sign in sign up
<<
Home , Exploratory laparotomy, Pancreaticoduodenectomy

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector HPB, 2008; 10: 464Á471

ORIGINAL ARTICLE

Striving for a better operative outcome: 101 Pancreaticoduodenectomies

A. W. C. KOWa, S. P. CHANb, A. EARNESTb, C. Y. CHANa, K. LIMa, S. Y. CHONGa, K. H. LIMa,C.K.HOa, S. P. CHEWa & K. H. LIAUa

aHepatobiliary & Pancreatic Surgery Service, Department of Surgery, Digestive Centre, Tan Tock Seng Hospital, Singapore and bClinical Research Unit, Tan Tock Seng Hospital, Singapore

Abstract (PD), once carried high morbidity and mortality, is now a routine operation performed for lesions arising from the pancreatico-duodenal complex. This study reviews the outcome of 101 pancreaticoduodenectomies performed after formalization of HepatoPancreatoBiliary (HPB) unit in the Department of Surgery. A prospective database comprising of patients who underwent PD was set up in 1999. Retrospective data for patients operated between 1996 and 1999 was included. One hundred and one cases accrued over 10 years from 1996 to 2006 were analysed using SPSS (Version 12.0). The mean age of our cohort of patients was 61912 years with male to female ratio of 2:1. The commonest clinical presentations were obstructive jaundice (64%) and (47%). Majority had malignant lesions (86%) with invasive adenocarcinoma of the head of being the predominant histopathology (41%). Median operative time was 315 (180Á945) minutes. Two-third of our patients had pancreaticojejunostomy (PJ) while the rest had pancreaticogas- trostomy (PG). There were five patients with pancreatico-enteric anastomotic leak (5%), three of whom (3%) were from PJ anastomosis. Overall, in-hospital and 30-day mortality were both 3%. The median post-operative length of stay (LOS) was 15 days. Using logistic regressions, the post-operative morbidity predicts LOS following operation (pB0.005). The strategy in improving the morbidity and mortality rates of pancreaticoduodenectomies lies in the subspecialization of surgical services with regionalization of such complex surgeries to high volume centers. The key success lies in the dedication of staffs who continues to refine the clinical care pathway and standardize management protocol.

Key Words: pancreaticoduodenectomy, Whipple operation

Introduction the development of subspecialization, regionalization of this complex operation and creation of high volume Since the first successful removal of a periampullary centers. Advancement in the operative techniques, carcinoma with a sleeve of by William Halsted in 1898 [1], many developments have taken improvement in peri-operative care and standardiza- place with significant improvement in outcome of tion of post-operative care using clinical care pathway pancreaticoduodenal surgery. Pancreaticoduodenect- have also contributed significantly to the improved omy (PD) was subsequently popularized by Dr Allen outcome of this operation. O. Whipple after his success in the initial three cases In our institution, we restructured and reorganized in 1935. He published the operation in a landmark the Department of into subspeciality- paper and this is now commonly known as ‘‘Whipple’s based surgical practice in 1996. Hepatopancreatobili- operation’’ [2]. ary (HPB) Surgery Unit was formalized and HPB Previously, this surgery was criticized by many as it surgeons and nurses were recruited. In the last decade, carried high rates of morbidity and mortality [3]. pancreaticoduodenectomy was performed by our HPB However, in recent years, the morbidity and mortality and upper (UGIT) surgeons. associated with PD have dropped significantly due to This study analyses our institution’s outcome with

Correspondence: Kui Hin Liau, Department of Surgery, Hepatobiliary & Pancreatic Surgery Service, Tan Tock Seng Hospital, 11, Jalan Tan Tock Seng, Singapore 304833. Tel (65)6357-7807, Fax (65)6357-7809. E-mail: [email protected]

(Received 29 March 2008; accepted 12 May 2008) ISSN 1365-182X print/ISSN 1477-2574 online # 2008 Informa UK Ltd. DOI: 10.1080/13651820802247094 101 Pancreaticoduodenectomies 465 the 101 pancreaticoduodenectomies performed over Polydiaxanone (PDS† Ethicon, Inc, Johnson & John- the last 10 years. son) suture is our usual practice [6,7]. Gastrojejunost- omy is performed with an omega loop in end-to-side fashion. For PPPD, the distal and are Methods preserved and duodenojejunostomy is reconstructed A prospective database comprising of patients who with an antecolic omega jejunal loop, 70 cm from the underwent PD was set up in our institution in 1999. hepaticojejunostomy. Through our hospital operation record book, we A standard post-operative care pathway is used for included retrospective data of patients operated be- post-operative management in the wards. Patients are tween 1996 and 1999. From 1996 to 2006, a total of kept nil by mouth with nasogastric tube to passive 101 pancreaticoduodenectomies were performed. drainage and aspiration at four hourly intervals. A The data was analysed using SPSS (Version 12.0) single dose of 200 mcg of subcutaneous sandostatin is and Stata version 9.2 (Stata Corp, Texas, USA). We administered during pancreatic transection and this is analysed the morbidity, mortality and length of stay continued for one week post-operatively. The dose is (LOS) after surgery. The linear regression model was dependent on the consistency of the pancreatic tissue used to examine factors associated with LOS. We assessed during operation. If the pancreas is soft or analyzed LOS on the natural logarithmic scale, as we pancreatic duct is B3 mm, 200 mcg at eight hourly found the residuals from the model to be not normally dosing interval is administered, otherwise 100 mcg distributed. Logistic regression was used to identify eight hourly is given [6]. Patients are allowed non- predictors for morbidity and mortality. The data was milk feeds if nasogastric output is B100 ml on first also divided into two periods, from 1999 to 2001 and post-operation day [POD] and nasogastric tube is from 2002 to 2006 for comparison. Year 2002 was removed on second POD if the output remains B100 selected as the cut-off as it coincided with the year ml. Feeding is graduated as tolerated. In general, by when HPB unit started implementing standardized third to fourth POD, patients will be taking full diet. management protocol in our institution. Drain fluid and serum amylase are performed on Cases that are indicated for PD are discussed at our first, third and fifth POD. The drain fluids from weekly multi-disciplinary Pancreato-Biliary Manage- surgical tubes placed in the subhepatic space and left ment Conference. Cases are deemed suitable for infracolic compartment are assayed. We defined resection when there is no evidence of or pancreatico-enteric anastomotic leakage when the peritoneal metastases, no gross involvement of major drain fluid amylase level is more than 3X serum blood vessels. We would consider trial of resection of amylase and drainage from pancreatic bed is more portal or superior mesenteric vein if pre-operative than 100 ml per day from the fifth post-operative day assessment showed close abutment of tumor to these [8Á16]. All complications were documented clearly veins and venous reconstruction was possible. Prior to and graded according to the classification proposed by operation, patients with significant cardiac and re- Clavien et al. in 1992. [17Á20]. spiratory conditions are assessed by cardiologists and All histology specimens were examined and re- respiratory physicians respectively. All patients would ported by our in-house pathologists. When the origin then undergo pre-operative chest physiotherapy and of the cancer at the ampullary region could not be incentive spirometry to minimize post-operative re- convincingly demonstrated, peri-ampullary carci- spiratory morbidity. Smokers are advised to cease noma is reported. Our post-discharge follow-up smoking two weeks prior to the operation for the same schedule comprises of three monthly clinic reviews reason mentioned above. in the first two years, half-yearly review in the Prior to 2006, only classical Whipple’s operation subsequent three years and yearly follow-up if patient was performed in our institution. Pylorus-preserving is disease free after five years from the surgery. pancreaticoduodenectomy (PPPD) was added to our Routine CT imaging is not mandatory unless patient operative repertoire in early 2006. Our classical is symptomatic or has elevated serum tumor marker Whipple’s operation is defined as resection of distal level on follow-up. third of stomach with transection of common hepatic duct just proximal to the junction with cystic duct. clearance is confined to the hepatoduo- Results denal ligament and retropancreatic areas up to super- Demography ior mesenteric vessels [4,5]. The posterior pancreatic resection margin is at the plane posterior to portal vein The mean age of our cohort of 101 patients was 619 and lateral to superior mesenteric vessels. is 12 years. Sixty-two percent were male. Seventy-eight transected at the first jejunal mesenteric branch. The percent of them were Chinese; Malay and Indian choice for the type of pancreatico-enteric anastomosis comprised 6% and 9%, respectively. Two-third of is based on surgeon’s preference. End-to-side hepati- them had one or more comorbidities. The two most cojejunostomy, about 10 cm distal to pancreaticojeju- common comorbidities were hypertension (29%) and nostomy, constructed using interrupted 4/0 or 5/0 mellitus (18%). All the patients had ASA 466 A. W. C. Kow et al. grade 53, slightly more than half (53%) of the patients were graded ASA 2. Table I. Pathological diagnoses of patients who underwent PD in our institution.

Presentation and pre-operative procedure Site of lesion No. (%)

The commonest clinical presentations were obstruc- Head of pancreas tive jaundice (64%) and abdominal pain (47%) for a Benign Serous cystadenoma 2 variety of diagnoses (Figure 1). Majority of patients Mucinous cystadenoma 1 had malignancy arising from the pancreaticoduodenal Cavernous lymphangioma 1 Inflammatory Chronic 4 complex (83%). Inflammatory pseudocyst 1 About one-third of our patients underwent pre- Malignant IPMT 4 operative endobiliary stenting (29%) for either in- Neuroendocrine tumour 2 fective or non-infective obstructive biliopathy. This Invasive adenocarcinoma 32 46.6 represented half of our patients who presented with Ampulla of Vater obstructive jaundice. Benign Dysplastic adenoma 2 Malignant Invasive adenocarcinoma 32 33.7 Periampullary Malignant Invasive adenocarcinoma 6 5.9 Operative data Duodenal All pancreaticoduodenectomies were performed by Benign AVM 1 either HPB or UGIT surgeon in our institution. All Malignant Invasive adenocarcinoma 5 5.9 Distal CBD were elective cases except for one patient who under- Malignant 3 3.0 went emergency PD for a traumatic pancreatic Others 5 5.0 transection following motor-vehicle accident. Ten patients (10%) underwent pylorus-preserving pan- creaticoduodenectomy (PPPD). Pathology The median length of operation was 315 (180Á945) Malignancies arising from the head of pancreas (32%) minutes. The outlier of the operation duration was a and ampulla of Vater (32%) were the two commonest man with who developed carci- pathology (Table I). Seventy-two percent of them noma of the head of pancreas. Because of his two were moderately differentiated in histological grade. previous upper abdominal operaions and chronic The median tumor size for malignant lesions was pancreatitis, 3Á4 hours were spent on adhesiolysis 27 (5Á160) mm. prior to pancreaticoduodenectomy. Median estimated intra-operative blood loss was 500 (150Á6400) ml. The median blood transfusion was 1.6 (0Á7.0) units. Post-operative outcome The same patient mentioned above had 6400 ml of intra-operative blood loss due to prolonged and Most of the patients were monitored in the surgical extensive dissection. Two-third of our patients high dependency (SHD) ward for a median of (69%) had pancreaticojejunostomy (PJ) while the 3(1Á10) days. The median post-operative LOS was rest (31%) had pancreaticogastrostomy (PG). Prior 11 (4Á90) days. One-third (37%) of our patients had to 2003, subcutaneous sandostatin was administered peri-operative morbidity. The majority of these are selectively and this practice was embedded as part of complications arising from wound infections, intra- our standardized peri-operative care protocol only abdominal complications, intestinal complications, from 2003 onwards. bleeding, respiratory complications and anastomotic

Obstructive 70 Jaundice, 63.9

60

Pain, 46.4 50

LOA, 37.1 LOW, 38.1 40

30

Percentage (%) 20 Mass, 10.3 Dyspepsia, 5.2 10 Gastric outlet obstruction, 1

0

Figure 1. Clinical presentation of patients prior to surgery. 101 Pancreaticoduodenectomies 467 Table II. Table of comparison for morbidities after PD surgery. Table III. Reasons for repeat exploratory in patients after pancreaticoduodenectomy. n101 cases % Operation and findings Wound Wound infection 4 4 Anastomotic leak 1. PG leak with laparotomy and refashioning Burst 1 1 related of PG and CJ/repair of burst abdomen Intra-abdominal complication 2. and evacuation of Intra-abd 5 5 Peripancreatic collection 1 1 3. Take down of PJ, drainage of collection, Other intra-abd collection 3 3 revision of HJ 4. Take down of PG and partial Anastomotic leak 5. DJ stenting across HJ leak and T-tube PJ leak 3 3 PG leak 2 2 Not related to leak 6. Gastric outlet obstruction with laparotomy HJ leak 1 1 and adhesiolysis done 7. Exploratory laparotomy for delayed gastric Intestinal complication emptying 4 4 8. Resection of tumor margin due to tumor Prolonged biliary to bowel transit 1 1 involvement Delayed gastric emptying 5 5 IO Á efferent loop partial obstruction 1 1 Gastric outlet obstruction 1 1 Overall, in-hospital and 30-day mortality were both Dumping syndrome 1 1 3%. There were two mortalities, one from acute renal Bleeding complication failure following postoperative sepsis, while another BGIT 3 3 Intra-abd bleeding 3 3 from acute myocardial infarction in the first week of operation. The third patient passed away from pan- Chest complication Pneumonia 4 4 creaticoduodenal artery haemorrhage on the fifth Pleural effusion 3 3 post-operative day. The diagnosis was made promptly Pulmonary oedema 1 1 but patient adamantly declined surgical intervention Prolonged ventilation 2 2 after failed attempts at angiographic dearterialization. Atelectasis 2 2 Other complication Enterocutaneous fistula 1 1 Predictors of outcome Using logistic regression, age (p0.05) and gender leak (Table II). Most of these morbidities were Grade (p0.05) did not influence post-operative LOS 2 complications (24%). There were five pancreatico- (Table IV). However, the presence of any morbidity enteric anastomotic leaks (5%) of which three (3%) significantly prolongs the post-operative LOS were from PJ anastomosis (Figure 2). There was one (pB0.05). As the post-operative LOS was not nor- bile leak from hepaticojejunostomy dehiscence. mally distributed, log transformation was performed Eight patients (8%) in our series required un- and subsequent linear regression showed that Grade 2 planned re-operations within two weeks after PD (pB0.001) and 3 (p0.003) complications were due to various complications (Table III). Five re- significantly associated with longer post-operative operations were performed to address complications LOS (Table V). When analyzing anastomotic leak as arising from anastomotic leak. Table IV. Predictors of outcome for pancreaticoduodenectomies.

80% Variables P-value 69% 70% Anastomosis Post-op LOS vs. 60% Leak rate Age 0.59 Gender 0.33 50% Anastomotic Leak vs. 40% Somatostatin administration 0.40 31% PJ vs. PG 0.90 30% SICU LOS 0.56 Percentage (%) 20% SHD LOS 0.38 Post-op LOS 0.20 10% 3% 2% Pre-op endobiliary stenting vs. 0% Infective complication 0.73 PJ PG Types of PE Anastomosis Note: P-value B0.05 as statistically significant. PJ, pancreaticoje- junostomy; PG, pancreaticogastrostomy; SICU, Surgical Intensive Figure 2. Types of pancreaticoenteric anastomosis and anastomotic Care Unit; SHD, Surgical High Dependency Unit; LOS, length of leak rate. stay. 468 A. W. C. Kow et al. Table V. Linear regression of post-operative LOS vs. grade of and morbidity between these two periods, the mor- complications (GOC). tality has dropped from 5 to 0% while overall morbidity dropped from 42 to 25%. The leak rate Log post-op LOS P-value Median p25 p75 has also decreased from 7 to 2%. Using Fisher’s exact GOC Grade 0 vs 9.5 8 12 test, these differences did not reached statistical GOC 1 0.056 17 11 31 significance. GOC 2 B0.001 18 11.5 27.5 GOC 3 0.003 17 9 41 GOC 4 0.82 10 6.5 22 Discussion

Note: p25, 25th percentile; p75, 75th percentile. GOC, Grade of Operative outcomes of pancreaticoduodenectomy complications based on Clavien et al. [17]. P-values obtained from (PD) have been widely published in the western linear regression model with LOS analyzed on the logarithmic scale. countries in the past decade but literatures from Asian centers are limited. There is paucity of outcome data an independent variable, it did not significantly for Whipple’s operation in Singapore. In this study, increase the LOS in Surgical Intensive Care Unit we share our one decade of experience in pancreati- (SICU), SHD or post-op LOS (Table IV). This could coduodenectomy and review the outcome of PD in a be due to the small sample size in our cohort of tertiary teaching hospital. patients. In our institution, this operation was initially In our series, the use of pre-operative endobiliary performed by general surgeons. However, as we stent in obstructed biliary track was also not found to progress toward subspecialization, this operation is increase post-operative infective complications (p now commonly performed by specialized trained 0.73). However, low post-operative hemoglobin level HPB surgical team since 1996 in our institution. (p0.04) and elevated pre-operative (p0.03) bilir- This trend is in tandem with the global phenomenon ubin level at presentation significantly correlated with of subspecialty-based surgical practice and regionali- higher morbidity. Although half of the patients had zation of major and complex operations to high- somatostatin analogue administration peri-opera- volume centers [21]. This has resulted in significant tively, this had not been shown to decrease the rate improvement in mortality and morbidity rates. Singa- of anastomotic leak (p0.40) (Table IV). We also did pore is a small island country with an average not find a significant difference in anastomotic leak population of about three million people. With the rate between PJ and PG anastomosis techniques influx of immigrants after the millennium, the popu- (p0.90). Again, the small sample size is a plausible lation now is about four million people. Our hospital explanation to this observation. has a healthcare catchment area of 1Á1.5 million of people, serving the Northern and North-eastern part of the country. Comparing outcomes of two periods Our overall in-hospital and 30-day mortality rate of To analyze the outcome after standardized peri- 3% is comparable to other specialized centers in this operative PD management protocol was implemented region. Poon et al. from Hong Kong reported a hospital in 2002, we divided the data into two periods, first five mortality rate of 2.9% in 140 patients over a 12-year- years (1996Á2001) and second five years (2002Á period [22]. Simliarly, these numbers were also con- 2006). While there were more PJ (n37) than PG sistent with the mortality rate quoted from various (n7) performed after year 2002, there was no high-volume centers around the world [23Á32]. difference in terms of age, gender, race, ASA grade, Birkmeyer et al. noticed that there was a trend toward operative time and intra-operative blood loss between regionalizing pancreaticoduodenectomy to high-vo- the two periods. lume centers in the USA due to better results and Further analysis of this data using student’s t-test lower in-hospital mortality rate. They divided the showed that the LOS in terms of total LOS and post- hospital’s average annual volume of pancreatocoduo- operation LOS were significantly shorter for PD done denectomies into four categories: very low (B1/year), after year 2002 (Table VI). Comparing the mortality low (B1Á2/year), medium (2Á5/year) and high (5/year). They observed that in-hospital mortality Table VI. Comparison of LOS before and after year 2001. rates at low and very low-volume hospitals were three to fourfold higher than at high-volume hospitals [31]. Median This was supported by Gordon et al who found that centers with high volume ( 20 cases per year) had Variables (Days) Before 2001 After 2001 P-value lower mortality rate for PD at 2.2% while low-volume LOS in SICU 0 0 0.27 centre (1Á5 cases per year) had higher mortality rate at LOS in HD 3 3 0.84 19.1% [32]. By Birkmeyer’s definition, our institution Total LOS 20 10 B0.001 falls into the high-volume center (medium volume by Post-op LOS 13.5 9 0.001 Gordon’s definition), and the mortality rate correlates (pB0.05 as statistically significant) well with the operative volume. 101 Pancreaticoduodenectomies 469 Many high-volume centers reported overall mor- protocol that was implemented, the outcome has bidity rate of PD of about 41Á50% [22,33Á41]. In improved remarkably. our series, 35% of our patients had one or more morbidities and this was again comparable to that Conclusion reported by Poon et al. from Hong Kong (38.6%) [22]. Majority of our morbidity was grade 2 com- The strategy in improving the morbidity and mortality plication. However, implementation of a standar- rates of pancreaticoduodenectomies lies in the sub- dized management protocol and care pathway has specialisation of surgical services with regionalization been shown to significantly decrease the risk of of such complex surgeries to high-volume centers. complications. Many centers have shown that proto- The key success lies in the dedication of staffs who colized management by an experienced team com- continue to refine the clinical care pathway and prising of surgeons, nurses and anesthesia medical standardize management protocol. Given such safety profile, it is therefore justified to extend the indication staffs can effectively improve clinical outcome. In of PD to benign conditions and presumed malignant specialized centers, post-operative and overall LOS pathology arising from pancreaticoduodenal complex. decreased with increased experience of the specia- lized team. Operation performed by a dedicated team helps to minimize leak rate and shorten LOS. This is due to familiarity with the operation and References better surgical techniques. [1] Halsted WS. Contributions to the surgery of the bile passages,

As mentioned earlier, the choice of pancreatico- especially of the common . Boston Med Surg J 1899;/ enteric anastomosis, i.e. PJ or PG, is entirely based on 141:645/ Á54. surgeon preference in our center. Randomized con- [2] Whipple AO, Parsons WB, Mullins CR. Treatment of

carcinoma of ampulla of Vater. Ann Surg 1935;102:/ 763/ Á79. trolled trials revealed no significant difference between [3] Janes RH, Niederhuber JE, Chmiel JS, Winchester DP, PJ and PG regarding overall postoperative complica- Ocwieja KC, Karnell LH. National patterns of care for tions, pancreatic fistula, intra-abdominal collection or : results of a survey by the commission on mortality [42Á44]. Although non-randomized obser- cancer. Ann Surg 1996;223:/ 268/ Á72. vational clinical studies showed significant results in [4] Pedrazzoli S, Beger HG, Obertop H, Andren-Sandberg A, Fernandez-Cruz L, Henne-Bruns D. A surgical and patholo- favor of PG in terms of reduction in pancreatic fistula gical based classification of resective treatment of pancreatic and mortality rate [45Á57], meta-analysis concluded cancer. Summary of an international workshop on surgical no significant difference between the two techniques procedures in pancreatic cancer. Dig Surg 1999;16:/ 337/ Á45. [58,59]. In our series, we did not observe any differ- [5] Ishikawa O, Ohhigashi H, Sasaki Y, Kabuto T, Fukuda I, Furukawa H. Practical usefulness of lymphatic and connective ence in outcome between PJ and PG. tissue clearance for the carcinoma of the pancreas head. Ann

The other strength of a matured surgical team is that, Surg 1998;208:/ 215/ Á20. as the team gathers more experience and momentum in [6] Friess H, Ho CK, Kleeff J, Buchler MW. Pancreaticoduode- performing PD, it allows them to explore and introduce nectomy, distal , segmental pancreatectomy, total pancreatectomy, and transduodenal resection of the improved modifications to the techniques of Whipple’s papilla of Vater: surgery of the liver, biliary tract, and pancreas. operation. Over the years, there have been many Fourth Edition 2007. creative modifications to performing PD. One such [7] Warshaw A, Thayer SP. Pancreaticoduodenectomy: how I do technique that has gained popularity as well as con- it. J Gastrointest Surg 2004;8(6):/ 733/ Á41. [8] Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki troversy is the pylorus-preserving pancreaticoduode- J, et al. For the international study group on pancreatic fistula nectomy (PPPD). Our institution started to perform definition. Postoperative pancreatic fistula: an international this operation in early 2006, nine years after our study group (ISGPF) definition. Surg 2005;138(1):/ 8/ Á13. subspecialized hepatobiliary surgery service was for- [9] Schinchi H, Wada K, Traverso W. The usefulness of drain data to identify a clinically relevant pancreatic anastomotic leak malised. We have also developed new modifications to

after pancreaticoduodenectomy? J Gastrointest Surg 2006;/

the technique of pancreatic anastomosis as we continue 10(4):490/ Á8. to explore this operation. Our current preferred PJ [10] Aranha GV, Aaron JM, Shoup M, Pickleman J. Current technique is duct-to-mucosa in two layers. management of pancreatic fistula after pancreaticoduodenect- Our 10 years of audited data has helped us to omy. Surg 2006;140(4):/ 561/ Á9. [11] Cullen JJ, Sarr MG, Ilstrup DM. Pancreatic anastomotic leak reshape the patient care protocol and refine the after pancreaticoduodenectomy: incidence, significance and

surgical techniques required in performing PD. As management. The Am J of Surg 1994;168:/ 295/ Á8. we strive to improve the outcome of our patients [12] Poon RTP, Lo SH, Fong D, Fan ST, Wong J. Prevention of undergoing PD, we periodically review our database pancreatic anastomotic leak after pancreaticoduodenectomy.

The Am J of Surg 2002;183:/ 42/ Á52. to explore further opportunity to improve our surgical [13] Grobmyer SR, Rivadeneira DE, Goodman CA, Mackrell P, techniques and care protocol. With the management Lieberman MD, Daly JM. Pancreatic anastomotic failure after 470 A. W. C. Kow et al.

pancreaticoduodenectomy. The Am J of Surg 2000;180:/ 117/ Á [33] Trede M, Schwall G, Saeger HD. Survival after pancreato- 20. duodenectomy: 118 consecutive resections without an opera-

[14] Shrikhande SV, Qureshi SS, Rakneesh N, Shukla PJ. Pan- tive mortality. Ann Surg 1990;221:/ 447/ Á58. creatic anastomoses after pancreaticoduodenectomy: do we [34] Fernandez DC, Rattner DW, Warshaw AL. Standards for

/ / need further studies? World J of Surg 2005;29:/ 1642/ Á849. pancreatic cancer resection in the 1990s. Arch Surg 1995;130: [15] Kazanjian KK, Hines OJ, Eibl G, Reber HA. Management of 295Á300. pancreatic fistulas after pancreaticoduodenectomy: results in [35] Imperato PJ, Nenner RP, Starr HA, Will TO, Rosenberg CR, Dearie MB. The effects of regionalization on clinical outcomes 437 consecutive patients. Arch of Surg 2005;140:/ 849/ Á55. [16] Ho CK, Kleeff J, Friess H, Buchler MW. Complications of for a high-risk surgical procedure: a study of the Whipple

procedure in New York State. Am J Med Qual 1996;11:/ 193/ Á7. pancreatic surgery. HPB 2005;7:/ 99/ Á108. [17] Clavien P, Sanabria J, Strasberg S. Proposed classification of [36] Gouma DJ, van Geenen RCI, van Gulik TM, de Haan RJ, de complication of surgery with examples of utility in cholecys- Wit LT, Busch ORC, Obertop H. Rates of complications and death after pancreaticoduodenectomies: risk factors and the

tectomy. Surg 1992;111:/ 518/ Á26. [18] Clavien P, Sanabria J, Mentha G. Recent results of elective impact of hospital volume. Ann Surg 2000;232(6):/ 786/ Á95. open in a North American and a European [37] Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand

consecutive pancreaticoduodenectomies. Ann Surg 2006;/

center. Ann Surg 1992;216:/ 618/ Á26.

244(1):10/ Á5. [19] Clavien P, Camargo C, Croxford R. Definition and classifica- [38] Miedema BW, Sarr MG, van Heerden JA, Nagorney DM, tion of negative outcomes in solid transplantation. Ann McIlrath DC, Ilstrup D. Complications following pancreati-

Surg 1994;220:/ 109/ Á20.

coduodenectomy. Current management. Arch of Surg 1992;/ [20] Dindo D, Demartines N, Clavien P. Classification of surgical

127(8):945/ Á949. complications: a new proposal with evaluation in a cohort of [39] Yeo CJ, Cameron JL, Sohn TA, Lillimoe KD, Pitt HA, 6336 patients and results of a survey. Ann Surg 2004;240(2):/ / Talamini MA, et al. Six hundred fifty consecutive pancreati- 205Á13. coduodenectomies in the 1990s. Ann Surg 1997;226(3):/ 248/ Á [21] Birkmeyer JD, Siewers AE, Finlayson EVA. Hospital volume 57. and surgical mortality in the United States. N Engl J Med [40] Schimdt CM, Powell ES, Yiannoutsos CT, Howard TJ,

2002;346:/ 1128/ Á37. Wiebke EA, Wiesernauer CA, et al. Pancreaticoduodenect- [22] Poon RTP, Fan ST, Chu KM, Poon JTC, Wong J. Standards omy: a 20-year experience in 516 patients. Arch Surg 2004;/ of pancreaticoduodenectomy in a tertiary referral centre in 139:718/ Á27. Hong Kong: retrospective case series. Hong Kong Med J [41] Dimick JB, Pronovost PJ, Cowan JA Jr, Lipsett PA, Stanley

2002;8(4):/ 249/ Á54. JC, Upchurch GR Jr. Variation in postoperative complication

[23] Winter JM, Cameron JL, Campbell KA, Arnold MA, Chang rates after high-risk surgery in the United States. Surg 2003;/

DC, Coleman J, et al. 1423 pancreaticoduodenectomies for 134:534/ Á41. pancreatic cancer: a single-institution experience. J Gastro- [42] Yeo CJ, Cameron JL, Maher MM. A prospective randomized intest Surg 2006;10(9):1199Á211. trial of pancreaticogastrostomy versus pancreaticojejunostomy

[24] Chew DKW, Attiyeh FF. Experience with the Whipple after PD. Ann Surg 1995;222:/ 580/ Á8. procedure (Pancreaticoduodenectomy) in a University-af- [43] Duffas JP, Suc B, Msika S. A controlled randomized multi-

fliated community hospital. The Am J of Surg 1997;174:/ / center trial of pancreaticogastrostomy or pancreaticojejunost-

312Á5. omy after PD. Am J Surg 2005;189:/ 720/ Á9. [25] Crist DW, Sitzmann JV, Cameron JL. Improved hospital [44] Bassi C, Falconi M, Molinari E. Reconstruction by pancrea- morbidity, mortality and survival after the Whipple procedure. ticojejunostomy versus pancreaticogastrostomy following pan-

createctomy: results of a comparative study. Ann Surg 2005;/ Ann Surg 1997;206(3):/ 358/ Á65.

[26] Lieberman MD, Kilburn H, Lindsey M, Brennan MF. 242:767/ Á71. Relationship of perioperative deanths to hospital volume [45] Ramesh H, Thomas PG. Pancreaticojejunostomy versus among patients undergoing pancreatic resection for malig- pancreaticogastrostomy in reconstruction following PD. Aust

N Z J Surg 1990;60:/ 973/ Á6.

nancy. Ann Surg 1995;222(5):/ 638/ Á45. [27] Sosa JA, Bowman HM, Gordon TA, Bass EB, Yeo CJ, [46] Miyagawa S, Makuuchi M, Lygidakis NJ. A retrospective comparative study of reconstructive methods following PD Lillemoe KD, et al. Importance of hospital volume in the Á pancreaticojejunostomy versus pancreaticogastrostomy. Hepa-

overall management of pancreatic cancer. Ann Surg 1998;/

togastroenterol 1992;39:/ 381/ Á4.

228(3):429/ Á38. [47] Morris DM, Ford RS. Pancreaticogastrostomy: preferred [28] Neoptolemos JP, Russel RCG, Bramhale S. Low mortality

reconstruction for Whipple resection. J Surg Res 1993;54:/ / following resection for pancreatic and periampullary tumours 122Á5. in 1026 patients. Br J Surg 1997;84:/ 1370./ [48] Mason GR, Freeark RJ. Current experience with pancreato- [29] Afsari A, Zhangdoug Z, Young S, Feruson L, Silapaswam S,

. Am J Surg 1995;169:/ 217/ Á9. Mittal V. Outcome analysis of pancreaticoduodenectomy at a [49] Gvalani AK. Pancreaticogastrostomy versus pancreaticojeju-

community hospital. Am Surg 2002;68(3):/ 281/ Á5. nostomy following PD for periampullary carcinoma. Indiana J [30] Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wenn- Gastroenterol 1996;15:/ 132/ Á4. berg DE, Lucas FL. Surgeon volume and operative mortality [50] Kim SW, Youk EG, Park YH. Comparison of pancreaticogas-

in the United States. N Engl J Med 2003;329(22):/ 2117/ Á27. trostomy and pancreaticojejunostomy after PD performed by [31] Birkmeyer JD, Finlayson SRG, Tosteson ANA, Sharp SM, one surgeon. World J Surg 1997;21:/ 640/ Á3. Warshaw AL, Fisher ES. Effect of hospital volume on in- [51] Arnaud JP, Tuech JJ, Cervi C, Bergamaschi R. Pancreatico-

hospital mortality with pancreaticoduodenectomy. Surg 1998;/ gastrostomy compared with pancreaticojejunostomy after PD.

/ 125(3):250Á6. Eur J Surg 1999;165:/ 357/ Á62. [32] Gordon TA, Burleyson GP, Tielsch JM, Cameron JL. The [52] Takano S, Ito Y, Watanabe Y. Pancreaticojejunostomy versus effects of regionalization on cost and outcome for one general pancreaticogastrostomy in reconstruction following PD. Br J

high-risk surgical procedure. Ann Surg 1995;221(1):/ 43/ Á9. Surg 2000;87:/ 423/ Á7. 101 Pancreaticoduodenectomies 471 [53] Fujino Y, Suzuki Y, Ajiki T. Risk factors influencing pancreatic [57] Oussoultzoglou E, Bachellier P, Bigourdan JM. Pancreatico- leakage and the mortality after PD in a medium-volume gastrostomy decreased relaparotomy caused by pancreatic

hospital. Hepatogastroenterol 2002;49:/ 1124/ Á9. fistula after PD compared with pancreaticojejunostomy. Arch [54] Schlitt HJ, Schimdt U, Simunec D. Morbidity and mortality Surg 2004;139:/ 327/ Á35. associated with pancreatigastrostomy and pancreatojejunost- [58] Wente MN, Shrikhande SV, Mu¨ller MW, Diener MK, Seiler omy following partial pancreaticoduodenectomy. Br J Surg CM, Friess H, Bu¨chler MW. Pancreaticojejunostomy versus

2002;89:/ 1245/ Á51. pancreaticogastrostomy: systematic review and meta-analysis. [55] Aranha GV, Hodul P, Golts E. A comparison of pancreatico- The Am J of Surg 2007;193:/ 171/ Á83. gastrostomy and pancreaticojejunostomy following PD. J [59] McKay A, Mackenzie S, Sutherland FR, Bathe OF, Doig C,

Gastrointest Surg 2003;7:/ 672/ Á82. Dort J, Vollmer CM, Dixon E. Meta-analysis of pancreatico- [56] Hoshal VL, Benedict MB, David LR, Kulick J. Personal versus pancreaticogastrostomy reconstruction experience with the Whipple operation: outcomes and lessons

after pancreaticoduodenectomy. BJS 2006;93:/ 929/ Á36.

learned. Am Surg 2004;70:/ 121/ Á5.