Welcome to the Journal of Hepato-Biliary-Pancreatic Sciences Virtual Issue page

View FREE virtual issue articles from Journal of Hepato-Biliary-Pancreatic Sciences

Pancreaticoduodenectomy: improved mortality, but still high morbidity.

Masayuki Ohtsuka
Department of General Surgery
Graduate School of Medicine, Chiba University
Chiba, Japan

One hundred twenty years have elapsed since Codivilla performed the first non-anatomical pancreaticoduodenectomy (abbreviated as PD) in 1898 that remains in the record (Dal Monte B. Rendiconto statistico della sezione chirurgica dell’ospedale d’Imola, anno 1898. Galeati, Imola. 1898, Gamberini C. Arch Ital Mal Appar Dig 1949;15:267-276), and 83 years have passed since Whipple published a report with three cases of two-stage PD (Ann Surg 1935;102:763-779). This publication by Whipple drew surgeons’ attention to the feasibility of this type of operation. After refinements to this procedure, he finally performed one-stage PD including pancreatoenteric anastomosis (Surg Gynecol Obstet 1946;82:623-631). Since then, even in a modern era, several modifications of this procedure has been continued to made, including pylorus preserving PD (PPPD) applied by Traverso and Longmire (Surg Gynecol Obstet 1978;146:959-962, Ann Surg 1980;192:306-309) to maintain gastrointestinal function, and also subtotal stomach preserving PD (SSPPD).

Pancreaticoduodenectomy is now widely performed as a standard procedure for the resection of tumors in the periampullary region. Due to the advancement of surgical instrumentation and techniques, perioperative care, and interventional radiology, the mortality of this procedure has declined dramatically: from 20-30% in the 1970s to less than 5% in the 2000s. Even zero mortality in 368 consecutive pancreaticoduodenectomy was reported. [1] According to the National Clinical Database (NCD) in Japan, the in-hospital mortality was 2.88% in 17,564 pancreaticoduodenectomy procedures conducted at 1,311 hospitals, containing both high and low volume centers, between 2011 and 2012. [2] Several analyses indicate that the mortality after pancreaticoduodenectomy is lowered in high volume centers, as compared with low volume centers. In fact, in the Massachusetts General Hospital series, the mortality in 1,000 consecutive pancreaticoduodenectomy was 1.8%. [3] Furthermore, even among the Japanese Society of Hepato-Biliary-Pancreatic Surgery board-certified training institutions, a significant difference of the 90-day mortality following pancreaticoduodenectomy was found between the A institutions (1.1%), where more than 50 high-level hepatobiliary pancreatic surgeries are performed annually, and the B institutions (1.5%), where more than 30 high-level hepatobiliary pancreatic surgeries are performed annually. [4]

Despite tremendous improvement of the mortality following pancreaticoduodenectomy, however, the morbidity still remains as high as 37-64%. While numerous reports identifying risk factors of postoperative complications and also showing novel strategies to reduce the morbidity after pancreaticoduodenectomy have been so far published, efforts by pancreatic surgeons to improve early outcome after pancreaticoduodenectomy has been continued. In the Journal of Hepato-Biliary-Pancreatic Sciences, several interesting papers regarding postoperative complications after pancreaticoduodenectomy has been published, even in recent years.

1. Pancreatic fistula (PF)

The International Study Group on Pancreatic Fistula (ISGPF) introduced the international consensus definition of postoperative PF with severity grading from A to C in 2005 (Bassi C, et al. Surgery 2005;138:8-13), allowing a reliable comparison among different surgical experiences. In 2016, updated grading system was proposed (Bassi C, et al. Surgery 2017;161:584-591) and a term “fistula” is now applied for only postoperative PF grades B and C (clinically relevant PF).

Clinically relevant PF is still a leading complication following pancreaticoduodenectomy, affecting between 13% and 39% of patients. These rates of PF have not significantly improved between 2006 and 2014. [5] Once PF is developed, serious complications such as hemorrhage due to the rupture of a pseudoaneurysm, surgical site infection, sepsis, and delayed gastric emptying (DGE) can occur consecutively.

There are many reports attempting to identify risk factors for clinically relevant PF. The factors consistently shown between the NCD in Japan [2] and the data from 53 institutions that participated in the Japanese Society of Pancreatic Surgery [5] were increased body mass index (BMI), a high American Society of Anesthesiologists (ASA) score, concomitant vascular resection and non-pancreatic cancer probably reflecting soft pancreatic texture, all of which have already been reported as risk factors for PF following pancreaticoduodenectomy.

On the other hand, pancreatic surgeons have always devised changes in peri- and intra-operative management in order to reduce the occurrence of PF. These include anastomostic technical aspect, how to use pancreatic duct stents, devices for drain management, reinforcement of anastomosis by biological sealants, [6] the usage of somatostatin analogue, and nutritional intervention. However, standardized management has not been established so far. Instead, lots of variations have been created as a result. In fact, Macedo et al [7] reported that significant variations in perioperative management and practice patterns related to pancreaticoduodenectomy across different geographical regions, institution types, and surgeons. Many issues are still under debate, [2] [7] so that it is far from the global standardization of peri- and intra-operative management of pancreaticoduodenectomy, although, in each institutional level, it was shown that standardization might be important in reducing post-pancreaticoduodenectomy complications. [5]

2. Surgical site infection (SSI)

SSI is another complication frequently encountered after pancreaticoduodenectomy, although the striking differences in the rates of SSI was found among literatures, with a range of 10 to 50%. These differences might be attributed to whether the overall rate of SSI refers to all levels incisional and organ/space SSI or only to the incisional SSI.

Several risk factors for the occurrence of SSI after pancreaticoduodenectomy have been reported to include malnutrition, preoperative biliary drainage (PBD), neoadjuvant chemotherapy, long operation time, red blood cell transfusion, concomitant vascular resection and also PF.

Among them, the impact of PBD on postoperative complications including SSI after pancreaticoduodenectomy has been discussed for a long time. Biliary drainage is performed to relieve obstructive jaundice, which could lead to hepatic dysfunction, coagulopathy and biliary infection. Several studies indicated that PBD is associated with an increased rate of morbidity and even mortality, while other studies showed improved morbidity and surgical outcome because of restoration of liver function and soothed inflammatory response. Shaib et al [8] showed, on the basis of the American College of Surgeons-the National Surgical Quality Improvement Program dataset, that sepsis and wound infection were more frequently observed in patients with endoscopic PBD than in patients without PBD. Furthermore, development of PF was also higher in patients with PBD, probably because of the increased risk of infection. In contrast, Sahora et al [3] indicated that PBD increased only the risk of wound infection after pancreaticoduodenectomy, but did not increase the risk of organ or space infection, sepsis, PF, or other major complications. The latter study was based on a single institutional, but a high volume center, experience. Importantly, both studies showed that PBD did not affect the mortality after pancreaticoduodenectomy. Although PBD is certainly associated with wound infection, it may not be necessary to hesitate to perform biliary drainage in cases required it, for example, cases planned neoadjuvant chemotherapy. SSI including wound infection could be decreased by preoperative bile culture-targeted administration of prophylactic antibiotics. [9] However, because most studies have been retrospective nature, a need is recognized for high quality randomized studies.

3. Delayed gastric emptying (DGE)

DGE causes delayed oral intake resulting in nutritional deterioration and prolonged hospital stay. Although exact mechanisms of DGE are not identified, several possible factors affecting DGE are proposed: gastric dysrhythmias due to postoperative PF, ischemia of the pyloric ring and antrum, gastric atony resulted from the absence of motilin, pylospasm due to vagotomy, and the torsion and angulation of the reconstruction.

For the reduction of the incidence of DGE, some technical modifications, including pylous ring resection, Billroth II reconstruction for the duodeno/gastrojejunostomy, and antecolic reconstruction for the duodeno/gastrojejunostomy, have been employed and some positive results have been observed both retrospectively and prospectively. In addition to these technical modifications, recent retrospective studies have shown beneficial effect of Braun anastomosis on reducing DGE after pancreaticoduodenectomy. Furthermore, a prospective randomized controlled trial demonstrated that Braun anastomosis could reduce clinically relevant DGE (Grades B and C according to the International Study group of Pancreatic Surgery consensus definition (Wente MN, et al. Surgery 2007;142:761-768)) with a marginally significant and that not receiving a Braun anastomosis was an independent risk factor affecting clinically relevant DGE following PPPD. [10] However, even if these technical modifications are adopted, DGE is still experienced in 3-22% of patients after pancreaticoduodenectomy.

DGE has not been believed as a life threatening complication after pancreaticoduodenectomy. However, it is reported that DGE may be associated with weight loss and poor nutritional status in patients undergoing pancreaticoduodenectomy in a long-term follow-up. In addition, Futagawa et al [11] reported that patients with DGE, especially with grade C DGE, had significantly poor survival. Although exact cause of this poor survival in patients with grade C DGE has not been clarified, it could be due to delayed initiation and decreased relative dose intensity of adjuvant chemotherapy, as well as prolonged malnutrition and consequent lowered immunity.

Since DGE is a serious complication affecting not only short-term but also long-term outcome following pancreaticoduodenectomy, elucidation of the underlying mechanism of DGE is required.

In addition to the abovementioned major complications, several types of serious systemic complication could be encountered after pancreaticoduodenectomy. [2] In order to reduce the morbidity that has not been improved for decades, more efforts should be taken for understanding of underlying mechanisms of complications and for developing innovative measures to prevent complications. The Journal of Hepato-Biliary-Pancreatic Sciences continues to welcome the submission of high quality papers regarding postoperative complications after pancreaticoduodenectomy.

References

1) Oguro S, Yoshimoto J, Imamura H, Ishizaki Y, Kawasaki S.
Three hundred and sixty-eight consecutive pancreaticoduodenectomies with zero mortality.
J Hepatobiliary Pancreat Sci. 2017;24:226-234.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.433

2) Aoki S, Miyata H, Konno H, Gotoh M, Motoi F, Kumamaru H, Wakabayashi G, Kakeji Y, Mori M, Seto Y, Unno M.
Risk factors of serious postoperative complications after pancreaticoduodenectomy and risk calculators for predicting postoperative complications: a nationwide study of 17,564 patients in Japan.
J Hepatobiliary Pancreat Sci. 2017;24:243-251.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.438

3) Sahora K, Morales-Oyarvide V, Ferrone C, Fong ZV, Warshaw AL, Lillemoe KD, Fernandez-del Castillo C.
Preoperative biliary drainage does not increase major complications in pancreaticoduodenectomy: a large center experience from the Massachusetts General Hospital.
J Hepatobiliary Pancreat Sci. 2016;23:181-187.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.322

4) Otsubo T, Kobayashi S, Sano K, Misawsa T, Ota T, Katagiri S, Yanaga K, Yamaue H, Kokudo N, Unno M, Fujimoto J, Miura F, Miyazaki M, Yamamoto M.
Safety-related outcomes of the Japanese Society of Hepato-Biliary-Pancreatic Surgery board certification system for expert surgeons.
J Hepatobiliary Pancreat Sci. 2017;24:252-261.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.444

5) Satoi S, Yamamoto T, Yoshitomi H, Motoi F, Kawai M, Fujii T, Wada K, Arimitsu H, Sho M, Matsumoto I, Hirano S, Yanagimoto H, Ohtsuka M, Unno M, Yamaue H, Kon M.
Developing better practices at the institutional level leads to better outcomes after pancreaticoduodenectomy in 3,378 patients: domestic audit of the Japanese Society of Pancreatic Surgery.
J Hepatobiliary Pancreat Sci. 2017;24:501-510.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.492

6) Kang JS, Han Y, Kim H, Kwon W, Kim SW, Jang JY.
Prevention of pancreatic fistula using polyethylene glycolic acid mesh reinforcement around pancreatojejunostomy: the propensity score-matched analysis.
J Hepatobiliary Pancreat Sci. 2017;24:169-175.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.428

7) Macedo FIB, Mowzoon M, Parikh J, Sathyanarayana SA, Jacobs MJ.
Disparities in the management and prophylaxis of surgical site infection and pancreatic fistula after pancreatoduodenectomy.
J Hepatobiliary Pancreat Sci. 2017;24:268-280.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.443

8) Shaib Y, Rahal MA, Rammal MO, Mailhac A, Tamim H.
Preoperative biliary drainage for malignant biliary obstruction: results from a national database.
J Hepatobiliary Pancreat Sci. 2017;24:637-642.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.505

9) Okamura K, Tanaka K, Miura T, Nakanishi Y, Noji T, Nakamura T, Tsuchikawa T, Okamura K, Shichinohe T, Hirano S.
Randomized controlled trial of perioperative antimicrobial therapy based on the results of preoperative bile cultures in patients undergoing biliary reconstruction.
J Hepatobiliary Pancreat Sci. 2017;24:382-393.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.453

10) Hwang HK, Lee SH, Han DH, Choi SH, Kang CM, Lee WJ.
Impact of Braun anastomosis on reducing delayed gastric emptying following pancreaticoduodenectomy: a prospective randomized controlled trial.
J Hepatobiliary Pancreat Sci. 2016;23:364-372.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.349

11) Futagawa Y, Kanehira M, Furukawa K, Kitamura H, Yoshida S, Usuba T, Misawa T, Okamoto T, Yanaga K.
Impact of delayed gastric emptying after pancreaticoduodenectomy on survival.
J Hepatobiliary Pancreat Sci. 2017;24:466-474.
https://onlinelibrary.wiley.com/doi/full/10.1002/jhbp.482