[Evaluation of plain abdominal radiography prescriptions in a university hospital center]

The aim of this study was to evaluate the prescription, impact and diagnostic utility of plain abdominal radiography in a University Hospital.

Plain abdominal radiography series performed over 15 consecutive days were evaluated for quality and agreement with the patient's clinical status. Agreement between the final diagnosis and the radiographic findings were also compared. Contribution of plain abdominal radiography to final diagnosis was assessed on the basis of the conclusions on the hospital discharge report or from those drawn from a telephone survey. The same relationship was also assessed in terms of clinician status (medical student, resident physician, senior physician).

The series included 175 radiography series (11.6 per day). In 86% of the cases, the choice of views was adapted to the clinical context. In 14% it was not. The quality of the x-rays was found to be unsatisfactory in 17%, good in 63% and excellent in 20%. The radiography was considered to be contributive to the final diagnosis in 13% of the cases and non-contributive in 87%. Clinician status was not correlated with the degree of usefulness for final diagnosis. Finally, the prescription was in agreement with standard guidelines in 28% of the cases and not in agreement in 72%.

Plain abdominal radiographs are neither sensitive nor specific, frequently misleading, and costly per specific and correct diagnosis. Better physician awareness is required to limit the number of unnecessary examinations.