消化系惡性腫瘤（膽囊癌）

膽囊癌的預後如何

對於膽囊癌的預後與存活率有些相關的報告可供參考。

Tsukada 等人報告膽囊癌已有淋巴轉移的病人接受手術治療經驗，在1981-1995.有111 例進行標準的外科手術方式，切除膽囊，切除部份肝臟與肝外膽管，並清除附近的淋巴結，Kaplan-Meier等人發現膽囊癌無淋巴轉移患者，在T2-T4期，估計五年存活率為42.5% ，若有淋巴轉移則約為31%。15位T1病人無淋巴結轉移，46位屬於T2中48%，25位T3中的72%有轉移現象，T4的20位更達80%。以分期來說，五年存活率第一、二、三與四期分別是91%、85%、40%及19%。有 35例存活超過五年其中 11 位有淋巴結轉移，10例為第一期，13例為第二期，10 例屬第三期，第四期的患者也有兩例。第三期與第四期若能接受治癒性切除手術，五年存活率則有52%，明顯比不能完全切除的患者 5%好很多。腫瘤侵犯的深度與是否有淋巴結移轉有密切關係，這又與預後有些關連，正確的 TNM 分期有助於比較手術效果，預測預後，與決定是否加入其它治療手段。標準的廓清手術對於侵犯範圍較厲害的膽囊癌患者，還是有所幫忙。

另外一篇報告，是比較1989-1990 與 1994-1995兩段時間裡膽囊癌患者，罹病的數目並未有改變，但發現愈能早期診斷出來，愈後愈佳。可惜早期膽囊癌的比例，並未有增加現象。這報告指出存活時間與膽囊癌侵犯到那個階段有相當密切的關係，第零期五年存活率可達60% 第一期約39%，第二期為15%，第三期僅有 5%，到了第四期更剩下 1%。而預後與是否接受積極性治療並無密切關連。

大致說來，假若膽囊癌是意外被發現，例如在切下的膽囊上發現癌症病灶，那治癒率通常會超過百分之八十以上，若是在手術前就有了症狀，懷疑有了癌症病灶時，多數的膽囊癌細胞都已穿入膽囊的肌肉層或是漿膜層，此時完全治癒的機會就小於百分之五了。

　

某些文獻來看，對於外科醫師治療膽囊癌時是必要時再次開刀，或做廓清式手術仍無共識。這篇文章主要是探討膽囊切除後意外發現膽囊癌時要再次開刀？ 與及進行性膽囊癌接受廓清手術的結果。回顧性統計了149例患者，接受剖腹，其中 23例可完全清除，包括九例切除了三葉肝，十例切除了膽管，17位是發現切除膽囊後發現膽囊癌再次剖腹。五年內有 51% 無發病現象，平均追蹤 48 個月。Eight patients are alive beyond 50 months. There were no operative deaths; the perioperative morbidity rate was 26%. Nodal status is the most powerful predictor of outcome. Two patients with T4, NO disease are alive without evidence of disease beyond 4 years. Thirteen of the 17 patients (76%) undergoing reoperation after simple cholecystectomy for T2 or T3 tumors had residual disease. CONCLUSIONS: Patients with nodal metastasis beyond the pericholedochal nodes should not be considered for curative resection. Tumors staged T4, NO should be included with stage III disease, and resection should be considered. Re-resection of T2 or T3 tumors after simple cholecystectomy is likely to include residual disease and should thus provide the only chance for long-term survival. (34 Refs)

　

BACKGROUND: The surgical management of gallbladder carcinoma is controversial, especially as regards the indications for radical resection. The aim of this study was to evaluate the results of surgical treatment for gallbladder carcinoma with special reference to the extent of its histological spread.

METHODS: Eighty-six patients from 25 French centers underwent resection for cure and were included in this study. They comprised 65 women and 21 men (mean age 65 21 years). Resection included radical resection in 21 patients (partial hepatectomy, regional lymphadenectomy, and common bile duct resection) and simple cholecystectomy in 65.

RESULTS: There were 3 postoperative deaths (3.5%). The mean follow-up period was 25 24 months. The overall 5-year actuarial survival rate was 26%. The 5-year actuarial survival rate was 27% for patients who had radical resection. Eight patients with nodal metastasis had a 5-year survival rate of 0%, but the rate for 13 patients without such metastasis was 43% (P <0.05). For patients undergoing simple cholecystectomy, the 5-year actuarial survival rate was 44% for stage I disease, 22% for stage II, and 0% for stage III (P <0.05).

CONCLUSIONS: In patients with stage I gallbladder carcinoma, outcome is good after cholecystectomy only. In stages II to IV, radical resection should only be considered in the absence of regional lymph node metastasis. [C] 1998 by Excerpta Medica, Inc.

Carcinoma of the gallbladder has an extremely poor prognosis. It is a highly malignant tumor and is characterized by early metastasis and a rapid clinical course. Resectability for cure has been reported to be only 20% to 40%, and 5-year survival rates range from 5% to 40%. (1-4) Because of the high frequency of advanced stage at presentation, long-term survivors are mostly patients who underwent cholecystectomy with a diagnosis of cholelithiasis, and in whom an early carcinoma confined to the mucosa or muscularis is found incidentally. (5) After the development of a method of hepatobiliary imaging that facilitates the detection of gallbladder carcinoma at a resectable stage, (6) extended operations combining liver resection and wide lymph node dissection were recently introduced. As a result, the prognosis for gallbladder carcinoma has begun to improve in recent years. (7,8) However, its surgical management is controversial, especially for the indication of radical resection. From the replies to a questionnaire sent to 76 prominent North American surgeons, Gagner et al (9) concluded that surgeons advocated different procedures for each stage of the disease and that there was no consensus of opinion in this respect.

The purpose of the present study was to evaluate the effect of surgical resection on outcome and to identify the appropriate surgical procedures, with special reference to the extent of the spread of gallbladder carcinoma, indicated by histology.

PATIENTS AND METHODS
The French cooperative group AURC (Association Universitaire de Recherche en Chirurgie) started a multicenter retrospective study of patients operated for histologically proven gallbladder carcinoma in 1975. In order to obtain a follow-up period of at least 5 years for all the patients, and to exclude patients who underwent cholecystectomy through a laparoscopic approach, only patients operated through laparotomy from 1975 to 1986 were included in this study.

We reviewed the data obtained from the replies to a questionnaire sent to different institutions. In this review, we analyzed data relating to patients' gender and age, symptoms, diagnostic tests, surgical procedure, pathological findings, and survival. Clinicopathological data were collected for 274 cases of carcinoma of the gallbladder operated on in 25 French centers. Of these cases, 97 had palliative treatment (ie, external or internal biliary drainage, gastrojejunostomy or biopsy) and were excluded from the study. Resection was performed in the remaining 177 patients, of whom but 91 had tumors that were not resectable for cure.

The other 86 patients underwent potentially curative resection and form the basis of this study. Curative resection consisted of complete removal of locoregional tumor spread without macroscopic residual tumor and free margins. There were 65 women and 21 men, whose mean age was 65 21 years (range 42 to 93). The most common presenting symptom was pain, experienced by 76 of 86 patients (88%), followed by weight loss in 21 (24%) and jaundice in 16 (19%). Most patients had multiple symptoms. On physical examination, a palpable mass was noted in 11 patients (13%). Gallbladder carcinoma was diagnosed in 17 patients before surgery, in 35 during surgery, and in 34 after pathological examination of the surgical specimen. No patients were reoperated for resection after the incidental finding of gallbladder carcinoma. Seventy-nine percent of tumors were associated with gallstones. Eight patients underwent postoperative radiation therapy and 5 had chemotherapy.

All the resected specimens were examined. The depth of the lesion and the presence or absence of residual tumor were determined on multiple sections of the whole lesion. Dissected lymph nodes were examined histologically to detect metastatic foci on a single representative section per lymph node. The extent of tumor spread on resected specimens was assessed by the pathological tumor-node-metastasis (pTNM) system (* Table I*). The tumors were classified by stage using the criteria of the American Joint Committee on Cancer (AJCC). (10)

Survival was measured from the day of operation until death from cancer or unrelated diseases, or to the last day of follow-up. Actuarial survival was calculated and curves were constructed the Kaplan-Meier method, and differences in survival were evaluated by the Log-rank test.

RESULTS
Pathological Findings
The diagnosis of carcinoma was confirmed in all 86 cases. Eighty-two patients had adenocarcinoma, 2 had adenosquamous tumors, and 2 had squamous carcinoma.

Tumor was confined to the mucosa or muscularis in 36 cases (T1), extended into perimuscular connective tissue in 32 (T2), invaded beyond the serosa in 9 (T3), and extended more than 2 cm into the liver, or into two or more adjacent organs, in 9 (T4). Direct extension into the liver was noted in 18 cases, and adjacent organs were involved in 7. Nodes were histologically uninvolved in 23 patients and involved in 14. The other 49 patients underwent simple cholecystectomy, and no information concerning their lymph node status was available. Thirty-six of them presented with a T1 tumor and were classified as N0, and the remaining 13 patients with T2 tumors were arbitrarily classified as stage II.

According to the AJCC classification, 36 patients (42%) had stage I disease, 26 (30%) stage II, 15 (17%) stage III, and 9 (11%) stage IV.

Surgical Procedures
Sixty-five patients underwent simple cholecystectomy. Six of them had common bile duct exploration for cholelithiasis and biliary-enteric anastomosis was performed in 4 cases.

The remaining 21 patients underwent radical resection, which consisted of cholecystectomy, resection of liver segment V and of the lower part of segment IV, regional lymphadenectomy of the hepatoduodenal ligament with skeletonization of the portal vein and hepatic artery, and resection of the extrahepatic bile duct. Additional procedures were performed in 6 patients, and included extended right lobectomy (n = 1), pancreaticoduodenectomy (n = 1) and colonic resection (n = 4).

Operative procedures for the 86 patients according to the depth of carcinoma invasion are shown in * Table II*. Most patients with stage I or II disease underwent simple cholecystectomy, and only a few had hepatic and bile duct resection. Among patients in stage III and IV, the number undergoing resection was much larger.

Results of Surgery
There were 3 postoperative deaths (3.5%). They respectively occurred on day 1 due to myocardial infarction, on day 26 after pulmonary embolism, and on day 20, from progressive heart failure. Two of these deaths occurred after simple cholecystectomy, and 1, after radical resection (NS).

Five patients (6%) had postoperative complications including wound infection in 3 and biliary fistulae in 2 patients, 1 of whom required reoperation. There was no significant difference between the perioperative morbidity rate for patients who had simple cholecystectomy and those who had radical resection.

Survival
Mean follow-up was 25 24 months (range 1 to 160). In the total group of 86 patients, actuarial survival rates were 63% at 1 year, 43% at 3 years, and 26% at 5 years. Median survival was 33 months. Ten patients survived more than 5 years. By the end of follow-up, 36 patients had died of carcinoma, and 13, of unrelated causes.

(* Figure 1*) shows the survival rates for 65 patients treated by simple cholecystectomy, according to the stage of disease. Thirty-six of these patients had stage I disease. Mean age was 72 14 years, which was higher than the rest of the population study. Their actuarial survival rates were 86% at 1 year, 63% at 3 years, and 44% at 5 years. Median survival was 73 months. Only 7 patients (19%) died of gallbladder carcinoma: 1 (8%) had a pT1a and 6 (26%) had pT1b tumors (NS). Furthermore, 10 patients (28%) died from unrelated disease. Among the patients with stage I disease, no significant difference in survival was observed between the 13 with pT1a tumor confined to the submucosa (45% at 5 years) and the 23 with a pT1b tumor invading the muscularis (44% at 5 years). Patients with stage II disease exhibited actuarial survival rates of 61% at 1 year, 30% at 3 years (P <0.02 versus stage I disease) and 22% at 5 years. The 5-year actuarial survival rate for the 6 patients with stage III disease was 0% (P <0.05 versus stage I disease). There was no significant difference in survival between stage II and stage III patients.

The actuarial survival of the 21 patients treated by radical resection was 33% at 1 year and 27% at 3 and 5 years. Median survival was 8 months. Two patients survived more than 5 years without any evidence of tumor recurrence. Both had stage II disease. At the end of follow-up, 13 patients (61%) died of the recurrence of carcinoma and 1 of an unrelated cause * Figure 2* shows the actuarial survival rates according to the stage of disease. The rate for patients with stage II disease was higher than for those with stage III or IV disease. However, because of the small number of patients in stages II to IV, the differences was not significant. Eight of the 21 resected patients had nodal metastasis and died of carcinoma within the first postoperative year. The 5-year survival rate for the remaining 13 patients without node metastasis was 43% (* Figure 3*, P <0.05 versus patients with nodal metastasis).

Of the 15 patients with stage III disease, 6 underwent simple cholecystectomy and 9, radical resection. The 5-year survival rates were 22% after treatment with radical resection and 0% after simple cholecystectomy. However, the difference between survival after each of these surgical procedures was not significant.

COMMENTS
Gallbladder carcinoma carries a poor prognosis, and the only chance for cure lies in early detection and complete surgical resection. However, the extent of resection required remains controversial, especially in advanced stages of the disease. One of the reasons is that no uniform classification of staging system exists for gallbladder carcinoma, so that comparison of data is impossible. Furthermore, in some studies, the results of simple cholecystectomy and more extended resection were not compared according to tumor stage. This led some authors to reach the apparently paradoxical conclusion that survival did not improve after extended resection versus simple cholecystectomy, (11-13) extended operations being reserved for the most advanced cases in the series concerned. Finally, in most of studies, the term "extended resections" included various surgical procedures ranging from limited hepatic resection to right lobectomy and common bile duct resection, thus making it difficult to assess the results. (2,7,13) The present investigation, in which the long-term results of two main surgical procedures were compared according to the stage of disease, demonstrated that simple cholecystectomy should be reserved for stage I gallbladder carcinoma only. In more advanced stages, radical resection should be considered, except when lymph nodes are present in the hepatoduodenal ligament.

Most surgeons agree that stage I disease does not require any operation other that simple cholecystectomy, which gives an overall 80% 5-year survival rate. (3-5,14) In our series, only a 44% 5-year survival was observed for patients with stage I disease. However, these patients were old, and among them, 28% died from unrelated disease, with only 19% dying from gallbladder carcinoma. Although very good long-term results are obtained for pT1a tumors, the results reported after simple cholecystectomy for pT1b tumors are not so good. (3) In our series, the 5-year survival rate of the 23 patients with pT1b tumor was only 44% and 6 of them died of gallbladder carcinoma. Some authors (3,4,15) have advocated more extended resection for patients with pT1b tumors, which resulted in a 66 to 100% 5-year survival. This aggressive approach is supported by the 28% lymphatic or venous invasion rate reported by Ouchi et al (3) for pT1b tumors.

We, as well as others (5,11,16,17) reported poor results after simple cholecystectomy for stage II disease (T2 N0 M0), and 5-year survival rates ranging from 10% to 22%. After radical resection, the 5-year survival rates was as high as 69% for one series of patients with stage II disease. (8) Shirai et al (181) demonstrated that 5-year survival for patients with stage II disease was significantly better after radical resection than after simple cholecystectomy (90% versus 40%). In our study, 2 of the 3 patients with stage II disease who underwent radical resection are still alive and displayed no evidence of recurrence at 5 years of follow-up. Therefore, as demonstrated both in previous reports (3,8,18,19) and in the present study, radical resection, including hepatic resection and locoregional lymph node dissection, should be recommended for patients with stage II gallbladder carcinoma. Furthermore, for patients having undergone cholecystectomy with any incidental finding of gallbladder cancer on a resected specimen, radical reoperation should be carried out for pT2 carcinoma. This is supported by the 46% incidence of lymph node metastasis reported in T2 patients. (8)

For stage III or IV disease, simple cholecystectomy cannot be considered a curative procedure because of tumor spreading outside the gallbladder. Thus, among our patients with stage III disease, all those treated by simple cholecystectomy died of carcinoma within 4 years whereas radical resection permitted an increase in survival, even though it was not significant. Consequently, only radical resection should be considered for patients with stage III and IV disease. Morbidity and mortality after major liver resection has decreased in recently reported series, even among the older population. (20) This justifies more aggressive management of advanced gallbladder cancer. In the present study, operative morbidity and mortality were similar after simple cholecystectomy and radical resection.

According to most authors, radical resection includes subsegmental resection of the hepatic bed of the gallbladder from segments IV and V with both macroscopic and microscopic margins. (7,8,13,19) However, the extent of hepatic resection remains controversial. Shirai et al (21) demonstrated that the extent of microscopic angiolymphatic portal tract invasion correlated well with the gross depth of direct invasion within the liver. However, it has never been reported that more extended hepatic resection improves the prognosis and survival rates, even in patients with a tumor extending more than 2 cm into the liver.

The need for associated common bile duct resection is still a matter of debate. We and others systematically performed common bile duct resection during radical resection. (7,13) Although no improvement in survival has been reported as a result of bile duct resection, it seems to us that it facilitates complete node removal from the hepatoduodenal ligament. Furthermore, when a tumor is located in the neck of the gallbladder, there is probably a high risk of common bile duct involvement due to direct intraductal extension or external invasion of the hepatoduodenal ligament.

Although few authors have reported long-term survivors among patients with involved nodes, (7,15,19) our study, as well as several other investigations, demonstrated that short-term survival is the rule, even after radical resection. (4,11,22-24) Thus, none of our 8 patients with involved nodes was alive 1 year after radical resection. Nodal status is the most powerful predictor of outcome in patients with gallbladder carcinoma. (8) It has been demonstrated (25) that tumor spreading within lymph nodes descended around the bile duct, and flowed into the cystic node, the pericholedochal nodes, and the lymph nodes posterior to the head and neck of the pancreas. Radical resection includes resection of the cystic and pericholedocal nodes, but leaves the retropancreatic nodes in position. For this reason, some surgeons (19) advocated more extended resection, including pancreato-duodenectomy. One of our patients with nodal metastasis around the head of the pancreas underwent this procedure but died of peritoneal dissemination 5 months later. Poor results have been reported with this procedure, with median survival ranging from 3 to 19 months after surgery. (7,15)

In conclusion, only patients with stage I and pT1a gallbladder carcinoma can be treated for cure by simple cholecystectomy. In more advanced stages, radical resection should be performed. However, despite the absence of any increase in terms of morbidity or mortality, only patients with no lymph node involvement in the hepatoduodenal ligament can benefit from such resection.

參考資料：

• Surgery 1996;120(5):816-21

• Ann Surg 1996;224(5):639-46

• Am J Surg. 1998;175:118-122. 

　

上一篇

下一篇