Plain abdominal radiographs and abdominal CT scans for nontraumatic abdominal pain--added value?

We conducted a retrospective descriptive study to determine the value of plain abdominal radiographs in emergency department (ED) patients also receiving abdominal computed tomography scans (CT) for the evaluation of nontraumatic abdominal, back and flank pain (NTAP). Cases were identified through radiology log books. Medical records and radiology reports were reviewed to determine whether the CT confirmed the findings of the plain abdominal radiographs, and whether the clinical course confirmed the results of either. Test characteristics for the plain abdominal radiograph and for the CT, using the clinical course including subsequent invasive procedures as the gold standard, were calculated. Of 177 patients who received CTs, 97 (55%) also received plain abdominal radiographs. Among the 74 patients who were admitted to the hospital and had complete data, the sensitivity and specificity for the plain abdominal radiographs were .43 and .75 respectively, compared to .91 and .94 for the CT scan (P(sens.) < .05, P(spec.) < .05). In 4 patients (5%), both studies failed to identify pathology shown in a subsequent procedure. In ED patients with NTAP, the plain abdominal radiograph may have some value as a screening tool; however, in patients in whom a CT is likely to be ordered anyway, a plain abdominal radiograph is unhelpful and often misleading.