- Email: [email protected]
Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review
HPB
http://dx.doi.org/10.1016/j.hpb.2017.08.002
REVIEW ARTICLE
Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review Niccolò Petrucciani1,2, Tarek Debs2, Giuseppe Nigri1, Giulia Giannini1, Elena Sborlini1, Radwan Kassir3, Imed Ben Amor2, Antonio Iannelli2, Stefano Valabrega1, Francesco D’Angelo1, Jean Gugenheim2 & Giovanni Ramacciato1 1
Department of Medical and Surgical Sciences and Translational Medicine, Faculty of Medicine and Psychology, Sapienza University of Rome, St Andrea Hospital, Rome, Italy, 2Division of Digestive Surgery and Liver Transplantation, Nice University Hospital, Nice, and 3 Department of General and Thoracic Surgery, St Etienne University Hospital, St Etienne, France
Abstract Background: Multivisceral resections combined with pancreatectomy have been proposed in selected patients with tumor invasion into adjacent organs, in order to allow complete tumor resection. Some authors have also reported multivisceral resection combined with metastasectomy in very selected cases. The utility of this practice is debated. The aim of the review is to compare the postoperative results and survival of pancreatectomies combined with multivisceral resections with those of standard pancreatectomies. Methods: A systematic literature search was performed to identify all studies published up to February 2017 that analyzed data of patients undergoing multivisceral and standard pancreatectomies. Clinical effectiveness was synthetized through a narrative review with full tabulation of results. Results: Three studies were retrieved, including 713 (80%) patients undergoing standard pancreatectomies and 176 (20%) undergoing multivisceral resections (MVR). Postoperative morbidity ranged from 37% to 50% after standard resections and from 56% to 69% after MVR. In-hospital mortality ranged from 4% after standard pancreatectomies to 10% after MVR. Median survival ranged from 20 to 23 months in standard resections and from 12 to 20 months after MVR, without significant differences. Discussion: The current literature suggests that multivisceral pancreatectomies are feasible and may increase the number of completely resected patients. Morbidity and mortality are higher than after standard pancreatectomies, and these procedures should be reserved to selected patients in referral centers. Further studies on the role of neoadjuvant therapy in this setting are advisable. Received 20 March 2017; accepted 12 August 2017
Correspondence Niccolò Petrucciani, Department of Medical and Surgical Sciences, and Translational Medicine, Faculty of Medicine and Psychology, Sapienza University of Rome, St Andrea Hospital, via di Grottarossa 1035-9, 00189, Rome, Italy. E-mail: [email protected]
Introduction Surgical resection with adjuvant chemotherapy represents the only potentially curative treatment for pancreatic malignancies. Five-year survival approaches 25% for patients undergoing complete surgical resection, in combination with chemotherapy.1,2 In recent time, several strategies have been tried in an effort to increase the proportion of patients with pancreatic
HPB 2017, -, 1–8
malignancies offered potentially curative surgical resection. These have included the use of neoadjuvant chemoradiation therapies based on the potential theoretical benefits of reduction of tumor volume and increase of complete resection rate.3–5 Portal and mesenteric vein resections are recommended in patients with venous invasion.6,7 Arterial resections have been also proposed to increase the R0 resection rate, but they are still controversial.8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
2
Multivisceral resections have been proposed in selected patients with invasion of adjacent organs, in order to allow a complete tumor resection.9 Some authors have emphasized the increase of postoperative morbidity and mortality after multivisceral resections during pancreatectomy,9 whereas others have reported postoperative outcomes similar to standard resections.10 Nevertheless, the utility of this practice is still debated. Neither meta-analyses nor randomized trials have been published on this subject. Therefore, the aim of this study was to perform a systematic review of the literature to evaluate the role of multivisceral resections combined with pancreatectomy for pancreatic malignancies based on the short and long term outcomes as compared to standard pancreatectomy.
Methods Study selection A systematic literature search was performed using Embase, Medline, Cochrane, and PubMed databases to identify all studies published up to and including February 2017 that analyzed the postoperative outcomes and the overall survival of patients undergoing pancreatectomy combined with multivisceral resection for pancreatic malignancies. The systematic review was conducted according the PRISMA guidelines.11 The following MESH search headings were used: “multivisceral OR extended OR adjacent organs”, “pancreatic cancer OR pancreatic carcinoma OR pancreatic adenocarcinoma OR pancreatic tumor”, “pancreatectomy OR pancreaticoduodenectomy OR pancreatoduodenectomy OR distal pancreatectomy OR duodenopancreatectomy OR pancreatic resection”. The “related articles” function was used to broaden the search, and all abstracts, studies, and citations scanned were reviewed. Using the criteria of the PRISMA statement, two authors (GG and ES) independently searched the literature for relevant studies. The abstract screening led to identification of the papers eligible to systematic review. A third author (NP) carried out conflicts. Inclusion and exclusion criteria Studies included were required to compare characteristics, perioperative outcomes and/or overall survival of patients undergoing pancreatectomy with or without multivisceral resection for pancreatic malignancies. Multivisceral resections were defined as resection of one or more adjacent organs that are usually not resected during pancreatectomy. Splenectomy associated with distal pancreatectomy was not considered as a multivisceral resection. Venous resections were not considered as multivisceral resection. Only patients with pancreatic malignancies were included; patients submitted to pancreatectomy associated with multivisceral resection for tumors originating from other organs or for metastases to the pancreas from other origins were excluded from the analysis. Non-pancreatic cancers were excluded because comparison of survival data would be biased. Furthermore, also loco-regional spreading patterns are
HPB 2017, -, 1–8
different between different types of tumors (vascular invasion, nodal invasion), and may have an impact on perioperative decisions and outcomes. The following types of studies were not considered for inclusion in the systematic review: (i) studies in which the outcomes of interest for pancreatectomies with multivisceral resections were not reported or were impossible to calculate; (ii) studies where no comparison with standard pancreatectomies was reported; (iii) “how I do it” articles, animal studies, case reports, and non-English language studies. Quality assessment Three reviewers, according to the Newcastle–Ottawa Scale (NOS) for non-randomized studies,12 reviewed all selected studies for methodological quality. Final scores were reached by general consensus. The Cochrane Collaboration’s tool for assessing risk of bias in individual studies was also used by three independent authors,13 and conflicts were ruled out by discussion. Data extraction Data were extracted on the base of a piloted form and registered in a spreadsheet for comprehensive analysis. Two reviewers independently extracted the following information from each study: first author, year of publication, study design, study population characteristics, tumors’ characteristics, surgical techniques, and survival outcomes, according to the magnitude of the surgery (standard or multivisceral pancreatectomy). Outcomes of interest and definition All studies were abstracted for the following relevant data: patient baseline characteristics (age, gender), tumor characteristics (histology, staging, tumor volume), type of procedure and adjacent organ resection, definitive resection margins, lymph node involvement, postoperative morbidity and mortality, survival data. A meta-analysis was considered not appropriate because studies were heterogeneous. Clinical effectiveness was synthetized through a narrative review with full tabulation of results of the included studies.
Results Study selection Systematic search process is showed in Fig. 1. Three studies evaluating the outcomes of standard and multivisceral pancreatic resections for malignancies of the pancreas met inclusion criteria. The included studies were published between 2009 and 201514–16 and included patients accrued between 1992 and 2014. A total of 889 patients were included of which 176 (20%) underwent MVR resection. All the included studies were single institution retrospective reports, from Germany. New Castle– Ottawa scale score was 7 for each of the three studies. One study16 matched patients with or without multivisceral resection using type of surgery, age, gender and histology.
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
3
Figure 1 PRISMA 2009 flow diagram
9,10,17–20
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and MetaAnalyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097. For more information, visit www.prismastatement.org.
In Table 1, patients’ and tumors’ characteristics of the two groups in each selected study are reported. In the study by Kulemann et al.14 distant metastases were considered a contraindication for resection. Concerning associated liver resections, Kulemann et al.14 reported liver resection only in case of direct contact/infiltration of the primary tumor but not in case of liver metastases. Hartwig et al.16 generally excluded patients with metastases. Patients with a solitary liver metastasis underwent pancreatectomy and metastasectomy only if they were young or in extremely good condition. Burdelski et al.15 performed liver resections for metastases only in young patients in extraordinarly good health, and hepatectomies varied from atypical resections to segmental resections. Surgical treatment and intraoperative variables are shown in Table 2. We reported the extent of multivisceral resection and the resected organs for each study in Table 3.
HPB 2017, -, 1–8
Postoperative outcomes and survival Postoperative outcomes are listed in Table 4. Few details regarding oncological treatment were reported. Hartwig et al.16 reported neoadjuvant therapy in 25 out of 202 patients (12%) in group 1 and 21 out of 101 (21%) in group 2 but did not specify the protocol. Burdelski et al.15 did not administer neoadjuvant treatment, but treated all their patients since 2004 with gemcitabine-based adjuvant chemotherapy. In the study by Kulemann et al.,14 30 patients (group not specified) out of a total of 228 had neoadjuvant chemotherapy (protocol not specified). In this study,14 postoperative treatment was heterogeneous, and the authors assert that in the early study period, patients did not routinely undergo adjuvant treatment. Lately, most patients underwent adjuvant chemotherapy after margin negative resections and chemoradiation after resections with a positive margin.14
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
4
Table
1 Patients’
and tumors’ characteristics for standard pancreatectomies and multivisceral resections associated with
pancreatectomies Variable
Standard pancreatectomies (%)
Multivisceral resections (%)
Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
Age, ya
66 (31–84)
nr
64 (33–88)
62 (40–89)
nr
63 (13–81)
Male gender
101 (49%)
nr
115 (57%)
9
nr
65 (64%)
1
nr
9 (3%)c
28 (14%)b,c
nr
0 (0%)c
1 (1%)b,c
2
nr
56 (18%)c
–
nr
0 (0%)c
T stage
3 4 N0
nr
209 (69%)
nr
c
45 (15%)
67 (32%)
c
165 (83%) 5 (3%)
117 (40%)
c c
c
nr
c
nr
66 (33%)
6
–
35 (64%)
c
87 (89%)c
20 (36%)
c
10 (10%)c
20 (36%)
c
34 (34%)
c
67 (66%)
N1
140 (68%)
180 (60%)
136 (67%)
14
35 (64%)
M1
0 (0%)
34 (11%)c
20 (10%)c
0
20 (36%)c
39 (39%)c
29 (1–130)
nr
nr
25 (18–60)
nr
nr
Tumor size, mm
a
N, number; Neoad., neoadjuvant; T, tumor stage according to Union of International Cancer Control/AJCC, N, nodal stage according to Union of International Cancer Control/AJCC; nr, not reported; y, years. a Data reported as median (range). b Data refers to T1 and T2 tumors. c Significant difference between standard and multivisceral resections.
Table 2 Surgical treatment, and intraoperative surgical variables for patients undergoing standard pancreatectomies and multivisceral
resections associated with pancreatectomies Variable
Standard pancreatectomies (%)
Multivisceral resections (%)
Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
PD
23 (11%)
nr
42 (21%)
2
30 (55%)
21 (21%)
PPPD
176 (85%)
nr
0 (0%)
11
0 (0%)
0 (0%)
SPD
0 (0%)
nr
0 (0%)
0
14 (25%)
0 (0%)
c
c
TP
9 (4%)
nr
40 (20%)
7
11 (20%)
20 (20%)
DP
0 (0%)
nr
120 (59%)
0
0 (0%)
60 (59%)
Op. time, mina
427 (184–720)c
340
260 (190–350)c
477 (346–760)c
351
320 (211–460)c
Intraop. Transf.a,b
0 (0–5)
2 (0–35)
nr
0 (0–2)
3 (0–44)
nr
Blood loss, La
nr
nr
1 (0–1)c
nr
nr
1 (1–2)c
N, number, PD, pancreaticoduodenectomy; PPPD, pylorus-preserving pancreaticoduodenectomy; SPD, subtotal pancreaticoduodenectomy; TP, total pancreatectomy; DP, distal pancreatectomy; Op., operative; min., minutes; Intraop, intraoperative; Transf., transfusions; L, liters. a Data reported as median (range). b Data reported as red blood packed cells. c Significant difference between standard and multivisceral resections.
Predictive and prognostic factors for short term morbidity an survival are shown in Table 5. No study identified risk factors for perioperative mortality.
Discussion Complete surgical resection with chemotherapy remains the only potentially curative treatment for pancreatic malignancies.21–23
HPB 2017, -, 1–8
Centralization of pancreatic surgery and advancement in surgical techniques and perioperative management had led surgeons to consider extended resections.24–29 Only a few studies have evaluated the role of multivisceral resection (MVR) for pancreatic cancer. A systematic search identified three heterogeneous series. Some authors have reported their results of multivisceral resections combined with pancreatectomy, including also portal or superior mesenteric vein resection into the multivisceral
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
5
group.18,19 Actually, venous resection has become part of the routine clinical practice. Approximately 20% of the patients who present for a pancreatic resection will concomitantly undergo a vascular resection according to national databases.20,21 Therefore, it is the authors opinion that venous resection should not be taken into account in the analysis of multivisceral pancreatectomies. Other series have analyzed multivisceral resection associated to pancreaticoduodenectomy20 or distal pancreatectomy.10 However the indications for resection did not include only pancreatic malignancies, but also various and heterogeneous benign and malignant diseases. It is the authors opinion that comparison of surgical resection of different tumors (bile duct, pancreas, duodenum, contiguous tumors) or even for chronic pancreatitis could result in bias not only concerning survival outcomes but also postoperative outcomes, because the technical problems and the spreading patterns of these tumors are different. The surgical approach and the potential morbidity may be influenced by the heterogeneity of such study population. The only study from a national database (National Surgical Quality Improvement Project, United Stated), by Bhayani et al.,9 analyzed 273 multivisceral resections, and highlighted the increased morbidity and mortality of MVR. This report included a heterogeneous population and analyzed data of patients operated on for several indications, such as pancreatic, periampullary and duodenal cancers and benign diseases of the pancreas.
Table 3 Numbers and types of resected organs in patients who
underwent pancreatectomies with multivisceral resections Kulemann
Burdelski
Hartwig
N. MVR
20 (9%)
55 (15%)
101 (33%)
1 additional resected organ
nr
13 (23%)
17 (17%)
2 additional resected organs
nr
24 (44%)
46 (47%)
3 additional resected organs
nr
11 (20%)
23 (23%)
4 or more additional resected organs
nr
7 (13%)
15 (15%)
Resected organs in multivisceral resections Stomach
6
32
34
Colon
7
22
38
Liver
5
24
19
Kidney
0
17
12
Diaphragm
0
11
0
Small bowel
1
5
7
Adrenal
1
0
28
Additional portal vein resection
12
nr
21
Additional arterial resection
nr
nr
17
nr, not reported; N., number, MVR, multivisceral resections.
Table 4 Short-term outcomes and survival for patients undergoing standard pancreatectomies and multivisceral resections associated with
pancreatectomies Variable
Standard pancreatectomies (%) Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
R0
157 (76%)b
225 (74%)b
122 (61%)
18b
30 (55%)b
58 (57%)
R1
49 (24%)b
58 (19%)b
64 (32%)
2b
21 (39%)b
29 (29%)
b
b
Multivisceral resections (%)
b
b
R2
0 (0%)
13 (7%)
0
4 (8%)
14 (14%)
Postop. transfa
nr
0 (0–38)c
nr
nr
1 (0–78)c
nr
Morbidity
104 (50%)
112 (37%)b
86 (43%)b
13
38 (69%)b
56 (56%)b
PPH
nr
19 (6%)
16 (8%)
nr
4 (7%)
10 (10%)
POPF
nr
27 (9%)
14 (7%)
nr
5 (9%)
11 (11%)
GI bleeding
nr
7 (2%)b
nr
nr
4 (7%)b
20 (6%)
Colonic fistula
nr
1 (0.3%)
Peritonitis
nr
nr
b
nr b
nr
nr
9 (16%)
3 (2%)
nr
Nr
6 (6%)
nr
DGE
nr
nr
31 (15%)
nr
Nr
17 (17%)
Re−laparotomy
19 (9%)
39 (13%)
17 (8%)b
3
10 (18%)
22 (22%)b
In hosp. mortality
8 (4%)
12 (4%)
7 (4%)
2
4 (7%)
7 (7%)
Survival, monthsa
20
21
23
12
16
20
N, number; R, resection margin according to Union of International Cancer Control/AJCC; Postop., postoperative; nr, not reported, POPF, postoperative pancreatic fistula, DGE, delayed gastric emptying, in hosp., in hospital, PPH, postoperative hemorrhage. Data reported as median (range). b Significant difference between standard and multivisceral resections. c Data reported as units of red blood packed cells, or # liters of transfused blood. a
HPB 2017, -, 1–8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
6
Table 5 Predictive factors of postoperative morbidity and negative prognostic factor for overall survival
Predictive factors of postoperative morbidity Type of analysis
Kulemann
Burdelski
Hartwig
Univariate
–
Intraop. transfusions
Absence of liver resection
Colic resection
Operative time >300 min
Kidney resection Liver resection Multivariable
–
Intraop. transfusions
Operative time >300 min
Kidney resection Negative prognostic factors for overall survival Type of analysis
Kulemanna
Burdelskib
Univariate
Nodal disease
T stage
Resection margin +
Kidney resection
Intraop. Blood transfusions
4 organs resected
Multivisceral resections
Postop. transfusions
Nodal disease
T stage
Hartwig
ICU stay >2 days Multivariable
Resection margin + Intraop. Blood transfusions Multivisceral resections Intraop., intraoperative; postop., postoperative. a The analyses refer to the entire population (group 1 + group 2). b The analyses refer to group 2.
Patients who underwent multivisceral resections had more advanced disease, with a higher stage in two of the studies15,16 and more frequent need of total pancreatectomy in the latter.14 It is interesting to note that in two out of three studies, liver resection for metastases was not an absolute contraindication, and it was reserved to oligometastatic young and/or very healthy selected patients.15,16 In these two studies the rate of M1 resection was 36%15 and 39%16 in patients undergoing multivisceral resections which is clearly outside what is reported in other studies. In these two studies, the rate of M1 disease was also considerable in patients having standard pancreatectomy, 11%15 and 10%,16 respectively.15,16 It is interesting to observe that survival results are not worst in the two studies including highly selected oligometastatic patients. Another interesting aspect was the use of neoadjuvant therapy in these series. Nowadays, the use of neoadjuvant therapy for borderline resectable pancreatic cancer has increased, and NCCN guidelines no longer recommend upfront resection in this setting.30–32 A recent meta-analysis has shown that 65% of borderline patients undergo resection after neoadjuvant therapy, with 57% achieving R0 status and a median survival of 26 months for resected patients.3 The utility of neoadjuvant therapy has also been advocated for resectable and for locally advanced pancreatic cancer.33–35 It is not specified if some patients included in the reviewed articles had resectable or borderline resectable or locally
HPB 2017, -, 1–8
advanced disease, but it seems logical to hypothesize that tumors requiring multivisceral resections have a higher risk of R1 or R2 resection, and may benefit from neoadjuvant therapies. Actually, only a minority of the patients included in this review had neoadjuvant treatment. For resection margins, results were discordant: Burdelski et al.15 report significant more frequent margin involvement in case of multivisceral resection, whereas in the study of Kulemann et al.,14 MVR had significantly lower rate of R+ (Not associated with a survival advantage, however). Hartwig et al.16 did not report significant differences. It is difficult to interpret these data, which may also reflect different protocols of pathological analysis, not always following the Leeds criteria, as stated by Kulemann et al.14 This review clearly shows that multivisceral resections have specific complications, which does not occur after standard pancreatectomies. Colonic fistula was as frequent as 16% of patients in the study by Burdelski et al.,15 who suggest liberal use of protective ileostomies to reduce the impact of such complication. The frequency and types of complications of multivisceral resection combined to pancreatectomy suggests the need of a tertiary referral centers with high level of expertise not only in pancreatic surgery but also in digestive surgery. The risk of these procedures are remarkable, with mortality rates higher than 5% and approaching 10%, and morbidity around 60%, data which should be taken into account in the patients’ counseling and decisional
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
7
process. Median survival after MVR ranged from 12 to 20 months. In all included studies survival was lower than after standard pancreatectomy, even if no significant differences were found. This systematic review shows that multivisceral pancreatectomies are feasible, even if postoperative morbidity and mortality are higher than after standard pancreatectomies. The overall survival was not significantly different in the two groups. This approach may increase the number of patients undergoing radical resection, and offers the possibility for curative resection to patients otherwise considered non resectable for infiltration of contiguous organs. Further studies analyzing the role of neoadjuvant therapies in patients with advanced tumor invading surrounding organs are advisable. Neoadjuvant therapy may permit better patients’ selection, excluding those who became non resectable for rapid tumor progression, and increase R0 rate, in patients with partial or complete tumor response, with potential survival benefit. Concerning patients’ management, it is important to highlight the high morbidity rate of MVR, and the potential complications of additional organ resections, which are frequent, in order to promptly detect and treat them.
standard pancreatectomies. The current literature suggests that multivisceral pancreatectomies are feasible and may increase the number of completely resected patients. Morbidity and mortality are higher than after standard pancreatectomies, and these procedures should be reserved to very selected patients in referral centers. Further studies on the role of neoadjuvant therapy in this setting are advisable.
Limits The study has several limitations. First of all, the included studies were all retrospective. However, all of them were conducted in referral centers for pancreatic cancer surgery. Second, two of the studies included a small percentage of patients with M1 pancreatic disease, which represent a contraindication for surgery, except in the setting of clinical studies. Third, only a low percentage of patients underwent neoadjuvant chemotherapy. Fourth, indications and protocols of neoadjuvant and adjuvant chemotherapies are not extensively detailed, and the treatments were heterogeneous. Fifth, all the studies were conducted in a single country (i.e., Germany), which may limit the generalizability of the data. Furthermore, clear definition of R0 margins was not clearly specified by the included studies (even if all the included institutions are referral pancreatic cancers, with expert pathologists). Strengths of the study include: (i) the methodological quality of the review, based on PRISMA guidelines, and including systematic literature search and quality assessment of the studies; (ii) the inclusion of referral centers; (iii) the high degree of concordance of the conclusion of each study which gave value of the review’s conclusion.
3. Tang K, Lu W, Qin W, Wu Y. (2016) Neoadjuvant therapy for patients
Funding No funding received. Conflict of interest None declared. References 1. D’Angelo FA, Antolino L, La Rocca M, Petrucciani N, Magistri P, Aurello P et al. (2016) Adjuvant and neoadjuvant therapies in resectable pancreatic cancer: a systematic review of randomized controlled trials. Med Oncol 33:28. 2. Strosberg JR, Cheema A, Weber JM, Ghayouri M, Han G, Hodul PJ et al. (2012) Relapse-free survival in patients with nonmetastatic, surgically resected pancreatic neuroendocrine tumors: an analysis of the AJCC and ENETS staging classifications. Ann Surg 256:321–325. with borderline resectable pancreatic cancer: a systematic review and meta-analysis of response and resection percentages. Pancreatology 16:28–37. 4. McClaine RJ, Lowy AM, Sussman JJ, Schmulewitz N, Grisell DL, Ahmad SA. (2010) Neoadjuvant therapy may lead to successful surgical resection and improved survival in patients with borderline resectable pancreatic cancer. HPB 12:73–79. 5. Khushman M, Dempsey N, Maldonado JC, Loaiza-Bonilla A, Velez M, Carcas L et al. (2015) Full dose neoadjuvant FOLFIRINOX is associated with prolonged survival in patients with locally advanced pancreatic adenocarcinoma. Pancreatology 15:667–673. 6. Ramacciato G, Nigri G, Petrucciani N, Pinna AD, Ravaioli M, Jovine E et al. (2016 Feb 18) Pancreatectomy with mesenteric and portal vein resection for borderline resectable pancreatic cancer: multicenter study of 406 patients. Ann Surg Oncol [Epub ahead of print]. 7. Ramacciato G, Mercantini P, Petrucciani N, Giaccaglia V, Nigri G, Ravaioli M et al. (2009) Does portal-superior mesenteric vein invasion still indicate irresectability for pancreatic carcinoma? Ann Surg Oncol 16:817–825. 8. Jegatheeswaran S, Baltatzis M, Jamdar S, Siriwardena AK. (2017) Superior mesenteric artery (SMA) resection during pancreatectomy for malignant disease of the pancreas: a systematic review. HPB 19: 483–490. 9. Bhayani NH, Enomoto LM, James BC, Ortenzi G, Kaifi JT, Kimchi ET et al. (2014) Multivisceral and extended resections during pancreato-
Conclusions
duodenectomy increase morbidity and mortality. Surgery 155:567–574.
The systematic review included three studies evaluating the utility of multivisceral resection combined with pancreatectomy for pancreatic malignancies. Patients treated with multivisceral resections had a morbidity rate ranging from 56% to 69% and a mortality rate ranging from 7% to 10%. Survival ranged from 12 to 20 and was not significantly different than those reported for
HPB 2017, -, 1–8
10. Irani JL, Ashley SW, Brooks DC, Osteen RT, Raut CP, Russell S et al. (2008) Distal pancreatectomy is not associated with increased perioperative morbidity when performed as part of a multivisceral resection. J Gastrointest Surg 12:2177–2182. 11. Preferred Reporting Items for Systematic Reviews and MetaAnalyses
(PRISMA)
website,
http://www.prisma-statement.org.
(accessed 7 February 2017).
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
8
12. The Newcastle-Ottawa Scale (NOS) for assessing the quality of
24. Gooiker GA, van der Geest LG, Wouters MW, Vonk M, Karsten TM,
nonrandomised studies in meta-analyses. http://www.ohri.ca/programs/
Tollenaar RA et al. (2011) Quality improvement of pancreatic surgery by
clinical_epidemiology/oxford.asp. (accessed 7 February 2017).
centralization in the western part of The Netherlands. Ann Surg Oncol
13. The Cochrane Collaboration’s tool for assessing risk of bias. http://ohg. cochrane.org/sites/ohg.cochrane.org/files/uploads/Risk%20of%20bias %20assessment%20tool.pdf. (accessed 7 March 2016). 14. Kulemann B, Hoeppner J, Wittel U, Glatz T, Keck T, Wellner UF et al. (2015) Perioperative and long-term outcome after standard pancreaticoduodenectomy, additional portal vein and multivisceral resection for pancreatic head cancer. J Gastrointest Surg 19:438–444. 15. Burdelski CM, Reeh M, Bogoevski D, Gebauer F, Tachezy M, Vashist YK et al. (2011) Multivisceral resections in pancreatic cancer: identification of risk factors. World J Surg 35:2756–2763. 16. Hartwig W, Hackert T, Hinz U, Hassenpflug M, Strobel O, Büchler MW et al. (2009) Multivisceral resection for pancreatic malignancies: riskanalysis and long-term outcome. Ann Surg 250:81–87. 17. Abu Hilal M, McPhail MJ, Zeidan BA, Jones CE, Johnson CD, Pearce NW. (2009) Aggressive multi-visceral pancreatic resections for locally advanced neuroendocrine tumours. Is it worth it? JOP 10:276–279. 18. Panzeri F, Marchegiani G, Malleo G, Malpaga A, Maggino L, Marchese T
18:1821–1829. 25. Massucco P, Ribero D, Sgotto E, Mellano A, Muratore A, Capussotti L. (2009) Prognostic significance of lymph node metastases in pancreatic head cancer treated with extended lymphadenectomy: not just a matter of numbers. Ann Surg Oncol 16:3323–3332. 26. Hartel M, Wente MN, Di Sebastiano P, Friess H, Büchler MW. (2004) The role of extended resection in pancreatic adenocarcinoma: is there good evidence-based justification? Pancreatology 4:561–566. 27. Stitzenberg KB, Watson JC, Roberts A, Kagan SA, Cohen SJ, Konski AA et al. (2008) Survival after pancreatectomy with major arterial resection and reconstruction. Ann Surg Oncol 15:1399–1406. 28. Hartwig W, Gluth A, Hinz U, Bergmann F, Spronk PE, Hackert T et al. (2015) Total pancreatectomy for primary pancreatic neoplasms: renaissance of an unpopular operation. Ann Surg 261:537–546. 29. Shrikhande SV, Kleeff J, Reiser C, Weitz J, Hinz U, Esposito I et al. (2007) Pancreatic resection for M1 pancreatic ductal adenocarcinoma. Ann Surg Oncol 14:118–127.
et al. (2016 Oct 27) Distal pancreatectomy associated with multivisceral
30. Blazer M, Wu C, Goldberg RM, Phillips G, Schmidt C, Muscarella P et al.
resection: results from a single centre experience. Langenbecks Arch
(2015) Neoadjuvant modified (m) FOLFIRINOX for locally advanced
Surg [Epub ahead of print].
unresectable (LAPC) and borderline resectable (BRPC) adenocarcinoma
19. Sasson AR, Hoffman JP, Ross EA, Kagan SA, Pingpank JF, Eisenberg BL. (2002) En bloc resection for locally advanced cancer of the pancreas: is it worthwhile? J Gastrointest Surg 6:147–157. 20. Nikfarjam M, Sehmbey M, Kimchi ET, Gusani NJ, Shereef S, Avella DM et al. (2009) Additional organ resection combined with pancreaticoduodenectomy does not increase postoperative morbidity and mortal-
of the pancreas. Ann Surg Oncol 22:1153–1159. 31. Rose JB, Rocha FG, Alseidi A, Biehl T, Moonka R, Ryan JA et al. (2014) Extended
chemotherapy
for
borderline
resectable
and survival. Ann Surg Oncol 21:1530–1537. 32. NCCN Clinical Practice Guidelines in Oncology, Pancreatic Adenocarcinoma,
ity. J Gastrointest Surg 13:915–921. 21. Sinn M, Striefler JK, Sinn BV, Sallmon D, Bischoff S, Stieler JM et al.
neoadjuvant
pancreatic cancer demonstrates promising postoperative outcomes
at
https://www.nccn.org/professionals/physician_gls/pdf/
pancreatic.pdf, (accessed 7 February 2017).
(2013) Does long-term survival in patients with pancreatic cancer
33. Christians KK, Heimler JW, George B, Ritch PS, Erickson BA,
really exist? Results from the CONKO-001 study. J Surg Oncol 108:
Johnston F et al. (2016) Survival of patients with resectable pancreatic cancer who received neoadjuvant therapy. Surgery 159:893–900.
398– 402. 22. Barugola G, Partelli S, Crippa S, Butturini G, Salvia R, Sartori N et al.
34. de Geus SW, Evans DB, Bliss LA, Eskander MF, Smith JK, Wolff RA
(2013) Time trends in the treatment and prognosis of resectable
et al. (2016) Neoadjuvant therapy versus upfront surgical strategies in
pancreatic cancer in a large tertiary referral centre. HPB 15:958–964.
resectable pancreatic cancer: a Markov decision analysis. Eur J Surg
23. Riall TS, Cameron JL, Lillemoe KD, Winter JM, Campbell KA,
Oncol 42:1552–1560.
Hruban RH et al. (2006) Resected periampullary adenocarcinoma:
35. Strobel O, Berens V, Hinz U, Hartwig W, Hackert T, Bergmann F et al.
5-year survivors and their 6- to 10-year follow-up. Surgery 140:
(2012 Sep) Resection after neoadjuvant therapy for locally advanced,
764–772.
“unresectable” pancreatic cancer. Surgery 152(3 Suppl. 1):S33–S42.
HPB 2017, -, 1–8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
http://dx.doi.org/10.1016/j.hpb.2017.08.002
REVIEW ARTICLE
Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review Niccolò Petrucciani1,2, Tarek Debs2, Giuseppe Nigri1, Giulia Giannini1, Elena Sborlini1, Radwan Kassir3, Imed Ben Amor2, Antonio Iannelli2, Stefano Valabrega1, Francesco D’Angelo1, Jean Gugenheim2 & Giovanni Ramacciato1 1
Department of Medical and Surgical Sciences and Translational Medicine, Faculty of Medicine and Psychology, Sapienza University of Rome, St Andrea Hospital, Rome, Italy, 2Division of Digestive Surgery and Liver Transplantation, Nice University Hospital, Nice, and 3 Department of General and Thoracic Surgery, St Etienne University Hospital, St Etienne, France
Abstract Background: Multivisceral resections combined with pancreatectomy have been proposed in selected patients with tumor invasion into adjacent organs, in order to allow complete tumor resection. Some authors have also reported multivisceral resection combined with metastasectomy in very selected cases. The utility of this practice is debated. The aim of the review is to compare the postoperative results and survival of pancreatectomies combined with multivisceral resections with those of standard pancreatectomies. Methods: A systematic literature search was performed to identify all studies published up to February 2017 that analyzed data of patients undergoing multivisceral and standard pancreatectomies. Clinical effectiveness was synthetized through a narrative review with full tabulation of results. Results: Three studies were retrieved, including 713 (80%) patients undergoing standard pancreatectomies and 176 (20%) undergoing multivisceral resections (MVR). Postoperative morbidity ranged from 37% to 50% after standard resections and from 56% to 69% after MVR. In-hospital mortality ranged from 4% after standard pancreatectomies to 10% after MVR. Median survival ranged from 20 to 23 months in standard resections and from 12 to 20 months after MVR, without significant differences. Discussion: The current literature suggests that multivisceral pancreatectomies are feasible and may increase the number of completely resected patients. Morbidity and mortality are higher than after standard pancreatectomies, and these procedures should be reserved to selected patients in referral centers. Further studies on the role of neoadjuvant therapy in this setting are advisable. Received 20 March 2017; accepted 12 August 2017
Correspondence Niccolò Petrucciani, Department of Medical and Surgical Sciences, and Translational Medicine, Faculty of Medicine and Psychology, Sapienza University of Rome, St Andrea Hospital, via di Grottarossa 1035-9, 00189, Rome, Italy. E-mail: [email protected]
Introduction Surgical resection with adjuvant chemotherapy represents the only potentially curative treatment for pancreatic malignancies. Five-year survival approaches 25% for patients undergoing complete surgical resection, in combination with chemotherapy.1,2 In recent time, several strategies have been tried in an effort to increase the proportion of patients with pancreatic
HPB 2017, -, 1–8
malignancies offered potentially curative surgical resection. These have included the use of neoadjuvant chemoradiation therapies based on the potential theoretical benefits of reduction of tumor volume and increase of complete resection rate.3–5 Portal and mesenteric vein resections are recommended in patients with venous invasion.6,7 Arterial resections have been also proposed to increase the R0 resection rate, but they are still controversial.8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
2
Multivisceral resections have been proposed in selected patients with invasion of adjacent organs, in order to allow a complete tumor resection.9 Some authors have emphasized the increase of postoperative morbidity and mortality after multivisceral resections during pancreatectomy,9 whereas others have reported postoperative outcomes similar to standard resections.10 Nevertheless, the utility of this practice is still debated. Neither meta-analyses nor randomized trials have been published on this subject. Therefore, the aim of this study was to perform a systematic review of the literature to evaluate the role of multivisceral resections combined with pancreatectomy for pancreatic malignancies based on the short and long term outcomes as compared to standard pancreatectomy.
Methods Study selection A systematic literature search was performed using Embase, Medline, Cochrane, and PubMed databases to identify all studies published up to and including February 2017 that analyzed the postoperative outcomes and the overall survival of patients undergoing pancreatectomy combined with multivisceral resection for pancreatic malignancies. The systematic review was conducted according the PRISMA guidelines.11 The following MESH search headings were used: “multivisceral OR extended OR adjacent organs”, “pancreatic cancer OR pancreatic carcinoma OR pancreatic adenocarcinoma OR pancreatic tumor”, “pancreatectomy OR pancreaticoduodenectomy OR pancreatoduodenectomy OR distal pancreatectomy OR duodenopancreatectomy OR pancreatic resection”. The “related articles” function was used to broaden the search, and all abstracts, studies, and citations scanned were reviewed. Using the criteria of the PRISMA statement, two authors (GG and ES) independently searched the literature for relevant studies. The abstract screening led to identification of the papers eligible to systematic review. A third author (NP) carried out conflicts. Inclusion and exclusion criteria Studies included were required to compare characteristics, perioperative outcomes and/or overall survival of patients undergoing pancreatectomy with or without multivisceral resection for pancreatic malignancies. Multivisceral resections were defined as resection of one or more adjacent organs that are usually not resected during pancreatectomy. Splenectomy associated with distal pancreatectomy was not considered as a multivisceral resection. Venous resections were not considered as multivisceral resection. Only patients with pancreatic malignancies were included; patients submitted to pancreatectomy associated with multivisceral resection for tumors originating from other organs or for metastases to the pancreas from other origins were excluded from the analysis. Non-pancreatic cancers were excluded because comparison of survival data would be biased. Furthermore, also loco-regional spreading patterns are
HPB 2017, -, 1–8
different between different types of tumors (vascular invasion, nodal invasion), and may have an impact on perioperative decisions and outcomes. The following types of studies were not considered for inclusion in the systematic review: (i) studies in which the outcomes of interest for pancreatectomies with multivisceral resections were not reported or were impossible to calculate; (ii) studies where no comparison with standard pancreatectomies was reported; (iii) “how I do it” articles, animal studies, case reports, and non-English language studies. Quality assessment Three reviewers, according to the Newcastle–Ottawa Scale (NOS) for non-randomized studies,12 reviewed all selected studies for methodological quality. Final scores were reached by general consensus. The Cochrane Collaboration’s tool for assessing risk of bias in individual studies was also used by three independent authors,13 and conflicts were ruled out by discussion. Data extraction Data were extracted on the base of a piloted form and registered in a spreadsheet for comprehensive analysis. Two reviewers independently extracted the following information from each study: first author, year of publication, study design, study population characteristics, tumors’ characteristics, surgical techniques, and survival outcomes, according to the magnitude of the surgery (standard or multivisceral pancreatectomy). Outcomes of interest and definition All studies were abstracted for the following relevant data: patient baseline characteristics (age, gender), tumor characteristics (histology, staging, tumor volume), type of procedure and adjacent organ resection, definitive resection margins, lymph node involvement, postoperative morbidity and mortality, survival data. A meta-analysis was considered not appropriate because studies were heterogeneous. Clinical effectiveness was synthetized through a narrative review with full tabulation of results of the included studies.
Results Study selection Systematic search process is showed in Fig. 1. Three studies evaluating the outcomes of standard and multivisceral pancreatic resections for malignancies of the pancreas met inclusion criteria. The included studies were published between 2009 and 201514–16 and included patients accrued between 1992 and 2014. A total of 889 patients were included of which 176 (20%) underwent MVR resection. All the included studies were single institution retrospective reports, from Germany. New Castle– Ottawa scale score was 7 for each of the three studies. One study16 matched patients with or without multivisceral resection using type of surgery, age, gender and histology.
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
3
Figure 1 PRISMA 2009 flow diagram
9,10,17–20
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and MetaAnalyses: The PRISMA Statement. PLoS Med 6(7): e1000097. doi:10.1371/journal.pmed1000097. For more information, visit www.prismastatement.org.
In Table 1, patients’ and tumors’ characteristics of the two groups in each selected study are reported. In the study by Kulemann et al.14 distant metastases were considered a contraindication for resection. Concerning associated liver resections, Kulemann et al.14 reported liver resection only in case of direct contact/infiltration of the primary tumor but not in case of liver metastases. Hartwig et al.16 generally excluded patients with metastases. Patients with a solitary liver metastasis underwent pancreatectomy and metastasectomy only if they were young or in extremely good condition. Burdelski et al.15 performed liver resections for metastases only in young patients in extraordinarly good health, and hepatectomies varied from atypical resections to segmental resections. Surgical treatment and intraoperative variables are shown in Table 2. We reported the extent of multivisceral resection and the resected organs for each study in Table 3.
HPB 2017, -, 1–8
Postoperative outcomes and survival Postoperative outcomes are listed in Table 4. Few details regarding oncological treatment were reported. Hartwig et al.16 reported neoadjuvant therapy in 25 out of 202 patients (12%) in group 1 and 21 out of 101 (21%) in group 2 but did not specify the protocol. Burdelski et al.15 did not administer neoadjuvant treatment, but treated all their patients since 2004 with gemcitabine-based adjuvant chemotherapy. In the study by Kulemann et al.,14 30 patients (group not specified) out of a total of 228 had neoadjuvant chemotherapy (protocol not specified). In this study,14 postoperative treatment was heterogeneous, and the authors assert that in the early study period, patients did not routinely undergo adjuvant treatment. Lately, most patients underwent adjuvant chemotherapy after margin negative resections and chemoradiation after resections with a positive margin.14
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
4
Table
1 Patients’
and tumors’ characteristics for standard pancreatectomies and multivisceral resections associated with
pancreatectomies Variable
Standard pancreatectomies (%)
Multivisceral resections (%)
Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
Age, ya
66 (31–84)
nr
64 (33–88)
62 (40–89)
nr
63 (13–81)
Male gender
101 (49%)
nr
115 (57%)
9
nr
65 (64%)
1
nr
9 (3%)c
28 (14%)b,c
nr
0 (0%)c
1 (1%)b,c
2
nr
56 (18%)c
–
nr
0 (0%)c
T stage
3 4 N0
nr
209 (69%)
nr
c
45 (15%)
67 (32%)
c
165 (83%) 5 (3%)
117 (40%)
c c
c
nr
c
nr
66 (33%)
6
–
35 (64%)
c
87 (89%)c
20 (36%)
c
10 (10%)c
20 (36%)
c
34 (34%)
c
67 (66%)
N1
140 (68%)
180 (60%)
136 (67%)
14
35 (64%)
M1
0 (0%)
34 (11%)c
20 (10%)c
0
20 (36%)c
39 (39%)c
29 (1–130)
nr
nr
25 (18–60)
nr
nr
Tumor size, mm
a
N, number; Neoad., neoadjuvant; T, tumor stage according to Union of International Cancer Control/AJCC, N, nodal stage according to Union of International Cancer Control/AJCC; nr, not reported; y, years. a Data reported as median (range). b Data refers to T1 and T2 tumors. c Significant difference between standard and multivisceral resections.
Table 2 Surgical treatment, and intraoperative surgical variables for patients undergoing standard pancreatectomies and multivisceral
resections associated with pancreatectomies Variable
Standard pancreatectomies (%)
Multivisceral resections (%)
Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
PD
23 (11%)
nr
42 (21%)
2
30 (55%)
21 (21%)
PPPD
176 (85%)
nr
0 (0%)
11
0 (0%)
0 (0%)
SPD
0 (0%)
nr
0 (0%)
0
14 (25%)
0 (0%)
c
c
TP
9 (4%)
nr
40 (20%)
7
11 (20%)
20 (20%)
DP
0 (0%)
nr
120 (59%)
0
0 (0%)
60 (59%)
Op. time, mina
427 (184–720)c
340
260 (190–350)c
477 (346–760)c
351
320 (211–460)c
Intraop. Transf.a,b
0 (0–5)
2 (0–35)
nr
0 (0–2)
3 (0–44)
nr
Blood loss, La
nr
nr
1 (0–1)c
nr
nr
1 (1–2)c
N, number, PD, pancreaticoduodenectomy; PPPD, pylorus-preserving pancreaticoduodenectomy; SPD, subtotal pancreaticoduodenectomy; TP, total pancreatectomy; DP, distal pancreatectomy; Op., operative; min., minutes; Intraop, intraoperative; Transf., transfusions; L, liters. a Data reported as median (range). b Data reported as red blood packed cells. c Significant difference between standard and multivisceral resections.
Predictive and prognostic factors for short term morbidity an survival are shown in Table 5. No study identified risk factors for perioperative mortality.
Discussion Complete surgical resection with chemotherapy remains the only potentially curative treatment for pancreatic malignancies.21–23
HPB 2017, -, 1–8
Centralization of pancreatic surgery and advancement in surgical techniques and perioperative management had led surgeons to consider extended resections.24–29 Only a few studies have evaluated the role of multivisceral resection (MVR) for pancreatic cancer. A systematic search identified three heterogeneous series. Some authors have reported their results of multivisceral resections combined with pancreatectomy, including also portal or superior mesenteric vein resection into the multivisceral
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
5
group.18,19 Actually, venous resection has become part of the routine clinical practice. Approximately 20% of the patients who present for a pancreatic resection will concomitantly undergo a vascular resection according to national databases.20,21 Therefore, it is the authors opinion that venous resection should not be taken into account in the analysis of multivisceral pancreatectomies. Other series have analyzed multivisceral resection associated to pancreaticoduodenectomy20 or distal pancreatectomy.10 However the indications for resection did not include only pancreatic malignancies, but also various and heterogeneous benign and malignant diseases. It is the authors opinion that comparison of surgical resection of different tumors (bile duct, pancreas, duodenum, contiguous tumors) or even for chronic pancreatitis could result in bias not only concerning survival outcomes but also postoperative outcomes, because the technical problems and the spreading patterns of these tumors are different. The surgical approach and the potential morbidity may be influenced by the heterogeneity of such study population. The only study from a national database (National Surgical Quality Improvement Project, United Stated), by Bhayani et al.,9 analyzed 273 multivisceral resections, and highlighted the increased morbidity and mortality of MVR. This report included a heterogeneous population and analyzed data of patients operated on for several indications, such as pancreatic, periampullary and duodenal cancers and benign diseases of the pancreas.
Table 3 Numbers and types of resected organs in patients who
underwent pancreatectomies with multivisceral resections Kulemann
Burdelski
Hartwig
N. MVR
20 (9%)
55 (15%)
101 (33%)
1 additional resected organ
nr
13 (23%)
17 (17%)
2 additional resected organs
nr
24 (44%)
46 (47%)
3 additional resected organs
nr
11 (20%)
23 (23%)
4 or more additional resected organs
nr
7 (13%)
15 (15%)
Resected organs in multivisceral resections Stomach
6
32
34
Colon
7
22
38
Liver
5
24
19
Kidney
0
17
12
Diaphragm
0
11
0
Small bowel
1
5
7
Adrenal
1
0
28
Additional portal vein resection
12
nr
21
Additional arterial resection
nr
nr
17
nr, not reported; N., number, MVR, multivisceral resections.
Table 4 Short-term outcomes and survival for patients undergoing standard pancreatectomies and multivisceral resections associated with
pancreatectomies Variable
Standard pancreatectomies (%) Kulemann
Burdelski
Hartwig
Kulemann
Burdelski
Hartwig
N
208 (91%)
303 (85%)
202 (67%)
20 (9%)
55 (15%)
101 (33%)
R0
157 (76%)b
225 (74%)b
122 (61%)
18b
30 (55%)b
58 (57%)
R1
49 (24%)b
58 (19%)b
64 (32%)
2b
21 (39%)b
29 (29%)
b
b
Multivisceral resections (%)
b
b
R2
0 (0%)
13 (7%)
0
4 (8%)
14 (14%)
Postop. transfa
nr
0 (0–38)c
nr
nr
1 (0–78)c
nr
Morbidity
104 (50%)
112 (37%)b
86 (43%)b
13
38 (69%)b
56 (56%)b
PPH
nr
19 (6%)
16 (8%)
nr
4 (7%)
10 (10%)
POPF
nr
27 (9%)
14 (7%)
nr
5 (9%)
11 (11%)
GI bleeding
nr
7 (2%)b
nr
nr
4 (7%)b
20 (6%)
Colonic fistula
nr
1 (0.3%)
Peritonitis
nr
nr
b
nr b
nr
nr
9 (16%)
3 (2%)
nr
Nr
6 (6%)
nr
DGE
nr
nr
31 (15%)
nr
Nr
17 (17%)
Re−laparotomy
19 (9%)
39 (13%)
17 (8%)b
3
10 (18%)
22 (22%)b
In hosp. mortality
8 (4%)
12 (4%)
7 (4%)
2
4 (7%)
7 (7%)
Survival, monthsa
20
21
23
12
16
20
N, number; R, resection margin according to Union of International Cancer Control/AJCC; Postop., postoperative; nr, not reported, POPF, postoperative pancreatic fistula, DGE, delayed gastric emptying, in hosp., in hospital, PPH, postoperative hemorrhage. Data reported as median (range). b Significant difference between standard and multivisceral resections. c Data reported as units of red blood packed cells, or # liters of transfused blood. a
HPB 2017, -, 1–8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
6
Table 5 Predictive factors of postoperative morbidity and negative prognostic factor for overall survival
Predictive factors of postoperative morbidity Type of analysis
Kulemann
Burdelski
Hartwig
Univariate
–
Intraop. transfusions
Absence of liver resection
Colic resection
Operative time >300 min
Kidney resection Liver resection Multivariable
–
Intraop. transfusions
Operative time >300 min
Kidney resection Negative prognostic factors for overall survival Type of analysis
Kulemanna
Burdelskib
Univariate
Nodal disease
T stage
Resection margin +
Kidney resection
Intraop. Blood transfusions
4 organs resected
Multivisceral resections
Postop. transfusions
Nodal disease
T stage
Hartwig
ICU stay >2 days Multivariable
Resection margin + Intraop. Blood transfusions Multivisceral resections Intraop., intraoperative; postop., postoperative. a The analyses refer to the entire population (group 1 + group 2). b The analyses refer to group 2.
Patients who underwent multivisceral resections had more advanced disease, with a higher stage in two of the studies15,16 and more frequent need of total pancreatectomy in the latter.14 It is interesting to note that in two out of three studies, liver resection for metastases was not an absolute contraindication, and it was reserved to oligometastatic young and/or very healthy selected patients.15,16 In these two studies the rate of M1 resection was 36%15 and 39%16 in patients undergoing multivisceral resections which is clearly outside what is reported in other studies. In these two studies, the rate of M1 disease was also considerable in patients having standard pancreatectomy, 11%15 and 10%,16 respectively.15,16 It is interesting to observe that survival results are not worst in the two studies including highly selected oligometastatic patients. Another interesting aspect was the use of neoadjuvant therapy in these series. Nowadays, the use of neoadjuvant therapy for borderline resectable pancreatic cancer has increased, and NCCN guidelines no longer recommend upfront resection in this setting.30–32 A recent meta-analysis has shown that 65% of borderline patients undergo resection after neoadjuvant therapy, with 57% achieving R0 status and a median survival of 26 months for resected patients.3 The utility of neoadjuvant therapy has also been advocated for resectable and for locally advanced pancreatic cancer.33–35 It is not specified if some patients included in the reviewed articles had resectable or borderline resectable or locally
HPB 2017, -, 1–8
advanced disease, but it seems logical to hypothesize that tumors requiring multivisceral resections have a higher risk of R1 or R2 resection, and may benefit from neoadjuvant therapies. Actually, only a minority of the patients included in this review had neoadjuvant treatment. For resection margins, results were discordant: Burdelski et al.15 report significant more frequent margin involvement in case of multivisceral resection, whereas in the study of Kulemann et al.,14 MVR had significantly lower rate of R+ (Not associated with a survival advantage, however). Hartwig et al.16 did not report significant differences. It is difficult to interpret these data, which may also reflect different protocols of pathological analysis, not always following the Leeds criteria, as stated by Kulemann et al.14 This review clearly shows that multivisceral resections have specific complications, which does not occur after standard pancreatectomies. Colonic fistula was as frequent as 16% of patients in the study by Burdelski et al.,15 who suggest liberal use of protective ileostomies to reduce the impact of such complication. The frequency and types of complications of multivisceral resection combined to pancreatectomy suggests the need of a tertiary referral centers with high level of expertise not only in pancreatic surgery but also in digestive surgery. The risk of these procedures are remarkable, with mortality rates higher than 5% and approaching 10%, and morbidity around 60%, data which should be taken into account in the patients’ counseling and decisional
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
7
process. Median survival after MVR ranged from 12 to 20 months. In all included studies survival was lower than after standard pancreatectomy, even if no significant differences were found. This systematic review shows that multivisceral pancreatectomies are feasible, even if postoperative morbidity and mortality are higher than after standard pancreatectomies. The overall survival was not significantly different in the two groups. This approach may increase the number of patients undergoing radical resection, and offers the possibility for curative resection to patients otherwise considered non resectable for infiltration of contiguous organs. Further studies analyzing the role of neoadjuvant therapies in patients with advanced tumor invading surrounding organs are advisable. Neoadjuvant therapy may permit better patients’ selection, excluding those who became non resectable for rapid tumor progression, and increase R0 rate, in patients with partial or complete tumor response, with potential survival benefit. Concerning patients’ management, it is important to highlight the high morbidity rate of MVR, and the potential complications of additional organ resections, which are frequent, in order to promptly detect and treat them.
standard pancreatectomies. The current literature suggests that multivisceral pancreatectomies are feasible and may increase the number of completely resected patients. Morbidity and mortality are higher than after standard pancreatectomies, and these procedures should be reserved to very selected patients in referral centers. Further studies on the role of neoadjuvant therapy in this setting are advisable.
Limits The study has several limitations. First of all, the included studies were all retrospective. However, all of them were conducted in referral centers for pancreatic cancer surgery. Second, two of the studies included a small percentage of patients with M1 pancreatic disease, which represent a contraindication for surgery, except in the setting of clinical studies. Third, only a low percentage of patients underwent neoadjuvant chemotherapy. Fourth, indications and protocols of neoadjuvant and adjuvant chemotherapies are not extensively detailed, and the treatments were heterogeneous. Fifth, all the studies were conducted in a single country (i.e., Germany), which may limit the generalizability of the data. Furthermore, clear definition of R0 margins was not clearly specified by the included studies (even if all the included institutions are referral pancreatic cancers, with expert pathologists). Strengths of the study include: (i) the methodological quality of the review, based on PRISMA guidelines, and including systematic literature search and quality assessment of the studies; (ii) the inclusion of referral centers; (iii) the high degree of concordance of the conclusion of each study which gave value of the review’s conclusion.
3. Tang K, Lu W, Qin W, Wu Y. (2016) Neoadjuvant therapy for patients
Funding No funding received. Conflict of interest None declared. References 1. D’Angelo FA, Antolino L, La Rocca M, Petrucciani N, Magistri P, Aurello P et al. (2016) Adjuvant and neoadjuvant therapies in resectable pancreatic cancer: a systematic review of randomized controlled trials. Med Oncol 33:28. 2. Strosberg JR, Cheema A, Weber JM, Ghayouri M, Han G, Hodul PJ et al. (2012) Relapse-free survival in patients with nonmetastatic, surgically resected pancreatic neuroendocrine tumors: an analysis of the AJCC and ENETS staging classifications. Ann Surg 256:321–325. with borderline resectable pancreatic cancer: a systematic review and meta-analysis of response and resection percentages. Pancreatology 16:28–37. 4. McClaine RJ, Lowy AM, Sussman JJ, Schmulewitz N, Grisell DL, Ahmad SA. (2010) Neoadjuvant therapy may lead to successful surgical resection and improved survival in patients with borderline resectable pancreatic cancer. HPB 12:73–79. 5. Khushman M, Dempsey N, Maldonado JC, Loaiza-Bonilla A, Velez M, Carcas L et al. (2015) Full dose neoadjuvant FOLFIRINOX is associated with prolonged survival in patients with locally advanced pancreatic adenocarcinoma. Pancreatology 15:667–673. 6. Ramacciato G, Nigri G, Petrucciani N, Pinna AD, Ravaioli M, Jovine E et al. (2016 Feb 18) Pancreatectomy with mesenteric and portal vein resection for borderline resectable pancreatic cancer: multicenter study of 406 patients. Ann Surg Oncol [Epub ahead of print]. 7. Ramacciato G, Mercantini P, Petrucciani N, Giaccaglia V, Nigri G, Ravaioli M et al. (2009) Does portal-superior mesenteric vein invasion still indicate irresectability for pancreatic carcinoma? Ann Surg Oncol 16:817–825. 8. Jegatheeswaran S, Baltatzis M, Jamdar S, Siriwardena AK. (2017) Superior mesenteric artery (SMA) resection during pancreatectomy for malignant disease of the pancreas: a systematic review. HPB 19: 483–490. 9. Bhayani NH, Enomoto LM, James BC, Ortenzi G, Kaifi JT, Kimchi ET et al. (2014) Multivisceral and extended resections during pancreato-
Conclusions
duodenectomy increase morbidity and mortality. Surgery 155:567–574.
The systematic review included three studies evaluating the utility of multivisceral resection combined with pancreatectomy for pancreatic malignancies. Patients treated with multivisceral resections had a morbidity rate ranging from 56% to 69% and a mortality rate ranging from 7% to 10%. Survival ranged from 12 to 20 and was not significantly different than those reported for
HPB 2017, -, 1–8
10. Irani JL, Ashley SW, Brooks DC, Osteen RT, Raut CP, Russell S et al. (2008) Distal pancreatectomy is not associated with increased perioperative morbidity when performed as part of a multivisceral resection. J Gastrointest Surg 12:2177–2182. 11. Preferred Reporting Items for Systematic Reviews and MetaAnalyses
(PRISMA)
website,
http://www.prisma-statement.org.
(accessed 7 February 2017).
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002
HPB
8
12. The Newcastle-Ottawa Scale (NOS) for assessing the quality of
24. Gooiker GA, van der Geest LG, Wouters MW, Vonk M, Karsten TM,
nonrandomised studies in meta-analyses. http://www.ohri.ca/programs/
Tollenaar RA et al. (2011) Quality improvement of pancreatic surgery by
clinical_epidemiology/oxford.asp. (accessed 7 February 2017).
centralization in the western part of The Netherlands. Ann Surg Oncol
13. The Cochrane Collaboration’s tool for assessing risk of bias. http://ohg. cochrane.org/sites/ohg.cochrane.org/files/uploads/Risk%20of%20bias %20assessment%20tool.pdf. (accessed 7 March 2016). 14. Kulemann B, Hoeppner J, Wittel U, Glatz T, Keck T, Wellner UF et al. (2015) Perioperative and long-term outcome after standard pancreaticoduodenectomy, additional portal vein and multivisceral resection for pancreatic head cancer. J Gastrointest Surg 19:438–444. 15. Burdelski CM, Reeh M, Bogoevski D, Gebauer F, Tachezy M, Vashist YK et al. (2011) Multivisceral resections in pancreatic cancer: identification of risk factors. World J Surg 35:2756–2763. 16. Hartwig W, Hackert T, Hinz U, Hassenpflug M, Strobel O, Büchler MW et al. (2009) Multivisceral resection for pancreatic malignancies: riskanalysis and long-term outcome. Ann Surg 250:81–87. 17. Abu Hilal M, McPhail MJ, Zeidan BA, Jones CE, Johnson CD, Pearce NW. (2009) Aggressive multi-visceral pancreatic resections for locally advanced neuroendocrine tumours. Is it worth it? JOP 10:276–279. 18. Panzeri F, Marchegiani G, Malleo G, Malpaga A, Maggino L, Marchese T
18:1821–1829. 25. Massucco P, Ribero D, Sgotto E, Mellano A, Muratore A, Capussotti L. (2009) Prognostic significance of lymph node metastases in pancreatic head cancer treated with extended lymphadenectomy: not just a matter of numbers. Ann Surg Oncol 16:3323–3332. 26. Hartel M, Wente MN, Di Sebastiano P, Friess H, Büchler MW. (2004) The role of extended resection in pancreatic adenocarcinoma: is there good evidence-based justification? Pancreatology 4:561–566. 27. Stitzenberg KB, Watson JC, Roberts A, Kagan SA, Cohen SJ, Konski AA et al. (2008) Survival after pancreatectomy with major arterial resection and reconstruction. Ann Surg Oncol 15:1399–1406. 28. Hartwig W, Gluth A, Hinz U, Bergmann F, Spronk PE, Hackert T et al. (2015) Total pancreatectomy for primary pancreatic neoplasms: renaissance of an unpopular operation. Ann Surg 261:537–546. 29. Shrikhande SV, Kleeff J, Reiser C, Weitz J, Hinz U, Esposito I et al. (2007) Pancreatic resection for M1 pancreatic ductal adenocarcinoma. Ann Surg Oncol 14:118–127.
et al. (2016 Oct 27) Distal pancreatectomy associated with multivisceral
30. Blazer M, Wu C, Goldberg RM, Phillips G, Schmidt C, Muscarella P et al.
resection: results from a single centre experience. Langenbecks Arch
(2015) Neoadjuvant modified (m) FOLFIRINOX for locally advanced
Surg [Epub ahead of print].
unresectable (LAPC) and borderline resectable (BRPC) adenocarcinoma
19. Sasson AR, Hoffman JP, Ross EA, Kagan SA, Pingpank JF, Eisenberg BL. (2002) En bloc resection for locally advanced cancer of the pancreas: is it worthwhile? J Gastrointest Surg 6:147–157. 20. Nikfarjam M, Sehmbey M, Kimchi ET, Gusani NJ, Shereef S, Avella DM et al. (2009) Additional organ resection combined with pancreaticoduodenectomy does not increase postoperative morbidity and mortal-
of the pancreas. Ann Surg Oncol 22:1153–1159. 31. Rose JB, Rocha FG, Alseidi A, Biehl T, Moonka R, Ryan JA et al. (2014) Extended
chemotherapy
for
borderline
resectable
and survival. Ann Surg Oncol 21:1530–1537. 32. NCCN Clinical Practice Guidelines in Oncology, Pancreatic Adenocarcinoma,
ity. J Gastrointest Surg 13:915–921. 21. Sinn M, Striefler JK, Sinn BV, Sallmon D, Bischoff S, Stieler JM et al.
neoadjuvant
pancreatic cancer demonstrates promising postoperative outcomes
at
https://www.nccn.org/professionals/physician_gls/pdf/
pancreatic.pdf, (accessed 7 February 2017).
(2013) Does long-term survival in patients with pancreatic cancer
33. Christians KK, Heimler JW, George B, Ritch PS, Erickson BA,
really exist? Results from the CONKO-001 study. J Surg Oncol 108:
Johnston F et al. (2016) Survival of patients with resectable pancreatic cancer who received neoadjuvant therapy. Surgery 159:893–900.
398– 402. 22. Barugola G, Partelli S, Crippa S, Butturini G, Salvia R, Sartori N et al.
34. de Geus SW, Evans DB, Bliss LA, Eskander MF, Smith JK, Wolff RA
(2013) Time trends in the treatment and prognosis of resectable
et al. (2016) Neoadjuvant therapy versus upfront surgical strategies in
pancreatic cancer in a large tertiary referral centre. HPB 15:958–964.
resectable pancreatic cancer: a Markov decision analysis. Eur J Surg
23. Riall TS, Cameron JL, Lillemoe KD, Winter JM, Campbell KA,
Oncol 42:1552–1560.
Hruban RH et al. (2006) Resected periampullary adenocarcinoma:
35. Strobel O, Berens V, Hinz U, Hartwig W, Hackert T, Bergmann F et al.
5-year survivors and their 6- to 10-year follow-up. Surgery 140:
(2012 Sep) Resection after neoadjuvant therapy for locally advanced,
764–772.
“unresectable” pancreatic cancer. Surgery 152(3 Suppl. 1):S33–S42.
HPB 2017, -, 1–8
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Petrucciani N, et al., Pancreatectomy combined with multivisceral resection for pancreatic malignancies: is it justified? Results of a systematic review, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.08.002