Prospective, blinded comparison of laparoscopic ultrasonography vs. contrast-enhanced computerized tomography for liver assessment in patients undergoing colorectal carcinoma surgery

To prospectively and blindly compare intraoperative laparoscopic ultrasonography to preoperative contrast-enhanced computerized tomography in detecting liver lesions in colorectal cancer patients. Additionally, we compared conventional (open) intraoperative ultrasonography with bimanual liver palpation to contrast-enhanced computerized tomography in a subset of patients. From December 1995 to March 1998, 77 consecutive patients underwent curative (n = 63) or palliative (n = 14) resections for colorectal cancer. All patients undergoing curative resections were randomized to either laparoscopic (n = 34) or conventional (n = 29) surgery after informed consent. All patients underwent contrast-enhanced computerized tomography, diagnostic laparoscopy, and laparoscopic ultrasonography before resection. In those patients who had conventional procedures, intraoperative ultrasonography with bimanual liver palpation was also done. All laparoscopic ultrasonography and intraoperative ultrasonography evaluations were performed by one of two radiologists who were blinded to the CT results. All hepatic segments were scanned using a standardized method. The yield of each modality was calculated using the number of lesions identified by each imaging modality divided by the total number of lesions identified. In 43 of the 77 patients, both the laparoscopic ultrasonography and CT scan were negative for any liver lesions. In 34 patients, a total of 130 lesions were detected by laparoscopic ultrasonography, CT, or both. When compared with laparoscopic ultrasonography, intraoperative ultrasonography with bimanual liver palpation identified one additional metastatic lesion and no additional benign lesions. laparoscopic ultrasonography identified two patients with mets who had negative preoperative contrast-enhanced computerized tomography. Laparoscopic ultrasonography of the liver at the time of primary resection of colorectal cancer yields more lesions than preoperative contrast-enhanced computerized tomography and should be considered for routine use during laparoscopic oncologic colorectal surgery.