Current management of pancreatic trauma

358 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma A Systematic Review M.I. Korontzi, Ch. Kontovounisios, Ch. C. Karaliotas, S. Lanitis, G.Sgourakis, T. Papakostantinou, C.Karaliotas Received 29/06/2010 Accepted 23/08/2010

Abstract Aim-Background: Albeit uncommon, pancreatic trauma is associated with high morbidity and mortality and strict criteria for its management is lacking. The objective of this study is to review and evaluate the role of intraoperative imaging, current management practice and the potential complications of both the surgical and non-surgical approach in regard to this entity. Methods: The present review pooled the data of studies published in the English literature mainly during the last decade. Fifteen reviews and eleven clinical studies were identified in the following databases: medline, pubmed, scholar google, scopus. Results: The patient’s haemodynamic status, integrity of the pancreatic duct, site of duct injury and presence of a concomitant duodenal injury are the major determinant factors in deciding final management. Intraoperative radiological techniques for the evaluation of the pancreatic duct integrity include: i) intraoperative cholangiogram, ii) intraoperative pancreatography by cannulation of the duct and iii) intraoperative Endoscopic Retrograde Cholangiopancreatography (ERCP) which is the most accurate method for defining ductal injury. The spectrum of pancreatic injuries ranges from simple contusions or lacerations which are managed nonoperatively or with external drainage, to major injuries with disruption of the pancreatic duct which are best treated by pancreatic resection. Complex pancreatoenteral anastomoses have also been advocated in the literature but are ill-advised as they are related to high rates of postoperative complications and are particularly time consuming at acute phase. In combined pancreatoduodenal injuries, each organ merits its own management depending on the severity of injury, with pancreatoduodenectomy being reserved for pancreatic head maceration and destruction of the ampulla of Vater.

2nd Surgical Department and Surgical Oncology Unit, General Hospital Korgialenio-Benakio Red Cross, Athens-Greece e-mail: [email protected]

In our systematic review, the overall morbidity of pancreatic trauma was found to be 40%. Morbidity increased with a higher grade of injury (grade I and II: 27%, grade III: 56%, grade IV and V: 69%). The rate of surgical complications for all grades was 46% (grade I and II: 7%, grade III: 57%, grade IV,V: 67%). The Non-Operative Management (NOM) morbidity rate for all grades of injury was found to be 33%. NOM is mainly reserved for low grade injuries and haemodynamically stable patients with no concomitant injuries. We found that NOM complication rates for low grade injuries reached 33%, while for high grade injuries this rose to 50%. Conclusion: The principles for the management of pancreatic trauma have not been clearly delineated. Existing reports are retrospective and single centre with a limited number of patients. Prospective, multicentre trials are warranted to ascertain strict criteria and evaluate the current recommendations.

Keywords

Pancreatic injury, Management of pancreatic trauma, Intraoperative pangreatography, Surgical techniques

Introduction The aim of this systematic review is to address the following three topics: i) the role of intraoperative imaging in the diagnosis and treatment of pancreatic injuries, ii) the current management practice of pancreatic trauma in relation to the grade of the injury, iii) the complications of both the surgical and non-surgical approach. The pancreas is the fourth most commonly injured organ and accounts for 5% of patients with blunt abdominal trauma, 6% of patients with gunshot wounds to the abdomen and 2% of stab wounds to the abdomen [1, 2]. Approximately 90% of pancreatic trauma is also associated with other intraabdominal injuries [3, 4]. Most commonly, blunt injury to the pancreas results from a direct blow to the mid-abdomen which compresses the pancreas against the vertebra, as typically occurs in traffic accidents where an unrestrained individual is thrust against the steering wheel of a car or the handle

35

36

359 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

bars of a bicycle or motorcycle [3]. Hwang et al, in a retrospective review of 75 patients with pancreatic trauma, identified the following prognostic determinants: i) the status of the main pancreatic duct, ii) surgical complexity, iii) associated abdominal venous injuries (only by univariate analysis), iv) associated colon and stomach injuries (only by univariate analysis), v) base deficit, vi) transfusion requirements (associated with mortality only) [5]. Bradley et al asserted that mortality and morbidity increased when detection of ductal injury was delayed [6], whereas Kao et al correlated the grade of pancreatic injury with morbidity and mortality [7]. The main factor to affect prognosis is pancreatic duct disruption and this is also the factor that most influences the management strategy [4]. The prompt and early diagnosis of pancreatic trauma is of paramount importance in decreasing morbidity and mortality. The triad of abdominal tenderness, leukocytosis, and hyperamylasaemia is neither sensitive nor specific for pancreatic injury [8]. Clinical findings that bear a high index of suspicion for pancreatic injury are lower rib fractures, soft tissue ecchymosis, supraumbilical seat belt sign and upper lumbar spine fractures [9]. The initial serum amylase level is not reliably accurate for the diagnosis of pancreatic trauma; its accuracy could possibly improve if amylase is measured more than three hours after injury [10]. Nevertheless, an elevated amylase level prompts further investigation with a contrast enhanced C/T or Endoscopic Retrograde Cholangiopancreatography (ERCP) [2]. The haemodynamically unstable patient must be taken to the operative room without unnecessary delay; the stable patient warrants a further diagnostic workup. Contrast enhanced C/T is the cornerstone of diagnostic evaluation of pancreatic trauma. Specific signs of pancreatic injury on C/T include: fracture, laceration, pancreatic enlargement with oedema, pancreatic haematoma, extravasation of IV contrast from peripancreatic arteries and fluid separation of the pancreas and splenic vein. C/T sensitivity for pancreatic trauma is reported to be 85% for the first 24 hours and 90% overall. In the first 12 hours after injury, C/T may appear normal but a continued vigilance with repeated C/T is warranted in stable patients suspicious for pancreatic trauma[11].The newer multidetector C/T (MDCT) offers much greater sensitivity and specificity (91%) for the diagnosis of pancreatic duct injury [12]. If the C/T is equivocal or there are signs of pancreatic injury, Endoscopic Retrograde Cholangiopancreatography (ERCP) is the gold standard method for defining the integrity of the pancreatic duct which

is the most important determinant of prognosis. ERCP can also offer a therapeutic advantage by means of stent placement [4]. Magnetic Resonance Cholangiopancreatography (MRCP) has equal and dynamic secretin-stimulated MRCP (DSS-MRCP) higher sensitivity compared to ERCP for the visualization of the pancreatic duct with the added advantage of being non-invasive [8]. MRCP can therefore precede ERCP when there is a high suspicion for duct involvement and C/T is equivocal [4]. The diversity of pancreatic injuries and the ambiguity of their clinical presentation poses a great challenge to the trauma surgeon. To date, there is no consensus regarding the management of pancreatic injuries and the series found in the literature are not sufficiently large to enable definitive guidelines to be devised

Methods An extensive review of the English literature was carried out, mainly of the last decade, using the following databases: medline, pubmed, scholar google, scopus. The key words used were pancreatic injury, management of pancreatic trauma , intraoperative pancreatography, surgical techniques, complications. Fifteen reviews and 11 clinical studies were identified.

Results The main limitations of the clinical studies found in the literature search concerned the small number of patients in the target groups, the retrospective design and the fact that most of these studies originated from a single centre. Another difficulty lay in the different classification systems for pancreatic injuries used by many authors. Table 1 Smego classification of pancreatic injuries Grade Criteria I

Contusion/hematoma with intact capsule and no parenchymal injury

II

Parenchymal injury without major duct injury.

III

Parenchymal disruption with presumed major ductal injury

IV

Massive parenchymal disruption

There are three main classification systems for pancreatic injuries (Tables 1, 2, 3). They all focus on the extent of parenchymal injury [13-15]. The Lucas system does not acknowledge the pancreatic duct as a prognostic determinant while Smego does not focus on the location of the parenchymal injury (proximal or distal). The most recent and widely used pancreatic classification system is that estab-

360 Hellenic Journal of Surgery 2010; 82: 6

lished by the American Association for the Surgery of Trauma (AAST) which acknowledges the significance of complex injuries affecting the pancreatic duct and the head of the pancreas (table III) [15]. For the homogeneity of our systematic review, all pancreatic injuries identified in the literature were converted to the AAST grading system. Table 2 LUCAS classification of pancreatic injury Class

Criteria

I

Simple superficial contusion or peripheral laceration with minimal parenchymal damage, any portion of the pancreas may be affected but intact main pancreatic duct

II

Deep laceration, perforation or transection of the neck, body or tail of the pancreas, with or without pancreatic duct injury.

III

Severe crush, perforation or transection of the head of the pancreas, with or without pancreatic duct injury

IV

Combined pancreaticoduodenal injuries: (a) minor pancreatic injury, (b) severe pancreatic and also duct injury

Table 3 The Organ Injury Scaling Committee of the American Association for the Surgery of Trauma for pancreatic trauma Class

Criteria

I

Simple contusion of the pancreas

II

Major contusion or laceration without tissue loss or involvement of the main pancreatic duct

III

Complete transection of the pancreas or a parenchymal injury with involvement of the major duct to the left of the superior mesenteric vein

IV

Ductal transection or a major parenchymal injury to the right of the superior mesenteric vein

V

Massive disruption of the head of the pancreas

The role of intraoperative imaging During laparotomy, choosing the appropriate surgical technique depends upon the patient’s haemodynamic status, pancreatic duct continuity and the presence or absence of concomitant duodenal injury [16]. Intraoperative radiological methods for the evaluation of the pancreatic duct integrity include: i) intraoperative cholangiogram by the cystic duct or the common bile duct. The intrapancreatic bile duct and the ampulla can be assessed; the pancreatic duct is visible if there is reflux of the contrast material [14, 17]. ii) intraoperative pancreatography by cannulation of the duct through the tail of the pancreas or via a transduodenal approach. Although older series have reported a reduction in mortality and a 15%-55% decrease in morbidity with the use of intraoperative pancreatography [18], these techniques are not advocated by more recent series. The

The Current Management of Pancreatic Trauma

transduodenal approach bears a high risk of duodenal leakage, while distal pancreatic transection to obtain a pancreatography via the tail is neither acceptable nor feasible, owing to the small diameter of the peripheral pancreatic duct [14, 17]. Jurkovich at al suggest that a careful inspection can offer the same accuracy as intraoperative pancreatography in regard to the diagnosis of ductal disruption [19, 20]. iii) intraoperative ERCP is the most accurate method for defining the extent and location of ductal injuries and conducive to treatment planning [21]. However, it is not without its limitations: it is not readily available, can only be performed when the patient is stable and concomitant hollow viscous injuries have been excluded, requires a skilled endoscopist and is associated with pancreatitis and infection risk [17, 21]. Although Degiannis et al do not advocate intraoperative evaluation of the pancreatic duct, they believe that intraoperative cholangiography is the most appropriate method for stable patients. If ductal injury is suspected, albeit not conclusively proven, they opt for distal pancreatectomy. If suspicion for ductal injury in an unstable patient is low, they recommend simple external drainage. Should the patient’s clinical status mandate it, ERCP is deferred until the postoperative period. [4, 17]. In line with this approach, Pata et al and Schroeppel and Croce also reserve ERCP for equivocal cases [22, 23]. On the other hand, Kao et al in their retrospective review of 75 patients with pancreatic injury concluded that intraoperative imaging was successful in demonstrating ductal injury in 75% of cases [7]. However, prospective studies are needed to define the role of intraoperative imaging in detecting ductal disruption. Treatment of American Association for the Surgery of Trauma (AAST) grade I and II pancreatic trauma Mattix et al conducted a retrospective multicentre trial of 173 paediatric trauma patients with blunt pancreatic injuries. Non-Operative Management (NOM) was successful in 74%. Of the 128 patients who did not sustain a pancreatic duct injury (AAST grade I and II), 45 failed NOM [24]. In an extensive review of the literature, Spaniolas and Velmahos concluded that NOM was only an option in pancreatic trauma without major duct involvement (AAST grade I and selected grade II ). The reported success rates vary from 74% to 95% [9]. Grade II injuries (no major duct involvement) can be managed conservatively with a higher incidence of complications, mainly pancreatitis and pseudocyst. The pseudocyst formation rate after NOM for pancreatic trauma reaches 45-50% [9].

37

38

361 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

Subramanian and Feliciano concur with the aforementioned findings and claim that if there is no evidence of pancreatic duct disruption on fine-cut C/T or ERCP, NOM is favoured [16]. Almond and Jaunoo concluded in their review that minor contusions and lacerations can be managed conservatively, but a high index of suspicion should be maintained and C/T should be repeated in 4-10 days or sooner if there is any clinical deterioration [4].On the contrary, Farrell et al. in a retrospective analysis of 51 patients with pancreatic trauma confirmed at laparotomy, found that fistulas occurred mainly in cases where the pancreatic duct injury was missed at laparotomy. Complications were more frequent in blunt trauma compared with gunshot or stab wounds and in higher grades of injury. Based on their results, the authors concluded that minor contusions and lacerations without major pancreatic duct involvement should be managed by external drainage [25]. The same notion is supported by Krige et al, who insist on control of bleeding and simple external drainage, leaving any capsular tears intact [14]. The repair of capsular lacerations in the presence of a missed ductal injury may lead to pseudocyst formation and must be avoided [26, 27]. In an attempt to stop bleeding, it is preferable to ligate bleeding vessels or use an omental plug as a haemostatic, rather

than suturing parenchymal lesions which can lead to pancreatic tissue necrosis [10]. In line with the above considerations, Al Ahmadi and Ahmed performed a retrospective study of 25 patients with pancreatic injury, including 18 AAST grade I and II injuries. Six were managed by haemostasis, debridement of devitalised tissue and external drainage. A complication rate of 2/6 was reported among the operative group (2 fistulas) and 1/11 among the non-operative group (1 pancreatitis). The 2 fistulas occurred after the repair of a grade II pancreatic injury. The limitations of this study are its small sample size, the retrospective analysis, the inability to follow up 24% of the patients included in the study and the fact that the morbidity data is not directly related to the grade of the injuries[26]. In an older review of the literature, Wilson and Moorehead postulated that simple drainage would suffice for contusions and lacerations as long as the main pancreatic duct was intact [28]. In conclusion, out of a total number of 180 patients with AAST grade I and II gathered from all the clinical series reviewed, 139 were treated conservatively with a complication rate of 33%. The complication rate for the remaining 41 patients that were treated surgically was 7%. The morbidity rate for low grade injury was 27% in total (Table 4 a,b).

Table 4a Management of low grade pancreatic injuries (AAST grade I, II) No of patients

AAST grade I, II

retrospective multicenter

173

128

128

Farrell et al [25]*

retrospective single center

51

nm

Al Ahmadi et Ahmed [26]*

retrospective single center

25

18

Author

Type of trial

Mattix et al [24]

Management NoM External drainage

Morbidity

Need for laparotomy

-

45/128 NOM

nm

nm

35

1pancreatitis/ 35ext.drainage

nm

11

6

1/11 NOM

nm

2 fistulae/6 ext.drainage

nm

*Pancreatic injuries were classified in the original articles according to Lucas classification. The results were converted to AAST classification.

Table 4b Morbidity for Operative and Non Operative Management (NOM) of low grade (AAST I and II) pancreatic injuries Management

No of patients

Complications

Operative

41

3 (7%)

NOM

139

46 (33%)

TOTAL

180

49 (27%)

362 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

sidered distal pancreatectomy with splenectomy as the best way to deal with grade III injuries stressing the risk of splenic vein thrombosis if the spleen is preserved [28]. Degiannis et al studied 48 patients with gunshot injuries to the pancreas and treated 14 parenchymal disruptions with presumed major ductal injury with debridement and external drainage. All patients developed a pancreatic duct leak that was treated conservatively. They recommend that most of these injuries can be debrided and drained unless a severe duodenal injury also exists, in which case resection is needed. The use of the Smego pancreatic grading system, the retrospective design and the limited number of patients are the drawbacks of this study [13]. In an older case series with 43 pancreatic injuries, Leppaniemi et al endeavoured to compare distal resection versus drainage alone for the treatment of distal pancreatic injuries with ductal involvement. They found a lower complication rate with distal pancreatectomy as opposed to that of drainage alone, but the number of patients in this study was

Treatment of grade Americal Association of the Surgery of Trauma (AAST) grade III pancreatic trauma Most authors agree that a distal pancreatectomy is indicated for grade III pancreatic injuries with transection of the pancreas to the left of the mesenteric vessels [16, 21]. In the case of a stable child under 10 years of age or a stable adult with an isolated pancreatic injury, splenic salvage should be considered. If the patient is haemodynamically unstable, distal pancreatectomy with splenectomy is the procedure of choice [16]. Distal pancreatectomies comprise 29% of operations for pancreatic trauma and the fistula rate is 14%, which explains why suction drains should be left in place [10]. Accordingly, in a retrospective study of 25 pancreatic trauma patients, Al-Ahmadi and Ahmed performed 4 distal pancreatectomies with a stapling device for grade III injuries with 2 postoperative pseudocysts. However, this case series was small and retrospective, the Lucas classification system was used and only 76% of the patients could be followed up [26]. In an older review, Wilson and Moorehead conTable 5a Management of AAST grade III pancreatic injuries Author

Type of trial

No of patients

AAST grade 3 4

Distal pancreatectomy

Management Roux-en-Y External drainage

Complications NOM

AlAhmadi et Ahmed [26]*

retrospective single centre

25

Degiannis et al [13]**

retrospective single centre

48

14

0

0

14

0

Leppaniemi et al [29]*

retrospective single centre

43

15

9

1

4

0

Vane et al [30]

retrospective single centre

14 children

4

0

0

4

0

Kao et al [7]

retrospective single centre

193

32

26

nm

4

nm

Fahy et al [31]

retrospective single centre

15

nm

15

0

0

0

9/15 (9 fistulae)

Farrell et al [25]*

retrospective single centre

51

6

6

0

0

0

1/6 (1 fistula)

Pata et al [22]

retrospective single centre

11

6

0

0

0

6

2/6

4

0

0

0 2/4 (2 pseudocysts) 18/14

12 fistulae 3 abscesses 2 pseudocysts 1 heamorrhage

Distal pancreatectomy: 2/9 (2 fistulae) Ext. drainage: 3/4 0 Distal pancreatectomy: 12/26 (12 fistulae) Ext.drainage:3/4 (3 fistulae)

1 abscess 1 fistula

nm: not mentioned *, ** : Pancreatic injuries were classified in the original articles according to Lucas* or Smego** classification. The results were converted to AAST classification.

39

40

363 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

Table 5b Morbidity for Operative and Non Operative Management (NOM) of AAST grade III pancreatic injuries Management

Operative NOM TOTAL

No of patients

Complications

87

50 (57%)

6

2 (33%)

93

52 (56%)

also limited [29]. In a paediatric population study, 4 grade III pancreatic injuries were treated with external drainage with excellent results. However, the limitations of this single centre study lie in its lack of randomisation, the confounding variables and its retrospective design [30]. In Kao’s at al retrospective series from 32 patients with grade III injuries, 4 were treated with drainage alone and 3 (75%) developed pancreatic fistulas. Twenty-six patients with grade III injury underwent distal pancreatectomy with or without splenectomy, and pancreatic fistulas developed in 12 (43%) [7]. Fahy et al present a higher rate of pancreatic fistula after distal pancreatectomy for trauma 9/15 [31]. This discrepancy could possibly be attributed to the different definition of pancreatic fistula . Farrel et al studied 51 patients with pancreatic trauma (blunt or penetrating) classified according to Lucas grading system. They performed 6 distal pancreatectomies for contusions, lacerations, stab or gunshot wounds in the body/tail with ductal involvemnent. A fistula developed in just one distal pancreatectomy in a patient who presented at the hospital 48h after injury [25]. A rarely performed alternative to distal pancreatectomy is Roux-en-Y distal pancreaticojejunostomy. Its only indication is for the stable patient with a transection to the neck of the pancreas or just right to the mesenteric vessels. In such a case, distal pancreatectomy results in resectioning 75-80% of the pancreas, which leads to an abnormal glucose tolerance or true hyperglycaemia in 50% of the patients [32]. Pata et al attempted to extend NOM to AAST grade III pancreatic injuries and treated 6 stable patients with grade III injury with NOM and octreotide. One patient developed a peripancreatic abscess and another a pancreatic fistula. Both complications were successfully treated conservatively. Again, the limitation of this study is the small size of the target population [22].

To conclude, out of the 93 cases of AAST grade III pancreatic trauma reported in the literature, 87 were surgically treated with a complication rate of 57%, whereas complications were shown in 2 of the 6 patients (33%) in the NOM group. The overall morbidity for AAST grade III pancreatic trauma was 56% (Table 5a,b). Management of AAST classified injuries grade IV and V Grade IV and V pancreatic injuries (Table 6a) are often accompanied by pancreaticoduodenal injuries. Treatment options depend on the severity of both pancreatic and duodenal trauma. The surgical technique is dictated by the extent of injury to the duodenum (Table 3), the integrity of the ampulla of Vater and the distal common bile duct [10]. Alternative treatments found in the literature for grade IV pancreatic injuries are many: i) Rouxen-Y distal pancreaticojejunostomy, ii) anterior Roux-en-Y pancreaticojejunostomy, iii) pancreatoduodenectomy, iv) endoscopically placed stents, v) damage control and simple drainage. Grade V injuries are routinely managed with pancreatoduodenectomy since the massive disruption of the pancreatic head does not permit reconstruction. The Whipple procedure is unavoidable when there is extensive trauma to the head of the pancreas, a severe combined pancreatoduodenal injury or destruction of the ampulla of Vater [2]. Severe injuries to the duodenum often mandate some kind of diversion procedure, in order to divert gastric and biliary contents away from the repair [14]. This is achieved by pyloric exclusion, duodenal diverticulisation, Ttube drainage [10] and triple tube drainage [2]. In a retrospective study of 193 pancreatic injuries identified by Kao et al, the seven grade IV injuries who survived for over 48h were not treated in a consistent manner: 3 were managed with drainage, 2 with distal panctreatectomy, 1 with resection plus internal drainage (Roux - en - Y pancreatojejunostomy) and 1 with pyloric exclusion. The pancreatic complication rate for grade IV injuries was 33%

364 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

Table 6a Management of AAST grade IV and V pancreatic injuries

retropective, single center

51

Al-Ahmadi and Ahmed [26]*

retropective, single center

Vane et al [30]

retropective, single center

Leppaniemi et al [29]*

retropective, single center

Lochan et al [3]

retropective, single center, (injuries not classified)

25

14 (children) 43

11

7

3

7

2

1

0

nm

0

1 (for grade V)

0

4

11

4 (for grade V)

0

4

3

1 (for grade IV)

0

4

0

2 (+ necrosectomy)

2

2

nm

2 (for grade IV) 0 2 (for grade IV)

0

1

0

0

1

0

0

grade V

3

Pancreatic Complications

2/7

1/3 grade V

7

Pyloric exclusion

5/7

6/7 grade V2

193

Rouxen-Y

1/2 1/1 fistula fistula 0

0

1 (for grade IV)

0

0

0

0

grade V

retropective, single center

Management NOM + Distal Octreotide pancreatectomy

External drainage

grade IV

Grade V

grade IV

Grade IV

grade IV2

Farrell et al [25]*

No of patients

grade IV

Kao et al [7]

Type of trial

nm

Author

3/3 fistulae

1 fistula (after surgery) 1 pseudocyst (after NOM)

nm: not mentioned 1 : died of multiple injuries after surgery 2 : kind of complications not explained *: Pancreatic injuries were classified in the original articles according to Lucas classification. The results were converted to AAST classification.

Table 6b Morbidity of Operative and Non Operative Management (NOM) of AAST grade IV and V pancreatic injuries Management

Operative NOM TOTAL

No of patients

Complications

30

20 (67%)

2

1 (50%)

32

22 (69%)

(2/7). The three patients with grade V injuries who survived for more than 24h were all treated with pancreatoduodenectomy and showed a pancreatic complication rate of 33% (1/3). The authors concluded that the pancreatic complications in relation to mortality were not statistically significant and that the AAST grade is a predictor for operative strategy, pancreatic morbidity and mortality [7]. In Farrell’s et al retrospective review of 51 patients with pancreatic trauma, 7 AAST grade IV and 7 AAST grade V injuries were identified: 2 involved

severe injuries to the pancreatic head with or without duct injury and 12 displayed combined panctreaticoduodenal injuries. The authors concluded that injuries to the head of the pancreas without devitalisation of pancreatic tissue should be treated with external drainage, providing any duodenal injury is amenable to simple repair. Internal drainage procedures (Roux - en - Y pancreatojejunostomy) are ill-advised since they are associated with significant morbidity and mortality. More severe devitalising combined pancreaticoduodenal injuries with dis-

41

42

365 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

Table 7 Overall morbidity of operative and non operative management (NOM) of pancreatic injuries Management

No of patients

Complications

Operative

158

73 (46%)

NOM

147

49 (33%)

TOTAL

305

122 (40%)

Table 8

Duodenal organ injury scale by the American Association for the Surgery of Trauma AAST. (D1 = first portion of the duodenum, D2 = second portion, D3 = third portion, D4 = fourth portion). Grade

I

II

III

IV

V

Criteria Hematoma

Involving a single portion of the duodenum

Laceration

Partial thickness without perforation

Hematoma

Involving more than one portion

Laceration

Disruption of > 50% of the circumference

Laceration

Disruption of 50–75% of the circumference of D2 Disruption of 50–100% of the circumference of D1, D3 or D4

Laceration

Disruption of > 75% of the circumference of D2 Disruption involving ampulla or common pancreatic duct

Laceration

ruption of the ampulla of Vater (AAST grade V) in stable patients are treated with a typical Whipple or pylorus preserving pancreatoduodenectomy. A total of 7 pancreatoduodenectomies were performed for grade V injuries, resulting in 0% mortality and 86% (6/7) complication rate. The pitfalls of this study concerned the limited number of patients, its retrospective design, the different pancreatic injury classification (Lucas) and the lack of available data regarding the mortality and complication rate of each operative procedure [25]. Al-Ahmadi and Ahmed identified 25 patients suffering from pancreatic trauma. The overall pancreatic morbidity rate in their series was 31,8%. Two patients with grade IV injuries were managed with distal pancreatectomies, while one grade V injury was managed with debridement, packing and external drainage in a two-stage procedure. These authors postulate that pancreatic head injuries are often associated with other life-threatening injuries and generally require damage control and external drainage [26]. If a proximal main pancreatic duct injury is detected by MRCP or ERCP, transpapillary pancreatic duct stenting is an advisable solution,

Massive disruption of pancreas-duodenum Vascular Devascularization of the duodenum

providing the patient is haemodynamically stable and asymptomatic [11, 21, 33, 34]; it is also a viable choice if surgical intervention is contraindicated due to the presence of comorbidities [9]. Patients with complete duct transection should be offered surgery [21]. For combined pancreaticoduodenal injuries with the distal common bile duct and ampulla intact, primary repair and drainage is the indicated option. The duodenal repair should be accompanied by pyloric exclusion or triple tube drainage due to the high rate of postoperative leakage. If the pancreatic head, duodenum, common bile duct and the ampulla are destroyed, a Whipple procedure may be required but should be performed in a two-stage fashion, after damage control surgery [4, 33]. Vane et al designed a paediatric population study with 14 cases of AAST grade II or higher pancreatic injury. Four patients had AAST grade IV injuries, all with main pancreatic duct involvement, who underwent external drainage without pancreatectomy with no complications. The only patient with grade V injury died of multiple injuries after surgery. The outcome for early drainage was excellent even for high grade injuries. The limitations to this study

366 Hellenic Journal of Surgery 2010; 82: 6

concern the fact that it is a retrospective, single centre study with no randomisation [30]. One older literature review from Wilson and Moorehead included 61 patients with pancreatic trauma, of whom 21 were treated with drainage only with 3 pancreatic and 6 non-pancreatic complications and a 9% mortality rate. The authors concluded that major injuries to the pancreatic head without duct destruction should be addressed by simple drainage. For duct destruction, they propose anterior Roux-en-Y pancreaticojejunostomy. They concur with the majority of authors that in regard to combined injuries, the Whipple procedure should be reserved for the most severe cases and propose safer options such as anterior Roux-en-Y and duodenal diversion by pyloric exclusion or duodenal diverticulisation [28]. Leppaniemi et al carried out a retrospective analysis of 43 patients with acute pancreatic trauma classified according to Lucas classification. Four patients had severe injuries to the head of the pancreas (AAST grade IV), one of whom was treated with drainage, 1 with Roux-en-Y anastomosis and 2 with distal resection. Three patients with combined pancreaticoduodenal injuries (AAST grade V) were all treated with pancreaticoduodenectomy. Owing to the high complication rate of pancreaticoduodenectomy (24-40%) in most series, the authors stress the need to apply strict criteria for the selection of patients who would benefit from a Whipple procedure [29]. Recently, Lochan et al treated 11 patients suffering pancreatic trauma conservatively with octreotide, 4 of whom had pancreatic head injuries with main pancreatic duct involvement (AAST grade IV). Two were successfully treated with octreotide, one of whom developed long term formation of pseudocysts. The other two patients underwent open necrosectomy and external drainage with one fistula formation. This study indicates that octreotide may be an alternative treatment option for pancreatic injury especially when patients present early post-injury [3]. Nonetheless, the target group of this study is limited and the pancreatic injuries are not classified according to AAST classification. In conclusion, out of a total of 32 patients with AAST grade IV and V pancreatic trauma found in the literature, 30 underwent some kind of surgical treatment and 2 were managed non-operatively. Morbidity among surgical patients was 67%, while the complication rate among non-surgical patients was 50%. The total complication rate for AAST IV and V pancreatic trauma reached 69% (Table 5b). In this systematic review, the overall morbidity of pancreatic trauma was 40% (Table 7).The mor-

The Current Management of Pancreatic Trauma

bidity rose according to the grade of injury (grade I and II: 27%, grade III: 56%, grade IV and V: 69%) These percentages are consistent with the morbidity presented by most authors [29, 35]. The rate of surgical complications for all grades was 46% which is also within the range of 26%-86% presented in the literature [2]. Postoperative complications presented in 7% of grade I and II injuries, 57% of grade III and 67% of grade IV and V injuries (Tables 4b, 5b, 6b,7). Our systematic review found NOM morbidity for all grades of injury to be 33% (Table 7). NOM is mainly reserved for low grade injuries and haemodynamically stable patients with no concomitant injuries. We found NOM complication rates to be 33% for low grade injuries (table 4b), 33% for grade III (Table 5b) and 50% for grade IV, V injuries (Table 6b). Since few patients with high grade injuries were treated non-operatively, no valid conclusion could be drawn regarding the value of this approach.

Discussion Pancreatic trauma presents challenging diagnostic and therapeutic dilemmas to the trauma surgeon due to the retroperitoneal location of the pancreas, the frequent lack of clinical signs and the absence of a sensitive biochemical abnormality. Associated organ injuries can mask pancreatic injury, both in imaging studies and intraoperatively, largely because of major bleeding or intestinal spillage [3, 4]. Pancreatic trauma is associated with 40%-60% morbidity and 12%- 31% mortality [2, 29, 36, 37]. Most published series report a mortality rate of 19% [29]. Pancreatic trauma can constitute a serious abdominal injury since in approximately 90% of cases it presents with concomitant intrabdominal injuries, especially to the liver and spleen [3, 4]. After the small bowel, the colon is the second most commonly affected organ in gunshot injuries, with the transverse colon being the most frequently involved segment. There is class III evidence that the coincidence of pancreatic and colon injuries is associated with an increased rate of septic complications and mortality. However, there is no evidence that this combination, per se, is a contraindication for primary repair or anastomosis of the colon.[38]. Recent studies report that most early deaths in pancreatic trauma are associated with massive haemorrhage from great vessel injury, while late deaths are usually related to concomitant injuries [7]. Associated injuries to major vessels, wounds of the head of the gland, and failure to adequately control the leakage of exocrine secretion have been previously reported in the literature as determinants of mortality [39].

43

44

367 The Current Management of Pancreatic Trauma

Isolated pancreatic trauma is associated with a 3%-10% mortality rate; mortality and morbidity are directly related to the degree of pancreatic injury and the integrity of the pancreatic duct, particularly if there is a delay in the detection of the pancreatic duct disruption [11, 17]. According to Leppaniemi et al, mortality in proximal pancreatic injuries (head and neck) reaches 19% and in distal injuries (tail) 17% [29]. Other reasons for the high mortality and morbidity of pancreatic trauma include diagnostic difficulty, the time that lapses between injury and diagnosis and the lack of a consensus approach regarding the management of pancreatic injuries [37]. The current trend of Non-Operative Management (NOM) of abdominal solid organ injuries could lead to a dramatic increase in missed pancreatic injuries with devastating consequences. Patients with abdominal trauma should be monitored diligently and a continued vigilance should be maintained. Published series up to now have not been sufficiently large to enable the development of definitive treatment guidelines for pancreatic trauma. The lack of a consensus approach gives rise to considerable disparity among authors regarding the appropriate treatment planning for each grade of pancreatic injury. The choice of the appropriate surgical technique depends upon the patient’s haemodynamic status, pancreatic duct continuity and the presence or absence of concomitant duodenal injury [16]. Some authors have supported NOM for stable patients with isolated blunt pancreatic injuries, providing there is no ductal disruption. In such cases, ductal integrity should be verified by ERCP or MRCP [11]. Patients who present as haemodynamically unstable or suffer from concomitant hollow viscous or vessel injury should undergo laparotomy without extensive preoperative workup. Intraoperatively, the first priority in multiple abdominal organ injuries with extrapancreatic haemorrhage or gastrointestinal spillage is to control the haemorrhage and isolate the site of gastrointestinal contamination, after which the pancreas is assessed. A high index of suspicion for pancreatic trauma is raised by the following findings: i) lesser sac fluid collection, ii) retroperitoneal bile staining, iii) haematoma overlying the pancreas, iv) fat necrosis of the omentum or retroperitoneum. Such findings demand complete mobilization of the pancreas and assessment of the duct continuity. Assessment of duct integrity is of paramount importance in the selection of the appropriate operative strategy to secure optimal outcome [17]. The intraoperative criteria of main pancreatic duct injury were first described by Heitsch et al in

Hellenic Journal of Surgery 2010; 82: 6

1976 and quoted by Degiannis et al in their extensive review: i) direct visualization of ductal violation, ii) complete transection of the pancreas, iii) laceration of more than half the diameter of the pancreas, iv) central perforation and severe maceration [14, 17, 40]. Injury to the pancreatic duct occurs in 15% of pancreatic trauma and is more frequent in penetrating injuries [10]. The full inspection of the body and tail of the pancreas calls for the division of the gastrocolic ligament and the entrance to the lesser sac, while an extended Kocher manoeuvre is required for exposure of the pancreatic head. The posterior aspect of the pancreas is exposed by dividing the retroperitoneum at the inferior edge of the pancreas, whereas the posterior aspect of the tail requires the mobilization of the spleen [16]. The trajectory of penetrating wounds should be traced all the way to exclude other visceral or vessel injuries [14]. Contusions and lacerations without duct disruption (grade I) represent 70% of pancreatic injuries [14]. If the patient is haemodynamically stable, minor pancreatic contusions (grade I), minor capsular injuries and traumatic pancreatitis can be treated by simple observation. Major contusions (grade II) are safely treated by external drainage. Lacerations without duct disruption (grade I and II) are often accompanied by bleeding; omental pancreatorrhaphy with simple external drainage avoids the suture necrosis that is associated with simple pancreatorrhaphy and also offers the advantage of wide drainage in the event of an undetected minor pancreatic duct injury [32]. Should the integrity of the main pancreatic duct be in doubt, some authors propose one of the following intraoperative imaging techniques: i) cholangiogram by the cystic duct or the common bile duct [17], ii) pancreatography by cannulation of the duct via the transected tail or transduodenally [20], though recent reports advise against this technique [17], iii) intraoperative ERCP [21]. Other authors prefer external drainage and postoperative ERCP for cases with low suspicion for ductal injury, proceeding to distal pancreatectomy only in the event of high suspicion [17]. There is much debate in the literature regarding whether a sump drain or a closed system suction is the more appropriate type of drainage. In a prospective randomised trial, sump drains have been associated with higher rates of bacterial colonisation and sepsis but this is not confirmed in other studies which found the two drain types to be equal [41, 42]. A closed suction drain is mostly preferred and offers better control of pancreatic excretions [14]. Distal transection or parenchymal injury with

368 Hellenic Journal of Surgery 2010; 82: 6

duct disruption to the left of the mesenteric vessels (grade III) mandates a distal pancreatectomy [2, 10, 43]. Splenic salvage is surrounded by considerable controversy in the literature. However, when the patient is stable and the injury is limited to the pancreas, splenic salvage can be considered [44]. This is particularly important for children under 10 years of age. In the haemodynamically unstable patient, splenectomy is indicated in conjunction with distal pancreatectomy [32]. Management of the pancreatic stump is also widely argued in the literature. One method advocates stapling the pancreatic stump. A small section of the gastrocolic ligament can be sewn to the stapling line, thus sealing small pancreatic branch ducts not closed by the staples. Another alternative is to tailor the anterior and posterior oblique planes of transition in fish mouth pattern and suture the two surfaces with horizontal mattress sutures. This technique avoids suturing with tension which can lead to necrosis and fistula and pseudocyst formation [16, 31]. A further option is to seal the pancreatic duct with sutures and ligate the duct separately [45]. Some reports advocate the drainage of the resection margin in a Roux-en-Y pancreatojejunostomy in an effort to reduce the increased rate of pancreatic fistula following distal pancreatectomy [28]. However, this procedure is associated with an increased risk of postoperative anastomotic leakage and therefore not recommended [14]. Other authors postulate that the method of stump closure is not related to the rate of fistula formation [46]. Roux-en-Y distal pancreatojejunostomy is an alternative to distal pancreatectomy and is rarely performed. Its only indication is in the stable patient with a transection at the neck of the pancreas or just proximal to the mesenteric vessels, in which case extended distal pancreatectomy may result in abnormal glucose tolerance. This procedure involves ligation of the proximal main pancreatic duct and mobilisation of the distal pancreatic stump and its end-to-end anastomosis with a Roux segment of jejunum which is mobilised via the transverse mesocolon to the right of the mesenteric vessels [32]. In grade IV injuries, if no duodenal injury exists and the main pancreatic duct is intact simple drainage is advocated. The same option is applied when the main pancreatic duct is injured, the ampulla is intact and there is no devitalisation [14]. The only indication for Roux-en-Y distal pancreatojejunostomy is the stable patient with a transection to the neck of the pancreas or just right to the mesenteric vessels. As already mentioned, this offers the advantage of avoiding the endocrine insufficiency related to extended distal pancreatectomy [47].

The Current Management of Pancreatic Trauma

Anterior Roux-en-Y pancreatojejunostomy can be performed when the pancreatic duct in the head is transected and the posterior parenchyma is preserved. During this procedure, a Roux limb is mobilised through the transverse mesocolon and anastomosed in an end-to-end fashion over the site of the injury [2].This technique is not indicated in the acute setting due to high rates of anastomotic breakdown [43], but if performed, extensive drainage is warranted [32]. Endoscopically placed transpapillary stents promote healing of duct disruptions by blocking the leaking duct, bridging the disruption, or transversing the pancreatic sphincter converting the high pressure pancreatic duct system to a low pressure system with preferential flow through the stent [21]. Patients eligible for this management practice are haemodynamically stable patients with partial and isolated pancreatic duct injuries [48]. Another indication for endoscopic stents is the trauma patient with significant brain or other severe injuries preventing complex surgical repairs [2]. Duct strictures are a possible complication of stent placement and the subject of ongoing research. Further studies are needed to validate the use of endoscopic stents in pancreatic trauma patients. In combined pancreatoduodenal injuries, each organ merits its own surgical treatment. Table 8 describes the duodenal injury scale as defined by the American Association for the Surgery of Trauma (AAST). Grade I or II duodenal haematomas (involving one or more portions of the duodenum) can be managed conservatively as long as there is no permanent obstruction or a full thickness laceration of the duodenal wall [9]. If a duodenal laceration does not involve an injury to the common bile duct or ampulla, the laceration is repaired and the pancreatic injury is treated according to the site of the injury. If there is evidence of duodenal ischaemia, the site of the ischaemia should be debrided and primary duodenal closure should be attempted. In this case, the anastomosis can be secured by placing the nasogastric tube in the duodenal loop beyond the anastomosis [14]. Should there be severe injury to the duodenum, one of the following diversion techniques can be performed: i) pyloric exclusion, in which technique the pyloric muscle is closed through a dependent gastrotomy with a number 1 polypropylene suture and an antecolic gastrojejunostomy is performed. In 95% of cases the pyloric muscle ring reopens in 2-3 weeks; alternatively, the sutures can be removed endoscopically. This allows a temporary diversion of gastric contents, allowing the duodenal and pancreatic injury to heal [2, 17,

45

46

369 The Current Management of Pancreatic Trauma

49]. Although pyloric exclusion was the preferred method of duodenal decompression in the last decade, a recent study demonstrated that this method displayed an increased rate of morbidity and did not improve clinical outcome, suggesting that simple repair of duodenum without pyloric exclusion is a safer method of management [50], ii) duodenal diverticulisation is another complex diversion technique that involves primary closure of the duodenal injury, a vagotomy, antrectomy with an end-to-end gastrojejunostomy, T-tube common bile duct drainage and a tube duodenostomy [14, 51], iii) the triple tube approach involves the placement of a gastrostomy tube for proximal decompression, a retrograde duodenostomy tube inserted through the jejunum for the decompression of the duodenum and an antegrade jejunostomy tube for enteral feeding. This technique is frequently complicated by fistulas and leaks [2]. The last two methods of diversion have fallen out of favour as they have proved both time consuming and inappropriate in a damage control situation [17], iv) T-tube drainage is performed by closing the injury over a T-tube and is indicated for combined injuries involving the second part of the duodenum [10]. Pancreaticoduodenectomy (Whipple procedure) has a mortality rate of 30-40% [2] and may be indicated for 2-3% of pancreatic trauma cases and 10% of combined pancreaticoduodenal injuries [29]. A grade V pancreatic injury with massive destruction of the pancreatic head routinely mandates pancreatoduodenectomy [2]. The criteria that have been proposed for pancreatoduodenectomy in trauma patients are: i) extensive damage to the head of the pancreas and duodenum, not amenable to reconstruction, ii) disruption of the duct in the pancreatic head together with injury to the distal common bile duct and the duodenum iii) destruction to the ampulla of Vater, iv) uncontrollable bleeding from vessels of the pancreatic head, v) exsanguinating retropancreatic portal or superior mesenteric vein injury [14]. The Whipple procedure can be performed on a stable patient during the original laparotomy, whereas in an unstable patient it should be done at second stage after the patient has been stabilised by a damage control procedure [2]. Pancreatic surgery for trauma presents a complication rate of 26%-86% in different series [2]. Operative and non-operative complications can be divided into three groups: i) Pancreatic complications: fistula, pseudocyst and pancreatitis, ii) Abdominal complications: intra-abdominal abscess, enterocutaneous fistula, wound infections, iii) ICU complications: acute respiratory distress syndrome,

Hellenic Journal of Surgery 2010; 82: 6

pneumonia, renal failure, multiple organ failure [5]. Combined injuries correlate with higher complication rates in most studies [10]. Sepsis and multiple organ failure are responsible for 30% of deaths following pancreatic surgery [17]. With an incidence rate ranging from 5%-37%, the pancreatic fistula constitutes the most common pancreatic complication after pancreatic surgery [2]. Most fistulae close spontaneously within 1-2 weeks or more, simply with external drainage, bowel rest and total parenteral nutrition [17]. While some reports show that octreotide can reduce pancreatic fistula output, a recent study postulates that somatostatin and its analogues do not seem to reduce healing time although they can reduce the daily output [52]. High output (>700ml) fistulae are indicative of major pancreatic duct disruption and may require surgical intervention or prolonged periods of drainage with nutritional support [10, 14]. Should the fistula persist in patients with partial duct disruption, transpapillary endoscopic stenting can be tried, while transmural drainage alone or in conjunction with transpapillary drainage is attempted in patients with large pseudocysts and complete pancreatic duct disruption [48]. If this also fails, distal pancreatectomy may be needed for fistulae in the neck, body and tail or a Roux-en-Y pancreaticojejunostomy for fistulae in the head of the pancreas [2, 14, 17]. Pseudocysts are usually the result of a missed pancreatic injury or a frequent complication of distal pancreatectomy. They are more often related to blunt injuries and may present weeks or months after the original pancreatic injury. The non-operative management of solid organ injuries has increased the rate of pseudocyst formation after a missed pancreatic injury [53]. With early diagnosis and intervention, the rate of pseudocysts should not surpass the 2%-3% [10]. The most important parameter in the management of pseudocyst is the integrity of the pancreatic ductal system, which explains why imaging with ERCP or MRCP is essential before any intervention. Percutaneous aspiration is the method of choice if the pancreatic duct is intact, but if the duct is disrupted, percutaneous aspiration will lead to chronic external fistula formation [10, 17]. In the case of a distal pseudocyst with ductal disruption, internal drainage (cystoenterostomy or cystogastrostomy) should be attempted, providing the wall of the cyst is mature, otherwise a distal pancreatectomy is warranted. If the pseudocyst is related to a proximal duct injury, endoscopic transpapillary stenting should be tried. If this fails and the wall of the cyst is mature, internal drainage is the optimal treatment; otherwise resection is indicated [16].

370 Hellenic Journal of Surgery 2010; 82: 6

Intra-abdominal abscess incidence ranges from 10% to 25% after pancreatic trauma [28] And is associated with higher grades of pancreatic injury and injuries to adjacent viscera [16, 28]. Percutaneous C/T guided drainage is often feasible in unilocular collections [14, 16]. Clinical signs of intra-abdominal sepsis warrant broad spectrum antibiotic coverage, pending the culture results from aspirated fluid [14]. Post-traumatic pancreatitis is usually caused by pancreatic fibrosis or a stricture in an injured or a stented duct [2]. Its management is similar to that indicated for any other pancreatitis and includes bowel rest and total parenteral nutrition [2, 16]. Most post-traumatic pancreatitis run a benign course [10]. Post traumatic leakage of pancreatic juice or abscess formation may lead to posttraumatic haemorrhage from erosion of adjacent vessels. In 70-100% of these cases, surgical intervention is mandated [28]. To maintain exocrine and endocrine function, it is essential to leave 10-20% of remnant pancreatic tissue. Pancreatectomies leaving less pancreatic tissue will result in exocrine and endocrine deficiency [10, 28]. Pancreatic trauma rehabilitation-Quality of life There is no report in the literature regarding the functional and socioeconomic outcome of pancreatic trauma in the patients’ quality of life after discharge from the hospital. Retrieved reports referred only to the quality of life after trauma in general. Major gastrointestinal complications, infections, as well as major pulmonary and musculoskeletal complications are associated with a significantly lower quality of life at 6- month follow-up [54]. The most common pancreatic complication, a pancreatic fistula may afford no other discomfort to the everyday life of some patients beyond that of maintaining a drain for several days following surgery, while in others the fistula can last for months and adversely affect their quality of life. Depending on the type of pancreatic fistula, antibiotics may be administered for several days and supplemental nutrition initiated; readmission may prove nesessary for percutaneous drainage and surgery [55]. Rodriguez postulates that nearly 75% of patients with a pancreatic fistula will need at least one hospital readmission, while up to 50% will require more than 4 postoperative C/T scans and nearly two and half months of post injury follow-up [56]. Most post-traumatic complications are identified by the second week after discharge from the hospital. Based on this finding, outpatient follow-up visits are ideally scheduled within 7-14 days. Protocols

The Current Management of Pancreatic Trauma

should be established for the follow-up of trauma patients with special focus on operated patients, patients who have suffered a penetrating injury and those over 65 years of age [57].

Conclusion Pancreatic trauma poses challenging diagnostic and therapeutic dilemmas to the trauma surgeon. Early and accurate diagnosis is of great importance and patients with suspicion of pancreatic injury should be monitored diligently. A delay in diagnosis could have devastating consequences. Once a pancreatic injury is identified during laparotomy, the basic steps include haemostasis, debridement of devitalised tissue and wide drainage. The patient’s haemodynamic status, integrity of the pancreatic duct, site of duct injury and presence of a concomitant duodenal injury will dictate the final management. Several intraoperative radiological methods can evaluate pancreatic duct integrity; intraoperative ERCP is the most accurate of these, but is not readily available and can only be performed on stable patients. The spectrum of pancreatic injuries ranges from simple contusions or lacerations, which are treated conservatively or with external drainage, to major injuries with disruption of the pancreatic duct best treated by distal pancreatectomy. Endoscopic transpapillary stenting, a recent addition to the armamentarium of therapeutic strategies, is indicated for proximal duct injuries. More complex techniques, including pancreatoenteral anastomosis, have also been advocated in an effort to preserve as much pancreatic tissue as possible. These complex techniques are contraindicated for the critical haemodynamically unstable patient and are falling out of favour since they are time consuming and result in high rates of postoperative complications. In combined pancreatoduodenal injuries, each organ can be managed separately according to the grade of injury. Pancreatoduodenectomy is indicated when there is maceration of the head of the pancreas, destruction of the ampulla of Vater or a severe combined pancreatoduodenal injury. In the unstable patient, a damage control procedure should precede. Pancreatic surgery for trauma presents a complication rate of 26%-86%, with the pancreatic fistula being the most common complication. Other pancreatic complications include pseudocyst, pancreatitis and abscess formation. To date, there is no consensus approach regarding the management of pancreatic injuries. Pancreatic trauma is rarely encountered in clinical practice and most studies are retrospective with a limited num-

47

48

371 The Current Management of Pancreatic Trauma

ber of patients. Large prospective trials are needed to enable the development of definitive guidelines for the management of pancreatic trauma. Conflict of interest The authors declare that they have no conflict of interest.

References

1. Jacombs, AS et al., Pancreatic trauma in children. J Pediatr Surg, 2004. 39(1): p. 96-99. 2. Subramanian, A, Dente CJ, and Feliciano DV, The management of pancreatic trauma in the modern era. Surg Clin North Am, 2007. 87(6): p. 1515-1532, x. 3. Lochan, R., et al., Management strategies in isolated pancreatic trauma. J Hepatobiliary Pancreat Surg, 2009. 16(2): p. 189196. 4. Almond LM, Janoo S., Diagnosis and management of pancreatic trauma. Eur Surg, 2009. 41 (5): p. 209-212. 5. Hwang, SY and Choi YC. Prognostic determinants in patients with traumatic pancreatic injuries. J Korean Med Sci, 2008. 23(1): p. 126-130. 6. Bradley EL et al. Diagnosis and initial management of blunt pancreatic trauma: guidelines from a multiinstitutional review. Ann Surg, 1998. 227(6): p. 861-869. 7. Kao LS et al. Predictors of morbidity after traumatic pancreatic injury. J Trauma, 2003. 55(5): p. 898-905. 8. Megibow A., Imaging of the pancreas. Springer, 2009. 9. Spaniolas K, Velmahos G. Nonoperative Management of Pancreato-Duodenal Injuries. Eur J Trauma Emerg Surg 2007. 33: p. 221-226. 10. Boffard KD and Brooks AJ. Pancreatic trauma--injuries to the pancreas and pancreatic duct. Eur J Surg, 2000. 166(1): p. 4-12. 11. Cirillo RL, Koniaris LG. Detecting Blunt Pancreatic Injuries. J Gastrointest Surg 2002. 6: p. 587-598. 12. Teh SH et al. Diagnosis and management of blunt pancreatic ductal injury in the era of high-resolution computed axial tomography. Am J Surg, 2007. 193(5): p. 641-643. 13. Degiannis E., et al. Gunshot injuries of the head of the pancreas: conservative approach. World J Surg, 1996. 20(1): p. 68-71. 14. Krige JE et al. The management of complex pancreatic injuries. S Afr J Surg, 2005. 43(3): p. 92-102. 15. Oniscu GC, Parks RW, Garden OJ. Classification of liver and pancreatic trauma. HPB (Oxford), 2006. 8(1): p. 4-9. 16. Subramanian A, Feliciano D. Pancreatic Trauma Revisited. Eur J Trauma Emerg Surg 2008. 34: p. 3-10. 17. Degiannis E. et al., Management of pancreatic trauma. Injury, 2008. 39(1): p. 21-9. 18. Davis JJ, Cohn I, and Nance FC. Diagnosis and management of blunt abdominal trauma. Ann Surg, 1976. 183(6): p. 672-678. 19. Jurkovich GJ and Carrico CJ. Pancreatic trauma. Surg Clin North Am, 1990. 70(3): p. 575-593. 20. Berni GA et al. Role of intraoperative pancreatography in patients with injury to the pancreas. Am J Surg, 1982. 143(5): p. 602-605. 21. Bhasin DK, Rana S, Rawal P. Endoscopic retrograde pancreatography in pancreatic trauma: Need to break the mental barrier. Journal of Gastroenterology and Hepatology 2009. 24: p. 720-728. 22. Pata G, et al. Extension of Nonoperative Management of Blunt Pancreatic Trauma to Include Grade III Injuries: A Safety Analysis. World J Surg, 2009. 33: p. 1611-1617. 23. Schroeppel TJ, Croce MA. Diagnosis and management of blunt abdominal solid organ injury. Current Opinion in Critical

Hellenic Journal of Surgery 2010; 82: 6

Care 2007 13(4): p. 399-404. 24. Mattix KD, et al. Pediatric pancreatic trauma: predictors of nonoperative management failure and associated outcomes. J Pediatr Surg, 2007. 42(2): p. 340-344. 25. Farrell RJ, et al. Operative strategies in pancreatic trauma. Br J Surg, 1996. 83(7): p. 934-937. 26. Al-Ahmadi K. and Ahmed N. Outcomes after pancreatic trauma: experience at a single institution. Can J Surg, 2008. 51(2): p. 118-124. 27. Lucas CE. Diagnosis and treatment of pancreatic and duodenal injury. Surg Clin North Am, 1977. 57(1): p. 49-65. 28. Wilson RH. and Moorehead RJ. Current management of trauma to the pancreas. Br J Surg, 1991. 78(10): p. 1196-1202. 29. Leppaniemi A. et al. Pancreatic trauma: acute and late manifestations. Br J Surg, 1988. 75(2): p. 165-167. 30. Vane DW, et al. Initial Resection of Potentially Viable Tissue is not Optimal Treatment for Grades II–IV Pancreatic Injuries. World J Surg 2009. 33: p. 221-227. 31. Fahy BN et al. Morbidity, mortality, and technical factors of distal pancreatectomy. Am J Surg, 2002. 183(3): p. 237-241. 32. Feliciano D. Abdominal trauma.Ιn: Schwartz SI, Ellis H, editors. Maingot’s abdominal operations. 9th edition. East Norwalk: Appleton & Lange. 1989. 33. Jurkovich GJ, Bulger EM. Duodenum and pancreas. In: Mattox KL, Moore ME, Feliciano DV, editors. Trauma.New York: McGraw-Hill Companies. 2004. 34. Kim HS, et al. The role of endoscopic retrograde pancreatography in the treatment of traumatic pancreatic duct injury. Gastrointest Endosc, 2001. 54(1): p. 49-55. 35. Akhrass R. et al. Pancreatic trauma: a ten-year multi-institutional experience. Am Surg, 1997. 63(7): p. 598-604. 36. Asensio JA. et al. Management of pancreatic injuries. Curr Probl Surg, 1999. 36(5): p. 325-419. 37. Mayer JM. et al. Pancreatic injury in severe trauma: early diagnosis and therapy improve the outcome. Dig Surg, 2002. 19(4): p. 291-297. 38. Beck DE, et al. Colon and Rectal Trauma and Rectal Foreign Bodies. The ASCRS Manual of Colon and Rectal Surgery 2009, 449-461, DOI: 10.1007/b12857_22,. ed: Springer. 39. Karl HW and Chandler JG. Mortality and morbidity of pancreatic injury. Am J Surg, 1977. 134(5): p. 549-554. 40. Heitsch RC, et al. Delineation of critical factors in the treatment of pancreatic trauma. Surgery, 1976. 80(4): p. 523-529. 41. Fabian TC, et al. Superiority of closed suction drainage for pancreatic trauma. A randomized, prospective study. Ann Surg, 1990. 211(6): p. 724-8; discussion 728-730. 42. Wisner DH., Wold RL, Frey CF. Diagnosis and treatment of pancreatic injuries. An analysis of management principles. Arch Surg, 1990. 125(9): p. 1109-1113. 43. Patton JH and Fabian TC. Complex pancreatic injuries. Surg Clin North Am, 1996. 76(4): p. 783-795. 44. Pachter HL, et al. Traumatic injuries to the pancreas: the role of distal pancreatectomy with splenic preservation. J Trauma, 1989. 29(10): p. 1352-1355. 45. Nowak MM, Baringer DC, Ponsky JL. Pancreatic injuries. Effectiveness of debridement and drainage for nontransecting injuries. Am Surg, 1986. 52(11): p. 599-602. 46. Sheehan MK, et al. Distal pancreatectomy: does the method of closure influence fistula formation? Am Surg, 2002. 68(3): p. 264-7; discussion 267-268. 47. Wind P, et al. Contribution of endoscopic retrograde pancreatography in management of complications following distal pancreatic trauma. Am Surg, 1999. 65(8): p. 777-783. 48. Bhasin DK., Rana SS, Nadkarni N. Protocol-based management strategy for post-endoscopic retrograde cholangiopancreatography pancreatitis: can it make a difference? J Gastroenterol Hepatol, 2008. 23(3): p. 344-347.

372 Hellenic Journal of Surgery 2010; 82: 6

49. Vaughan GD, et al. The use of pyloric exclusion in the management of severe duodenal injuries. Am J Surg, 1977. 134(6): p. 785-790. 50. Seamon MJ, et al. A ten-year retrospective review: does pyloric exclusion improve clinical outcome after penetrating duodenal and combined pancreaticoduodenal injuries? J Trauma, 2007. 62(4): p. 829-833. 51. Berne CJ, et al. Duodenal “diverticulization” for duodenal and pancreatic injury. Am J Surg, 1974. 127(5): p. 503-507. 52. Vasquez JC, et al. Management of penetrating pancreatic trauma: an 11-year experience of a level-1 trauma center. Injury, 2001. 32(10): p. 753-759. 53. Kudsk KA. et al. Post-traumatic pancreatic sequestrum: recognition and treatment. J Trauma, 1986. 26(4): p. 320-324. 54. Holbrook TL, Hoyt DB, and Anderson JP. The impact of major in-hospital complications on functional outcome and quality of life after trauma. J Trauma, 2001. 50(1): p. 91-95. 55. Pratt WB, et al. Clinical and economic validation of the International Study Group of Pancreatic Fistula (ISGPF) classification scheme. Ann Surg, 2007. 245(3): p. 443-451. 56. Rodriguez JR. et al. Implications and cost of pancreatic leak following distal pancreatic resection. Arch Surg, 2006. 141(4): p. 361-365; discussion 366. 57. Battistella FD, Torabian SZ, and Siadatan KM. Hospital readmission after trauma: an analysis of outpatient complications. J Trauma, 1997. 42(6): p. 1012-1016.

The Current Management of Pancreatic Trauma

49

50

373 Hellenic Journal of Surgery 2010; 82: 6

The Current Management of Pancreatic Trauma

H Σύγχρονη Αντιμετώπιση του Παγκρεατικού Τραύματος

Συστηματική Ανασκόπιση Μ. Ι. Κοροντζή, Χ. Κοντοβουνήσιος, Χ.Κ. Καραλιώτας, Σ. Λανίτης, Γ. Σγουράκης, Τ. Παπακωσνταντίνου & Κ. Καραλιώτας

Περίληψη

Σκοπός: Η αξιολόγηση της χρησιμότητας του διεγχειρητικού ακτινολογικού ελέγχου και των σύγχρονων πρακτικών στην αντιμετώπιση του παγκρεατικού τραύματος καθώς και ανασκόπηση των επιπλοκών της συντηρητικής και της χειρουργικής θεραπείας. Οι κακώσεις του παγκρέατος, αν και σπάνιες εμφανίζουν υψηλή νοσηρότητα και θνητότητα, ενώ μέχρι σήμερα δεν έχουν καθιερωθεί αυστηρά κριτήρια για την αντιμετώπισή τους. Μεθοδολογία: Tα δεδομένα αντλήθηκαν κυρίως από τις πρόσφατα δημοσιευμένες μελέτες στην αγγλόφωνη βιβλιογραφία. Δεκαπέντε ανασκοπήσεις και έντεκα κλινικές μελέτες συνελέγησαν προς επεξεργασία. Αποτελέσματα: H αιμοδυναμική κατάσταση του ασθενούς, η ακεραιότητα του παγκρεατικού πόρου, η εντόπιση του τραύματος και η συνύπαρξη ή όχι δωδεκαδακτυλικού τραύματος καθορίζουν την επιλογή της θεραπείας. Διάφορες διεγχειρητικές ακτινολογικές τεχνικές είναι διαθέσιμες για την εκτίμηση της ακεραιότητας του παγκρεατικού πόρου: i)η διεγχειρητική χολαγγειογραφία, ii) η διεγχειρητική παγκρεατογραφία με καθετηριασμό του παγκρεατικού πόρου και iii) η διεγχειρητική ενδοσκοπική ανάστροφη χολαγγειοπαγκρεατογραφία (ERCP) η οποία είναι και η πιο ακριβής για τη εκτίμηση της κατάστασης του παγκρεατικού πόρου. Το φάσμα του παγκρεατικού τραύματος περιλαμβάνει από απλές θλάσεις ή διασχίσεις που αντιμετωπίζονται συντηρητικά ή με εξωτερική παροχέτευση μέχρι σοβαρές βλάβες με καταστροφή του παγκρεατικού πόρου που αντιμετωπίζονται κατά προτίμηση με παγκρεατεκτομή. Σύνθετες παγκρεατοεντερικές αναστομώσεις επίσης αναφέρονται στη βιβλιογραφία αλλά είναι προτιμότερο να αποφεύγονται καθώς σχετίζονται με υψηλά ποσοστά μετεγχειρητικών επιπλοκών, είναι ιδιαίτερα χρονοβόρες και δεν αποτελούν καλή επιλογή κυρίως στην οξεία φάση. Μία σημαντική εξέλιξη στη διάγνωση και θεραπεία του παγκρεατικού τραύματος αποτελεί η ενδοσκοπική

ανάστροφη χολαγγειοπαγκρεατογραφία (ERCP) και η διασφιγκτηριακή τοποθέτηση stent προκειμένου για τους εγγύς τραυματισμούς του μείζονος παγκρεατικού πόρου. Στους σύνθετους τραυματισμούς παγκρέατος και δωδεκαδακτύλου κάθε όργανο αντιμετωπίζεται ουσιαστικά ξεχωριστά ανάλογα με τον βαθμό του τραυματισμού όπως αυτός ορίζεται από την κατάταξη AAST (American Association for the Surgery of Trauma).H παγκρεατοδωδεκαδακτυλεκτομή διενεργείται μόνο σε περιπτώσεις πλήρους καταστροφής της κεφαλής του παγκρέατος και του φύματος του Vater. Στη συστηματική ανασκόπηση της βιβλιογραφίας που διενεργήσαμε, οι συνολικές επιπλοκές του παγκρεατικού τραύματος ήταν 40%. Οι επιπλοκές αυξάνονταν με τον βαθμό του τραυματισμού (grade I και II: 27%,grade III: 56%, grade IV, V: 69%). Συνολικά, το ποσοστό των μετεγχειρητικών επιπλοκών ήταν 46%, ενώ οι επιπλοκές της συντηρητικής θεραπείας ήταν 33%. Η συντηρητική αντιμετώπιση εφαρμόζεται κυρίως στους χαμηλού βαθμού τραυματισμούς (grade I και II), σε αιμοδυναμικά σταθερούς ασθενείς, χωρίς άλλους σύγχρονους τραυματισμούς. Για τους χαμηλού βαθμού τραυματισμούς , το ποσοστό των επιπλοκών της συντηρητικής θεραπείας στην ανασκόπηση μας ανέρχονταν στο 33%, ενώ για τις υψηλόβαθμες βλάβες στο 50%. Συμπέρασμα: Οι αρχές της αντιμετώπισης του παγκρεατικού τραύματος δεν έχουν ακόμη πλήρως καθορισθεί. Οι υπάρχουσες μελέτες είναι αναδρομικές, προερχόμενες από μεμονωμένα κέντρα τραύματος, με περιορισμένο αριθμό ασθενών. Προοπτικές, πολυκεντρικές μελέτες είναι απαραίτητες προκειμένου να θεσπισθούν κατευθυντήριες οδηγίες.

Λέξεις Κλειδιά Πάγκρεας, Τραύμα, Παγκρεατικό τραύμα

‘Β Χειρουργική Κλινική και Μονάδα Χειρουργικής Ογκολογίας, Νοσοκομείο Ερυθρού Σταυρού “Κοργιαλένειο Μπενάκειο” Αθήνα