ORIGINAL ARTICLE. The Impact of Laparoscopy and Laparoscopic Ultrasonography on the Management of Pancreatic Cancer

Transcription

1 ORIGINAL ARTICLE The Impact of Laparoscopy and Laparoscopic Ultrasonography on the Management of Pancreatic Cancer Pinhas P. Schachter, MD; Yona Avni, MD; Mordechai Shimonov, MD; Gabriela Gvirtz, MD; Ada Rosen, MD; Abraham Czerniak, MD Hypothesis: Laparoscopy and laparoscopic ultrasonographic (LAPUS) examinations combined with a biopsy of the pancreatic lesion contribute significantly in the determination of resectability of pancreatic cancer. Design: A prospective evaluation of the impact of laparoscopy and LAPUS on surgical decision making in patients with pancreatic cancer. Setting: A general community hospital; the department of surgery serves as referral for pancreatic surgery. Patients: During a 36-month period, 94 patients with pancreatic lesions were prospectively examined. Twenty-seven patients were found to have advanced disease. The remaining 67 patients were examined by laparoscopy and LAPUS to determine the resectability of the pancreatic tumor. Results: Laparoscopy and LAPUS contributed new, additional data in 40 patients (60%). Advanced disease was found in 30 patients, precluding curative resection. The study indicated potentially resectable tumors in 37 patients (55%), including 3 defined by conventional imaging studies as probably unresectable, and these patients were operated on with the intention of curative resection. Thirty-three patients underwent resection, and 4 (6%) were found to have nonresectable disease and form the false-positive group of the study. A summary of the results shows that the study resulted in a change of the decision regarding surgical intervention in 24 patients (36%) and avoided unnecessary laparotomies in 21 (31%). The study had a sensitivity of 100%, a specificity of 88%, and a false-positive rate of 6%. The positive predictive value of the study is 89%, and the negative predictive value is 100%. Conclusions: Although rather invasive procedures that require general anesthesia and hospitalization, laparoscopy and LAPUS significantly contribute to the staging of patients with potentially resectable pancreatic cancer, avoiding unnecessary explorative laparotomies. These procedures should be performed in all patients with potentially resectable pancreatic cancer before explorative laparotomy. Arch Surg. 2000;135: From the Departments of Surgery A (Drs Schachter, Shimonov, Rosen, and Czerniak), Gastroenterology (Dr Avni), and Radiology (Dr Gvirtz), E. Wolfson Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel. PANCREATIC CANCER presents a difficult diagnostic challenge since clinical signs and symptoms are vague and appear late in the course of the disease. 1-6 Surgical resection is the only prospect for cure, yet most patients have advanced disease at the time of diagnosis precluding curative surgery. Although pancreatic imaging and endoscopic nonsurgical techniques for the diagnosis and treatment of biliary obstruction have improved significantly, many patients still undergo explorative laparotomy to find an irresectable tumor Surgery in these patients is associated with significant morbidity and mortality, in particular in those not in need of a palliative bypass. The introduction of laparoscopy and laparoscopic ultrasonography (LAPUS) for the staging of pancreatic cancer has been shown to avoid unnecessary explorations This study evaluates prospectively the impact of LAPUS on staging and surgical decision making in patients with pancreatic tumors. RESULTS Laparoscopy and LAPUS were performed in 67 patients (31 men and 36 women; age range, years; mean age, 63.3 years). The average time of the study was 30 minutes (range, minutes). Patients were hospitalizedfor1to2days(mean,1.2days). No postoperative complications were noted. Mostlesionswereintheheadofthepancreas (in 48 patients [72%], including 3 with major duodenal papilla lesions and 5 with lesions of the uncinate process), while 19 were in the body or tail of the pancreas. 1303

2 PATIENTS AND METHODS During a 36-month period (April 1, 1996, through March 31, 1999), 94 patients with pancreatic tumors were screened for this study. Twenty-seven patients were found to have advanced disease, precluding curative resection. The remaining 67 patients with pancreatic tumors scheduled for explorative laparotomy were included in the study. All patients underwent a transabdominal ultrasonographic (US) study; contrast-enhanced, thin-sliced computed tomography; and, in some cases, endoscopic retrograde cholangiopancreatography. Ten patients with inconclusive computed tomographic results with regard to resectability were also examined via endoscopic US. Patients considered candidates for surgery were offered the option of laparoscopy and LAPUS as separate procedures, before the final determination of treatment. Laparoscopy with LAPUS was performed under general anesthesia, as a separate procedure by a standard technique. Briefly, 2 disposable, 10-mm cannulas were introduced: one at the umbilicus for the 30 telescope and the other at the right upper abdominal quadrant for the US probe. Additional ports were entered as needed for dissection of adhesions or to create an opening into the gastrocolic ligament to reach the surface of the pancreas (ie, the lesser sac). Laparoscopic US was performed with a 10- mm, 8-MHz sectoral contact US probe (laparoscopic ultrasound system; Sharplan Usight, Tel-Aviv, Israel). The system uses a simultaneous view of the laparoscopic and US images, and has the capability of a lateral US view of up to 90, a Doppler system to differentiate between blood vessels (vein vs artery) and bile ducts or cysts, and a LAPUSguided biopsy system using an 18-gauge biopsy needle (Quick core biopsy needle; Cook, Bloomington, Ind). All patients were prospectively studied using a protocol (Table 1). Laparoscopic and US examination of the body and tail of the pancreas was performed through an aperture in the gastrocolic ligament and entrance into the lesser sac. Special attention was directed to define involvement of portal and mesenteric vessels (Figure 1), and to identify enlarged lymph nodes ( 1 cm) in the peripancreatic region, the celiac axis region, the hepatoduodenal ligament, the mesenteric root, and the liver hilus (Figure 2). The liver was examined systematically for deep, occult metastases. Information obtained by LAPUS was compared with findings on conventional imaging studies with regard to additional data and the impact of this information on treatment strategy. Absolute criteria for irresectability were metastatic spread in the peritoneal cavity or the liver and involvement of the superior mesenteric artery or vein. Demonstration of regional lymph node involvement (coalescence of lymph nodes into a metastatic mass) was considered a contraindication for resection; however, enlargement of sporadic lymph nodes in itself was not considered an absolute contraindication for explorative laparotomy. The finding of tumors larger than 5 cm with no other variables of irresectability was not considered an absolute contraindication for exploration, particularly in instances of cystic lesions or endocrine tumors. False-positive results for laparoscopy and LAPUS are those patients defined resectable by LAPUS and found to have a nonresectable tumor at laparotomy. Predictive values for positive and negative test results were calculated. A flow chart (Figure 3) was used in the decision tree for the management of patients with pancreatic tumors. Table 1. Study Protocol Laparoscopic evaluation Peritoneal metastatic spread Superficial metastases: liver, spleen, lymph nodes, and other organs Direct invasion of adjacent organs LAPUS evaluation* Size and anatomic location of the tumor Ultrasonographic features: solid, cystic Relation to portal vein and mesenteric vessels Additional lesions in the pancreas Enlargement of lymph nodes Peripancreatic, hepatoduodenal ligament, liver hilus, celiac axis, mesenteric root, and splenic hilus Deep liver metastases *LAPUS indicates laparoscopic ultrasonography. Preoperative imaging studies defined 55 patients (82%) as suitable for resection and 12 (18%) as having locally advanced disease that might preclude resection (tumors 5 cm or suspected involvement of the portal vein). The LAPUS study revealed new, additional information in 40 patients (60%) (Table 2). Advanced disease was found by laparoscopy alone in 8 patients (12%) (liver metastases, peritoneal spread, or local invasion). Laparoscopic US revealed an additional 22 patients (33%) with advanced disease (involvement of mesenteric vessels and the portal vein, enlarged regional lymph nodes, or deep liver metastases), precluding curative resection (Table 3). In 3 of these patients, LAPUS-guided biopsy of the lesion revealed metastatic tumors in the pancreas (breast, renal, and colon origin), and another patient from this group was found to have pancreatic lymphoma. Nine of these 30 patients with irresectable tumors had symptoms of gastric outlet obstruction and underwent surgical bypass. The remaining 21 patients were treated conservatively. The LAPUS study indicated potentially resectable tumors in 37 patients (55%), and these patients were operated on with the intention of resection. Four patients (6%) were found to have nonresectable disease at explorative laparotomy (invasion of superior mesenteric vessels [n=2] and involvement of the celiac trunk [n=2]) and form the false-positive study group. The remaining 33 patients underwent pancreatic resection (pancreaticoduodenectomy, 22; distal pancreatectomy, 8; and local excisions of tumors of the major duodenal papilla, 3). More important, tumors in 3 patients of the group of 12 defined as probably unresectable by conventional preoperative studies were found resectable by LAPUS. Two underwent pancreaticoduodenectomy with resection of the involved portal vein segment, and the third had a large (7 6 cm), nonfunctional endocrine tumor in the head of the pancreas (T2 N0 M0 tumor). The study group included 4 patients with focal pancreatitis in the head of the pancreas. One patient was diagnosed by LAPUSguided tru-cut biopsy as having focal pancreatitis, and 1304

3 A A V T B B A P S C Figure 2. Laparoscopic ultrasonographic demonstration of lymph node involvement. A, A small ( 1-cm) lymph node (arrow) at the mesenteric root. B, A 1.2-cm lymph node at the liver hilus (arrow) adjacent to the portal vein (P) and the hepatic artery (A). In this patient, no other findings were noted. At exploration, a biopsy of the lymph node was performed, and the results were normal. The patient underwent curative resection. Figure 1. Laparoscopic ultrasonographic demonstration of vascular involvement by carcinoma of the head of the pancreas. A, Demonstration of tumor (T) involving the portal vein (V). The line (arrow) indicates the route of the biopsy needle. B, Demonstration of involvement of the superior mesenteric vein (S). C, A Doppler study showing the disturbed flow (arrow) in the superior mesenteric vein. was not operated on. The other 3 patients underwent pancreaticoduodenectomy and were found to have focal pancreatitis on histopathologic examination. In one patient diagnosed as having a pancreatic cyst, LAPUS with biopsy of the cystic wall and aspiration of the cystic content revealed a malignant neoplasm, and at surgery carcinoma of the pancreas was found. Endoscopic US performed in 10 patients correctly diagnosed one patient with a nonresectable tumor in the head of the pancreas as having involvement of the duodenum and encasement of the mesenteric vein. Endoscopic US was helpful in identifying a periampullary lesion and another small (1 cm) lesion in the head of the pancreas and in defining the proximity and possible involvement of the portal vein in 3 other patients. Endoscopic US contributed to the planning of the surgical intervention, yet did not alter the decision regarding explorative laparotomy. A summary of the results shows that use of LAPUS resulted in a change of decision regarding surgical intervention in 24 (36%) of the 67 patients. Thus, unnecessary laparotomy was avoided in 21 patients (31%). More important, it changed the surgical decision toward operation in 3 patients defined as potentially unresectable by conventional imaging studies. The study has a falsepositive rate of 6%. The predictive value for a positive test result is 89%; and for a negative test result, 100%. COMMENT Pancreatic cancer is one of the leading causes of cancerrelated death, with an increasing incidence over past years. 16 Resection of the tumor offers the only hope for prolonging survival, yet resectability rates are extremely low in these patients and palliative procedures do not alter their poor prognosis. 17 Modern imaging techniques (dynamic contrast-enhanced computed tomographic scanning) have improved significantly the assessment of resectability; however, based on imaging alone, the resectability rates vary between 50% and 88%. 18,19 Moreover, interventional radiological and endoscopic techniques offer nonoperative alternatives for alleviating the biliary obstruction, emphasizing the challenge to identify the patients who will benefit from resection, palliative surgery, interventional therapy, or supportive treatment only. 20 Laparoscopic diagnostic procedures have been adopted as a more precise method of staging pancreatic 1305

4 Tumor of Pancreas (n = 94) Table 2. A Summary of the Laparoscopic and Laparoscopic Ultrasonographic Findings (n = 27) Conventional Imaging Studies Resectable (n = 55) Probably (n = 12) Laparoscopy and Laparoscopic Ultrasound (n = 67) Finding No. of Patients Laparoscopy 14 Milliary peritoneal spread (ascites) 4 Superficial liver metastases 4 Direct invasion 6 Laparoscopic ultrasonography 40* Tumor size 5cm 11 Vascular involvement 18 Enlarged lymph nodes 1cm 15 Deep liver metastases 10 *A patient may have more than 1 positive finding. (n = 30) Resectable (n = 37) (n = 4) Explorative Laparotomy (n = 37) RESECTION Total (n = 33) Figure 3. A flow chart for the management of patients with pancreatic tumors. The figures outside the boxes represent patient numbers in our series. One asterisk indicates 3 patients with metastases to the pancreas from other sites and 1 with lymphoma of the pancreas; 2 asterisks, 3 patients from the group defined by conventional imaging studies as probably unresectable ; and 3 asterisks, the false-positive group. cancer. Studies 4,14,17,21 have reported a significant change in the staging of patients with pancreatic cancer at laparoscopy. However, laparoscopic evaluation alone is limited in the locoregional assessment of pancreatic cancer, and in the identification of small, occult liver metastases. 22,23 In our series, laparoscopic findings alone that defined irresectability were present in only 8 patients. Laparoscopy combined with peritoneal cytologic features will disclose an additional 8% of patients having positive cytologic features and advanced unresectable pancreatic cancer. 24,25 The addition of US is crucial to the laparoscopic staging of pancreatic lesions, enabling a precise evaluation of the primary tumor and its relations to major blood vessels; the demonstration of peripancreatic lymph nodes; and the determination of small, deep liver metastases. 18,20 Indeed, in our series, LAPUS examination enabled a correct staging of irresectability in an additional 22 patients. Twelve patients have had questionable results on conventional imaging regarding resectability. Following LAPUS, 2 of these underwent resection with vascular reconstruction and 1 underwent resection of a large endocrine tumor. These findings point at the high sensitivity of the study in determining involvement of portal and mesenteric vessels, enabling us to decide on and prepare properly for resection of the tumor with the involved segment of the portal vein. The results of endoscopic US are encouraging, especially in the identification of small pancreatic and periampullary tumors ( 3 cm) and in defining their relation to the portal vein. However, like others, 24 we found endoscopic US limited in the evaluation of superior mesenteric artery Table 3. Laparoscopic and Laparoscopic Ultrasonographic Findings Precluding Resection Finding No. of Patients* Laparoscopic 8 Milliary peritoneal spread (ascites) 4 Superficial liver metastases 4 Direct invasion 6 Laparoscopic ultrasonography 22 Tumor size 5cm 10 Vascular involvement 16 Enlarged lymph nodes 1cm 12 Deep liver metastases 10 *A patient may have more than 1 positive finding. Tumor size or enlarged lymph nodes alone were not considered a contraindication for exploration. involvement, due to either tumor size or distance from the transducer and suboptimal imaging of the liver and impossible detection of small liver metastases. Indeed, our study shows that LAPUS improved staging, sparing an unnecessary explorative laparotomy in a third of the patients. Others 14,18,20,26 reported similar results in predicting the resectability of pancreatic cancer. However, with the improvement in quality and resolution of imaging techniques in particular, the computed tomographic scan and the advent of endoscopic US the percentage of laparoscopically avoided laparotomies shows a decreasing tendency. 8,20,23 Most of the LAPUS systems use a linear US probe. The use of a sectoral US system rather than a linear one enables a direct, US-guided, 18-gauge core biopsy of pancreatic and hepatic lesions and enlarged lymph nodes at the liver hilum or celiac trunk. It is also possible to perform a fine-needle aspiration cytologic study of lesions adjacent to blood vessels and aspiration of cystic lesions. The necessity and importance of positive histologic features before surgery are debatable. In our series, histologic results prevented pancreatic resection in 4 patients (3 with metastatic disease to the pancreas and 1 with focal pancreatitis). Surgery may have been avoided in an additional 3 patients with focal pancreatitis. Histologic examination was attempted in 19 patients, and the results were correct in 17. The option of LAPUSguided tru-cut biopsy may be used for patients with suspected focal pancreatitis since other imaging studies, in- 1306

5 cluding LAPUS, do not offer conclusive differentiation from malignant masses. Indeed, 3 patients with focal pancreatitis underwent pancreaticoduodenectomy in the present series. The additional use of a Doppler system significantly enhances correct diagnosis and identification of blood vessels and dilated bile ducts and enables safe biopsy of cystic lesions. The performance of LAPUS demands technical expertise and is time-consuming, particularly in its early stage, having its own learning curve. In the last 6 months of the study, the average time of the examination was 21 minutes, while it was 39 minutes in the first period of the study. The approach we have presently adopted is to perform the diagnostic LAPUS as a separate procedure. All available information is then discussed at a multidisciplinary meeting, where management is jointly agreed on. We believe that this approach enables better planning and preparation of the operation and improves the costeffectiveness of operating room time. In conclusion, LAPUS offers an accurate, costeffective, and safe method of staging patients with pancreatic and periampullary cancer; in our opinion, it should be performed in all patients with potentially resectable pancreatic cancer before explorative laparotomy. Reprints: Pinhas P. Schachter, MD, Department of Surgery A, E. Wolfson Medical Center, Holon 58100, Israel ( nir-sc@inter.net.il). REFERENCES 1. Warshaw AL, Gu ZY, Wittenberg J, Waltman AC. Preoperative staging and assessment of resectability of pancreatic cancer. Arch Surg. 1990;125: Watt I, Stewart I, Anderson D, et al. Laparoscopy, ultrasound and computed tomography in cancer of the oesophagus and gastric cardia: a prospective comparison for detecting intra-abdominal metastases. Br J Surg. 1989;76: Cuschieri A, Hall AW, Clark J. Value of laparoscopy in the diagnosis and management of pancreatic carcinoma. Gut. 1978;19: Cuschieri A. Laparoscopy for pancreatic cancer: does it benefit the patient? Eur J Surg Oncol. 1988;14: Easter DW, Cuschieri A, Nathanson LK, Lavelle-Jones M. The utility of diagnostic laparoscopy for abdominal disorders: audit of 120 patients. Arch Surg. 1992; 127: Warshaw AL, Fernandez del Castillo F. Pancreatic carcinoma. N Engl J Med. 1992; 326: Singh SM, Longmire WP, Reber HA. Surgical palliation for pancreatic cancer: the UCLA experience. Ann Surg. 1990;212: Huibregtse K, Katon RM, Coene PP, Tytgat GNJ. Endoscopic palliative treatment in pancreatic cancer. Gastrointest Endosc. 1986;32: Shimi S, Banting S, Cuschieri A. Laparoscopy in the management of pancreatic cancer: endoscopic cholecystjejunostomy for advanced disease. Br J Surg. 1992; 79: Fletcher DR, Jones RM. Laparoscopic cholecystjejunostomy as palliation for obstructive jaundice in inoperable carcinoma of the pancreas. Surg Endosc. 1992; 6: Mouiel J, Katkhouda N, White S, Dumas R. Endolaparoscopic palliation of pancreatic cancer. Surg Laparosc Endosc. 1992;2: Nishizaki T, Matsumata T, Adachi E, Sugimachi K. Laparoscopy preferable to imaging procedures in detecting metastases of a pancreas carcinoma to the liver. Surg Endosc. 1994;8: Fernandez del Castillo F, Warshaw AL. Peritoneal metastases in pancreatic cancer. Hepatogastroenterology. 1993;40: 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: Bemelman WA, DeWit 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: Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, CA Cancer J Clin. 1996;65: Conlon KC, Dougherty EC, Klimstra DS, et al. The value of minimal access surgery in the staging of patients with potentially resectable peripancreatic malignancy. Ann Surg. 1996;223: Minnard EA, Conlon KC, Hoos A, et al. Laparoscopic ultrasound enhances standard laparoscopy in the staging of pancreatic cancer. Ann Surg. 1998;228: Fuhrman GM, Charnsangavej C, Abbruzzese JL, et al. Thin-section contrastenhanced computed tomography accurately predicts the resectability of malignant pancreatic neoplasms. Am J Surg. 1994;167: Friess H, Kleeff J, Silva JC, Sadowski C, Baer HU, Buchler MW. The role of diagnostic laparoscopy in pancreatic and periampullary malignancies. J Am Coll Surg. 1998;186: Warshaw AL, Tepper JE, Shipley WU. Laparoscopy in the staging and planning of therapy for pancreatic cancer. Am J Surg. 1986;151: Watanabe M, Takatori Y, Ueki K, et al. Pancreatic biopsy under visual control in conjunction with laparoscopy for diagnosis of pancreatic cancer. Endoscopy. 1989; 21: Ishida H, Dohzono T, Furukawa Y, Kobayashi M, Tsuneoka K. Laparoscopy and biopsy in the diagnosis of malignant intra-abdominal tumors. Endoscopy. 1984; 16: Merchant NB, Conlon KC, Saigo P, Dougherty E, Brennan MF. Positive peritoneal cytology predicts unresectability of pancreatic adenocarcinoma. J Am Coll Surg. 1999;188: Fernandez-del-Castillo CL, Warshaw AL. Pancreatic cancer: laparoscopic staging and peritoneal cytology. Surg Oncol Clin N Am. 1998;7: Callery MP, Strasberg SM, Doherty GM, et al. Staging laparoscopy with laparoscopic ultrasonography: optimizing resectability in hepatobiliary and pancreatic malignancy. J Am Coll Surg. 1997;185: