Is preoperative ultrasound cancer survey necessary in patients with normal liver function tests?

Abdominal ultrasound examination is so easy that it may be overused. To reduce medical costs, we investigated whether preoperative ultrasonic abdominal survey was necessary for oncologic patients with normal liver function tests. We used the ultrasound diagnosis as the final standard to test the value of liver function tests for the prediction of liver metastases. Five hundred forty-four patients with various primary malignancies were referred for preoperative ultrasonic survey of the abdomen. The results showed that for a single test, serum albumin was most sensitive (87.2%) but least specific (29.4%), and serum bilirubin was least sensitive (13.0%) but most specific (97.2%), for the prediction of liver metastases. For a combination of five tests or more, the sensitivity was 95.9% and the specificity 29.5%. In patients with normal liver function tests, liver metastases were detected by ultrasound in two of 106 (1.9%) patients. One hundred of 544 (18.4%) patients having major abnormalities that may influence clinical management were also detected during the ultrasound examination. We conclude that ultrasound is convenient, cheap, and without side effects. It can provide additional information affecting patient management, even in the presence of normal liver function tests.