- Email: [email protected]
A systematic review of radical antegrade modular pancreatosplenectomy for adenocarcinoma of the body and tail of the pancreas
HPB
http://dx.doi.org/10.1016/j.hpb.2016.07.014
REVIEW ARTICLE
A systematic review of radical antegrade modular pancreatosplenectomy for adenocarcinoma of the body and tail of the pancreas Yanming Zhou1, Bin Shi2, Lupeng Wu1 & Xiaoying Si1 1
Department of Hepatobiliary & Pancreatovascular Surgery, First Affiliated Hospital of Xiamen University, Xiamen, and 2General Intensive Care Unit, Songjiang Central Hospital, First People’s Hospital of Shanghai Jiaotong University, Shanghai, China
Abstract Background: To assess the published evidence on clinical outcomes following radical antegrade modular pancreatosplenectomy (RAMPS) for adenocarcinoma in the body or tail of the pancreas. Method: PubMed and Chinese Biomedical Literature databases were searched. The results of comparisons between RAMPS and standard retrograde pancreatosplenectomy (SRPS) were analyzed by meta-analytical techniques. Results: The literature search identified 13 observational studies involving 354 patients undergoing RAMPS. The overall morbidity and 30-day mortality was 40% and 0% respectively. The R0 resection rate was 88%; the median number of retrieved lymph nodes was 21; and the median 5-year overall survival rate was 37%. The result of meta-analysis showed that RAMPS was associated with a significantly less intraoperative bleeding [weighted mean difference −195.2 (95% confidence interval (CI) −223.27 to −167.13); P < 0.001], a greater number of retrieved lymph nodes [odds ratio (OR) 6.19 (95% CI 3.72 to 8.67); P < 0.001] and a higher percentage of R0 resection [OR 2.46 (95% CI 1.13 to 5.35); P = 0.02] as compared with SRPS. Conclusion: The current literature provides supportive evidence that RAMPS is a safe and effective procedure for adenocarcinoma in the body or tail of the pancreas, and is oncologically superior to SRPS. Received 28 June 2016; accepted 28 July 2016
Correspondence Yanming Zhou, Department of Hepatobiliary & Pancreatovascular Surgery, First Affiliated Hospital of Xiamen University, 55 Zhenhai Road, Xiamen 361003, FJ, China. Tel: +86 0592 2139708. Fax: +86 0592 2137289. E-mail: [email protected]
Introduction Adenocarcinoma in the body or tail of the pancreas is conventionally resected by the standard retrograde pancreatosplenectomy (SRPS) performed in the left-to-right direction with mobilization of the spleen first, and then resection of the posterior aspect of the pancreas from the tail to the body. However, SRPS is associated with a high positive tangential margin rate, devoid of the described lymph node drainage of the organ. To overcome these problems, Strasberg et al.1 in 2003 introduced a modified technique of SRPS called radical antegrade modular pancreatosplenectomy (RAMPS) in which division of the neck of the pancreas and splenic vessels and a celiac node dissection are Yanming Zhou and Bin Shi contributed equally to this work.
HPB 2017, 19, 10–15
performed first, followed by dissection proceeding from right-toleft in 1 of the 2 posterior dissection planes, depending on the extent of penetration of the tumor. However, only a few studies have reviewed the experience and practice of RAMPS.2–10 In addition, data comparing this procedure with SRPS are limited, and therefore the potential value of RAMPS has not been clearly demonstrated. The aim of this systematic review was to assess the published evidence on clinical outcomes following RAMPS for adenocarcinoma in the body or tail of the pancreas.
Methods This study was performed in accordance with the guidelines of preferred reporting items for systematic reviews and metaanalyses (PRISMA) 2009.11
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
11
Literature search strategy and study identification An electronic search was performed of the Pubmed and Chinese Biomedical Literature database from the date of the earliest report of RAMPS in 2003 to May 2016 using the following keyword: “radical antegrade modular pancreatosplenectomy.” No language restriction was applied. Reference lists of relevant articles were further searched manually to check for additional studies. Studies reporting the outcomes following the RAMPS in patients with pancreatic adenocarcinoma were included for analysis. To ensure that the series reviewed reflect consistent surgical approach, only study involving more than 5 patients were included in the systematic review of overall outcome of RAMPS. Reviews, conference abstracts, non-human studies, case report were excluded. In cases of duplicated studies with overlapping patients, only the most recent publication with accumulating numbers of patients or increased lengths of follow-up was considered. Two investigators (YZ and BS) independently appraised each eligible article using predefined criteria. Discrepancies between the two reviewers were resolved by discussion and consensus. Data were extracted on the first author, country, year of publication, sample size, study design, population characteristics,
duration of operation, estimated blood loss, proportion of radical (R0) resection, morbidity, incidence and severity of pancreatic fistula as defined by the International Study Group on Pancreatic Fistula (ISGPF),12 30-day mortality, and survival. The level of evidence of each study was categorized according to the Evidence-Based Medicine Levels of Evidence.13 Statistical analysis Descriptive statistics were performed and data are expressed as mean or median (interquartile range) where appropriate. A meta-analysis of the comparative studies of RAMPS and SRPS was undertaken with Review Manager (RevMan) software, version 5.1 (The Cochrane Collaboration, Software Update, Oxford). Odds ratio (OR) or weighted mean difference (WMD) with a 95% confidence interval (95% CI) were calculated for dichotomous variables and continuous variables respectively. Heterogeneity was assessed using the c2 test and I2. When the heterogeneity was not significant (I2 <50%), a fixed-effects model was used for the pooled analysis. Otherwise, a random-effects model was used. Statistical significance was set at P < 0.05.
Figure 1 Study selection
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
12
Results
Discussion
Systematic review As shown in Fig. 1, the literature search initially yielded 41 articles, of which 13 studies comprising a total of 354 patients fulfilling the eligibility criteria were included in this systematic review and are summarized in Table 1.4–10,14–19 Agreement for these studies was 100% between the two reviewers. No randomised controlled trials (RCT) or meta-analyses were identified. All these studies were observational in nature and classified as level-4 evidence. Five studies were conducted in Japan,4,8,15,16,18 three in Korea,5,9,10 two in USA,13,19 one in Italy,6 one in France,7 and one in China.17 The surgical outcomes of the 13 studies are summarized in Table 2. Of the 354 patients, 330 underwent open RAMPS and 24 underwent a laparoscopic or robotic approach. The overall mortality was 0%. Pancreatic fistula, the most critical complication following pancreatectomy, was seen in 70 (20%) patients. Fifty-five pancreatic fistulas were graded according to the ISGPF. Most were grade A (20, 36%) or B (23, 42%); only five (1%) were grade C and seven (13%) were graded as B–C.
This is the first review to assess existing evidence on clinical outcomes following RAMPS for adenocarcinoma in the body or tail of the pancreas. The results show that the perioperative mortality rate was zero, and the median 5-year OS was 37%, confirming the safety and efficacy of this procedure. The result of meta-analysis shows that the operating time was comparable between RAMPS and SRPS, though a longer operative time was observed in some smaller series6,10 of RAMPS group, suggesting that this may reflect the learning curve of the surgeons. As dissection during RAMPS commences from rightto-left with early division of the neck of the pancreas, it provides more superior access to control the major blood vessels including the splenic, adrenal and renal veins, blood loss was therefore reduced in RAMPS compared with SRPS. Although no survival benefit of an extended lymphadenectomy has been shown in pancreatic adenocarcinoma resection,20 thorough pathologic evaluation of lymph nodes may contribute to accurate staging. A recent review of 499 patients reported significant impact of the total number of examined lymph nodes on the estimation of stage-based survival after curative pancreatectomy for pancreatic adenocarcinoma.21 Survival for nodenegative (pN0) patients with <11 lymph nodes examined was worse than for pN0 patients with 11 lymph nodes with a 3-year survival rate of 32% vs. 50%, suggesting that metastatic nodes were missed by sampling insufficient lymph nodes. These investigators concluded that pathologic assessment of 11 lymph nodes in resected specimens is needed for accuracy of pancreatic adenocarcinoma staging. During SRPS, nodes at the roots of celiac or superior mesenteric artery regions are not removed. In contrast, RAMPS permits resection of these nodes in a controlled manner.1 The current systematic review demonstrates that the median number of total retrieved lymph nodes after RAMPS was 21 (range, 11–30), suggesting that RAMPS fulfills the criteria of adequate node dissection. R0 resection is an important factor determining survival in patients with pancreatic adenocarcinoma. The posterior margin is the predominant site of margin involvement of pancreatic body-tail adenocarcinoma.22 Published literatures with large sample size (n > 100) of distal adenocarcinoma demonstrate R0 resection rate of 50%–74%.23,24 The R0 resection rate of up to 88% after RAMPS in this systematic review is relatively high. In comparative studies, the cumulative R0 resection rate was 90% after RAMPS and 80% after SRPS with a pooled OR of 2.46 (95% CI 1.13–5.35). In the SRPS procedure, dissection proceeds from left-to-right, making it difficult to obtain a clear concept of where the posterior plane of dissection is, especially whenever the patient is deep-chested or obese.1 On the other hand, RAMPS enables the surgeon to set up the posterior plane of dissection more easily, thus allowing for more radical resection by early identification of the renal vein and the anterior surface of the adrenal vein.
Meta-analysis of RAMPS vs. SRPS Four studies compared RAMPS (n = 105) with SRPS (n = 131).6,10,14,18 Table 3 summarizes the outcomes of the metaanalysis. Compared with the SRPS group, the RAMPS group exhibited less intraoperative bleeding, a greater number of total retrieved lymph nodes, and a higher percentage of R0 resection (Fig. 2a–c). No significant difference was seen in other outcomes of interest. The funnel plot for retrieved lymph nodes was symmetric, indicating the absence of publication bias (Fig. 3). Table 1 Baseline characteristics of the studies included in system-
atic review Reference
Study period
Ikegami et al.4 Chang et al.
5
Age (year)a
Men, n TS (mm)a
2009–2011 6
67
4
30 41
2005–2009 24
60
13
Lartorre et al.6
2003–2011 8
61
5
–
Rosso et al.7
2008–2012 10
62
3
46
Kitagawa et al.8
2007–2012 24
67
15
35
Lee et al.9
2007–2010 12
64
7
28
Park et al.10
2007–2010 38
63
23
31
Trottman et al.14
2004–2011 6
–
–
–
Kawabata et al.15
2013–2014 11
68
6
33
Murakawa et al.16 2000–2014 49
68
31
38
Wu et al.17
2013–2014 35
65
13
46
2000–2014 53
69
31
–
Grossman et al.19 1999–2013 78
67
32
47
Abe et al.
18
Total/Median or % a
N
354 66 183 37 (9–43) (62–68) (53%) (31–46)
Median or mean; TS, tumor size.
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
13
Table 2 The surgical outcomes of radical antegrade modular pancreatosplenectomy.
Reference Ikegami et al.4 Chang et al.
5
Lartorre et al.6 Rosso et al.7
Operative time (min)a
Blood loss (mL)a BT n
258
226
–
Morbidity n PLN, n 4
–
RLN, na
R0 R n
MS (months) 5-year OS (%)
–
6
11
–
–
–
1
9
17
21
22
18
–
315
–
2
2
–
21
7
14
26
424
–
1
4
7
17
9
21
–
387
371
–
14
13
24
21
–
53
Lee et al.9
324
446
2
3
3
11
12
–
56
Park et al.10
210
325
–
7
22
14
34
–
40
Trottman et al.14
300
500
–
5
–
12
4
–
–
15
423
500
0
4
10
26
8
–
–
Murakawa et al.16
278
850
11
20
27
15
41
23
27
Wu et al.17
160
190
–
10
–
30
32
–
–
267
485
0
19
28
28
48
47
37
252
629
17
41
37
20
66
25
25
Kitagawa et al.
8
Kawabata et al.
Abe et al.
18
Grossman et al.19
Total/Median or % 289 (254–371) 465 (300–532)
34 (14%) 142 (40%)
164 (55%) 21 (14–26) 310 (88%) 21 (14–25)
37 (26–53)
a
Median or mean; BT, blood transfusion; PLN, number of patients with positive lymph nodes; RLN, retrieved lymph nodes; R0 R, R0 resection; MS, median survival; OS, overall survival.
Table 3 Results of the meta-analysis.
Outcome of interest
No. of studies
No. of patients
OR/WMD (95% CI)
P-value
I2 (%)
Operation time (min)
3
211
−17.03 (−94.46, 60.40)
0.67
94
Blood loss (mL)
3
211
−195.2 (−223.27, −167.13)
<0.001
0
Overall morbidity
4
236
0.97 (0.54, 1.77)
0.93
0
Overall PF
4
236
0.45 (0.19, 1.10)
0.08
0
ISGPF B + C PF
2
185
0.50 (0.18, 1.38)
0.18
0
Retrieved lymph nodes
4
236
6.19 (3.72, 8.67)
<0.001
0
R0 resection
4
223
2.46 (1.13, 5.35)
0.02
0
5-year OS (%)
3
210
1.77 (0.53, 5.93)
0.35
67
WMD, weighted mean difference; OR, odds ratio; CI, confidence interval; PF, pancreatic fistula; ISGPF, International Study Group of Pancreatic Fistula; OS, overall survival.
The presence of circulating tumor cells in the portal vein has been found to be associated with a higher rate of liver metastasis after pancreatic adenocarcinoma surgery.25 Left-to-right mobilization and handling of the pancreato-splenic specimen without early ligation of the draining blood vessels during SRPS could increase the risk of cancer cell shedding into the portal vein. In this context, RAMPS complies with the principle of no-touch isolation. CellSearch™ method might provide the evidence for the biological benefit of this procedure.25 The result of the present meta-analysis shows that RAMPS does not seem to improve survival as compared with SRPS. It is apparent that the number of patients enrolled in the current study is small with insufficient power to detect a possible difference. Mitchem et al. calculated that to compare two treatments with 5-year survival rates of 20% and 35% at the 95% CI level, a total recruitment of 556 patients was required.3 HPB 2017, 19, 10–15
The main limitation of this review is the relatively low level of evidence owing to the lack of RCT on the subject. All of the included studies were observational nature with an inherent risk of bias. There are many differences between the studies in terms of the patient selection, disease stage, surgical skill, and perioperative care. The definition of R0 resection was inhomogeneous; 12 studies simply reported that ‘surgical margins’ were negative without specified criteria,4–6,14,16–18 whereas others referred to the ‘absence of microscopic tumor cells’7,10 or ‘the tumor is >1 mm from all inked margins’.15,19 Meta-analysis of observational studies may either exaggerate or underestimate the magnitude of measured effect.26 Another limitation is the small sample size. As adenocarcinoma of the body and tail of the pancreas is usually detected in a relatively late stage because of its delayed presentation, it is often unresectable at the time of confirmed diagnosis. In addition, as
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
14
Figure 2 Results of the meta-analysis: (a) intraoperative bleeding; (b) number of total retrieved lymph nodes; (c) R0 resection
prospective RCTs comparing RAMPS with SRPS are required to obtain a more convincing conclusion. Conflicts of interest The authors declare that they have no potential competing interests. The authors received no funding for this work. References 1. Strasberg SM, Drebin JA, Linehan D. (2003) Radical antegrade modular pancreatosplenectomy. Surgery 133:521–527. http://dx.doi.org/10.1067/ msy.2003.146. 2. Strasberg SM, Linehan DC, Hawkins WG. (2007) Radical antegrade modular pancreatosplenectomy procedure for adenocarcinoma of the Figure 3 Funnel plot shows symmetry for retrieved lymph nodes
suggesting the absence of publication bias
body and tail of the pancreas: ability to obtain negative tangential margins. J Am Coll Surg 204:244–249. http://dx.doi.org/10.1016/ j.jamcollsurg.2006.11.002. 3. Mitchem JB, Hamilton N, Gao F, Hawkins WG, Linehan DC,
RAMPS is a novel surgical approach, surgeons are often reluctant to employ it, instead awaiting evidence of its safety and efficacy. In conclusion, RAMPS is a safe and effective procedure for adenocarcinoma in the body or tail of the pancreas and is oncologically superior to SRPS. Although there are no data supporting a prolonged survival, more lymph nodes and an apparently greater R0 rate of resection seem evident. As all the evidence in this meta-analysis comes from observational studies involving a relatively small number of patients, larger-sample HPB 2017, 19, 10–15
Strasberg SM. (2012) Long-term results of resection of adenocarcinoma of the body and tail of the pancreas using radical antegrade modular pancreatosplenectomy procedure. J Am Coll Surg 214:46–52. http:// dx.doi.org/10.1016/j.jamcollsurg.2011.10.008. 4. Ikegami T, Maeda T, Oki E, Kayashima H, Ohgaki K, Sakaguchi Y et al. (2011) Antegrade en bloc distal pancreatectomy with plexus hanging maneuver. J Gastrointest Surg 15:690–693. http://dx.doi.org/10.1007/ s11605-010-1382-9. 5. Chang YR, Han SS, Park SJ, Lee SD, Yoo TS, Kim YK et al. (2012) Surgical outcome of pancreatic cancer using radical antegrade modular
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
15
pancreatosplenectomy
procedure.
World
J
Gastroenterol
18:
body and tail of the pancreas. BMC Surg 15:120. http://dx.doi.org/
6. Latorre M, Ziparo V, Nigri G, Balducci G, Cavallini M, Ramacciato G.
17. Wu WG, Wu XS, Li ML, Wang XA, Shu YJ, Weng H et al. (2015) Ante-
(2013) Standard retrograde pancreatosplenectomy versus radical
grade pancreatosplenectomy for curative resection of adenocarcinoma
5595–5600. http://dx.doi.org/10.3748/wjg.v18.i39.5595.
10.1186/s12893-015-0107-0.
antegrade modular pancreatosplenectomy for body and tail pancreatic
of the left pancreas. Chin J Pract Surg 35:296–298. 18. Abe T, Ohuchida K, Miyasaka Y, Ohtsuka T, Oda Y, Nakamura M. (2016)
adenocarcinoma. Am Surg 79:1154–1158. 7. Rosso E, Langella S, Addeo P, Nobili C, Oussoultzoglou E, Jaeck D
Comparison of surgical outcomes between radical antegrade modular
et al. (2013) A safe technique for radical antegrade modular pancrea-
pancreatosplenectomy (RAMPS) and standard retrograde pancreatos-
tosplenectomy with venous resection for pancreatic cancer. J Am Coll
plenectomy (SPRS) for left-sided pancreatic cancer. World J Surg.
Surg 217:e35–39. http://dx.doi.org/10.1016/j.jamcollsurg.2013.08.007.
http://dx.doi.org/10.1007/s00268-016-3526-x [Epub ahead of print].
8. Kitagawa H, Tajima H, Nakagawara H, Makino I, Miyashita T,
19. Grossman JG, Fields RC, Hawkins WG, Strasberg SM. (2016) Single
Terakawa H et al. (2014) A modification of radical antegrade modular
institution results of radical antegrade modular pancreatosplenectomy
pancreatosplenectomy for adenocarcinoma of the left pancreas: sig-
for adenocarcinoma of the body and tail of pancreas in 78 patients.
nificance of en bloc resection including the anterior renal fascia. World J
J Hepatobiliary Pancreat Sci. http://dx.doi.org/10.1002/jhbp.362 [Epub
Surg 38:2448–2454. http://dx.doi.org/10.1007/s00268-014-2572-5.
ahead of print].
9. Lee SH, Kang CM, Hwang HK, Choi SH, Lee WJ, Chi HS. (2014)
20. Shrikhande SV, Barreto SG. (2010) Extended pancreatic resections and
Minimally invasive RAMPS in well-selected left-sided pancreatic cancer
lymphadenectomy: an appraisal of the current evidence. World J
within Yonsei criteria: long-term (>median 3 years) oncologic outcomes.
Gastrointest Surg 2:39–46. http://dx.doi.org/10.4240/wjgs.v2.i2.39.
Surg Endosc 28:2848–2855. http://dx.doi.org/10.1007/s00464-014-
21. Huebner M, Kendrick M, Reid-Lombardo KM, Que F, Therneau T, Qin R
3537-3.
et al. (2012) Number of lymph nodes evaluated: prognostic value in
10. Park HJ, You DD, Choi DW, Heo JS, Choi SH. (2014) Role of radical antegrade modular pancreatosplenectomy for adenocarcinoma of the body and tail of the pancreas. World J Surg 38:186–193. http:// dx.doi.org/10.1007/s00268-013-2254-8.
pancreatic adenocarcinoma. J Gastrointest Surg 16:920–926. http:// dx.doi.org/10.1007/s11605-012-1853-2. 22. Okada K, Kawai M, Tani M, Hirono S, Miyazawa M, Shimizu A et al. (2013) Surgical strategy for patients with pancreatic body/tail carci-
11. Moher D, Liberati A, Tetzlaff J Altman DG, &, PRISMA Group. (2009)
noma: who should undergo distal pancreatectomy with en-bloc celiac
Preferred reporting items for systematic reviews and meta-analyses: the
axis
PRISMA
j.surg.2012.07.036.
statement.
BMJ
339:b2535.
http://dx.doi.org/10.1136/
resection?
Surgery
153:365–372.
http://dx.doi.org/10.1016/
23. de Rooij T, Tol JA, van Eijck CH, Boerma D, Bonsing BA, Bosscha K
bmj.b2535. 12. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J et al.
et al. (2016) Outcomes of distal pancreatectomy for pancreatic ductal
(2005) Postoperative pancreatic fistula: an international study group
adenocarcinoma in The Netherlands: a nationwide retrospective anal-
(ISGPF)
http://dx.doi.org/10.1016/
ysis. Ann Surg Oncol 23:585–591. http://dx.doi.org/10.1245/s10434-
13. Zhu JC, Yan TD, Morris DL. (2008) A systematic review of radio-
24. Kooby DA, Hawkins WG, Schmidt CM, Weber SM, Bentrem DJ,
definition.
Surgery
138:8–13.
015-4930-4.
j.surg.2005.05.001. frequency ablation for lung tumors. Ann Surg Oncol 15:1765–1774.
Gillespie TW et al. (2010) A multicenter analysis of distal pancreatec-
http://dx.doi.org/10.1245/s10434-008-9848-7.
tomy for adenocarcinoma: is laparoscopic resection appropriate? J Am
14. Trottman P, Swett K, Shen P, Sirintrapun J. (2014) Comparison of standard distal pancreatectomy and splenectomy with radical ante15. Kawabata Y, Hayashi H, Takai K, Kidani A, Tajima Y. (2015) Superior artery-first
approach
in
radical
antegrade
12.033. 25. Bissolati M, Sandri MT, Burtulo G, Zorzino L, Balzano G, Braga M.
grade modular pancreatosplenectomy. Am Surg 80:295–300. mesenteric
Coll Surg 210:779–787. http://dx.doi.org/10.1016/j.jamcollsurg.2009.
modular
pancreatosplenectomy for borderline resectable pancreatic cancer: a technique to obtain negative tangential margins. J Am Coll Surg 220: e49–54. http://dx.doi.org/10.1016/j.jamcollsurg.2014.12.054. 16. Murakawa M, Aoyama T, Asari M, Katayama Y, Yamaoku K, Kanazawa A et al. (2015) The short- and long-term outcomes of radical
(2015) Portal vein-circulating tumor cells predict liver metastases in patients with resectable pancreatic cancer. Tumour Biol 36:991–996. http://dx.doi.org/10.1007/s13277-014-2716-0. 26. MacLehose RR, Reeves BC, Harvey IM, Sheldon TA, Russell IT, Black AM. (2000) A systematic review of comparisons of effect sizes derived from randomised and non-randomised studies. Health Technol Assess 4:1–154. http://dx.doi.org/10.3310/hta4340.
antegrade modular pancreatosplenectomy for adenocarcinoma of the
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.hpb.2016.07.014
REVIEW ARTICLE
A systematic review of radical antegrade modular pancreatosplenectomy for adenocarcinoma of the body and tail of the pancreas Yanming Zhou1, Bin Shi2, Lupeng Wu1 & Xiaoying Si1 1
Department of Hepatobiliary & Pancreatovascular Surgery, First Affiliated Hospital of Xiamen University, Xiamen, and 2General Intensive Care Unit, Songjiang Central Hospital, First People’s Hospital of Shanghai Jiaotong University, Shanghai, China
Abstract Background: To assess the published evidence on clinical outcomes following radical antegrade modular pancreatosplenectomy (RAMPS) for adenocarcinoma in the body or tail of the pancreas. Method: PubMed and Chinese Biomedical Literature databases were searched. The results of comparisons between RAMPS and standard retrograde pancreatosplenectomy (SRPS) were analyzed by meta-analytical techniques. Results: The literature search identified 13 observational studies involving 354 patients undergoing RAMPS. The overall morbidity and 30-day mortality was 40% and 0% respectively. The R0 resection rate was 88%; the median number of retrieved lymph nodes was 21; and the median 5-year overall survival rate was 37%. The result of meta-analysis showed that RAMPS was associated with a significantly less intraoperative bleeding [weighted mean difference −195.2 (95% confidence interval (CI) −223.27 to −167.13); P < 0.001], a greater number of retrieved lymph nodes [odds ratio (OR) 6.19 (95% CI 3.72 to 8.67); P < 0.001] and a higher percentage of R0 resection [OR 2.46 (95% CI 1.13 to 5.35); P = 0.02] as compared with SRPS. Conclusion: The current literature provides supportive evidence that RAMPS is a safe and effective procedure for adenocarcinoma in the body or tail of the pancreas, and is oncologically superior to SRPS. Received 28 June 2016; accepted 28 July 2016
Correspondence Yanming Zhou, Department of Hepatobiliary & Pancreatovascular Surgery, First Affiliated Hospital of Xiamen University, 55 Zhenhai Road, Xiamen 361003, FJ, China. Tel: +86 0592 2139708. Fax: +86 0592 2137289. E-mail: [email protected]
Introduction Adenocarcinoma in the body or tail of the pancreas is conventionally resected by the standard retrograde pancreatosplenectomy (SRPS) performed in the left-to-right direction with mobilization of the spleen first, and then resection of the posterior aspect of the pancreas from the tail to the body. However, SRPS is associated with a high positive tangential margin rate, devoid of the described lymph node drainage of the organ. To overcome these problems, Strasberg et al.1 in 2003 introduced a modified technique of SRPS called radical antegrade modular pancreatosplenectomy (RAMPS) in which division of the neck of the pancreas and splenic vessels and a celiac node dissection are Yanming Zhou and Bin Shi contributed equally to this work.
HPB 2017, 19, 10–15
performed first, followed by dissection proceeding from right-toleft in 1 of the 2 posterior dissection planes, depending on the extent of penetration of the tumor. However, only a few studies have reviewed the experience and practice of RAMPS.2–10 In addition, data comparing this procedure with SRPS are limited, and therefore the potential value of RAMPS has not been clearly demonstrated. The aim of this systematic review was to assess the published evidence on clinical outcomes following RAMPS for adenocarcinoma in the body or tail of the pancreas.
Methods This study was performed in accordance with the guidelines of preferred reporting items for systematic reviews and metaanalyses (PRISMA) 2009.11
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
11
Literature search strategy and study identification An electronic search was performed of the Pubmed and Chinese Biomedical Literature database from the date of the earliest report of RAMPS in 2003 to May 2016 using the following keyword: “radical antegrade modular pancreatosplenectomy.” No language restriction was applied. Reference lists of relevant articles were further searched manually to check for additional studies. Studies reporting the outcomes following the RAMPS in patients with pancreatic adenocarcinoma were included for analysis. To ensure that the series reviewed reflect consistent surgical approach, only study involving more than 5 patients were included in the systematic review of overall outcome of RAMPS. Reviews, conference abstracts, non-human studies, case report were excluded. In cases of duplicated studies with overlapping patients, only the most recent publication with accumulating numbers of patients or increased lengths of follow-up was considered. Two investigators (YZ and BS) independently appraised each eligible article using predefined criteria. Discrepancies between the two reviewers were resolved by discussion and consensus. Data were extracted on the first author, country, year of publication, sample size, study design, population characteristics,
duration of operation, estimated blood loss, proportion of radical (R0) resection, morbidity, incidence and severity of pancreatic fistula as defined by the International Study Group on Pancreatic Fistula (ISGPF),12 30-day mortality, and survival. The level of evidence of each study was categorized according to the Evidence-Based Medicine Levels of Evidence.13 Statistical analysis Descriptive statistics were performed and data are expressed as mean or median (interquartile range) where appropriate. A meta-analysis of the comparative studies of RAMPS and SRPS was undertaken with Review Manager (RevMan) software, version 5.1 (The Cochrane Collaboration, Software Update, Oxford). Odds ratio (OR) or weighted mean difference (WMD) with a 95% confidence interval (95% CI) were calculated for dichotomous variables and continuous variables respectively. Heterogeneity was assessed using the c2 test and I2. When the heterogeneity was not significant (I2 <50%), a fixed-effects model was used for the pooled analysis. Otherwise, a random-effects model was used. Statistical significance was set at P < 0.05.
Figure 1 Study selection
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
12
Results
Discussion
Systematic review As shown in Fig. 1, the literature search initially yielded 41 articles, of which 13 studies comprising a total of 354 patients fulfilling the eligibility criteria were included in this systematic review and are summarized in Table 1.4–10,14–19 Agreement for these studies was 100% between the two reviewers. No randomised controlled trials (RCT) or meta-analyses were identified. All these studies were observational in nature and classified as level-4 evidence. Five studies were conducted in Japan,4,8,15,16,18 three in Korea,5,9,10 two in USA,13,19 one in Italy,6 one in France,7 and one in China.17 The surgical outcomes of the 13 studies are summarized in Table 2. Of the 354 patients, 330 underwent open RAMPS and 24 underwent a laparoscopic or robotic approach. The overall mortality was 0%. Pancreatic fistula, the most critical complication following pancreatectomy, was seen in 70 (20%) patients. Fifty-five pancreatic fistulas were graded according to the ISGPF. Most were grade A (20, 36%) or B (23, 42%); only five (1%) were grade C and seven (13%) were graded as B–C.
This is the first review to assess existing evidence on clinical outcomes following RAMPS for adenocarcinoma in the body or tail of the pancreas. The results show that the perioperative mortality rate was zero, and the median 5-year OS was 37%, confirming the safety and efficacy of this procedure. The result of meta-analysis shows that the operating time was comparable between RAMPS and SRPS, though a longer operative time was observed in some smaller series6,10 of RAMPS group, suggesting that this may reflect the learning curve of the surgeons. As dissection during RAMPS commences from rightto-left with early division of the neck of the pancreas, it provides more superior access to control the major blood vessels including the splenic, adrenal and renal veins, blood loss was therefore reduced in RAMPS compared with SRPS. Although no survival benefit of an extended lymphadenectomy has been shown in pancreatic adenocarcinoma resection,20 thorough pathologic evaluation of lymph nodes may contribute to accurate staging. A recent review of 499 patients reported significant impact of the total number of examined lymph nodes on the estimation of stage-based survival after curative pancreatectomy for pancreatic adenocarcinoma.21 Survival for nodenegative (pN0) patients with <11 lymph nodes examined was worse than for pN0 patients with 11 lymph nodes with a 3-year survival rate of 32% vs. 50%, suggesting that metastatic nodes were missed by sampling insufficient lymph nodes. These investigators concluded that pathologic assessment of 11 lymph nodes in resected specimens is needed for accuracy of pancreatic adenocarcinoma staging. During SRPS, nodes at the roots of celiac or superior mesenteric artery regions are not removed. In contrast, RAMPS permits resection of these nodes in a controlled manner.1 The current systematic review demonstrates that the median number of total retrieved lymph nodes after RAMPS was 21 (range, 11–30), suggesting that RAMPS fulfills the criteria of adequate node dissection. R0 resection is an important factor determining survival in patients with pancreatic adenocarcinoma. The posterior margin is the predominant site of margin involvement of pancreatic body-tail adenocarcinoma.22 Published literatures with large sample size (n > 100) of distal adenocarcinoma demonstrate R0 resection rate of 50%–74%.23,24 The R0 resection rate of up to 88% after RAMPS in this systematic review is relatively high. In comparative studies, the cumulative R0 resection rate was 90% after RAMPS and 80% after SRPS with a pooled OR of 2.46 (95% CI 1.13–5.35). In the SRPS procedure, dissection proceeds from left-to-right, making it difficult to obtain a clear concept of where the posterior plane of dissection is, especially whenever the patient is deep-chested or obese.1 On the other hand, RAMPS enables the surgeon to set up the posterior plane of dissection more easily, thus allowing for more radical resection by early identification of the renal vein and the anterior surface of the adrenal vein.
Meta-analysis of RAMPS vs. SRPS Four studies compared RAMPS (n = 105) with SRPS (n = 131).6,10,14,18 Table 3 summarizes the outcomes of the metaanalysis. Compared with the SRPS group, the RAMPS group exhibited less intraoperative bleeding, a greater number of total retrieved lymph nodes, and a higher percentage of R0 resection (Fig. 2a–c). No significant difference was seen in other outcomes of interest. The funnel plot for retrieved lymph nodes was symmetric, indicating the absence of publication bias (Fig. 3). Table 1 Baseline characteristics of the studies included in system-
atic review Reference
Study period
Ikegami et al.4 Chang et al.
5
Age (year)a
Men, n TS (mm)a
2009–2011 6
67
4
30 41
2005–2009 24
60
13
Lartorre et al.6
2003–2011 8
61
5
–
Rosso et al.7
2008–2012 10
62
3
46
Kitagawa et al.8
2007–2012 24
67
15
35
Lee et al.9
2007–2010 12
64
7
28
Park et al.10
2007–2010 38
63
23
31
Trottman et al.14
2004–2011 6
–
–
–
Kawabata et al.15
2013–2014 11
68
6
33
Murakawa et al.16 2000–2014 49
68
31
38
Wu et al.17
2013–2014 35
65
13
46
2000–2014 53
69
31
–
Grossman et al.19 1999–2013 78
67
32
47
Abe et al.
18
Total/Median or % a
N
354 66 183 37 (9–43) (62–68) (53%) (31–46)
Median or mean; TS, tumor size.
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
13
Table 2 The surgical outcomes of radical antegrade modular pancreatosplenectomy.
Reference Ikegami et al.4 Chang et al.
5
Lartorre et al.6 Rosso et al.7
Operative time (min)a
Blood loss (mL)a BT n
258
226
–
Morbidity n PLN, n 4
–
RLN, na
R0 R n
MS (months) 5-year OS (%)
–
6
11
–
–
–
1
9
17
21
22
18
–
315
–
2
2
–
21
7
14
26
424
–
1
4
7
17
9
21
–
387
371
–
14
13
24
21
–
53
Lee et al.9
324
446
2
3
3
11
12
–
56
Park et al.10
210
325
–
7
22
14
34
–
40
Trottman et al.14
300
500
–
5
–
12
4
–
–
15
423
500
0
4
10
26
8
–
–
Murakawa et al.16
278
850
11
20
27
15
41
23
27
Wu et al.17
160
190
–
10
–
30
32
–
–
267
485
0
19
28
28
48
47
37
252
629
17
41
37
20
66
25
25
Kitagawa et al.
8
Kawabata et al.
Abe et al.
18
Grossman et al.19
Total/Median or % 289 (254–371) 465 (300–532)
34 (14%) 142 (40%)
164 (55%) 21 (14–26) 310 (88%) 21 (14–25)
37 (26–53)
a
Median or mean; BT, blood transfusion; PLN, number of patients with positive lymph nodes; RLN, retrieved lymph nodes; R0 R, R0 resection; MS, median survival; OS, overall survival.
Table 3 Results of the meta-analysis.
Outcome of interest
No. of studies
No. of patients
OR/WMD (95% CI)
P-value
I2 (%)
Operation time (min)
3
211
−17.03 (−94.46, 60.40)
0.67
94
Blood loss (mL)
3
211
−195.2 (−223.27, −167.13)
<0.001
0
Overall morbidity
4
236
0.97 (0.54, 1.77)
0.93
0
Overall PF
4
236
0.45 (0.19, 1.10)
0.08
0
ISGPF B + C PF
2
185
0.50 (0.18, 1.38)
0.18
0
Retrieved lymph nodes
4
236
6.19 (3.72, 8.67)
<0.001
0
R0 resection
4
223
2.46 (1.13, 5.35)
0.02
0
5-year OS (%)
3
210
1.77 (0.53, 5.93)
0.35
67
WMD, weighted mean difference; OR, odds ratio; CI, confidence interval; PF, pancreatic fistula; ISGPF, International Study Group of Pancreatic Fistula; OS, overall survival.
The presence of circulating tumor cells in the portal vein has been found to be associated with a higher rate of liver metastasis after pancreatic adenocarcinoma surgery.25 Left-to-right mobilization and handling of the pancreato-splenic specimen without early ligation of the draining blood vessels during SRPS could increase the risk of cancer cell shedding into the portal vein. In this context, RAMPS complies with the principle of no-touch isolation. CellSearch™ method might provide the evidence for the biological benefit of this procedure.25 The result of the present meta-analysis shows that RAMPS does not seem to improve survival as compared with SRPS. It is apparent that the number of patients enrolled in the current study is small with insufficient power to detect a possible difference. Mitchem et al. calculated that to compare two treatments with 5-year survival rates of 20% and 35% at the 95% CI level, a total recruitment of 556 patients was required.3 HPB 2017, 19, 10–15
The main limitation of this review is the relatively low level of evidence owing to the lack of RCT on the subject. All of the included studies were observational nature with an inherent risk of bias. There are many differences between the studies in terms of the patient selection, disease stage, surgical skill, and perioperative care. The definition of R0 resection was inhomogeneous; 12 studies simply reported that ‘surgical margins’ were negative without specified criteria,4–6,14,16–18 whereas others referred to the ‘absence of microscopic tumor cells’7,10 or ‘the tumor is >1 mm from all inked margins’.15,19 Meta-analysis of observational studies may either exaggerate or underestimate the magnitude of measured effect.26 Another limitation is the small sample size. As adenocarcinoma of the body and tail of the pancreas is usually detected in a relatively late stage because of its delayed presentation, it is often unresectable at the time of confirmed diagnosis. In addition, as
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
14
Figure 2 Results of the meta-analysis: (a) intraoperative bleeding; (b) number of total retrieved lymph nodes; (c) R0 resection
prospective RCTs comparing RAMPS with SRPS are required to obtain a more convincing conclusion. Conflicts of interest The authors declare that they have no potential competing interests. The authors received no funding for this work. References 1. Strasberg SM, Drebin JA, Linehan D. (2003) Radical antegrade modular pancreatosplenectomy. Surgery 133:521–527. http://dx.doi.org/10.1067/ msy.2003.146. 2. Strasberg SM, Linehan DC, Hawkins WG. (2007) Radical antegrade modular pancreatosplenectomy procedure for adenocarcinoma of the Figure 3 Funnel plot shows symmetry for retrieved lymph nodes
suggesting the absence of publication bias
body and tail of the pancreas: ability to obtain negative tangential margins. J Am Coll Surg 204:244–249. http://dx.doi.org/10.1016/ j.jamcollsurg.2006.11.002. 3. Mitchem JB, Hamilton N, Gao F, Hawkins WG, Linehan DC,
RAMPS is a novel surgical approach, surgeons are often reluctant to employ it, instead awaiting evidence of its safety and efficacy. In conclusion, RAMPS is a safe and effective procedure for adenocarcinoma in the body or tail of the pancreas and is oncologically superior to SRPS. Although there are no data supporting a prolonged survival, more lymph nodes and an apparently greater R0 rate of resection seem evident. As all the evidence in this meta-analysis comes from observational studies involving a relatively small number of patients, larger-sample HPB 2017, 19, 10–15
Strasberg SM. (2012) Long-term results of resection of adenocarcinoma of the body and tail of the pancreas using radical antegrade modular pancreatosplenectomy procedure. J Am Coll Surg 214:46–52. http:// dx.doi.org/10.1016/j.jamcollsurg.2011.10.008. 4. Ikegami T, Maeda T, Oki E, Kayashima H, Ohgaki K, Sakaguchi Y et al. (2011) Antegrade en bloc distal pancreatectomy with plexus hanging maneuver. J Gastrointest Surg 15:690–693. http://dx.doi.org/10.1007/ s11605-010-1382-9. 5. Chang YR, Han SS, Park SJ, Lee SD, Yoo TS, Kim YK et al. (2012) Surgical outcome of pancreatic cancer using radical antegrade modular
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
HPB
15
pancreatosplenectomy
procedure.
World
J
Gastroenterol
18:
body and tail of the pancreas. BMC Surg 15:120. http://dx.doi.org/
6. Latorre M, Ziparo V, Nigri G, Balducci G, Cavallini M, Ramacciato G.
17. Wu WG, Wu XS, Li ML, Wang XA, Shu YJ, Weng H et al. (2015) Ante-
(2013) Standard retrograde pancreatosplenectomy versus radical
grade pancreatosplenectomy for curative resection of adenocarcinoma
5595–5600. http://dx.doi.org/10.3748/wjg.v18.i39.5595.
10.1186/s12893-015-0107-0.
antegrade modular pancreatosplenectomy for body and tail pancreatic
of the left pancreas. Chin J Pract Surg 35:296–298. 18. Abe T, Ohuchida K, Miyasaka Y, Ohtsuka T, Oda Y, Nakamura M. (2016)
adenocarcinoma. Am Surg 79:1154–1158. 7. Rosso E, Langella S, Addeo P, Nobili C, Oussoultzoglou E, Jaeck D
Comparison of surgical outcomes between radical antegrade modular
et al. (2013) A safe technique for radical antegrade modular pancrea-
pancreatosplenectomy (RAMPS) and standard retrograde pancreatos-
tosplenectomy with venous resection for pancreatic cancer. J Am Coll
plenectomy (SPRS) for left-sided pancreatic cancer. World J Surg.
Surg 217:e35–39. http://dx.doi.org/10.1016/j.jamcollsurg.2013.08.007.
http://dx.doi.org/10.1007/s00268-016-3526-x [Epub ahead of print].
8. Kitagawa H, Tajima H, Nakagawara H, Makino I, Miyashita T,
19. Grossman JG, Fields RC, Hawkins WG, Strasberg SM. (2016) Single
Terakawa H et al. (2014) A modification of radical antegrade modular
institution results of radical antegrade modular pancreatosplenectomy
pancreatosplenectomy for adenocarcinoma of the left pancreas: sig-
for adenocarcinoma of the body and tail of pancreas in 78 patients.
nificance of en bloc resection including the anterior renal fascia. World J
J Hepatobiliary Pancreat Sci. http://dx.doi.org/10.1002/jhbp.362 [Epub
Surg 38:2448–2454. http://dx.doi.org/10.1007/s00268-014-2572-5.
ahead of print].
9. Lee SH, Kang CM, Hwang HK, Choi SH, Lee WJ, Chi HS. (2014)
20. Shrikhande SV, Barreto SG. (2010) Extended pancreatic resections and
Minimally invasive RAMPS in well-selected left-sided pancreatic cancer
lymphadenectomy: an appraisal of the current evidence. World J
within Yonsei criteria: long-term (>median 3 years) oncologic outcomes.
Gastrointest Surg 2:39–46. http://dx.doi.org/10.4240/wjgs.v2.i2.39.
Surg Endosc 28:2848–2855. http://dx.doi.org/10.1007/s00464-014-
21. Huebner M, Kendrick M, Reid-Lombardo KM, Que F, Therneau T, Qin R
3537-3.
et al. (2012) Number of lymph nodes evaluated: prognostic value in
10. Park HJ, You DD, Choi DW, Heo JS, Choi SH. (2014) Role of radical antegrade modular pancreatosplenectomy for adenocarcinoma of the body and tail of the pancreas. World J Surg 38:186–193. http:// dx.doi.org/10.1007/s00268-013-2254-8.
pancreatic adenocarcinoma. J Gastrointest Surg 16:920–926. http:// dx.doi.org/10.1007/s11605-012-1853-2. 22. Okada K, Kawai M, Tani M, Hirono S, Miyazawa M, Shimizu A et al. (2013) Surgical strategy for patients with pancreatic body/tail carci-
11. Moher D, Liberati A, Tetzlaff J Altman DG, &, PRISMA Group. (2009)
noma: who should undergo distal pancreatectomy with en-bloc celiac
Preferred reporting items for systematic reviews and meta-analyses: the
axis
PRISMA
j.surg.2012.07.036.
statement.
BMJ
339:b2535.
http://dx.doi.org/10.1136/
resection?
Surgery
153:365–372.
http://dx.doi.org/10.1016/
23. de Rooij T, Tol JA, van Eijck CH, Boerma D, Bonsing BA, Bosscha K
bmj.b2535. 12. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J et al.
et al. (2016) Outcomes of distal pancreatectomy for pancreatic ductal
(2005) Postoperative pancreatic fistula: an international study group
adenocarcinoma in The Netherlands: a nationwide retrospective anal-
(ISGPF)
http://dx.doi.org/10.1016/
ysis. Ann Surg Oncol 23:585–591. http://dx.doi.org/10.1245/s10434-
13. Zhu JC, Yan TD, Morris DL. (2008) A systematic review of radio-
24. Kooby DA, Hawkins WG, Schmidt CM, Weber SM, Bentrem DJ,
definition.
Surgery
138:8–13.
015-4930-4.
j.surg.2005.05.001. frequency ablation for lung tumors. Ann Surg Oncol 15:1765–1774.
Gillespie TW et al. (2010) A multicenter analysis of distal pancreatec-
http://dx.doi.org/10.1245/s10434-008-9848-7.
tomy for adenocarcinoma: is laparoscopic resection appropriate? J Am
14. Trottman P, Swett K, Shen P, Sirintrapun J. (2014) Comparison of standard distal pancreatectomy and splenectomy with radical ante15. Kawabata Y, Hayashi H, Takai K, Kidani A, Tajima Y. (2015) Superior artery-first
approach
in
radical
antegrade
12.033. 25. Bissolati M, Sandri MT, Burtulo G, Zorzino L, Balzano G, Braga M.
grade modular pancreatosplenectomy. Am Surg 80:295–300. mesenteric
Coll Surg 210:779–787. http://dx.doi.org/10.1016/j.jamcollsurg.2009.
modular
pancreatosplenectomy for borderline resectable pancreatic cancer: a technique to obtain negative tangential margins. J Am Coll Surg 220: e49–54. http://dx.doi.org/10.1016/j.jamcollsurg.2014.12.054. 16. Murakawa M, Aoyama T, Asari M, Katayama Y, Yamaoku K, Kanazawa A et al. (2015) The short- and long-term outcomes of radical
(2015) Portal vein-circulating tumor cells predict liver metastases in patients with resectable pancreatic cancer. Tumour Biol 36:991–996. http://dx.doi.org/10.1007/s13277-014-2716-0. 26. MacLehose RR, Reeves BC, Harvey IM, Sheldon TA, Russell IT, Black AM. (2000) A systematic review of comparisons of effect sizes derived from randomised and non-randomised studies. Health Technol Assess 4:1–154. http://dx.doi.org/10.3310/hta4340.
antegrade modular pancreatosplenectomy for adenocarcinoma of the
HPB 2017, 19, 10–15
© 2016 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.