacter pylori in 13), but no patients had gastric ulcers or oesophagitis. Among neurologically impaired children, one had a Helicobacter pylori-negative duodenal ulcer, and four had gastric ulcers (three were Helicobacter pylori-positive). Twenty-four neurologically impaired children had oesophagitis. Neurologically impaired children had significantly more oesophagitis and gastric ulcers (P<0.001 and P=0.004, respectively) but less duodenal ulcers (P=0.024). In 111 children who had gastric biopsies, the Helicobacter pylori infection rate was 35% (31% in 'normal' children and 43% in the neurologically impaired). The diagnostic value was 37% in 'normal' children and 81% in the neurologically impaired (P<0.001). The overall diagnostic value of oesophagogastroduodenoscopy was 50%. CONCLUSION The clinical presentation and endoscopic findings in 'normal' and neurologically impaired children were discrepant. Oesophagogastroduodenoscopy appeared to confer greater diagnostic value in neurologically impaired than 'normal' children. Diagnostic values in our unit were comparable to reports from western studies.