Click here to read this article as a CME course
The Oncologist, Vol. 12, No. 6, 654-663, June 2007; doi:10.1634/theoncologist.12-6-654
© 2007 AlphaMed Press
Extended Resection for Pancreatic Adenocarcinoma
Srinevas K. Reddy,
Douglas S. Tyler,
Theodore N. Pappas,
Bryan M. Clary
Duke University Medical Center, Durham, North Carolina, USA
Key Words. Pancreatic neoplasms • Pancreatic ductal carcinoma • Pancreaticoduodenectomy • Pancreatectomy Lymph node excision • Portal vein
Correspondence: Bryan Clary, M.D., Duke University Medical Center, Box 3247, Durham, North Carolina 27710, USA. Telephone: 919-684-6553; Fax: 919-681-7508; e-mail: clary001{at}mc.duke.edu
Received November 1, 2006;
accepted for publication March 1, 2007.
 |
Learning Objectives
|
|---|
After completing this course, the reader will be able to:- Discuss the evidence in favor of and against the role of surgical resection in the management of pancreatic adenocarcinoma.
- Explain how the components of an extended pancreatectomy, including total pancreatectomy, extended lymph node dissection, and major intestinal arterial resection, do not improve survival above what is provided after pancreaticoduodenectomy and are associated with substantial morbidity.
- Discuss how portal/mesenteric vein resection to achieve negative resection margins does not add to the morbidity of pancreaticoduodenectomy, does not worsen survival after resection, and is most often performed because of tumor location.
Access and take the CME test online and receive 1 AMA PRA Category 1 CreditTM at CME.TheOncologist.com
|
ABSTRACT
|
|---|
Adenocarcinoma of the pancreas presents a number of therapeutic challenges. Given the poor long-term outcomes after pancreaticoduodenectomy (PD), many surgeons have sought to improve survival via a radical or "extended" pancreatectomy which may include (a) total pancreatectomy (TP), (b) extended lymph node dissection (ELND), and (c) portal/mesenteric vascular resections. These themes of "extended" resection are addressed in this review.
TP should not be performed for most cases of adenocarcinoma of the pancreatic head because of the nominal incidence of lymph node involvement along the body and tail of the pancreas, the scarcity of multicentric disease, and the better management of pancreatic leaks after PD. Most studies show no difference in long-term survival and demonstrate greater postoperative morbidity after TP than after PD. Performing ELND in addition to PD is not worthwhile because most studies do not demonstrate any long-term benefits from ELND and the circumferential dissection around the mesenteric vessels required to harvest distant lymph nodes increases postoperative morbidity. Major arterial resection increases postoperative morbidity after PD and worsens long-term survival as the need for arterial resection to achieve negative resection margins indicates more aggressive disease. In contrast, portal and/or mesenteric venous resection does not increase the morbidity after PD or impact long-term survival as venous resection is often performed because of tumor location and not extent of disease. The disappointing experience with extended resections underscores the need for better adjuvant systemic strategies and the interdisciplinary care of patients with pancreatic adenocarcinoma.
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION
|
|---|
Adenocarcinoma of the pancreas presents a number of therapeutic challenges. On presentation, 80%–90% of patients have locally advanced or disseminated disease, precluding "curative" resection [1, 2]. The 5-year survival rate after resection is 10%–20% in larger series [3, 4]. The illusion of "localized" disease is betrayed by the high frequency of regional (46%–83%) [3, 5] and distal (15%–30%) [6, 7] nodal involvement, positive pancreatic and/or retroperitoneal margins following resection (20%–36%) [3, 5, 8], and the presence of distant micrometastases that frequently become clinically apparent shortly after resection. The high rate of positive resection margins reflects the frequency of tumor extension beyond the pancreatic capsule to adjacent retroperitoneal soft tissue.
Resection of adenocarcinoma of the pancreatic head typically requires a pancreaticoduodenectomy (PD). Standard PD usually includes resection of the gallbladder, lower common bile duct, pancreatic head, duodenum, and distal stomach (variably), and extirpation of soft tissue including lymph nodes associated with the lower bile duct, pylorus, and anterior and posterior surfaces of the pancreatic head. PD can be performed safely with a mortality of 0.7%–3% and morbidity of 36%–41% in experienced centers [3, 8–10]. Given the poor outcomes historically associated with PD, many surgeons have sought to improve survival using more radical excisions. Fortner [11, 12] pioneered this strategy through an en bloc resection of the pancreas with the common bile duct, gallbladder, distal stomach, duodenum, spleen, and transverse mesocolon with middle colic vessels, and a retroperitoneal node dissection from the diaphragm to the origin of the inferior mesenteric artery and laterally from the renal hilum to the aorta. The portal vein, common hepatic artery, celiac axis, and superior mesenteric artery were skeletonized. This resection was done without (type 0) or with (type 1) portal vein resection and occasionally with hepatic or superior mesenteric artery (SMA) (type 2) resection for extensive cancers. Results from 56 cases (30 total pancreatectomy and 26 subtotal pancreatectomy) revealed a postoperative mortality rate of 8.9% and morbidity rate of 82%. These results were remarkable at that time given the prevalence of 
25% mortality rates after PD. The actual 5-year survival rate after this "regional pancreatectomy" was 14%; patients with tumors <2.5 cm had a greater 5-year survival rate (33% versus 12%) [13]. Contemporary components of a radical or "extended" pancreatectomy may include (a) total pancreatectomy (TP), (b) wider fields of lymph node/soft tissue resection, and (c) portal/mesenteric vascular resections. These themes of "extended" resection as pertaining to ductal adenocarcinoma of the pancreas are addressed. Malignant transformation associated with mucinous tumors or pancreatic endocrine tumors is not addressed. Each topic focuses on prospective randomized trials. When prospective randomized trials do not exist, large retrospective case series are reviewed. Biases in patient selection and variations in surgical skill, disease stage, and the use of adjuvant chemoradiotherapies, and the lack of standardization in preoperative selection and postoperative care are all key limitations of comparisons of patients within and across these studies.
|
RATIONALE FOR RESECTION
|
|---|
Despite widespread acceptance, only two prospective randomized trials evaluated resection of pancreatic cancer. Imamura et al. [14] randomized 42 patients with tumor invasion of the serosal or retroperitoneal pancreatic surfaces or with extension into the intrapancreatic portal vein without complete obstruction to resection without adjuvant chemoradiotherapy (n = 20) or radiochemotherapy with or without digestive/biliary bypass (n = 22). Therapy consisted of 5,040 cGy in 28 fractions with continuous 5-fluorouracil (5-FU) at 200 mg/m2 per day during radiotherapy followed by weekly 5-FU at 500 mg/m2. Exclusion criteria included intraoperative discovery of SMA or common hepatic artery tumor involvement, liver or para-aortic nodal metastases, peritoneal seeding, and tumor spread to adjacent organs other than the transverse mesocolon, duodenum, or common bile duct. These tumors corresponded to T3N0M0 (stage IIA) and T3N1M0 (stage IIB) by the American Joint Committee on Cancer staging system (Sixth Edition) [15]. This trial was prematurely terminated because of slow accrual and financial difficulty. Of resected patients, 15 underwent PD and four underwent distal pancreatectomy. After a mean follow-up of 13 and 10 months in the resection and radiochemotherapy groups, respectively, resected patients had a longer mean survival time (17.8 versus 11 months) and greater 1-year survival rate (65.2% versus 30%), p < .05. Resection was the only independent predictor of survival (p = .04).
Lygidakis et al. [16] randomized 56 patients with carcinoma of the head of the pancreas 5 cm in diameter with extension to the neck and with regional vascular involvement of the superior mesenteric or portal vein to en bloc splenopancreaticoduodenectomy (n = 27) or gastric biliary bypass (n = 29). Resection included splenectomy, TP, duodenectomy, distal gastrectomy, and portal vein resection. Patients who underwent standard PD, had distant metastases, and did not have evidence of vascular invasion during dissection were excluded. All patients were subsequently treated with maintenance chemoimmunotherapy for 5 years after resection via an SMA catheter. Drug regimens included a 5-day course of gemcitabine, carboplatin, mitoxantrone, mitomycin-C, and 5-FU and a 10-day course of interleukin-2. Nine (30%) resected patients had multicentric disease. The overall 5-year survival rate for resected patients was 18.5%. The 1-year overall survival rate for bypassed patients was 45% and no patient survived 2 years after surgery (p = .0001). Importantly, the advanced disease regarded for inclusion that would not be cured by standard PD limits the application of these results to all patients eligible for resection.
To critique these prospective trials and retrospective analyses of large case series, Gudjonsson [17–19] argued that long-term survival after pancreatic adenocarcinoma resection has been grossly exaggerated. Repeated publications from the same institution and migration of patients to several treatment centers have resulted in repetition of the number of resections and long-term survivors. Reduction of the subset on which calculations are based also inflates survival outcomes. Instead of calculating survival relative to all patients, subsets of resected patients, patients who did not suffer postoperative mortality, and/or patients with a particular tumor stage are used as the denominator. Hence, survival rates increase not because of a growing number of long-term survivors but because of smaller subsets of patients included in the study population. Actuarial survival rates exaggerate true survival because patients lost to follow-up are censored and consequently ignored in Kaplan-Meier survival analyses. There also have been several reports of 5-year nonresected survivors, discrediting the notion that resection is required for long-term survival. After correcting for these biases, Gudjonsson estimated the true number of 5-year resected survivors to be 300–350 patients—resulting in an actual survival rate <0.4%. Thus, pancreatic resections have a minimal impact on survival and are not worthwhile given limited resources.
Results of prospective randomized trials suggest that, in the absence of disseminated disease, resection of T3 pancreatic cancer is associated with a longer overall survival duration compared with bypass alone in settings with and without adjuvant chemotherapy. However, there have been no prospective randomized trials evaluating resection for smaller tumors with no apparent portal/mesenteric venous invasion. Also, survival data from large retrospective series likely exaggerate the number of true long-term survivors after resection.
|
TOTAL PANCREATECTOMY (TP)
|
|---|
Arguments for TP include: (a) it eliminates potential leakage at the pancreaticojejunal anastomosis and subsequent sepsis, (b) pancreatic cancer is multicentric in up to 30% of cases and a standard PD or distal pancreatectomy would not resect these multiple sites, (c) TP reaps the oncological benefits of more extensive lymphadenectomy around the entire pancreas, and (d) TP lessens the incidence of a positive margin at the site of glandular transection. Disadvantages to TP include obligate diabetes mellitus with frequent severe hypoglycemic episodes, impaired immunologic response resulting from splenectomy, marginal ulceration at the gastrojejunostomy anastomosis resulting from the loss of pancreatic bicarbonate secretion, and an increase in hepatic steatosis likely as a result of malabsorption of fatty acids [20]. While no prospective trials have been performed comparing TP with PD for pancreatic cancer, several retrospective case series detailing results of TP have been published (Table 1) [20–25]. However, most studies do not account for differences in lymph node status, tumor differentiation, stage, or other factors shown to predict survival. Most of these series consist of procedures performed in an older era with higher mortality and morbidity after pancreatic resections compared with today. Both Karpoff et al. [21] and Ihse et al. [22] reported worse survival after TP than after a more limited pancreatectomy, whereas Sarr et al. [20] did not note any survival differences. Only Ihse et al. [22] accounted for differences in stage and lymph node status between groups of patients that underwent TP and subtotal pancreatectomy. They demonstrated an independent survival benefit for subtotal pancreatectomy versus TP (relative risk, 0.63; 95% confidence interval, 0.40–0.99) with less postoperative mortality (3% versus 27%). The only report that showed a survival benefit from TP was the Veterans Affairs study [25]. This benefit was seen for all patients (Table 1). When broken down by stage, there was a trend for longer survival with TP for stage I and II disease (mean, 25.7 versus 14.9 months; p = .06) and no difference with stage III disease (mean, 10.0 versus 8.9 months). They also reported no differences in postoperative morbidity or mortality between TP and PD.
TP should not be recommended for most cases of pancreatic adenocarcinoma. Many of the advantages noted for TP are no longer evident. Improvements in management of pancreatic leaks and fistulas have drastically reduced the incidence of sepsis and death [9, 10, 26]. More accurate determination of true multicentric disease by gene mutations rather than histological analyses has revealed the incidence to be much less frequent than previously believed [27]. Moreover, studies have shown that nodes along the body and tail of the pancreas are less likely to be involved in adenocarcinoma of the head of the pancreas than retroperitoneal nodes [28]. Finally, most studies reveal no survival benefit to TP above that provided by limited pancreatectomy and report morbidity and mortality rates that exceed those of PD or distal pancreatectomy. Despite this, there are some rare situations in which TP should be considered [20]. These include a soft pancreas such that a subsequent pancreatoenterostomy is likely to leak, a continued positive margin on frozen section as resection is carried into the pancreatic tail, the presence of papillomatous changes or dysplasia throughout the pancreatic duct, and in patients with a well-documented family history of multicentric disease.
|
EXTENDED LYMPH NODE DISSECTION
|
|---|
Because up to 30% of patients have retroperitoneal nodal disease that would not be resected with standard PD, some surgeons hypothesized that extended lymph node dissection (ELND) with PD may have survival benefits for pancreatic adenocarcinoma. Japanese retrospective studies have noted higher long-term survival rates after ELND than after standard lymphadenectomy (SL) [29–31]. As these benefits were relative to historical data or nonmatched controls, a number of prospective trials have been performed to evaluate the benefits of PD with ELND compared with SL (Table 2).
View this table:
[in this window]
[in a new window]
|
Table 2. Prospective trials of pancreaticoduodenectomy with extended lymph node dissection (ELND) versus standard lymph node dissection
|
|
A German group conducted a nonrandomized prospective study comparing results after ELND and SL in addition to PD for pancreatic cancer [32–35]. ELND (n = 70) was the routine procedure and SL (n = 21) was performed in patients with advanced age or comorbidity. SL in this study was more extensive than in other studies—all lymph nodes at the hepatoduodenal ligament, the celiac trunk including the first 3 cm of the splenic artery, and the ventral surface of both the renal veins and the vena cava were removed. ELND included additional removal of all tissue along the left side of the SMA and along the aorta between the inferior mesenteric artery (IMA) and the diaphragm. More nodes were harvested after ELND (median, 24 versus 14). Twelve percent of ELND patients had positive aortic and/or juxtarenal lymph nodes. Positive para-aortic lymph nodes were found in four patients; all had positive regional lymph nodes. One patient did have a positive gastric lymph node with negative regional and para-aortic nodes. Patients in each group received a pylorus-preserving PD or a distal gastrectomy. No patient received adjuvant chemotherapy. While there were no differences in postoperative mortality or morbidity, 76% of ELND patients experienced some gastrointestinal motility disorder (most often diarrhea) during the first postoperative year (Table 2). There was no survival benefit after ELND. Quality of life (QOL) analyses revealed conflicting results; ELND patients had higher physical functioning scores but worse diarrhea 24 months after PD. Weaknesses of this study include its nonrandomized design, low number of SL patients, lack of adjuvant chemotherapy among all patients, and the inconsistent use of enteroenteric anastomosis.
The largest prospective randomized trial comparing ELND with SL has been conducted by the Johns Hopkins group [9, 36, 37]. That trial included patients with any periampullary carcinoma (n = 294), of which 167 had pancreas adenocarcinoma. Pylorus preservation was preferred for patients randomized to SL and a 30%–40% distal gastrectomy was performed for ELND patients. SL was limited because the entire hepatoduodenal ligament and region surrounding the common hepatic artery were not harvested. ELND was also more restricted in that the superior margin of the dissection was the portal vein (and not the diaphragm or celiac axis) and the left lateral border of resection was the aorta (and not the left renal hilum). In addition, lymphatic and neural tissues were cleared around approximately one half instead of the entire circumference of the SMA. Despite these limitations, the median number of harvested nodes (28.5) was comparable with those of other studies and was greater than with SL [17]. Thirteen of 83 (16%) patients with pancreatic cancer who underwent ELND had positive distant retroperitoneal nodes; all had positive regional nodes. While there was no difference in post-operative mortality, morbidity was greater in ELND patients receiving a distal gastrectomy (Table 2). Whereas there was a trend toward a higher 5-year survival rate among ELND patients, the authors note that this is likely a result of the higher incidence of R1 resections among SL patients (21% versus 5%) and not a result of the ELND itself. There were no differences in survival between ELND and SL for node-positive (median, 18 versus 19 months) or node-negative (median, 44 versus 43 months) patients. ELND was not independently associated with overall survival. There were no differences in patients' degrees of perceived weight loss, pancreatic enzyme usage, or number of bowel movements per day 6 months after PD. At a mean of 2 years after PD, there were no differences in QOL scores [38]. Strengths of this trial include the standardization of operation and pathologic examination and the fact that the majority of patients underwent adjuvant chemotherapy, similar to how most patients are treated in the current era. However, this study has several limitations, including a lesser ELND than in other studies, inclusion of all periampullary carcinomas, inconsistent enteroenteric anastomosis among SL and ELND patients, and the absence of disease relapse data.
The other prospective randomized trial based in the U.S. is from the Mayo Clinic [10]. Seventy-two patients with pancreatic cancer were randomized into SL and ELND groups. ELND in that study was extensive, because retroperitoneal soft tissue was removed from both renal hila, the hepatoduodenal ligament was skeletonized up to the hepatic hilum, para-aortic nodes were harvested from the celiac axis to the IMA, and circumferential dissection of the hepatic artery, celiac axis, and SMA was performed. This resulted in a larger lymph node harvest (median, 36 versus 15) and greater likelihood of one unit of blood transfusion (23% versus 5%) for ELND patients. Ten of 34 ELND patients (29%) had positive second-echelon nodes and there were two patients with positive N2 nodes and negative N1 nodes (skip nodal metastasis). All patients in this trial underwent distal gastrectomy. While mortality was similar, there were some trends toward worse morbidity among ELND patients (Table 2). The median length of hospital stay was similar (both 11 days). There was no significant difference in overall patient survival, which was maintained when subsets of patients with positive (median, 17 versus 19 months) or negative (median, 42 months each) lymph nodes were examined. While there were more grade IV neoplasms in the ELND group (15% versus 0%), the extent of lymph node dissection was not independently associated with overall survival. Overall and subset QOL scores did not differ. At 4 months after PD, 53% and 42% of patients who underwent ELND had little control of their bowels and reported diarrhea, respectively, compared with 9% and 8% for SL patients. After 8 months, diarrhea and lack of bowel control in ELND patients decreased to that observed in SL patients. The authors hypothesized that disruption of the celiac and SMA nerve plexuses resulting from circumferential dissection was the likely cause of the diarrhea. Strengths of this study are standardization of operations by one surgeon, uniform specimen examination, exclusion of patients with diagnoses other than pancreatic cancer, and the fact that >75% of the patients underwent adjuvant chemotherapy. Weaknesses include the low patient number in each group and the absence of disease relapse data.
The first randomized prospective trial on the role of ELND was a multicenter study based in Italy [39]. Eighty-one patients with pancreatic cancer were randomized, 40 to SL and 41 to ELND. The choice of pylorus preservation or distal gastrectomy was left to the operating surgeon and no patient received adjuvant chemotherapy. Nineteen patients (allocated equally to ELND and SL) received intraoperative radiation therapy. ELND in this study was similar to that in the Mayo Clinic study; a key difference was that the superior margin extended up to the diaphragmatic hiatus. This resulted in a median retrieval of 19.8 nodes, versus 13.3 nodes for SL, without greater operative time, blood product transfusions, morbidity (Table 2), or length of hospital stay (mean, 22.7 versus 19.3 days). While there was no difference in overall survival between ELND and SL for all patients, an a posteriori analysis not planned at the time of study design revealed node-positive patients to have better overall survival after ELND (median, 18 versus 11 months; p < .05). However, the extent of lymph node dissection was not independently associated with survival. These authors concluded that ELND should be considered in selected patients as it was not associated with greater morbidity and because it may be associated with longer survival among node-positive patients. Criticisms of this trial include the low overall survival rate after PD (22.2% at 2 years), small patient numbers, absence of adjuvant therapy (rare for most patients in the current era), relatively low number of lymph nodes harvested, discretionary use of pylorus preservation, small difference in number of harvested nodes between ELND and SL, the a posteriori nature of the analysis that revealed a survival difference for node-positive patients, and the lack of difference in survival after ELND for patients with micrometastatic lymph node disease versus grossly positive nodes.
Nimura et al. [40] have published the only Japanese randomized prospective clinical trial evaluating radical PD with ELND versus SL. One hundred one patients, none of whom received adjuvant chemotherapy, were analyzed. ELND was performed in a manner similar to that used in the Italian study. The mean node retrieval (40 versus 13) was higher with ELND. The incidence of postoperative diarrhea was higher after ELND (Table 2). There were no differences in postoperative mortality, mean length of hospital stay (42 versus 44 days), and QOL 1 year after PD. The 3-year overall and disease-free survival rates (12.0% versus 6.7%) did not differ after ELND compared with SL. Survival also did not differ for node-positive and node-negative patients.
As a critique to these trials, Pawlik et al. [41] created a model to estimate the sample size needed to detect a survival benefit for patients undergoing an ELND. They assumed that the percentage of patients who would benefit from ELND was equal to the percentage who underwent R0 resection multiplied by the percentage who had positive second-order nodes multiplied by the percentage who had no visceral metastatic disease. Applying this model to their experience of 158 patients who underwent PD and separate ELND (which they defined to include nodes adjacent to the proximal hepatic artery and anterior to the great vessels), only 0.3% of patients would have the opportunity to benefit from ELND because 10.2% of patients had positive second-order nodes, 47.1% of patients with positive second-order nodes underwent an R0 resection, and 5.9% had no evidence of occult visceral metastatic disease (defined by the lack of disease recurrence after resection). Using their 5-year actuarial overall survival rate of 27.30%, they deduced that ELND would only increase survival to 27.52% assuming that all patients who would benefit from ELND had a 100% long-term survival rate after resection. To detect this small survival difference, a prospective study would require an unfeasible 202,000 study patients in each arm. Importantly, this model does not allow for any survival benefit from adjuvant chemotherapy for a microscopically positive (R1) but grossly negative resection. Moreover, this model excludes those patients who benefit with a long disease-free survival interval after resection, because five of 17 patients had disease-free survival times >1 year and overall survival times in the range of 23.1–76.3 months after resection were noted in this series.
Based on these prospective trials, ELND with PD is not worthwhile for patients with pancreatic adenocarcinoma. No study has shown an improvement in long-term survival with ELND for all patients. The Italian study did show some benefit for node-positive patients; however, that conclusion was based on an a posteriori analysis that was not intended in that trial. Moreover, that benefit has been refuted by all other studies. ELND increases morbidity as a result of, in part, the circumferential dissection around the mesenteric vessels. The paucity of patients that would theoretically benefit from an ELND does not justify this greater morbidity. These studies also revealed that ELND will not offer more accurate staging information because patients rarely have skip nodal metastasis.
|
PORTAL/MESENTERIC VASCULAR RESECTION
|
|---|
There are two main rationales for vein resection (VnR) of the portal vein (PV), superior mesenteric vein (SMV), or superior mesenteric-portal vein confluence (SMPV) as part of pancreatic cancer resection. First, VnR is performed in order to achieve a negative resection margin because the tumor invades the vessel or inflammatory adhesions around the tumor preclude adequate separation from the vein. Second, VnR is performed to achieve the benefits of an "extended pancreatectomy" via clearance of second-order lymph nodes and surrounding connective tissue. The following discussion pertains to the former reason. There are no prospective randomized trials comparing pancreatectomy with and without VnR. Most retrospective studies demonstrate that while VnR is associated with longer operative times and larger blood loss, the morbidity and mortality of pancreatectomy with VnR is similar to those of pancreatectomy alone (Table 3) [42–54]. There was a high incidence of venous invasion (50%–82%) among patients who underwent VnR. No study demonstrated that VnR was associated with a higher R0 resection rate—in fact, positive resection margins were more likely after VnR in three studies. This is likely a result of larger tumor size and tumor location as opposed to the VnR itself. In a multivariate analysis, Tseng et al. [42] showed that tumor size and lack of neoadjuvant therapy, and not the need for VnR, predicted positive retroperitoneal margin status. Despite great diversity in overall survival, most studies also showed that the addition of VnR was not associated with longer overall survival, with three studies noting shorter survival after VnR [47, 51, 54].
View this table:
[in this window]
[in a new window]
|
Table 3. Results of resection of pancreatic adenocarcinoma with portal and/or mesenteric vein resection (VnR) as compared with pancreatectomy without VnR
|
|
In a comprehensive review of synchronous PV/SMV resection for pancreatic cancer, Siriwardana and Siriwardena [55] identified 1,646 nonduplicated patients from 52 studies. This was 26% of the total number of patients undergoing pancreatic cancer resection. Seventy-eight percent underwent PD, 24% underwent TP, and 2% each underwent subtotal pancreatectomy and distal pancreatectomy. Seventy-three percent underwent circumferential VnR; 11% had lateral wedge resections. Primary anastomosis was performed in 89%, 10% had an autologous vein graft, and 2% had a synthetic vein graft. The median morbidity and mortality rates after VnR were 42% (9%–78%) and 5.9% (0%–33%). The mortality rate for all patients was 5.2%. VnR did not add to mortality or morbidity. PV invasion was detected in 63% of patients, with 40% of patients having a positive resection margin (0%–85%). For 917 patients with available data, the median overall survival time was 13 months and the 5-year survival rate was 5.8%. Pooling results from 23 studies providing outcome data on pancreatectomy with and without VnR demonstrated no significant difference in the 5-year survival rate.
These studies show that resection of the SMV, PV, and/or SMPV confluence does not add to the morbidity or mortality of pancreatectomy and may be successfully performed to achieve a margin-negative resection in selected patients. While VnR itself does not predict a positive resection margin, the need for a VnR indicates that an R0 resection will be less likely because of tumor size and location. Whereas VnR does not diminish long-term survival after pancreatectomy, resection of portal/mesenteric vein invaded by tumor does not prolong survival.
Compared with VnR, resection of the mesenteric, celiac, or hepatic arteries is more rarely performed. Of 1,646 patients who underwent PV/SMV resection, 16% had adjacent arterial resection (common hepatic, SMA, and celiac axis were most common) [55]. In their series of 141 patients who underwent vascular resection with PD, Tseng et al. [42] reported that only 17 (12%) underwent hepatic arterial resection. Similarly, Nakao et al. [54] performed 14 arterial resections among 201 total cases of curative pancreatectomy with vascular resection. Postoperative mortality was higher among patients who underwent arterial resection (36%) compared with VnR (3%) and no vascular resection (1%). Long-term survival was lower after combined PV and arterial resection than after PV resection alone (2-year survival rate, 0% versus 17%; p = .02) and was similar to that of unresected patients. The authors argued that this low survival rate after combined arterial and venous resection was a result of advanced tumor stage and the high incidence of positive peripancreatic resection margins. Because of the high associated postoperative mortality, lack of survival benefit, and difficulty in achieving negative resection margins, they concluded that indications for arterial resection with pancreatectomy are limited.
|
CONCLUSION
|
|---|
From the limited randomized data available, it appears that resection of pancreatic cancer is essential for long-term survival. However, reported survival benefits are likely exaggerated. Components of an "extended pancreatectomy," including TP, ELND, and major arterial resection, do not improve survival above what is provided after standard PD and are likely associated with greater morbidity. These dismal survival rates suggest that the presence of disease outside the region encompassed by a standard PD indicates an aggressive tumor with widespread distant micrometastases at the time of presentation. Therefore, an extended resection to remove all regional disease should be viewed with similar circumspect as resection of gross distant metastatic disease. The exception to this is portal/mesenteric VnR, which has no added morbidity to that after a standard PD and is most often performed to achieve negative resection margins because of tumor location and not aggressive tumor biology. The disappointing experience with extended resections underscores the need for better adjuvant systemic strategies and interdisciplinary patient care.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
|
|---|
The authors indicate no potential conflicts of interest.
|
REFERENCES
|
|---|
- Sener SF, Fremgen A, Menck HR et al. Pancreatic cancer: A report of treatment and survival trends for 100,313 patients diagnosed from 1985–1995 using the National Cancer Database. J Am Coll Surg 1999;189:1–7.[CrossRef][Medline]
- Bramhall SR, Allum WH, Jones AG et al. Treatment and survival in 13,560 patients with pancreatic cancer, and incidence of the disease, in the West Midlands: An epidemiological study. Br J Surg 1995;82:111–115.[Medline]
- Cameron JL, Riall TS, Coleman J et al. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006;244:10–15.[CrossRef][Medline]
- Fernandez-del Castillo C, Rattner DW, Warshaw AL. Standards for pancreatic resection in the 1990s. Arch Surg 1995;130:295–299; discussion 299–300.[Abstract/Free Full Text]
- Neoptolemos JP, Stocken DD, Dunn JA et al. Influence of resection margins on survival for patients with pancreatic cancer treated by adjuvant chemoradiation and/or chemotherapy in the ESPAC-1 randomized controlled trial. Ann Surg 2001;234:758–768.[CrossRef][Medline]
- Riall TS, Cameron JL, Lillemoe KD et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma–part 3: Update on 5-year survival. J Gastrointest Surg 2005;9:1191–1206.[CrossRef][Medline]
- Farnell MB, Pearson RK, Sarr MG et al. A prospective randomized trial comparing standard pancreatoduodenectomy with pancreatoduodenectomy with extended lymphadenectomy in resectable pancreatic head adenocarcinoma. Surgery 2005;138:618–630.[CrossRef][Medline]
- Brennan MF, Kattan MW, Klimstra D et al. Prognostic nomogram for patients undergoing resection for adenocarcinoma of the pancreas. Ann Surg 2004;240:293–298.[Medline]
- Gouma DJ, van Geenen RC, van Gulik TM et al. Rates of complications and death after pancreaticoduodenectomy: Risk factors and the impact of hospital volume. Ann Surg 2000;232:786–795.[CrossRef][Medline]
- Buchler MW, Wagner M, Schmied BM et al. Changes in morbidity after pancreatic resection: Towards the end of completion pancreatectomy. Arch Surg 2003;138:1310–1314; discussion 1315.[Abstract/Free Full Text]
- Fortner JG. Regional resection of cancer of the pancreas: A new surgical approach. Surgery 1973;73:307–320.[Medline]
- Fortner JG, Kim DK, Cubilla A et al. Regional pancreatectomy: En bloc pancreatic, portal vein and lymph node resection. Ann Surg 1976;186:42–50.
- Fortner JG, Klimstra DS, Senie RT et al. Tumor size is the primary prognosticator for pancreatic cancer after regional pancreatectomy. Ann Surg 1996;223:147–153.[CrossRef][Medline]
- Imamura M, Doi R, Imaizumi T et al. A randomized multicenter trial comparing resection and radiochemotherapy for resectable locally invasive pancreatic cancer. Surgery 2004;136:1003–1011.[CrossRef][Medline]
- American Joint Committee on Cancer. AJCC Cancer Staging Manual. Sixth Edition. New York: Springer, 2002:157-164.
- Lygidakis NJ, Singh G, Bardaxoglou E et al. Mono-bloc total spleno-pancreaticoduodenectomy for pancreatic head carcinoma with portal-mesenteric venous invasion. A prospective randomized study. Hepatogastroenterology 2004;51:427–434.[Medline]
- Gudjonsson B. Carcinoma of the pancreas: Critical analysis of costs, results of resections, and the need for standardized reporting. J Am Coll Surg 1995;181:483–503.[Medline]
- Gudjonsson B. Pancreatic cancer: The need for critical reassessment. J Clin Gastroenterol 1996;23:2–6.[CrossRef][Medline]
- Gudjonsson B. Survival statistics gone awry: Pancreatic cancer, a case in point. J Clin Gastroenterol 2002;35:180–184.[CrossRef][Medline]
- Sarr MG, Behrns KE, van Heerden JA. Total pancreatectomy: An objective analysis of its use in pancreatic cancer. Hepatogastroenterology 1993;40:418–421.[Medline]
- Karpoff HM, Klimstra DS, Brennan MF et al. Results of total pancreatectomy for adenocarcinoma of the pancreas. Arch Surg 2001;136:44–47; discussion 48.[Abstract/Free Full Text]
- Ihse I, Anderson H, Andren-Sandberg A. Total pancreatectomy for cancer of the pancreas: Is it appropriate? World J Surg 1996;20:288–294.[CrossRef][Medline]
- Sugiyama M, Atomi Y. Pylorus-preserving total pancreatectomy for pancreatic cancer. World J Surg 2000;24:66–71.[CrossRef][Medline]
- Launois B, Franci J, Bardaxoglou E et al. Total pancreatectomy for ductal adenocarcinoma of the pancreas with special reference to resection of the portal vein and multicentric cancer. World J Surg 1993;17:122–127.[CrossRef][Medline]
- Swope TJ, Wade TP, Neuberger TJ et al. A reappraisal of total pancreatectomy for pancreatic cancer: Results from U.S. Veterans Affairs hospitals, 1987–1991. Am J Surg 1994;168:582–586.[CrossRef][Medline]
- Balcom JH 4th, Rattner DW, Warshaw AL et al. Ten-year experience with 733 pancreatic resections: Changing indications, older patients, and decreasing length of hospitalization. Arch Surg 2001;136:391–398.[Abstract/Free Full Text]
- Motojima K, Urano T, Nagata Y et al. Detection of point mutations in the Kirsten-ras oncogene provides evidence for multicentricity of pancreatic carcinoma. Ann Surg 1993;217:138–143.[Medline]
- Cooperman AM, Herter FP, Marboe CA et al. Pancreatoduodenal resection and total pancreatectomy—an institutional overview. Surgery 1981;90:707–712.[Medline]
- Ishikawa O, Ohhigashi H, Sasaki Y et al. Practical usefulness of lymphatic and connective tissue clearance for the carcinoma of the pancreas head. Ann Surg 1988;208:215–220.[Medline]
- Satake K, Nishiwaki H, Yokomatsu H et al. Surgical curability and prognosis for standard versus extended resection for T1 carcinoma of the pancreas. Surg Gynecol Obstet 1992;175:259–265.[Medline]
- Tsuchiya R, Oribe T, Noda T. Size of the tumor and other factors influencing prognosis of carcinoma of the head of the pancreas. Am J Gastroenterol 1985;80:459–462.[Medline]
- Schniewind B, Bestmann B, Henne-Bruns D et al. Quality of life after pancreaticoduodenectomy for ductal adenocarcinoma of the pancreatic head. Br J Surg 2006;93:1099–1107.[CrossRef][Medline]
- Kremer B, Vogel I, Luttges J et al. Surgical possibilities for pancreatic cancer: Extended resection. Ann Oncol 1999;10(suppl 4):S252–S256.
- Henne-Bruns D, Vogel I, Luttges J et al. Surgery for ductal adenocarcinoma of the pancreatic head: Staging, complications, and survival after regional versus extended lymphadenectomy. World J Surg 2000;24:595–602.[CrossRef][Medline]
- Henne-Bruns D, Vogel I, Luttges J et al. Ductal adenocarcinoma of the pancreas head: Survival after regional versus extended lymphadenectomy. Hepatogastroenterology 1998;45:855–866.[Medline]
- Yeo CJ, Cameron JL, Sohn TA et al. Pancreaticoduodenectomy with or without extended retroperitoneal lymphadenectomy for periampullary adenocarcioma: Comparison of morbidity and mortality and short-term outcome. Ann Surg 1999;229:613–624.[CrossRef][Medline]
- Yeo CJ, Cameron JL, Lillemoe KD et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcioma, part 2: Randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002;236:355–368.[CrossRef][Medline]
- Nguyen TC, Sohn TA, Cameron JL et al. Standard vs. radical pancreaticoduodenectomy for periampullary adenocarcinoma: A prospective, randomized trial evaluating quality of life in pancreaticoduodenectomy survivors. J Gastrointest Surg 2003;7:1–11.[CrossRef][Medline]
- Pedrazzoli S, DiCarlo V, Dionigi R et al. Standard versus extended lymphadenectomy associated with pancreatoduodenectomy in the surgical treatment of adenocarcioma of the head of the pancreas: A multicenter, prospective, randomized study. Lymphadenectomy Study Group. Ann Surg 1998;228:508–517.[CrossRef][Medline]
- Nimura Y, Nagino M, Kato H et al. Regional versus extended lymph node dissection in radical pancreatoduodenectomy for pancreatic cancer: A multicenter, randomized controlled trial. HPB Off J Int Hepato Pancreato Biliary Assoc 2004;6(suppl 1):2.
- Pawlik TM, Abdalla EK, Barnett CC et al. Feasibility of a randomized trial of extended lymphadenectomy for pancreatic cancer. Arch Surg 2005;140:584–591.[Abstract/Free Full Text]
- Tseng JF, Raut CP, Lee JE et al. Pancreaticoduodenectomy with vascular resection: Margin status and survival duration. J Gastrointest Surg 2004;8:935–950.[CrossRef][Medline]
- Harrison LE, Klimstra DS, Brennan MF. Isolated portal vein involvement in pancreatic adenocarcinoma: A contraindication for resection? Ann Surg 1996;224:342–349.[CrossRef][Medline]
- Hartel M, Neidergethmann M, Farag-Soliman M et al. Benefit of venous resection for ductal adenocarcinoma of the pancreatic head. Eur J Surg 2002;168:707–712.[CrossRef][Medline]
- Capussotti L, Massucco P, Ribero D et al. Extended lymphadenectomy and vein resection for pancreatic head cancer: Outcomes and implications for therapy. Arch Surg 2003;138:1316–1322.[Abstract/Free Full Text]
- Bachellier P, Nakano H, Oussoultzoglou PD et al. Is pancreaticoduodenectomy with mesentericoportal venous resection safe and worthwhile? Am J Surg 2001;182:120–129.[CrossRef][Medline]
- Roder JD, Stein HJ, Siewert JR. Carcinoma of the periampullary region: Who benefits from portal vein resection? Am J Surg 1996;171:170–175.[CrossRef][Medline]
- Takahashi S, Ogata Y, Tsuzuki T. Combined resection of the pancreas and portal vein for pancreatic cancer. Br J Surg 1994;81:1190–1193.[Medline]
- Klempnauer J, Ridder GJ, Bektas H et al. Extended resections of ductal pancreatic cancer—impact on operative risk and prognosis. Oncology 1996;53:47–53.[CrossRef][Medline]
- Shibata C, Kobari M, Tsuchiya T et al. Pancreatectomy combined with superior mesenteric-portal vein resection for adenocarcinoma in pancreas. World J Surg 2001;25:1002–1005.[CrossRef][Medline]
- Allema JH, Reinders ME, van Gulik TM et al. Portal vein resection in patients undergoing pancreatoduodenectomy for carcinoma of the pancreatic head. Br J Surg 1994;81:1642–1646.[Medline]
- Riediger H, Makowiec F, Fischer E et al. Postoperative morbidity and long-term survival after pancreaticoduodenectomy with superior mesenterico-portal vein resection. J Gastrointest Surg 2006;10:1106–1115.[CrossRef][Medline]
- Carrere N, Sauvanet A, Goere D et al. Pancreaticoduodenectomy with mesentericoportal vein resection for adenocarcinoma of the pancreatic head. World J Surg 2006;30:1526–1535.[CrossRef][Medline]
- Nakao A, Takeda S, Inoue S et al. Indications and techniques of extended resection for pancreatic cancer. World J Surg 2006;30:976–982; discussion 983–984.[CrossRef][Medline]
- Siriwardana HPP, Siriwardena AK. Systematic review of outcome of synchronous portal-superior mesenteric vein resection during pancreatectomy for cancer. Br J Surg 2006;93:662–673.[CrossRef][Medline]
Top
Learning Objectives
Abstract
