The staging of gastrointestinal malignancies aims to ensure that each patient receives the most appropriate treatment with minimal morbidity and in a cost-effective manner. Thus, the goal of staging is to distinguish patients who have a potentially resectable localized tumor from those with advanced disease and/or distant metastases. Accurate staging has become increasinglyimportant in view of the increasing alternatives for neoadjuvant therapy and nonoperative palliative procedures.
The options available for the evaluation of patients suspected of having a gastrointestinal tract malignancy are numerous and often expensive. To identify and stage a solid organ malignancy, noninvasive tests, such as ultrasonography or computed tomography (CT), are rational initial studies. Since a tissue diagnosis is nearly always necessary, endoscopy is clearly the first line of investigation for patients who may have a carcinoma originating in a hollow viscus. Once a malignancy is identified, a multitude of additional studies may be obtained to determine the stage and resectability of the disease.
Despite current technologic sophistication, preoperative imaging methods are still far from optimal. Significant disparity often exists between the results of preoperative investigations and findings at exploratory laparotomy. Preoperative understaging may result in the termination of a planned "curative" resection, thereby subjecting the patient to some additional physical and psychological morbidity. It is clear that the ideal staging modality approaches the sensitivity and specificity of exploratory laparotomy and yet is less invasive and has minimal risks.
With the resurgence and improvement of laparoscopy over the past 10 years, this technology is evolving as an ideal tool for staging intraabdominal malignancies. Laparoscopy may be useful in procuring tissue for the definitive diagnosis of malignancy, as well as in determining the extent of disease following the identification of a tumor. Although laparoscopy is more invasive than other imaging modalities, it has better sensitivity and specificity. Moreover, with the advent of laparoscopic ultrasonography, the accuracy of laparoscopy in the assessment of solid organs is even greater.
Enthusiasm for laparoscopic staging should be tempered by the view that it should be performed only if the information provided will influence the treatment plan, however. Any patient in whom a laparotomy must be performed, regardless of the findings of laparoscopy should not be considered a candidate for this staging technique.
This article reviews the current status of laparoscopy in the staging of gastrointestinal malignancies and attempts to demonstrate how the advancement of this technology is beneficial to the field of oncology and, most importantly, to the cancer patient. New developments and future applications are also discussed, since the rapid progression of laparoscopic techniques will undoubtedly lead to many future improvements in the diagnosis, staging, and treatment of cancer.
Table 1 lists the basic instruments needed to perform laparoscopic staging procedures.
The periumbilical area is the usual insertion site of the initial trocar for the creation of a pneumoperitoneum. Although a history of prior abdominal surgery is not a contraindication to laparoscopy, an alternate initial trocar site may be necessary if there are surgical scars, masses, or organomegaly. The Veress needle may be used for most patients, but insertion of a blunt cannula under direct vision using the "open" technique is the preferred method for patients who have undergone a previous laparotomy.
In patients with ascites, the Trendelenburg position should not be used for the establishment of a pneumoperitoneum since, in this position, loops of the small intestine are likely to be located adjacent to the anterior wall of the lower abdomen. Instead, the patient should be placed in the reverse Trendelenburg position to allow the small intestine to float cephalad to the surface of the ascitic fluid, thereby limiting the possibility of intestinal injury when the Veress needle is introduced.
Inspection of the Peritoneal Cavity
The initial step in staging laparoscopy is meticulous inspection of the four quadrants of the peritoneal cavity, which, if done properly, can identify serosal implants as small as 1 mm. To facilitate visualization, any ascitic fluid should be aspirated and sent for cytology prior to inspection. However, if laparoscopic ultrasonography is planned during the evaluation, it should be performed before aspiration, as the ascitic fluid is an excellent medium for the transduction of ultrasonic sound waves.
Changing the position of the operating table facilitates visualization of most areas of the peritoneal cavity, including the paracolic gutters, esophageal hiatus, greater curvature of the stomach, spleen, and pelvis. The insertion of two or three additional trocars permits the manipulation of the stomach and transverse colon so that the lesser sac can be entered and the retrogastric area, including the body and tail of the pancreas, can be inspected. The remainder of the retroperitoneum and pouch of Douglas are also assessed at this time.
In female patients, transvaginal insertion of a uterine manipulator may be necessary for complete inspection of the pelvis. As in open operations, the small intestine and its mesentery are routinely inspected from the ligament of Treitz to the ileocecal valve. Recently introduced flexible-tip laparoscopes now permit visualization of the hepatic surface up to the dome of the diaphragm.
Peritoneal washings should be obtained, especially if the patient is being evaluated for pancreatic cancer, as the presence of malignant cells has been shown to influence resectability and prognosis. Approximately 100 mL of 0.9% normal saline is instilled into the subhepatic space, and the fluid is dispersed by agitating the abdomen. The right side of the table is then tilted downward, and the fluid is aspirated and sent for cytologic analysis.
When a parenchymal lesion, peritoneal implant, or suspicious lymph node is identified, laparoscopic biopsy can be done with great precision. Depending on the size and location, tissue may be obtained by a variety of techniques, including fine-needle, core-needle, forceps, and incisional or excisional biopsy. If bleeding develops, it is usually easily controlled using electrosurgery, clip application, or laparoscopic suture ligation.
Table 2 summarizes the laparoscopic staging procedures that are applicable to the various regions of the gastrointestinal tract.
Laparoscopic techniques are well-suited to confirm the diagnosis of hepatocellular carcinoma or metastatic liver lesions, in that they can be used to visualize a surface lesion and perform a biopsy. Laparoscopic techniques have the additional advantage of permitting direct control of hemorrhage, if necessary. Specimens for cytologic analysis can also be obtained by laparoscopic methods.
Furthermore, laparoscopy is a reliable method for detecting additional small lesions and peritoneal seeding, which are usually missed by other imaging methods. For example, transcutaneous abdominal ultrasonography is known to have a false-negative rate for liver metastasis that exceeds 50% in some studies. Moreover, a multi-institution study demonstrated that 42% of patients with hepatic metastases from colorectal carcinoma who were deemed resectable by CT scan were, in fact, un-resectable at the time of laparotomy.
Primary Liver Tumors
In the 1980s, the potential benefits of staging laparoscopy for primary liver tumors were reported by Lightdale and Jeffers et al, who, using laparoscopy, identified unresectability due to multifocal tumors and peritoneal implantation in 83% of the 41 patients in their combined series.[4,5] Thus, staging laparoscopy spared these patients an unnecessary laparotomy. In a recent series of 50 consecutive patients diagnosed with potentially resectable liver tumors, laparoscopy alone demonstrated unresectability in 46%, and nontherapeutic laparotomy was avoided in these patients.
Despite these data, laparoscopic evaluation alone has been the subject of continued criticisms. These include the inability of laparoscopy to: (1) palpate the liver parenchyma, (2) assess perihilar lymph nodes and retroperitoneal structures, and (3) exclude portal vein thrombosis, all of which are critical to accurate tumor staging.
These objections can be addressed by considering the additional contributions of intraoperative ultrasonography to the staging of hepatic malignancies. Numerous studies have shown that intraoperative ultrasound during open operations is more sensitive for the detection of liver lesions than preoperative imaging modalities, including ultrasound, CT, CT portography, and magnetic resonance imaging (MRI).[8,9]
When Clarke and colleagues compared intraoperative ultrasonography to preoperative ultrasonography, CT, and angiography in 54 patients, intraoperative ultrasound showed 25% to 35% of additional liver lesions. Moreover, 40% of the 167 lesions detected by intraoperative ultrasound were not visible or palpable at the time of exploratory laparotomy. Based on the results of intraoperative ultrasonography, the operative plan was altered in one-third of the study group. Intraoperative ultrasonography also may be used to evaluate surrounding structures, including the portal vein and perihilar structures.
With the advent of laparoscopic ultrasound (Figure 1), accuracy comparable to that of intraoperative ultrasound can be obtained without a major abdominal incision. Preliminary data suggested that, in 48% of patients undergoing a planned hepatic resection for malignancy, the combination of
laparoscopy and laparoscopic ultrasonography averted a nontherapeutic laparotomy. In a study of 50 patients by John and colleagues, laparoscopic staging using ultrasound significantly increased the resectability rate of hepatic malignancies from 58% to 93%, when compared with as-sessment without laparoscopy. More recent small series in the literature have reported similar promising results.[12,13]
Intraoperative ultrasound may also be used during primary resection of colorectal cancer to identify liver metastases that are not detected by preoperative ultrasound or CT. In a prospective analysis of 189 patients with colorectal carcinoma, Machi et al[14,15] demonstrated that intraoperative ultrasound had a 82% sensitivity for the identification of lesions that went undetected by preoperative studies. These lesions later progressed to a size detectable by abdominal ultrasound and/or CT.
Early identification of patients with liver metastasis may serve to direct adjuvant therapy or earlier surgical treatment for metastatic disease (Figure 2). An obvious extension of this is the use of laparoscopic ultrasonography of the liver during laparoscopic surgery for colorectal cancer. At the Cleveland Clinic, we have demonstrated that laparoscopic ultrasonography is a safe and rapid technique that permits systematic scanning of all eight hepatic segments through a single cannula site.
In order to better evaluate the efficacy of this technique, we are currently conducting a prospective, blindedinvestigation comparing laparoscopic ultrasonography to conventional preoperative imaging modalities in patients undergoing resection of colorectal malignancies.