Surg Endosc DOI 10.1007/s00464-014-3997-5
and Other Interventional Techniques
DYNAMIC MANUSCRIPT
Lateral approach in laparoscopic distal pancreatectomy is safe and potentially beneficial compared to the traditional medial approach Matt Strickland • Julie Hallet • Daniel Abramowitz • Shuyin Liang • Calvin H. L. Law • Shiva Jayaraman
Received: 4 May 2014 / Accepted: 5 November 2014 Ó Springer Science+Business Media New York 2014
Abstract Introduction Laparoscopic distal pancreatectomy has become widely accepted for the treatment of left-sided pancreatic lesions. Traditionally, a medial laparoscopic distal pancreatectomy (MDLP) has been employed, with division of the gland followed by medial to lateral mobilization. Recent technical reports of lateral laparoscopic distal pancreatectomy (LLDP) suggest that it offers easier access and more precise dissection. Data on this technique remain sparse and inconclusive, with no formal comparison with MLDP. We sought to compare outcomes of LLDP to MLDP. Methods We reviewed the charts of patients undergoing laparoscopic distal pancreatectomy at two academic institutions, from July 2009 to June 2013. Primary outcomes were operating time and estimated blood loss. Secondary outcomes included success of spleen-preserving procedures, length of sacrificed pancreas parenchyma, margins Electronic supplementary material The online version of this article (doi:10.1007/s00464-014-3997-5) contains supplementary material, which is available to authorized users. M. Strickland (&) J. Hallet D. Abramowitz S. Liang C. H. L. Law S. Jayaraman Division of General Surgery, University of Toronto, Toronto, ON, Canada e-mail: [email protected] S. Jayaraman e-mail: [email protected] J. Hallet C. H. L. Law Division of General Surgery, Sunnybrook Health Sciences Centre, Toronto, ON, Canada S. Jayaraman Division of General Surgery, St. Joseph’s Health Centre, 30 The Queensway, Toronto, ON M6R 1B5, Canada
status, 30-day major morbidity (Clavien grade 3–5 complications), and length of stay. We reported data as proportions and medians. We performed comparative analysis using Chi square test or Fisher’s exact test for categorical variables, and Mann–Whitney U test for continuous variables. Results We retrieved 43 cases (19 LLDP, 24 MLDP). Median operative time was shorter (166 vs 190 min; p = 0.03) and estimated blood loss lower (50 vs 250 mL; p \ 0.01) with LLDP. No margin was positive with LLDP compared to 2 (8.3 %) with MLDP. Major morbidity did not differ (LLDP 21.0 % vs MLDP 25.0 %; p = 0.76). Trends toward lower conversion rate (16.7 vs 5.3 %; p = 0.36) and shorter length of stay (5 vs 4 days; p = 0.35) were not significant. Conclusion LLDP is a feasible and safe approach for distal lesions of the pancreatic tail, associated with shorter operative time and decreased blood loss compared to traditional MLDP. Potential of decreased conversion rate and length of stay exists. These hypotheses need to be confirmed in larger prospective studies. Keywords Distal pancreatectomy Laparoscopic Lateral Medial Spleen preserving Laparoscopic distal pancreatectomy has become a widely accepted surgical approach for the treatment of left-sided benign, borderline, and some malignant pancreatic tumors. Compared to open distal pancreatectomy, the laparoscopic technique offers decreased morbidity and shorter length of stay [1–3]. The traditional approach involves medial dissection of the upper and lower borders of the pancreas lateral to the superior mesenteric vessels for control of the splenic artery and vein, and medial division of the pancreas, with the patient in the supine position. The dissection
123
Surg Endosc
is then carried from medial to lateral toward the splenic hilum [1]. However, this approach may be challenging for lesions situated distally in the tail of the pancreas. There are emerging reports of a lateral approach for distal pancreatectomy [4–6], whereby dissection is carried out from lateral to medial after dividing the lateral splenic ligaments, with the patient in the right lateral decubitus position. Proponents claim that this technique provides easier access to and enables more precise dissection of distal pancreatic tail tumors with potential benefits in terms of operative outcomes, and pancreatic parenchymal and splenic preservation. Thus far, literature on lateral laparoscopic distal pancreatectomy (LLDP) is limited to technical reports or small single-centre case series [4–6] In an effort to formally assess the benefits of LLDP, we sought to compare the outcomes of lateral laparoscopic distal pancreatectomy (LLDP) to medial laparoscopic distal pancreatectomy (MLDP).
Methods We conducted a retrospective cohort study to compare the outcomes of LLDP and MLDP. This study was reviewed and approved by the Research Ethics Board at Sunnybrook Health Sciences Centre and St. Joseph’s Health Centre. Selection of participants All laparoscopic distal pancreatectomies performed from July 2009 to June 2013 by two hepato-pancreato-biliary surgeons at two institutions (Sunnybrook Health Sciences Centre and St. Joseph’s Health Centre, Toronto, Ontario) were retrieved using institutional HPB databases. We included all adult patients (C18-years old) with a benign or malignant pancreatic tumor. The choice of laparoscopic vs open, decision to proceed with LLDP or MLDP, and whether or not to aim for spleen preservation was left to the surgeon’s discretion. Spleen preservation was not attempted for malignant cases.
lesion. The pancreas was divided using a laparoscopic surgical stapler. If splenectomy was performed, the splenic artery and vein were also divided with a laparoscopic stapler, and the dissection was carried out from medial to lateral before dividing the short gastric vessels and lateral spleen attachments. For spleen preservation, the pancreatic gland is elevated from the splenic vein from medial to lateral. For LLPD, patients are placed in right lateral decubitus, with the surgeon on the right side of the operating table. A 4- or 5-trocar technique is used to perform distal pancreatectomy with or without spleen preservation (Fig. 1). A 30-degree camera is inserted through a 10-mm port 1 cm above and to the left of the umbilicus using a Veress needle in the left upper quadrant and optical trocar. After confirming the absence of extra-pancreatic disease, the dissection begins by lowering the splenic flexure. The inferior border of the pancreas is then identified and mobilized, followed by mobilization of the superior border of the pancreas which includes division of the most distal short gastric vessels and entrance into the lesser sac (Fig. 2A). With the pancreas fully exposed, the gland is elevated from the splenic vessels in a lateral to medial fashion, for a spleen-sparing procedure (Fig. 2B, C). Once well medial to the lesion, the pancreas is transected using a laparoscopic stapler (Fig. 2D). For distal pancreato-splenectomy, the lateral splenic attachments are divided to mobilize the spleen, and the pancreas is elevated from the retroperitoneum while staying posterior to the splenic vessels. Prior to pancreatic transection, the splenic vessels are ligated and divided. The specimen is extracted in a specimen retrieval bag through a Pfannenstiel incision. Surgical drains are not routinely used. As a dynamic supplement to this paper, videos of the mobilization [Supplementary material 1] and isolation of splenic vessels and transection [Supplementary material 2] are available online.
Technical information The two participating hepato-pancreato-biliary surgeons performed both LLDP and MLPD. MLDP was performed in the usual fashion [7–9], with the patient in the supine position. A 4- to 5-trocar technique was used, with a 30-degree camera inserted in a 10-mm umbilical port. The gastro-colic ligament was opened to enter the lesser sac, and the stomach was retracted superiorly and medially to expose the anterior aspect of the pancreas. The retroperitoneum was opened inferior to the pancreas to create a tunnel proximally to the pancreatic
123
Fig. 1 Port placement for lateral laparoscopic distal pancreatectomy
Surg Endosc Fig. 2 Operative steps in a lateral laparoscopic distal pancreatectomy. A, B The inferior border and then superior border of the pancreas are identified and mobilized. The distal short gastric vessels are ligated and the lesser sac is entered. C The pancreas is elevated from the splenic vessels in a lateral to medial fashion, for a spleen-sparing procedure. D Once well medial to the lesion, the pancreas is transected using a laparoscopic stapler
Outcome measures and data collection The primary outcomes were operating time and estimated blood loss. Secondary outcomes included success of spleen-preserving procedures (number of successful/number planned spleen preservations), length of pancreas parenchyma resected, resection margins status, 30-day major morbidity defined as Clavien–Dindo grade 3, 4, and 5 [10], and grade C pancreatic fistula as per the International Study Group on Pancreatic Fistula classification [11]. Reviewers not involved in the treatment process collected clinical and pathological data from patients’ charts. In addition to demographics such as age and gender, baseline assessment of patients’ comorbidity was obtained using Charlson Comorbidity Score [12]. Total length of pancreas resected was defined as the length of the pancreatic specimen as measured in pathology, and length of healthy pancreas resected was computed by subtracting the tumor length to the specimen length.
Fig. 3 Patient distribution within the two groups
Statistical analysis Results Analysis was performed using SPSS 21.0 (IBM Corp., Amonk, NY). Due to the small sample size, continuous data are expressed as medians with inter-quartile range (IQR). Categorical data are reported as absolute numbers (n) and proportions (%). We compared LLDP and MLDP groups using the Mann–Whitney U test, Pearson Chi square test, or Fisher’s exact test, as appropriate. Results were considered significant at p \ 0.05.
We identified 43 patients meeting our inclusion criteria, including 24 MLDP and 19 LLDP (Fig. 3). Table 1 summarizes patient demographics and clinical characteristics of the two groups. Of note, patients undergoing MLPD had a higher mean Charlson score (p = 0.03) and more tumors localized in the proximal third of the pancreatic tail (16.7 vs 0 %) than with LLDP.
123
Surg Endosc Table 1 Patient demographics and clinical characteristics Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
Spleen preservation Planned spleen preservation—n (%)
15 (62.5)
13 (68.4)
0.75
Successful spleen preservation
13 (86.7)
10 (76.9)
0.64
Operating time (min)—median (IQR)
190 (166.25–215.5)
Estimated blood loss (mL)—median (IQR)
300 (175–750)
Conversion to open procedure—n (%)
166 (150–196) 50 (20–100)
4 (16.7)
1 (5.3)
0.03 0.01 0.36
Proportion computed with number of planned spleen preservation as denominator
IQR inter-quartile range
The operative characteristics of both groups are summarized in Table 2. There was no significant difference in the percentage of patients with planned spleen preservation, successful spleen preservation, or rate of conversion to open procedure. The median operating time in the LLDP group was 24 min shorter (p = 0.03), and median estimated blood loss was 250 mL less (p \ 0.01) than that with MLDP. There were fewer conversions to open surgery with LLDP (16.7 vs 5.3 %; p = 0.36), but this was not statistically significant. Post-operative outcomes are detailed in Table 3. All surgical margins were negative with LLDP compared to 2 (8.3 %) microscopically positive margins with MLDP. Median length of total or healthy resected pancreatic parenchyma did not differ. There was a shorter median length of stay with LLDP (4 vs 5 days; p = 0.35), but this was not statistically significant. No difference was observed in major morbidity (MLDP—25.0 % vs LLDP— 21.1 %; p = 0.76).
Discussion This is the first study comparing the traditional medial approach to laparoscopic distal pancreatectomy to the lateral approach. We observed that while obtaining a similar oncological resection, the lateral approach benefited patients in terms of shorter operating time (166 vs 190 min; p = 0.03) and lower estimated blood loss (50 vs 300 mL; p = 0.01). Although not reaching statistical significance, we identified the potential for positive impact on need for conversion to open surgery and length of stay. Since its introduction in the early 1990s [13], laparoscopic distal pancreatectomy has become widely accepted and is now an established operation among hepato-pancreato-biliary surgeons. Resection of the distal pancreas lends itself well to a laparoscopic approach because it involves resection without reconstruction. The medial approach was the first described and has become the
123
standard technique. Several other techniques have been reported including the lateral approach examined in this study [4–6], retroperitoneal [14] and hand-assisted versions [15]. In this still-evolving field, it remains important to critically examine alternative surgical techniques. Our group has been performing both MLDP and LLDP since 2009. Our willingness to try the lateral approach has mainly been driven by several perceived technical advantages that may potentially translate into clinical benefits. Positioning the patient in the right lateral decubitus position allows gravity to help retract both the stomach and the spleen, facilitating access and dissection by a single skilled surgeon. Easier access and greater exposure allow for more precision and quicker dissection that can potentially reduce operative time and blood loss. With this approach, we believe that the splenic vessels are easier to visualize, identify, and isolate at the tail of the pancreas than at the isthmus. Spleen preservation requires complete sparing of the splenic artery and vein that are often embedded in the supero-posterior aspect of the pancreatic body [7, 8]. Separation of the splenic vein from the pancreatic gland during MLDP is responsible for most bleeding during that surgery, due to the challenge imposed by multiple short branches coursing into the pancreatic parenchyma [9]. By finding the vessels laterally and then dissecting them medially, this potentially improves the chances of spleen preservation and decreases the amount of blood loss [16]. With LLDP, the pancreas is divided more laterally which could theoretically preserve pancreatic parenchyma, and lessen the risk of post-operative pancreatic insufficiency. So far, LLDP has been assessed mainly in technical reports or small non-comparative case series [4–6]. In 2012, a group from India reported treatment of four insulinoma cases using the lateral approach for enucleation (n = 3) and spleen-preserving distal pancreatectomy (n = 1) [5]. The only comparative cohort was reported by Nakamura et al., in 2011, with 8 lateral laparoscopic spleen-preserving distal pancreatectomy compared to 15 medial approaches to spleen preservation [6]. Patients who
Surg Endosc Table 2 Operative characteristics Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
Post-operative morbidity (Clavien–Dindo classification)—n (%) Grade 1
0 (0)
1 (5.3)
–
Grade 2
1 (4.1)
2 (10.5)
0.57
Grade 3
5 (20.8)
3 (15.8)
1.00
Grade 4
1 (4.2)
1 (5.3)
1.00
Pancreatic leak (ISGPF Grade C)—n (%)
2 (8.3)
0 (0)
–
Post-operative mortality—n (%) Length of stay (days)—median (IQR)
0 (0) 5 (4–6)
0 (0) 4 (3.25–5)
– 0.35
Neuroendocrine tumor
8 (33.3)
4 (21.0)
0.5
Serous cystadenoma
4 (16.7)
2 (10.5)
0.68
Mucinous cystadenoma
4 (16.7)
2 (10.5)
0.68
Intraductal papillary mucinous neoplasia
3 (12.5)
1 (5.3)
0.62
Solid pseudopapillary tumor
0 (0)
3 (15.8)
–
Pancreatic intraepithelial neoplasia
0 (0)
2 (10.5)
–
Pancreatic adenocarcinoma
1 (4.2)
1 (5.3)
1.00
Pancreatic renal cell carcinoma Metastasis
1 (4.2)
0 (0)
–
Chronic pancreatitis
0 (0)
3 (15.8)
–
Retention cyst
1 (4.2)
0 (0)
–
Epithelial cyst
1 (4.2)
0 (0)
–
Histologic diagnosis—n (%)
Cystic teratoma Accessory spleen embedded in pancreas Tumor size (cm)—median (IQR)
1 (4.2) 0 (0) 2.9 (2–3.6)
0 (0)
–
1 (5.3) 3.5 (2.5–5)
– 0.52 –
Negative resection margins—n (%)
22 (91.7)
19 (100)
Total length of pancreas resected (cm)—median (IQR)
7.1 (4.9–9.0)
6.9 (4.6–9.2)
0.70
Length of healthy pancreas resected (cm)—median (IQR)
3.1 (1.5–6)
3 (1.5–5.5)
0.79
ISGPF International Study Group on pancreatic fistula, IQR inter-quartile range
underwent a lateral operation experienced fewer conversions to open procedure (0 vs 2), and lower blood loss (125 vs 1025 mL; p = 0.07), while operative time and postoperative morbidity did not differ. This report focused on spleen-preserving operations, including 8 patients in the medial group that underwent the Warshaw procedure, as a method of spleen preservation [17]. By including all patients undergoing laparoscopic distal pancreatectomy with or without spleen preservation, our study provides a larger and more pragmatic appreciation of the adoption of LLDP in everyday surgical practice, while yielding similar conclusions in terms of morbidity outcomes. However, we did not observe a difference in the feasibility of spleen preservation. This may be related to the experience and specialized expertise of the participating hepato-pancreatobiliary surgeons who were able to consistently complete spleen-preserving distal pancreatectomy with the medial approach. This review of our experience also points to additional advantages, including a significant decrease in operative
time and blood loss with LLDP. Trends toward decreased need for conversion and length of stay were also observed, although not reaching statistical significance likely due to low statistical power of the study. Perioperative morbidity and mortality were not found to be significantly different, indicating safety of LLDP. Although not reaching statistical significance, the LLDP group included more tumors in the distal third of the pancreatic tail and none in the proximal third. Those patients would appear to be the best candidates for LLDP and could explain the potential benefits surfacing in this analysis. In studies such as this, examining refinements of surgical techniques, differences in post-operative morbidity may not always be appreciable. However, other outcomes such as operative time and blood loss may better reflect small but important differences in terms of technical ease. An easier technique may also translate into broader applicability of laparoscopic distal pancreatectomy for more lesions and by more surgeons when the medial approach might prove too challenging, thus extending the benefits of laparoscopic surgery to more
123
Surg Endosc Table 3 Post-operative outcomes
Age—median (IQR) Male gender—n (%) Body mass index—median (IQR) Charlson Comorbidity Score—median (IQR)
Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
59.3 (48.7–66.7)
58.3 (48.2–64.5)
0.58
6 (25.0)
6 (31.6)
28.1 (26.4–29.2)
27.8 (25.0–31.7)
4.5 (2–6)
0.74 0.96
2 (1–4)
0.03
0 (0)
–
6 (31.6) 7 (36.8)
0.37 0.17
Pre-operative diagnosis—n (%) Pancreatic pseudocyst
4 (16.7)
Cystic neoplasm Neuroendocrine tumor
11 (45.8) 4 (16.7)
Solid pseudopapillary tumor
0 (0)
2 (10.5)
–
Pancreatic adenocarcinoma
2 (8.3)
1 (5.3)
1.0
Pancreatic renal cell carcinoma metastasis
1 (4.1)
0 (0)
–
Pancreatic mass not yet diagnosed
2 (8.3)
3 (15.8)
0.64
Tumor localization—n (%) Proximal third
4 (16.7)
0 (0)
–
Middle third
5 (20.8)
3 (15.8)
1.0
Distal third
15 (62.5)
16 (84.2)
0.17
IQR inter-quartile range
patients. Therefore, we believe that the lateral approach could prove superior to MLDP for selected patients with distal pancreatic tail lesions and perhaps more broadly adoptable. We acknowledge that as a retrospective cohort series, this study presents inherent limitations. Due to small numbers and the intrinsic nature of pancreatic tail lesions, there was a great deal of heterogeneity in the pre-operative diagnosis and eventual pathology. The Charlson comorbidity index was higher in the MLDP group, pointing toward potential selection bias by experienced hepatopancreato-biliary surgeons. The difference in tumor localization also highlights the importance of selection in obtaining benefits of LLDP over MLDP, which could not be further explored through sub-group analyses due to the small number of patients. Finally, the small sample size may have hampered potential benefits as indicated by some trends that are not found to be statistically significant. Nonetheless, this study offers a strong appraisal of the benefits of LLDP over traditional MLDP in the largest cohort to date. Our results indicate that LLDP seems to hold promise in further improving outcomes and potentially broadening the use of laparoscopic distal pancreatectomy.
Conclusions The lateral approach is a feasible and safe approach for distal lesions of the pancreatic tail. The improved exposure and ease of finding the splenic vessels offer the potential
123
for shorter operative time and decreased blood loss compared to traditional medial approach, and even possible reduction in conversion rate and length of stay. These hypotheses need to be confirmed in larger prospective studies. Acknowledgments The authors would like to thank Mrs. Laura Maaske ([email protected]) for medical illustrations. Disclosures Matt Strickland—GestSure Technologies (founder, equity interest). Shiva Jayaraman—Astellas (speaker honorarium), Ethicon Endosurgery (consultant and academic support) and Roche (academic support). Julie Hallet, Daniel Abramowitz, Shuyin Liang, and Calvin HL Law have no conflict of interest or financial ties to disclose.
References 1. Kooby DA, Gillespie T, Bentrem D, Nakeeb A, Schmidt MC, Merchant NB et al (2008) Left-sided pancreatectomy. Ann Surg 126:88–96 2. Lebedyev A, Zmora O, Kuriansky J, Rosin D, Khaikin M, Shabtai M et al (2004) Laparoscopic distal pancreatectomy. Surg Endosc 18(10):1427–1430 3. Jayaraman S, Gonen M, Brennan MF, D’Angelica MI, DeMatteo RP, Fong Y, Jarnagin WR, Allen PJ (2010) J Am Coll Surg 211(4):503–509 4. Thomas JK, Abraham D, Joseph P, Paul MJ (2012) Lateral laparoscopic approach to pancreatic tail insulinomas. World J Endocr Surg 4(1):3–7 5. Honore C, Honore P, Meurisse M (2007) Laparoscopic spleenpreserving distal pancreatectomy: description of an original posterior approach. J Laparoendosc Adv Surg Technol 17(5): 686–689
Surg Endosc 6. Nakamura M, Nagayoshi Y, Kono H et al (2011) Lateral approach for laparoscopic splenic vessel-preserving distal pancreatectomy. Surgery 150(2):326–331 7. Boutros C, Espat NJ, Somasundar P (2010) Completely laparoscopic subtotal pancreatectomy with splenic artery preservation. J Gastrointest Surg 14(1):171–174 8. Suzuki O, Tanaka E, Hirano S, Suzuoki M, Hashida H, Ichimura T et al (2009) Efficacy of the electrothermal bipolar vessel sealer in laparoscopic spleen-preserving distal pancreatectomy with conservation of the splenic artery and vein. J Gastrointest Surg 13(1):155–158 9. Melotti G, Butturini G, Piccoli M, Casetti L, Bassi C, Mullineris B et al (2007) Laparoscopic distal pancreatectomy: results on a consecutive series of 58 patients. Ann Surg 246(1):77–82 10. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240:205–213 11. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J et al (2005) International Study Group on Pancreatic Fistula. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 138:8–13
12. Charlson M, Pompei P, Ales K, MacKenzie C (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373–383 13. Soper NJ, Brunt LM, Dunnegan DL, Meininger TA (1994) Laparoscopic distal pancreatectomy in the porcine model. Surg Endosc 8:57–60 14. Takada M, Ichihara T, Toyama H, Suzuki Y, Kuroda Y (2004) Retroperitoneoscopic laparoscopic distal pancreatectomy with spleen salvage. Hepatogastroenterology 51:925–927 15. Shinchi H, Takao S, Noma H, Mataki Y, Iino S, Aikou T (2001) Hand-assisted laparoscopic distal pancreatectomy with minilaparotomy for distal pancreatic cystadenoma. Surg Laparosc Endosc Percutan Technol 11:139–143 16. Yoon YS, Lee KH, Han HS, Cho JY, Ahn KS (2009) Patency of splenic vessels after laparoscopic spleen and splenic vessel-preserving distal pancreatectomy. Br J Surg 96:633–640 17. Adam JP, Jacquin A, Laurent C, Collet D, Masson B, Ferna´ndezCruz L et al (2013) Laparoscopic spleen-preserving distal pancreatectomy: splenic vessel preservation compared with the Warshaw technique. JAMA Surg 148(3):246–252
123
and Other Interventional Techniques
DYNAMIC MANUSCRIPT
Lateral approach in laparoscopic distal pancreatectomy is safe and potentially beneficial compared to the traditional medial approach Matt Strickland • Julie Hallet • Daniel Abramowitz • Shuyin Liang • Calvin H. L. Law • Shiva Jayaraman
Received: 4 May 2014 / Accepted: 5 November 2014 Ó Springer Science+Business Media New York 2014
Abstract Introduction Laparoscopic distal pancreatectomy has become widely accepted for the treatment of left-sided pancreatic lesions. Traditionally, a medial laparoscopic distal pancreatectomy (MDLP) has been employed, with division of the gland followed by medial to lateral mobilization. Recent technical reports of lateral laparoscopic distal pancreatectomy (LLDP) suggest that it offers easier access and more precise dissection. Data on this technique remain sparse and inconclusive, with no formal comparison with MLDP. We sought to compare outcomes of LLDP to MLDP. Methods We reviewed the charts of patients undergoing laparoscopic distal pancreatectomy at two academic institutions, from July 2009 to June 2013. Primary outcomes were operating time and estimated blood loss. Secondary outcomes included success of spleen-preserving procedures, length of sacrificed pancreas parenchyma, margins Electronic supplementary material The online version of this article (doi:10.1007/s00464-014-3997-5) contains supplementary material, which is available to authorized users. M. Strickland (&) J. Hallet D. Abramowitz S. Liang C. H. L. Law S. Jayaraman Division of General Surgery, University of Toronto, Toronto, ON, Canada e-mail: [email protected] S. Jayaraman e-mail: [email protected] J. Hallet C. H. L. Law Division of General Surgery, Sunnybrook Health Sciences Centre, Toronto, ON, Canada S. Jayaraman Division of General Surgery, St. Joseph’s Health Centre, 30 The Queensway, Toronto, ON M6R 1B5, Canada
status, 30-day major morbidity (Clavien grade 3–5 complications), and length of stay. We reported data as proportions and medians. We performed comparative analysis using Chi square test or Fisher’s exact test for categorical variables, and Mann–Whitney U test for continuous variables. Results We retrieved 43 cases (19 LLDP, 24 MLDP). Median operative time was shorter (166 vs 190 min; p = 0.03) and estimated blood loss lower (50 vs 250 mL; p \ 0.01) with LLDP. No margin was positive with LLDP compared to 2 (8.3 %) with MLDP. Major morbidity did not differ (LLDP 21.0 % vs MLDP 25.0 %; p = 0.76). Trends toward lower conversion rate (16.7 vs 5.3 %; p = 0.36) and shorter length of stay (5 vs 4 days; p = 0.35) were not significant. Conclusion LLDP is a feasible and safe approach for distal lesions of the pancreatic tail, associated with shorter operative time and decreased blood loss compared to traditional MLDP. Potential of decreased conversion rate and length of stay exists. These hypotheses need to be confirmed in larger prospective studies. Keywords Distal pancreatectomy Laparoscopic Lateral Medial Spleen preserving Laparoscopic distal pancreatectomy has become a widely accepted surgical approach for the treatment of left-sided benign, borderline, and some malignant pancreatic tumors. Compared to open distal pancreatectomy, the laparoscopic technique offers decreased morbidity and shorter length of stay [1–3]. The traditional approach involves medial dissection of the upper and lower borders of the pancreas lateral to the superior mesenteric vessels for control of the splenic artery and vein, and medial division of the pancreas, with the patient in the supine position. The dissection
123
Surg Endosc
is then carried from medial to lateral toward the splenic hilum [1]. However, this approach may be challenging for lesions situated distally in the tail of the pancreas. There are emerging reports of a lateral approach for distal pancreatectomy [4–6], whereby dissection is carried out from lateral to medial after dividing the lateral splenic ligaments, with the patient in the right lateral decubitus position. Proponents claim that this technique provides easier access to and enables more precise dissection of distal pancreatic tail tumors with potential benefits in terms of operative outcomes, and pancreatic parenchymal and splenic preservation. Thus far, literature on lateral laparoscopic distal pancreatectomy (LLDP) is limited to technical reports or small single-centre case series [4–6] In an effort to formally assess the benefits of LLDP, we sought to compare the outcomes of lateral laparoscopic distal pancreatectomy (LLDP) to medial laparoscopic distal pancreatectomy (MLDP).
Methods We conducted a retrospective cohort study to compare the outcomes of LLDP and MLDP. This study was reviewed and approved by the Research Ethics Board at Sunnybrook Health Sciences Centre and St. Joseph’s Health Centre. Selection of participants All laparoscopic distal pancreatectomies performed from July 2009 to June 2013 by two hepato-pancreato-biliary surgeons at two institutions (Sunnybrook Health Sciences Centre and St. Joseph’s Health Centre, Toronto, Ontario) were retrieved using institutional HPB databases. We included all adult patients (C18-years old) with a benign or malignant pancreatic tumor. The choice of laparoscopic vs open, decision to proceed with LLDP or MLDP, and whether or not to aim for spleen preservation was left to the surgeon’s discretion. Spleen preservation was not attempted for malignant cases.
lesion. The pancreas was divided using a laparoscopic surgical stapler. If splenectomy was performed, the splenic artery and vein were also divided with a laparoscopic stapler, and the dissection was carried out from medial to lateral before dividing the short gastric vessels and lateral spleen attachments. For spleen preservation, the pancreatic gland is elevated from the splenic vein from medial to lateral. For LLPD, patients are placed in right lateral decubitus, with the surgeon on the right side of the operating table. A 4- or 5-trocar technique is used to perform distal pancreatectomy with or without spleen preservation (Fig. 1). A 30-degree camera is inserted through a 10-mm port 1 cm above and to the left of the umbilicus using a Veress needle in the left upper quadrant and optical trocar. After confirming the absence of extra-pancreatic disease, the dissection begins by lowering the splenic flexure. The inferior border of the pancreas is then identified and mobilized, followed by mobilization of the superior border of the pancreas which includes division of the most distal short gastric vessels and entrance into the lesser sac (Fig. 2A). With the pancreas fully exposed, the gland is elevated from the splenic vessels in a lateral to medial fashion, for a spleen-sparing procedure (Fig. 2B, C). Once well medial to the lesion, the pancreas is transected using a laparoscopic stapler (Fig. 2D). For distal pancreato-splenectomy, the lateral splenic attachments are divided to mobilize the spleen, and the pancreas is elevated from the retroperitoneum while staying posterior to the splenic vessels. Prior to pancreatic transection, the splenic vessels are ligated and divided. The specimen is extracted in a specimen retrieval bag through a Pfannenstiel incision. Surgical drains are not routinely used. As a dynamic supplement to this paper, videos of the mobilization [Supplementary material 1] and isolation of splenic vessels and transection [Supplementary material 2] are available online.
Technical information The two participating hepato-pancreato-biliary surgeons performed both LLDP and MLPD. MLDP was performed in the usual fashion [7–9], with the patient in the supine position. A 4- to 5-trocar technique was used, with a 30-degree camera inserted in a 10-mm umbilical port. The gastro-colic ligament was opened to enter the lesser sac, and the stomach was retracted superiorly and medially to expose the anterior aspect of the pancreas. The retroperitoneum was opened inferior to the pancreas to create a tunnel proximally to the pancreatic
123
Fig. 1 Port placement for lateral laparoscopic distal pancreatectomy
Surg Endosc Fig. 2 Operative steps in a lateral laparoscopic distal pancreatectomy. A, B The inferior border and then superior border of the pancreas are identified and mobilized. The distal short gastric vessels are ligated and the lesser sac is entered. C The pancreas is elevated from the splenic vessels in a lateral to medial fashion, for a spleen-sparing procedure. D Once well medial to the lesion, the pancreas is transected using a laparoscopic stapler
Outcome measures and data collection The primary outcomes were operating time and estimated blood loss. Secondary outcomes included success of spleen-preserving procedures (number of successful/number planned spleen preservations), length of pancreas parenchyma resected, resection margins status, 30-day major morbidity defined as Clavien–Dindo grade 3, 4, and 5 [10], and grade C pancreatic fistula as per the International Study Group on Pancreatic Fistula classification [11]. Reviewers not involved in the treatment process collected clinical and pathological data from patients’ charts. In addition to demographics such as age and gender, baseline assessment of patients’ comorbidity was obtained using Charlson Comorbidity Score [12]. Total length of pancreas resected was defined as the length of the pancreatic specimen as measured in pathology, and length of healthy pancreas resected was computed by subtracting the tumor length to the specimen length.
Fig. 3 Patient distribution within the two groups
Statistical analysis Results Analysis was performed using SPSS 21.0 (IBM Corp., Amonk, NY). Due to the small sample size, continuous data are expressed as medians with inter-quartile range (IQR). Categorical data are reported as absolute numbers (n) and proportions (%). We compared LLDP and MLDP groups using the Mann–Whitney U test, Pearson Chi square test, or Fisher’s exact test, as appropriate. Results were considered significant at p \ 0.05.
We identified 43 patients meeting our inclusion criteria, including 24 MLDP and 19 LLDP (Fig. 3). Table 1 summarizes patient demographics and clinical characteristics of the two groups. Of note, patients undergoing MLPD had a higher mean Charlson score (p = 0.03) and more tumors localized in the proximal third of the pancreatic tail (16.7 vs 0 %) than with LLDP.
123
Surg Endosc Table 1 Patient demographics and clinical characteristics Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
Spleen preservation Planned spleen preservation—n (%)
15 (62.5)
13 (68.4)
0.75
Successful spleen preservation
13 (86.7)
10 (76.9)
0.64
Operating time (min)—median (IQR)
190 (166.25–215.5)
Estimated blood loss (mL)—median (IQR)
300 (175–750)
Conversion to open procedure—n (%)
166 (150–196) 50 (20–100)
4 (16.7)
1 (5.3)
0.03 0.01 0.36
Proportion computed with number of planned spleen preservation as denominator
IQR inter-quartile range
The operative characteristics of both groups are summarized in Table 2. There was no significant difference in the percentage of patients with planned spleen preservation, successful spleen preservation, or rate of conversion to open procedure. The median operating time in the LLDP group was 24 min shorter (p = 0.03), and median estimated blood loss was 250 mL less (p \ 0.01) than that with MLDP. There were fewer conversions to open surgery with LLDP (16.7 vs 5.3 %; p = 0.36), but this was not statistically significant. Post-operative outcomes are detailed in Table 3. All surgical margins were negative with LLDP compared to 2 (8.3 %) microscopically positive margins with MLDP. Median length of total or healthy resected pancreatic parenchyma did not differ. There was a shorter median length of stay with LLDP (4 vs 5 days; p = 0.35), but this was not statistically significant. No difference was observed in major morbidity (MLDP—25.0 % vs LLDP— 21.1 %; p = 0.76).
Discussion This is the first study comparing the traditional medial approach to laparoscopic distal pancreatectomy to the lateral approach. We observed that while obtaining a similar oncological resection, the lateral approach benefited patients in terms of shorter operating time (166 vs 190 min; p = 0.03) and lower estimated blood loss (50 vs 300 mL; p = 0.01). Although not reaching statistical significance, we identified the potential for positive impact on need for conversion to open surgery and length of stay. Since its introduction in the early 1990s [13], laparoscopic distal pancreatectomy has become widely accepted and is now an established operation among hepato-pancreato-biliary surgeons. Resection of the distal pancreas lends itself well to a laparoscopic approach because it involves resection without reconstruction. The medial approach was the first described and has become the
123
standard technique. Several other techniques have been reported including the lateral approach examined in this study [4–6], retroperitoneal [14] and hand-assisted versions [15]. In this still-evolving field, it remains important to critically examine alternative surgical techniques. Our group has been performing both MLDP and LLDP since 2009. Our willingness to try the lateral approach has mainly been driven by several perceived technical advantages that may potentially translate into clinical benefits. Positioning the patient in the right lateral decubitus position allows gravity to help retract both the stomach and the spleen, facilitating access and dissection by a single skilled surgeon. Easier access and greater exposure allow for more precision and quicker dissection that can potentially reduce operative time and blood loss. With this approach, we believe that the splenic vessels are easier to visualize, identify, and isolate at the tail of the pancreas than at the isthmus. Spleen preservation requires complete sparing of the splenic artery and vein that are often embedded in the supero-posterior aspect of the pancreatic body [7, 8]. Separation of the splenic vein from the pancreatic gland during MLDP is responsible for most bleeding during that surgery, due to the challenge imposed by multiple short branches coursing into the pancreatic parenchyma [9]. By finding the vessels laterally and then dissecting them medially, this potentially improves the chances of spleen preservation and decreases the amount of blood loss [16]. With LLDP, the pancreas is divided more laterally which could theoretically preserve pancreatic parenchyma, and lessen the risk of post-operative pancreatic insufficiency. So far, LLDP has been assessed mainly in technical reports or small non-comparative case series [4–6]. In 2012, a group from India reported treatment of four insulinoma cases using the lateral approach for enucleation (n = 3) and spleen-preserving distal pancreatectomy (n = 1) [5]. The only comparative cohort was reported by Nakamura et al., in 2011, with 8 lateral laparoscopic spleen-preserving distal pancreatectomy compared to 15 medial approaches to spleen preservation [6]. Patients who
Surg Endosc Table 2 Operative characteristics Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
Post-operative morbidity (Clavien–Dindo classification)—n (%) Grade 1
0 (0)
1 (5.3)
–
Grade 2
1 (4.1)
2 (10.5)
0.57
Grade 3
5 (20.8)
3 (15.8)
1.00
Grade 4
1 (4.2)
1 (5.3)
1.00
Pancreatic leak (ISGPF Grade C)—n (%)
2 (8.3)
0 (0)
–
Post-operative mortality—n (%) Length of stay (days)—median (IQR)
0 (0) 5 (4–6)
0 (0) 4 (3.25–5)
– 0.35
Neuroendocrine tumor
8 (33.3)
4 (21.0)
0.5
Serous cystadenoma
4 (16.7)
2 (10.5)
0.68
Mucinous cystadenoma
4 (16.7)
2 (10.5)
0.68
Intraductal papillary mucinous neoplasia
3 (12.5)
1 (5.3)
0.62
Solid pseudopapillary tumor
0 (0)
3 (15.8)
–
Pancreatic intraepithelial neoplasia
0 (0)
2 (10.5)
–
Pancreatic adenocarcinoma
1 (4.2)
1 (5.3)
1.00
Pancreatic renal cell carcinoma Metastasis
1 (4.2)
0 (0)
–
Chronic pancreatitis
0 (0)
3 (15.8)
–
Retention cyst
1 (4.2)
0 (0)
–
Epithelial cyst
1 (4.2)
0 (0)
–
Histologic diagnosis—n (%)
Cystic teratoma Accessory spleen embedded in pancreas Tumor size (cm)—median (IQR)
1 (4.2) 0 (0) 2.9 (2–3.6)
0 (0)
–
1 (5.3) 3.5 (2.5–5)
– 0.52 –
Negative resection margins—n (%)
22 (91.7)
19 (100)
Total length of pancreas resected (cm)—median (IQR)
7.1 (4.9–9.0)
6.9 (4.6–9.2)
0.70
Length of healthy pancreas resected (cm)—median (IQR)
3.1 (1.5–6)
3 (1.5–5.5)
0.79
ISGPF International Study Group on pancreatic fistula, IQR inter-quartile range
underwent a lateral operation experienced fewer conversions to open procedure (0 vs 2), and lower blood loss (125 vs 1025 mL; p = 0.07), while operative time and postoperative morbidity did not differ. This report focused on spleen-preserving operations, including 8 patients in the medial group that underwent the Warshaw procedure, as a method of spleen preservation [17]. By including all patients undergoing laparoscopic distal pancreatectomy with or without spleen preservation, our study provides a larger and more pragmatic appreciation of the adoption of LLDP in everyday surgical practice, while yielding similar conclusions in terms of morbidity outcomes. However, we did not observe a difference in the feasibility of spleen preservation. This may be related to the experience and specialized expertise of the participating hepato-pancreatobiliary surgeons who were able to consistently complete spleen-preserving distal pancreatectomy with the medial approach. This review of our experience also points to additional advantages, including a significant decrease in operative
time and blood loss with LLDP. Trends toward decreased need for conversion and length of stay were also observed, although not reaching statistical significance likely due to low statistical power of the study. Perioperative morbidity and mortality were not found to be significantly different, indicating safety of LLDP. Although not reaching statistical significance, the LLDP group included more tumors in the distal third of the pancreatic tail and none in the proximal third. Those patients would appear to be the best candidates for LLDP and could explain the potential benefits surfacing in this analysis. In studies such as this, examining refinements of surgical techniques, differences in post-operative morbidity may not always be appreciable. However, other outcomes such as operative time and blood loss may better reflect small but important differences in terms of technical ease. An easier technique may also translate into broader applicability of laparoscopic distal pancreatectomy for more lesions and by more surgeons when the medial approach might prove too challenging, thus extending the benefits of laparoscopic surgery to more
123
Surg Endosc Table 3 Post-operative outcomes
Age—median (IQR) Male gender—n (%) Body mass index—median (IQR) Charlson Comorbidity Score—median (IQR)
Medial laparoscopic distal pancreatectomy n = 24
Lateral laparoscopic distal pancreatectomy n = 19
p value
59.3 (48.7–66.7)
58.3 (48.2–64.5)
0.58
6 (25.0)
6 (31.6)
28.1 (26.4–29.2)
27.8 (25.0–31.7)
4.5 (2–6)
0.74 0.96
2 (1–4)
0.03
0 (0)
–
6 (31.6) 7 (36.8)
0.37 0.17
Pre-operative diagnosis—n (%) Pancreatic pseudocyst
4 (16.7)
Cystic neoplasm Neuroendocrine tumor
11 (45.8) 4 (16.7)
Solid pseudopapillary tumor
0 (0)
2 (10.5)
–
Pancreatic adenocarcinoma
2 (8.3)
1 (5.3)
1.0
Pancreatic renal cell carcinoma metastasis
1 (4.1)
0 (0)
–
Pancreatic mass not yet diagnosed
2 (8.3)
3 (15.8)
0.64
Tumor localization—n (%) Proximal third
4 (16.7)
0 (0)
–
Middle third
5 (20.8)
3 (15.8)
1.0
Distal third
15 (62.5)
16 (84.2)
0.17
IQR inter-quartile range
patients. Therefore, we believe that the lateral approach could prove superior to MLDP for selected patients with distal pancreatic tail lesions and perhaps more broadly adoptable. We acknowledge that as a retrospective cohort series, this study presents inherent limitations. Due to small numbers and the intrinsic nature of pancreatic tail lesions, there was a great deal of heterogeneity in the pre-operative diagnosis and eventual pathology. The Charlson comorbidity index was higher in the MLDP group, pointing toward potential selection bias by experienced hepatopancreato-biliary surgeons. The difference in tumor localization also highlights the importance of selection in obtaining benefits of LLDP over MLDP, which could not be further explored through sub-group analyses due to the small number of patients. Finally, the small sample size may have hampered potential benefits as indicated by some trends that are not found to be statistically significant. Nonetheless, this study offers a strong appraisal of the benefits of LLDP over traditional MLDP in the largest cohort to date. Our results indicate that LLDP seems to hold promise in further improving outcomes and potentially broadening the use of laparoscopic distal pancreatectomy.
Conclusions The lateral approach is a feasible and safe approach for distal lesions of the pancreatic tail. The improved exposure and ease of finding the splenic vessels offer the potential
123
for shorter operative time and decreased blood loss compared to traditional medial approach, and even possible reduction in conversion rate and length of stay. These hypotheses need to be confirmed in larger prospective studies. Acknowledgments The authors would like to thank Mrs. Laura Maaske ([email protected]) for medical illustrations. Disclosures Matt Strickland—GestSure Technologies (founder, equity interest). Shiva Jayaraman—Astellas (speaker honorarium), Ethicon Endosurgery (consultant and academic support) and Roche (academic support). Julie Hallet, Daniel Abramowitz, Shuyin Liang, and Calvin HL Law have no conflict of interest or financial ties to disclose.
References 1. Kooby DA, Gillespie T, Bentrem D, Nakeeb A, Schmidt MC, Merchant NB et al (2008) Left-sided pancreatectomy. Ann Surg 126:88–96 2. Lebedyev A, Zmora O, Kuriansky J, Rosin D, Khaikin M, Shabtai M et al (2004) Laparoscopic distal pancreatectomy. Surg Endosc 18(10):1427–1430 3. Jayaraman S, Gonen M, Brennan MF, D’Angelica MI, DeMatteo RP, Fong Y, Jarnagin WR, Allen PJ (2010) J Am Coll Surg 211(4):503–509 4. Thomas JK, Abraham D, Joseph P, Paul MJ (2012) Lateral laparoscopic approach to pancreatic tail insulinomas. World J Endocr Surg 4(1):3–7 5. Honore C, Honore P, Meurisse M (2007) Laparoscopic spleenpreserving distal pancreatectomy: description of an original posterior approach. J Laparoendosc Adv Surg Technol 17(5): 686–689
Surg Endosc 6. Nakamura M, Nagayoshi Y, Kono H et al (2011) Lateral approach for laparoscopic splenic vessel-preserving distal pancreatectomy. Surgery 150(2):326–331 7. Boutros C, Espat NJ, Somasundar P (2010) Completely laparoscopic subtotal pancreatectomy with splenic artery preservation. J Gastrointest Surg 14(1):171–174 8. Suzuki O, Tanaka E, Hirano S, Suzuoki M, Hashida H, Ichimura T et al (2009) Efficacy of the electrothermal bipolar vessel sealer in laparoscopic spleen-preserving distal pancreatectomy with conservation of the splenic artery and vein. J Gastrointest Surg 13(1):155–158 9. Melotti G, Butturini G, Piccoli M, Casetti L, Bassi C, Mullineris B et al (2007) Laparoscopic distal pancreatectomy: results on a consecutive series of 58 patients. Ann Surg 246(1):77–82 10. Dindo D, Demartines N, Clavien PA (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240:205–213 11. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J et al (2005) International Study Group on Pancreatic Fistula. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 138:8–13
12. Charlson M, Pompei P, Ales K, MacKenzie C (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373–383 13. Soper NJ, Brunt LM, Dunnegan DL, Meininger TA (1994) Laparoscopic distal pancreatectomy in the porcine model. Surg Endosc 8:57–60 14. Takada M, Ichihara T, Toyama H, Suzuki Y, Kuroda Y (2004) Retroperitoneoscopic laparoscopic distal pancreatectomy with spleen salvage. Hepatogastroenterology 51:925–927 15. Shinchi H, Takao S, Noma H, Mataki Y, Iino S, Aikou T (2001) Hand-assisted laparoscopic distal pancreatectomy with minilaparotomy for distal pancreatic cystadenoma. Surg Laparosc Endosc Percutan Technol 11:139–143 16. Yoon YS, Lee KH, Han HS, Cho JY, Ahn KS (2009) Patency of splenic vessels after laparoscopic spleen and splenic vessel-preserving distal pancreatectomy. Br J Surg 96:633–640 17. Adam JP, Jacquin A, Laurent C, Collet D, Masson B, Ferna´ndezCruz L et al (2013) Laparoscopic spleen-preserving distal pancreatectomy: splenic vessel preservation compared with the Warshaw technique. JAMA Surg 148(3):246–252
123
