Management of biliary complications following living donor liver transplantation—a single center experience
- 101 Downloads
Biliary complications (BC) after living donor liver transplantation (LDLT) are reported in up to 32%. We retrospectively reviewed the biliary reconstruction after 95 LDLT.
Eighty-one right hemiliver grafts and 14 left lateral section grafts were transplanted. Bile duct anastomoses were performed as duct-to-duct (DD) or bilioenteric anastomosis (RYHJ); multiple bile ducts were anastomosed using a ductoplasty or multiple direct bile duct anastomoses.
After right hemiliver LDLT, a total of 45.5% of BC was observed, with an incidence of 27.7% in case of DD anastomosis and 18.8% in case of RYHJ. After DD anastomosis, strictures were successfully treated endoscopically in 50%; insufficiencies mainly required reoperations.
BC still account for a high percentage of morbidity and mortality after LDLT. DD anastomoses are performed more frequently and are feasible in cases with simple biliary anatomy; RYHJ is the gold standard for the reconstruction of multiple bile ducts.
KeywordsBiliary complications Living donor liver transplantation Management of BC Liver transplantation
Liver transplantation is an established, safe, and definitive therapeutic option for the treatment of end-stage liver diseases. Organ shortage is the limiting factor preventing physicians from offering this life-saving procedure to more patients in need. Since the introduction of the MELD score, the rate of patients dying while waiting for an organ has significantly decreased . Nevertheless, up to 10% of patients die while waiting for a suitable liver graft. Living donor liver transplantation (LDLT) has become a promising option to ameliorate the effects of organ shortage. In July 1989, the first successful LDLT was performed in an 11-month-old child using a left lateral section donated by its mother . The left lateral section liver has been used since 1996 for adult liver transplantation . Increasing experience leads to the transplantation of right hemiliver, which in our days is the most commonly used procedure in adult LDLT [4, 5]. Beside the known risk for the donor, complications in the recipient are frequently observed due to the technical challenge of the whole procedure. The biliary anastomosis is technically the most challenging surgical step as well as for deceased and especially living donor transplantation. Complication rates for this anastomosis are reported up to 20% for deceased full organ liver transplantations [6, 7, 8]. Due to various factors (e.g., frequent variations of the bile duct anatomy and the frequent occurrence (up to 60%) of multiple biliary orifices of the right hemiliver [9, 10, 11]), the incidence of biliary complications (BCs) in LDLT is significantly higher  and frequently underestimated in the past. BCs include anastomotic complications (e.g., insufficiencies and strictures) as well as complications of the dissection of the parenchyma at the cutting surface that result in bile leakage and bilioma. The incidence of anastomotic leakage and stricture after right hemiliver LDLT are reported up to 32%, respectively . Several surgical approaches to performing bile duct anastomoses in LDLT are available. In left lateral section LDLT, a Roux en Y hepaticojejunostomy (RYHJ) is usually preferred. Using the right hemiliver for LDLT, a single duct-to-duct choledochocholedochostomy (DD) is performed with a single bile duct or in cases of closely located multiple biliary orifices after ductoplasty. If there are two or three biliary orifices distantly located, biliary reconstruction is performed predominantly with two separate DD anastomoses using the left and right recipient bile ducts, cystic bile duct, or a combination of DD and RYHJ .
We retrospectively reviewed data from 95 LDLT procedures at our center and report our management of postoperative biliary complications.
Patients and methods
Between December 1999 and February 2005, 95 living donor liver transplantations (LDLTs) were performed in 94 patients. The median age was 48 years (2 months–70 years), and the male to female ratio was 60:34. A total of 81 right hemilivers with segments IVa, V, VI, VII, and VIII and 14 left lateral sections with segments II and III were transplanted. Beside a general health check, donor evaluation included the evaluation of donor liver morphology and anatomy with a CT scan with a virtual reconstruction of the vessel anatomy. To get more detailed information about the biliary and vascular anatomy of the donor, an MRCP was performed. With these non-invasive diagnostic tools, liver volume, percentage of steatosis, and fibrotic changes were assessed. In our series, the calculated liver volume differed from the intraoperatively measured liver volume by between 3.9% and 12.5% .
Immediately after laparotomy, an intraoperative cholangiography using the cystic duct was performed to confirm the preoperatively assessed number of bile ducts. After splitting the liver tissue and explantation, the graft was perfused with HTK.
Right hemiliver transplantation, reconstruction with direct duct-to-duct anastomosis, patients with biliary reconstruction including three orifices, two separate DD anastomosis, or a combination of DD and RYHJ are summarized as complex reconstructions
n = 37
n = 17
n = 3
Two duct-to-duct anastomoses
n = 5
Two duct-to-duct anastomoses including cystic bile duct
n = 2
Duct-to-duct and bilioenteric anastomosis
n = 1
n = 65
Anastomotic BC all appeared in different patients. If there were more types of anastomotic BC in one patient, the first one was statistically evaluated and data are given in the text.
Standard immunosuppressive therapy after transplantation consisted of daclizumab, methyprednisolone, and tacrolimus. Data were assessed in regard to bile duct complications, including leakages and strictures, t-tube associated problems, and ischemic type biliary lesions (i.e., non-anastomotic strictures). Additionally, biliomas and leakages in the area of the cutting surface were documented.
All patients provided written consent that their data and pictures could be used in clinical research.
Single DD anastomoses were performed in 57 patients, with one, two, and three orifices in 37, 17, and three patients, respectively. Two separate bile duct anastomoses were performed in eight patients; the right and left hepatic bile ducts of the recipient were used in five patients, the hepatic bile duct and cystic bile duct were used in two patients, and a DD anastomosis combined with a RYHJ was used in one patient (Table 1). These eight patients and the three patients with three orifices were summarized as complex reconstructions. All patients routinely received a t-tube (Table 1).
Right hemiliver transplantation, reconstruction with bilioenteric anastomosis
n = 9
n = 6
n = 1
n = 16
RYHJ was also performed in all patients after left lateral section LDLT. No BCs occurred in these 14 patients.
Treatment of patients with biliary complications after right hemiliver LDLT
Endoscopy and reoperation
CT drain and endoscopy
Complication after duct-to-duct anastomosis
Stenosis of anastomosis (n = 4)
Insufficiency of anastomosis (n = 9)
Bilioma (n = 3)
Insufficiency of t-tube (n = 2)
Complication after bilioenteric anastomosis
Stenosis of anastomosis (n = 1)
Insufficieny of anastomosis (n = 1)
Bilioma (n = 1)
Anastomotic complications in patients with DD anastomosis and one orifice after right hemiliver LDLT occurred in five patients, including two cases of anastomotic stricture and three cases of anastomotic leaks. Five patients with a DD anastomosis and two orifices developed complications at the biliary anastomosis (three leakages and two strictures). After complex reconstruction, three patients experienced anastomotic leakages but no anastomotic strictures occurred (Fig. 1).
Strictures of DD anastomosis could be sufficiently treated with endoscopy in 50% of cases (n = 2). The other two of these patients had to be reoperated on after endoscopic treatment failure due to consecutive leaks or insufficient endoscopic treatment. One patient died after reoperation.
Anastomotic insufficiencies could be treated in one patient by endoscopy, three patients underwent direct relaparotomy and in four of the nine patients endoscopic treatment was combined with surgery. One patient was retransplanted due to primary graft non-function. In six patients, BCs resulted in the need for long-term endoscopic treatment.
Three patients with bile duct insufficiency died while hospitalized; two died due to biliary peritonitis and consecutive MOF, and one died due to gastrointestinal bleeding.
Three patients (18.8%) developed a BC after right hemiliver LDLT with bilioenteric anastomosis. These included one patient each with an anastomotic stricture, leakage, and bilioma at the cutting surface.
Approximately 4.9% (n = 4) of all patients after right hemiliver LDLT developed a bilioma at the cutting surface. All these patients were treated successfully with CT-guided drainage and an ERC with a papillotomy.
Two patients (3.1%) showed t-tube-related complications. One was treated conservatively, while the other was treated with an ERC and stenting.
We compared the occurrence of BCs between patients with reconstruction with different numbers of orifices after LDLT and DD anastomosis. Although there was a trend towards fewer BCs in cases of one orifice, there were no significant differences between one and two orifices or complex reconstructions.
The incidence of bile duct stricture after right hemiliver LDLT is reported to be as high as 32%, and the incidence of biliary leakage is reported to be between 5% and 18% . In our series, we observed biliary strictures after right hemiliver LDLT in 6.2% and leakages in 12.3% of patients. Thus, the total incidence of strictures in our series is less than that in the literature, and the incidence of leakages is within the reported ranges.
With regard to the different bile duct anastomoses (DD and RYHJ), BCs after DD reconstructions are reported to occur in 24–34% of cases. Leakages have an incidence ranging from 4.7% to 7.3%, and strictures occur in between 24% and 31.7% of patients [10, 14]. In our experience, bile leaks after DD anastomoses occurred in 13.8% and strictures in 6.2% of patients. These numbers reflect our low incidence of strictures, which can cause considerable long-term morbidity and might result in retransplantation. The relatively high number of diagnosed biliary leakages might be a result of our standard procedure, which involves routine bile duct drainage and control X-rays 5 days after transplantation even if patients have had an uneventful clinical course . As a result of our routine clinical procedures (i.e., performance of a routine cholangiogram via a t-tube on postoperative days 5–7), leakages might be over-diagnosed in comparison to the published results.
After RYHJ and right hemiliver LDLT, published values show that leakages occur in 12–18% and strictures in 8–16% of cases [9, 10, 16]. Our series shows leaks and strictures each occurring in 6.3% of patients. Once again, RYHJ leakages are within the published ranges and strictures are below ranges reported in the literature.
In our series, RYHJs were performed via interrupted sutures with PDS 5-0 to 7-0 (depending on the size of the bile duct). In all cases, a transanastomotic drainage system was used. This drainage system was not removed earlier than 6 weeks after the operation. Other series have described the technique of RYHJ inconsistently, with interrupted or continuous sutures as well as the presence or absence of biliary drainage depending on surgeon preference. The consequent training and use of interrupted sutures with the use of biliary drainage may explain the relatively low rate of strictures in our series.
Due to the high anatomical variability of the right hemihepatic graft, over 60% of transplantations have more than one bile duct that must be anastomosed [17, 18]. The anastomosis was performed either as a ductoplasty (when bile ducts were close to each other), multiple single anastomoses (including the left and right hepatic bile ducts), or combination of DD and RYHJ . Whether there is a correlation between the presence of multiple bile ducts and the occurrence of BCs is not yet clear [19, 20]. However, several reports suggest the increased incidence of complications in cases with multiple bile ducts [9, 17]. In our results, there is an increased incidence of BCs both in cases of multiple bile duct orifices as well as after complex biliary reconstructions. In case of DD anastomosis and one orifice, five patients out of 37 developed a BC, with two orifices five out of 17 patients, and after complex reconstruction three out of 11 patients developed a BC. However, these differences were not statistically significant.
Due to the increased incidence of BCs in cases of multiple bile ducts that require complex reconstructions, it seems advisable to perform biliary anastomosis in these instances with RYHJ. RYHJ has the disadvantages of a longer operative time, the loss of physiological bilioenteric continuity, and a delay in the return of gastrointestinal function. In contrast, DD anastomosis has the advantages of preserving the function of the sphincter Oddi as well as physiological bilioenteric continuity . In addition, the access to DD anastomosis is easier to perform via ERC than RYHJ, which requires a PTC when intervention is necessary. However, RYHJ produces fewer BCs than DD anastomosis without compromising the long-term outcome .
Two t-tube-related complications were observed in our series. This is in obvious contrast to earlier published results regarding the usage of t-tubes after deceased donor transplantation . These complications were minor, and all were treated conservatively or by endoscopy without further intervention. In our hands, the t-tube is an excellent tool for diagnosing and monitoring BCs after transplantation. We also note the potential reduction of the postoperative biliary pressure on the anastomotic region. Operative denervation of the sphincter Oddi is reported to cause ampullary dysfunction in 2–5% of patients after OLT and can cause papillar stenosis. Endoscopic treatment (e.g., transpapillary stenting or endoscopic sphincterotomy) should be considered as an initial approach after diagnosis of the dysfunction .
Biliary leakages with symptomatic biliomas due to parenchymal fistulas can generally be managed successfully by percutaneous drainages . In our series, biliomas were no serious problem and could be managed routinely by percutaneous drainages without surgical intervention. In contrast, anastomotic BCs in our series required either reoperations or endoscopic interventions. These findings are in accordance with previously published results .
Anastomotic BCs contribute significantly to postoperative morbidity and mortality. Our first therapeutic approach to manage anastomotic strictures was endoscopy, which was successful over the long term in 50% of patients. On the other hand, surgical reintervention was associated with high mortality and morbidity rates. One patient died 5 months after transplantation, and a second was reoperated 1 year after transplantation due to persistent strictures.
In cases of insufficiencies, the endoscopic approach was applied in five out of nine cases; however, it was successful only in one case. Seven of these patients were reoperated on; four patients received a RYHJ, and in three the bile duct was reanastomosed. Three of the patients suffering of insufficiencies died in the first 2 months after transplantation. In three patients, anastomotic leakage produced secondary anastomotic stricture that resulted in long-term endoscopic stenting. Reoperations in case of insufficiencies are associated with a high rate of complications due to the intraoperative findings, which often show a necrotic bile duct, where the reconstruction requires reanastomosis or a RYHJ. The first therapeutic approach should be endoscopically, although realizing that insufficiencies need more reoperations than strictures and contribute themselves to a higher rate of morbidity and mortality. Anastomotic BCs after RYHJ occurred in only two patients and could be treated conservatively.
The overall 3-year survival rates of all the patients with right hemiliver LDLT are shown in Fig. 3. BC after RYHJ were minor and consisted of one patient with cholestasis and one with insufficiency, which both could be treated conservatively. One patient had a leak of a solitary bile duct of segment VII, which became atrophic. So BCs did not contribute to mortality in case of RYHJ, and survival was 100% after 3 years in these three patients.
BC after DD anastomosis did contribute to the mortality, and survival rates were lower in case of BC compared to the group without BC, however not reaching statistical significance.
These findings are in accordance with previously published data, in which RYHJ was associated with fewer BCs. This might be due to the fact that the blood supply of the bile duct in a right hemiliver graft comes only from the liver side. RYHJ is more beneficial than DD from the viewpoint of arterial collateral formation to the graft duct stump. This may permit a better blood supply for the anastomosis .
ITBL after LDLT has not been reported previously. We report one case in our series in which this occurred. Known risk factors for developing ITBL include hypoxia, I/R related injury, immunological factors, and cytotoxic injury induced by bile acids [26, 27]. In general, the cold ischemic time in LDLT is reduced to a minimum. Nevertheless, I/R injury contributes to damage at the bile duct. Additionally, immunological factors (e.g., ABO incompatibility, autoimmune hepatitis, and chronic rejection) may contribute to the development of ITBL in the setting of LDLT. In our case, no singular risk factor could be identified despite extensive diagnostic evaluation. We hypothesize that the summation of various risk factors contributed to the development of ITBL in this patient. However, ITBL is a multifactorial process that may rarely occur after LDLT as reported here.
BCs after LDLT still account for a high percentage of morbidity and mortality. Due to improvements in surgical techniques in recent years, DD anastomoses have been performed increasingly more frequently. Some recent reports suggest that DD anastomosis is feasible in cases with simple biliary anatomy, but RYHJ remains the gold standard for the reconstruction of multiple bile ducts due to its low incidence of subsequent BCs.
- 21.Liu CL, Lo CM, Fan ST (2005) What is the best technique for right hemiliver living donor liver transplantation? With or without the middle hepatic vein? Duct-to-duct biliary anastomosis or Roux-en-Y hepaticojejunostomy? J Hepatol 43(1):17–22. doi:10.1016/j.jhep.2005.05.003 CrossRefPubMedGoogle Scholar
- 27.Heidenhain C, Heise M, Jonas S et al (2006) Retrograde reperfusion via vena cava lowers the risk of initial nonfunction but increases the risk of ischemic-type biliary lesions in liver transplantation—a randomized clinical trial. Transpl Int 19(9):738–748. doi:10.1111/j.1432-2277.2006.00347.x CrossRefPubMedGoogle Scholar