Letters to the Editor

To the Editor:

We read with interest the article by Rahusen et al ¹ about the use of laparoscopic ultrasonography in the preoperative study of patients with colorectal liver metastases. In our opinion, the role of laparoscopic ultrasonography has been overemphasized in this study. In effect, the authors describe scantly or not at all the results and the methodology (technical features and imaging evaluation) of the preoperative imaging studies. The computed tomography (CT) results especially are very poor and somewhat surprising. We wonder how the authors were able to compute the true negatives of imaging studies, which are potentially infinite. There is a very large difference between the proportion of patients considered to be candidates for surgical resection on the basis of the previous imaging studies and the proportion of patients eligible after laparoscopic ultrasonography (29/47 [61%]).

An important flaw in this series is that the technique of preoperative abdominal CT scan (although only partially described) is clearly suboptimal. Accurate technique is critical to improve detection of liver metastases with CT. The authors studied the liver with 10-mm collimation. This collimation is definitely insufficient for an adequate study of the liver metastases, and the disappointing results of CT are thus not surprising. Most authors agree that helical CT of the liver should be performed with 5-mm collimation. ^2,3

Moreover, the authors do not describe the dose and the rate of contrast material. Again, this point is essential to obtain high-quality diagnostic studies. In our institution, preoperative staging of patients with colorectal liver metastases is performed with helical CT. Details of the technique and results have been reported previously. ⁴ Briefly, scanning protocol includes 5-mm collimation and 1.5 pitch and subsequent reconstruction at 5-mm intervals. Either ionic or nonionic contrast material (170 mL) is injected at a rate of 3 mL per second and acquisition starts at 60 to 70 seconds.

Between October 1995 and December 1998, 119 patients with suspected liver metastases from colorectal cancer were operated on in our hospital. Preoperative staging was performed in all patients with helical CT. In all cases, an experienced hepatic surgeon performed the intraoperative ultrasound. Helical CT findings were correlated with pathologic findings on a lesion-by-lesion basis. Results of intraoperative ultrasound, liver palpation, and histologic study disclosed 288 metastatic lesions. Helical CT correctly detected 246 metastatic lesions. The overall detection rate for helical CT was 85.4% and the positive predictive value was 96%. The false-positive rate was 3.9% (10/256). In 11 patients (9%), surgical resection was not performed due to undetected extrahepatic disease (n = 5), no tumor (n = 2), more metastases than previously detected (n = 2), and location near vessels (n = 2).

Additionally, as a standard preoperative work-up, colonoscopy and CT of the chest and pelvis were performed in these patients to rule out disseminated disease. In the series by Rahusen and coworkers, ¹ six (13%) patients were operated on and discarded for resection after intraoperative ultrasonography. Of course we agree that intraoperative ultrasonography of the liver by an experienced surgeon is the gold standard, and that any preoperative study must be compared with it. Bimanual surgical palpation and intraoperative ultrasound disclosed 42 additional metastatic lesions. One hundred and eight patients were submitted to liver resection; therefore, the resectability rate, taking the group as a whole, was 91%. Selection criteria for liver resection were medical fitness for major surgery and no signs of disseminated disease on preoperative imaging. We had no predefined criteria of resectability with regard to the number or size of the tumors, or to locoregional invasion, provided that resection could be complete and macroscopically curative. Postoperative mortality was 4.2% (5 patients). The median survival calculated from the time of liver resection was 44 months. The actuarial survival rate at 1, 2, 3, and 4 years was 88%, 74%, 57%, and 48%, respectively. In the majority of our patients (5/11 [5%]), peritoneal carcinomatosis was the reason they were not candidates for surgical resection after laparotomy. Of course, diagnostic laparoscopy could be very helpful in ruling out these patients, but minilaparotomy is another option to be considered.

The majority of patients with colorectal liver metastases have had previous abdominal surgery for resection of the primary tumor. In this condition, intraperitoneal adhesions occur frequently, and probably for this reason, the median time to accomplish the laparoscopy procedure was 70 minutes. This is a long time, and having to undergo two surgical procedures, laparoscopy and liver resection, certainly increases the patient’s anxiety and discomfort.

In this paper, the resectability rate was very low (23/47 [48%]). The authors compare their results with a similar study from the literature, ⁵ but those data are rather old (1991) and the technique and results of the imaging studies have changed substantially during the intervening years. Our resectability rate of 91%, considering the entire group, is probably more realistic. More recent studies from the Memorial Sloan-Kettering Cancer Center in New York ⁶ showed data similar to our results (329/416 patients resected [79%]). Another reason for the low resectability rate of the authors could be very restrictive indications for resection. But considering that there is no other effective therapy for colorectal liver metastases, we believe that surgical resection should be considered the standard therapy whenever possible.

In conclusion, we believe that in this article, the role of laparoscopic ultrasonography (if any) is overemphasized, probably because the potential utility of the preoperative helical CT is underestimated.

In our experience, adequate selection of patient candidates for surgery can be done, in most cases, with a high-quality preoperative study of the liver with helical CT. The main reason for discrepancy with surgical findings is peritoneal metastases (11–18% of the patients) and these lesions can be ruled out with a minilaparotomy just before resection, or alternatively, with a diagnostic laparoscopy.

Moreover, this diagnostic approach is cost-effective because preoperative staging is performed on an outpatient basis, with low cost and a noninvasive technique. We consider that in this era of cost-containment in medicine, diagnostic strategies should be designed to consider both the accuracy and the cost as well as the patient’s comfort.

August 7, 1999

Juan Figueras MD, PhD
Carlos Valls MD, PhD