Laparoscopy-Assisted Distal Gastrectomy with Systemic Lymph Node Dissection for Early Gastric Carcinoma: A Review of 43 Cases

Transcription

1 Laparoscopy-Assisted Distal Gastrectomy with Systemic Lymph Node Dissection for Early Gastric Carcinoma: A Review of 43 Cases Michitaka Fujiwara, MD, Yasuhiro Kodera, MD, Yasushi Kasai, MD, Yasuaki Kanyama, MD, Kenji Hibi, MD, Katsuki Ito, MD, Seiji Akiyama, MD, Akimasa Nakao, MD, FACS BACKGROUND: STUDY DESIGN: RESULTS: CONCLUSIONS: Recently, laparoscopy and laparoscopy-assisted surgery have been used increasingly as lessinvasive alternatives to conventional open surgery. But the use of this approach in gastric carcinoma has received little attention, possibly from the low incidence of early-stage disease in the West and the relative complexity of the surgical procedure. A prospective feasibility study of laparoscopy-assisted distal gastrectomy was performed in patients with histologically confirmed gastric carcinoma located in the lower or middle third of the stomach. Patients whose preoperative evaluations, including endoscopic ultrasonography and computerized tomography, led to a diagnosis of T1 N0 stage disease, and who had no advanced disease discovered during laparoscopy, were eligible. Intraoperative blood loss, time of operation, mortality, and morbidity were assessed, along with the number of lymph nodes retrieved and shortterm survival. Between 1998 and 1999, 43 patients were enrolled. Laparoscopy-assisted distal gastrectomy was converted to an open procedure in one patient. There were no operative or in-hospital deaths, but the incidence of anastomotic leakage was 14% (6 of 43). The mean blood loss was 239 ml, the time of operation was 225 minutes, and lymph node retrieval was 20.2 nodes. These results are comparable with a series of conventional open operations. One patient died of recurrent disease, and all other patients remain disease-free to date. Port-site recurrence was not observed. Although laparoscopy-assisted distal gastrectomy was equivalent to open surgery in several clinical parameters, the relatively high morbidity was a drawback. Its appropriateness to gastric cancer surgery must be verified by further studies. ( J Am Coll Surg 2003;196: by the American College of Surgeons) No competing interests declared. Received February 12, 2002; Revised May 28, 2002; Accepted August 19, From the Department of Surgery II, Nagoya University School of Medicine (Fujiwara, Kodera, Kanyama, Hibi, Ito, Akiyama, Nakao), and the Department of Surgery, Chunichi Hospital (Kasai), Nagoya/Aichi, Japan. Correspondence address: Yasuhiro Kodera, MD, Department of Surgery II, Nagoya University School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya Gastric carcinoma remains one of the leading causes of cancer death. Despite the well-documented shift of adenocarcinoma to the cardia and lower esophagus in the western hemisphere, 1 cancer of the lower and middle third stomach is still common in Japan. Historical data from several institutions worldwide indicate that earlystage gastric cancer often can be cured by surgery. 2 In Japan, an aggressive surgical approach of gastrectomy plus extended systemic lymphadenectomy (D2 resection) has long been the standard treatment, even for superficial cancers. But investigators have recently begun to explore less invasive options to provide patients with an improved quality of life without compromising curability. Endoscopic mucosal resection has already been established in Japan as a standard treatment for a small subset of mucosal cancers. 3 This procedure is relatively uncommon outside of Japan because of the comparative rarity of early-stage disease, and differences in the diagnostic criteria of gastric mucosal cancer between Japan and the West. 4 Laparoscopic surgery is considered less invasive than open surgery, resulting in faster recovery and, consequently, an earlier return to work. The benefits observed for laparoscopic cholecystectomy have prompted surgeons to expand this approach to other procedures, including herniorrhaphy, 5 colorectal surgery, 6 and surgery for gastroesophageal reflux disease by the American College of Surgeons ISSN /03/$21.00 Published by Elsevier Science Inc. 75 PII S (02)

2 76 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy J Am Coll Surg Figure 1. The extent of resection achieved by D1 alpha resection. The # denotes lymph node numbers as defined by the Japanese Classification of Gastric Carcinoma. 13 Laparoscopic wedge resection of the stomach has been performed, but is appropriate for a limited number of patients because systemic lymphadenectomy for gastric cancer cannot be performed concomitantly. 8 The use of laparoscopy-assisted distal gastrectomy (LADG) for gastric cancer, first reported by Kitano and colleagues in 1994, 9 and subsequently explored by other investigators, is another less-invasive alternative to open surgery. Because lymphadenectomy can be performed concomitantly with this procedure, LADG might be applicable to a greater subset of patients with gastric carcinoma than endoscopic mucosal resection. This article describes a feasibility study on the use of LADG for gastric carcinoma performed between January 1998 and December Because there is no evidence that extended lymphadenectomy can be performed as well under laparoscopy as it can be by conventional open surgery, only patients at low risk for lymph node metastasis were enrolled. METHODS Eligibility Patients who had been diagnosed preoperatively as having histologically proved T1 N0-stage gastric cancer (invasion of the mucosa or submucosa with no apparent lymph node involvement), no previous abdominal surgery, and who were sufficiently fit to tolerate conventional open surgery, were eligible for enrollment. The T category was determined based on barium contrast studies, endoscopy, and endoscopic ultrasonography, and the N category was based on computed tomography and endoscopic ultrasonography. Patients who had been referred for salvage surgery after failure to achieve complete resection by endoscopic mucosal resection were also considered eligible. There were no restrictions based on the gross appearance (elevated or depressed) or histopathology (differentiated or undifferentiated). Conversion to open surgery was indicated when invasion of the gastric serosa or gross lymphadenopathy was detected laparoscopically, in which case extended lymphadenectomy was performed, or in the event of uncontrollable bleeding or other major complications. This study was recently approved by the Institutional Review Board (IRB) of Nagoya University, but not at the onset of the study. This is because it was not customary for the IRB of Nagoya University to review studies that were not pharmaceutical industry-related at the time the study was initiated. Nevertheless, the study was conducted under a rigid protocol with rigorous selection criteria without violating general human rights. Informed consent was obtained from all patients for participation in this study in accordance with the standards at Nagoya University in regard to surgical investigations. Extent of systemic lymphadenectomy LADG was accompanied by either D1 or D1 alpha lymph node dissection. D1 dissection denotes concomitant resection of perigastric lymph nodes (lymph node nos. 3, 4, 5, and 6) through dissection of the major gastric arteries at their base 13 and is a term in common usage. D1 alpha in this study denotes dissection of the lymph nodes along the right side of the cardia (no. 1), at the base of the left gastric artery (no. 7), along the common hepatic artery (no. 8), and at the base of the splenic artery (no. 11), in addition to the perigastric nodes (Fig. 1). D1 was performed in patients with a preoperative diagnosis of mucosal cancer or for patients undergoing salvage surgery after endoscopic mucosal resection, and D1 alpha was performed for gastric cancer when submucosal invasion was detected preoperatively.

3 Vol. 196, No. 1, January 2003 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy 77 Figure 3. Representative laparoscopic view of the right gastroepiploic artery isolated adjacent to the pancreas. RGEA, right gastroepiploic artery. Figure 2. Placement of the five trocars used for laparoscopyassisted distal gastrectomy. Surgical procedure Pneumoperitoneum was created through an Optiview trocar (Ethicon Endo-Surgery, Cincinnati, OH) inserted in the subumbilical position. A laparoscope (3CCD Video System SX-2, Olympus, Tokyo, Japan) was introduced through this port, and four other trocars (two 10-mm and two 5-mm ports) were placed as shown in Figure 2. The omentum was divided between the greater curvature of the stomach and the transverse colon, and omental branches from the gastroepiploic vessels were coagulated using laparoscopic coagulating shears (Harmonic Scarpel, Ethicon Endo-Surgery). The left gastroepiploic artery and vein were divided with a linear stapler (Endopath ETS Flex45, Ethicon Endo- Surgery). The right gastroepiploic vein was divided with the coagulating shears after being clamped on either side with surgical clips (Ligaclip ERCA, Ethicon Endo- Surgery), and the right gastroepiploic artery was similarly divided at the lower edge of the pancreas (Fig. 3). The infrapyloric artery and other less prominent branches were then divided to expose the duodenum. The antrum was elevated and the gastroduodenal artery was visualized. The right gastric artery was then divided along with the corresponding vein at its origin from the gastroduodenal or the proper hepatic artery (Fig. 4). The duodenum was divided distal to the pyloric ring using a linear stapler (Endopath ETS Flex45, Ethicon Endo- Surgery). Adipose tissue at the anterosuperior border of the pancreas was dissected, and the left gastric vein was exposed either posterior to the common hepatic artery or anterior to the splenic artery and divided using clips and coagulating shears. At this point, the common hepatic artery was exposed and the no. 8 lymph nodes were dissected in case of D1 alpha resection. The exposure, double clipping, and dissection of the left gastric artery at its origin were performed as follows. The lesser omen- Figure 4. Laparoscopic view during dissection of the right gastric artery using laparoscopic coagulating shears. The surgical clips are already in place. CHA, common hepatic artery; GDA, gastroduodenal artery; PHA, proper hepatic artery; RGA, right gastric artery.

4 78 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy J Am Coll Surg tum was divided with dissection of the right paracardial lymph nodes and vagal trunks. A 4- to 5-cm minilaparotomy incision was made and covered with a surgical drape (Steridrape, 3M, St Paul, MN). The distal stomach was delivered through the incision and dissected to an appropriate line of resection using a linear stapler (Proximate, Ethicon Endo-Surgery). Macroscopic confirmation of the adequacy of the resection line through gastrotomy was performed before the resection when the lesion was located in the middle third of the stomach. This procedure was performed exclusively outside the abdominal cavity, with the surgical wound protected by a surgical drape. A Billroth type I anastomosis was performed at the end of the staple line, using a circular stapler in most cases (Endopath ILS, Ethicon Endo-Surgery) introduced through an incision in the anterior wall of the remnant stomach. The incision was closed by another linear stapler (Linearstapler TL-60, Ethicon Endo-Surgery). The anastomosis was performed by hand when the remnant stomach was too small to enable the smooth introduction and withdrawal of the circular stapler at an adequate angle via the minilaparotomy. Histopathology of the surgical specimens The surgical specimens were examined macroscopically by the surgeons, and the lymph nodes were divided and classified into perigastric nodes (nos. 1, 3, 4, 5, and 6) and other regional lymph nodes (nos. 7, 8, 11). The stomach and lymph nodes were fixed in formalin and stained with hematoxylin and eosin for histopathologic evaluation of the depth of invasion and identification of lymph node metastasis. The final stage was based on the UICC stage classification criteria. 14 Postoperative monitoring of patients Blood loss, time of operation, length of postoperative hospital stay, and surgical complications were recorded. The followup regimen consisted of interim history, physical examination, hematology, and blood chemistry panels, including tests for carcinoembyonic antigen and carbohydrate antigen 19 9, performed every three months for the first postoperative year, and every six months after. Either abdominal ultrasonography or computerized tomography was performed every six months. RESULTS Forty-three patients with gastric carcinoma who fulfilled inclusion criteria underwent LADG at Nagoya University Hospital from 1998 to The degree of lymph node dissection was D1 in 15 patients and D1 alpha in 28 patients. There were 32 men and 11 women, with the mean age ( standard deviation) of years (range, 35 to 85 years). Mean height and weight of the patients were cm and kg. The mean time of operation was minutes (range, 115 to 370 minutes) and the intraoperative blood loss was ml (range, 16 to 1752 ml). One patient required conversion to open surgery because the oral margin of resection was positive on frozen section examination. Consequently, this operation was the longest of all cases at 370 minutes. No case was converted to open surgery because advanced disease was discovered. Two patients had blood loss in excess of 1000 ml observed in two patients. In one case, the proper hepatic artery was injured, but no discrete source of bleeding was identified in the second case. This patient was somewhat obese (body mass index 26 kg/m 2 ), and oozing from raw surfaces might have been responsible for the unexpected amount of blood loss. There were no operative or inhospital deaths. Postoperative complications occurred in seven patients (16%) and included six cases of anastomotic leakage, one case of wound infection, and a case of hematemesis from anastomotic ulcer. Five of the six cases of leakage occurred at the gastroduodenostomy of the Billroth type I reconstruction. Drainage tubes inserted at the time of surgery functioned well in all five cases and oral intake was resumed within a few weeks. Leakage in the remaining case occurred at the gastrotomy staple line in the stomach remnant through which the circular stapler had been inserted. This patient had to be treated by placement of an ultrasonographyguided drainage, but recovered well. No notable cardiovascular or respiratory complication were observed. The mean hospital stay was days (range, 11 to 65 days). The diameter of the tumor measured on the surgical specimens was 23 mm (range, 7 to 80 mm). The depth of invasion was T1 in 38 patients (mucosal in 23 patients and submucosal in 15 patients) and T2 (invasion of the muscularis propria and subserosa) in 5 patients. This indicates that although most patients were either correctly diagnosed or overdiagnosed as having submucosal

5 Vol. 196, No. 1, January 2003 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy 79 invasion rather than being a mucosal cancer, the depth of invasion was underestimated in five cases (12%), and the lesion was T2, not T1. The total number of regional lymph nodes retrieved was (range, 4 to 51); for D1 resection and for D1 alpha resection. The yield of perigastric lymph nodes was (range, 2 to 29); for D1 resection and for D1 alpha resection. The number of other regional lymph nodes (second-tier nodes) resected among patients who underwent D1 alpha resection was (range, 2 to 13). Lymph node metastases were discovered in two patients; in four nodes along the lesser curvature (no. 3) in one patient with submucosal cancer, and in one node at the base of the left gastric artery (no. 7) in another patient with T2 stage cancer that had invaded the subserosa. The pathologic stage was IA in 37 patients, IB in 5 patients, and II in one patient. One patient with a poorly differentiated T1 lesion without nodal metastasis (Stage IA) died of rapidly progressive recurrent disease with extensive systemic lymph node metastasis and peritoneal seeding two years after operation. All other patients are alive and disease-free after a followup of months. DISCUSSION The indications for minimally invasive surgery have been expanded to include resection of neoplastic disease. Colon cancer has been treated using this approach extensively 15 and is technically feasible and compares favorably with open colectomy in terms of surgical morbidity, mortality, 16 and survival, 17 although a clear benefit in terms of quality of life was not always evident. 18 The experience with colorectal surgery has been encouraging, and laparoscopic techniques have been developed to include surgery for gastric cancer. The evaluation of LADG remains at a preliminary phase. The first LADG with suboptimal (D0) lymph node dissection was performed for mucosal gastric carcinoma at Nagoya University Hospital in early Two additional D0 LADGs and several laparoscopic wedge resections for mucosal cancer and stromal tumors were performed before recruitment for the current study commenced in Operations were performed exclusively by a single surgeon with extensive experience both in gastric cancer surgery and general laparoscopic surgery. Despite these considerations, it is likely that most operators will experience a learning curve, as evidenced by high rate of anastomotic leakage and the two cases of major intraoperative hemorrhage. Although there is currently no data specific for LADG, analyses of the learning curves in laparoscopic herniorrhaphy and colectomy suggest that somewhere in the range of 30 to 50 cases is required for technical proficiency. 5,15 Laparoscopic approaches have been criticized in terms of time of operation, complications, unnecessary recurrence, and cost. Unfortunately, some of these criticisms are apt in the current study. Time of operation was comparable with several series of open surgery in Japan, 19,20 but should be considered as slightly prolonged given that more extensive lymphadenectomy (D2 resection) had been performed in most open procedures. On the other hand, the mean operative blood loss was less than 250 ml, even including the two cases of massive hemorrhage. These data are comparable with the one other series of LADG 21 and with data obtained from open surgery. 19,20 Anastomotic leakage, which developed in 14% of the patients (6 of 43), occurred throughout the 2-year period but has decreased recently (data not shown), indicating that it might be related to inexperience. Minor injury to the duodenal wall when manipulating the laparoscopic coagulating shears during isolation of the duodenum is one potential cause. Although we now perform thorough examination of the duodenum through the surgical wound after creating the anastomosis, this type of injury might have been unnoticed in the earlier cases. Excessive duodenal stump devascularization might have been another cause of anastomotic failure. Devascularization is kept to a minimum during open surgery, but it is often more difficult to decide where to stop the dissection under laparoscopy. Finally, tension on the anastomosis could be considered as a causative factor, given that a Kocher maneuver was not performed in this series. Our large experience with open surgery suggests that with most Japanese patients, who are relatively lean, Billroth type I anastomosis without a Kocher maneuver rarely causes any problems. The number of lymph nodes in the surgical specimen has been equated with the adequacy of node sampling. Although the lymph node harvest averaged 20 lymph nodes, the yield in 13 patients (30%) fell short of 15 lymph nodes, which is considered the minimal number required for evaluating pn in the TNM classification. 14 Lymph node harvest was comparable with that in another series of LADG 21 and several other series of open

6 80 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy J Am Coll Surg surgery at Japanese institutions. 19,20 Second-tier lymph nodes (lymph node nos. 7, 8, and 11) were also dissected by the laparoscopic approach and a mean of six lymph nodes were harvested. Asao and colleagues 12 recommend that this part of the operation be performed through the minilaparotomy incision after removal of the stomach. Unfortunately, lack of detailed information regarding lymph node retrieval in their series precludes comparison between their approach and ours. Survival after LADG was excellent. This is attributable to the strict eligibility criteria that resulted in a majority of patients having confirmed Stage IA disease. It should be borne in mind that one patient with T1 N0 stage died of recurrent cancer two years after surgery. Whether this should be interpreted as a pattern of failure peculiar to the laparoscopic procedure and pneumoperitoneum or can be dismissed as an unfortunate patient whose lesion had exceptionally high malignant potential cannot be decided at this time. A new series of LADG currently under way that includes evaluation of minimal residual disease and micrometastasis by immunostaining 22 and reverse transcriptase-polymerase chain reaction 23,24 might provide some clues to solve this query. A problem related to the indications for LADG was that 12% of the patients (5 of 43) who were considered eligible for this procedure had been understaged preoperatively and were found to have had T2 disease. The percentage of errors was similar in the other series. 12 Whether this turns out to be a major problem depends in part on whether extended lymphadenectomy more than D1 alpha is necessary for T2 disease. This question might be less of an issue in the West where D1 rather than D2 lymphadenectomy is usually performed. 25 To expand the indications for laparoscopic surgery to include T2 disease, further feasibility studies of this subset are needed to determine the incidence of unexpected tumor dissemination and port-site recurrence. Finally, staplers and other disposable instruments designed for and frequently used in laparoscopic surgery are much more expensive in Japan than in the West, which precludes their use in general hospitals with limited resources for performing experimental procedures. The median postoperative hospital stay of 15 days, although relatively short given the current standard of medical practice in Japan, does not seem to have been shortened sufficiently to justify the cost of these surgical instruments. LADG is currently performed in only a few centers. Although some successes have been reported, this study also identified various potential shortcomings. Further analysis in a phase II setting to establish whether LADG is as effective as open surgery in preventing recurrences and a randomized trial to establish whether clear qualityof-life benefits exist is needed before this procedure can be recommended as a standard treatment. Author Contributions Study conception and design: Kodera, Nakao Acquisition of data: Fujiwara, Kasai, Kanyama Analysis and interpretation of data: Fujiwara, Kodera Drafting of manuscript: Kodera Critical revision: Hibi, Akiyama, Ito Obtaining funding: Nakao Supervision: Nakao REFERENCES 1. Devesa S, Fraumeni JF. The rising incidence of gastric cardia cancer. J Natl Cancer Inst 1999;91: Everett SM, Axon ATR. Early gastric cancer in Europe. Gut 1997;41: Ono H, Kondo H, Gotoda T, et al. Endoscopic mucosal resection for treatment of early gastric cancer. Gut 2001;48: Schlemper RJ, Itabashi M, Kato Y, et al. Differences in diagnostic criteria for gastric carcinoma between Japanese and Western pathologists. Lancet 1997;349: DeTurris SV, Cacchione RN, Mungara A, et al. Laparoscopic herniorrhaphy: beyond the learning curve. J Am Coll Surg 2002;194: Phillips EH, Franklin M, Carroll BJ, et al. Laparoscopic colectomy. Ann Surg 1992;216: Cuschieri A. Laparoscopic antireflux surgery and repair of hiatal hernia. World J Surg 1993;17: Ohgami M, Otani Y, Kumai K, et al. Curative laparoscopic surgery for early gastric cancer: five years experience. World J Surg 1999;23: Kitano S, Iso Y, Moriyama M, Sugimachi K. Laparoscopyassisted Billroth I gastrectomy. Surg Laparo Endosc 1994;4: Hayashi K, Munekata Y. Laparoscopic assisted distal gastrectomy with regional lymphadenectomy. Surg Endosc 1998;12: Shiraishi N, Adachi Y, Kitano S, et al. Indication for an outcome of laparoscopy-assisted Billroth I gastrectomy. Br J Surg 1999; 86: Asao T, Hosouchi Y, Nakabayashi T, et al. Laparoscopically assisted total or distal gastrectomy with lymph node dissection for early gastric cancer. Br J Surg 2001;88: Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma. 2nd English Edition. Gastric Cancer 1998;1: Sobin LH, Wittekind CH. TNM classification of malignant tumors. 5th ed. Heidelberg: Springer-Verlag; 1997.

7 Vol. 196, No. 1, January 2003 Fujiwara et al Laparoscopy-Assisted Distal Gastrectomy Lin KM, Ota DM. Laparoscopic colectomy for cancer: an oncologic feasible option. Surg Oncol 2000;9: Delgato S, Lacy AM, Garcia Valdecasas JC, et al. Could age be an indication for laparoscopic colectomy in colorectal cancer? Surg Endosc 2000;14: Santoro E, Carlini M, Feroce A. Colorectal carcinoma: laparoscopic versus traditional open surgery. A clinical trial. Hepatogastroenterology 1999;46: Weeks JC, Nelson H, Gelber S, et al. Short-term quality-of-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 2002;287: Adachi W, Kobayashi M, Koike S, et al. The influence of excess body weight on the surgical treatment of patients with gastric cancer. Surg Today 1995;25: Dhar DK, Kubota H, Tachibana M, et al. Body mass index determines the success of lymph node dissection and predicts the outcome of gastric carcinoma patients. Oncology 2000;59: Adachi Y, Shiraishi N, Shiromizu A, et al. Laparoscopy-assisted Billroth I gastrectomy compared with conventional open gastrectomy. Arch Surg 2000;165: Fukagawa T, Sasako M, Mann GB, et al. Immunohistochemically detected micrometastases of the lymph nodes in patients with gastric carcinoma. Cancer 2001;92: Noguchi S, Hiratsuka M, Furukawa H, et al. Detection of gastric cancer micrometastases in lymph nodes by amplification of keratin 19 mrna with reverse transcriptase-polymerase chain reaction. Jpn J Cancer Res 1997;87: Kodera Y, Nakanishi H, Yamamura Y, et al. Prognostic value and clinical implication of disseminated cancer cells in the peritoneal cavity detected by reverse transcriptase-polymerase chain reaction and cytology. Int J Cancer 1998;79: Bonenkamp JJ, Hermans J, Sasako M, et al. Extended lymphnode dissection for gastric cancer. N Engl J Med 1999;340: