- Email: [email protected]
Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy
HPB
https://doi.org/10.1016/j.hpb.2019.07.016
ORIGINAL ARTICLE
Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy Brigitta Globke1, Lea Timmermann1, Fritz Klein1, Uli Fehrenbach2, Johann Pratschke1, Marcus Bahra1 & Thomas Malinka1 Department of Surgery, Charité – Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, Germany, and 2Department of Radiology, Charité – Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, Germany 1
Abstract Background: Recent studies have suggested acute pancreatitis as a separate pancreatic-specific complication following pancreaticoduodenectomy. However, data on necrotizing pancreatitis of the pancreatic remnant is limited. This study aimed to evaluate parameters of patients undergoing completion pancreatectomy (CP) after initial pancreaticoduodenectomy (PD) and compare those with or without necrosis of the pancreatic remanent. Methods: Patients who underwent CP following PD between January 2005 and December 2017 were identified from a prospectively collected database. Perioperative parameters were recorded, and patients were divided into those with or without histological evidence of necrosis of the pancreatic remnant. Results: Postoperative acute necrotizing pancreatitis (POANP) was histologically detected in 33 (41%) of 79 patients after CP. Serum CRP levels on POD 2 and the day of revision were significantly higher in the POANP group (p < 0.001 for each). POANP was reflected by higher APACHE II and SOFA scores after PD (P < 0.001 for each). Although patients with POANP had an earlier revision, length of ICU and total hospital stay was prolonged (p < 0.001 for each). POANP was associated with more major complications (Clavien-Dindo 3) and more often necessitated reoperations within 30 days (p < 0.001 for each). Conclusion: Patients requiring CP following PD for POANP have an increased risk of major complications, and longer hospital stay. CRP levels, APACHE II and SOFA score, seem to correlate with the severity and might predict POANP. Universally accepted definitions with a clinically validated grading system of severity for POAP and POANP are needed to facilitate appropriate treatment strategies and enable comparison of future studies. Received 7 May 2019; accepted 27 July 2019
Correspondence Thomas Malinka, Department of Surgery, Charité Campus Mitte and Charité Campus Virchow Klinikum, Charité Universitätsmedizin Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany. E-mail: [email protected]
Introduction Postoperative complications after pancreatoduodenectomy (PD) are mainly related to problems with the pancreatoenteral anastomosis. Common complications include postoperative pancreatic fistula (POPF), postpancreatectomy haemorrhage (PPH), or intra-abdominal abscesses. Mortality in patients with clinically relevant POPF is reported to be as high as 45%.1,2
HPB xxxx, xxx, xxx
Recent reports have suggested that postoperative acute pancreatitis (POAP) may be responsible for the complications mentioned above.3,4 In this regard, manifestations of POAP may range from self-resolving inflammation to fulminant pancreatitis with associated multiorgan failure and pancreatic necrosis.3 Perioperative complications after partial pancreatic resection may require completion pancreatectomy (CP) of the pancreatic remnant as a salvage procedure.5 The impact of postoperative
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
2
acute necrotizing pancreatitis of the pancreatic remnant (POANP) on the patients’ postoperative course and clinical outcome are not well understood. This study aimed to review those patients who had undergone CP following PD and compare the perioperative clinical presentation and outcome of those who did or did not have necrotizing pancreatitis within the pancreatic remnant.
Methods Patients’ inclusion criteria and data collection Retrospective single-centre analysis conducted at the Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum in Berlin, Germany. Data of all patients undergoing pancreatic resections between January 2005 and December 2017 were entered into a prospective database, and an independent ethics committee approved the study design. Patient characteristics were extracted from the medical records and further analysed. Patients were included for further analysis if they underwent elective PD and subsequent salvage CP for complications. Definition of postoperative acute necrotizing pancreatitis (POANP) Necrotizing pancreatitis after pancreatic resection was defined by histological verification of necrosis involving the pancreatic parenchyma or the peripancreatic tissue. In this context, parenchymal necrosis refers to nonviable pancreatic tissue, while peripancreatic necrosis involves the peripancreatic fat. Histological evidence of an acute inflammatory cell infiltrate with oedema and fibrinous exudate, without detection of necrosis within the resected remnant, was defined as postoperative acute interstitial pancreatitis (POAP). All resected pancreatic remnants were individually assessed and selectively subdivided into those with or without necrotizing pancreatitis. The definition of necrotizing pancreatitis was limited to the histological assessment given the time variable delay in the radiological diagnosis of the pancreatic necrosis.6 Peri-operative course A multimodal therapeutic approach was individually formulated for every oncologic patient by a multidisciplinary tumour board, which consisted of radiologists, medical oncologists, specialised radiation therapists, and surgeons. Experienced surgeons performed all included operations at the study site, and all procedures were performed by an open surgical approach according to international standards at that time. Somatostatin analogues were not prescribed. A pancreaticogastrostomy or pancreaticojejunostomy was performed based on the surgeon’s preference. For both techniques, absorbable monofilament sutures were used. Non-suction silicon drains were placed as pancreatic duct stents to prevent postoperative, oedema associated occlusion of the pancreatic duct. Pancreatectomy for suspected carcinoma included a radical lymphadenectomy
HPB xxxx, xxx, xxx
according to international standards.7 Postoperatively, daily serum and drain lipase levels were performed. In the absence of signs of POPF, oral food intake was started depending on the clinical presentation and tolerance. Indications for CP after PD were leakage of the pancreaticoenteric anastomosis, bleeding at the anastomosis, sepsis, or organ failure. The Clavien-Dindo (CD) classification was used to grade postoperative complications.8 Major complications were defined as CD 3. Complications and death within 90 days determined postoperative morbidity and mortality. All resected specimens were histologically examined to identify the underlying entity and to verify necrotizing pancreatitis of the pancreatic remnant. Data collection Preoperative data such as patients demographics, characteristics and the use of neoadjuvant chemotherapy were recorded. Perioperatively, the procedure and operative details including blood loss, need for transfusion, associated resection and pancreatic texture, observed during initial PD, were recorded. Postoperatively, routine laboratory testing included leucocyte count, C-reactive protein (CRP), serum lipase levels, and lipase levels in drains at various time points. Furthermore, the Acute Physiology and Chronic Health Evaluation (APACHE II) and the Sequential Organ Failure Assessment (SOFA) score, complications, length of intensive care and total hospital stay and reoperation and readmission within 30 days were recorded. An independent radiologist reviewed all available CT scans after initial PD and stated the CT severity index (CTSI). Statistics Data are presented as median (range) or numbers unless indicated otherwise. Nonparametric Mann-Whitney-Test or Wilcoxon test compared continuous parameters between groups of patients as appropriate. Chi-square tests or Fisher’s exact tests compared categorical data as appropriate. Significance tests were two-sided, and p < 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS version 25.0 (SPSS, Chicago, IL, USA).
Results Patients characteristics Between January 2005 and December 2017, 1750 patients underwent pancreatic resections at the Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum in Berlin, Germany. A total of 79 (7%) patients required CP after initial PD (n = 1172). Table 1 presents patient and baseline characteristics by the presence or absence of necrosis within the pancreatic remnant. POANP was histologically detected in 33 (41%) patients after CP. Isolated parenchymal or peripancreatic necrosis occurred in 4 and 2 patients, respectively. Combined parenchymal and peripancreatic necrosis occurred in 27/33 (82%) of patients. The remaining 46 (58%) patients had POAP without histological
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
3
Table 1 Patients and baseline characteristics
Table 2 Perioperative Characteristics for Patients divided into those
POANP after PD (n [ 33)
No POANP after PD (n [ 46)
p-value
69 (39–82)
65 (38–84)
0.108
Male
22
37
0.196
Female
11
9
25 (18–35)
24 (16–41)
Pancreatic adenocarcinoma
24
36
Periampullary carcinomaa
3
Characteristics
Age (years)
necrotizing pancreatitis of the pancreatic remnant (POANP) after PD
Gender
BMI (kg/m2)
Characteristics
POANP after PD No POANP after p-value (n [ 33) PD (n [ 46)
Texture of pancreatic gland (assessed during PD)
0.723
Histology/Indication for PD
4
IPMN
0.412
Malignant
1
–
Benign
–
1
Chronic pancreatitis
5
5
Endocrine insufficiency before PD
4
7
0.697
Pancreatitis history before PD
7
7
0.494
None
27
38
0.928
Neoadjuvant
6
8
Soft
13
15
Hard
20
31
0.635
Portal vein/SMV resection during PD
8
8
0.458
Multivisceral resection during PD
7
9
0.858
Median CTSI after PD
4,5 (2–10)
4 (2–10)
0.980
Hemodynamic instability
4
7
0.634
Uncontrolled leakage/sepsis
29
39
Indication for CP
Chemotherapy before PD
ASA score before PD I
–
–
II
13
18
III
17
26
IV
3
2
Preoperative Albumin level before PD (g/l)
39 (19–45)
41 (14–50)
0.490
Preoperative Bilirubin level before PD (mg/dl)
0.8 (0.1–13.1)
0.6 (0.1–13.3)
0.723
Preoperative ERCP before PD
12
15
0.894
Preoperative Stenting before PD
8
11
0.908
0.680
POANP, postoperative acute necrotizing pancreatitis; PD, pancreaticoduodenectomy; BMI, body mass index; IPMN, intraductal papillary mucinous neoplasia; ASA, American Society of Anaesthesiologists; ERCP, Endoscopic retrograde cholangiopancreatography. a Pancreatic head cancer excluded.
evidence of necrosis within the resected remnant. A total of 70 (89%) patients, 29/33 CP with POANP and 41/46 CP without POANP underwent a computerized tomography (CT) in the time period between PD and CP. The distribution of perioperative characteristics divided by histological evidence of POANP is shown in Table 2.
HPB xxxx, xxx, xxx
with or without histological evidence of postoperative acute
Operation time of CP 244 (170–330) (min)*
202 (90–290)
<0.001
Intraoperative blood 1300 (800–2500) 900 (500–2000) loss during CP (ml) *
<0.001
7 (2–14) Intraoperative transfusion* during CP
4 (2–10)
<0.001
Splenectomy during CP
30
40
0.588
Delayed gastric emptying after CP
12
18
0.804
Bleeding complication after CP
3
3
0.673
Abdominal collections after CP
6
4
0.214
Major complications 12 Clavien-Dindo 3 after CP
8
0.037
Reoperation within 30 days after CP
11
9
0.044
90-day mortality after CP
12
15
0.730
CP, completion pancreatectomy; POANP, postoperative acute necrotizing pancreatitis; PD, pancreaticoduodenectomy; CTSI, CT severity index; SMV, superior mesenteric vein.
Fig. 1 demonstrates the distribution of time to revision, length of ICU and total hospital stay by subgroup. Patients in the CP cohort with POANP revealed significant differences with regard to median (range) time to revision [POD 8 (4–15) vs POD 11 (3–21), P < 0.014], length of intensive care stay [6 (3–12 days) vs
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
4
Figure 1 Timepoint of operative revision (A) intensive care stay (B) and overall hospital stay (C) for patients with CP, subdivided by POANP
4 (2–11 days), P < 0.002], and indicated prolonged total hospital stay [31 (16–50 days) vs 24 (14–40 days), P < 0.001]. Fig. 2 provides data for the analysis of further perioperative findings. There were no differences in the CP cohort concerning leucocyte count, serum lipase levels, and lipase levels in the drains by time point and the presence or absence of POANP (Fig. 2). However, patients who underwent CP for POANP had significantly higher serum median (IQR) CRP level on POD 2 after initial PD [190 (130–241) vs 130 (88–164), P < 0.001] and on the day of revision [226 (186–260) vs 139 (107–185), P < 0.001]. Patients with CP and POANP had significantly higher median (IQR) APACHE II and SOFA scores on POD 4 after PD [9 (4–15) vs 5 (5–12), P < 0.001] and [4 (2.5–8) vs 2 (2–4), P < 0.001], the day before CP [10 (7–14) vs 4 (3–8), P < 0.001] and [4 (1–7) vs 2 (1–3), P < 0.001], day of CP [11 (5–19) vs 6 (3–10), P < 0.001] and [4 (I2–6) vs 2 (1–4), P < 0.001], and day after CP [12 (7–14) vs 7 (4–9), P < 0.001] and [5 (4–6) vs 2 (1–3), P < 0.001], respectively.
Discussion The purpose of the present study was to compare the perioperative outcomes of patients who had to undergo CP after PD as a salvage procedure in order to control postoperative complications at a high-volume medical centre. The results of the current study underline that POANP is associated with an increased SIRS response and that patients with POANP often do not show adequate recovery after the initial operation. POANP is associated with an increased rate of perioperative additional severe complications, and prolonged total hospital stay. Furthermore, CRP levels, APACHE II and SOFA scores seem to correlate with the severity and might predict POANP. Improvements in perioperative management, combined with an ongoing centralisation of patients to high volume centres, have reduced mortality associated with PD.9 However, PD remains associated with anastomosis-related complications, and recent studies report postoperative morbidity rates ranging from
HPB xxxx, xxx, xxx
46 to 54%.10,11 POPF and PPH are challenging complications after pancreatic surgery, which may lead to reintervention, relaparotomy and delayed recovery or even death.1,12 Contemporary rates of clinically relevant POPF (grades B/C) are still reported between 8 and 17%.13 The aetiology of POPF is likely multifactorial, including changes in blood supply, tissue ischemia, infection, anatomical factors, and surgical technique.14,15 In 2016, Connor proposed that a proportion of POPF was due to POAP and suggested POAP as a new pancreatic specific complication following pancreatic resection.3 In contrast, Kriger et al. consider POAP mainly as an indirect sign of POPF.4 The definition of Connor depends on a biochemical elevation in serum amylase levels above the upper limit of normal. However, an evaluation of pancreatic necrosis of the remnant was missing. So far, the revised Atlanta classification for acute pancreatitis has been suggested as a way to define POAP.16,17 In this context, the definition of acute pancreatitis includes abdominal pain and serum amylase or lipase levels of more than three times the upper limit of normal. Accordingly, the system subdivides the clinical course of acute pancreatitis into early (<1 week) and late (>1 week) phases. Pancreatic necrosis is further divided into three morphologic subtypes, depending on whether it involves pancreatic parenchyma only, peripancreatic tissues, or both pancreatic parenchyma and peripancreatic tissues.18 Although biochemical and radiological criteria exist, the accurate prediction of the severity of pancreatitis remains challenging. How these apply in the postoperative setting remains unclear. In the current analysis, 79 patients underwent CP after initial PD as a salvage procedure. Based on the histological assessment of the pancreatic remnant, further evaluation was subdivided by the presence or absence of POANP. Importantly all patients who underwent CP demonstrated evidence of POAP with 33 demonstrating evidence of POANP. Given that all patients in the current analysis met the criteria for POPF, and it would seem that reclassification of POPF is required. Patients with POANP had significantly higher CRP levels on POD 2 and the day of CP. Although Alfonso et al. proposed that higher CRP levels may
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
5
Figure 2 Median leucocyte count (A) median serum CRP levels (B) median serum lipase levels (C) median lipase levels in drain (D) median
APACHE II score (E) and median SOFA score (F) for patients with CP, subdivided by POANP
predict the probability of the presence of pancreatic necrosis, more recent reviews suggest further well-designed studies are required to validate the predictive significance of CRP.19,20
HPB xxxx, xxx, xxx
Furthermore, APACHE II and SOFA score were significantly elevated in the postoperative course after PD in patients with POANP and may indicate more severe POAP. An essential
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
6
finding of the current study demonstrated that CP was performed significantly earlier in patients with POANP and that ICU-stay and total hospital stay were significantly prolonged. The findings of the study help demonstrate that POANP is associated with a systemic, exaggerated inflammatory response syndrome, associated with systemic organ dysfunction, immunosuppression, and organ failure, and more severe course of the disease. Patients with POANP do not show adequate recovery after the initial operation and remain systemically unwell in the early postoperative period. Even after early CP, disease prolonged recovery is required. Risk factors for POAP and POANP remain ill-understood. Bannone et al. investigated risk factors for POAP and indicated additional resection of the pancreatic stump margin, preoperative exocrine insufficiency, neoadjuvant therapy, soft pancreatic texture, and small main pancreatic duct diameter as independent predictors.21 Normal pancreatic tissue appears to be extremely susceptible to ischemia, and transient hypoperfusion can consequently lead to pancreatic necrosis.21 However, in this analysis, we did not observe any differences concerning the pancreatic texture that might indicate POAP or POANP. Overall, further prospective studies will be required to evaluate predictive risk factors of POAP or POANP. The present study is limited by common biases associated with retrospective studies. Precision and completeness of data acquisition are difficult to control, especially over such an extended study period. Moreover, patients were not stratified according to different surgical techniques to avoid low statistical power. Unfortunately, information concerning the main pancreatic duct diameter is mostly missing. Therefore, the Fistula Risk Score (FRS) could not be calculated. In conclusion, the current study demonstrates that POAP has a ubiquitous histological presence in patients undergoing CP following PD with 41% demonstrating evidence of POANP. Patients requiring with CP for POANP demonstrated a greater SIRS response in the postoperative period and required a longer recovery associated with more complications. POAP and POANP would appear to be distinct postpancreatectomy-specific complications that require further classification. In this regard, a universally accepted definition associated with a clinically validated grading system of severity is needed to define appropriate management and classification criteria as part of further prospective studies.
pancreaticoduodenectomy: clinical predictors and patient outcomes. HPB Surg 2009:404520. 2. Eckardt AJ, Klein F, Adler A, Veltzke-Schlieker W, Warnick P, Bahra M et al. (2011) Management and outcomes of haemorrhage after pancreatogastrostomy versus pancreatojejunostomy. Br J Surg 98:1599–1607. 3. Connor S. (2016) Defining post-operative pancreatitis as a new pancreatic specific complication following pancreatic resection. HPB 18:642–651. 4. Kriger
AG,
Kubishkin
VA,
Karmazanovskii
GG,
Svitina
KA,
Kochatkov AV, Berelavichus SV et al. (2012) The postoperative pancreatitis after the pancreatic surgery. Khirurgiia (Mosk), 14–19. 5. Gueroult S, Parc Y, Duron F, Paye F, Parc R. (2004) Completion pancreatectomy for postoperative peritonitis after pancreaticoduodenectomy: early and late outcome. Arch Surg 139:16–19. 6. Bollen TL, Singh VK, Maurer R, Repas K, van Es HW, Banks PA et al. (2012) A comparative evaluation of radiologic and clinical scoring systems in the early prediction of severity in acute pancreatitis. Am J Gastroenterol 107:612–619. 7. Jones L, Russell C, Mosca F, Boggi U, Sutton R, Slavin J et al. (1999) Standard kausch-whipple pancreatoduodenectomy. Dig Surg 16: 297–304. 8. Clavien PA, Barkun J, de Oliveira ML, Vauthey JN, Dindo D, Schulick RD et al. (2009) The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 250:187–196. 9. Balzano G, Zerbi A, Capretti G, Rocchetti S, Capitanio V, Di Carlo V. (2008) Effect of hospital volume on outcome of pancreaticoduodenectomy in Italy. Br J Surg 95:357–362. 10. Addeo P, Delpero JR, Paye F, Oussoultzoglou E, Fuchshuber PR, Sauvanet A et al. (2014) Pancreatic fistula after a pancreaticoduodenectomy for ductal adenocarcinoma and its association with morbidity: a multicentre study of the French Surgical Association. HPB 16:46–55. 11. Hata T, Motoi F, Ishida M, Naitoh T, Katayose Y, Egawa S et al. (2016) Effect of hospital volume on surgical outcomes after pancreaticoduodenectomy: a systematic review and meta-analysis. Ann Surg 263: 664–672. 12. Ahmad SA, Edwards MJ, Sutton JM, Grewal SS, Hanseman DJ, Maithel SK et al. (2012) Factors influencing readmission after pancreaticoduodenectomy: a multi-institutional study of 1302 patients. Ann Surg 256:529–537. 13. Denbo JW, Orr WS, Zarzaur BL, Behrman SW. (2012) Toward defining grade C pancreatic fistula following pancreaticoduodenectomy: incidence, risk factors, management and outcome. HPB 14:589–593. 14. McMillan MT, Vollmer CM, Jr., Asbun HJ, Ball CG, Bassi C, Beane JD et al. (2016) The characterization and prediction of ISGPF grade C fistulas following pancreatoduodenectomy. J Gastrointest Surg 20:262–276. 15. Ryska M, Rudis J. (2014) Pancreatic fistula and postoperative pancreatitis after pancreatoduodenectomy for pancreatic cancer. Hepatobiliary
Funding
Surg Nutr 3:268–275.
None.
16. Kuhlbrey CM, Samiei N, Sick O, Makowiec F, Hopt UT, Wittel UA. (2017) Pancreatitis after pancreatoduodenectomy predicts clinically relevant
Conflicts of interest All authors declare no commercial associations that might represent conflicts
Postoperative serum amylase predicts pancreatic fistula formation following pancreaticoduodenectomy. J Gastrointest Surg 18:348–353.
References 1. Schmidt CM, Choi J, Powell ES, Yiannoutsos CT, Zyromski NJ, Nakeeb
postoperative pancreatic fistula. J Gastrointest Surg 21:330–338. 17. Cloyd JM, Kastenberg ZJ, Visser BC, Poultsides GA, Norton JA. (2014)
of interest with this article and have nothing to disclose.
A
HPB xxxx, xxx, xxx
et
al.
(2009)
Pancreatic
fistula
following
18. Banks PA, Bollen TL, Dervenis C, Gooszen HG, Johnson CD, Sarr MG et al. (2013) Classification of acute pancreatitis–2012: revision of the
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
7
Atlanta classification and definitions by international consensus. Gut 62:
diagnosis of pancreatic necrosis. Cochrane Database Syst Rev 4:
102–111.
CD012645.
19. Alfonso V, Gomez F, Lopez A, Moreno-Osset E, del Valle R, Anton MD
21. Bannone E, Andrianello S, Marchegiani G, Masini G, Malleo G, Bassi C
et al. (2003) Value of C-reactive protein level in the detection of necrosis
et al. (2018) Postoperative acute pancreatitis following pancreatico-
in acute pancreatitis. Gastroenterol Hepatol 26:288–293.
duodenectomy: a determinant of fistula potentially driven by the intra-
20. Komolafe O, Pereira SP, Davidson BR, Gurusamy KS. (2017) Serum C-
operative fluid management. Ann Surg 268:815–822.
reactive protein, procalcitonin, and lactate dehydrogenase for the
HPB xxxx, xxx, xxx
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
https://doi.org/10.1016/j.hpb.2019.07.016
ORIGINAL ARTICLE
Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy Brigitta Globke1, Lea Timmermann1, Fritz Klein1, Uli Fehrenbach2, Johann Pratschke1, Marcus Bahra1 & Thomas Malinka1 Department of Surgery, Charité – Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, Germany, and 2Department of Radiology, Charité – Universitätsmedizin Berlin, Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin Institute of Health, Germany 1
Abstract Background: Recent studies have suggested acute pancreatitis as a separate pancreatic-specific complication following pancreaticoduodenectomy. However, data on necrotizing pancreatitis of the pancreatic remnant is limited. This study aimed to evaluate parameters of patients undergoing completion pancreatectomy (CP) after initial pancreaticoduodenectomy (PD) and compare those with or without necrosis of the pancreatic remanent. Methods: Patients who underwent CP following PD between January 2005 and December 2017 were identified from a prospectively collected database. Perioperative parameters were recorded, and patients were divided into those with or without histological evidence of necrosis of the pancreatic remnant. Results: Postoperative acute necrotizing pancreatitis (POANP) was histologically detected in 33 (41%) of 79 patients after CP. Serum CRP levels on POD 2 and the day of revision were significantly higher in the POANP group (p < 0.001 for each). POANP was reflected by higher APACHE II and SOFA scores after PD (P < 0.001 for each). Although patients with POANP had an earlier revision, length of ICU and total hospital stay was prolonged (p < 0.001 for each). POANP was associated with more major complications (Clavien-Dindo 3) and more often necessitated reoperations within 30 days (p < 0.001 for each). Conclusion: Patients requiring CP following PD for POANP have an increased risk of major complications, and longer hospital stay. CRP levels, APACHE II and SOFA score, seem to correlate with the severity and might predict POANP. Universally accepted definitions with a clinically validated grading system of severity for POAP and POANP are needed to facilitate appropriate treatment strategies and enable comparison of future studies. Received 7 May 2019; accepted 27 July 2019
Correspondence Thomas Malinka, Department of Surgery, Charité Campus Mitte and Charité Campus Virchow Klinikum, Charité Universitätsmedizin Berlin, Campus Virchow-Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany. E-mail: [email protected]
Introduction Postoperative complications after pancreatoduodenectomy (PD) are mainly related to problems with the pancreatoenteral anastomosis. Common complications include postoperative pancreatic fistula (POPF), postpancreatectomy haemorrhage (PPH), or intra-abdominal abscesses. Mortality in patients with clinically relevant POPF is reported to be as high as 45%.1,2
HPB xxxx, xxx, xxx
Recent reports have suggested that postoperative acute pancreatitis (POAP) may be responsible for the complications mentioned above.3,4 In this regard, manifestations of POAP may range from self-resolving inflammation to fulminant pancreatitis with associated multiorgan failure and pancreatic necrosis.3 Perioperative complications after partial pancreatic resection may require completion pancreatectomy (CP) of the pancreatic remnant as a salvage procedure.5 The impact of postoperative
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
2
acute necrotizing pancreatitis of the pancreatic remnant (POANP) on the patients’ postoperative course and clinical outcome are not well understood. This study aimed to review those patients who had undergone CP following PD and compare the perioperative clinical presentation and outcome of those who did or did not have necrotizing pancreatitis within the pancreatic remnant.
Methods Patients’ inclusion criteria and data collection Retrospective single-centre analysis conducted at the Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum in Berlin, Germany. Data of all patients undergoing pancreatic resections between January 2005 and December 2017 were entered into a prospective database, and an independent ethics committee approved the study design. Patient characteristics were extracted from the medical records and further analysed. Patients were included for further analysis if they underwent elective PD and subsequent salvage CP for complications. Definition of postoperative acute necrotizing pancreatitis (POANP) Necrotizing pancreatitis after pancreatic resection was defined by histological verification of necrosis involving the pancreatic parenchyma or the peripancreatic tissue. In this context, parenchymal necrosis refers to nonviable pancreatic tissue, while peripancreatic necrosis involves the peripancreatic fat. Histological evidence of an acute inflammatory cell infiltrate with oedema and fibrinous exudate, without detection of necrosis within the resected remnant, was defined as postoperative acute interstitial pancreatitis (POAP). All resected pancreatic remnants were individually assessed and selectively subdivided into those with or without necrotizing pancreatitis. The definition of necrotizing pancreatitis was limited to the histological assessment given the time variable delay in the radiological diagnosis of the pancreatic necrosis.6 Peri-operative course A multimodal therapeutic approach was individually formulated for every oncologic patient by a multidisciplinary tumour board, which consisted of radiologists, medical oncologists, specialised radiation therapists, and surgeons. Experienced surgeons performed all included operations at the study site, and all procedures were performed by an open surgical approach according to international standards at that time. Somatostatin analogues were not prescribed. A pancreaticogastrostomy or pancreaticojejunostomy was performed based on the surgeon’s preference. For both techniques, absorbable monofilament sutures were used. Non-suction silicon drains were placed as pancreatic duct stents to prevent postoperative, oedema associated occlusion of the pancreatic duct. Pancreatectomy for suspected carcinoma included a radical lymphadenectomy
HPB xxxx, xxx, xxx
according to international standards.7 Postoperatively, daily serum and drain lipase levels were performed. In the absence of signs of POPF, oral food intake was started depending on the clinical presentation and tolerance. Indications for CP after PD were leakage of the pancreaticoenteric anastomosis, bleeding at the anastomosis, sepsis, or organ failure. The Clavien-Dindo (CD) classification was used to grade postoperative complications.8 Major complications were defined as CD 3. Complications and death within 90 days determined postoperative morbidity and mortality. All resected specimens were histologically examined to identify the underlying entity and to verify necrotizing pancreatitis of the pancreatic remnant. Data collection Preoperative data such as patients demographics, characteristics and the use of neoadjuvant chemotherapy were recorded. Perioperatively, the procedure and operative details including blood loss, need for transfusion, associated resection and pancreatic texture, observed during initial PD, were recorded. Postoperatively, routine laboratory testing included leucocyte count, C-reactive protein (CRP), serum lipase levels, and lipase levels in drains at various time points. Furthermore, the Acute Physiology and Chronic Health Evaluation (APACHE II) and the Sequential Organ Failure Assessment (SOFA) score, complications, length of intensive care and total hospital stay and reoperation and readmission within 30 days were recorded. An independent radiologist reviewed all available CT scans after initial PD and stated the CT severity index (CTSI). Statistics Data are presented as median (range) or numbers unless indicated otherwise. Nonparametric Mann-Whitney-Test or Wilcoxon test compared continuous parameters between groups of patients as appropriate. Chi-square tests or Fisher’s exact tests compared categorical data as appropriate. Significance tests were two-sided, and p < 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS version 25.0 (SPSS, Chicago, IL, USA).
Results Patients characteristics Between January 2005 and December 2017, 1750 patients underwent pancreatic resections at the Charité – Universitätsmedizin Berlin, Campus Virchow-Klinikum in Berlin, Germany. A total of 79 (7%) patients required CP after initial PD (n = 1172). Table 1 presents patient and baseline characteristics by the presence or absence of necrosis within the pancreatic remnant. POANP was histologically detected in 33 (41%) patients after CP. Isolated parenchymal or peripancreatic necrosis occurred in 4 and 2 patients, respectively. Combined parenchymal and peripancreatic necrosis occurred in 27/33 (82%) of patients. The remaining 46 (58%) patients had POAP without histological
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
3
Table 1 Patients and baseline characteristics
Table 2 Perioperative Characteristics for Patients divided into those
POANP after PD (n [ 33)
No POANP after PD (n [ 46)
p-value
69 (39–82)
65 (38–84)
0.108
Male
22
37
0.196
Female
11
9
25 (18–35)
24 (16–41)
Pancreatic adenocarcinoma
24
36
Periampullary carcinomaa
3
Characteristics
Age (years)
necrotizing pancreatitis of the pancreatic remnant (POANP) after PD
Gender
BMI (kg/m2)
Characteristics
POANP after PD No POANP after p-value (n [ 33) PD (n [ 46)
Texture of pancreatic gland (assessed during PD)
0.723
Histology/Indication for PD
4
IPMN
0.412
Malignant
1
–
Benign
–
1
Chronic pancreatitis
5
5
Endocrine insufficiency before PD
4
7
0.697
Pancreatitis history before PD
7
7
0.494
None
27
38
0.928
Neoadjuvant
6
8
Soft
13
15
Hard
20
31
0.635
Portal vein/SMV resection during PD
8
8
0.458
Multivisceral resection during PD
7
9
0.858
Median CTSI after PD
4,5 (2–10)
4 (2–10)
0.980
Hemodynamic instability
4
7
0.634
Uncontrolled leakage/sepsis
29
39
Indication for CP
Chemotherapy before PD
ASA score before PD I
–
–
II
13
18
III
17
26
IV
3
2
Preoperative Albumin level before PD (g/l)
39 (19–45)
41 (14–50)
0.490
Preoperative Bilirubin level before PD (mg/dl)
0.8 (0.1–13.1)
0.6 (0.1–13.3)
0.723
Preoperative ERCP before PD
12
15
0.894
Preoperative Stenting before PD
8
11
0.908
0.680
POANP, postoperative acute necrotizing pancreatitis; PD, pancreaticoduodenectomy; BMI, body mass index; IPMN, intraductal papillary mucinous neoplasia; ASA, American Society of Anaesthesiologists; ERCP, Endoscopic retrograde cholangiopancreatography. a Pancreatic head cancer excluded.
evidence of necrosis within the resected remnant. A total of 70 (89%) patients, 29/33 CP with POANP and 41/46 CP without POANP underwent a computerized tomography (CT) in the time period between PD and CP. The distribution of perioperative characteristics divided by histological evidence of POANP is shown in Table 2.
HPB xxxx, xxx, xxx
with or without histological evidence of postoperative acute
Operation time of CP 244 (170–330) (min)*
202 (90–290)
<0.001
Intraoperative blood 1300 (800–2500) 900 (500–2000) loss during CP (ml) *
<0.001
7 (2–14) Intraoperative transfusion* during CP
4 (2–10)
<0.001
Splenectomy during CP
30
40
0.588
Delayed gastric emptying after CP
12
18
0.804
Bleeding complication after CP
3
3
0.673
Abdominal collections after CP
6
4
0.214
Major complications 12 Clavien-Dindo 3 after CP
8
0.037
Reoperation within 30 days after CP
11
9
0.044
90-day mortality after CP
12
15
0.730
CP, completion pancreatectomy; POANP, postoperative acute necrotizing pancreatitis; PD, pancreaticoduodenectomy; CTSI, CT severity index; SMV, superior mesenteric vein.
Fig. 1 demonstrates the distribution of time to revision, length of ICU and total hospital stay by subgroup. Patients in the CP cohort with POANP revealed significant differences with regard to median (range) time to revision [POD 8 (4–15) vs POD 11 (3–21), P < 0.014], length of intensive care stay [6 (3–12 days) vs
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
4
Figure 1 Timepoint of operative revision (A) intensive care stay (B) and overall hospital stay (C) for patients with CP, subdivided by POANP
4 (2–11 days), P < 0.002], and indicated prolonged total hospital stay [31 (16–50 days) vs 24 (14–40 days), P < 0.001]. Fig. 2 provides data for the analysis of further perioperative findings. There were no differences in the CP cohort concerning leucocyte count, serum lipase levels, and lipase levels in the drains by time point and the presence or absence of POANP (Fig. 2). However, patients who underwent CP for POANP had significantly higher serum median (IQR) CRP level on POD 2 after initial PD [190 (130–241) vs 130 (88–164), P < 0.001] and on the day of revision [226 (186–260) vs 139 (107–185), P < 0.001]. Patients with CP and POANP had significantly higher median (IQR) APACHE II and SOFA scores on POD 4 after PD [9 (4–15) vs 5 (5–12), P < 0.001] and [4 (2.5–8) vs 2 (2–4), P < 0.001], the day before CP [10 (7–14) vs 4 (3–8), P < 0.001] and [4 (1–7) vs 2 (1–3), P < 0.001], day of CP [11 (5–19) vs 6 (3–10), P < 0.001] and [4 (I2–6) vs 2 (1–4), P < 0.001], and day after CP [12 (7–14) vs 7 (4–9), P < 0.001] and [5 (4–6) vs 2 (1–3), P < 0.001], respectively.
Discussion The purpose of the present study was to compare the perioperative outcomes of patients who had to undergo CP after PD as a salvage procedure in order to control postoperative complications at a high-volume medical centre. The results of the current study underline that POANP is associated with an increased SIRS response and that patients with POANP often do not show adequate recovery after the initial operation. POANP is associated with an increased rate of perioperative additional severe complications, and prolonged total hospital stay. Furthermore, CRP levels, APACHE II and SOFA scores seem to correlate with the severity and might predict POANP. Improvements in perioperative management, combined with an ongoing centralisation of patients to high volume centres, have reduced mortality associated with PD.9 However, PD remains associated with anastomosis-related complications, and recent studies report postoperative morbidity rates ranging from
HPB xxxx, xxx, xxx
46 to 54%.10,11 POPF and PPH are challenging complications after pancreatic surgery, which may lead to reintervention, relaparotomy and delayed recovery or even death.1,12 Contemporary rates of clinically relevant POPF (grades B/C) are still reported between 8 and 17%.13 The aetiology of POPF is likely multifactorial, including changes in blood supply, tissue ischemia, infection, anatomical factors, and surgical technique.14,15 In 2016, Connor proposed that a proportion of POPF was due to POAP and suggested POAP as a new pancreatic specific complication following pancreatic resection.3 In contrast, Kriger et al. consider POAP mainly as an indirect sign of POPF.4 The definition of Connor depends on a biochemical elevation in serum amylase levels above the upper limit of normal. However, an evaluation of pancreatic necrosis of the remnant was missing. So far, the revised Atlanta classification for acute pancreatitis has been suggested as a way to define POAP.16,17 In this context, the definition of acute pancreatitis includes abdominal pain and serum amylase or lipase levels of more than three times the upper limit of normal. Accordingly, the system subdivides the clinical course of acute pancreatitis into early (<1 week) and late (>1 week) phases. Pancreatic necrosis is further divided into three morphologic subtypes, depending on whether it involves pancreatic parenchyma only, peripancreatic tissues, or both pancreatic parenchyma and peripancreatic tissues.18 Although biochemical and radiological criteria exist, the accurate prediction of the severity of pancreatitis remains challenging. How these apply in the postoperative setting remains unclear. In the current analysis, 79 patients underwent CP after initial PD as a salvage procedure. Based on the histological assessment of the pancreatic remnant, further evaluation was subdivided by the presence or absence of POANP. Importantly all patients who underwent CP demonstrated evidence of POAP with 33 demonstrating evidence of POANP. Given that all patients in the current analysis met the criteria for POPF, and it would seem that reclassification of POPF is required. Patients with POANP had significantly higher CRP levels on POD 2 and the day of CP. Although Alfonso et al. proposed that higher CRP levels may
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
5
Figure 2 Median leucocyte count (A) median serum CRP levels (B) median serum lipase levels (C) median lipase levels in drain (D) median
APACHE II score (E) and median SOFA score (F) for patients with CP, subdivided by POANP
predict the probability of the presence of pancreatic necrosis, more recent reviews suggest further well-designed studies are required to validate the predictive significance of CRP.19,20
HPB xxxx, xxx, xxx
Furthermore, APACHE II and SOFA score were significantly elevated in the postoperative course after PD in patients with POANP and may indicate more severe POAP. An essential
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
6
finding of the current study demonstrated that CP was performed significantly earlier in patients with POANP and that ICU-stay and total hospital stay were significantly prolonged. The findings of the study help demonstrate that POANP is associated with a systemic, exaggerated inflammatory response syndrome, associated with systemic organ dysfunction, immunosuppression, and organ failure, and more severe course of the disease. Patients with POANP do not show adequate recovery after the initial operation and remain systemically unwell in the early postoperative period. Even after early CP, disease prolonged recovery is required. Risk factors for POAP and POANP remain ill-understood. Bannone et al. investigated risk factors for POAP and indicated additional resection of the pancreatic stump margin, preoperative exocrine insufficiency, neoadjuvant therapy, soft pancreatic texture, and small main pancreatic duct diameter as independent predictors.21 Normal pancreatic tissue appears to be extremely susceptible to ischemia, and transient hypoperfusion can consequently lead to pancreatic necrosis.21 However, in this analysis, we did not observe any differences concerning the pancreatic texture that might indicate POAP or POANP. Overall, further prospective studies will be required to evaluate predictive risk factors of POAP or POANP. The present study is limited by common biases associated with retrospective studies. Precision and completeness of data acquisition are difficult to control, especially over such an extended study period. Moreover, patients were not stratified according to different surgical techniques to avoid low statistical power. Unfortunately, information concerning the main pancreatic duct diameter is mostly missing. Therefore, the Fistula Risk Score (FRS) could not be calculated. In conclusion, the current study demonstrates that POAP has a ubiquitous histological presence in patients undergoing CP following PD with 41% demonstrating evidence of POANP. Patients requiring with CP for POANP demonstrated a greater SIRS response in the postoperative period and required a longer recovery associated with more complications. POAP and POANP would appear to be distinct postpancreatectomy-specific complications that require further classification. In this regard, a universally accepted definition associated with a clinically validated grading system of severity is needed to define appropriate management and classification criteria as part of further prospective studies.
pancreaticoduodenectomy: clinical predictors and patient outcomes. HPB Surg 2009:404520. 2. Eckardt AJ, Klein F, Adler A, Veltzke-Schlieker W, Warnick P, Bahra M et al. (2011) Management and outcomes of haemorrhage after pancreatogastrostomy versus pancreatojejunostomy. Br J Surg 98:1599–1607. 3. Connor S. (2016) Defining post-operative pancreatitis as a new pancreatic specific complication following pancreatic resection. HPB 18:642–651. 4. Kriger
AG,
Kubishkin
VA,
Karmazanovskii
GG,
Svitina
KA,
Kochatkov AV, Berelavichus SV et al. (2012) The postoperative pancreatitis after the pancreatic surgery. Khirurgiia (Mosk), 14–19. 5. Gueroult S, Parc Y, Duron F, Paye F, Parc R. (2004) Completion pancreatectomy for postoperative peritonitis after pancreaticoduodenectomy: early and late outcome. Arch Surg 139:16–19. 6. Bollen TL, Singh VK, Maurer R, Repas K, van Es HW, Banks PA et al. (2012) A comparative evaluation of radiologic and clinical scoring systems in the early prediction of severity in acute pancreatitis. Am J Gastroenterol 107:612–619. 7. Jones L, Russell C, Mosca F, Boggi U, Sutton R, Slavin J et al. (1999) Standard kausch-whipple pancreatoduodenectomy. Dig Surg 16: 297–304. 8. Clavien PA, Barkun J, de Oliveira ML, Vauthey JN, Dindo D, Schulick RD et al. (2009) The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 250:187–196. 9. Balzano G, Zerbi A, Capretti G, Rocchetti S, Capitanio V, Di Carlo V. (2008) Effect of hospital volume on outcome of pancreaticoduodenectomy in Italy. Br J Surg 95:357–362. 10. Addeo P, Delpero JR, Paye F, Oussoultzoglou E, Fuchshuber PR, Sauvanet A et al. (2014) Pancreatic fistula after a pancreaticoduodenectomy for ductal adenocarcinoma and its association with morbidity: a multicentre study of the French Surgical Association. HPB 16:46–55. 11. Hata T, Motoi F, Ishida M, Naitoh T, Katayose Y, Egawa S et al. (2016) Effect of hospital volume on surgical outcomes after pancreaticoduodenectomy: a systematic review and meta-analysis. Ann Surg 263: 664–672. 12. Ahmad SA, Edwards MJ, Sutton JM, Grewal SS, Hanseman DJ, Maithel SK et al. (2012) Factors influencing readmission after pancreaticoduodenectomy: a multi-institutional study of 1302 patients. Ann Surg 256:529–537. 13. Denbo JW, Orr WS, Zarzaur BL, Behrman SW. (2012) Toward defining grade C pancreatic fistula following pancreaticoduodenectomy: incidence, risk factors, management and outcome. HPB 14:589–593. 14. McMillan MT, Vollmer CM, Jr., Asbun HJ, Ball CG, Bassi C, Beane JD et al. (2016) The characterization and prediction of ISGPF grade C fistulas following pancreatoduodenectomy. J Gastrointest Surg 20:262–276. 15. Ryska M, Rudis J. (2014) Pancreatic fistula and postoperative pancreatitis after pancreatoduodenectomy for pancreatic cancer. Hepatobiliary
Funding
Surg Nutr 3:268–275.
None.
16. Kuhlbrey CM, Samiei N, Sick O, Makowiec F, Hopt UT, Wittel UA. (2017) Pancreatitis after pancreatoduodenectomy predicts clinically relevant
Conflicts of interest All authors declare no commercial associations that might represent conflicts
Postoperative serum amylase predicts pancreatic fistula formation following pancreaticoduodenectomy. J Gastrointest Surg 18:348–353.
References 1. Schmidt CM, Choi J, Powell ES, Yiannoutsos CT, Zyromski NJ, Nakeeb
postoperative pancreatic fistula. J Gastrointest Surg 21:330–338. 17. Cloyd JM, Kastenberg ZJ, Visser BC, Poultsides GA, Norton JA. (2014)
of interest with this article and have nothing to disclose.
A
HPB xxxx, xxx, xxx
et
al.
(2009)
Pancreatic
fistula
following
18. Banks PA, Bollen TL, Dervenis C, Gooszen HG, Johnson CD, Sarr MG et al. (2013) Classification of acute pancreatitis–2012: revision of the
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016
HPB
7
Atlanta classification and definitions by international consensus. Gut 62:
diagnosis of pancreatic necrosis. Cochrane Database Syst Rev 4:
102–111.
CD012645.
19. Alfonso V, Gomez F, Lopez A, Moreno-Osset E, del Valle R, Anton MD
21. Bannone E, Andrianello S, Marchegiani G, Masini G, Malleo G, Bassi C
et al. (2003) Value of C-reactive protein level in the detection of necrosis
et al. (2018) Postoperative acute pancreatitis following pancreatico-
in acute pancreatitis. Gastroenterol Hepatol 26:288–293.
duodenectomy: a determinant of fistula potentially driven by the intra-
20. Komolafe O, Pereira SP, Davidson BR, Gurusamy KS. (2017) Serum C-
operative fluid management. Ann Surg 268:815–822.
reactive protein, procalcitonin, and lactate dehydrogenase for the
HPB xxxx, xxx, xxx
© 2019 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Globke B et al., Postoperative acute necrotizing pancreatitis of the pancreatic remnant (POANP): a new definition of severe pancreatitis following pancreaticoduodenectomy, HPB, https://doi.org/10.1016/j.hpb.2019.07.016