Intraoperative ultrasound imaging of the liver at the time of colorectal cancer resection.

OBJECTIVES

To evaluate the accuracy of intraoperative ultrasound (IOUS) liver imaging at the time of primary colorectal cancer resection, which might eliminate incurable patients from adjuvant chemotherapy trials or permit earlier resection of curable metastases.

DESIGN

A prospective trial of routine IOUS liver imaging during resections of primary colorectal cancer. The rate of detection of occult metastases by IOUS imaging alone and the false-negative rate over 22.7 months of follow-up were determined.

SETTING

A tertiary care referral center in Boston, Mass.

PATIENTS

Fifty-five patients undergoing 56 operations for colorectal carcinoma between May 1990 and June 1992.

MAIN OUTCOME MEASURES

The rate of detection, by IOUS imaging alone, of otherwise occult hepatic metastases, the total number of patients with metastases detected at any time during follow-up, and the rate of false-negative findings on IOUS imaging and direct examination.

RESULTS

Occult hepatic metastases were detected by IOUS imaging alone in 5% of patients. Restriction of IOUS imaging to patients with T3 or T4 lesions or recurrent cancers would have identified all metastases and increased the detection rate to 10%. Occult metastases were detected by IOUS imaging alone in 12.5% of patients with T3, N0 lesions. The rate of false-negative findings on IOUS imaging was 13% overall, 0% for patients with T1 or T2 lesions, 3% for patients with node-negative findings, and 7% for patients with T3, N0 lesions.

CONCLUSIONS

The small increment in the detection of occult metastases by IOUS liver imaging does not warrant its use in all patients with colorectal cancer. Selective use in patients with T3 or T4 lesions or recurrent cancers increased the incremental gain in detection. The observed frequency of occult metastases in patients with T3, N0 lesions is sufficient to impact on results of adjuvant chemotherapy trials. Longer follow-up in more patients is needed to determine whether a negative IOUS study is an additional favorable prognosticator in patients with T1 and T2 lesions and node-negative findings.