Assessment of the accuracy of double contrast gastroduodenal radiology.

The report presents an analysis of routine double contrast gastroduodenal radiology based on our experience with 1500 consecutive examinations. The value of this technique is illustrated by the demonstration of subtle abnormalities such as erosions, linear ulcers, and ulcer scars which are usually not demonstrated by the standard barium study. Two hundred and twenty-five of these patients have also been examined by endoscopy and radiological errors were found in 7%. This represents a marked improvement over the results obtained with the standard method. In the last 128 patients, a confidence level ranging from 3 (certainty) to 1 (uncertain) was assigned to the radiological diagnosis before endoscopy. It was found that the majority of radiological errors occurred in the few studies given the lowest confidence level. These findings indicate that the double contrast study of the stomach and duodenum can be a very accurate examination and that endoscopy can be applied more selectively to patients with inconclusive double contrast studies.