Objective: The objective of this study was to evaluate the diagnostic performance and role of abdominal sonography in excluding abdominal malignancy in the initial workup of patients with abdominal complaints.
Materials and methods: The sonographic report and follow-up data of 494 patients who had undergone a primary sonographic examination were retrospectively reviewed. Sensitivity and specificity of sonography for the diagnosis of an abdominal malignancy-that is, a primary tumor or metastasis-were determined. Multivariate logistic regression analysis was performed to determine the incremental value of sonography, and a prediction rule was derived.
Results: An abnormality on sonography--that is, a mass, ascites, pleural effusion, hydronephrosis, or focal intraparenchymal heterogeneity suggestive of a mass--had a sensitivity for abdominal malignancy of 86% and a specificity of 94%. In the multivariate analysis, the sonographic findings were found to have significant incremental value (odds ratio = 74) after adjustment for the clinical determinants. In patients younger than 38 years with no previous malignancy, no palpable mass, normal liver function test results, and negative findings on sonographic examination, the risk of an abdominal malignancy was less than 1 in 500.
Conclusion: Our results suggest that sonography may be useful in excluding an abdominal malignancy when used in a primary care setting in patients with abdominal complaints who are at low risk for a malignancy.