Pancreatico-duodenectomy: 5-year experience

Pancreatico-duodenectomy: 5-year experience

Alaa H Abdel-Razek, MD; Khaled M Katri, MD; Wael N Abdel-Salam, MD;

Elsaid A Elkayal, MD

Department of Surgery, Alexandria University, Egypt.

Abstract

Background: Pancreatico-duodenectomy is the only potentially curative treatment of pancreatic cancer and is considered one of the most technically demanding and challenging procedures. Survival after surgery of pancreatic cancer is still poor.

Aim of the study: was to estimate  the resectability rate of pancreatic  and periampullary tumors presenting with jaundice and to assess the outcome of tumors treated by pancreatico­ duodenectomy.

Patients and methods: Throughout a 5-year period from 2006 to 2011, 378 patients presented with tumors involving the head of pancreas or periampullary region. Pancreatico-duodenectomy was performed for 65 (17.2%) patients with resectable tumors. Follow up was done for 3 years. Univariate and multivariate analyses were done for factors affecting the postoperative survival.

Results: Resectability was feasible in 17.2% of patients. After pancreatico-duodenectomy, complications occurred in 32.3% of patients. Pylorus preservation did not affect the rate of complications. The 1-, 2- and 3-year survival rates were 64.6%, 43.1% and 35.4% respectively. Survival more than 3 years was significantly related to tumor diameter up to 3cm,free resection margin, absence of lymph nodes involvement, well differentiation of the tumors and periampullary tumors. By multivariate  regression analysis, tumor diameter not exceeding 3cm and absence of lymph nodes involvement were found to be the most predictive factors correlated to survival more than 3 years.

Conclusion: The malignant pancreatic tumors have a low resectability rate. Most of pancreatic tumors present in late stages.   Tumors not exceeding 3cm in diameter and absence of lymph nodes involvement were predictive of probable 3 year survival.

Key   words: Pancreatic carcinoma, jaundice, resectability,  surgical  outcome.

Introduction:

Pancreatic cancer is one of the most lethal human tumors and is currently the fifth leading cause of cancer death for men and the sixth for women. Ductal pancreatic adenocarcinoma is the most common pathological  type (90% of pancreatic  cancer) and is one of the most aggressive   human  malignancies1,2.

Pancreatico-duodenectomy (PD) is the only potentially curative  treatment of pancreatic cancer  and  is considered one  of the  most technically demanding and  challenging procedures. Although there has been constant progress in surgical techniques and advances in perioperative care with a modern interdisciplinary approach,  prognosis  is still life-threatening even after curative resection.3,4

The  aim of this  study  was  to estimate the

resectability rate of  pancreatic and periampullary tumors presenting with jaundice and to assess the outcome of the tumors treated by  pancreaticoduodenectomy.

Patients and methods:

This study was approved by the Committee of Ethics in Faculty ofMedicine, University of  Alexandria. This study was performed from  March  2006  to   October  2011.

Three  hundred  and   seventy  eight

consecutive patients with obstructive jaundice due  to mass  in the pancreatic head  or periampullary region were studied prospectively. The  studied patients were subjected to detailed medical history, full clinical examination and  laboratory investigations including liver function tests

Ain-Shams J Surg 2012; 5(1):9-16

and  tumor markers {CA  19-9 and carcinoembryonic antigen). Imaging included abdominal ultrasonography, abdominal CT and cholangiography for all patients. ERCP was performed for 302 {79.9%) patients and MRCP for 120 {31.7%) patients. Both ERCP and MRCP were performed for 44 {11.6%) patients. Preoperative endoscopic stents were inserted  in 26 {6.8%) patients, with serum bilirubin> 20mg/dl, to alleviate the jaundice and improve the patient fitness.

The data obtained by the previous workup were used for preoperative assessment of the patients. Diagnostic laparoscopy was performed for  9 patients with suspicious resectability.

Pylorus preserving pancreatico­ duodenectomy was performed for 19 patients while conventional pancreatico-duodenectomy was performed for 46 patients according to the preference of the surgeon. Pancreatico-jejunal anastomosis by duct to mucosa method was the preferred technique. Dunking method was used for patients with small caliber of pancreatic duct  and soft  pancreas. Postoperative histopathological examination was done for excised specimens.

Postoperative follow up was done every 3 months. This included clinical  assessment, liver function tests, tumor markers, abdominal ultrasonography and CT.

Statistics:

Data are presented, when appropriate, as

median and percentage. The Fisher's exact and chi-squared tests were performed for univariate analyses of categorical values.  Logistic regression was performed for multivariate model with P values and 95% confidence intervals estimated by the Wald method. A value ofp < 0.05 was considered statistically significant. All data analyses were performed with the Statistical Package for the Social Sciences version  18 software {SPSS, Inc., Chicago, IL).

Results:

The studied  patients  were 228 {60.3%) males and 150 (39.7 %) females. Their median age was 56 years. Their clinical presentations included jaundice in 378 (100 %) patients, dyspepsia in 320 (84.6%) patients, abdominal

pain in 356 (94.2%) patients, weight loss in

312 (82.5%)  patients and abdominal  mass

{palpable liver or gall bladder) in 108{ 28.5%)

patients.The laboratory workup showed anemia

{less than 70% of normal) in 66 (17.5%) patients, elevated serum bilirubin in 378 (1000/o) patients with a median of 10.51mg/dl, elevated tumor   markers:   CA 19-9   and/or carcinoembryonic  antigen  (CEA)  in

301(79.6%) patients. The  abdominal ultrasonography showed  the  presence  of pancreatic mass in263 (69.6%) patients, dilated biliary tract in 375 (99.2%) patients and ascites in 24 (6.3%) patients. Advanced unresectable tumors could be diagnosed by ultrasonography in 85(22.5%) patients  who showed liver metastases or ascites. Abdominal CT revealed pancreatic head tumor in 352 {93.1%) patients and periampullary tumor in26 (6.9%) patients, vascular invasion of major vessels {superior mesenteric vessels or portal vein) in 223(59

%) patients, enlarged peri-pancreatic and/or

para-aortic lymph nodes in193{51%) patients and liver metastases in 56 (14.8%) patients. Associated co-morbidities were present in 45

{11.9%) patients including diabetes mellitus in 23 patients, hypertension in 14 patients, liver  cirrhosis  in  5   patients  and chronic pulmonary disease in 3 patients. Exploration was done for 70 patients. On exploration, five patients were found unresectable due to major vascular  invasion  in 2 patients,  peritoneal seeding in 2 patients and liver metastases in one  patient.  These  operative findings  of unresectability were missed by preoperative imaging. Pancreatico-duodenectomy could be performed for 65 (17.2%) patients. Pancreatico­ jejunostomy was performed by duct to mucosa method in 57 (87.7% of operations) patients and by dunking method in 8 {12.3%) patients. Operative time ranged from 290 to 430 minutes with a median  of   325 minutes;  while the median intraoperative blood loss was 700ml. The hospital stay ranged from 9 to 34 days with  a median12  days  for uncomplicated patients and 26 days for complicated patients.

Postoperative complications occurred in 21

{32.3 %) patients. These complications  are shown in Table(l). Five patients (7.7%) died in the perioperative period. Two patients died due to complicated  pancreatic  fistula, one

patient died due to severe bleeding, one patient died due to severe renal shutdown on top of diabetic nephropathy and another patient died due to cholestatic jaundice complicated by liver failure. Pancreatic fistula occurred in 9 patients (13.8%). No significant difference was found in rates of postoperative complications between pylorus-preserving pancreatico­ duodenectomy and conventional pancreatico­ duodenectomy. These were 31.5% and 32.6% respectively. Follow up revealed that I-, 2- and 3-year survivals were 64.6%, 43.I% and

35.4% respectively.This is shown inFigure(l).

The cause of disease-related mortality was tumor recurrence.Recurrence was revealed by

abdominal CT as a mass at the resection area or pancreatic remnant, multiple para-aortic lymph nodes, liver metastases and/ or omental deposits.

By univariate analysis, survival more than

3 years was significantly  related to tumor

diameter not exceeding 3cm, free resection margin, negative lymph nodes, well differentiation of the tumors and periampullary tumors. This is shown in Table(l). By multivariate regression analysis, tumor diameter not exceeding 3cm and absence oflymph nodes involvement were found to be the most predictive factors correlated to survival more than  3 years.  This  is shown  in Table(3).

Table (1): Postoperative complications (morbidity).

Table (1): Relationship between tumor pathological criteria and survivaL

p1: p value between non survivor and >3 years survival

p2: p value between <1 year survival with (1-2 year survival, 2-3 years survival and>3years

survival)

p: p value for Chi-square test

MCp: p value for Monte Carlo test

FEp: p value for Fisher Exact test

*: Statistically significant at p0.05

Table (3): Multivariate regression analysis for tumor diameter,tumor location, resection margin,lymph nodes and tumor differentiation.

Discussion:

Time (month$)

Figure(1):Kaplan Meiersurvival curve.

patients with unresectable disease the risks

Surgical treatment of pancreatic cancer is

the only modality that can offer a chance of long-term survival. Many studies found that potentially  curative surgery is an option for only 15% of all   patients  with pancreatic adenocarcinoma.s In this studyt 65 patients (17.2 %) had resectable tumors.Complications occurred in33.90A. of patients; while 1-t 2- and

3-year survival rates were 64.6%, 43.1% and

35.4% respectively.

Despite improvements in imaging modalities, a high incidence of unresectable diseaset is still found at the time of surgery. It

is important to identify  patients  who are

candidates for curative resection and to spare

-&iiihl!fiil!ffNtiiDii*lll

associated with surgery and allow them proper

palliation.5,6 Staging laparoscopy may identify locally advanced disease and small peritoneal or hepatic lesions not identifiedby noninvasive imaging.6 In this study,diagnostic laparoscopy was used in 9 patients with suspicious resectability and was found beneficial to detect unresectable tumors in 4 patients.

A multivariate analysis showed  that neoplasms larger than 4.Scm, low leucocytic count  (<9500/mm), high  bilirubin levels (>137micromol/L) and tomographic tumor invasion are independent factors predicting unresectability.7

In this study,  313 ( 82.8%)  patients  had unresectable tumors.  The  reasons for unresectability in  these patients were involvement of major vessels (superior mesenteric, portal vein, or celiac trunk), local spread and/or metastasis (hepatic or peritoneal).

The  high  incidence of  lymphatic and perineural invasion of pancreatic cancer results in  poor  loco-regional control. In addition, sentinel lymph node mapping is not technically feasible in pancreatic cancer. Radical pancreatico-duodenectomy may achieve better loco-regional control, but is accompanied by increasing morbidity.8

Complete resection (RO) remains the only potentially curative  treatment of pancreatic ductal adenocarcinoma. R1 resection was defined as microscopic evidence of tumor

< or =1mm from a resection margin. The rate

of microscopic margin involvement (R1) varies markedly in the literature (from 5 to 85%) and Rl resections are frequently  underreported. Involvement of transection margins in contrast to mobilization margins defines a group whose outcome is significantly worse. Better survival in the resection group and similar perioperative risk  would  support  the decision to perform pancreatico-duodenectomy even when there is the possibility  of incomplete microscopic clearance.  In most of the patients, complete loco-regional surgical eradication of malignant tissue  is impossible. Patients selected for resection have been  shown to have  an advantage over operative bypass in terms of length of survivai.9-12

Pancreatic surgery has  been   improved dramatically. Perioperative mortality rates after Whipple's procedure in the 1980s exceeded

20%, but nowadays it is less than 5% in high volume centers. Currently, postoperative morbidity is considerable, about 30-50% and survival rate is still low. OverallS-year survival rate  was   reported   less   than  4%.2,13

Postoperative complications ofPD are still frequent  and severe. The pancreatic fistula represents the most relevant complication. The standard  and meticulous surgical  technique with good postoperative care would support early detection of complications and improvement of outcome. There is evidence that   delayed gastric  emptying   can   be

responsible for 50% of morbidity  after PD. Preservation of the pylorus and extent oflymph node removal have no impact on the incidence of delayed gastric emptying)4,15 The method of pancreatic anastomosis is crucial. The isolated defunctioned duct-to-mucosa pancreatico-jejunostomy is a safe procedure offering  good functional  results)6,17 In this study,  incidence of pancreatic fistula  was significantly less after  duct to mucosa  than dunking  pancreatico-j ejunostomy.

Preoperative biliary drainage by endoscopic stenting is associated with a high incidence of infective complications. There is no support of the routine use of biliary stenting prior to pancreatico-duodenectomy.18 In this study, preoperative stenting was used in 26 patients with serum bilirubin> 20mg/dl.

In this study, both pylorus-preserving and conventional  pancreatico-duodenectomy had no different impact  on the outcome. Other studies showed no evidence of relevant differences in morbidity, survival and quality of life between both techniques, if the principles of  viable and  tumor free  margins are followed.I'19

In this study, 1-, 2- and 3-year survival rates after pancreatico-duodenectomy were 64.6%,

43.1% and 35.4% respectively. Other studies reported median survival of RO, Rl and palliative bypass groups to be 27.2, 15.6 and

6.5 months respectively.  While 1-, 2- and 5- year survival of RO were  79 %, 48.3%and

21.5%  respectively; and  1-, 2- and 5-year survival of  Rl were  70%,  39.1  and  9.9% respectively. The  l-and 2-year  survival of palliative surgical bypass group were reported to   be   34%   and   0%   respectively)0,20

Many factors are determinant of the outcome and  survival after  PD.   Resection margin involvement is an important  determinant of overall survival. The cut-off  value  of 2cm tumor diameter was  not independently associated with outcome, however, tumor size was strongly associated with the risk of other adverse prognostic factors.21,22 Even  after complete resection and adjuvant chemotherapy, the 5-year survival rate does not exceed  20-

25%.23

In this study, survival more than 3 years was significantly related to tumor diameter not

exceeding 3cm, free resection margin, absence of  lymph nodes involvement, well differentiation of the tumors and periampullary tumors. By multivariate regression analysis, tumor diameter not exceeding 3cm and absence of lymph nodes involvement were found to be the most predictive factors correlated to survival more than 3 years.

In patients with vascular invasion, partial resection of superior mesenteric or portal vein can be done followed by application of a vein graft. It was found that postoperative morbidity and survival in patients with partial resection of superior mesenteric vein or portal vein were similar to those  patients without vein resection.24,25

Conclusion:

The malignant pancreatic tumors have a

low resectability rate. Most of pancreatic tumors present in late stages. Tumors not exceeding

3cm in diameter and absence of lymph nodes involvement were predictive of probable 3 year survival after pancreatico-duodenectomy.

References:

1-    Germano s    S,   Gourgiotis   S, Stavrothanaspoulou A, Alepas P, Zampitis N, Panteli A: Diagnostic and therapeutic approach to pancreatic adenocarcinoma.J GastrointestinLiver Dis 2006; 15(3): 257-

263.

2- K.leeff J, Michalski C, Friess H, Buchler MW: Pancreatic cancer: From bench to 5- year survival. Pancreas 2006; 33(2): 111-

118.

3- Loos M, Friess H, Kleeff J: Surgery of malignant pancreatic tumors. Radiologe

2009; 49(2): 137-143.

4- Makowiec F, Hopt UT, Adam U, zurHausen A, Wellner U, Keck T, et al: The lymph node ratio is the strongest prognostic factor after resection of pancreatic cancer. J Gastrointest Surg 2009; 13(7): 1337-1344.

5- Benzoni E, Rossit L, Coiutti A, Favero A, Saccomano  E, Zompicchiatti A, et al: Analysis of survival after pancreatic resection for oncological pathologies. Chir lta/2007; 59(1): 17-25.

6- Croome KP, Jayaraman S, Schlachta CM:

Preoperative staging   of cancer  of the

pancreatic head:  Is there  room  for improvement?  Can J Surg 2010; 53(3):

171-174.

7- Kama NA, Doganay M, Tutuncu T, Gozalan U, Daglar G, Yuksek YN:A scoring system for preoperative assessment of respectability in periampullary neoplasms. Bratisl  Lek Listy 2010; 111(5): 275-279.

8- Kocher HM, Sohail M, Benjamin IS, Patel AG: Technical limitations of lymph node mapping in pancreatic cancer. Eur J Surg Onco/2007; 33(7): 887-891.

9-  Delpero JR,   Turrini 0: Ductal adenocarcinoma of the head of the pancreas: A critical study ofR1 resection rates. Bull Cancer 2008;95(12):1193-8.

10-Fusai G, Warnaar N, Sabin CA, Archibong

S, Davidson BR: Outcome ofR1 resection in  patients undergoing pancreatico­ duodenectomy for pancreatic cancer. Eur J Surg Onco/2008; 34(12): 1309-1315.

11-Wakeman CJ, Martin IG, Robertson RW, Dobbs BR, Fizelle FA: Pancreatic cancer: Management and survival.  ANZ J Surg

2004; 74(11): 941-944.

12-Al-Ghnaniem R, Camprodon RA, Kocher

HM, Portmann B, Al-Nawab M, Shaikh H, et al: Strategy  to reduce the risk of positive pancreatic resection margin at pancreatico-duodenectomy. ANZ J Surg

2008; 78(4): 237-239.

13-Dellaportas D, Tympa  A, Nastos  C, Psychogiou V, Karakatsanis A, Polydorou A, et al: An ongoing  dispute in the management of severe pancreatic fistula: Pancreatectomy  or   not? World J Gastrointest  Surg 2010; 2(11): 381-384.

14-Molino D, Perrotti P, Napoli V, Antropoli C, Perrotta S, Presta L, et al: Surgical complications following pancreatico­ duodenectomy: Results of a single center experience. G Chir 2002; 23(11-12): 405-

412.

15-DeOliveira ML, Winter JM, Schafer M,

Cunningham SC, Cameron JL, Yeo CJ, et al: Assessment  of complications after pancreatic surgery. A novel grading system applied to 633 patients undergoing pancreaticoduodenectomy. Ann Surg 2006;

244(6): 931-939.

16-Kingsnorth  AN: Safety and function of