Diagnostic and therapeutic role of endoscopic retrograde pancreatography in the management of traumatic pancreatic duct injury patients: Single center experience for 34 years

Accepted Manuscript Diagnostic and therapeutic role of endoscopic retrograde pancreatography in the management of traumatic pancreatic duct injury patients: Single center experience for 34 years Seongyup Kim, MD, Jae Woo Kim, MD, PhD, Pil Young Jung, MD, Hye Youn Kwon, MD, Hongjin Shim, MD, Ji Young Jang, MD, Keum Seok Bae, MD, PhD PII:

S1743-9191(17)30274-1

DOI:

10.1016/j.ijsu.2017.03.054

Reference:

IJSU 3689

To appear in:

International Journal of Surgery

Received Date: 13 January 2017 Revised Date:

6 March 2017

Accepted Date: 17 March 2017

Please cite this article as: Kim S, Kim JW, Jung PY, Kwon HY, Shim H, Jang JY, Bae KS, Diagnostic and therapeutic role of endoscopic retrograde pancreatography in the management of traumatic pancreatic duct injury patients: Single center experience for 34 years, International Journal of Surgery (2017), doi: 10.1016/j.ijsu.2017.03.054. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Diagnostic and therapeutic role of endoscopic retrograde pancreatography in the management of traumatic pancreatic duct injury patients: single center experience for 34 years

Hongjin Shim, MD1; Ji Young Jang, MD1; Keum Seok Bae, MD, PhD1

RI PT

Seongyup Kim, MD1; Jae Woo Kim, MD, PhD2; Pil Young Jung, MD1; Hye Youn Kwon, MD1;

Department of Surgery, Yonsei University Wonju College of Medicine

2

Department of Internal Medicine, Yonsei University Wonju College of Medicine

SC

1

Seongyup Kim, MD: [email protected]

M AN U



Jae Woo Kim, MD, PhD: [email protected]

Pil Young Jung, MD: [email protected] Hye Youn Kwon, MD: [email protected]

TE D

Hongjin Shim, MD: [email protected] Ji Young Jang, MD: [email protected]

Keum Seok Bae, MD, PhD: [email protected]

EP

Conflicts of interest: none

AC C

Disclosure of funding: none

**Correspondence & Request reprints to Keum Seok Bae

Tel:+82-33-741-0884 Fax:+82-33-741-1205 Department of Surgery, Yonsei University Wonju College of Medicine, 220-701 Ilsan-ro, Wonju-si, Gangwon-do, Republic of Korea E-mail: [email protected]

ACCEPTED MANUSCRIPT Diagnostic and therapeutic role of endoscopic retrograde pancreatography in the management of traumatic pancreatic duct injury patients: single center experience for 34 years Abstract

RI PT

Background: Traumatic pancreatic injuries are rare and present diagnostic and therapeutic difficulties. We evaluated the usefulness of endoscopic retrograde pancreatography and

SC

transpapillary pancreatic stent in the diagnosis and treatment of pancreatic trauma patients. Methods: We reviewed medical records of 83 pancreas trauma patients who underwent

M AN U

endoscopic retrograde pancreatography for the suspicion of pancreas duct injury between January 1983 and December 2016. Patient notes, radiologic findings, laboratory investigations, operative records, and endoscopic retrograde pancreatography reports were reviewed.

TE D

Results: Pre-endoscopic retrograde pancreatography abdominal computed tomography was performed in 52 patients. Twenty-seven patients of them were diagnosed with a major pancreatic duct injury by subsequent endoscopic retrograde pancreatography. Eleven major duct

injuries(40.7%)

EP

pancreatic

were

missed

pancreatography abdominal CT. Major pancreatic duct

by

the

pre-endoscopic

retrograde

injury was confirmed by endoscopic

AC C

retrograde pancreatography in 43 patients. These 43 major pancreatic duct injury patients were classified into the Operative(n=21), Stent(n=15), and Conservative(n=7) groups according to the first treatment modality which was chosen by attending surgeon.

Age,

initial white blood cell count, initial serum amylase, associated injury, and major pancreatic duct injury site were similar among groups, while the rate of parenchymal leakage(Dye leakage confined to pancreatic capsule) on endoscopic retrograde pancreatography findings 1

ACCEPTED MANUSCRIPT differed. Parenchymal leakage was most common in the Conservative group. Pancreas related mortality occurred in 1(4.76%) Operative group, 2(13.33%) Stent group, and 0(0%) Conservative group. Pancreas related complication occurred in 16(76.19%) Operative group,

RI PT

10(66.67%), Stent group and 5(71.43%) Conservative group. There were no statistically significant differences in the occurrence rate of pancreas related complications and mortalities among three groups.

SC

Conclusion: Endoscopic retrograde pancreatography helps clinicians choose a treatment modality for major pancreatic duct injury since it provides information about the precise

M AN U

condition of the major pancreatic duct injury. Endoscopic retrograde pancreatography with transpapillary pancreatic stenting also shows promise as a substitute for laparotomy or

Introduction

TE D

pancreatic resection in selected patients.

Traumatic pancreatic injuries are rare and present diagnostic and therapeutic difficulties [1,

EP

2]. The deep retroperitoneal location of the pancreas protects it from less severe trauma, but also renders the diagnosis of injury more difficult. Early detection of pancreatic trauma is

AC C

essential to preventing subsequent complications. Abdominal computed tomography (CT) is the modality of choice for evaluating pancreatic injury in polytrauma patients. Abdominal CT has variable sensitivity and specificity for detecting pancreatic trauma [3, 4].

In addition,

findings of early pancreatic injury are often subtle on CT images and can be missed unless specifically sought [5]. Endoscopic retrograde pancreatography (ERP) is considered the traditional gold standard for the diagnosis of pancreatic duct injuries and provides therapeutic 2

ACCEPTED MANUSCRIPT options such as pancreatic duct stenting and pancreatic sphincterotomy for pancreatic fistula, pseudocyst, and stricture [2, 6-9]. In the present study, we analyzed the diagnostic and

RI PT

therapeutic role of ERP in patients with traumatic pancreatic injuries.

Patients and Methods

The medical records of all abdominal trauma patients who underwent ERP with the clinical

SC

impression of pancreatic injury and were treated at Wonju Severance Christian Hospital, a

M AN U

tertiary institution in Gangwondo, Korea, between January 1983 and December 2016, were reviewed. Patient notes, radiologic findings, laboratory investigations, operative records, and ERP reports were reviewed. Data of patients whose major pancreatic duct (MPD) injury was confirmed by ERP were analyzed. Patients were categorized into three groups by intention to treat for the MPD injury. The Operative group underwent surgery, Stent group underwent

TE D

transpapillary pancreatic duct stent insertion, and Conservative group underwent neither surgery nor transpapillary pancreatic duct stent insertion as the initial treatment method for MPD injury. The treatment methods were chosen by the attending surgeon. Patient mortality

EP

and complications were compared among three groups. Mortality and complication cases

AC C

were divided into two categories, pancreas-related and-unrelated, and then evaluated. Pancreatic fistula was defined as follows: (1) an abnormal communication between the pancreatic ductal epithelium and another epithelial surface containing pancreas derived, enzyme-rich fluid, (2) a failure of healing/sealing of a pancreatic-enteric anastomosis, or a parenchymal leak not directly related to an anastomosis such as one originating from the raw pancreatic surface. In this case, there is a leak from the pancreatic ductal system into and around the pancreas and not necessarily to another epithelialized surface [10]. MPD injuries 3

ACCEPTED MANUSCRIPT were divided into two groups MPD leakage confined to the pancreatic capsule (parenchymal leakage) and MPD leakage beyond the pancreatic capsule by ERP findings. MPD leakage confined to the pancreatic capsule indicates less severe injury [7, 11].

The correlation

RI PT

between MPD stricture and MPD cut-off findings on ERP image was analyzed. In addition, the correlation between MPD stricture and stent placement duration was analyzed.

Quantitative data are reported as mean, standard deviation (SD) and statistic analysis was

SC

performed by Mann-Whitney test or Kruskal-Wallis test. Qualitative data are reported as frequency and percentage and statistic analysis was performed by Fisher’s exact test. All data

M AN U

were analyzed using SPSS version 21.

Results

A total of 83 consecutive patients (65 men, 18 women: mean age, 35.16 ± 14.82 years)

TE D

underwent ERP at Wonju Severance Christian Hospital under the diagnostic suspicion of an MPD injury. Blunt abdominal trauma was the injury mechanism in 82 patients versus a stab injury in one patient. All patients were hemodynamically stable. MPD injury was confirmed

EP

by ERP in 43 patients. The MPD injury site was the pancreatic head in 16 patients, neck in

AC C

four patients, body in 17 patients, and tail in six patients. The median time from trauma to the first ERP was 6 days (range, 0–361 days). Table 1 shows the distribution of time from trauma to ERP. A total of 44 (53.0%) patients underwent ERP < 7 days after the trauma. A total of 120 ERP procedures were attempted; of them, 116 were successful. Pre-ERP abdominal CT was performed in 52 patients. The median time from pre-ERP abdominal CT to ERP was 2 days (range, 0–211 days). Twenty-seven patients were diagnosed 4

ACCEPTED MANUSCRIPT with an MPD injury by subsequent ERP. Eleven MPD injuries (40.7%) were missed by the pre-ERP abdominal CT. Pre-ERP abdominal ultrasonography (USG) was performed in 17 patients. The median time from pre-ERP abdominal USG to ERP was 4 days (range, 0–49

injuries (72.7%) were missed by the pre-ERP abdominal USG.

RI PT

days). Eleven patients were diagnosed with MPD injury by subsequent ERP. Eight MPD

Twenty-one, fifteen, and seven patients were in the Operative, Stent, and Conservative

SC

groups, respectively. Age, initial white blood cell count, initial serum amylase, associated injury, and MPD injury site were similar among groups, while the rate of parenchymal

M AN U

leakage on ERP findings differed (Table 2). Pancreatic injuries for which ERP findings showed parenchymal leakage were most common in the Conservative group and there was a statistically significant difference between the operative and Conservative groups. This finding indicates that MPD injuries in the Conservative group were less severe than those in

TE D

the Operative group.

There were nine distal pancreatectomies, six external drainages, two near-total pancreatectomies, two Whipple operations, one partial pancreatectomy, and one

EP

cystogastrostomy in the operative group. Nine patients in the Operative group underwent pre-

AC C

ERP laparotomy for the surgical indication of hemoperitoneum in eight patients and generalized peritonitis in one. These nine patients underwent postoperative ERP with the following indications: postoperative serum amylase elevation and fluid collection on image study in five, postoperative pancreatic fistula in two, postoperative pseudocyst in one, and equivocal surgical finding suggestive of an MPD injury to the proximal pancreas in one. Four of these nine patients underwent a subsequent external drainage operation, three underwent transpapillary pancreatic stent insertion, one underwent a distal pancreatectomy, and one 5

ACCEPTED MANUSCRIPT underwent a percutaneous drainage procedure. The Operative group had a total of 20 pancreas-related complications in 16 patients: eight postoperative pancreatic fistulas, seven pseudocysts, three peripancreatic abscesses, one duct stenosis, and one patient with diabetes

RI PT

mellitus. There were nine pancreas-unrelated complications in nine patients: enterocutaneous fistulas in three, acute renal failure in one, postoperative ileus in one, cerebral infarction in one, left-side pleural effusion in one, duodenal leakage in one, and multi-organ failure in one.

SC

There were two mortalities in the Operative group: pancreas-unrelated in one and pancreasrelated in one. The cause of pancreas-unrelated mortality was cerebral infarction and

M AN U

pancreas-unrelated was peritonitis due to pancreatic juice leakage.

The Stent group had 17 pancreas-related complications in 11 patients: eight pseudocysts, four MPD strictures, three pancreas atrophies distal to the MPD injury site, and two pancreatic fistulas. There were two pancreas-related mortalities in the Stent group. These two

TE D

patients died of sepsis due to peritonitis with pancreatic juice leakage. One of the mortality patients underwent pancreatic stent removal after 1 week because of severe pancreatitis and then developed severe peritonitis with sepsis due to pancreatic juice leakage. The other

EP

patient had a medical condition of severe alcoholic liver cirrhosis and contracted sepsis resulting from a rapidly progressing intra-abdominal infection. Although these two patients

AC C

underwent subsequent explorative laparotomy and external drainage, they could not recover from the sepsis. There were three stent-related complications: two of mild pancreatitis and one of proximal stent migration. Two mild pancreatitis cases were successfully treated conservatively. The proximally migrated stent was retrieved endoscopically. Three patients required surgery in the Stent group: two were the aforementioned mortality cases, while the third patient contracted peritonitis because of pancreatic juice leakage after stent insertion for 6

ACCEPTED MANUSCRIPT which proximal duct ligation and a pancreaticojejunostomy of the distal pancreas were performed. However, the patient underwent distal pancreatectomy 1 year after the injury for a severe pancreaticojejunostomy site stricture and recurrent pancreatitis. Twelve patients in the

RI PT

Stent group ultimately avoided surgery. The Conservative group had six pancreas-related complications in five patients: four

pseudocysts, one MPD stenosis, and one atrophic pancreas distal to the MPD injury site.

SC

There was no case of pancreas-related mortality in the Conservative group. No pancreasunrelated complications or mortality occurred in the Conservative group. There were four

M AN U

parenchymal leakage findings on ERP; of them, only one contracted a pancreatic pseudocyst. No percutaneous drainage procedures were required in those four patients. Meanwhile, three patients in the Conservative group showed dye leakage beyond the pancreatic parenchyma, and all of them contracted a pancreatic pseudocyst. Two patients underwent a percutaneous

TE D

drainage procedure. One patient visited our hospital 3 months after the injury and showed signs of distal pancreas atrophy and a well-established 2-cm pancreatic pseudocyst, which was managed conservatively.

EP

Table 3 compares the treatment results among the three groups. Although pancreas-related

AC C

complications were most frequent in the Operative group and pancreas-related mortality was most frequent in the Stent group, there was no statistically significant intergroup difference. Overall, there was no significant difference in treatment results among the three groups. Subgroup analysis was performed with regard to pancreas head or neck injury patients. Patients with head or neck MPD injury in ERP findings was 42.86% (n=9) in Operative group, 40% (n=6) in Stent group, and 71.43% (n=5) in Conservative group (Table 2). In these 7

ACCEPTED MANUSCRIPT patients, the pancreas related complications were 55.6% (n=5) in Operative subgroup, 66.7% (n=4) in Stent subgroup, and 0% in Conservative subgroup. The pancreas related mortality was 11.1% (n=1) in Operative subgroup, 16.7% (n=1) in Stent subgroup, and 0% in

RI PT

Conservative subgroup. There were no statistically significant differences in pancreas related complications rate and pancreas related mortality rate among three subgroups. Parenchymal leakage was 33.3% (n=3) in Operative subgroup, 16.7% (n=1) in Stent subgroup, and 60.6%

SC

(n=3) in Conservative subgroup. There were no statistically significant differences in parenchymal leakage among three subgroups. The results of surgery in the Operative

M AN U

subgroup were as follows; there were two Whipple operations. Type of derivative procedure was pancreaticojeunostomy. One patient, aged 64 years old, who underwent Whipple operation died of cerebral infarction at postoperative 14th day; three patients underwent external drainage, two of them suffered from pancreatic fistula and the third one died of

TE D

sepsis; two patients underwent near total pancreatectomy and one patient suffered from diabetes mellitus; two patients underwent distal pancreatectomy and one suffered from diabetes mellitus.

EP

Follow-up images were collected (abdominal CT in five, ultrasonography in two, and endoscopic ultrasonography in one) at least 1 month after stent removal (range, 47–3,820

AC C

days; median, 147.50 days) in eight patients in the Stent group. The MPD stricture group showed more frequent MPD cutoff signs and longer stent duration. However, there was no statistical significant difference among groups. Table 4 details the stent cutoff sign frequency and stent duration. Discussion

8

ACCEPTED MANUSCRIPT The present study showed the 34-year experience of a single center about ERP and transpapillary pancreatic stents in the treatment of MPD injuries. ERP showed an important role as the traditional gold standard in the diagnosis of MPD injury. MPD injuries missed by

RI PT

abdominal CT, ultrasonography, and laparotomy were detected by ERP in our series, whereas MPD injuries missed by ERP were not detected with abdominal CT, ultrasonography, or laparotomy.

SC

CT is the primary imaging modality used in the diagnosis of blunt pancreatic injury because it is easier and faster than ERP. CT also enables surgeons to evaluate other

M AN U

simultaneous intra-abdominal organ injuries. However, as with serum amylase, CT is an imperfect test for the early diagnosis of pancreatic injury [12]. There were some reports that the pancreas may appear normal in 20–40% of patients when CT is performed within 12 hours after trauma, probably due to obscuration of the fracture plane, hemorrhage, and close

TE D

apposition of the pancreatic fragments [3, 13-16]. Moreover, CT has a variable reported sensitivity (54–100%) and specificity (91–94.8%) for detecting MPD injuries [16-19]. Considering the unfavorable results of missed MPD injury in trauma patients, the sensitivity

EP

of CT for diagnosing MPD injuries is somewhat insufficient. The present study also showed a relatively low sensitivity of CT for diagnosing MPD injuries. Eleven of 27 (40.74%) MPD

AC C

injuries were missed by pre-ERP CT. This low diagnostic accuracy of CT may result from the long study period duration because CT images were unable to demonstrate MPD injury in the early study period. We compared the missed duct injury rate by pre-ERP CT in 1983–1999 and 2000–2015. The missed duct injury rate was 50% (six of 12 patients) in the former period and 33.3% (five of 15 patients) in the latter period. We assumed that this result might be due to the development of CT technology during the long study period. 9

ACCEPTED MANUSCRIPT ERP was the original diagnostic tool for MPD injury, but CT has gradually replaced it as technology advanced. Meanwhile, there is an emerging role of ERP in the management of MPD injury patients. Transpapillary pancreatic stent insertion with endoscopy has emerged as

RI PT

an attractive treatment method in the management of MPD injuries. Recent reports have described the successful results of transpapillary pancreatic stent insertion for the

management of pancreatic injuries. Thomson et al. reported seven patients who underwent

SC

transpapillary pancreatic stent insertion for pancreatic duct injuries among 47 patients who underwent ERP after pancreatic trauma [2]. The indications were pancreatic fistula in six

M AN U

patients and MPD stricture in one. They reported that all of the patients with pancreatic fistula successfully avoided surgery. The treatment results of patients with MPD strictures were not reported in that study. Kong et al. compared endoscopic and non-endoscopic management groups of patients who underwent non-operative management for MPD injuries

TE D

[20]. The endoscopic management group underwent ERP or ERP with intervention; 76% of those patients underwent nasopancreatic drain placement or transpapillary pancreatic stent placement. Pancreas-related complications occurred in 46% of the patients in the non-

EP

endoscopic management group and 26% in the endoscopic management group. The nonoperative management failure rate was 30% in the non-endoscopic management group and 9%

AC C

in the endoscopic management group. Although not all patients in the endoscopic management group underwent MPD stent insertion, the better treatment results seen in the endoscopic management group may have resulted from the high MPD stent insertion rate. Rogers et al. reported the treatment results of 26 MPD injuries confirmed by ERP. Among them, nine underwent MPD stent insertion; of them, seven avoided further surgery [6]. Our study also showed the usefulness of the endoscopic treatment of MPD injury patients.

10

ACCEPTED MANUSCRIPT Although three of 15 patients underwent surgery after MPD stent insertion, 12 ultimately avoided surgery. In this study, the treatment group was further divided into three groups and analyzed. The

RI PT

patient characteristics were similar between the Operative and Stent groups. Although there was no statistical difference, the parenchymal leakage rate was higher in the Stent group (operative group, 14.29%; stent group, 26.67%). Takishima et al. reported that either

SC

nonsurgical treatment or laparotomy without pancreatic resection was satisfactory for patients in whom dye leakage was confined to the pancreatic capsule on ERP findings [11]. As such,

M AN U

the parenchymal leakage rate difference indicates that patients with more severe MPD injuries were allocated to the Operative group in the present study. Nevertheless, the pancreas-related mortality rate was 4.76% in the Operative group and 13.3% in the Stent group. Although there was no statistically significant intergroup difference, the Stent group

TE D

showed a higher mortality rate. One of the mortality cases in the Stent group had underlying severe alcoholic liver cirrhosis, a potent risk factor of intra-abdominal infection, and ultimately died of sepsis. It may be assumed that such a high-risk patient would die regardless

EP

of treatment method. Another mortality case in the Stent group underwent stent removal 8 days after insertion because of severe pancreatitis. This patient underwent subsequent

AC C

drainage operation due to generalized peritonitis with sepsis induced by pancreatic juice leakage and died 1 month after stent removal. These two pancreas-related mortality cases in the Stent group show the importance of prudent case selection and close observation after stent insertion when transpapillary pancreatic stent insertion is selected as the primary treatment modality for patients with MPD injury. In 2005, our hospital implemented a practice policy to perform CT evaluations before pancreatic stent insertion whenever possible 11

ACCEPTED MANUSCRIPT to detect pancreatic juice spillage extended to the peritoneal cavity and physical examinations to detect generalized peritonitis. In cases of severe intra-abdominal fluid collection or signs of generalized peritonitis, we treat the patient surgically. Since 2005, there have been no cases of

RI PT

mortality in patients who underwent transpapillary pancreatic stent insertion as the primary treatment modality for MPD injury.

There were no cases of pancreas-related mortality in the Conservative group. The ERP

SC

findings of the Conservative group regarding MPD injury were parenchymal leakage in five, communication with pseudocyst without peritoneal cavity or retroperitoneal space leakage in

M AN U

one, and duct stenosis at the neck to body MPD in one. The patients with ERP findings of MPD leakage communicated with pseudocyst visited our hospital 7 days after the trauma. The last patient visited our hospital 3 months after the trauma. We assumed that since there was no MPD leakage extended to the peritoneal cavity or retroperitoneal space in any patients

TE D

in the Conservative group that no cases of pancreas-related mortality occurred. Theoretically, Whipple operation or pylorus preserving pancreaticoduodenectomy is the definite surgery of pancreas head injury. However, the complex and time consuming nature,

EP

and high postoperative morbidity of Whipple operation or pylorus preserving

AC C

pancreaticoduodenectomy frequently prevent the application of these ‘definite surgery’ in pancreas head injury patients as pancreatic injury patients usually have severe associated injuries and hemodynamic unstability. In the present study, 9 proximal MPD injury patients (head or neck injury in ERP findings) underwent initially surgical treatment. However, only two patients underwent Whipple operation and there was no pancreas related morbidity in these two patients. Meanwhile, 5 of 7 (71.4%) patients who underwent external drainage 12

ACCEPTED MANUSCRIPT (n=3) or near total pancreatectomy (n=2) or distal pancreatectomy (n=2) suffered from pancreas related morbidity.

Although no pancreatic duct occlusion technique after

pancreaticoduodenectomy without derivative procedure (pancreaticojejunostomy or

RI PT

pancreaticogastrostomy) was applied in our series, the pancreatic duct occlusion technique may be not a reasonable substitute for Whipple operation or pylorus preserving

pancreaticoduodenectomy. Conzo et al made a conclusion based on extensive literature

SC

review and review of their own series that as pancreatic duct occlusion techniques after

pancreaticoduodenectomy are associated with highest rate of postoperative complications,

M AN U

especially pancreatic fistula, so derivative procedure should be preferred except in selected high risk elderly patients in oncologic surgery [21, 22]. Meanwhile, the complication rate of Stent group with proximal MPD injury was 66.7% in this study and the incidence of complication is slightly lower than that of proximal MPD injury patients who did not

TE D

underwent Whipple operation or pylorus preserving pancreaticduodenectomy. In addition, Conzo et al reported morbidity of derivative procedure after pancreaticoduodenectomy was ranged from 22.6% to 66.7% which was comparable with the result of Stent group with

EP

proximal MPD injury in this study [21]. Although the present study is retrospective study and the study population was too small to enable the detection of statistical differences, the result

AC C

of this study showed the promise of ERP with transpapillary pancreatic stenting as a substitute for Whipple operation or pylorus preserving pancreaticoduodenectomy. In the present study, transpapillary pancreatic stent was inserted to treat postoperative pancreatic fistula in two cases after left pancreatectomy in the operative group. Two patients were treated successfully. Grobmyer et al. reported that eight patients who developed refractory Grade C postoperative pancreatic fistula after left pancreatectomy were 13

ACCEPTED MANUSCRIPT successfully treated with transpapillary pancreatic stent insertion [23]. Reddymasu et al. also reported similar results in eight patients with Grade B pancreatic fistula after left pancreatectomy. All of the patients were successfully treated with transpapillary pancreatic

RI PT

stent insertion [24]. These results are similar to those of our study with regard to postoperative pancreatic fistula management. Recently, the use of preoperative pancreatic duct stenting to prevent postoperative pancreatic fistula has been evaluated [25-28].

SC

However, the results of these studies are not universally applicable because these studies did not include trauma patients.

M AN U

MPD stricture is a major long-term complication after pancreatic trauma [9]. This complication can result from the trauma itself or the pancreatic stent [29]. Some authors assert that longer stent placement has an increased possibility of ductal stricture [9, 30]. In our study, the first stent duration was longer in the MPD stricture group than in the no MPD

TE D

stricture group. Total stent duration was also longer in the MPD stricture group. In addition, MPD cutoff on ERP findings was more frequent in the MPD stricture group. We could not identify whether stent duration or MPD cutoff had a more dominant influence on MPD

EP

stricture occurrence.

AC C

Our study has some limitations. First, it is retrospective study and has an excessively long study period. Retrospective studies routinely have problems with data collection and group comparability. A long study period aggravates such problems. Our study period, 34 years, is long enough to change the clinical outcomes of certain diseases as advances occur in medication, surgical techniques, and postoperative care. Second, the study population was too small to enable the detection of statistical differences. Traumatic pancreatic injuries are uncommon, so we believe that multicenter data collection will be necessary to surmount this 14

ACCEPTED MANUSCRIPT problem. Conclusion ERP is very useful for the diagnosis and treatment of MPD injury. ERP helps clinicians

RI PT

choose a treatment modality for MPD injury since it provides information about the precise condition of the MPD injury. ERP with transpapillary pancreatic stenting also shows promise

M AN U

SC

as a substitute for laparotomy or pancreatic resection in selected patients.

Conflict of Interest: The authors declare no conflicts of interest Funding Source: None

[1].

AC C

References

EP

TE D

Ethical Approval: None required

Iacono C, Zicari M, Conci S, Valdegamberi A, De Angelis M, Pedrazzani C, et al.

Management of pancreatic trauma: A pancreatic surgeon's point of view. Pancreatology. 2016;16(3):302-8. [2].

Thomson DA, Krige JE, Thomson SR, Bornman PC. The role of endoscopic

retrograde pancreatography in pancreatic trauma: a critical appraisal of 48 patients treated at

15

ACCEPTED MANUSCRIPT a tertiary institution. J Trauma Acute Care Surg. 2014;76(6):1362-6. [3].

Gupta A, Stuhlfaut JW, Fleming KW, Lucey BC, Soto JA. Blunt trauma of the

pancreas and biliary tract: a multimodality imaging approach to diagnosis. Radiographics.

[4].

RI PT

2004;24(5):1381-95. Stuhlfaut JW, Anderson SW, Soto JA. Blunt abdominal trauma: current imaging

techniques and CT findings in patients with solid organ, bowel, and mesenteric injury. Semin

[5].

Kumar A, Panda A, Gamanagatti S. Blunt pancreatic trauma: A persistent diagnostic

M AN U

conundrum? World J Radiol. 2016;8(2):159-73. [6].

SC

Ultrasound CT MR. 2007;28(2):115-29.

Rogers SJ, Cello JP, Schecter WP. Endoscopic retrograde cholangiopancreatography

in patients with pancreatic trauma. J Trauma. 2010;68(3):538-44. [7].

Kim HS, Lee DK, Kim IW, Baik SK, Kwon SO, Park JW, et al. The role of

TE D

endoscopic retrograde pancreatography in the treatment of traumatic pancreatic duct injury. Gastrointest Endosc. 2001;54(1):49-55. [8].

Kozarek RA. Endoscopic therapy of complete and partial pancreatic duct disruptions.

[9].

EP

Gastrointest Endosc Clin N Am. 1998;8(1):39-53. Lin BC, Liu NJ, Fang JF, Kao YC. Long-term results of endoscopic stent in the

[10].

AC C

management of blunt major pancreatic duct injury. Surg Endosc. 2006;20(10):1551-5. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J, et al. Postoperative

pancreatic fistula: an international study group (ISGPF) definition. Surgery. 2005;138(1):8-13. [11].

Takishima T, Hirata M, Kataoka Y, Asari Y, Sato K, Ohwada T, et al.

Pancreatographic classification of pancreatic ductal injuries caused by blunt injury to the pancreas. J Trauma. 2000;48(4):745-51; discussion 51-2.

16

ACCEPTED MANUSCRIPT [12].

Potoka DA, Gaines BA, Leppaniemi A, Peitzman AB. Management of blunt

pancreatic trauma: what's new? Eur J Trauma Emerg Surg. 2015;41(3):239-50. [13].

Jeffrey RB, Jr., Federle MP, Crass RA. Computed tomography of pancreatic trauma.

[14].

RI PT

Radiology. 1983;147(2):491-4. Cirillo RL, Jr., Koniaris LG. Detecting blunt pancreatic injuries. J Gastrointest Surg.

2002;6(4):587-98.

Venkatesh SK, Wan JM. CT of blunt pancreatic trauma: a pictorial essay. Eur J

SC

[15].

Radiol. 2008;67(2):311-20.

Teh SH, Sheppard BC, Mullins RJ, Schreiber MA, Mayberry JC. Diagnosis and

M AN U

[16].

management of blunt pancreatic ductal injury in the era of high-resolution computed axial tomography. Am J Surg. 2007;193(5):641-3; discussion 3. [17].

Wong YC, Wang LJ, Fang JF, Lin BC, Ng CJ, Chen RJ. Multidetector-row computed

TE D

tomography (CT) of blunt pancreatic injuries: can contrast-enhanced multiphasic CT detect pancreatic duct injuries? J Trauma. 2008;64(3):666-72. [18].

Phelan HA, Velmahos GC, Jurkovich GJ, Friese RS, Minei JP, Menaker JA, et al. An

EP

evaluation of multidetector computed tomography in detecting pancreatic injury: results of a multicenter AAST study. J Trauma. 2009;66(3):641-6; discussion 6-7. Panda A, Kumar A, Gamanagatti S, Bhalla AS, Sharma R, Kumar S, et al. Evaluation

AC C

[19].

of diagnostic utility of multidetector computed tomography and magnetic resonance imaging in blunt pancreatic trauma: a prospective study. Acta Radiol. 2015;56(4):387-96. [20].

Kong Y, Zhang H, He X, Liu C, Piao L, Zhao G, et al. Endoscopic management for

pancreatic injuries due to blunt abdominal trauma decreases failure of nonoperative management and incidence of pancreatic-related complications. Injury. 2014;45(1):134-40.

17

ACCEPTED MANUSCRIPT [21]. Conzo G, Gambardella C, Tartaglia E, Sciascia V, Mauriello C, Napolitano S, et al. Pancreatic fistula following pancreatoduodenectomy. Evaluation of different surgical approaches in the management of pancreatic stump. Literature review. International journal

RI PT

of surgery. 2015;21 Suppl 1:S4-9. [22]. Mauriello C, Polistena A, Gambardella C, Tartaglia E, Orditura M, De Vita F, et al. Pancreatic stump closure after pancreatoduodenectomy in elderly patients: a retrospective

[23].

SC

clinical study. Aging clinical and experimental research. 2017;29(Suppl 1):35-40.

Grobmyer SR, Hunt DL, Forsmark CE, Draganov PV, Behrns KE, Hochwald SN.

M AN U

Pancreatic stent placement is associated with resolution of refractory grade C pancreatic fistula after left-sided pancreatectomy. Am Surg. 2009;75(8):654-7; discussion 7-8. [24].

Reddymasu SC, Pakseresht K, Moloney B, Alsop B, Oropezia-Vail M, Olyaee M.

Incidence of pancreatic fistula after distal pancreatectomy and efficacy of endoscopic therapy

TE D

for its management: results from a tertiary care center. Case Rep Gastroenterol. 2013;7(2):332-9. [25].

Fischer CP, Bass B, Fahy B, Aloia T. Transampullary pancreatic duct stenting

EP

decreases pancreatic fistula rate following left pancreatectomy. Hepatogastroenterology. 2008;55(81):244-8.

Sugiyama M, Abe N, Yamaguchi Y, Yamato T, Nakaura H, Matsuoka H, et al.

AC C

[26].

Preoperative endoscopic pancreatic stenting for safe local pancreatic resection. Hepatogastroenterology. 2001;48(42):1625-7. [27].

Hirota M, Kanemitsu K, Takamori H, Chikamoto A, Ohkuma T, Komori H, et al.

Local pancreatic resection with preoperative endoscopic transpapillary stenting. Am J Surg. 2007;194(3):308-10; discussion 11-2. 18

ACCEPTED MANUSCRIPT [28].

Frozanpor F, Lundell L, Segersvard R, Arnelo U. The effect of prophylactic

transpapillary pancreatic stent insertion on clinically significant leak rate following distal pancreatectomy: results of a prospective controlled clinical trial. Ann Surg.

Kozarek RA. Pancreatic stents can induce ductal changes consistent with chronic

pancreatitis. Gastrointest Endosc. 1990;36(2):93-5. [30].

Ikenberry SO, Sherman S, Hawes RH, Smith M, Lehman GA. The occlusion rate of

AC C

EP

TE D

M AN U

pancreatic stents. Gastrointest Endosc. 1994;40(5):611-3.

SC

[29].

RI PT

2012;255(6):1032-6.

19

ACCEPTED MANUSCRIPT

TABLE 1. Distribution of Time from Trauma to ERP* Number of patients (n = 83)

0 days

7 (8.43%)

1 day

12 (14.46%)

2 days

10 (12.05%)

3 days

7 (8.43%)

RI PT

Time to first ERP

4 days

3 (3.61%)

5 days

2 (2.41%)

6 days

3 (3.61%)

14 (16.87%)

15–21 days

SC

7–14 days

10 (12.05%)

>21 days

15 (18.07%)

AC C

EP

TE D

M AN U

*ERP, endoscopic retrograde pancreatography

ACCEPTED MANUSCRIPT

Patients’ Clinical Variables by Treatment Group Operative group

Stent group

(n = 21)

(n = 15)

Conservative group

Overall

Initial white blood cell count (/L)

15,163 ± 5,662

Initial serum amylase (U/L) Associated

stent group

conservative

conservative

P value**

group

group

P value**

P value**

0.924

0.836

0.783

0.347

0.026

0.162

0.419

0.297

0.535

33.6 ± 17.1

0.952

13,293 ± 6,082

8,535 ± 4,798

0.082

1,450 ± 3,301

935 ± 1,107

1,175 ± 1,033

7 (33.33%)

6 (40.00%)

1 (14.29%)

0.485

0.474

0.633

0.350

Associated thorax injury

2 (9.52%)

2 (13.33%)

1 (14.29%)

0.913

1.000

1.000

1.000

Associated pelvic bone injury

0 (0.00%)

1 (6.67%)

0 (0.00%)

0.385

0.417

1.000

1.000

Associated CNS injury

0 (0.00%)

0 (0.00%)

0 (0.00%)

1.000

1.000

1.000

1.000

Associated long bone injury

1 (4.76%)

3 (20.00%)

0 (0.00%)

0.195

0.287

1.000

0.523

MPD injury site

9 (42.86%)

6 (40%)

5 (71.43%)

0.428

1.000

0.385

0.361

4 (57.14%)

0.087

0.418

0.043

0.343

Dye leakage confined to pancreatic capsule (parenchymal leakage)

EP

AC C

(head and neck)

TE D

other intra-abdominal organ injury

3 (14.29%)

4 (26.67%)

SC

36.5 ± 14.8

Stent vs

0.460

M AN U

36.4 ± 14.6

Operative vs

P value*

(n = 7)

Age (yr)

Operative vs

RI PT

TABLE 2.

CNS, central nervous system; MPD, major pancreatic duct *Statistics were analyzed by Kruskal-Wallis test. **Statistics were analyzed by Mann-Whitney test.

ACCEPTED MANUSCRIPT

Operative group

Stent group

Conservative

(n = 21)

(n = 15)

group

Overall

P value*

(n = 7)

Operative vs

Stent vs

stent group

conservative

conservative

P value**

group

group

P value**

P value**

1.000

1.000

1.000

0.559

1.000

1.000

2 (13.33%)

0 (0%)

Pancreas-related mortality

1 (4.76%)

2 (13.33%)

0 (0%)

Pancreas-unrelated mortality

1 (4.76%)

0 (0%)

0 (0%)

1.000

1.000

1.000

1.000

Overall complications

18 (85.71%)

11 (73.33%)

5 (71.43%)

0.532

0.418

0.574

1.000

Pancreas-related complications

16 (76.19%)

10 (66.67%)

5 (71.43%)

0.899

0.709

1.000

1.000

Pancreas-unrelated complications

6 (28.57%)

4 (26.67%)

0 (0%)

0.313

1.000

0.288

0.263

Follow-up duration (months)

39.7 ± 58.3

38.2 ± 52.8

51.1 ± 102.6

SC

1.000

Operative vs

2 (9.52%)

0.745

M AN U

TE D

EP

Overall mortality

RI PT

TABLE 3. Treatment Results by Group

AC C

*Statistics were analyzed by Kruskal-Wallis test. **Statistics were analyzed by Mann-Whitney test.

ACCEPTED MANUSCRIPT

TABLE 4. The Relation Between MPD stricture and MPD cut off/Stent duration

P value*

3 (75%)

0 (0%)

0.143

61.75 ± 24.68

47.75 ± 58.36

0.386

315.75 ± 340.91

87.50 ± 91.38

0.248

4.75 ± 3.10

2.75 ± 1.71

0.304

findings First stent duration

Total stent duration

M AN U

(days)**

(days)*** Number of ERP

RI PT

No MPD stricture (n = 4)

SC

MPD cut off on ERP

MPD stricture (n = 4)

TE D

procedures performed

MPD, major pancreatic duct; ERP, endoscopic retrograde pancreatography *Statistics were analyzed by Fisher’s exact test and Mann-Whitney test **First stent duration: time between first stent insertion and removal

AC C

EP

***Total stent duration: time between first stent insertion and last stent removal

ACCEPTED MANUSCRIPT Highlights Endoscopic retrograde pancreatography is very useful for the diagnosis and treatment of major pancreatic duct injury. Endoscopic retrograde pancreatography helps clinicians choose

RI PT

a treatment modality for major pancreatic duct injury since it provides information about the precise condition of the major pancreatic duct injury. Endoscopic retrograde pancreatography with transpapillary pancreatic stenting also shows promise as a substitute for laparotomy or

AC C

EP

TE D

M AN U

SC

pancreatic resection in selected patients.