- Email: [email protected]
Staging laparoscopy and laparoscopic ultrasonography in more than 400 patients with upper gastrointestinal carcinoma1
Staging Laparoscopy and Laparoscopic Ultrasonography in More Than 400 Patients with Upper Gastrointestinal Carcinoma Els J Nieveen van Dijkum, MD, Laurens Th de Wit, MD, Otto M van Delden, MD, Philip M Kruyt, MD, Jan JB van Lanschot, MD, Erik AJ Rauws, MD, Hugo Obertop, MD, Dirk J Gouma, MD
and pancreatic body and tail tumors. The value of laparoscopic staging for patients with periampullary tumors is not as great as stated in previous studies and is still the subject of investigation. (J Am Coll Surg 1999;189: 459–465. © 1999 by the American College of Surgeons)
Background: Resection offers the only chance of cure to patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary tumors. Staging is essential to select patients who will benefit from operation because palliation can also be performed nonoperatively. Several studies, including limited numbers of patients, have shown that laparoscopic staging prevents unnecessary laparotomies, but it is doubtful whether general application of this staging method can be advised. The aim of this study was to assess the benefit of diagnostic laparoscopy for staging patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary tumors.
Accurate staging of gastrointestinal tumors selects patients who will benefit from curative surgery. The additional value of diagnostic laparoscopy in the preoperative workup has been described.1-12 Small superficial liver metastases and peritoneal deposits can be detected and biopsied under direct view.1-5 Using laparoscopic ultrasonography small intrahepatic metastases and tumor ingrowth in surrounding tissue can be diagnosed.6-11 Adding laparoscopy and laparoscopic ultrasonography to conventional staging protocols prevents unnecessary laparotomies in 11% to 48% of patients with gastrointestinal tumors.1,9,10 This approach is valid only when nonoperative palliation is feasible and an unnecessary laparotomy can be prevented, such as in patients with a malignancy of the esophagus, pancreas, liver, and bile ducts. Patients with other gastrointestinal tumors, for example, distal gastric and colorectal malignancies, almost always undergo laparotomy because operative palliation is preferred when a curative resection is impossible. Pilot studies performed by our group in patients with esophageal and gastroesophageal junction tumors and in patients with periampullary tumors showed that laparoscopic staging prevented unnecessary laparotomies in 3%, 11%, and 19% of the patients, respectively.13,14 It was decided to use diagnostic laparoscopy routinely for staging periampullary tumors.14 On the other hand, laparoscopy was of little value in staging esophageal tumors, and since
Study Design: Between June 1992 and December 1996, 420 patients with a resectable tumor after conventional staging underwent diagnostic laparoscopy combined with laparoscopic ultrasonography. Histologic proof of metastases or ingrowth was used to cancel laparotomy. Results: Laparoscopic staging avoided laparotomy in 20% of patients (sensitivity 0.70): 5% with an esophageal tumor, 20% with a gastroesophageal junction tumor, 15% with a periampullary tumor, 40% with a proximal bile duct tumor, 35% with a liver tumor, and 40% with a pancreatic body or tail tumor. Complications and port-site metastases were seen in 4% and 2% of patients, respectively. Conclusions: Laparoscopic staging is a safe procedure with low morbidity and without mortality in this series. It has shown no benefit in esophageal cancer, but seems beneficial for staging tumors located at the gastroesophageal junction, proximal bile duct tumors, liver tumors, No competing interests declared. Received May 28, 1999; Accepted July 7, 1999. From the Departments of Surgery (Nieveen van Dijkum, de Wit, Kruyt, van Lanschot, Obertop, Gouma), Radiology (van Delden), and Gastroenterology (Rauws), Academic Medical Center, Amsterdam, The Netherlands. Correspondence address: EJM Nieveen van Dijkum, Department of Surgery, Reinier de Graaf Gasthuis, Postbus 5011, 2600 GA Delft, The Netherlands. © 1999 by the American College of Surgeons Published by Elsevier Science Inc.
459
ISSN 1072-7515/99/$21.00 PII S1072-7515(99)00186-6
460
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
then this procedure has been omitted in the workup of those patients, but has been continued in patients with tumors at the gastroesophageal junction.13 In this study the results of laparoscopic staging of more than 400 patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary (HPB) tumors were evaluated in an attempt to further elucidate the general applicability of diagnostic laparoscopy for staging gastrointestinal malignancies. The value of laparoscopic staging is expressed in specificity, sensitivity, and positive and negative predictive values for detection of metastases and tumor ingrowth, and in the number of prevented laparotomies. The morbidity and mortality of the procedure and the appearance of port-site metastases were analyzed. METHODS Between June 1992 and December 1996, patients with esophageal, gastroesophageal junction, or HPB tumors were included in this study, after noninvasive staging showed that the tumor was locally resectable and without distant metastases. Patients who were not fit for an extensive surgical procedure were excluded. Since June 1994, only patients with a tumor of the gastroesophageal junction were included because the pilot study showed a limited effect of laparoscopic staging for patients with tumors of the proximal and mid-esophagus.13 At the same time another pilot study showed a substantial benefit of laparoscopic staging for periampullary tumors, so diagnostic laparoscopy was included in the standard workup of these patients.14 The results of the pilot studies are shown separately. Staging before laparoscopy Patients with a tumor at the gastroesophageal junction were preoperatively staged with ultrasonography of the neck, a chest x-ray, and ultrasonography combined with color-Doppler of the abdomen. The liver and celiac axis were carefully examined for hepatic or lymph node metastases, and biopsies were performed under ultrasonographic guidance for histology or cytologic confirmation. Local tumor extension was assessed by endoscopic ultrasonography and in case of proximal tumors, by bronchoscopy. Indirect laryngoscopy was done to exclude tumor ingrowth in the recurrent laryngeal nerves. Patients with HPB tumors underwent transabdominal ultrasonography combined with color-
J Am Coll Surg
Doppler. Hepatic or lymph node metastases were biopsied under ultrasound or CT guidance. Venous occlusion or arterial encasement were clear signs of local tumor ingrowth. ERCP was performed in all patients, and in case of obstructive jaundice, the common bile duct was stented with a polyethylene endoprosthesis for drainage. CT scans were often made in the referring hospital and were always reviewed by experienced radiologists. Patients with hepatic malignancies underwent three-phase CT scans. Since September 1995, helical CT scans were performed in all patients with a distal bile duct stenosis. Visceral angiography was seldom used to exclude vascular involvement. Diagnostic laparoscopy and laparoscopic ultrasonography Laparoscopy was performed under general anesthesia, as described above.6,7,13,14 A CO2 pneumoperitoneum was induced and, using 3 10- to 11-mm cannulas (umbilical, left, and right subcostal), the abdominal cavity was investigated for peritoneal and hepatic deposits and for malignant infiltration of the hepatoduodenal ligament and the mesocolon. The abdominal cavity was irrigated with 500mL isotonic saline, which was sampled for cytologic examination. The cytology results are still subjects of investigation and did not influence treatment strategy. A 7.5-MHz linear array US-probe (UST-5522-7.5; Aloka Co, Tokyo, Japan) was used to examine the liver for intrahepatic metastases, to evaluate the pancreas and the portal and superior mesenteric vessels, and to investigate the celiac axis for lymph node metastases. Biopsies of suspected metastatic lesions were taken under direct laparoscopic or ultrasonographic guidance using biopsy forceps or True-cut (Travenol; Baxter Healthcare Corporation, Deerfield, IL) and Rotex (Ursus Konsult AB; Stockholm, Sweden) biopsy needles. Laparoscopy findings were recorded on a separate form by both surgeon and radiologist. Complications and treatment after laparoscopy were registered. Minor complications were those that subsided without additional treatment; patients with major complications needed additional treatment. All tumors were restaged after laparoscopy. Patients with tumors judged to be resectable after laparoscopy underwent a laparotomy. A tumor was considered unresectable if metastatic disease or tumor ingrowth was detected during laparoscopy and was proved with a biopsy. If metastatic disease or tumor ingrowth shown at laparoscopy could not be proved
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
461
Table 1. Patient Characteristics and Reasons for Exclusion During and After 1995 Tumor localization
n
Insufficient examination*
Exclusions
n
Male: female
Age mean (SD)
Esophagus before 199513 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Liver primary Liver secondary Pancreas body or tail Total
56
1
3†
52
40:16
63(11)
36 73 141 53 10 36 15 420
1 1 4 3 0 3 0 13
0 2‡ 4§ 3㛳 0 0 0 12
35 70 133 47 10 33 15 395
28:8 42:28 72:61 27:20 6:4 14:19 8:7 237:163
61(10) 61(10) 63(10) 57(13) 69(17) 56(14) 56(11) 60(11)
*Insufficient examination caused by adhesions after previous abdominal surgery. † One patient refused laparotomy, two patients with tumor ingrowth in the bronchus that became clear during thoracotomy. ‡ One patient with cholecystitis, one patient with liver cirrhosis and portal hypertension. § Two patients died of cholangitis and sepsis before laparotomy, two patients were explored more than two months after laparoscopy. 㛳 Three patients with primary sclerosing cholangitis.
with histology, an explorative laparotomy was performed. Laparotomy was performed preferably within 4 weeks after laparoscopy. Almost all patients with obstructive jaundice underwent preoperative internal biliary drainage. Palliative therapy Patients restaged with an unresectable tumor after laparoscopy underwent nonoperative palliation. This indicated a Tygon-tube (Schneider Europe AG, Bulack, Switzerland) or a Wallstent (Schneider) for esophageal and gastroesophageal junction tumors, and a biliary polyethylene endoprothesis or a Wallstent for patients with obstructive jaundice. Bypass surgery was performed when periampullary tumors were unresectable during laparotomy. All patients were discussed in a multidisciplinary oncology meeting for adjuvant chemo- or radiotherapy within ongoing European Organization for Registration of Treatment for Cancer trials or local study protocols. Curative therapy When curative resection was judged to be feasible, explorative laparotomy was performed. Patients with adenocarcinoma of the esophagus or gastroesophageal junction were included in a study protocol and randomized for transhiatal or transthoracic resection. Patients with proximal bile duct tumors underwent a hilar resection with or without a partial hepatectomy and patients with a liver tumor underwent a segmentectomy or hemihepatectomy. Patients with pancreatic corpus or tail tumors underwent distal pancreatectomy, and patients with periampullary tumors underwent a pylorus-preserving pancreatoduodenectomy.
RESULTS Diagnostic laparoscopy combined with laparoscopic ultrasonography was performed in 420 patients; 92 patients with an esophageal-gastric tumor, 214 patients with a periampullary tumor, 53 patients with a proximal bile duct tumor, 46 patients with a liver tumor, and 15 patients with a pancreatic body or tail tumor. A total of 395 patients were evaluated; 25 patients were excluded. Patient characteristics and reasons for exclusion are mentioned in Table 1. Laparoscopic detection of metastatic disease Metastatic disease proved with a biopsy was found at laparoscopy in 81 of 395 patients (21%) (Table 2). One patient with a metastatic tumor of the gastroesophageal junction underwent resection of the tumor and the metastasis in the triangular ligament. The other 80 patients (20%) did not undergo explorative laparotomy. Of the patients diagnosed as having a metastatic proximal bile duct obstruction, 6 had a metastatic gallbladder carcinoma, 1 had a hilar metastasis of a gastric tumor, 1 had a hilar metastasis of an ovarian tumor, and 11 had a proximal bile duct tumor with metastases. Three of the patients with metastatic liver tumors had a cirrhotic liver, a finding not expected before laparoscopy. Ten patients (3%) had laparoscopically detected metastases but the biopsies were negative. These 10 patients underwent an exploration and were all proved to have a metastatic tumor. During exploration in 26 patients, metastases were detected that were missed during laparoscopy (Table 2). The specificity and positive predictive value of diagnostic laparoscopy were 100% for me-
462
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
J Am Coll Surg
Table 2. Laparoscopic Detection of Metastases Before 1995
Tumor localization
Laparoscopy
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Laparoscopydetected metastases histology (ⴙ) n
%
n
n
%
n
%
52
3
5
49
3
6
1
2
35 70 133 47 10 33 15 395
8 12 18 19 4 11 6 81
20 17 14 40 40 33 40 21
28 58 115 28 6 22 9 315
3 4 — — — — — 10
11 6 — — — — — 3
2 5 12 1 1 3 0 25
7 7 9 4 17 14 — 8
Laparoscopic detection of local tumor ingrowth Tumor ingrowth was seen laparoscopically but was not proved by pathology in 46 of the 315 patients (15%) who underwent explorative laparotomy (Table 4). Three patients with an esophageal tumor had tumor ingrowth in the diaphragm that could be resected along with the tumor. Diagnostic laparoscopy correctly detected a total of 32 of 37 patients with a locally unresectable periampullary tumor, 5 patients were false-positively staged as unresectable: 1 patient after previous radiotherapy of the abdomen for a testicular tumor, 1 patient with a Wallstent that caused shadows on the laparoscopic ultrasound that were mistaken for tumor ingrowth, and 3 patients with pancreatitis surrounding the tumor and giving the impression of tumor ingrowth in the superior mesTable 3. Sensitivity and Negative Predictive Value of Laparoscopic Detection of Metastases Before 1995
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Metastasis at laparotomy
n
tastases detection, because all metastases were histologically proved. Sensitivity was 70% and negative predictive value 89% (Table 3).
Tumor localization
Exploration
Laparoscopydetected metastases histology (ⴚ)
Sensitivity
Negative predictive value
0.42
0.92
0.62 0.57 0.60 0.95 0.80 0.79 1 0.70
0.81 0.84 0.90 0.96 0.83 0.86 1 0.89
enteric or portal vein. These five patients underwent a resection with tumor-negative resection margins. Tumor ingrowth was not detected during laparoscopy in 48 of 315 patients (15%) who appeared unresectable during laparotomy (Table 4). The specificity and positive predictive value of detection of tumor ingrowth with laparoscopic staging were 100% for all tumors except periampullary tumors with a specificity of 95% and a positive predictive value of 86% (Table 5). The sensitivity of laparoscopically detected tumor ingrowth was 46% (range 30% to 100%) for the different tumors and a negative predictive value of 82% (range 72% to 100%) (Table 5). Complications In 15 of 395 patients (4%) a complication was registered. Three patients (1%) had a major complication: anaphylactic shock, bile leakage after a liver biopsy, and a small bowel perforation. Twelve patients (3%) had minor complications: wound infections (four), wound hematoma (two), pain in the upper abdomen (three), aspiration pneumonia (one), urinary retention (one), and incisional hernation (one). No mortality occurred after laparoscopy. Followup Port-site metastases were diagnosed in 8 patients (2%), at a median of 5 months (range 2 to 10 months) after laparoscopy. Port-site metastases developed only in the laparoscopy scars and not in the laparotomy scars. All patients were in an end stage of their disease with extensive peritoneal spread of the tumor in the abdomen at the time of diagnosis of the
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
463
Table 4. Laparoscopic Detection of Tumor Ingrowth Before 1995 Tumor localization
Laparoscopy-detected histology (ⴚ)
Tumor ingrowth during laparotomy
n
n
%
n
%
49
3
6
0
—
28 58 115 28 6 22 9 315
0 13 24 3 2 1 0 46
— 22 21 11 33 5 — 15
4 9 25 7 0 1 2 48
15 15 22 25 — 5 22 15
Exploration
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
port-site metastases. One patient received radiotherapy on the port-site metastasis because of pain. DISCUSSION The benefit of staging gastrointestinal tumors with laparoscopy and laparoscopic ultrasonography has already been reported.1,4,5,9,10,15-18 Most studies were performed in a relatively small number of patients and so far only one publication describes more extensive experience with laparoscopic staging.3 After pilot studies by our group, patients with tumors at the gastroesophageal junction or periampullary region are routinely staged with laparoscopy and laparoscopic ultrasonography.13,14 In the present study laparoscopic staging performed as a routine examination in a large group of patients with gastrointestinal malignancies is evaluated. Laparoscopy combined with laparoscopic ultrasonography could prove metastatic disease and prevent unnecessary laparotomies in 20% of patients, ranging from 5% to 40% for the different tumors. Tumor ingrowth in surrounding structures or organs could be diagnosed during lapa-
roscopy or laparoscopic ultrasonography and was detected in 15% of patients (range 5% to 33%). Because tumor ingrowth was not proved by biopsy, these patients all underwent laparotomy to examine this feature. Laparoscopic staging is described to prevent laparotomies in 14% to 16% of patients with distal esophageal and gastric-cardia tumors, which is in accordance with our results.15,18 Our group has reported the low benefit of laparoscopic staging in patients with esophageal tumors; laparotomy could be prevented in only 1 of 38 patients (3%) with a middle- or lower-esophageal tumor, but in 2 of 18 patients (11%) with a gastroesophageal junction tumor.13 When laparoscopic staging was only performed in patients with tumors at the gastroesophageal junction, the benefit increased to 20%. Patients with proximal bile duct tumors benefit the most from laparoscopic staging (40%). This can be partly explained by a change in diagnosis after laparoscopy; often a hilar stenosis is not caused by a bile duct tumor but by a gallbladder tumor or a tu-
Table 5. Results of Laparoscopic Detection of Tumor Ingrowth Before 1995
Tumor localization
Sensitivity
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Specificity
Positive predictive value
Negative predictive value
1
1
1
1
— 0.59 0.43 0.30 1 0.50 — 0.46
1 0.96 0.94 1 1 1 1 0.98
— 0.93 0.83 1 1 1 — 0.89
0.85 0.74 0.73 0.72 1 0.95 0.78 0.82
464
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
mor of the cystic duct growing into the hepatoduodenal ligament and with liver metastases.19 At followup, 15% of patients who did not undergo explorative laparotomy after laparoscopy required operative palliation.20 Despite this, laparoscopic staging still prevents a high percentage of unnecessary laparotomies. This is especially important for patients with gallbladder and cystic duct carcinoma, because adequate nonoperative palliation is available and minimal survival is expected. Laparoscopic staging of patients with liver tumors could prove metastases in 40% of primary and 33% of secondary liver tumors. In the literature, a comparable benefit of diagnostic laparoscopy is reported.1,10,21 Assessment of liver cirrhosis seems essential in patients designated for liver resection. Laparoscopic liver biopsies have a 100% diagnostic yield; “blind” liver biopsies are inconclusive in 20% of patients.22 Complications of laparoscopic staging are low and the benefit for the patient is high because surgical palliation is not an option.22 In the few patients with a pancreatic body or tail tumor, laparoscopic staging prevented unnecessary laparotomy in 40%. One important reason is that a tumor in the more proximal part of the pancreas is usually discovered in a later phase of the disease, increasing the chance of laparoscopically detectable metastases or tumor ingrowth. The initial high benefit of laparoscopic staging of periampullary tumors as described in the earlier studies could not be reproduced.3,9 In our first analysis, laparotomy was prevented in 19% of the patients.14 In this study diagnostic laparoscopy prevented 14% of unnecessary laparotomies, with a sensitivity of 60%. This decrease is also mentioned by Fernandezdel Castillo and colleagues, who had an initial benefit of 30% in the patients until 1989, and after 1989 this benefit dropped to 18%.3 Possible causes are improvement of preoperative staging with helical CT scans and color-Doppler or diagnosis in an earlier phase of disease.23-25 In this study 9% of metastases were found during laparotomy and missed with laparoscopy, indicating that a higher yield is possible. Tumor ingrowth detected during laparoscopy but not proved with a biopsy is still considered insufficient evidence of unresectability. In 5 of 37 patients with periampullary tumors, the diagnosis of tumor ingrowth was incorrect. These false positives indicate the need for careful interpretation of laparoscopic ultrasound results in patients with periampullary tumors. If tumor ingrowth diagnosed during laparo-
J Am Coll Surg
scopic ultrasonography is accepted as a standard to cancel laparotomy, some patients with curative tumors will be excluded from exploration. When increasing experience of the radiologist and surgeon decreases the percentage of false-positive results of laparoscopic staging of tumor ingrowth, tumor ingrowth seen with laparoscopic ultrasonography may become an exclusion criterion for explorative laparotomy, and the benefit of laparoscopic staging will increase. This will also depend heavily on the philosophy toward the possibility to perform a radical resection in more borderline cases. In our opinion the total of five false-positive findings of tumor ingrowth in the present series makes this a weak criterion of unresectability. If tumor ingrowth detected with laparoscopic ultrasonography becomes an exclusion criterion for laparotomy, 22% of patients with tumor ingrowth are still missed. Another issue in the choice for diagnostic laparoscopy is the approach toward the best palliative treatment. Although excellent results can be achieved by insertion of an endoprosthesis, a considerable number of patients need a laparotomy because of insufficient nonoperative palliation at a later date.20 It has also been shown that surgical palliation has a low morbidity and mortality and is indicated in patients who are expected to survive more than 6 months.26,27 In this study of more than 400 patients, laparoscopic staging was a safe method with low morbidity and without mortality. A special problem is the appearance of port-site metastases in the port-site scars, in this study in 8 of 395 patients (2%).28 The mechanism of these port-site metastases is unknown. The pneumoperitoneum leading to a flow of tumor cells, repeated introduction of trocars through the port sites with possible contamination of tumor cells, and increased manipulation in the abdomen caused by the lack of direct hand contact are all possible causative mechanisms.29-32 In our series port-site metastases developed exclusively in the laparoscopy scars and not in the laparotomy scar. Until now only patients in an end stage of their disease developed port-site metastases as part of their intraabdominal tumor spread. Our opinion was not to take biopsies of the primary tumor as long as it is thought to be resectable to prevent tumor spread in patients with resectable lesions. So biopsies were only taken to prove unresectable disease. This approach can be advised, but may not prevent all port-site metastases. It can be concluded from this study that diagnos-
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
tic laparoscopy combined with laparoscopic ultrasonography has value as a staging technique and should be performed in patients with liver tumors, suspicion of proximal bile duct tumors, pancreatic body and tail tumors, and gastroesophageal junction tumors. The need for laparoscopic staging of patients with periampullary tumors should be further investigated because the number of laparotomies prevented is rather low. It is unclear whether the preferred palliative treatment is nonoperative in these patients. Because the morbidity and mortality of bypass surgery has become very low, the question arises if patients with periampullary tumors should be extensively staged or just have a CT scan and, if no overt metastatic disease is shown, undergo an explorative laparotomy with a resection or double bypass. References 1. Babineau TJ, Lewis WD, Jenkins RL, et al. Role of staging laparoscopy in the treatment of hepatic malignancy. Am J Surg 1994; 167:151–155. 2. Cuschieri A. Laparoscopic management of cancer patients. J R Coll Surg Edinb 1995;40:1–9. 3. Fernandez-del Castillo C, Rattner DW, Warshaw AL. Further experience with laparoscopy and peritoneal cytology in the staging of pancreatic cancer. Br J Surg 1995;82:1127–1129. 4. Hemming AW, Nagy AG, Scudamore CH, Edelmann K. Laparoscopic staging of intraabdominal malignancy. Surg Endosc 1995; 9:325–328. 5. Watt I, Stewart I, Anderson D, et al. Laparoscopy, ultrasound and computed tomography in cancer of the esophagus and gastric cardia: a prospective comparison for detecting intra-abdominal metastases. Br J Surg 1989;76:1036–1039. 6. Delden OMv, de Wit LT, Nieveen van Dijkum EJM, et al. Value of laparoscopic ultrasonography in staging of proximal bile duct tumors. J Ultrasound Med 1997;16:7–12. 7. Delden OMv, Smits NJ, Bemelman WA, et al. Comparison of laparoscopic and transabdominal ultrasonography in staging of cancer of the pancreatic head region. J Ultrasound Med 1996;15: 207–212. 8. Hunerbein M, Rau B, Schlag PM. Laparoscopy and laparoscopic ultrasound for staging of upper gastrointestinal tumors. Eur J Surg Oncol 1995;21:50–55. 9. John TG, Greig JD, Carter DC, Garden OJ. Carcinoma of the pancreatic head and periampullary region. Tumor staging with laparoscopy and laparoscopic ultrasonography. Ann Surg 1995; 221:156–164. 10. John TG, Greig JD, Crosbie JL, et al. Superior staging of liver tumors with laparoscopy and laparoscopic ultrasound. Ann Surg 1994;220:711–719. 11. John TG, Garden OJ. Laparoscopic ultrasonography: extending the scope of diagnostic laparoscopy. Br J Surg 1994;81:5–6. 12. Gouma DJ, de Wit LT, Nieveen van Dijkum EJM, et al. Laparoscopic ultrasonography for staging gastrointestinal malignancy. Scand J Gastroenterol 1996;31:43–49.
Laparoscopy in Gastrointestinal Cancer
465
13. Bemelman WA, van Delden OM, van Lanschot JJ, et al. Laparoscopy and laparoscopic ultrasonography in staging of carcinoma of the esophagus and gastric cardia. J Am Coll Surg 1995;181:421– 425. 14. Bemelman WA, de Wit LT, van Delden OM, et al. Diagnostic laparoscopy combined with laparoscopic ultrasonography in staging of cancer of the pancreatic head region. Br J Surg 1995;82: 820–824. 15. O’Brien MG, Fitzgerald EF, Lee G, et al. A prospective comparison of laparoscopy and imaging in the staging of esophagogastric cancer before surgery. Am J Gastroenterol 1995;90:2191–2194. 16. Cuschieri A. Laparoscopy for pancreatic cancer: does it benefit the patient? Eur J Surg Oncol 1988;14:41–44. 17. Cuesta MA, Meijer S, Borgstein PJ, et al. Laparoscopic ultrasonography for hepatobiliary and pancreatic malignancy. Br J Surg 1993;80:1571–1574. 18. Anderson DN, Campbell S, Park KGM. Accuracy of laparoscopic ultrasonography in the staging of upper gastrointestinal malignancy. Br J Surg 1996;83:1424–1428. 19. Iida F, Kajikawa S, Horigome N. Evaluation of imaging examination for hepatic invasion of carcinoma of the gallbladder and postoperative patient outcome. J Am Coll Surg 1995;180:72–76. 20. Nieveen van Dijkum EJM, de Wit LT, Delden OMv, et al. The efficacy of laparoscopic staging in patients with upper gastrointestinal tumors. Cancer 1997;79:1315–1319. 21. Ishida H, Dohzono T, Furukawa Y, et al. Laparoscopy and biopsy in the diagnosis of malignant intra-abdominal tumors. Endoscopy 1984;16:140–142. 22. Vargas C, Jeffers LJ, Bernstein D, et al. Diagnostic laparoscopy: a 5-year experience in a hepatology training program. Am J Gastroenterol 1995;90:1258–1262. 23. Tomiyama T, Ueno N, Tano S, et al. Assessment of arterial invasion in pancreatic cancer using color Doppler ultrasonography. Am J Gastroenterol 1996;91:1410–1416. 24. Fuhrman GM, Charnsangavej C, Abbruzzese JL, et al. Thinsection contrast-enhanced computed tomography accurately predicts the resectability of malignant pancreatic neoplasms. Am J Surg 1994;167:104–113. 25. Hommeyer SC, Freeny PC, Crabo LG. Carcinoma of the head of the pancreas: evaluation of the pancreaticoduodenal veins with dynamic CT—potential for improved accuracy in staging. Radiology 1995;196:233–238. 26. Bosch RPv, Schelling GPv, Klinkenbijl JH, et al. Guidelines for the application of surgery and endoprostheses in the palliation of obstructive jaundice in advanced cancer of the pancreas. Ann Surg 1994;219:18–24. 27. Wagensveld BAv, Coene PPLO, Gulik TM, et al. Biliary and gastric bypass surgery in the palliation of pancreatic head carcinoma: outcome of 126 patients. Br J Surg 1997;84:1402–1406. 28. Nieveen van Dijkum EJM, de Wit LT, Obertop H, Gouma DJ. Port-site metastases following diagnostic laparoscopy. Br J Surg 1996;83:1793–1794. 29. Cook TA, Dehn TCB. Port-site metastases in patients undergoing laparoscopy for gastrointestinal malignancy. Br J Surg 1996;83: 1419–1420. 30. Bouvy ND, Marquet RL, Jeekel H, Bonjer HJ. Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Ann Surg 1996;224:694–701. 31. Wexner SD, Cohen SM. Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 1995;82:295– 298. 32. Childers JM, Aqua KA, Surwit EA, et al. Abdominal-wall tumor implantation after laparoscopy for malignant conditions. Obstet Gynecol 1994;84:765–769.
and pancreatic body and tail tumors. The value of laparoscopic staging for patients with periampullary tumors is not as great as stated in previous studies and is still the subject of investigation. (J Am Coll Surg 1999;189: 459–465. © 1999 by the American College of Surgeons)
Background: Resection offers the only chance of cure to patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary tumors. Staging is essential to select patients who will benefit from operation because palliation can also be performed nonoperatively. Several studies, including limited numbers of patients, have shown that laparoscopic staging prevents unnecessary laparotomies, but it is doubtful whether general application of this staging method can be advised. The aim of this study was to assess the benefit of diagnostic laparoscopy for staging patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary tumors.
Accurate staging of gastrointestinal tumors selects patients who will benefit from curative surgery. The additional value of diagnostic laparoscopy in the preoperative workup has been described.1-12 Small superficial liver metastases and peritoneal deposits can be detected and biopsied under direct view.1-5 Using laparoscopic ultrasonography small intrahepatic metastases and tumor ingrowth in surrounding tissue can be diagnosed.6-11 Adding laparoscopy and laparoscopic ultrasonography to conventional staging protocols prevents unnecessary laparotomies in 11% to 48% of patients with gastrointestinal tumors.1,9,10 This approach is valid only when nonoperative palliation is feasible and an unnecessary laparotomy can be prevented, such as in patients with a malignancy of the esophagus, pancreas, liver, and bile ducts. Patients with other gastrointestinal tumors, for example, distal gastric and colorectal malignancies, almost always undergo laparotomy because operative palliation is preferred when a curative resection is impossible. Pilot studies performed by our group in patients with esophageal and gastroesophageal junction tumors and in patients with periampullary tumors showed that laparoscopic staging prevented unnecessary laparotomies in 3%, 11%, and 19% of the patients, respectively.13,14 It was decided to use diagnostic laparoscopy routinely for staging periampullary tumors.14 On the other hand, laparoscopy was of little value in staging esophageal tumors, and since
Study Design: Between June 1992 and December 1996, 420 patients with a resectable tumor after conventional staging underwent diagnostic laparoscopy combined with laparoscopic ultrasonography. Histologic proof of metastases or ingrowth was used to cancel laparotomy. Results: Laparoscopic staging avoided laparotomy in 20% of patients (sensitivity 0.70): 5% with an esophageal tumor, 20% with a gastroesophageal junction tumor, 15% with a periampullary tumor, 40% with a proximal bile duct tumor, 35% with a liver tumor, and 40% with a pancreatic body or tail tumor. Complications and port-site metastases were seen in 4% and 2% of patients, respectively. Conclusions: Laparoscopic staging is a safe procedure with low morbidity and without mortality in this series. It has shown no benefit in esophageal cancer, but seems beneficial for staging tumors located at the gastroesophageal junction, proximal bile duct tumors, liver tumors, No competing interests declared. Received May 28, 1999; Accepted July 7, 1999. From the Departments of Surgery (Nieveen van Dijkum, de Wit, Kruyt, van Lanschot, Obertop, Gouma), Radiology (van Delden), and Gastroenterology (Rauws), Academic Medical Center, Amsterdam, The Netherlands. Correspondence address: EJM Nieveen van Dijkum, Department of Surgery, Reinier de Graaf Gasthuis, Postbus 5011, 2600 GA Delft, The Netherlands. © 1999 by the American College of Surgeons Published by Elsevier Science Inc.
459
ISSN 1072-7515/99/$21.00 PII S1072-7515(99)00186-6
460
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
then this procedure has been omitted in the workup of those patients, but has been continued in patients with tumors at the gastroesophageal junction.13 In this study the results of laparoscopic staging of more than 400 patients with esophageal, gastroesophageal junction, and hepatopancreatobiliary (HPB) tumors were evaluated in an attempt to further elucidate the general applicability of diagnostic laparoscopy for staging gastrointestinal malignancies. The value of laparoscopic staging is expressed in specificity, sensitivity, and positive and negative predictive values for detection of metastases and tumor ingrowth, and in the number of prevented laparotomies. The morbidity and mortality of the procedure and the appearance of port-site metastases were analyzed. METHODS Between June 1992 and December 1996, patients with esophageal, gastroesophageal junction, or HPB tumors were included in this study, after noninvasive staging showed that the tumor was locally resectable and without distant metastases. Patients who were not fit for an extensive surgical procedure were excluded. Since June 1994, only patients with a tumor of the gastroesophageal junction were included because the pilot study showed a limited effect of laparoscopic staging for patients with tumors of the proximal and mid-esophagus.13 At the same time another pilot study showed a substantial benefit of laparoscopic staging for periampullary tumors, so diagnostic laparoscopy was included in the standard workup of these patients.14 The results of the pilot studies are shown separately. Staging before laparoscopy Patients with a tumor at the gastroesophageal junction were preoperatively staged with ultrasonography of the neck, a chest x-ray, and ultrasonography combined with color-Doppler of the abdomen. The liver and celiac axis were carefully examined for hepatic or lymph node metastases, and biopsies were performed under ultrasonographic guidance for histology or cytologic confirmation. Local tumor extension was assessed by endoscopic ultrasonography and in case of proximal tumors, by bronchoscopy. Indirect laryngoscopy was done to exclude tumor ingrowth in the recurrent laryngeal nerves. Patients with HPB tumors underwent transabdominal ultrasonography combined with color-
J Am Coll Surg
Doppler. Hepatic or lymph node metastases were biopsied under ultrasound or CT guidance. Venous occlusion or arterial encasement were clear signs of local tumor ingrowth. ERCP was performed in all patients, and in case of obstructive jaundice, the common bile duct was stented with a polyethylene endoprosthesis for drainage. CT scans were often made in the referring hospital and were always reviewed by experienced radiologists. Patients with hepatic malignancies underwent three-phase CT scans. Since September 1995, helical CT scans were performed in all patients with a distal bile duct stenosis. Visceral angiography was seldom used to exclude vascular involvement. Diagnostic laparoscopy and laparoscopic ultrasonography Laparoscopy was performed under general anesthesia, as described above.6,7,13,14 A CO2 pneumoperitoneum was induced and, using 3 10- to 11-mm cannulas (umbilical, left, and right subcostal), the abdominal cavity was investigated for peritoneal and hepatic deposits and for malignant infiltration of the hepatoduodenal ligament and the mesocolon. The abdominal cavity was irrigated with 500mL isotonic saline, which was sampled for cytologic examination. The cytology results are still subjects of investigation and did not influence treatment strategy. A 7.5-MHz linear array US-probe (UST-5522-7.5; Aloka Co, Tokyo, Japan) was used to examine the liver for intrahepatic metastases, to evaluate the pancreas and the portal and superior mesenteric vessels, and to investigate the celiac axis for lymph node metastases. Biopsies of suspected metastatic lesions were taken under direct laparoscopic or ultrasonographic guidance using biopsy forceps or True-cut (Travenol; Baxter Healthcare Corporation, Deerfield, IL) and Rotex (Ursus Konsult AB; Stockholm, Sweden) biopsy needles. Laparoscopy findings were recorded on a separate form by both surgeon and radiologist. Complications and treatment after laparoscopy were registered. Minor complications were those that subsided without additional treatment; patients with major complications needed additional treatment. All tumors were restaged after laparoscopy. Patients with tumors judged to be resectable after laparoscopy underwent a laparotomy. A tumor was considered unresectable if metastatic disease or tumor ingrowth was detected during laparoscopy and was proved with a biopsy. If metastatic disease or tumor ingrowth shown at laparoscopy could not be proved
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
461
Table 1. Patient Characteristics and Reasons for Exclusion During and After 1995 Tumor localization
n
Insufficient examination*
Exclusions
n
Male: female
Age mean (SD)
Esophagus before 199513 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Liver primary Liver secondary Pancreas body or tail Total
56
1
3†
52
40:16
63(11)
36 73 141 53 10 36 15 420
1 1 4 3 0 3 0 13
0 2‡ 4§ 3㛳 0 0 0 12
35 70 133 47 10 33 15 395
28:8 42:28 72:61 27:20 6:4 14:19 8:7 237:163
61(10) 61(10) 63(10) 57(13) 69(17) 56(14) 56(11) 60(11)
*Insufficient examination caused by adhesions after previous abdominal surgery. † One patient refused laparotomy, two patients with tumor ingrowth in the bronchus that became clear during thoracotomy. ‡ One patient with cholecystitis, one patient with liver cirrhosis and portal hypertension. § Two patients died of cholangitis and sepsis before laparotomy, two patients were explored more than two months after laparoscopy. 㛳 Three patients with primary sclerosing cholangitis.
with histology, an explorative laparotomy was performed. Laparotomy was performed preferably within 4 weeks after laparoscopy. Almost all patients with obstructive jaundice underwent preoperative internal biliary drainage. Palliative therapy Patients restaged with an unresectable tumor after laparoscopy underwent nonoperative palliation. This indicated a Tygon-tube (Schneider Europe AG, Bulack, Switzerland) or a Wallstent (Schneider) for esophageal and gastroesophageal junction tumors, and a biliary polyethylene endoprothesis or a Wallstent for patients with obstructive jaundice. Bypass surgery was performed when periampullary tumors were unresectable during laparotomy. All patients were discussed in a multidisciplinary oncology meeting for adjuvant chemo- or radiotherapy within ongoing European Organization for Registration of Treatment for Cancer trials or local study protocols. Curative therapy When curative resection was judged to be feasible, explorative laparotomy was performed. Patients with adenocarcinoma of the esophagus or gastroesophageal junction were included in a study protocol and randomized for transhiatal or transthoracic resection. Patients with proximal bile duct tumors underwent a hilar resection with or without a partial hepatectomy and patients with a liver tumor underwent a segmentectomy or hemihepatectomy. Patients with pancreatic corpus or tail tumors underwent distal pancreatectomy, and patients with periampullary tumors underwent a pylorus-preserving pancreatoduodenectomy.
RESULTS Diagnostic laparoscopy combined with laparoscopic ultrasonography was performed in 420 patients; 92 patients with an esophageal-gastric tumor, 214 patients with a periampullary tumor, 53 patients with a proximal bile duct tumor, 46 patients with a liver tumor, and 15 patients with a pancreatic body or tail tumor. A total of 395 patients were evaluated; 25 patients were excluded. Patient characteristics and reasons for exclusion are mentioned in Table 1. Laparoscopic detection of metastatic disease Metastatic disease proved with a biopsy was found at laparoscopy in 81 of 395 patients (21%) (Table 2). One patient with a metastatic tumor of the gastroesophageal junction underwent resection of the tumor and the metastasis in the triangular ligament. The other 80 patients (20%) did not undergo explorative laparotomy. Of the patients diagnosed as having a metastatic proximal bile duct obstruction, 6 had a metastatic gallbladder carcinoma, 1 had a hilar metastasis of a gastric tumor, 1 had a hilar metastasis of an ovarian tumor, and 11 had a proximal bile duct tumor with metastases. Three of the patients with metastatic liver tumors had a cirrhotic liver, a finding not expected before laparoscopy. Ten patients (3%) had laparoscopically detected metastases but the biopsies were negative. These 10 patients underwent an exploration and were all proved to have a metastatic tumor. During exploration in 26 patients, metastases were detected that were missed during laparoscopy (Table 2). The specificity and positive predictive value of diagnostic laparoscopy were 100% for me-
462
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
J Am Coll Surg
Table 2. Laparoscopic Detection of Metastases Before 1995
Tumor localization
Laparoscopy
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Laparoscopydetected metastases histology (ⴙ) n
%
n
n
%
n
%
52
3
5
49
3
6
1
2
35 70 133 47 10 33 15 395
8 12 18 19 4 11 6 81
20 17 14 40 40 33 40 21
28 58 115 28 6 22 9 315
3 4 — — — — — 10
11 6 — — — — — 3
2 5 12 1 1 3 0 25
7 7 9 4 17 14 — 8
Laparoscopic detection of local tumor ingrowth Tumor ingrowth was seen laparoscopically but was not proved by pathology in 46 of the 315 patients (15%) who underwent explorative laparotomy (Table 4). Three patients with an esophageal tumor had tumor ingrowth in the diaphragm that could be resected along with the tumor. Diagnostic laparoscopy correctly detected a total of 32 of 37 patients with a locally unresectable periampullary tumor, 5 patients were false-positively staged as unresectable: 1 patient after previous radiotherapy of the abdomen for a testicular tumor, 1 patient with a Wallstent that caused shadows on the laparoscopic ultrasound that were mistaken for tumor ingrowth, and 3 patients with pancreatitis surrounding the tumor and giving the impression of tumor ingrowth in the superior mesTable 3. Sensitivity and Negative Predictive Value of Laparoscopic Detection of Metastases Before 1995
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Metastasis at laparotomy
n
tastases detection, because all metastases were histologically proved. Sensitivity was 70% and negative predictive value 89% (Table 3).
Tumor localization
Exploration
Laparoscopydetected metastases histology (ⴚ)
Sensitivity
Negative predictive value
0.42
0.92
0.62 0.57 0.60 0.95 0.80 0.79 1 0.70
0.81 0.84 0.90 0.96 0.83 0.86 1 0.89
enteric or portal vein. These five patients underwent a resection with tumor-negative resection margins. Tumor ingrowth was not detected during laparoscopy in 48 of 315 patients (15%) who appeared unresectable during laparotomy (Table 4). The specificity and positive predictive value of detection of tumor ingrowth with laparoscopic staging were 100% for all tumors except periampullary tumors with a specificity of 95% and a positive predictive value of 86% (Table 5). The sensitivity of laparoscopically detected tumor ingrowth was 46% (range 30% to 100%) for the different tumors and a negative predictive value of 82% (range 72% to 100%) (Table 5). Complications In 15 of 395 patients (4%) a complication was registered. Three patients (1%) had a major complication: anaphylactic shock, bile leakage after a liver biopsy, and a small bowel perforation. Twelve patients (3%) had minor complications: wound infections (four), wound hematoma (two), pain in the upper abdomen (three), aspiration pneumonia (one), urinary retention (one), and incisional hernation (one). No mortality occurred after laparoscopy. Followup Port-site metastases were diagnosed in 8 patients (2%), at a median of 5 months (range 2 to 10 months) after laparoscopy. Port-site metastases developed only in the laparoscopy scars and not in the laparotomy scars. All patients were in an end stage of their disease with extensive peritoneal spread of the tumor in the abdomen at the time of diagnosis of the
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
463
Table 4. Laparoscopic Detection of Tumor Ingrowth Before 1995 Tumor localization
Laparoscopy-detected histology (ⴚ)
Tumor ingrowth during laparotomy
n
n
%
n
%
49
3
6
0
—
28 58 115 28 6 22 9 315
0 13 24 3 2 1 0 46
— 22 21 11 33 5 — 15
4 9 25 7 0 1 2 48
15 15 22 25 — 5 22 15
Exploration
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
port-site metastases. One patient received radiotherapy on the port-site metastasis because of pain. DISCUSSION The benefit of staging gastrointestinal tumors with laparoscopy and laparoscopic ultrasonography has already been reported.1,4,5,9,10,15-18 Most studies were performed in a relatively small number of patients and so far only one publication describes more extensive experience with laparoscopic staging.3 After pilot studies by our group, patients with tumors at the gastroesophageal junction or periampullary region are routinely staged with laparoscopy and laparoscopic ultrasonography.13,14 In the present study laparoscopic staging performed as a routine examination in a large group of patients with gastrointestinal malignancies is evaluated. Laparoscopy combined with laparoscopic ultrasonography could prove metastatic disease and prevent unnecessary laparotomies in 20% of patients, ranging from 5% to 40% for the different tumors. Tumor ingrowth in surrounding structures or organs could be diagnosed during lapa-
roscopy or laparoscopic ultrasonography and was detected in 15% of patients (range 5% to 33%). Because tumor ingrowth was not proved by biopsy, these patients all underwent laparotomy to examine this feature. Laparoscopic staging is described to prevent laparotomies in 14% to 16% of patients with distal esophageal and gastric-cardia tumors, which is in accordance with our results.15,18 Our group has reported the low benefit of laparoscopic staging in patients with esophageal tumors; laparotomy could be prevented in only 1 of 38 patients (3%) with a middle- or lower-esophageal tumor, but in 2 of 18 patients (11%) with a gastroesophageal junction tumor.13 When laparoscopic staging was only performed in patients with tumors at the gastroesophageal junction, the benefit increased to 20%. Patients with proximal bile duct tumors benefit the most from laparoscopic staging (40%). This can be partly explained by a change in diagnosis after laparoscopy; often a hilar stenosis is not caused by a bile duct tumor but by a gallbladder tumor or a tu-
Table 5. Results of Laparoscopic Detection of Tumor Ingrowth Before 1995
Tumor localization
Sensitivity
Esophagus before 1995 Gastroesophageal junction after 1995 Periampullary before 199514 Periampullary after 1995 Proximal bile duct Primary liver Secondary liver Pancreas body or tail Total 13
Specificity
Positive predictive value
Negative predictive value
1
1
1
1
— 0.59 0.43 0.30 1 0.50 — 0.46
1 0.96 0.94 1 1 1 1 0.98
— 0.93 0.83 1 1 1 — 0.89
0.85 0.74 0.73 0.72 1 0.95 0.78 0.82
464
Nieveen van Dijkum et al
Laparoscopy in Gastrointestinal Cancer
mor of the cystic duct growing into the hepatoduodenal ligament and with liver metastases.19 At followup, 15% of patients who did not undergo explorative laparotomy after laparoscopy required operative palliation.20 Despite this, laparoscopic staging still prevents a high percentage of unnecessary laparotomies. This is especially important for patients with gallbladder and cystic duct carcinoma, because adequate nonoperative palliation is available and minimal survival is expected. Laparoscopic staging of patients with liver tumors could prove metastases in 40% of primary and 33% of secondary liver tumors. In the literature, a comparable benefit of diagnostic laparoscopy is reported.1,10,21 Assessment of liver cirrhosis seems essential in patients designated for liver resection. Laparoscopic liver biopsies have a 100% diagnostic yield; “blind” liver biopsies are inconclusive in 20% of patients.22 Complications of laparoscopic staging are low and the benefit for the patient is high because surgical palliation is not an option.22 In the few patients with a pancreatic body or tail tumor, laparoscopic staging prevented unnecessary laparotomy in 40%. One important reason is that a tumor in the more proximal part of the pancreas is usually discovered in a later phase of the disease, increasing the chance of laparoscopically detectable metastases or tumor ingrowth. The initial high benefit of laparoscopic staging of periampullary tumors as described in the earlier studies could not be reproduced.3,9 In our first analysis, laparotomy was prevented in 19% of the patients.14 In this study diagnostic laparoscopy prevented 14% of unnecessary laparotomies, with a sensitivity of 60%. This decrease is also mentioned by Fernandezdel Castillo and colleagues, who had an initial benefit of 30% in the patients until 1989, and after 1989 this benefit dropped to 18%.3 Possible causes are improvement of preoperative staging with helical CT scans and color-Doppler or diagnosis in an earlier phase of disease.23-25 In this study 9% of metastases were found during laparotomy and missed with laparoscopy, indicating that a higher yield is possible. Tumor ingrowth detected during laparoscopy but not proved with a biopsy is still considered insufficient evidence of unresectability. In 5 of 37 patients with periampullary tumors, the diagnosis of tumor ingrowth was incorrect. These false positives indicate the need for careful interpretation of laparoscopic ultrasound results in patients with periampullary tumors. If tumor ingrowth diagnosed during laparo-
J Am Coll Surg
scopic ultrasonography is accepted as a standard to cancel laparotomy, some patients with curative tumors will be excluded from exploration. When increasing experience of the radiologist and surgeon decreases the percentage of false-positive results of laparoscopic staging of tumor ingrowth, tumor ingrowth seen with laparoscopic ultrasonography may become an exclusion criterion for explorative laparotomy, and the benefit of laparoscopic staging will increase. This will also depend heavily on the philosophy toward the possibility to perform a radical resection in more borderline cases. In our opinion the total of five false-positive findings of tumor ingrowth in the present series makes this a weak criterion of unresectability. If tumor ingrowth detected with laparoscopic ultrasonography becomes an exclusion criterion for laparotomy, 22% of patients with tumor ingrowth are still missed. Another issue in the choice for diagnostic laparoscopy is the approach toward the best palliative treatment. Although excellent results can be achieved by insertion of an endoprosthesis, a considerable number of patients need a laparotomy because of insufficient nonoperative palliation at a later date.20 It has also been shown that surgical palliation has a low morbidity and mortality and is indicated in patients who are expected to survive more than 6 months.26,27 In this study of more than 400 patients, laparoscopic staging was a safe method with low morbidity and without mortality. A special problem is the appearance of port-site metastases in the port-site scars, in this study in 8 of 395 patients (2%).28 The mechanism of these port-site metastases is unknown. The pneumoperitoneum leading to a flow of tumor cells, repeated introduction of trocars through the port sites with possible contamination of tumor cells, and increased manipulation in the abdomen caused by the lack of direct hand contact are all possible causative mechanisms.29-32 In our series port-site metastases developed exclusively in the laparoscopy scars and not in the laparotomy scar. Until now only patients in an end stage of their disease developed port-site metastases as part of their intraabdominal tumor spread. Our opinion was not to take biopsies of the primary tumor as long as it is thought to be resectable to prevent tumor spread in patients with resectable lesions. So biopsies were only taken to prove unresectable disease. This approach can be advised, but may not prevent all port-site metastases. It can be concluded from this study that diagnos-
Vol. 189, No. 5, November 1999 Nieveen van Dijkum et al
tic laparoscopy combined with laparoscopic ultrasonography has value as a staging technique and should be performed in patients with liver tumors, suspicion of proximal bile duct tumors, pancreatic body and tail tumors, and gastroesophageal junction tumors. The need for laparoscopic staging of patients with periampullary tumors should be further investigated because the number of laparotomies prevented is rather low. It is unclear whether the preferred palliative treatment is nonoperative in these patients. Because the morbidity and mortality of bypass surgery has become very low, the question arises if patients with periampullary tumors should be extensively staged or just have a CT scan and, if no overt metastatic disease is shown, undergo an explorative laparotomy with a resection or double bypass. References 1. Babineau TJ, Lewis WD, Jenkins RL, et al. Role of staging laparoscopy in the treatment of hepatic malignancy. Am J Surg 1994; 167:151–155. 2. Cuschieri A. Laparoscopic management of cancer patients. J R Coll Surg Edinb 1995;40:1–9. 3. Fernandez-del Castillo C, Rattner DW, Warshaw AL. Further experience with laparoscopy and peritoneal cytology in the staging of pancreatic cancer. Br J Surg 1995;82:1127–1129. 4. Hemming AW, Nagy AG, Scudamore CH, Edelmann K. Laparoscopic staging of intraabdominal malignancy. Surg Endosc 1995; 9:325–328. 5. Watt I, Stewart I, Anderson D, et al. Laparoscopy, ultrasound and computed tomography in cancer of the esophagus and gastric cardia: a prospective comparison for detecting intra-abdominal metastases. Br J Surg 1989;76:1036–1039. 6. Delden OMv, de Wit LT, Nieveen van Dijkum EJM, et al. Value of laparoscopic ultrasonography in staging of proximal bile duct tumors. J Ultrasound Med 1997;16:7–12. 7. Delden OMv, Smits NJ, Bemelman WA, et al. Comparison of laparoscopic and transabdominal ultrasonography in staging of cancer of the pancreatic head region. J Ultrasound Med 1996;15: 207–212. 8. Hunerbein M, Rau B, Schlag PM. Laparoscopy and laparoscopic ultrasound for staging of upper gastrointestinal tumors. Eur J Surg Oncol 1995;21:50–55. 9. John TG, Greig JD, Carter DC, Garden OJ. Carcinoma of the pancreatic head and periampullary region. Tumor staging with laparoscopy and laparoscopic ultrasonography. Ann Surg 1995; 221:156–164. 10. John TG, Greig JD, Crosbie JL, et al. Superior staging of liver tumors with laparoscopy and laparoscopic ultrasound. Ann Surg 1994;220:711–719. 11. John TG, Garden OJ. Laparoscopic ultrasonography: extending the scope of diagnostic laparoscopy. Br J Surg 1994;81:5–6. 12. Gouma DJ, de Wit LT, Nieveen van Dijkum EJM, et al. Laparoscopic ultrasonography for staging gastrointestinal malignancy. Scand J Gastroenterol 1996;31:43–49.
Laparoscopy in Gastrointestinal Cancer
465
13. Bemelman WA, van Delden OM, van Lanschot JJ, et al. Laparoscopy and laparoscopic ultrasonography in staging of carcinoma of the esophagus and gastric cardia. J Am Coll Surg 1995;181:421– 425. 14. Bemelman WA, de Wit LT, van Delden OM, et al. Diagnostic laparoscopy combined with laparoscopic ultrasonography in staging of cancer of the pancreatic head region. Br J Surg 1995;82: 820–824. 15. O’Brien MG, Fitzgerald EF, Lee G, et al. A prospective comparison of laparoscopy and imaging in the staging of esophagogastric cancer before surgery. Am J Gastroenterol 1995;90:2191–2194. 16. Cuschieri A. Laparoscopy for pancreatic cancer: does it benefit the patient? Eur J Surg Oncol 1988;14:41–44. 17. Cuesta MA, Meijer S, Borgstein PJ, et al. Laparoscopic ultrasonography for hepatobiliary and pancreatic malignancy. Br J Surg 1993;80:1571–1574. 18. Anderson DN, Campbell S, Park KGM. Accuracy of laparoscopic ultrasonography in the staging of upper gastrointestinal malignancy. Br J Surg 1996;83:1424–1428. 19. Iida F, Kajikawa S, Horigome N. Evaluation of imaging examination for hepatic invasion of carcinoma of the gallbladder and postoperative patient outcome. J Am Coll Surg 1995;180:72–76. 20. Nieveen van Dijkum EJM, de Wit LT, Delden OMv, et al. The efficacy of laparoscopic staging in patients with upper gastrointestinal tumors. Cancer 1997;79:1315–1319. 21. Ishida H, Dohzono T, Furukawa Y, et al. Laparoscopy and biopsy in the diagnosis of malignant intra-abdominal tumors. Endoscopy 1984;16:140–142. 22. Vargas C, Jeffers LJ, Bernstein D, et al. Diagnostic laparoscopy: a 5-year experience in a hepatology training program. Am J Gastroenterol 1995;90:1258–1262. 23. Tomiyama T, Ueno N, Tano S, et al. Assessment of arterial invasion in pancreatic cancer using color Doppler ultrasonography. Am J Gastroenterol 1996;91:1410–1416. 24. Fuhrman GM, Charnsangavej C, Abbruzzese JL, et al. Thinsection contrast-enhanced computed tomography accurately predicts the resectability of malignant pancreatic neoplasms. Am J Surg 1994;167:104–113. 25. Hommeyer SC, Freeny PC, Crabo LG. Carcinoma of the head of the pancreas: evaluation of the pancreaticoduodenal veins with dynamic CT—potential for improved accuracy in staging. Radiology 1995;196:233–238. 26. Bosch RPv, Schelling GPv, Klinkenbijl JH, et al. Guidelines for the application of surgery and endoprostheses in the palliation of obstructive jaundice in advanced cancer of the pancreas. Ann Surg 1994;219:18–24. 27. Wagensveld BAv, Coene PPLO, Gulik TM, et al. Biliary and gastric bypass surgery in the palliation of pancreatic head carcinoma: outcome of 126 patients. Br J Surg 1997;84:1402–1406. 28. Nieveen van Dijkum EJM, de Wit LT, Obertop H, Gouma DJ. Port-site metastases following diagnostic laparoscopy. Br J Surg 1996;83:1793–1794. 29. Cook TA, Dehn TCB. Port-site metastases in patients undergoing laparoscopy for gastrointestinal malignancy. Br J Surg 1996;83: 1419–1420. 30. Bouvy ND, Marquet RL, Jeekel H, Bonjer HJ. Impact of gas (less) laparoscopy and laparotomy on peritoneal tumor growth and abdominal wall metastases. Ann Surg 1996;224:694–701. 31. Wexner SD, Cohen SM. Port site metastases after laparoscopic colorectal surgery for cure of malignancy. Br J Surg 1995;82:295– 298. 32. Childers JM, Aqua KA, Surwit EA, et al. Abdominal-wall tumor implantation after laparoscopy for malignant conditions. Obstet Gynecol 1994;84:765–769.