Telephone: 214-820-2050; FAX: 214-820-4527
Twenty years of follow-up of aortohepatic conduits in liver transplantation†
Article first published online: 29 SEP 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 14, Issue 10, pages 1486–1490, October 2008
How to Cite
Nikitin, D., Jennings, L. W., Khan, T., Sanchez, E. Q., Chinnakotla, S., Randall, H. B., McKenna, G. J., Goldstein, R. M., Levy, M. F. and Klintmalm, G. B. (2008), Twenty years of follow-up of aortohepatic conduits in liver transplantation. Liver Transpl, 14: 1486–1490. doi: 10.1002/lt.21575
See Editorial on Page 1412
- Issue published online: 29 SEP 2008
- Article first published online: 29 SEP 2008
- Manuscript Accepted: 15 MAY 2008
- Manuscript Received: 19 JAN 2008
Arterial problems remain a formidable challenge in liver transplantation. In many situations, an aortohepatic conduit can provide a solution. No long-term results (over 5 years) have been reported. This study was designed to assess the impact of aortohepatic conduits on graft survival after liver transplantation and the safety of aortohepatic conduits and to establish the long-term results (up to 20 years) of aortohepatic conduits. Data from 2346 adult liver transplants were prospectively collected into the computerized database and analyzed. In the majority of cases, arterial conduits were constructed from the donor iliac artery obtained at the liver retrieval. Aortohepatic conduits were required in 149 (6.4%) first transplants. The long-term graft survival after liver transplantation using aortohepatic conduits was excellent and comparable to that of the control group. The graft survival was 59% with the conduit versus 67% without the conduit at 5 years of follow-up, 50% versus 52% at 10 years, and 33% versus 35% at 15 years. With up to 20 years of follow-up, there was no statistically significant difference in graft survival, patient survival, hepatic artery complications, or biliary complications. For the same time period, there was no statistically significant difference in graft survival or patient survival for the retransplants with and without aortohepatic conduits. In conclusion, in experienced hands, aortohepatic conduits can be used safely for liver transplantation with no negative impact on long-term graft survival, patient survival, hepatic artery complications, or biliary complications. Excellent long-term results can be obtained. Liver Transpl 14:1486–1490, 2008. © 2008 AASLD.
The aortohepatic conduit is an important tool in the armamentarium of the liver transplant surgeon. This technique has been used more frequently in more complex cases. There is considerable controversy in the assessment of the safety and indications of aortohepatic conduits. No long-term results (over 5 years) have been reported.
Cases likely to need the conduit include patients with celiac trunk stenosis or occlusion, arcuate ligament syndrome, severe atherosclerotic disease (more common in older donors and older recipients), retransplants, split liver, auxiliary liver, or living-donor transplants, hepatic artery intimal dissection, stenosis or thrombosis, or hemodynamic instability.
The goal of this study was to assess the impact of aortohepatic conduits on graft survival after liver transplantation and the safety of aortohepatic conduits in liver transplantation and to establish the long-term results (up to 20 years) of aortohepatic conduits.
PATIENTS AND METHODS
Data from 2346 adult liver transplants performed at the Baylor Regional Transplant Institute were prospectively collected into a computerized database and analyzed. Aortohepatic conduits placed at the time of the transplant were analyzed separately from retransplants. This study was approved by the Baylor University Medical Center Institutional Review Board.
The aortohepatic conduit technique has been described by our group in previous publications.1 In the overwhelming majority of cases, arterial conduits are constructed from the donor iliac artery obtained during liver procurement and preserved in preservation solution. The conduits are preferentially anastomosed to the infrarenal aorta except when retroperitoneal varices make the dissection dangerous, extensive adhesions prevent safe dissection, or the aorta is severely calcified. In these instances, the conduit is anastomosed to the iliac artery. In most cases, an antepancreatic approach is used. We reflect the colon cranially and reflect the small bowel to the right. Then, at the root of the small bowel mesentery, we identify and expose the infrarenal aorta. A self-retaining retractor is used for exposure. A side biting clamp is applied to the aorta to partially occlude it. The aorta is opened with a #11 blade; 5-0 running prolene is used to create the anastomosis between the conduit and the aorta. Self-retaining clamps are applied to the conduit, and the aortic clamp is removed from the aorta. Any leaks are controlled. Then, the window is made in the mesocolon and behind the stomach to pass the conduit anterior to the pancreas, but posterior to the stomach and the colon, the duodenum stays to the right of the conduit. Then, the anastomosis is performed between the conduit and the donor hepatic artery or celiac trunk or aortic patch. After completion of the anastomosis, the clamps are removed, and the arterial flow in the liver is restored. The flow in the hepatic artery is measured by an electromagnetic method. Currently, for an adult liver transplant, a flow of 400 mL/minute or more is considered acceptable.2–4
The patency of the hepatic artery is assessed by ultrasound on postoperative day 1, later at 1 year, 2 years, 5 years, 10 years, 15 years, and 20 years, and when clinically indicated. If there is a suggestion of compromised arterial flow, then an angiogram is used to further delineate the arterial inflow to the liver. If there is hepatic artery stenosis, it is surgically corrected. We do not perform repeated percutaneous interventional arterial procedures and prefer to correct the arterial inflow surgically. We have not seen any pseudoaneurysms. Kinks are included in the category of stenosis and are corrected surgically as well. We have not observed stenosis at the level of inflow into the conduit.
Continuous variables have been compared with the 2-sample Wilcoxon test. Categorical variables have been compared with the 2-tailed Fisher's exact test for 2-by-2 tables and with the likelihood ratio chi-square test for larger tables. Actuarial survival has been estimated by the Kaplan-Meier method with significance testing with the log-rank test. Results with a P value less than 0.05 have been considered statistically significant. All analysis has been performed with SAS 9.1.3.
The long-term graft survival after first liver transplantation using aortohepatic conduits was excellent and comparable to that of the control group (Fig. 1 and Table 1). Graft survival was 72% (102) with a conduit versus 82% (1758) without a conduit at 1 year of follow-up, 65% (75) versus 74% (1416) at 3 years, 59% (64) versus 67% (1092) at 5 years, 50% (40) versus 52% (545) at 10 years, 33% (18) versus 35% (158) at 15 years, and 24% (2) versus 25% (16) at 20 years. There was no statistically significant difference in graft survival, patient survival, hepatic artery complications, or biliary complications (up to 20 years of follow-up; Table 2). As indicated in Table 2, hepatic artery stenosis was observed in 3.4% of patients with a conduit and 5.2% of patients without a conduit. The difference was not statistically significant (P = 0.44). Hepatic artery thrombosis was observed in 4.7% of patients with a conduit and in 2.9% of patients without a conduit. The difference was not statistically significant (P = 0.21). We did not see any pseudoaneurysms. Kinks were included in the category of stenosis and were corrected surgically as well. We did not observe stenosis at the level of inflow into the conduit. No embolic complications were noted.
|n||1 Year||3 Years||5 Years||10 Years||15 Years||20 Years|
|Parameter||Conduit [n (%)]||No Conduit [n (%)]||P Value|
|Hepatic artery stenosis||5 (3.36)||114 (5.19)||0.4391|
|Hepatic artery thrombosis||7 (4.70)||63 (2.87)||0.2080|
|Pancreatitis||3 (2.01)||22 (1.00)||0.2092|
|Abdominal ileus||6 (4.03)||27 (1.23)||0.0157|
|Biliary stricture||4 (2.68)||110 (5.01)||0.2410|
|Biliary: anastomotic leak||10 (6.71)||94 (4.28)||0.1525|
|Respiratory failure||16 (10.74)||107 (4.87)||0.0064|
|Renal failure||11 (7.38)||54 (2.46)||0.0021|
Aortohepatic conduits were performed at the time of first liver transplant in 149 patients (6.4%).
An aortohepatic conduit was more frequently used in female recipients (8.0% versus 5.1%, P = 0.0047; Table 3). No statistically significant difference was found between groups with and without aortohepatic conduits in donor age, donor body mass index, donor days in intensive care unit, or donor liver function tests (Table 4). Recipients who received conduits had a smaller body mass index (median, 25.6 versus 26.5, P = 0.05), a higher Model for End-Stage Liver Disease score (median, 18 versus 16, P = 0.01), and a higher percentage of fulminant hepatic failure as the preoperative diagnosis (8.7 versus 3.8, P = 0.009). It is a policy at our institution to use conduits in hemodynamically unstable patients as a conduit is not subject to vasoconstriction, thus securing the arterial inflow.
|Parameter||Total||Conduit||No Conduit||P Value|
|Donor sex: female||903||66||7.3||837||92.7||0.1381|
|Donor sex: male||1433||82||5.7||1351||94.3|
|Donor race: white||1869||125||6.7||1744||93.3||0.2073|
|Donor race: other||477||24||5.0||453||95.0|
|Donor cardiac arrest: yes||107||12||11.2||95||98.8||0.0699|
|Donor cardiac arrest: no||2167||133||6.1||2034||93.9|
|Recipient sex: male||1334||68||5.1||1266||94.9||0.0047|
|Recipient sex: female||1012||81||8.0||931||92.0|
|Parameter||Conduit||No Conduit||P Value|
|Donor age (years)||148||33.50||2187||35.00||0.9947|
|Donor BMI (kg/m2)||125||24.58||2019||24.24||0.3040|
|Donor days in ICU||145||2.00||2144||2.00||0.4370|
|Recipient age (years)||149||50.00||2197||50.00||0.3483|
|Recipient BMI (kg/m2)||147||25.64||2176||26.51||0.0486|
|Cold ischemia time (hours)||140||8.37||2063||8.60||0.7446|
|Warm ischemia time (hours)||115||0.87||1615||0.90||0.2287|
|OR time (hours)||124||7.57||1936||6.00||<0.0001|
|PRBC: intraoperative total (units)||88||5.00||1647||4.00||0.0049|
The use of aortohepatic conduits was more frequently associated with longer operative room time (7.57 versus 6.0 hours, P = 0.0001), more units of packed red blood cells transfused during transplant (5.0 versus 4.0, P = 0.0049), and more units of fresh frozen plasma required at operation (6.0 versus 5.0, P = 0.0082).
In the postoperative period, the patients with conduits were more likely to develop an ileus (4% versus 1%, P = 0.016), respiratory failure (11% versus 5%, P = 0.006), and renal failure (7% versus 2%, P = 0.002). The incidence of postoperative renal failure was higher with the conduits. We do not perform the supraceliac aortohepatic conduit; therefore, we do not have data to compare the incidence of postoperative renal failure between the supraceliac and infrarenal conduits.
A separate analysis was conducted for retransplants. Two hundred twenty-nine livers were retransplanted. Conduits were used in 100 transplants (44%). Graft survival was 61% (61) with a conduit versus 64% (82) without a conduit at 1 year of follow-up, 54% (47) versus 52% (60) at 3 years, 48% (33) versus 47% (52) at 5 years, 34% (16) versus 36% (16) at 10 years, 23% (6) versus 27% (15) at 15 years, and 23% (2) versus 22% (4) at 20 years (Table 5). There was no statistically significant difference in graft survival (P = 0.91) or patient survival (P = 0.70; Fig. 2).
|n||1 Year||3 Years||5 Years||10 Years||15 Years||20 Years|
The utilization of aortohepatic conduits increased from 2.8% in 1997-1998 to 10.1% in 2005-2006.
Our study shows that liver transplantation using an aortohepatic conduit has long-term results similar to those of standard liver transplantation. Prior reports of poor results after liver transplantation using aortohepatic conduits likely reflected additional factors (retransplants, placement of the conduit after the hepatic artery thrombosis, and patients with a hypercoagulable state).5–8 When the influence of the confounding factors is minimized, the results of liver transplantation using aortohepatic conduits become very close to those of the control group. The trend of lower graft survival in the first 5 years and the higher incidence of postoperative renal failure may be related to a sicker recipient population (a higher Model for End-Stage Liver Disease score, more patients with fulminant hepatic failure, more hemodynamically unstable patients, and a higher rate of respiratory failure). Excellent long-term results prove the longevity of aortohepatic conduits using the iliac artery of the donor.
Thrombosis of the hepatic artery and the conduit remains the major problem.9–13 We believe that meticulous surgical technique, aggressive surveillance, and early intervention are very important in preventing and treating this complication. We also by policy place such recipients on salicylic acid (80 mg once daily) as prophylaxis.
This study is important because this is the first report in the literature on the long-term results (up to 20 years) of aortohepatic conduits. It is based on extensive information from a single center, and this provides consistency in management and technique. The results of this study may also be applicable to other fields of surgery (the use of biological conduits for vascular reconstruction).
Our study has a number of limitations. It is a retrospective analysis of a prospectively maintained database and is subject to selection bias, observer bias, and bias of the confounding factors.
In conclusion, in experienced hands, aortohepatic conduits can be used safely for liver transplantation with no negative impact on long-term graft survival, patient survival, hepatic artery complications, or biliary complications. Excellent long-term results can be obtained.