- Email: [email protected]
Extended central pancreatic resection as an alternative for extended left or extended right resection for appropriate pancreatic neoplasms
Extended central pancreatic resection as an alternative for extended left or extended right resection for appropriate pancreatic neoplasms Guellue Cataldegirmen, MD, Claus G. Schneider, MD, Dean Bogoevski, MD, Alexandra Koenig, MD, Jussuf T. Kaifi, MD, Maximilian Bockhorn, MD, Lena S. Deutsch, MD, Yogesh Vashist, MD, Jakob R. Izbicki, MD, and Emre F. Yekebas, MD, Hamburg, Germany
Background. Whether patients with focal pancreatic lesions of benign or borderline pathology should be treated by extended central pancreatectomy rather than by extended classic resectional procedures, such as extended right and left resections, is controversial. Methods. Between 1992 and 2007, 105 patients underwent operation for focal pancreatic lesions of borderline or benign neuroendocrine neoplasms, cystadenoma, intraductal papillary mucinous neoplasia (IPMN), and secondary metastasis. In all, 35 patients were subjected to extended central pancreatectomy, whereas the remaining 70 patients were treated by an extended classic right resection or an extended classic left resection. Groups were matched according to age, sex, and histopathology. Results. No peri-operative mortality occurred after extended central pancreatectomy and extended classic left resection (n = 35, each). Two (6%) patients died after extended classic right resection. Overall, inhospital morbidity was 26% after extended central pancreatectomy, 43% after extended classic right resection, and 37% after extended classic left resection. After a median follow-up of 48 months, a local recurrence rate of 17% after extended central pancreatectomy was similar to the corresponding rates of 9% after extended classic left resection and 14% after extended classic right resection. Endocrine and exocrine impairment was less pronounced after extended central pancreatectomy (6% and 9%) than after extended classic left resection (34% and 29%) and extended classic right resection (28% and 24%; P < .05). Conclusion. Extended central pancreatectomy for appropriate pancreatic neoplasms is associated with less peri-operative morbidity and mortality than after extended classic left and extended classic right resection. Long-term local recurrence after extended central pancreatectomy is similar to the recurrence rates after extended classic right and classic left resection. Our results suggest that appropriately selected patients will benefit from extended central pancreatectomy because of the maintenance of endocrine and exocrine function. (Surgery 2010;147:331-8.) From the Department of General, Viszeral and Thoracic Surgery, University Medical Center HamburgEppendorf, University of Hamburg, Hamburg, Germany
ALTHOUGH MORTALITY AFTER MAJOR PANCREATIC SURGERY has decreased to less than 3%,1-4 morbidity remains considerably high, ranging from 18% to 52%.4-8 Organ-sparing pancreatic resections, such as enucleation and central pancreatic resection, are suggested as suitable procedures for borderline Accepted for publication October 6, 2009. Reprint requests: Emre F. Yekebas, MD, Department of General, Visceral, and Thoracic Surgery, University Medical Centre Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2010 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2009.10.027
and benign neuroendocrine neoplasms located in the neck and body of the gland. These types of resections offer an alternative to extended classic resections with the main advantage of preserving unaffected pancreatic parenchyma.9,10 Subsequent arguments in favor of central pancreatectomy are the preservation of the spleen, thereby minimizing the splenectomy-associated morbidity related to infectious and/or thrombotic complications, which can occur when performing standard resections to the left of the mesenterico-portal axis. The preservation of both gastroduodenal and biliary continuity is an advantage of central pancreatectomy in patients with appropriately selected right-sided lesions.11,12 SURGERY 331
332 Cataldegirmen et al
Since the first publication by Guillemin and Bessot in 1957,13 central pancreatectomy has been applied increasingly to benign and borderline lesions of the pancreas neck and body.14 Nonetheless, surgeons’ attitude toward central pancreatectomy is often characterized by reluctance,9,10,15-17 which is derived from doubts concerning several aspects of the procedure. First, complete tumor resection may require special technical expertise limited to specialized centers. Second, because of the limitations of current staging examinations, the accuracy of preoperative diagnosis is often questioned. Last, because central pancreatectomy results in 2 pancreatic stumps that must be dealt with, this procedure is assumed to carry a greater risk of operative morbidity associated with pancreatic fistula. Our institutional experience reflects the value of extended central pancreatectomy for treatment of nonmalignant pancreatic lesions and intrapancreatic metastasis of nonpancreatic malignancies. Two major issues are addressed: first, whether extended central pancreatectomy is safe and effective with regard to early postoperative morbidity and mortality, and, second, the longterm endocrine and exocrine function, recurrence rates, and survival associated with extended central pancreatectomy were compared with extended classic left and right resectional procedures. PATIENTS AND METHODS Patient characteristics. Between 1992 and 2007, 1,780 patients undergoing operations for pancreatic disorders were included in a prospective database; 742 patients underwent major resectional procedures, such as classic pylorus-preserving pancreatoduodenectomy, distal pancreatectomy with splenectomy, or total pancreatectomy, including 200 patients who underwent extended proximal pancreatoduodenectomy and extended distal pancreatectomy with splenectomy. In 620 patients, organ-preserving operations for chronic pancreatitis; benign, borderline, and other nonmalignant diseases; and metastasis of extrapancreatic malignancies were performed. In all, 105 patients had a size $5 cm, whereas 71 had lesions <5 cm. Among the latter group, 25 patients with small, peripheral neoplasms underwent simple enucleation, whereas in 46 patients, because of an involvement of the main pancreatic duct, more extensive procedures, such as pancreatoduodenectomy, distal pancreatectomy with splenectomy, and central pancreatectomy were performed. Among the subset of 105 patients with a tumor size $5 cm, 35 patients underwent extended central pancreatectomy.
Surgery March 2010
The diagnosis of focal pancreatic lesions was established by computed tomography (CT) or magnetic resonance imaging (MRI). Magnetic resonance cholangiopancreaticography (MRCP) or endoscopic retrograde cholangiopancreaticography (ERCP) were obtained routinely to evaluate concomitant ductal irregularities, especially in the pancreatic head (see below). In contrast to lesions highly suspicious for primary pancreatic or periampullary cancer, our institutional policy, before subjecting patients to central pancreatectomy, consisted of performing routinely an endoscopic ultrasonography (EUS)-guided biopsy of the mass, as reported previously.18,19 Contra-indications to central pancreatic resection included biopsy-proven pancreatic ductal adenocarcinoma. The indication for extended central pancreatectomy for treatment of malignancies was restricted to the metastasis of extrapancreatic malignancies. Definition of extended central pancreatectomy, extended classic right resection, and extended classic left resection. Our institutional differentiation of ‘‘conventional’’ central pancreatectomy from extended central pancreatectomy is based on the transverse size of the pancreatic defect after tumor resection. In the event of central lesions with a tumor size of $5 cm that, because of a safety margin of 1 cm at the cut surfaces resulted in a transversal defect of 7 cm or more, the term ‘‘extended central pancreatectomy’’ was used. In contrast, any segmental resection with a transversal defect of <7 cm was termed ‘‘conventional central pancreatectomy.’’ The extended classic right-sided resection was defined as any classic or pylorus-preserving pancreatoduodenectomy with a pancreatic transection margin $3 cm to the left of the superior mesenteric vein. Extended classic left-sided resection was defined as a right-sided extension of distal pancreatectomy with the pancreatic transection margin located greater than $1.5 cm to the right of the superior mesenteric vein. Indications. Until 2002, no clear institutional attitude existed regarding intra-operative decision making about whether to perform extended central pancreatectomy or to subject patients to extended right or left resections. From April 2002 on, we standardized our institutional policy and performed extended central pancreatectomy for appropriately selected central lesions. For predominantly left-sided lesions, a tail segment of >2.5 cm could be preserved; for predominantly right-sided lesions, the duodenal blood supply and the distal common bile duct were preserved.
Surgery Volume 147, Number 3
Overall, 35 patients underwent extended central pancreatectomy; 16 were operated before 2002 and 19 were operated afterward. Technique of central pancreatectomy. After transection of the gastrocolic ligament and the lesser sac, the anterior surface of the pancreas and the lesion were exposed. Where necessary, patients underwent intraoperative ultrasonography to verify the relationship of the lesion to retropancreatic vascular structures. Usually, a Kocher maneuver was performed, especially when the pathology extended to the right of the mesenterico-portal axis. The posterior surface of the pancreas was dissected off the mesenterico-portal axis. At its superior aspect, the pancreas was freed from the splenic vein and artery (to the left) and from the hepatic artery (to the right). A transection of the gland was performed with a scalpel toward the pancreatic tail and usually with a linear stapler toward the head of the gland (Fig 1). In case of a thick gland or a diameter of the main pancreatic duct $4 mm, the cephalic portion was transected with the scalpel. Frozen sections were performed routinely of the resected neoplasm to confirm appropriate pathology for a central pancreatectomy and of the operative margins to confirm the lack of histopathologic involvement. For reconstruction, a retrocolic, Roux-en-Y limb of jeujunum was used. First, the distal pancreatic remnant was anastomosed to the jejunum using a running, end-to-side technique using monofilament sutures (Monocryl 4-0; Ethicon, Somerville, NJ). Management of the pancreatic head remnant depended on several aspects; when the proximal transection had been performed with a linear stapler, the capsule of the pancreatic head remnant was simply covered with the serosa of the Rouxen-Y limb, which was used also for the distal pancreaticojejunal anastomosis. For patients in whom stapler transection was not feasible or in whom preoperative MRCP or ERCP showed a dilated main pancreatic duct of $4 mm, a second, running, end-to-side anastomosis between the pancreatic head remnant and the jejunumn just distal to the pancreatic jejunostomy to the distal pancreatic remnant was performed. Alimentary continuity was reconstructed by an end-to-side Roux-en-Y jejuno-jejunostomy, approximately 30 cm distal to the ligament of Treitz. Routinely, 2 abdominal drains were placed close to the pancreatic anastomosis. Peri-operative and long-term outcome parameters. Informed consent was obtained from all patients. Peri-operative outcome parameters included operating time, postpancreatectomy hemorrhage
Cataldegirmen et al 333
(PPH), pancreatic fistula, intra-abdominal abscess/ infection, insufficiency of the pancreatic anastomosis, need for reoperation, and postoperative duration of hospital stay, respectively. For pancreatic fistula, the definition of the International Study Group for Pancreatic Fistula was adopted, with a drain output of any measurable volume of fluid on or after the third postoperative day with an amylase content greater than 3 times the serum amylase activity.21 PPH was adopted to the definition of the International Study Group for Pancreatic Surgery.37 Follow-up was performed by interviewing the patient’s general practitioners or by interviewing the patients in our institution on an outpatient basis. Exocrine pancreatic function was assessed by fecal chymotrypsin levels (normal >40 mg/g feces, pathologic <40 mg/g feces) and the pancreolauryl test (normal >30%, intermediate 20--30%, pathologic <20%). In all patients who were not on oral antidiabetic agents or insulin, an oral glucose tolerance test was performed, and the results were classified as normal or representative of diabetes mellitus according to the criteria set forth by the German Diabetes Society in 2002.20 STATISTICAL ANALYSIS We used SPSS for Windows (SPSS Inc., Chicago, IL) for statistical analysis. Associations between categorical variables were assessed by the Fisher exact test. The Kaplan-Meier method was used to estimate the occurrence probability of an event, including death, relapse, local recurrence, and observation of distant metastasis. Point and interval estimates of survival rates at 60 months were calculated. For comparison purposes, log-rank tests and exact stratified log-rank tests were performed. Significance statements refer to P values of 2-tailed tests that are less than .05. Matched-pair analysis. To compare peri-operative valuables and long-term outcome after extended central pancreatectomy, the database was checked for all patients undergoing extended classic resections. Among a total of 200 patients undergoing extended classic procedures, 70 patients, who were distributed equally to extended classic right resection and extended classic left resection groups (n = 35, each), were identified and matched with respect to age, sex, and histopathology to the extended central pancreatectomy group. Patients in whom extended left or right pancreatectomy was performed based on oncologic reasons were excluded from matched-pair analysis. The reason for this exclusion was that in
334 Cataldegirmen et al
Surgery March 2010
Fig 1. Example of an appropriate lesion located in the body of the pancreas (left) and schematic situs after the resection of the central lesion (right).
such patients, central pancreatectomy was not considered as an alternative to extended oncologic resection. RESULTS Peri-operative results. A total of 105 patients (50 male and 55 female) underwent either extended central pancreatectomy, extended classic right resection, or extended classic left resection (n = 35, each). The median age at time of operation was 58 years (range, 29--88). Definitive histopathology and intra-operative frozen sections excluded malignant pancreatic primaries and confirmed clear resection margins. The median follow-up time was 48 months (range, 3--156). The management of the pancreatic head remnant after extended central pancreatectomy in 19 patients (54%) involved stapled closure followed by covering the proximal remnant with the serosal aspect of the Roux-en-Y limb distal to the left-sided pancreaticojejunostomy. In contrast, 16 (46%) patients underwent a second, more distal end-toside anastomosis to the pancreatic head stump, either because of the thickness of the transection site in the pancreatic head or because the size of the pancreatic duct was $4 mm, or both. There were no differences in pancreatic fistula rates between these 2 groups (6% vs 9%). Table I shows histopathologic findings in patients undergoing extended central pancreatectomy. Most patients had borderline neoplasms (eg, intraductal papillary mucinous neoplasia [IPMN] with high-grade dysplasia, cystic lesions, and neuroendocrine neoplasms, respectively). A near-identical distribution pattern of underlying diseases was found in patients undergoing extended classic right resection and extended classic left resection (Table II). Only 6 patients in the study cohort had metastasis from extrapancreatic malignancies.
The median postoperative duration of hospital stay was less after extended central pancreatectomy than that after extended classic right- and leftsided resections (12 vs 24 and 18 days, respectively). Also, the median operating time and the number of required blood units were less after extended central pancreatectomy versus extended classic right- and left-sided resections (235 vs 350 and 260 min, respectively; 2 vs 1 and 2, respectively; Table II). Postoperative morbidity/mortality. Morbidity rates related to specific complications and mortality rates are shown in Table II. The pancreatic fistula rate after extended central pancreatectomy was only slightly different from that after extended classic right and left resections (14% vs 11% and 17%, respectively; P = not significant). The overall in-hospital morbidity rate after extended central pancreatectomy was similar to that after extended classic right and left resections (26% vs 43% and 37%, respectively; P = .05). Six patients had to be reoperated: 2 because of early and 1 because of late postpancreatectomy hemorrhage, 2 as a result of intra-abdominal abscess and 1 as a result of biliary leakage (Table II). Impact of pancreatic resections on pancreatic function. For the assessment of postoperative functional changes, only patients who had normal endocrine and exocrine function pre-operatively were evaluated, whereas those with abnormal pancreatic function preoperatively were excluded from analysis. Because of this limitation, the analysis of endocrine and exocrine impairment was based on 34 (endocrine function) and 35 (exocrine function) patients with extended central pancreatectomy, 32 and 34 patients with extended classic right resection, and 32 and 34 patients undergoing extended classic left resection (Table II). Impairment of the exocrine
Cataldegirmen et al 335
Surgery Volume 147, Number 3
Table I. Clinical features of the 35 patients undergoing extended central pancreatectomy Sex, male/female Age Histology Cystadenoma Borderline NEN Renal cancer metastasis Breast cancer metastasis Symptoms/pain Exocrine insufficiency Endocrine insufficiency Operative time (min) Transfused blood units Postoperative complications Hemorrhage Pancreatic fistula Pneumonia Management Conservative Drainage Embolization Medical Reoperation Radiologic drainage Median follow-up (months)
17/18 60 (29–76) 11 10 11 2 1 18 1 235 0 9 3 5 1 4 1 1 1 2 1 48
(31%) (29%) (31%) (6%) (3%) (51%) 0 (3%) (110–425) (0–4) (26%) (9%) (14%) (3%) (11%) (3%) (3%) (3%) (6%) (3%) (3–156)
NEN indicates benign neuroendocrine neoplasia and neuroendocrine tumors with high-grade dysplasia. Cystadenoma includes both serous and mucinous tumors. Borderline tumors indicate IPMN and borderline neuroendocrine neoplasia.
function measured by fecal chymotrypsin and pancreolauryl tests was less after extended central pancreatectomy compared with extended classic right and left resections (9% vs 24% and 29%, respectively). The prevalence of operation-related impairment of endocrine function, detected by fasting blood glucose concentration, was less after extended central pancreatectomy than after extended right and left resections (6% vs 28% and 34%, respectively). No difference was found after extended classic right and left resections concerning the impairment of endocrine and exocrine function. The postoperative body weight (BW) was unaffected after extended central pancreatectomy and extended classic left resection, whereas a decrease was observed after extended classic right resection (P < .05; Table II). Tumor recurrence. Tumor recurrence was restricted to patients in whom histopathology confirmed a diagnosis of ‘‘benign’’ or borderline-type neuroendocrine neoplasia and IPMN. After a median follow-up of 48 months, the tumor recurrence rate and incidence of liver metastasis after
extended central pancreatectomy was 17% (n = 6), after extended classic right resection 14% (n = 5), and after extended classic left resection 9% (n = 3; P = not significant). In the extended central pancreatectomy group, all patients with local recurrence underwent reoperation except for 1 patient who showed additional diffuse liver metastasis. The following salvage operations were performed: completion pancreatoduodenectomy (n = 1), classic (n = 3), and pylorus preserving Whipple (n = 1) resection. In 1 patient with a solitary liver metastasis, classic Whipple resection was combined with nonanatomic liver resection (metastasectomy). In the group of patients who underwent extended classic right and extended classic left resections, 2 patients developed liver metastases in the follow-up and underwent nonanatomic hepatic resections. Four patients with local recurrence were subjected to total pancreatoduodenectomy (n = 3) and partial gastrectomy (n = 1). One patient with local recurrence, who did not undergo reoperation died 11 years after the initial operation from a low-grade neuroendocrine carcinoma. Disease-related mortality from nonresectable local recurrence and distant liver metastasis after extended central pancreatectomy was 9% (n = 3), and after extended classic left and right resections, it was 17% (n = 6, each; P = not significant; Fig 2). DISCUSSION The rationale of organ-sparing resection for small benign or borderline lesions and metastases from nonpancreatic malignancies is to preserve as much endocrine and exocrine pancreatic tissue as possible. Surgical studies addressing appropriate lesions with a size of $5 cm are scarce. Depending on the relationship to the mesenterico-portal axis, appropriate lesions >5 cm are usually treated either by an extended Whipple’s resection to the left or by extension of the left pancreatic resection to the right of the superior mesenteric vein. These procedures are burdened with a loss of unaffected pancreatic parenchyma, thereby resulting in a substantially increased incidence of functional impairment of up to 85%.22-26 The major goal of the current study was to compare the perioperative and long-term outcomes of patients undergoing extended central pancreatectomy with those undergoing extended classic resections. Because no clear institutional policy regarding the treatment of such selected lesions existed before 2002, only patients in whom intra-operative decision making was made in favor of extended classic
336 Cataldegirmen et al
Surgery March 2010
Table II. Matched-pairs analysis comparing patients with extended central pancreatectomy, extended right resection, and extended left resection Patient characteristics Median age Female Male Histology Borderline Cystadenoma Benign NEN Metastasis Peri-operative results Median transfused blood units Median operating time (min) Median hospital stay (days) Morbidity Pancreatic fistula Hemorrhage Abscess Biliary leakage Other Overall Mortality Reoperation Pancreatic function{{ Pre-operative pathologic OGTT Postoperative pathologic OGTT Pre-operative exocrine insufficiency Postoperative exocrine insufficiency Body weight (% decrease after surgery)
ECP (n = 35)
ERR (n = 35)
ELR (n = 35)
58 (29–76) 18 17
57 (30–78) 19 16
58 (34–88) 18 17
10 11 11 3
11 11 11 2
11 12 11 1
0 (0–4) 235 (110–425) 12 (7–30)
2 (0–22) 350 (125–600) 24 (11–47)
1 (0–4) 260 (120–570) 18 (7–57)
5 (14%) 3 (9%) — — 1 (3%) 9 (26%) — 1 (3%)
4 2 2 2 5 15 2 3
(11%) (6%) (6%) (6%) (14%) (43%) (6%) (9%)
6 (17%) 2 (6%) 1 (3%) — 4 (11%) 13 (37%) — 2 (6%)
1 (3%) 2 (6%) — 3 (9%) 0.4%
3 9 1 8
(9%) (28%) (3%) (24%) 6.6%
3 11 1 10
(9%) (34%) (3%) (29%) 1.3%
P value
.01* .01y and .02z .01§ and .05k ns
.05{ and .05# ns ns ns .02** and .05yy ns .03zz and .06§§ .05kk
*P = .01 for ERR versus ECP and for ELR versus ECP. yP = .01 for ERR versus ECP and for ERR versus ELR. zP = .02 for ECP versus ELR. §P = .01 for ERR versus ECP. kP = .05 for ECP versus ELR. {P = .05 for ERR versus ECP. #P = .05 for ELR versus ECP. **P = .02 for ERR versus ECP. yyP = .05 for ELR versus ECP. zzP = .03 for ERR versus ECP. §§P = .06 for ELR versus ECP. kkP = .05 for ERR. {{Values indicating postoperative OGTT and exocrine function refer to ‘‘de novo’’ findings, whereas patients with pre-operative derangements (ECP = 1, ECRR = 3, and ECRR = 3) were not considered in functional follow-up investigations. ECP, Extended central pancreatectomy; ELR, extended left resection; ERR, extended right resection; ns, not significant.
resectional procedures based on a questionable frozen section on unpredictable intra-operative findings were selected. This approach ensured that only patients in whom an extended central pancreatectomy represented a potential surgical alternative were included. Central pancreatectomy was first introduced in 1957 and was performed in a patient with chronic pancreatitis.13 In truth, the Beger procedure, which represents one of the milestones in the surgical armamentarium for chronic pancreatitis, is a modified right-sided central pancreatectomy
characterized by the complete transection of the pancreas, hereby resulting in 2 cut surfaces to be drained enterically.27,28 Central pancreatectomy has become the procedure of choice for small benign and borderline lesions located in the pancreatic body and neck. It has also been suggested for low-grade malignancies and metastases from extrapancreatic malignancies.19 To identify patients in whom classic resection would constitute operative ‘‘overtreatment’’ and, in turn, to protect those patients from inadequate organ-sparing pancreatic
Cataldegirmen et al 337
Surgery Volume 147, Number 3
1,00
Central vs. Right p=0.35 Central vs. Left p=0.296 Rght vs. Left p=0.516i
0,95 0,90
Cum Survival
0,85
Extended Central Pancreatic Resection
0,80 0,75 0,70
Extended Left Resection
0,65 0,60 Extended Right Resection
0,55 0,50 0
12 24 36 48 60 72 84 96 108 120 132 144 156 168 180
OverallSurvival in Months
Fig 2. Kaplan-Meier overall cumulative survival curves comparing patients with extended central pancreatic resection, extended left resection, and extended right resection (n = 35, each).
resection who are likely to have malignant lesions, preoperative evaluation using endoscopic, ultrasonography-guided, fine-needle aspiration biopsy is mandatory. Staging accuracy by endoscopic, ultrasonography-guided, fine-needle aspiration in the diagnostic workup of pancreatic lesions has improved considerably; its sensitivity in diagnosing questionable pancreatic lesions correctly is reported to be greater than 80%.18,19 Nonetheless, a gap remains between ‘‘formally’’ accepted indications for central pancreatectomy and the frequency of its use in the clinical routine. This gap is reflected by the fact that only 600 patients have reportedly undergone central pancreatectomy, and only 5 published series have a sample size exceeding 30 patients.9,10,29-32 Central pancreatectomy is usually performed for lesions with a transverse size smaller than 3 cm. So far, institutional studies scrutinizing the outcome of patients subjected to extended central pancreatectomy for appropriate lesions with a minimal size of $5 cm that were addressed in the presented series have not been published. Besides technical aspects, one may speculate that the major reason for the reluctance to perform central pancreatectomy derives from concerns regarding the accuracy of a preoperative histopathologic diagnosis and the inherent risk of ‘‘false negative’’ findings. An inadequate resection of a malignant lesion that has been identified preoperatively as a benign or borderline lesion is a well-known oncologic pitfall inherent to all organpreserving procedures. This possibility is especially pertinent to IPMNs, mucinous cystadenomas, and
neuroendocrine neoplasms, which may harbor malignant islets of tumor within ‘‘benign’’ or ‘‘borderline’’ lesions. Even frozen sections performed during the operation carry a considerable risk of overlooking invasive, malignant areas in such neoplasms, as was the case in 1 patient in the presented series. Another aspect contributing to the reluctant surgical attitude regarding central pancreatectomy is the assumption that it is associated with an increased incidence of pancreatic fistula. The procedure results in 2 pancreatic stumps of which at least the left side has to be anastomosed, whereas the cephalic remnant may be stapled, suture ligated, or covered with the serosal aspect of the distal jejunal limb that is used for the left-sided remnant. In the presented series, the prevalence of pancreatic fistula (14%) after extended central pancreatectomy did not differ from the incidence after extended left pancreatectomy (17%) and extended right pancreatectomy (11%). Nonetheless, despite the slightly greater incidence of pancreatic fistula after extended central pancreatectomy compared with extended classic left resection, substantial benefits regarding overall perioperative morbidity, median operative time, and need for blood transfusions have been found in comparison with classic resectional procedures. We also report on improved functional results after extended central pancreatectomy compared with extended classic right and left resection. The development of diabetes mellitus, exocrine insufficiency, and weight loss are leading causes of longterm morbidity of classic resectional procedures, such as pancreatoduodenal and distal splenopancreatectomy. The findings in this series demonstrate that postoperative endocrine function after extended central pancreatectomy was better than after extended classic right resection and extended classic left resections. These findings are consistent with previous studies.22,23 Significant benefits were also found in terms of protection from exocrine insufficiency in favor of extended central pancreatectomy compared with extended classic right resection and extended classic left resection. These findings are consistent with previous studies reporting the maintenance of exocrine function after organ-sparing pancreatic resection.23 Fecal chymotrypsin and pancreolauryl levels after extended central pancreatectomy were substantially greater than after extended classic right resection and extended classic left resection. Moreover, when compared with extended classic right resection, the functional benefits of extended central pancreatectomy resulted in a substantial protection from body weight loss.
338 Cataldegirmen et al
In conclusion, extended central pancreatectomy is a safe and technically feasible option for the operative treatment of benign, borderline, or low-grade malignant neoplasms of the pancreas. It is not associated with an increased likelihood of pancreatic fistula as compared with extended classic left resection, but overall, postoperative morbidity is less than after extended standard resection. The risk of developing diabetes mellitus and exocrine insufficiency is negligible. Extended central pancreatectomy should be considered as a less invasive alternative to extended classic resection in appropriately selected patients. REFERENCES 1. Craighead CC, Lien RC. Pancreatoduodenal resection; comments on indications, operative diagnosis, staged procedures, morbid and lethal factors, and survival. Ann Surg 1958;147:931-4. 2. Lillemoe KD, Kaushal S, Cameron JL, Sohn TA, Pitt HA, Yeo CJ. Distal pancreatectomy: indications and outcomes in 235 patients. Ann Surg 1999;229:693-8. 3. Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248-60. 4. Neoptolemos JP, Russell RC, Bramhall S, Theis B. Low mortality following resection for pancreatic and periampullary tumors in 1026 patients: UK survey of specialist pancreatic units. UK Pancreatic Cancer Group. Br J Surg 1997;84: 1370-6. 5. Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006;244:10-5. 6. Brennan FM, Kattan MW, Klimstra D, Conlon K. Prognostic nomogram for patients undergoing resections for adenocarcinoma of the pancreas. Ann Surg 2004;240:293-8. 7. Tien YW, Lee PH, Yang CY, Ho MC, Chiu YF. Risk factors of massive bleeding related to pancreatic leak after pancreaticoduodenectomy. J Am Coll Surg 2005;201:554-9. 8. Gouma DJ, van Geenen RC, van Gulik TM, de Haan RJ, de Wit LT, Busch OR, et al. Rates of complication and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000; 232:786-95. 9. Mu¨ller MW, Friess H, Kleeff J, Hinz U, Wente MN, Paramythiotis D, et al. Middle segmental pancreatic resection: an option to treat benign pancreatic body lesions. Ann Surg 2006;233:909-20. 10. Sauvanet A, Partensky C, Sastre B, Gigot JF, Fagniez PL, Tuech JJ, et al. Medial pancreatectomy: a multi-institutional retrospective study of 53 patients by the French Pancreas Club. Surgery 2002;132:836-43. 11. Carrere N, Abid S, Julio CH, Bloom E, Prade`re B. Spleenpreserving distal pancreatectomy with excision of splenic artery and vein: a case-matched comparison with conventional distal pancreatectomy with splenectomy. World J Surg 2007; 31:375-82. 12. Shoup M, Brennan MF, McWhite K, Leung DH, Klimstra D, Conlon KC. The value of splenic preservation with distal pancreatectomy. Arch Surg 2002;137:164-8.
Surgery March 2010
13. Guillemin P, Bessot M. Chronic calcifying pancreatitis in renal tuberculosis: pancreatojejunostomy using an original technique. Mem Acad Chir (Paris) 1957;83:869-71. 14. Letton AH, Wilson JP. Traumatic severance of pancreas treated by Roux-Y anastomosis. Surg Gynecol Obstet 1959;109:473-8. 15. Warshaw AL, Rattner DW, Ferna´ndez-del Castillo C, Z’graggen K. Middle segment pancreatectomy: a novel technique for conserving pancreatic tissue. Arch Surg 1998;133:327-31. 16. Iacono C, Bortolasi L, Serio G. Indications and technique of central pancreatectomy---early and late results. Langenbecks Arch Surg 2005;390:266-71. 17. Sperti C, Pasquali C, Ferronato A, Pedrazzoli S. Median pancreatectomy for tumors of the neck and body of the pancreas. J Am Coll Surg 2000;190:711-6. 18. Fritscher-Ravens A, Izbicki JR, Sriram PV, Krause C, Knoefel WT, Topalidis T, et al. Endosonography-guided, fine-needle aspiration cytology extending the indication for organ-preserving pancreatic surgery. Am J Gastroenterol 2000;95: 2255-60. 19. Schurr PG, Strate T, Rese K, Kaifi JT, Reichelt U, Petri S, et al. Aggressive surgery improves long-term survival in neuroendocrine pancreatic tumors: an institutional experience. Ann Surg 2007;245:273-81. 20. Brueckel J, Koebberling J. Definition, klassifikation und diagnostik des diabetes mellitus. Diabetes Stoffwechsel 2002; 11:6-39. 21. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J. International Study Group on Pancreatic Fistula Definition. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8-13. 22. Kendall DM, Sutherland DE, Najarian JS, Goetz FC, Robertson RP. Effects of hemipancreatectomy on insulin secretion and glucose tolerance in healthy humans. N Engl J Med 1990;322:898-903. 23. Wittingen J, Frey CF. Islet concentration in the head, body, tail and uncinate process of the pancreas. Ann Surg 1974; 179:412-4. 24. Sperti C, Pasquali C, Ferronato A, Pedrazzoli S. Median pancreatectomy for tumors of the neck and body of the pancreas. J Am Coll Surg 2000;190:711-6. 25. Yasuda H, Takada T, Toyota N, Amano H, Yoshida M, Takada Y, et al. Limited pancreatectomy: significance of postoperative maintenance of pancreatic exocrine function. J Hepatobiliary Pancreat Surg 2000;7:466-72. 26. Warshaw AL, Rattner DW, Ferna´ndez-del Castillo C, Z’graggen K. Middle segment pancreatectomy: a novel technique for conserving pancreatic tissue. Arch Surg 1998;133:327-31. 27. Beger HG, Bu¨chler M, Bittner RR, Oettinger W, Roscher R. Duodenum-preserving resection of the head of the pancreas in severe chronic pancreatitis. Early and late results. Ann Surg 1989;209:272-8. 28. Frey CF, Child CG, Fry W. Pancreatectomy for chronic pancreatitis. Ann Surg 1976;184:403-13. 29. Adham M, Giunippero A, Hervieu V, Courbie`re M, Partensky C. Central pancreatectomy: single-center experience of 50 cases. Arch Surg 2008;143:175-81. 30. Balzano G, Zerbi A, Veronesi P, Cristallo M, Di Carlo V. Surgical treatment of benign and borderline neoplasms of the pancreatic body. Dig Surg 2003;20:506-10. 31. Crippa S, Bassi C, Warshaw AL, Falconi M, Partelli S, Thayer SP, et al. Middle pancreatectomy: indications, short- and long-term operative outcomes. Ann Surg 2007;246:69-76. 32. Wente MN, Veit JA, Bassi C. Postpancreatectomy hemorrhage (PPH)-An International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007;142:20-5.
Background. Whether patients with focal pancreatic lesions of benign or borderline pathology should be treated by extended central pancreatectomy rather than by extended classic resectional procedures, such as extended right and left resections, is controversial. Methods. Between 1992 and 2007, 105 patients underwent operation for focal pancreatic lesions of borderline or benign neuroendocrine neoplasms, cystadenoma, intraductal papillary mucinous neoplasia (IPMN), and secondary metastasis. In all, 35 patients were subjected to extended central pancreatectomy, whereas the remaining 70 patients were treated by an extended classic right resection or an extended classic left resection. Groups were matched according to age, sex, and histopathology. Results. No peri-operative mortality occurred after extended central pancreatectomy and extended classic left resection (n = 35, each). Two (6%) patients died after extended classic right resection. Overall, inhospital morbidity was 26% after extended central pancreatectomy, 43% after extended classic right resection, and 37% after extended classic left resection. After a median follow-up of 48 months, a local recurrence rate of 17% after extended central pancreatectomy was similar to the corresponding rates of 9% after extended classic left resection and 14% after extended classic right resection. Endocrine and exocrine impairment was less pronounced after extended central pancreatectomy (6% and 9%) than after extended classic left resection (34% and 29%) and extended classic right resection (28% and 24%; P < .05). Conclusion. Extended central pancreatectomy for appropriate pancreatic neoplasms is associated with less peri-operative morbidity and mortality than after extended classic left and extended classic right resection. Long-term local recurrence after extended central pancreatectomy is similar to the recurrence rates after extended classic right and classic left resection. Our results suggest that appropriately selected patients will benefit from extended central pancreatectomy because of the maintenance of endocrine and exocrine function. (Surgery 2010;147:331-8.) From the Department of General, Viszeral and Thoracic Surgery, University Medical Center HamburgEppendorf, University of Hamburg, Hamburg, Germany
ALTHOUGH MORTALITY AFTER MAJOR PANCREATIC SURGERY has decreased to less than 3%,1-4 morbidity remains considerably high, ranging from 18% to 52%.4-8 Organ-sparing pancreatic resections, such as enucleation and central pancreatic resection, are suggested as suitable procedures for borderline Accepted for publication October 6, 2009. Reprint requests: Emre F. Yekebas, MD, Department of General, Visceral, and Thoracic Surgery, University Medical Centre Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2010 Mosby, Inc. All rights reserved. doi:10.1016/j.surg.2009.10.027
and benign neuroendocrine neoplasms located in the neck and body of the gland. These types of resections offer an alternative to extended classic resections with the main advantage of preserving unaffected pancreatic parenchyma.9,10 Subsequent arguments in favor of central pancreatectomy are the preservation of the spleen, thereby minimizing the splenectomy-associated morbidity related to infectious and/or thrombotic complications, which can occur when performing standard resections to the left of the mesenterico-portal axis. The preservation of both gastroduodenal and biliary continuity is an advantage of central pancreatectomy in patients with appropriately selected right-sided lesions.11,12 SURGERY 331
332 Cataldegirmen et al
Since the first publication by Guillemin and Bessot in 1957,13 central pancreatectomy has been applied increasingly to benign and borderline lesions of the pancreas neck and body.14 Nonetheless, surgeons’ attitude toward central pancreatectomy is often characterized by reluctance,9,10,15-17 which is derived from doubts concerning several aspects of the procedure. First, complete tumor resection may require special technical expertise limited to specialized centers. Second, because of the limitations of current staging examinations, the accuracy of preoperative diagnosis is often questioned. Last, because central pancreatectomy results in 2 pancreatic stumps that must be dealt with, this procedure is assumed to carry a greater risk of operative morbidity associated with pancreatic fistula. Our institutional experience reflects the value of extended central pancreatectomy for treatment of nonmalignant pancreatic lesions and intrapancreatic metastasis of nonpancreatic malignancies. Two major issues are addressed: first, whether extended central pancreatectomy is safe and effective with regard to early postoperative morbidity and mortality, and, second, the longterm endocrine and exocrine function, recurrence rates, and survival associated with extended central pancreatectomy were compared with extended classic left and right resectional procedures. PATIENTS AND METHODS Patient characteristics. Between 1992 and 2007, 1,780 patients undergoing operations for pancreatic disorders were included in a prospective database; 742 patients underwent major resectional procedures, such as classic pylorus-preserving pancreatoduodenectomy, distal pancreatectomy with splenectomy, or total pancreatectomy, including 200 patients who underwent extended proximal pancreatoduodenectomy and extended distal pancreatectomy with splenectomy. In 620 patients, organ-preserving operations for chronic pancreatitis; benign, borderline, and other nonmalignant diseases; and metastasis of extrapancreatic malignancies were performed. In all, 105 patients had a size $5 cm, whereas 71 had lesions <5 cm. Among the latter group, 25 patients with small, peripheral neoplasms underwent simple enucleation, whereas in 46 patients, because of an involvement of the main pancreatic duct, more extensive procedures, such as pancreatoduodenectomy, distal pancreatectomy with splenectomy, and central pancreatectomy were performed. Among the subset of 105 patients with a tumor size $5 cm, 35 patients underwent extended central pancreatectomy.
Surgery March 2010
The diagnosis of focal pancreatic lesions was established by computed tomography (CT) or magnetic resonance imaging (MRI). Magnetic resonance cholangiopancreaticography (MRCP) or endoscopic retrograde cholangiopancreaticography (ERCP) were obtained routinely to evaluate concomitant ductal irregularities, especially in the pancreatic head (see below). In contrast to lesions highly suspicious for primary pancreatic or periampullary cancer, our institutional policy, before subjecting patients to central pancreatectomy, consisted of performing routinely an endoscopic ultrasonography (EUS)-guided biopsy of the mass, as reported previously.18,19 Contra-indications to central pancreatic resection included biopsy-proven pancreatic ductal adenocarcinoma. The indication for extended central pancreatectomy for treatment of malignancies was restricted to the metastasis of extrapancreatic malignancies. Definition of extended central pancreatectomy, extended classic right resection, and extended classic left resection. Our institutional differentiation of ‘‘conventional’’ central pancreatectomy from extended central pancreatectomy is based on the transverse size of the pancreatic defect after tumor resection. In the event of central lesions with a tumor size of $5 cm that, because of a safety margin of 1 cm at the cut surfaces resulted in a transversal defect of 7 cm or more, the term ‘‘extended central pancreatectomy’’ was used. In contrast, any segmental resection with a transversal defect of <7 cm was termed ‘‘conventional central pancreatectomy.’’ The extended classic right-sided resection was defined as any classic or pylorus-preserving pancreatoduodenectomy with a pancreatic transection margin $3 cm to the left of the superior mesenteric vein. Extended classic left-sided resection was defined as a right-sided extension of distal pancreatectomy with the pancreatic transection margin located greater than $1.5 cm to the right of the superior mesenteric vein. Indications. Until 2002, no clear institutional attitude existed regarding intra-operative decision making about whether to perform extended central pancreatectomy or to subject patients to extended right or left resections. From April 2002 on, we standardized our institutional policy and performed extended central pancreatectomy for appropriately selected central lesions. For predominantly left-sided lesions, a tail segment of >2.5 cm could be preserved; for predominantly right-sided lesions, the duodenal blood supply and the distal common bile duct were preserved.
Surgery Volume 147, Number 3
Overall, 35 patients underwent extended central pancreatectomy; 16 were operated before 2002 and 19 were operated afterward. Technique of central pancreatectomy. After transection of the gastrocolic ligament and the lesser sac, the anterior surface of the pancreas and the lesion were exposed. Where necessary, patients underwent intraoperative ultrasonography to verify the relationship of the lesion to retropancreatic vascular structures. Usually, a Kocher maneuver was performed, especially when the pathology extended to the right of the mesenterico-portal axis. The posterior surface of the pancreas was dissected off the mesenterico-portal axis. At its superior aspect, the pancreas was freed from the splenic vein and artery (to the left) and from the hepatic artery (to the right). A transection of the gland was performed with a scalpel toward the pancreatic tail and usually with a linear stapler toward the head of the gland (Fig 1). In case of a thick gland or a diameter of the main pancreatic duct $4 mm, the cephalic portion was transected with the scalpel. Frozen sections were performed routinely of the resected neoplasm to confirm appropriate pathology for a central pancreatectomy and of the operative margins to confirm the lack of histopathologic involvement. For reconstruction, a retrocolic, Roux-en-Y limb of jeujunum was used. First, the distal pancreatic remnant was anastomosed to the jejunum using a running, end-to-side technique using monofilament sutures (Monocryl 4-0; Ethicon, Somerville, NJ). Management of the pancreatic head remnant depended on several aspects; when the proximal transection had been performed with a linear stapler, the capsule of the pancreatic head remnant was simply covered with the serosa of the Rouxen-Y limb, which was used also for the distal pancreaticojejunal anastomosis. For patients in whom stapler transection was not feasible or in whom preoperative MRCP or ERCP showed a dilated main pancreatic duct of $4 mm, a second, running, end-to-side anastomosis between the pancreatic head remnant and the jejunumn just distal to the pancreatic jejunostomy to the distal pancreatic remnant was performed. Alimentary continuity was reconstructed by an end-to-side Roux-en-Y jejuno-jejunostomy, approximately 30 cm distal to the ligament of Treitz. Routinely, 2 abdominal drains were placed close to the pancreatic anastomosis. Peri-operative and long-term outcome parameters. Informed consent was obtained from all patients. Peri-operative outcome parameters included operating time, postpancreatectomy hemorrhage
Cataldegirmen et al 333
(PPH), pancreatic fistula, intra-abdominal abscess/ infection, insufficiency of the pancreatic anastomosis, need for reoperation, and postoperative duration of hospital stay, respectively. For pancreatic fistula, the definition of the International Study Group for Pancreatic Fistula was adopted, with a drain output of any measurable volume of fluid on or after the third postoperative day with an amylase content greater than 3 times the serum amylase activity.21 PPH was adopted to the definition of the International Study Group for Pancreatic Surgery.37 Follow-up was performed by interviewing the patient’s general practitioners or by interviewing the patients in our institution on an outpatient basis. Exocrine pancreatic function was assessed by fecal chymotrypsin levels (normal >40 mg/g feces, pathologic <40 mg/g feces) and the pancreolauryl test (normal >30%, intermediate 20--30%, pathologic <20%). In all patients who were not on oral antidiabetic agents or insulin, an oral glucose tolerance test was performed, and the results were classified as normal or representative of diabetes mellitus according to the criteria set forth by the German Diabetes Society in 2002.20 STATISTICAL ANALYSIS We used SPSS for Windows (SPSS Inc., Chicago, IL) for statistical analysis. Associations between categorical variables were assessed by the Fisher exact test. The Kaplan-Meier method was used to estimate the occurrence probability of an event, including death, relapse, local recurrence, and observation of distant metastasis. Point and interval estimates of survival rates at 60 months were calculated. For comparison purposes, log-rank tests and exact stratified log-rank tests were performed. Significance statements refer to P values of 2-tailed tests that are less than .05. Matched-pair analysis. To compare peri-operative valuables and long-term outcome after extended central pancreatectomy, the database was checked for all patients undergoing extended classic resections. Among a total of 200 patients undergoing extended classic procedures, 70 patients, who were distributed equally to extended classic right resection and extended classic left resection groups (n = 35, each), were identified and matched with respect to age, sex, and histopathology to the extended central pancreatectomy group. Patients in whom extended left or right pancreatectomy was performed based on oncologic reasons were excluded from matched-pair analysis. The reason for this exclusion was that in
334 Cataldegirmen et al
Surgery March 2010
Fig 1. Example of an appropriate lesion located in the body of the pancreas (left) and schematic situs after the resection of the central lesion (right).
such patients, central pancreatectomy was not considered as an alternative to extended oncologic resection. RESULTS Peri-operative results. A total of 105 patients (50 male and 55 female) underwent either extended central pancreatectomy, extended classic right resection, or extended classic left resection (n = 35, each). The median age at time of operation was 58 years (range, 29--88). Definitive histopathology and intra-operative frozen sections excluded malignant pancreatic primaries and confirmed clear resection margins. The median follow-up time was 48 months (range, 3--156). The management of the pancreatic head remnant after extended central pancreatectomy in 19 patients (54%) involved stapled closure followed by covering the proximal remnant with the serosal aspect of the Roux-en-Y limb distal to the left-sided pancreaticojejunostomy. In contrast, 16 (46%) patients underwent a second, more distal end-toside anastomosis to the pancreatic head stump, either because of the thickness of the transection site in the pancreatic head or because the size of the pancreatic duct was $4 mm, or both. There were no differences in pancreatic fistula rates between these 2 groups (6% vs 9%). Table I shows histopathologic findings in patients undergoing extended central pancreatectomy. Most patients had borderline neoplasms (eg, intraductal papillary mucinous neoplasia [IPMN] with high-grade dysplasia, cystic lesions, and neuroendocrine neoplasms, respectively). A near-identical distribution pattern of underlying diseases was found in patients undergoing extended classic right resection and extended classic left resection (Table II). Only 6 patients in the study cohort had metastasis from extrapancreatic malignancies.
The median postoperative duration of hospital stay was less after extended central pancreatectomy than that after extended classic right- and leftsided resections (12 vs 24 and 18 days, respectively). Also, the median operating time and the number of required blood units were less after extended central pancreatectomy versus extended classic right- and left-sided resections (235 vs 350 and 260 min, respectively; 2 vs 1 and 2, respectively; Table II). Postoperative morbidity/mortality. Morbidity rates related to specific complications and mortality rates are shown in Table II. The pancreatic fistula rate after extended central pancreatectomy was only slightly different from that after extended classic right and left resections (14% vs 11% and 17%, respectively; P = not significant). The overall in-hospital morbidity rate after extended central pancreatectomy was similar to that after extended classic right and left resections (26% vs 43% and 37%, respectively; P = .05). Six patients had to be reoperated: 2 because of early and 1 because of late postpancreatectomy hemorrhage, 2 as a result of intra-abdominal abscess and 1 as a result of biliary leakage (Table II). Impact of pancreatic resections on pancreatic function. For the assessment of postoperative functional changes, only patients who had normal endocrine and exocrine function pre-operatively were evaluated, whereas those with abnormal pancreatic function preoperatively were excluded from analysis. Because of this limitation, the analysis of endocrine and exocrine impairment was based on 34 (endocrine function) and 35 (exocrine function) patients with extended central pancreatectomy, 32 and 34 patients with extended classic right resection, and 32 and 34 patients undergoing extended classic left resection (Table II). Impairment of the exocrine
Cataldegirmen et al 335
Surgery Volume 147, Number 3
Table I. Clinical features of the 35 patients undergoing extended central pancreatectomy Sex, male/female Age Histology Cystadenoma Borderline NEN Renal cancer metastasis Breast cancer metastasis Symptoms/pain Exocrine insufficiency Endocrine insufficiency Operative time (min) Transfused blood units Postoperative complications Hemorrhage Pancreatic fistula Pneumonia Management Conservative Drainage Embolization Medical Reoperation Radiologic drainage Median follow-up (months)
17/18 60 (29–76) 11 10 11 2 1 18 1 235 0 9 3 5 1 4 1 1 1 2 1 48
(31%) (29%) (31%) (6%) (3%) (51%) 0 (3%) (110–425) (0–4) (26%) (9%) (14%) (3%) (11%) (3%) (3%) (3%) (6%) (3%) (3–156)
NEN indicates benign neuroendocrine neoplasia and neuroendocrine tumors with high-grade dysplasia. Cystadenoma includes both serous and mucinous tumors. Borderline tumors indicate IPMN and borderline neuroendocrine neoplasia.
function measured by fecal chymotrypsin and pancreolauryl tests was less after extended central pancreatectomy compared with extended classic right and left resections (9% vs 24% and 29%, respectively). The prevalence of operation-related impairment of endocrine function, detected by fasting blood glucose concentration, was less after extended central pancreatectomy than after extended right and left resections (6% vs 28% and 34%, respectively). No difference was found after extended classic right and left resections concerning the impairment of endocrine and exocrine function. The postoperative body weight (BW) was unaffected after extended central pancreatectomy and extended classic left resection, whereas a decrease was observed after extended classic right resection (P < .05; Table II). Tumor recurrence. Tumor recurrence was restricted to patients in whom histopathology confirmed a diagnosis of ‘‘benign’’ or borderline-type neuroendocrine neoplasia and IPMN. After a median follow-up of 48 months, the tumor recurrence rate and incidence of liver metastasis after
extended central pancreatectomy was 17% (n = 6), after extended classic right resection 14% (n = 5), and after extended classic left resection 9% (n = 3; P = not significant). In the extended central pancreatectomy group, all patients with local recurrence underwent reoperation except for 1 patient who showed additional diffuse liver metastasis. The following salvage operations were performed: completion pancreatoduodenectomy (n = 1), classic (n = 3), and pylorus preserving Whipple (n = 1) resection. In 1 patient with a solitary liver metastasis, classic Whipple resection was combined with nonanatomic liver resection (metastasectomy). In the group of patients who underwent extended classic right and extended classic left resections, 2 patients developed liver metastases in the follow-up and underwent nonanatomic hepatic resections. Four patients with local recurrence were subjected to total pancreatoduodenectomy (n = 3) and partial gastrectomy (n = 1). One patient with local recurrence, who did not undergo reoperation died 11 years after the initial operation from a low-grade neuroendocrine carcinoma. Disease-related mortality from nonresectable local recurrence and distant liver metastasis after extended central pancreatectomy was 9% (n = 3), and after extended classic left and right resections, it was 17% (n = 6, each; P = not significant; Fig 2). DISCUSSION The rationale of organ-sparing resection for small benign or borderline lesions and metastases from nonpancreatic malignancies is to preserve as much endocrine and exocrine pancreatic tissue as possible. Surgical studies addressing appropriate lesions with a size of $5 cm are scarce. Depending on the relationship to the mesenterico-portal axis, appropriate lesions >5 cm are usually treated either by an extended Whipple’s resection to the left or by extension of the left pancreatic resection to the right of the superior mesenteric vein. These procedures are burdened with a loss of unaffected pancreatic parenchyma, thereby resulting in a substantially increased incidence of functional impairment of up to 85%.22-26 The major goal of the current study was to compare the perioperative and long-term outcomes of patients undergoing extended central pancreatectomy with those undergoing extended classic resections. Because no clear institutional policy regarding the treatment of such selected lesions existed before 2002, only patients in whom intra-operative decision making was made in favor of extended classic
336 Cataldegirmen et al
Surgery March 2010
Table II. Matched-pairs analysis comparing patients with extended central pancreatectomy, extended right resection, and extended left resection Patient characteristics Median age Female Male Histology Borderline Cystadenoma Benign NEN Metastasis Peri-operative results Median transfused blood units Median operating time (min) Median hospital stay (days) Morbidity Pancreatic fistula Hemorrhage Abscess Biliary leakage Other Overall Mortality Reoperation Pancreatic function{{ Pre-operative pathologic OGTT Postoperative pathologic OGTT Pre-operative exocrine insufficiency Postoperative exocrine insufficiency Body weight (% decrease after surgery)
ECP (n = 35)
ERR (n = 35)
ELR (n = 35)
58 (29–76) 18 17
57 (30–78) 19 16
58 (34–88) 18 17
10 11 11 3
11 11 11 2
11 12 11 1
0 (0–4) 235 (110–425) 12 (7–30)
2 (0–22) 350 (125–600) 24 (11–47)
1 (0–4) 260 (120–570) 18 (7–57)
5 (14%) 3 (9%) — — 1 (3%) 9 (26%) — 1 (3%)
4 2 2 2 5 15 2 3
(11%) (6%) (6%) (6%) (14%) (43%) (6%) (9%)
6 (17%) 2 (6%) 1 (3%) — 4 (11%) 13 (37%) — 2 (6%)
1 (3%) 2 (6%) — 3 (9%) 0.4%
3 9 1 8
(9%) (28%) (3%) (24%) 6.6%
3 11 1 10
(9%) (34%) (3%) (29%) 1.3%
P value
.01* .01y and .02z .01§ and .05k ns
.05{ and .05# ns ns ns .02** and .05yy ns .03zz and .06§§ .05kk
*P = .01 for ERR versus ECP and for ELR versus ECP. yP = .01 for ERR versus ECP and for ERR versus ELR. zP = .02 for ECP versus ELR. §P = .01 for ERR versus ECP. kP = .05 for ECP versus ELR. {P = .05 for ERR versus ECP. #P = .05 for ELR versus ECP. **P = .02 for ERR versus ECP. yyP = .05 for ELR versus ECP. zzP = .03 for ERR versus ECP. §§P = .06 for ELR versus ECP. kkP = .05 for ERR. {{Values indicating postoperative OGTT and exocrine function refer to ‘‘de novo’’ findings, whereas patients with pre-operative derangements (ECP = 1, ECRR = 3, and ECRR = 3) were not considered in functional follow-up investigations. ECP, Extended central pancreatectomy; ELR, extended left resection; ERR, extended right resection; ns, not significant.
resectional procedures based on a questionable frozen section on unpredictable intra-operative findings were selected. This approach ensured that only patients in whom an extended central pancreatectomy represented a potential surgical alternative were included. Central pancreatectomy was first introduced in 1957 and was performed in a patient with chronic pancreatitis.13 In truth, the Beger procedure, which represents one of the milestones in the surgical armamentarium for chronic pancreatitis, is a modified right-sided central pancreatectomy
characterized by the complete transection of the pancreas, hereby resulting in 2 cut surfaces to be drained enterically.27,28 Central pancreatectomy has become the procedure of choice for small benign and borderline lesions located in the pancreatic body and neck. It has also been suggested for low-grade malignancies and metastases from extrapancreatic malignancies.19 To identify patients in whom classic resection would constitute operative ‘‘overtreatment’’ and, in turn, to protect those patients from inadequate organ-sparing pancreatic
Cataldegirmen et al 337
Surgery Volume 147, Number 3
1,00
Central vs. Right p=0.35 Central vs. Left p=0.296 Rght vs. Left p=0.516i
0,95 0,90
Cum Survival
0,85
Extended Central Pancreatic Resection
0,80 0,75 0,70
Extended Left Resection
0,65 0,60 Extended Right Resection
0,55 0,50 0
12 24 36 48 60 72 84 96 108 120 132 144 156 168 180
OverallSurvival in Months
Fig 2. Kaplan-Meier overall cumulative survival curves comparing patients with extended central pancreatic resection, extended left resection, and extended right resection (n = 35, each).
resection who are likely to have malignant lesions, preoperative evaluation using endoscopic, ultrasonography-guided, fine-needle aspiration biopsy is mandatory. Staging accuracy by endoscopic, ultrasonography-guided, fine-needle aspiration in the diagnostic workup of pancreatic lesions has improved considerably; its sensitivity in diagnosing questionable pancreatic lesions correctly is reported to be greater than 80%.18,19 Nonetheless, a gap remains between ‘‘formally’’ accepted indications for central pancreatectomy and the frequency of its use in the clinical routine. This gap is reflected by the fact that only 600 patients have reportedly undergone central pancreatectomy, and only 5 published series have a sample size exceeding 30 patients.9,10,29-32 Central pancreatectomy is usually performed for lesions with a transverse size smaller than 3 cm. So far, institutional studies scrutinizing the outcome of patients subjected to extended central pancreatectomy for appropriate lesions with a minimal size of $5 cm that were addressed in the presented series have not been published. Besides technical aspects, one may speculate that the major reason for the reluctance to perform central pancreatectomy derives from concerns regarding the accuracy of a preoperative histopathologic diagnosis and the inherent risk of ‘‘false negative’’ findings. An inadequate resection of a malignant lesion that has been identified preoperatively as a benign or borderline lesion is a well-known oncologic pitfall inherent to all organpreserving procedures. This possibility is especially pertinent to IPMNs, mucinous cystadenomas, and
neuroendocrine neoplasms, which may harbor malignant islets of tumor within ‘‘benign’’ or ‘‘borderline’’ lesions. Even frozen sections performed during the operation carry a considerable risk of overlooking invasive, malignant areas in such neoplasms, as was the case in 1 patient in the presented series. Another aspect contributing to the reluctant surgical attitude regarding central pancreatectomy is the assumption that it is associated with an increased incidence of pancreatic fistula. The procedure results in 2 pancreatic stumps of which at least the left side has to be anastomosed, whereas the cephalic remnant may be stapled, suture ligated, or covered with the serosal aspect of the distal jejunal limb that is used for the left-sided remnant. In the presented series, the prevalence of pancreatic fistula (14%) after extended central pancreatectomy did not differ from the incidence after extended left pancreatectomy (17%) and extended right pancreatectomy (11%). Nonetheless, despite the slightly greater incidence of pancreatic fistula after extended central pancreatectomy compared with extended classic left resection, substantial benefits regarding overall perioperative morbidity, median operative time, and need for blood transfusions have been found in comparison with classic resectional procedures. We also report on improved functional results after extended central pancreatectomy compared with extended classic right and left resection. The development of diabetes mellitus, exocrine insufficiency, and weight loss are leading causes of longterm morbidity of classic resectional procedures, such as pancreatoduodenal and distal splenopancreatectomy. The findings in this series demonstrate that postoperative endocrine function after extended central pancreatectomy was better than after extended classic right resection and extended classic left resections. These findings are consistent with previous studies.22,23 Significant benefits were also found in terms of protection from exocrine insufficiency in favor of extended central pancreatectomy compared with extended classic right resection and extended classic left resection. These findings are consistent with previous studies reporting the maintenance of exocrine function after organ-sparing pancreatic resection.23 Fecal chymotrypsin and pancreolauryl levels after extended central pancreatectomy were substantially greater than after extended classic right resection and extended classic left resection. Moreover, when compared with extended classic right resection, the functional benefits of extended central pancreatectomy resulted in a substantial protection from body weight loss.
338 Cataldegirmen et al
In conclusion, extended central pancreatectomy is a safe and technically feasible option for the operative treatment of benign, borderline, or low-grade malignant neoplasms of the pancreas. It is not associated with an increased likelihood of pancreatic fistula as compared with extended classic left resection, but overall, postoperative morbidity is less than after extended standard resection. The risk of developing diabetes mellitus and exocrine insufficiency is negligible. Extended central pancreatectomy should be considered as a less invasive alternative to extended classic resection in appropriately selected patients. REFERENCES 1. Craighead CC, Lien RC. Pancreatoduodenal resection; comments on indications, operative diagnosis, staged procedures, morbid and lethal factors, and survival. Ann Surg 1958;147:931-4. 2. Lillemoe KD, Kaushal S, Cameron JL, Sohn TA, Pitt HA, Yeo CJ. Distal pancreatectomy: indications and outcomes in 235 patients. Ann Surg 1999;229:693-8. 3. Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, et al. Six hundred fifty consecutive pancreaticoduodenectomies in the 1990s: pathology, complications, and outcomes. Ann Surg 1997;226:248-60. 4. Neoptolemos JP, Russell RC, Bramhall S, Theis B. Low mortality following resection for pancreatic and periampullary tumors in 1026 patients: UK survey of specialist pancreatic units. UK Pancreatic Cancer Group. Br J Surg 1997;84: 1370-6. 5. Cameron JL, Riall TS, Coleman J, Belcher KA. One thousand consecutive pancreaticoduodenectomies. Ann Surg 2006;244:10-5. 6. Brennan FM, Kattan MW, Klimstra D, Conlon K. Prognostic nomogram for patients undergoing resections for adenocarcinoma of the pancreas. Ann Surg 2004;240:293-8. 7. Tien YW, Lee PH, Yang CY, Ho MC, Chiu YF. Risk factors of massive bleeding related to pancreatic leak after pancreaticoduodenectomy. J Am Coll Surg 2005;201:554-9. 8. Gouma DJ, van Geenen RC, van Gulik TM, de Haan RJ, de Wit LT, Busch OR, et al. Rates of complication and death after pancreaticoduodenectomy: risk factors and the impact of hospital volume. Ann Surg 2000; 232:786-95. 9. Mu¨ller MW, Friess H, Kleeff J, Hinz U, Wente MN, Paramythiotis D, et al. Middle segmental pancreatic resection: an option to treat benign pancreatic body lesions. Ann Surg 2006;233:909-20. 10. Sauvanet A, Partensky C, Sastre B, Gigot JF, Fagniez PL, Tuech JJ, et al. Medial pancreatectomy: a multi-institutional retrospective study of 53 patients by the French Pancreas Club. Surgery 2002;132:836-43. 11. Carrere N, Abid S, Julio CH, Bloom E, Prade`re B. Spleenpreserving distal pancreatectomy with excision of splenic artery and vein: a case-matched comparison with conventional distal pancreatectomy with splenectomy. World J Surg 2007; 31:375-82. 12. Shoup M, Brennan MF, McWhite K, Leung DH, Klimstra D, Conlon KC. The value of splenic preservation with distal pancreatectomy. Arch Surg 2002;137:164-8.
Surgery March 2010
13. Guillemin P, Bessot M. Chronic calcifying pancreatitis in renal tuberculosis: pancreatojejunostomy using an original technique. Mem Acad Chir (Paris) 1957;83:869-71. 14. Letton AH, Wilson JP. Traumatic severance of pancreas treated by Roux-Y anastomosis. Surg Gynecol Obstet 1959;109:473-8. 15. Warshaw AL, Rattner DW, Ferna´ndez-del Castillo C, Z’graggen K. Middle segment pancreatectomy: a novel technique for conserving pancreatic tissue. Arch Surg 1998;133:327-31. 16. Iacono C, Bortolasi L, Serio G. Indications and technique of central pancreatectomy---early and late results. Langenbecks Arch Surg 2005;390:266-71. 17. Sperti C, Pasquali C, Ferronato A, Pedrazzoli S. Median pancreatectomy for tumors of the neck and body of the pancreas. J Am Coll Surg 2000;190:711-6. 18. Fritscher-Ravens A, Izbicki JR, Sriram PV, Krause C, Knoefel WT, Topalidis T, et al. Endosonography-guided, fine-needle aspiration cytology extending the indication for organ-preserving pancreatic surgery. Am J Gastroenterol 2000;95: 2255-60. 19. Schurr PG, Strate T, Rese K, Kaifi JT, Reichelt U, Petri S, et al. Aggressive surgery improves long-term survival in neuroendocrine pancreatic tumors: an institutional experience. Ann Surg 2007;245:273-81. 20. Brueckel J, Koebberling J. Definition, klassifikation und diagnostik des diabetes mellitus. Diabetes Stoffwechsel 2002; 11:6-39. 21. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J. International Study Group on Pancreatic Fistula Definition. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8-13. 22. Kendall DM, Sutherland DE, Najarian JS, Goetz FC, Robertson RP. Effects of hemipancreatectomy on insulin secretion and glucose tolerance in healthy humans. N Engl J Med 1990;322:898-903. 23. Wittingen J, Frey CF. Islet concentration in the head, body, tail and uncinate process of the pancreas. Ann Surg 1974; 179:412-4. 24. Sperti C, Pasquali C, Ferronato A, Pedrazzoli S. Median pancreatectomy for tumors of the neck and body of the pancreas. J Am Coll Surg 2000;190:711-6. 25. Yasuda H, Takada T, Toyota N, Amano H, Yoshida M, Takada Y, et al. Limited pancreatectomy: significance of postoperative maintenance of pancreatic exocrine function. J Hepatobiliary Pancreat Surg 2000;7:466-72. 26. Warshaw AL, Rattner DW, Ferna´ndez-del Castillo C, Z’graggen K. Middle segment pancreatectomy: a novel technique for conserving pancreatic tissue. Arch Surg 1998;133:327-31. 27. Beger HG, Bu¨chler M, Bittner RR, Oettinger W, Roscher R. Duodenum-preserving resection of the head of the pancreas in severe chronic pancreatitis. Early and late results. Ann Surg 1989;209:272-8. 28. Frey CF, Child CG, Fry W. Pancreatectomy for chronic pancreatitis. Ann Surg 1976;184:403-13. 29. Adham M, Giunippero A, Hervieu V, Courbie`re M, Partensky C. Central pancreatectomy: single-center experience of 50 cases. Arch Surg 2008;143:175-81. 30. Balzano G, Zerbi A, Veronesi P, Cristallo M, Di Carlo V. Surgical treatment of benign and borderline neoplasms of the pancreatic body. Dig Surg 2003;20:506-10. 31. Crippa S, Bassi C, Warshaw AL, Falconi M, Partelli S, Thayer SP, et al. Middle pancreatectomy: indications, short- and long-term operative outcomes. Ann Surg 2007;246:69-76. 32. Wente MN, Veit JA, Bassi C. Postpancreatectomy hemorrhage (PPH)-An International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007;142:20-5.