Fibre-optic endoscopy and radiology in the investigation of the ipper gastrointestinatract.

A comparative study is reported of endoscopy and the barium meal in the investigation of the oesophagus, stomach and duodenum, based on 281 endoscopies. There was total agreement between the two methods in 239 instances, including 101 normals, 80 peptic ulcers and nine cancers. In nine instances with unequivocal follow-up information there was definite disagreement, radiology being wrong in two (incorrect confident diagnoses of carcinoma) and endoscopy being wrong in seven (three carcinomas and four ulcers not diagnosed). Of 116 normal barium meals, 15 definite abnormalities were demonstrated by endoscopy, including six peptic ulcers. 'Probable' radiological diagnoses were confirmed by endoscopy in 44 instances. In those where endoscopy disagreed, radiology was definitely shown to be wrong in five, and endoscopy in five. A high degree of accuracy can be achieved by the selected use of both techniques, the information obtained from each considered of equal importance, and neither being regarded as the final arbiter.