Risk factors for postoperative bile leakage: a retrospective single-center analysis of 411 hepatectomies

Fabrizio Panaro, Lisa Hacina, Hassan Bouyabrine, Al-Warith Al-Hashmi, Astrid Herrero and Francis NavarroMontpellier, France

?

Risk factors for postoperative bile leakage: a retrospective single-center analysis of 411 hepatectomies

Fabrizio Panaro, Lisa Hacina, Hassan Bouyabrine, Al-Warith Al-Hashmi, Astrid Herrero and Francis Navarro
Montpellier, France

Author Affiliations: Department of General and Liver Transplant Surgery, Montpellier University Hospital, College of Medicine, 80 Avenue Augustin Fliche 34295, Montpellier-Cedex 5, France (Panaro F, Hacina L, Bouyabrine H, Al-Hashmi A, Herrero A and Navarro F)

? 2016, Hepatobiliary Pancreat Dis Int. All rights reserved.

Published online September 17, 2015.

BACKGROUND: The primary focus of the study was to analyze the risk factors for bile leakage after hepatectomy for benign or malignant tumors.

METHODS: A total of 411 patients who had undergone hepatectomy between December 2006 and December 2011 were retrospectively analyzed. The severity of bile leakage was graded according to the ISGLS classification. Twenty-eight pre- and postoperative parameters were analyzed.

RESULTS: The overall bile leakage incidence was 10.2% (42/411). The severity of the leakage was classified according to the ISGLS classification. Bile leakage was detected early in case of abdominal drainage (11.4% vs 1.9%, P=0.034). It prolonged the time of hospitalization (16 vs 9 days, P=0.001). In all patients, wedge resection was associated with a higher incidence of bile leakage in contrast to anatomical resections (25.6% vs 4.1%, P<0.0001) regardless of the underlying liver disease. Furthermore, total vascular exclusion increased risk of bile leakage (P=0.008).

CONCLUSIONS: Bile leakage as a major issue after hepatic resection is related to the postoperative morbidity and the hospitalization time. It is associated with non-anatomical resection and a total vascular exclusion.

(Hepatobiliary Pancreat Dis Int 2016;15:81-86)

KEY WORDS:liver resection; bile leakage; wedge; total vascular exclusion

IntroductionMethods

Definitions

Bile leakage was defined as the bilirubin concentration in the drain fluid being at least 3 times higher than the serum bilirubin concentration on or after postoperative day 3 or as the need for radiologic or operative intervention resulting from biliary collections or bile peritonitis. Using this criterion, we classified the severity of bile leakage according to its impact on patients’ clinical manage-ment. Grade A bile leakage causes no change in patients’clinical management. Grade B bile leakage requires active therapeutic intervention but is manageable without relaparotomy, whereas in grade C, bile leakage re-laparotomy is required.[1]

Surgical resection

Anatomic hepatectomy: if the line of resection follows the limits of one or more functional segments of the liver as defined by the Couinaud classification.[4, 8]Non-anatomic (wedge) hepatectomy: if the line of resection does not follow the limits of one or more functional segments of the liver as defined by the Couinaud classification.[4, 8]Major hepatectomy is defined as the resection of at least three segments of the liver.[4, 8-10]

Patient selection

After approval from the Institutional Review Board of Montpellier University Hospital, retrospectively collected clinico-pathological data of 411 liver resections performed in our institution between December 2006 and December 2011 were reviewed.

All the consecutive patients who had undergone hepatectomy were included, except for those who had undergone liver biopsy, biliary surgery, fenestration or resection of benign cyst, hepatic resection associated with hepatico-jejunostomy and liver transplantation.

The demographic characteristics of the patients were identified. The operative indications were collected and classified as benign tumor, hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma, colorectal and non-colorectal liver metastasis, and other primary malignant tumors. The surgical technique, the type of hepatectomy (minor, major, anatomical, non-anatomical), the surgical devices utilized for liver transection, the type and duration of vascular clamping, the hemostatic agent used, and the type and length of drainage were also recorded. The medical records of these patients were retrospectively analyzed for bile leakage-related factors [presentation, length, quantity, quality, management (medical, surgical, endoscopic and/or radiological), complications (fever, pain, abscess, sepsis, pleural effusion) and recurrence]. The length of hospitalization and the mortality rate were collected.

Surgical technique

Routinely, we did not perform magnetic resonance imaging (MRI) preoperatively for the biliary tree. However, in case of cholestasis (increase of gamma glutamyl transpeptidase, alkaline phosphatase, and total bilirubin serum level) or biliary abnormalities (dilated biliary tree at ultrasound or computed tomography), an MRI was performed.

Intraoperative ultrasound was routinely performed in order to detect the lesions described in the preoperative imaging and analyze their contiguity with the vascular structures. During the resection, the hepatic pedicle control was routinely performed. The pedicle clamping was complete in 27.3% of the patients, and selective (right/left) in 4.9%. In only 2.7% of the patients, we performed a complete vascular exclusion of the liver (segment I hypertrophy, giant lesions). The pedicle clamping was intermittent in all patients. The clamping time ranged from 10 to 60 minutes according to the type of resection, the state of underlying liver parenchyma and the instrument utilized for the resection.

During the study period, four senior surgeons performed 411 liver resections (excluding resection of Klatskin tumors and hepatectomy associated with biliary or pancreatic resection). The devices used for liver resection were different according to the practice of surgeons. Hemostasis and bile control were performed by ligation with no absorbable suture (polypropylene 4/0 and 5/0), and/or metal clips. The trans-cystic methylene blue test and/or a trans-cystic cholangiography after surgery were not systematically performed except in case of doubt about a bile duct injury or anatomic abnormalities. A withe lap test was applied for a few minutes on the cut surface of liver resection, to detect biliary leakage in all patients. The hemostatic agents were not applied routinely on the liver transection surface after surgery. A passive drainage is placed close to the cut surface of the liver at the end of the operation. This drain was left in place, on average, 3 to 5 days except in case of complication.

Statistical analysis

Parametric and nonparametric data were expressed as mean±standard deviation and median (range), respectively. Primary endpoints included patient survival. Statistical analysis was performed according to the Kaplan-Meier method and resulting curves were compared using the log-rank test. Univariate analysis was performed to identify factors associated with the incidence of bile leakage. A multivariate logistic regression analysis was performed, including variables associated with the occurrence of bile leakage with a P value less than 0.20. A difference was considered statistically significant when a P value less than 0.05. All statistical analyses were performed using the SPSS ? statistical package (SAS Institute, Cary, NC, USA).

Results

Study population

The cohort consisted of 411 patients including 240 (58.4%) males and 171 (41.6%) females. The median age was 60years (range 18-84 years). Hepatectomy was indicated for benign liver tumor (adenoma, focal nodular hyperplasia, hemangioma, hydatid cyst, cystadenoma) in 98 (23.8%) patients, for HCC in 136 (33.1%) patients, for intrahepatic cholangiocarcinoma in 20 (4.9%) patients, for colorectal and non-colorectal liver metastasis in 154 (37.5%) patients and for other rare primary malignant tumor (sarcoma) in 3 (0.7%) patients. The underlying liver parenchyma was histologically normal in 115 patients（ 28.0%) who received neoadjuvant chemotherapyy,, normal in 156 patients (38.0%) without chemotherapy, steato-hepatitis or sinusoidal obstructive syndrome in 9.2% (n=38) of the patients and cirrhotic in 24.8% (n=102) of the patients. Of the 153 patients that received neoadjuvant chemotherapy, 99 (64.7%) also received bevacizumab as anti-cancer drug. These data are summarized in Table 1.

Table 1. Demographics and data of the patients (n, %)___

Hepatectomies were performed by a laparotomy (subcostal, median) in 86.9% (n=357) of the patients and by a laparoscopic approach in 13.1% (n=54) of the patients. Liver resections were minor (one or two segments) in 63.3% (n=260) of the patients and major (beyond three segments) in 36.7% (n=151) of the patients. The surgical devices utilized for hepatic transection included an ultrasonic surgical aspirator Cavitron ? CUSA (Dissectron?, Integra NeuroSciences Limited Newbury Road Andover - Hampshire SP10 DR, UK) in 39.4% (n=162) of the patients, the bipolar sealer (TissueLink? or Aquamantys?, Medtronic, Minneapolis, MN, 55432-5604, USA) in 24.1% (n=99) of the patients, and vessels sealing instruments (LigaSure?, Covidien, Medtronic, MN, Minnesota, 55432-5604, USA; or Ultracsion? Harmonic scalpel, Ethicon US, LLC) in 32.1% (n=132) of the patients, and crush clamping in the remnants 4.4% (n=18). Parenchymal resection and hemostasis were performed according to the surgeons’ preference and technique. Approximately 42.3% (n=174) of the patients received ligatures and titanium clips or polypropylene 5/0 and 4/0 ligature only, 38.2% (n=157) received, in addition to ligatures, hemostatic collagen sponge soaked with thrombin and human fibrinogen, 16.1% (n=66) of the patients received other hemostatic agents and 3.4% (n=14) received biological glue (Table 2).

Table 2. Intraoperative data (n, %)

In total, 268 (65.2%) patients did not require vascular clamping during hepatectomy, but 112 (27.3%) had a complete pedicle clamping, 20 (4.9%) had a selective clamping and 11 (2.7 %) had a total vascular exclusion (TVE). The length of clamping was less than 20 minutes in 84 (58.7%) of the 143 clamped patients. Passive drainage in contact with the liver surface was utilized in 359 (87.3%) of the patients and the duration of drainage was less than 5 days in 287 (80.0%) of the patients. The median length of hospitalization was 10 days (range 1-66 days).

Incidence of bile leakage and outcome

Bile leakage was seen in 42 (10.2%) of the 411 patients within 30 days postoperatively. Nine (21.4%) patients were classified as grade A, 27 (64.3%) as grade B and 6 (14.3%) as grade C. Bile leakage was diagnosed by the presence of bile into the abdominal drain in 19 (45.2%) patients, and the presence of intra-abdominal collection containing bile in 22 (52.4%) of the patients (Table 3). The median time to onset of bile leakage was 5 days after surgery (range 2-21 days). In the group of atypical hepatic resections (n=117), bile leakage occurred in 30 (25.6%) of the patients vs 12 (4.1%) of the 294 in patients undergoing anatomic resection (P<0.0001). Among the various intra-operative types of vascular controls (selective, hepatic pedicle, TVE), univariate analysis revealed a predominance of bile leakage in case of TVE clamping (P=0.008) (Tables 4, 5).

The postoperative mortality (within 30 days) of the patients was 3.6% (n=15), including only 1 patient with bile leakage. The patient died of liver failure secondary to a cirrhosis decompensation. The length of hospitalization was longer in patients with bile leakage (median 16 days) than that in patients without bile leakage (median 9 days) (P<0.001). The management of bile leakage was purely medical in 9 (21.4%, grade A) of the 42 patients. Three patients (7.1%, grade B) required endoscopic treatment(two sphincterotomies and one biliary stent placement associated with sphincterotomy), 24 (57.1%, grade B) required treatment by radiological means (22 underwent single percutaneous drainage and two internal-external biliary drainage), and 6 (14.3%, grade C) required surgical intervention to optimize drainage of the bile.

Predictive factors

The factors relating to bile leakage are summarized in Table 5. In the 117 patients with non-anatomical hepatectomies, 30 experienced bile leakage but in 294 patients who underwent anatomical resection, 12 had bile leakage (P<0.0001). Univariate analysis showed a predominance of bile leakage after use of TVE (P=0.008). Other factors such as the quality of the underlying liver (normal, steatosis, chemotherapy toxicity and cirrhosis), pathology (benign or malignant tumor), devices utilized for parenchymotomy, and minor or major hepatectomy showed no significant difference in occurrence of bile leakage after hepatic resection. Surgical re-intervention was required in 14.3% of the patients with bile leakage (6/42).

Table 3. Bile leakage and its complications (n, %)______

Table 4. Grade and treatment of bile leakage according to the ISGLS classification

Table 5. Variables associated with bile leakage after liver resection

In case of bile leakage refractory to conservative medical treatment (9 patients, 21.4%) sphincterotomy/endoprothesis was performed in 3 (7.1%) patients, alternativelya radiologic procedure (internal/external drainage) was performed in 24 (57.1%) patients. In our series, radiological treatment was often prescribed because of the high volume of fluid collection (biloma). Laparotomy was done in 6 (14.3%) patients because of the radiological inefficacity.

Discussion

The results of this study suggest that non-anatomical hepatic resections and TVE are associated with the increased rate of bile leakage after hepatectomy and therefore a longer hospitalization time.

In our patients, the incidence of bile leakage after hepatic resection was 10.2%, and it is comparable to that reported previously.[5, 11]Therefore, bile leakage after hepatectomy remains a serious complication lengthening the hospitalization time and a causal factor for decompensation of liver function especially in cirrhotic patients.[8-18]

In our patients, the quality of the remnant liver parenchyma was not associated with the occurrence of postoperative bile leakage contrary to what was found in the study by Guillaud et al.[19]In fact, the patients with colorectal liver metastases who benefited from bevacizumab based chemotherapy showed a higher rate of bile leakage after hepatectomy. We found that atypical liver resection was associated with a significantly higher rate of postoperative bile leakage compared to anatomical resection. This can be explained by the injury of biliary and the vascular structures during the surgery as well reported in different studies.[19-25]Concerning the devices used for the transection of the liver parenchyma, we did not find a causal link between the device utilized and the incidence of bile leakage. This is probably due to the miscellaneous methods used in our center. TVE of the liver during hepatectomy was associated (despite the small number of cases) with the occurrence of bile leakage after surgery, in contrast to the recently published study[19]where the selective vascular clamping is more associated with the occurrence of bile leakage. Moreover, unlike what has been reported in some studies[6, 13](advocate the protective effect of biological agents on the bile leak appearance), we did not find any benefit of their use (thrombin human and fibrinogen patch agents, glue, etc.) compared to the standard ligature and clips. Two randomized trials[9, 19]reported the benefit of “l(fā)eakage tests”(methylene blue and/or cholangiography) in detecting bile leakage; in our study no any benefit was found. In our study, the presence of an abdominal drain allowed to perform an early diagnosis of bile leakage and therefore to perform a prompt medical treatment. Despite the limitations such as retrospective analysis, selection bias criteria, miscellaneous indications and procedures, and different surgeons, we report a large single center “real life” experience.

In conclusion, bile leakage is a common complication after liver resection and it has a significant effect on postoperative morbidity and cost of care (lengthening of the hospitalization). In our study, the occurrence of bile leakage is associated with non-anatomical (wedge) resection and TVE.

Contributors: PF and HL proposed the study. PF, HL, AA and HA wrote the first draft. PF, HL and BH collected the data. PF, HL, AA and NF contributed to the design and interpretation of the study and to further drafts. PF and HL contributed equally to this article. NF is the guarantor.

Funding: None.

Ethical approval: This study was approved by the Ethics Committee of the Montpellier University Hospital School of Medicine (2014-0041).

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

References

1 Koch M, Garden OJ, Padbury R, Rahbari NN, Adam R, Capussotti L, et al. Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery 2011;149:680-688.

2 Imamura H, Seyama Y, Kokudo N, Maema A, Sugawara Y, Sano K, et al. One thousand fifty-six hepatectomies without mortality in 8 years. Arch Surg 2003;138:1198-1206.

3 Jarnagin WR, Gonen M, Fong Y, DeMatteo RP, Ben-Porat L, Little S, et al. Improvement in perioperative outcome after hepatic resection: analysis of 1,803 consecutive cases over the past decade. Ann Surg 2002;236:397-407.

4 Yamashita Y, Hamatsu T, Rikimaru T, Tanaka S, Shirabe K, Shimada M, et al. Bile leakage after hepatic resection. Ann Surg 2001;233:45-50.

5 Tsao JI, Loftus JP, Nagorney DM, Adson MA, Ilstrup DM. Trends in morbidity and mortality of hepatic resection for malignancy. A matched comparative analysis. Ann Surg 1994;220: 199-205.

6 Capussotti L, Ferrero A, Viganò L, Sgotto E, Muratore A, Polastri R. Bile leakage and liver resection: Where is the risk? Arch Surg 2006;141:690-695.

7 Tanaka S, Hirohashi K, Tanaka H, Shuto T, Lee SH, Kubo S, et al. Incidence and management of bile leakage after hepatic resection for malignant hepatic tumors. J Am Coll Surg 2002;195:484-489.

8 Nagano Y, Togo S, Tanaka K, Masui H, Endo I, Sekido H, et al. Risk factors and management of bile leakage after hepatic resection. World J Surg 2003;27:695-698.

9 Ijichi M, Takayama T, Toyoda H, Sano K, Kubota K, Makuuchi M. Randomized trial of the usefulness of a bile leakage test during hepatic resection. Arch Surg 2000;135:1395-1400.

10 Lam CM, Lo CM, Liu CL, Fan ST. Biliary complications during liver resection. World J Surg 2001;25:1273-1276.

11 Nakai T, Koh K, Funai S, Kawabe T, Okuno K, Yasutomi M. Comparison of controlled and Glisson’s pedicle transections of hepatic hilum occlusion for hepatic resection. J Am Coll Surg 1999;189:300-304.

12 Lo CM, Fan ST, Liu CL, Lai EC, Wong J. Biliary complications after hepatic resection: risk factors, management, and outcome. Arch Surg 1998;133:156-161.

13 Noun R, Elias D, Balladur P, Bismuth H, Parc R, Lasser P, et al. Fibrin glue effectiveness and tolerance after elective liver resection: a randomized trial. Hepatogastroenterology 1996;43:221-224.

14 Norton L, Moore G, Eiseman B. Liver failure in the postoperative patient: the role of sepsis and immunologic deficiency. Surgery 1975;78:6-13.

15 Yanaga K, Kanematsu T, Takenaka K, Sugimachi K. Intraperitoneal septic complications after hepatectomy. Ann Surg 1986;203:148-152.

16 Pace RF, Blenkharn JI, Edwards WJ, Orloff M, Blumgart LH, Benjamin IS. Intra-abdominal sepsis after hepatic resection. Ann Surg 1989;209:302-306.

17 Davids PH, Rauws EA, Tytgat GN, Huibregtse K. Postoperative bile leakage: endoscopic management. Gut 1992;33:1118-1122.

18 Sherman S, Shaked A, Cryer HM, Goldstein LI, Busuttil RW. Endoscopic management of biliary fistulas complicating liver transplantation and other hepatobiliary operations. Ann Surg 1993;218:167-175.

19 Guillaud A, Pery C, Campillo B, Lourdais A, Sulpice L, Boudjema K. Incidence and predictive factors of clinically relevant bile leakage in the modern era of liver resections. HPB (Oxford) 2013;15:224-229.

20 Erdogan D, Busch OR, van Delden OM, Rauws EA, Gouma DJ, van Gulik TM. Incidence and management of bile leakage after partial liver resection. Dig Surg 2008;25:60-66.

21 Tsim N, Healey AJ, Frampton AE, Habib NA, Bansi DS, Wasan H, et al. Two-stage resection for bilobar colorectal liver metastases: R0 resection is the key. Ann Surg Oncol 2011;18:1939-1946.

22 Zimmitti G, Vauthey JN, Shindoh J, Tzeng CW, Roses RE, Ribero D, et al. Systematic use of an intraoperative air leak test at the time of major liver resection reduces the rate of postoperative biliary complications. J Am Coll Surg 2013;217:1028-1037.

23 Strücker B, Stockmann M, Denecke T, Neuhaus P, Seehofer D. Intraoperative placement of external biliary drains for prevention and treatment of bile leaks after extended liver resection without bilioenteric anastomosis. World J Surg 2013;37:2629-2634.

24 Liu Z, Jin H, Li Y, Gu Y, Zhai C. Randomized controlled trial of the intraoperative bile leakage test in preventing bile leakage after hepatic resection. Dig Surg 2012;29:510-515.

25 Zimmitti G, Roses RE, Andreou A, Shindoh J, Curley SA, Aloia TA, et al. Greater complexity of liver surgery is not associated with an increased incidence of liver-related complications except for bile leak: an experience with 2,628 consecutive resections. J Gastrointest Surg 2013;17:57-65.

Received March 16, 2015

Accepted after revision July 14, 2015

Original Article / Biliary

doi:10.1016/S1499-3872(15)60424-6

Corresponding Author:Fabrizio Panaro, MD, PhD, Department of Surgery, Division of Transplantation, Montpellier University Hospital, College of Medicine, 80 Avenue Augustin Fliche 34295, Montpellier-Cedex 5, France (Tel: +33-0467336733; Fax: +33-0467337623; Email: f-panaro@ chu-montpellier.fr)