| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Surgery (SNH), University of Florida College of Medicine, Gainesville, Florida; the Department of Surgery (MRW, MFB, KCC), Memorial Sloan-Kettering Cancer Center, New York, New York; and Escola Paulista de Medicina (RC), Universidade Federal de Sao Paulo, Sao Paulo, Brazil.
Correspondence: Address correspondence and reprint requests to: Kevin C. Conlon, MD, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021; Fax: 212-717-3097.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods: A retrospective analysis from 1993 to 1999 revealed 48 patients who underwent surgical evaluation for NFI cell tumors. Of these, 34 (71%) patients underwent laparoscopy and CT for either diagnostic purposes or tumor staging. CT and laparoscopic criteria for curative resectability were defined and the sensitivity, specificity, and predictive value of both modalities in determining resectability were calculated.
Results: The most frequent tumor location and presenting symptoms were pancreatic head (n = 27, 56%) and abdominal pain (n = 31, 65%), respectively. Median tumor size was 4.0 cm. In the laparoscopy group, curative resection was performed in 20 cases (59%). CT followed by laparoscopy was more sensitive than CT alone in predicting resectability (93% vs. 50%, P = 0.03) with similar specificity (both 100%). The predictive value for tumor resectability was 74% for CT alone and 95% for CT followed by laparoscopy. Reasons for unresectability identified at laparoscopy but not indicated by CT were liver metastases (n = 6) or nodal disease (n = 1). Four of these patients were spared a laparotomy while the other three patients underwent surgical palliation and all are alive with disease (AWD). In those not undergoing laparoscopy (n = 14), curative resection was performed in 64% (n = 9). Four of these patients underwent resection, despite having metastases, and three are AWD.
Conclusions: NFI cell tumors of the pancreas present as large masses with frequent metastases. Despite metastatic disease, prolonged survival is often achieved with or without open surgical treatment. Laparoscopy can be used in diagnosis and accurately identifies metastases not seen on CT, thus sparing laparotomy in some patients.
Key Words: Laparoscopy • Pancreatic islet cell tumors • Resectability
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In multiple series that evaluated patients with nonfunctioning islet cell tumors of the pancreas, metastatic disease was found at exploratory laparotomy in > 50% of cases, which resulted in low resectability rates.2–5 Unfortunately, most of these series span long time frames, and the availability of high-quality radiological imaging to assess metastatic disease is unknown. A number of studies have demonstrated improved accuracy of laparoscopy in predicting extent of disease for pancreatic adenocarcinoma.6,7 In this population, laparoscopy can detect occult metastases and spare patients from unnecessary laparotomy. Due to the uncommon incidence of nonfunctioning islet cell tumors of the pancreas, the role of laparoscopy in the treatment algorithm for this disease is unknown.
Our goal was to compare the efficacy of computed tomography (CT) alone vs. CT followed by laparoscopy in determining resectability of nonfunctioning islet cell tumors of the pancreas. In addition, the impact of laparoscopy in the management of patients with these tumors was evaluated.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CT scans, obtained in all patients before the operation, consisted of contrast-enhanced dynamic scans in most patients during the first few years of this review and contrast-enhanced helical CT scans in the later patients. Laparoscopy was performed via standard multiport technique, and documentation of metastatic disease was obtained by biopsy.
We compared the ability of CT alone vs. CT followed by laparoscopy to predict curative resectability. Tumors were determined to be unresectable based on local arterial invasion or distant metastases. Resectability was defined as potentially removable disease with gross negative resection margins. Distant metastases were defined as clear evidence of nodal disease (> 1 cm) outside the field of proposed resection, liver metastases, or peritoneal disease. Local invasion was defined as evidence of encasement of a major arterial structure (e.g., hepatic or superior mesenteric arteries). This was determined by CT scan or, if necessary, at the time of laparoscopy via laparoscopic ultrasound. The sensitivity, specificity, and negative predictive value of CT alone and CT followed by laparoscopy in predicting resectability were determined. True-negatives were defined as those patients who were resected and had no evidence of local arterial invasion or distant metastases on CT or at laparoscopy. True-positives were defined as those patients for whom CT or laparoscopy revealed evidence of local blood vessel encasement or intra-abdominal metastases and who subsequently were confirmed at operation. False-negatives were believed to be resectable by CT or laparoscopic criteria and were not found to be resectable at operation. False-positives were those believed not to be resectable by CT or laparoscopic criteria and subsequently were resected. Statistical analysis was performed with Student’s t-test or 
2 analysis, and significance was defined as P < .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
CT followed by laparoscopy was significantly more sensitive than CT alone in predicting resectability (93% vs. 50%, P = .03) with similar specificity (both 100%). The major difference was in the number of false-negatives for metastatic disease with CT. Laparoscopy identified seven patients with metastatic disease not seen with CT scan (Table 2). This consisted of liver metastases in six patients and nodal disease in one patient. CT identified one patient with metastastic disease not seen with laparoscopy. The predictive value for tumor resectability was 74% for CT alone and 95% for CT followed by laparoscopy.
|
|
In the group of patients who did not have laparoscopic evaluation (n = 14), CT had a sensitivity of 80% and specificity of 100% in predicting a curative resection. Four patients were known to have metastatic disease before operation and underwent laparotomy with attempted resection of the primary tumor and the liver metastases. This was accomplished in three patients, and in one patient the primary tumor was resected and liver metastases were not because they were believed to be too numerous and scattered. Of the three patients who underwent resection of the primary tumor in the pancreas and metastases from the liver, two are alive at 54 and 17 months of follow-up. The third patient died from liver failure after this aggressive approach in the postoperative period. The patient who underwent a resection of the primary tumor without metastectomy is AWD at 52 months. Of the 14 patients in this group, only one patient did not get any type of resection. This patient had an islet cell tumor in the body of the pancreas with liver and mesenteric nodal metastases and underwent an exploratory laparotomy with biopsy. After 37 months, this patient developed gastric outlet obstruction and underwent a gastric bypass. The patient then died of disease at 62 months after the original exploratory laparotomy (Fig. 1).
Overall median length of stay was 9.0 days in the laparoscopy group and 9.5 days in the no-laparoscopy group (P = .48). For those patients undergoing laparoscopic exam alone (n = 7), the median length of stay was < 1 day. The patient in the no-laparoscopy group, who underwent a laparotomy without any palliative procedure, had a 5-day postoperative stay.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Laparoscopy in many centers now is employed widely for tumors of the pancreas. For the most common pancreatic tumor type, adenocarcinoma, data indicate that laparoscopy improves the detection of liver and peritoneal metastases not seen on high-quality CT scans.6 The median survival of patients with metastatic adenocarcinoma of the pancreas is 6 months. Due to the poor prognosis of these patients, diagnosing stage 4 disease at laparoscopy and avoiding a laparotomy is beneficial and has been shown to decrease hospital length of stay.6
In this study, CT followed by laparoscopy had a significantly improved sensitivity as compared with CT alone for the detection of metastases from nonfunctioning islet cell tumors. This led to changes in management in four patients who were managed with laparoscopy and biopsy alone. All four patients are alive, and none have required operations for palliation of biliary or gastric obstruction. Three patients in whom liver metastases were identified at laparoscopy underwent laparotomy for palliative resection or biliary bypass. These patients are alive with disease at 21, 43, and 44 months of follow-up. Palliative biliary and/or gastric bypass in patients with metastatic nonfunctioning islet cell tumors appears reasonable when patients present with jaundice or signs of gastric outlet obstruction. Indeed, there is some evidence to support this approach in the literature, because disease-specific survival after palliative bypass is associated with long-term survival.3,10 However, patients with nonfunctioning pancreatic islet cell tumors do not often present with biliary or gastric outlet obstruction. Indeed, in this study, only 9 of 48 (19%) patients presented with jaundice and fewer with vomiting. In addition, recent improvements in biliary and duodenal stenting, which include the use of metal stents, may make palliative bypass unnecessary. As indicated in Figure 1, the identification of metastatic disease at either laparoscopy or laparotomy was associated with prolonged survival, with little correlation between intervention and subsequent outcome.
In this study, the decision whether to perform laparoscopy was at the discretion of the attending surgeon. Only a small percentage of patients presented with jaundice and, therefore, most patients did not need a palliative procedure at the time of initial presentation. Laparoscopy was performed in 22 of 27 (81%) patients with pancreatic head lesions as compared with only 12 of 19 (63%) patients with pancreatic body or tail lesions. A potential explanation is that removal of a lesion in the pancreatic head via pancreaticoduodenectomy is a more extensive and morbid operation than a distal pancreatectomy for a tail lesion. Therefore, most surgeons have a greater incentive to evaluate for metastases with the use of laparoscopy before embarking on the resection of a pancreatic head lesion.
Laparoscopy for diagnosis and biopsy was performed in three patients who had probable liver metastases on CT. This was performed for lesions that were either inaccessible or too small to accurately biopsy with interventional radiological techniques. The benefit of laparoscopy in these cases is that laparoscopy can provide useful information about the extent and location of intra-abdominal metastatic disease and can be used to provide sufficient tissue for definitive pathologic diagnosis. For those lesions that are deep in the liver, laparoscopic biopsy can be performed under laparoscopic ultrasound guidance.
Two basic approaches prevail in the management of the patient with a recently diagnosed pancreatic mass. In one approach, all patients undergo surgical exploration by laparotomy soon after the clinical diagnosis has been established. In the other, a series of tests are carried out in an attempt preoperatively to stage the cancer, and tailored treatment is instituted based on these findings. In our opinion, if the presentation and radiological imaging are consistent with a potentially resectable non-hormone-producing pancreatic mass, then we advocate an operative approach. In many cases this will be in the absence of a histological diagnosis. These patients undergo laparoscopy before laparotomy. This can be performed during the same anesthesia or at a separate setting. Using a similar approach, laparoscopic staging has been found to be cost-effective in both primary and secondary hepatobiliary malignancies and adenocarcinoma of the pancreas.11,12
However, in the absence of metastatic disease, if patient presentation and imaging are consistent with certain functioning islet cell tumors, such as insulinoma, then the surgeon should proceed directly to laparotomy. For those patients with metastastic functional islet cell tumors, a number of treatment options exist such as chemotherapy, tumor embolization, hormonal therapy, and surgery. Certainly, with almost all pancreatic islet cell tumors, in the presence of metastases, biliary and/or gastric bypass or resection to alleviate symptoms can be justified. The value of resecting asymptomatic metastatic nonfunctioning islet cell tumor remains controversial. Currently, we do not routinely advocate resection in this setting, because the natural history of the disease can be long and, therefore, the morbidity of the treatment will exceed its benefit. However, in certain patients in whom resection of the primary tumor can be performed with acceptable morbidity, concomitant resection of small volume metastases may be indicated.
Received for publication November 22, 1999. Accepted for publication September 19, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. C. Conlon and R. L. McMahon Minimally Invasive Surgery in the Diagnosis and Treatment of Upper Gastrointestinal Tract Malignancy Ann. Surg. Oncol., October 1, 2002; 9(8): 725 - 737. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
