Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor

HPB

https://doi.org/10.1016/j.hpb.2018.07.024

ORIGINAL ARTICLE

Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor Jenny Rystedt1, Bobby Tingstedt1, Christoph Ansorge2, Johan Nilsson3 & Bodil Andersson1 1

Department of Surgery, Skåne University Hospital; Clinical Sciences, Lund University, Lund, 2Division of Surgery, CLINTEC, Karolinska Institutet, Karolinska University Hospital, Stockholm, and 3Cardiothoracic Surgery, Skåne University Hospital; Clinical Sciences, Lund University, Lund, Sweden

Abstract Background: Pancreatoduodenectomy is associated with a high risk of complications. The aim was to identify preoperative risk factors for major intraoperative bleeding. Methods: Patients registered for pancreatoduodenectomy in the Swedish National Pancreatic and Periampullary Cancer Registry, 2011 to 2016, were included. Major intraoperative bleeding was defined as 1000 ml. Univariable and multivariable analysis of preoperative parameters were performed. Results: In total, 1864 patients were included. The median blood loss was 600 ml, and 502 patients (27%) had registered bleeding of 1000 ml. Preoperative independent risk factors associated with major bleeding were male sex (p < 0.001), body mass index (BMI) 25 kg/m2 (p < 0.001), preoperative biliary drainage (PBD) (p < 0.001), C-reactive protein (CRP) 12 mg/L (p = 0.006) and neo-adjuvant chemotherapy treatment (NAT) (p = 0.002). Postoperative intensive care (p < 0.001), reoperation (p = 0.035), surgical infections (p = 0.036), and bile leakage (p = 0.045) were more common in the group with major bleeding, and the 30-day mortality was higher (4.9% vs 1.6%; p < 0.001). Conclusion: Most predictive parameters for major intraoperative bleeding are not modifiable. PBD is an independent predictor for major intraoperative bleeding and to reduce the risk, patients with resectable periampullary tumors should, if possible, be subject to surgery without preoperative biliary drainage. Received 22 March 2018; accepted 22 July 2018

Correspondence J. Rystedt, Department of Surgery, Skåne University Hospital, 221 85, Lund, Sweden. E-mail: jenny. [email protected]

Introduction Pancreatoduodenectomy is the only possible cure for malignant periampullary tumors, but also associated with a high risk for intraoperative and postoperative complications. An increasing number of patients with locally advanced disease or borderline resectable tumors are offered surgery by means of concomitant vascular resection of the portal vein or superior mesenteric vein.1,2 In the presence of arterial involvement, neo-adjuvant chemotherapy is advocated to improve the possibility of resection.3 For jaundiced patients with resectable pancreatic cancer, surgery rather than preoperative biliary drainage is advocated to relieve biliary obstruction.4 Pancreatoduodenectomy is a complex surgical procedure in itself where intraoperative blood loss is

HPB 2018, -, 1–7

anticipated and the etiology is multifactorial. It is however reported that major intra-operative bleeding may have a negative impact on short term outcome and long-term survival.5–7 In Sweden, having a population of approximately 10 million inhabitants, the centralization of pancreatic cancer surgery to six high volume centers has been in progress during the past decade. In 2010, the profession instituted a national registry for all periampullary tumors subject to pancreatoduodenectomy, the Swedish National Pancreatic and Periampullary Cancer Registry. The validation process of the registry was completed in 2016 showing a national coverage of more than 90% for recent years and 87% in total since the start, in comparison with the national gold standard, the Swedish Cancer Registry.8,9

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

2

The aim of this study was to identify preoperative risk factors, modifiable as well as non-modifiable, for patients with major intraoperative blood loss, and to outline the intra- and postoperative outcomes.

Methods Swedish National Pancreatic and Periampullary Cancer registry All data are registered prospectively into a secure web site by the operating surgeon. The postoperative data are collected from the medical records or at an outpatient visit. A complete list of variables is available on the web page of the register.10 National and regional data are analysed and the results are e-published once a year.8 All consecutive registered patients who underwent pancreatoduodenectomy from January 2011 to December 2016 were included. Registration is performed on-line in six different forms (A-F) including preoperative descriptive variables, intraoperative and postoperative variables as well as 1-year follow up. Preoperative variables included were age, sex, BMI (body mass index), involuntary weight-loss, smoking, diabetes, ASA (American Society of Anesthesiologists) Physical Status Classification System, hemoglobin, C-reactive protein (CRP), bilirubin, heart disease, preoperative biliary drainage (PBD) tumor size and neo-adjuvant chemotherapy (NAT). Heart disease was defined as the use of antiarrhythmic- or antihypertensive medication, warfarin or other anticoagulant medication, and/or symptoms/presence or treatment of ischemic heart disease, congestion or increased pulmonary pressure. Intraoperative variables included bleeding as the mandatory study variable, operating time, intraoperative transfusion and vascular resection. Postoperative variables

included were; postoperative transfusion, ICU (Intensive Care Unit) stay, morbidity registered according to Clavien-Dindo; including delayed gastric emptying (DGE), surgical infections, bile leakage, post-operative pancreatic fistula (POPF), postpancreatic hemorrhage (PPH), reoperations and mortality.11,12 Major intraoperative bleeding was defined as 1000 ml (corresponding to the 4th quartile).7,13 Statistical analysis Baseline characteristics between patients with and without major bleeding were compared using the unpaired Mann–Whitney U tests for continuous variables and the chi-squared (c2) for categorical variables. Multivariable analysis was performed using logistic regression. Inclusion criteria for the full model was p < 0.25. A p-value < 0.05 was considered significant. Statistical analyses were performed using Stata MP statistical package version 15.1 (2017) (StataCorp LP, College Station, Texas, USA. Ethics Approval for the study was obtained from the Lund Regional Ethics Committee, Sweden; Dnr2015/393.

Results A total of 1864 patients were included. The mean age was 66.7 ± 9.7 years and time to surgery from final decision regarding best treatment was in median 27 (19–40) days. The median blood loss was 600 (300–1000) ml, and 502 patients had bleeding of 1000 ml or more, Fig. 1. The pre- and perioperative variables in relation to the estimated intraoperative bleeding (IOB) are listed in Table 1. The operative time was longer and vascular resection more

Figure 1 Intraoperative bleeding in milliliters for patients subjected to pancreatoduodenectomy (n = 1864)

HPB 2018, -, 1–7

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

3

Table 1 Pre- and perioperative variables in relation to the estimated intraoperative bleeding (IOB)

N

Total (n = 1864)

IOB <1000 ml (n = 1362)

IOB 1000 ml (n = 502)

Age (years)

1854

68 (62–73)

68 (62–74)

68 (62–73)

0.123

Male sex

1864

992 (54)

681 (50)

321 (64)

<0.001

p

BMI

1788

25 (22–28)

24 (22–27)

26 (23–29)

<0.001

Weight loss

1829

953 (52)

691 (52)

262 (53)

0.515

Smoking

1786

330 (19)

234 (18)

96 (20)

0.340

Diabetes

1851

354 (19)

240 (18)

114 (23)

0.016

Heart disease

1844

594 (32)

404 (30)

190 (38)

0.001

ASA 2

1849

1447 (78)

1055 (78)

392 (78)

0.992

Bilirubin

1779

15 (8–34)

15 (8–34)

16 (3–30)

0.275

Preop biliary drainage

1850

1215 (66)

845 (62)

370 (74)

<0.001

CRP

1728

5 (2–8)

5 (2–11)

8 (3–16)

<0.001

Pancreas T3

932

180 (19)

135 (75)

45 (25)

0.030

Duodenum T3

101

69 (68)

58 (84)

11 (16)

0.176

Neoadjuvant chemotherapy

1855

52 (2.8)

27 (2.0)

25 (5.0)

<0.001

Vascular resection

1855

320 (17)

179 (13)

141 (28)

<0.001

Operating time (min)

1060

385 (336–444)

372 (325–426)

420 (374–491)

<0.001

Intraop bleeding (ml)

1864

600 (300–1000)

450 (250–600)

1500 (1100–2000)

a

N, number of patients with non-missing values. Qualitative data are expressed as n (%), quantitative as mean (±SD) or median (IQR), as appropriate. Body mass index kg/m2 (BMI); American Society of Anesthesiologists (ASA) Physical Status; Bilirubin mmol/L(Bilirubin); C-reactive protein mg/L (CRP).

common in patients with major bleeding, Table 1. The intraoperative transfusion rate in the group with major bleeding was 56%, and in the control group 9% (p < 0.001). Preoperative independent risk factors associated with major bleeding were male sex, BMI 25 kg/m2, preoperative biliary drainage, CRP 12 mg/L and neo-adjuvant chemotherapy treatment, Table 2. No difference could be shown between the groups for the level of bilirubin or if the patient was a smoker or not, Table 2. A box-plot of intraoperative bleeding volumes for patients with a BMI above and below 25 as well as patients who had received and not received preoperative neoadjuvant chemotherapy are displayed in Fig. 2 a and b. In the group of Table 2 Univariable

patients that had received preoperative chemotherapy (52/1855; 2.8%) preoperative hemoglobin level was lower, 125 (117–131) g/L versus 130 (120–139) g/L (p = 0.029) and vascular resection were more common, 33 versus 17% (p = 0.005). Postoperative ICU stay, reoperation, surgical infections and bile leakage were more common in the group with major bleeding, Table 3. Other known complications such as postoperative pancreatectomy fistula, delayed gastric emptying, postoperative pancreatic hemorrhage, and medical complications (including deep venous thrombosis, pulmonary embolism and cardiac complications) did not significantly differ between the groups, Table 3. Histology was complete for 1698

and multivariable analysis to identify predictors for major intraoperative bleeding (1000 ml) during

pancreatoduodenectomy Event/Total

Univariable analysis

Multivariable analysis

OR

95% CI

p-value

OR

95% CI

p-value

Male sex

1002/1864

1.77

1.44–2.19

<0.001

1.87

1.49–2.35

<0.001

BMI 25

838/1788

1.77

1.43.–2.19

<0.001

1.72

1.37–2.15

<0.001

Diabetes

862/1854

1.36

1.06–1.75

0.016

Heart disease

594/1844

1.42

1.15–1.76

0.001

PBD

1215/1850

1.67

1.33–2.09

<0.001

1.59

1.25–2.04

<0.001

CRP 12

452/1728

1.47

1.16–1.85

0.001

1.41

1.10–1.80

0.006

NAT

52/1855

2.60

1.49–4.53

0.001

2.70

1.43–5.10

0.002

Odds ratio (OR); Confidence interval (CI); Body mass index kg/m2 (BMI); Preoperative biliary drainage (PBD); Preoperative biliary drainage (PBD); C-reactive protein mg/L (CRP); Neoadjuvant chemotherapy (NAT).

HPB 2018, -, 1–7

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

4

Figure 2 (a) Intraoperative bleeding in milliliters for patients subjected to pancreatoduodenectomy divided in two groups according to BMI (body

mass index) < 25 (n = 950) or 25 kg/m2 (n = 838). (b) Intraoperative bleeding in milliliters for patients subjected to pancreatoduodenectomy divided in two groups according to if neoadjuvant chemotherapy (NAT, n = 52) were given or not (No-NAT, n = 1803)

patients and showed pancreatic cancer for 899 (53%), ampullary cancer for 20612 and duodenal cancer for 104 (6.1%) patients. Patients with ampullary cancer and duodenal cancer bled significantly less than patients with pancreatic cancer (OR

HPB 2018, -, 1–7

0.41 CI 0.27–0.61, p < 0.001 and OR 0.33 CI 0.19–0.60, p < 0.001). The 90-day mortality was higher in the group with major bleeding, 5.3% compared with 2.7% in the control group (p = 0.007), Table 3.

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

5

Table 3 Postoperative outcome variables in relation to the estimated intraoperative bleeding (IOB)

N

Total (n = 1732)

IOB <1000 ml (n = 1252)

IOB 1000 ml (n = 480)

p

1731

512 (39)

320 (26)

192 (40)

<0.001

510

3.8 (2–4)

3.5 (2–4)

4.4 (2–4)

0.128

Postop ICU

1732

130 (7.5)

73 (5.8)

57 (12)

<0.001

Clavien 2 DGE

1732

356 (21)

260 (21)

96 (20)

0.740

Surg. inf.

1732

310 (18)

209 (17)

101 (21)

0.036

Bile leak.

1732

73 (4.2)

45 (3.6)

28 (5.8)

0.045

POPF

1732

292 (17)

223 (18)

69 (14)

0.099

PPH

1732

152 (8.8)

106 (8.5)

46 (9.6)

0.450

Postop tranfusion Units

Reop

1732

180 (10)

118 (9.4)

62 (13)

0.035

Mortality 90-days

1830

62 (3.4)

36 (2.7)

26 (5.3)

0.007

N, number of patients with non-missing values. Qualitative data are expressed as n (%), quantitative as median (IQR). Intensive care unit (ICU); Delayed gastric emptying (DGE); Surgical infection (Surg. Inf.); Bile leakage (Bile leak.); Postoperative pancreatic fistula (POPF); Postpancreatectomy haemorrhage (PPH); Reoperation (Reop).

Discussion This national study on resectable periampullary tumors shows that the only modifiable risk factor for major intraoperative bleeding was preoperative biliary drainage (PBD). This emphasizes the need for a fast-track preoperative evaluation to reduce the time to surgery for resectable patients.14 Historically major surgery on severely jaundiced patients was associated with increased morbidity and mortality. Extensive experimental studies have also shown a positive effect of PBD on liver function, nutritional status and immune function but this has not been possible to repeat in clinical studies.15 In the modern era, it has instead been shown that the risks of PBD are higher than the risk of operating a jaundiced patient.14,16 Several centers report increased infectious complications with PBD and Roberts et al. could also show that early surgery, without PBD, resulted in an increased resection rate, compared to patients subject to PBD.14 The delay caused by PBD, and possible complications, may result in tumor progression and a subsequent reduced resection rate. BMI is indeed a modifiable risk factor in the long-term but in the situation where need for surgery is very imminent, as in the presence of pancreatic cancer, no intervention is possible to normalize a high BMI. In 2004 Chang et al. demonstrated BMI to be the only independent predictor of intraoperative bleeding in radical prostatectomy.17 With a cut-off of 25 kg/m2, overweight and obese patients had a significantly higher risk for major intraoperative blood loss. Krane and colleagues also showed an increased intraoperative bleeding among overweight and obese patients in laparoscopic surgery for inflammatory bowel disease, but could not verify any difference in postoperative complications.18 Sandini et al. showed that not BMI but sarcopenic obesity as a strong predictor for major complications after pancreatoduodenectomy19 Male sex is a known risk factor for major intraoperative bleeding in hepatectomy, and was a risk factor in the present HPB 2018, -, 1–7

study.20 It is also a risk factor for post pancreatectomy hemorrhage (PPH).21 Preoperatively patient-related modifiable risk factors are for example tobacco use and over consumption of alcohol, and these risk factors should be addressed before surgery, since two weeks of abstinence results in lower morbidity also in terms of less intraoperative bleeding.22 In this study we found no difference in intraoperative bleeding whether the patient was a smoker or not, and data on alcohol consumption were not available for analysis. An elevated CRP, an expression for increased inflammatory response, was associated with an increased risk for major intraoperative bleeding. Many argue that inflammation is “the fuel that feeds the flame” once the genetic damage has occurred.23 A trend to lower inflammatory markers among patients with longer survival has been shown and elevated CRP levels is probably a marker for more advanced disease and worse prognosis.24 Also, elevated CRP may co-vary with PBD since elevated CRP is associated with infectious complications and post-ERC pancreatitis, but multivariable analysis in this study proved CRP and PBD to be independent risk factors. Neo-adjuvant chemotherapy in the presence of border-line resectable disease has been increasingly utilized during the past decade, and some reports suggest that preoperative systemic therapy should be considered for all resectable ductal adenocarcinomas, irrespective of size, since also small tumors have proven to give rise to systemic (micro-) metastasis.25 For patients needing neoadjuvant chemotherapy, biliary drainage is not optional, possibly adding risk for major bleeding during later surgery. The increased risk of major intraoperative bleeding associated with neo-adjuvant chemotherapy demonstrated in this study is probably due to more advanced disease, since tumor stage 3 and 4 were more frequent among patients with major bleeding. Longer operating times is reported by many to be associated with increased estimated intraoperative bleeding.26 This was also

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

6

found in the present study. Locally advanced disease and more complex procedures with vascular resection should require more time. But time-extending meticulous dissection techniques increase the risk for hypothermia and bleeding and though surgeons strive for minimal blood-loss, longer operating times infallibly increases total intraoperative bleeding.27 Keeping patients warm have been reported to not only reduce the intraoperative bleeding but also to give a threefold decrease in wound infections.27 In this study patients with major bleeding after pancreatoduodenectomy experienced, as expected, more frequent reoperations and an increased 90-day mortality. An important outcome-measure after pancreatoduodenectomy is the frequency of post-operative pancreatic fistula. Callery et al. could demonstrate an almost six-fold increased risk for a clinically significant fistula with an intraoperative blood loss of 1000 ml or more.7 Surprisingly, no correlation between major bleeding and post-operative pancreatic fistula could be found in this study (p = 0.099). This study has limitations inherent to the retrospective analysis of registry data. Further, in the present study cohort, the total volume of bleeding has been estimated, not measured, and should be regarded as a subjective categorization, not an objective finding, hence local traditions could have importance for the specified volume. In conclusion, we found that most predictive factors for major intraoperative bleeding during pancreatoduodenectomy are not modifiable. Male sex, overweight, neo-adjuvant chemotherapy treatment and CRP 12 mg/L were all independently associated with an increased risk of major intraoperative bleeding. The only modifiable risk factor for major intraoperative bleeding was preoperative biliary drainage (PBD). In jaundiced patients with resectable periampullary tumors, PBD should, if possible, be avoided in favor of fast-track surgery since this is the only modifiable factor to reduce the risk for major intraoperative bleeding.

3. Bockhorn M, Uzunoglu FG, Adham M, Imrie C, Milicevic M, Sandberg AA et al. (2014) Borderline resectable pancreatic cancer: a consensus statement by the international study group of pancreatic surgery (ISGPS), 2014 Surgery 155:977–988. 4. NICE. (2018) National institute for helath and care excellence. Available from: https://www.nice.org.uk/guidance/ng85/chapter/Recommendations# relieving-biliary-and-duodenal-obstruction. [Accessed 22 May 2018]. 5. Kazanjian KK, Hines OJ, Duffy JP, Yoon DY, Cortina G, Reber HA. (2008) Improved survival following pancreaticoduodenectomy to treat adenocarcinoma of the pancreas: the influence of operative blood loss. Arch Surg 143:1166–1171. 6. Braga M, Capretti G, Pecorelli N, Balzano G, Doglioni C, Ariotti R et al. (2011) A prognostic score to predict major complications after pancreaticoduodenectomy. Ann Surg 254:702–707. discussion 7-8. 7. Callery MP, Pratt WB, Kent TS, Chaikof EL, Vollmer CM, Jr.. (2013) A prospectively validated clinical risk score accurately predicts pancreatic fistula after pancreatoduodenectomy. J Am Coll Surg 216: 1–14. 8. Annual Report. (2016) Swedish national pancreatic and periampullary cancer registry, 2016 [Available from: http://www.cancercentrum.se/ samverkan/cancerdiagnoser/bukspottkortel/kvalitetsregister/. [Accessed 22 May 2018]. 9. NBHW. (2017) Swedish cancer registry. Available from: https://www. socialstyrelsen.se/register/halsodataregister/cancerregistret/inenglish. [Accessed 22 May 2018]. 10. Register variables. (2017) Swedish pancreatic and periampullary cancer registry.

Available

from:

http://www.cancercentrum.se/samverkan/

cancerdiagnoser/bukspottkortel/kvalitetsregister/dokument/.

[Accessed

22 May 2018]. 11. Dindo D, Demartines N, Clavien PA. (2004) Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 240:205–213. 12. 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. 13. Chen JS, Huang JQ, Chen XL, Zhan GF, Feng JT. (2015) Risk factors associated with intraoperative major blood loss during resection of hepatocellular carcinoma. Hepato-Gastroenterology 62: 790 – 793. 14. Roberts KJ, Prasad P, Steele Y, Marcon F, Faulkner T, Cilliers H et al.

Acknowledgement

(2017) A reduced time to surgery within a ’fast track’ pathway for

Thanks are due to the HPB surgeons of Sweden for reporting data to the Swedish National Registry for Pancreatic and Periampullary Cancer and to all persons involved in maintaining the registry. This research was supported by Government grant for clinical research (http://www.skane.se/fou/alf) and Erik and Angelica Sparres research foundation.

periampullary malignancy is associated with an increased rate of pancreatoduodenectomy. HPB 19:713–720. 15. Tol JA, Busch OR, van der Gaag NA, van Gulik TM, Gouma DJ. (2012) The quandary of preresection biliary drainage for pancreatic cancer. Cancer J 18:550–554. 16. van der Gaag NA, Rauws EA, van Eijck CH, Bruno MJ, van der Harst E,

Conflict of interest

Kubben FJ et al. (2010) Preoperative biliary drainage for cancer of the

None declared.

head of the pancreas. NEJM 362:129–137. 17. Chang SS, Duong DT, Wells N, Cole EE, Smith JA, Jr., Cookson MS.

References 1. Zhou Y, Zhang Z, Liu Y, Li B, Xu D. (2012) Pancreatectomy combined

(2004) Predicting blood loss and transfusion requirements during radical

with superior mesenteric vein-portal vein resection for pancreatic

prostatectomy: the significant negative impact of increasing body mass

cancer: a meta-analysis. World J Surg 36:884–891.

index. J Urol 171:1861–1865.

2. Kleive D, Sahakyan MA, Berstad AE, Verbeke CS, Gladhaug IP, Edwin B

18. Krane MK, Allaix ME, Zoccali M, Umanskiy K, Rubin MA, Villa A et al.

et al. (2017) Trends in indications, complications and outcomes for

(2013) Does morbid obesity change outcomes after laparoscopic sur-

venous resection during pancreatoduodenectomy. Br J Surg 104:

gery for inflammatory bowel disease? Review of 626 consecutive cases.

1558–1567.

J Am Coll Surg 216:986–996.

HPB 2018, -, 1–7

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024

HPB

7

19. Sandini M, Bernasconi DP, Fior D, Molinelli M, Ippolito D, Nespoli L et al. (2016) A high visceral adipose tissue-to-skeletal muscle ratio as a determinant of major complications after pancreatoduodenectomy for cancer. Nutrition 32:1231–1237. (2013) Predictors of intraoperative blood loss in patients undergoing hepatectomy. Surg Today 43:485–493.

protein and the neutrophil-lymphocyte ratio in resectable pancreatic cancer: a systematic review. HPB 17:285–291. 25. Ansari D, Bauden M, Bergstrom S, Rylance R, Marko-Varga G,

21. Wellner UF, Kulemann B, Lapshyn H, Hoeppner J, Sick O, Makowiec F et al. (2014) Postpancreatectomy hemorrhage-incidence, treatment, and risk factors in over 1,000 pancreatic resections. J Gastrointest Surg 18:464–475.

Andersson R. (2017) Relationship between tumour size and outcome in pancreatic ductal adenocarcinoma. Br J Surg 104:600–607. 26. Nanashima A, Abo T, Hamasaki K, Wakata K, Kunizaki M, Nakao K et al. (2013) Predictive parameters of intraoperative blood loss in patients

22. Tonnesen H, Rosenberg J, Nielsen HJ, Rasmussen V, Hauge C, Pedersen IK et al. (1999) Effect of preoperative abstinence on poor postoperative outcome in alcohol misusers: randomised controlled trial.

HPB 2018, -, 1–7

virchow? Lancet 357:539–545. 24. Stevens L, Pathak S, Nunes QM, Pandanaboyana S, Macutkiewicz C, Smart N et al. (2015) Prognostic significance of pre-operative C-reactive

20. Nanashima A, Abo T, Hamasaki K, Wakata K, Kunizaki M, Tou K et al.

BMJ 318:1311–1316.

23. Balkwill F, Mantovani A. (2001) Inflammation and cancer: back to

who

underwent

pancreatectomy.

Hepato-Gastroenterology

60:

1217–1221. 27. Sessler DI. (1997) Mild perioperative hypothermia. NEJM 336: 1730–1737.

© 2018 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Rystedt J, et al., Major intraoperative bleeding during pancreatoduodenectomy - preoperative biliary drainage is the only modifiable risk factor, HPB (2018), https://doi.org/10.1016/j.hpb.2018.07.024