LIVER TRANSPLANTATION 21:1051–1055, 2015
ORIGINAL ARTICLE
Technical Options for Outflow Reconstruction in Domino Liver Transplantation: A Single European Center Experience Sofıa De la Serna, Laura Llado, Emilio Ramos, Joan Fabregat, Carme Baliellas, Juli Busquets, Lluis Secanella, Nuria Pelaez, Jaume Torras, and Antoni Rafecas Biome`dica de Bellvitge, Liver Transplantation Unit, Hospital Universitari de Bellvitge, Institut d’Investigacio Barcelona, Spain
Venous outflow is critical to the success of liver transplantation (LT). In domino liver transplantation (DLT), the venous cuffs should be shared between the donor and the recipient, and the length can be compromised. The aim of this study was to describe and compare the technical options for outflow reconstruction used at our institution. This was a retrospective analysis of 39 consecutive DLT recipients between January 1997 and May 2013. Twenty-seven men and 12 women (mean age, 61.8 6 4.3 years) underwent LT and consented to receive a liver from a donor with familial amyloid polyneuropathy (FAP). The main indications were hepatocellular carcinoma and hepatitis C virus cirrhosis. All recipients underwent transplantation by a piggyback technique. Liver procurement in the FAP donors was performed with the classic technique in 22 patients and with the piggyback technique in the last 17. In these latter cases, for vascular outflow reconstruction, a cadaveric venous graft was interposed between the hepatic vein (HV) stump of the FAP liver and the recipient HV in 11 cases (28%). Since 2011, we have employed arterial grafts to be interposed between the vessels stumps: a tailored arterial graft in 5 patients and an aortic graft in 1 case. There was no postoperative mortality. Arterial and portal complications presented in 2 (5.1) and 4 patients (10.3), respectively. Postoperative outflow complications (post-LT subacute Budd-Chiari syndrome) occurred in 4 patients, and all of them had received a venous interposed graft for reconstruction. The incidence of outflow complications tended to be higher among patients with venous grafts than those with arterial graft interposition. Overall patient survival at 1, 3, 5, and 10 years was 97%, 79%, respectively. Arterial grafts constitute a feasible and safe option for vascular outflow reconstruction in DLT because they are associated with a relatively low incidence of complications. The recently proposed Bellvitge arterial graft technique should be added to the current range of available surgical modalities. C 2015 AASLD. Liver Transpl 21:1051-1055, 2015. V Received December 23, 2014; accepted March 30, 2015. In liver transplantation (LT), ensuring optimal venous outflow is critical to avoiding graft congestion and post-LT Budd-Chiari syndrome (BCS). BCS has a reported incidence in transplant patients of 1% to 7% and is associated with mortality rates up to 24%; therefore, it represents a significant and relatively common complication.1,2 Domino liver transplantation (DLT), also known as sequential LT, is a wellknown technique useful for expanding the number of
potential donors. Although its safety has already been established, special issues remain to be addressed. The main sources of DLT donors are patients with familial amyloid polyneuropathy (FAP). Traditionally, liver procurement in FAP donors has been performed with the classic technique of caval replacement. However, the increased hemodynamic lability of FAP patients has led many centers to progressively replace the classic technique with the double piggyback
Abbreviations: BCS, Budd-Chiari syndrome; DLT, domino liver transplantation; FAP, familial amyloid polyneuropathy; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HV, hepatic vein; IVC, inferior vena cava; LT, liver transplantation; MELD, Model for End-Stage Liver Disease. Potential conflict of interest: Nothing to report. Address reprint requests to Laura Llado, M.D., Ph.D., Liver Transplantation Unit, Hospital Universitari de Bellvitge, Feixa Llarga s/n, 08907 Barcelona, Spain. Telephone: 134-93-260-7940; FAX: 134-93-260-7603; E-mail: [email protected] DOI 10.1002/lt.24143 View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
C 2015 American Association for the Study of Liver Diseases. V
1052 DE LA SERNA ET AL.
technique in both the donor and the recipient.3,4 Nevertheless, venous reconstruction remains a major challenge of DLT, especially when the double piggyback technique is employed. The need to share the vascular stumps between the graft and the donor might compromise the length of the vessels and render the outflow reconstruction extremely laborious. Direct end-to-end anastomosis between the hepatic veins (HVs) of the graft and the recipient have also been associated with technical difficulties,5 and therefore, septoplasties or bridge venoplasties between the HV stumps in the FAP graft are advisable whenever possible.6-8 To overcome the problem of short cuffs, interposed grafts are often required to obtain a suitable operative field for the reconstruction, for which several venous graft options have been proposed in the literature.1,915 To this end, our group has recently proposed a novel technique using arterial grafts to create a tailored patch that can be anastomosed between the HV stumps of the FAP graft and the recipient.16 The aim of this study was to review the technical options for outflow reconstruction in DLT in a European single center and to analyze the short- and longterm results.
PATIENTS AND METHODS Study Design This was a retrospective study of 39 consecutive recipients of DLT performed at Hospital Universitari de Bellvitge (Barcelona, Spain) from January 1997 to May 2013. We performed LT 974 times at our institution during the study period. Thirty-nine patients (4%) required LT for FAP disease. Since 1999, when the DLT program was implemented at our institution, 33 patients with FAP have consented to donate their livers for transplantation. Among these 33 potential FAP donors, 2 were rejected preoperatively: one because of a history of breast cancer 7 years earlier (although the patient was free of tumor recurrence) and the other because of advanced age (67 years old) and the presence of liver fibrosis confirmed by elastography. In 1 patient, the donor liver was discarded intraoperatively because of a prolonged warm ischemia time. Another 9 FAP livers were shipped from other hospitals, and this produced a final case series of 39. The selection criteria for recipients of FAP livers were an age equal or greater than 60 years and hepatocellular carcinoma (HCC) within the Milan criteria.17 Patients less than 60 years old were considered if particular criteria were met.
Data Analysis Data were obtained from our prospectively collected LT database and retrospectively analyzed after data anonymization. Preoperative graft and recipient characteristics were evaluated along with the intraoperative details and postoperative outcomes of recipients.
LIVER TRANSPLANTATION, August 2015
TABLE 1. Preoperative Characteristics of the Recipients Recipient Characteristics Male/female, n Age, years Indication, n (%) HCC HCV cirrhosis Alcoholic cirrhosis HBV cirrhosis Primary biliary cirrhosis Insulinoma liver metastasis Re-LT Chronic rejection Ischemic cholangitis MELD
Values 27/12 61.8 6 4.3 18 (46.2) 9 (23.1) 5 (12.8) 1 (2.6) 1 (2.6) 1 (2.6) 4 (10.3) 3 (7.7) 1 (2.6) 16.8 6 5.7
Statistical Analysis Continuous variables were expressed as means and standard deviations. A chi-square test or Fisher’s exact test was employed to compare categorical variables. Patient actuarial survival was calculated with the Kaplan-Meier method. A P value < 0.05 was considered significant in all tests. All statistical analyses were performed with the PASW statistical package (version 18.0 for Windows; SPSS, Chicago, IL).
RESULTS Graft Characteristics In the context of DLT, we assessed the data for 39 FAP livers from 23 men and 16 women with a mean age of 42.9 6 14.7 years. Livers were retrieved from the FAP donor by the classic technique in 22 patients (56%); 17 were performed during the first 5 years of the program (before the first description of the double piggyback technique). Since 2005, the piggyback technique was considered the gold standard procedure for explanting FAP grafts at our institute.
Recipient Characteristics (Table 1) Recipients were predominantly male (27 males and 12 females) with a mean age of 61.8 6 4.3 years and a mean Model for End-Stage Liver Disease (MELD) score of 16.8 6 5.7. HCC within the Milan criteria16 was the most frequent indication for LT (46.2%), and it was followed by HCV cirrhosis (23.1%) and alcoholic cirrhosis (12.8%). Four patients underwent transplantation in the context of re-orthotopic liver transplantation (3 because of chronic rejection and 1 because of ischemic cholangitis). Eight patients had previous abdominal surgery, and nontumoral portal vein thrombosis was evident in 15.3%.
LIVER TRANSPLANTATION, Vol. 21, No. 8, 2015
DE LA SERNA ET AL. 1053
TABLE 2. Type of Outflow Reconstruction in DLT Recipients and Incidence of Outflow Complications Outflow Type of Venous Outflow Reconstruction (n 5 39)
n (%)
Complications, n (%)
0 (0) 6 (15) 16 (41)
0 (0) 0 (0) 0 (0)
11 (28)
4 (10.2)
IVC Auricle
4 2
1 0
Y-shaped IVC-iliac bifurcation
5
3
6 (15.3)
0 (0)
Aorta
1
0
“Bellvitge” arterial graft
5
0
Classic technique (donor) Classic anastomosis Cavocaval side-to-side anastomosis End-to-side anastomosis without graft Piggyback technique (donor) End-to-side anastomosis with venous graft interposition
End-to-side anastomosis with arterial graft interposition
Recipient Intraoperative Data The piggyback technique with caval preservation was performed in all recipients to remove the native liver.17,18 The total operating time was 363.5 6 87.9 minutes, and the total ischemia time was 558.5 6 137.2 minutes (the arterial ischemia time was 24.2 6 19.6 minutes). In total, 46.2% of the patients were transfused, and this required a median of 1 pack of red blood cells (range, 0-19 packs). For venous outflow reconstruction, among the FAP livers procured by the classic technique, we accomplished a cavocaval side-to-side anastomosis in 6 patients (15%) and an end-to-side anastomosis without graft interposition in 16 cases (41%). For the 17 FAP livers retrieved by the piggyback technique, we performed an end-to-side anastomosis with an interposed graft in all of the recipients. In 11 cases (28%), a cadaveric venous graft was interposed between the stump of the 3 HVs of the recipient and the graft HV to enlarge the end-to-side anastomosis. The types of venous grafts employed were as follows: inferior vena cava (IVC; n 5 4), auricles (n 5 2), and Yshaped IVC-iliac bifurcation (n 5 5; Table 2). Since 2011, we have employed arterial grafts for venous outflow reconstruction. In 1 case, we used a cadaveric aortic graft segment, and in 5 patients, we created a tailored patch with a cadaveric pulmonary or iliac artery. The technique has been described elsewhere16; briefly, the artery is longitudinally opened on the back table and sutured to the junction of the 3 HVs of the FAP graft before both ends are joined transversally. End-to-end portal vein anastomosis was performed in all cases, and it required an interposed venous graft in 2 patients. Arterial anastomosis was adapted
to the arterial anatomical variants of the hepatic graft, and this required arterial reconstruction during bench surgery in 25% of the patients. Biliary anastomosis was completed in an end-to-end fashion without a T-tube in all cases.
Postoperative Outcomes Among the 39 recipients, there was no primary nonfunction, and there was no in-hospital mortality. Three patients (7.7%) underwent relaparotomy because of hemoperitoneum. The mean intensive care unit stay was 3.9 6 4.6 days (range, 1-29 days), and the mean hospital stay was 16.6 6 9.1 days (range, 845 days). After a mean follow-up of 81.4 6 53.3 months, 6 patients (15.4%) developed biliary stenosis. As for arterial complications, 1 patient presented with arterial stenosis (2.6%) 2 months after re-LT, and 1 patient developed late thrombosis (2.5%) 12 months after LT. All of them were solved after endovascular and antiplatelet therapy. Additionally, portal thrombosis occurred in 2 patients (5.1%) 30 and 54 months after LT. One of them had a pre-LT portal thrombosis and had required an interposed graft for portal anastomosis. Two patients (5.1%) developed portal stenosis 51 and 14 months after LT. All portal complications were solved after endovascular and anticoagulant therapy. As for outflow complications, 4 patients (10.3 %) showed clinical features of subacute BCS (moderated ascites and mild perturbation of hepatic laboratory tests) after a median delay of 7 months (range, 4.9-84.4 months). The diagnosis was established by contrastenhanced imaging technique and the measurement of
1054 DE LA SERNA ET AL.
Figure 1. Overall survival. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
venous pressure gradients. All patients subsequently underwent percutaneous balloon dilatation and stent placement. It should be noted that all these patients who presented with BCS had received an interposed venous graft for outflow reconstruction, and none of the 6 patients who had received an interposed arterial graft developed any outflow complications at their last follow-up (Table 2). Thus, when we compared outcomes between venous and arterial graft interposition for outflow reconstruction, the incidence of outflow complications tended to be higher among the venous group versus the arterial group [venous graft, 4 (36%), versus arterial graft, 0 (0%); P 5 0.12]. At the end of follow-up (mean, 81.4 6 53.3 months), 15 recipients had died, mostly because of hepatitis C recurrence and cirrhosis complications (6 patients or 15.4%). HCC recurrence and ductopenic rejection each led to death in 2 patients (5.1%). At 1, 3, 5, and 10 years, overall patient survival was 97%, 79%, 76%, and 67%, respectively (Fig. 1).
DISCUSSION DLT is a validated technique that has been used to expand the donor pool with excellent results.19 The younger ages of donors with FAP, the shorter ischemia times, and the comparable morbidity in these recipients versus those with cadaveric donors20 are major advantages of DLT. The procedure notably takes advantage of the low risk of developing clinically relevant de novo systemic amyloidosis in the recipient. However, because this might occur earlier than thought, as already published by our group,21 it probably remains justifiable to restrict the use of these types of grafts to select, older “risky” recipients. Donors with FAP should be treated as living liver donors, and the safest surgical procedure should be
LIVER TRANSPLANTATION, August 2015
guaranteed. Although it has already been demonstrated that there are no differences in morbidity and survival rates between FAP donors and FAP nondonors,20 all efforts should be directed toward keeping this risk as low as possible. To this end, the double piggyback technique (in both donors and recipients) described by Pena et al.3 has progressively replaced the classic procedure of caval replacement in many centers. This avoids the use of venovenous bypass and attends to the particular problem of hemodynamic lability in this population. In DLT, vessels should be shared between the FAP donor liver and the FAP graft; thus, vascular stumps are usually short, and this renders the anastomosis more difficult, especially when the double piggyback technique is employed. This is particularly critical in venous anastomosis when one is trying to avoid BCS, which has a reported mortality rate up to 24% after LT in the literature.2 It occurred in 4 patients in our series, and fortunately, none of them died because of this complication. Venous outflow obstruction following LT has been related to technical factors. Although conflicting results have been published on the incidence after the piggyback technique versus the classic technique, it is clear that the side-to-side cavocaval approach diminishes the risk, and the number of HV stumps used for the end-to-end anastomosis seems to be highly related.1 To overcome the problem of short vascular stumps for outflow reconstruction in the special scenario of DLT, several technical options have been proposed in the literature. Direct end-to-end anastomosis between the HV stumps is technically difficult, and venoplasties are usually performed.6 The use of vascular grafts is also highly recommended, and several venous graft options have been proposed, including the use of a cadaveric Y-shaped IVC-iliac or portal vein bifurcation,9,10,13 a longitudinally opened iliac vein,11,12 or an autologous portal vein7,13 or umbilical vein.14 Cescon et al.15 have even described a cadaveric IVC patch that is sutured to each venous opening in the FAP graft to create a circular stump that can be anastomosed to the recipient’s IVC. Prosthetic materials have also been also employed for venous anastomosis but are associated with a higher incidence of thrombosis because of the lowpressure circulation system.22 In living donor LT, the use of arterial grafts is a common procedure used for venous reconstruction that overcomes the problem of venous tributaries. Excellent results have been published with the iliac bifurcation and aorta used to reconstruct the venous outflow.23-25 Nevertheless, using arterial grafts for venous reconstruction is an uncommon approach in DLT, and our group recently published the first description of the technique (Bellvitge arterial graft).16 The bench surgery technique does not differ from the conventional interposed venous graft reconstruction, and we believe that it provides superior strength to the anastomosis, which could have contributed to the lower incidence of BCS observed among our patients receiving an interposed arterial graft in this series.
LIVER TRANSPLANTATION, Vol. 21, No. 8, 2015
Some authors consider arterial grafts the best option for venous reconstruction in living donor LT.23 Although arteries usually shrink when employed in low-pressure systems, their patency mainly depends on the diameter of the artery.25 Therefore, the arterial graft that we propose is somewhat “tailored,” in that it is calibrated and adapted to the diameter of the HV stump in the recipient to ensure that it provides sufficient caliber to remain patent and avoid outflow complications. Indeed, in our experience, the incidence of postoperative BCS was comparable between patients with venous and arterial interposed grafts and tended to be lower among patients receiving arterial grafts. The small sample size and the retrospective nature of the review, which are important limitations of our study, could possibly explain the lack of statistical significance despite the clinically relevant difference. Larger series are needed to confirm our results. In conclusion, arterial grafts constitute a feasible and safe option for vascular outflow reconstruction in DLT because they are associated with a relatively low incidence of outflow complications. The recently proposed Bellvitge arterial graft technique should be added to the current range of available surgical modalities.
ACKNOWLEDGMENT We want to thank Michael Maudsley for reviewing the English in this article.
REFERENCES 1. Horton JD, San Miguel FL, Membreno F, Wright F, Paima J, Foster P, Ortiz JA. Budd-Chiari syndrome: illustrated review of current management. Liver Int 2008;28:455466. 2. Navarro F, Le Moine MC, Fabre JM, Belghiti J, Cherqui D, Adam R, et al. Specific vascular complications of orthotopic liver transplantation with preservation of the retrohepatic vena cava: review of 1361 cases. Transplantation 1999;68:646-650. 3. Pena JR, Barroso E, Martins A, Andrade JR, Pereira JP. Sequential whole liver transplant resected as piggy-back from FAP patients. Liver Transpl 2002;8:C24. 4. Escobar B, Taura P, Barreneche N, Beltran J, Balust J, Martinez-Palli G, et al. The influence of the explant technique on the hemodynamic profile during sequential domino liver transplantation in familial amyloid polyneuropathy patients. Liver Transpl 2009;15:869-875. 5. Inomata Y, Zeled on ME, Asonuma K, Okajima H, Takeichi T, Ishiko T, Ando Y. Whole-liver graft without the retrohepatic inferior vena cava for sequential (domino) living donor liver transplantation. Am J Transplant 2007;7:1629-1632. 6. Chan SC, Lo CM, Ng KK, Chok KS, Fan ST. Simplifying hepatic venous outflow reconstruction in sequential living donor liver transplantation. Liver Transpl 2009;15: 1514-1518. 7. Soin A, Kumaran V, Mohanka R, Mehta N, Mohan N, Nundy S. Bridge venoplasty: a new technique to simplify venous outflow reconstruction in living donor domino liver transplantation. Surgery 2010;148:155-157. 8. Lacerda CM, Miranda LE, Amorim AG, Sabat BD, de Melo PS, Neto OF, et al. Reconstruction of venous outflow of domino liver graft using deceased donor iliac/cava veins. Transplant Proc 2008;40:811-813.
DE LA SERNA ET AL. 1055
9. Pacheco-Moreira LF, de Oliveira ME, Balbi E, da Silva AC, Miecznikowski R, de Faria LJ, et al. A new technical option for domino liver transplantation. Liver Transpl 2003;9:632-633. 10. Jabbour N, Gagandeep S, Genyk Y, Selby R, Mateo R. Caval preservation with reconstruction of the hepatic veins using caval-common iliac bifurcation graft for domino liver transplantation. Liver Transpl 2006;12:342-345. 11. Liu C, Loong CC, Hsia CY, Tsou MY, Tsai HL, Wei CF. Venoplasty of hepatic venous outflow with a venous patch in domino liver transplantation. Liver Transpl 2008;14:1378-1379. 12. Pinheiro RS, Lai Q, Dahrenmoller C, Lerut J. Complex hepatic outflow reconstruction in domino liver transplantation. Hepatobiliary Pancreat Dis Int 2014;13:98-100. 13. Cerqueira A, Pacheco-Moreira L, Enne M, Alves J, Amil R, Balbi E, Martinho JM. Outflow reconstruction in domino liver transplantation with interposition of autologous portal vein graft. A new technical option in living donor domino liver transplant scenario. Liver Transpl 2006;12: 1298-1300. 14. Mergental H, Gouw AS, Slooff MJ, de Jong KP. Venous outflow reconstruction with surgically reopened obliterated umbilical vein in domino liver transplantation. Liver Transpl 2007;13:769-772. 15. Cescon M, Grazi GL, Ravaioli M, Cucchetti A, Ercolani G, Pinna AD. Modified outflow reconstruction with a venous patch in domino liver transplantation. Liver Transpl 2007;13:1756-1757. 16. Llado L, Ramos E, De LaSerna S, Fabregat J. Outflow reconstruction with arterial patch in domino liver transplantation: a new technical option. Hepatobiliary Pancreat Dis Int 2014;13:551-554. 17. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693-699. 18. Llad o L, Figueras J. Techniques of orthotopic liver transplantation. HPB (Oxford) 2004;6:69-75. 19. Department of Transplant Surgery, Karolinska University Hospital Huddinge. The familial amyloidotic polyneuropathy world transplant registry. www.fapwtr.org. Accessed May 1, 2015. 20. Tincani G, Hoti E, Andreani P, Ricca L, Pittau G, Vitale V, et al. Operative risks of domino liver transplantation for the familial amyloid polyneuropathy liver donor and recipient: a double analysis. Am J Transplant 2011;11: 759-766. 21. Llad o L, Baliellas C, Casasnovas C, Ferrer I, Fabregat J, Ramos E, et al. Risk of transmission of systemic transthyretin amyloidosis after domino liver transplantation. Liver Transpl 2010;16:1386-1392. 22. Hwang S, Jung DH, Ha TY, Ahn CS, Moon DB, Kim KH, et al. Usability of ringed polytetrafluoroethylene grafts for middle hepatic vein reconstruction during living donor liver transplantation. Liver Transpl 2012;18:955-965. 23. Hwang S, Lee SG, Ahn CS, Park KM, Kim KH, Moon DB, Ha TY. Cryopreserved iliac artery is indispensable interposition graft material for middle hepatic vein reconstruction of right liver grafts. Liver Transpl 2005;11:644649. 24. Dar FS, Faraj W, Heaton ND, Rela M. Variation in the venous drainage of left lateral segment liver graft requiring reconstruction of segment III vein with donor iliac artery. Liver Transpl 2008;14:576-579. 25. Kilic M, Aydin U, Sozbilen M, Ozer I, Tamsel S, Demirpolat G, et al. Comparison between allogenic and autologous vascular conduits in the drainage of anterior sector in right living donor liver transplantation. Transpl Int 2007;20:697-701.
ORIGINAL ARTICLE
Technical Options for Outflow Reconstruction in Domino Liver Transplantation: A Single European Center Experience Sofıa De la Serna, Laura Llado, Emilio Ramos, Joan Fabregat, Carme Baliellas, Juli Busquets, Lluis Secanella, Nuria Pelaez, Jaume Torras, and Antoni Rafecas Biome`dica de Bellvitge, Liver Transplantation Unit, Hospital Universitari de Bellvitge, Institut d’Investigacio Barcelona, Spain
Venous outflow is critical to the success of liver transplantation (LT). In domino liver transplantation (DLT), the venous cuffs should be shared between the donor and the recipient, and the length can be compromised. The aim of this study was to describe and compare the technical options for outflow reconstruction used at our institution. This was a retrospective analysis of 39 consecutive DLT recipients between January 1997 and May 2013. Twenty-seven men and 12 women (mean age, 61.8 6 4.3 years) underwent LT and consented to receive a liver from a donor with familial amyloid polyneuropathy (FAP). The main indications were hepatocellular carcinoma and hepatitis C virus cirrhosis. All recipients underwent transplantation by a piggyback technique. Liver procurement in the FAP donors was performed with the classic technique in 22 patients and with the piggyback technique in the last 17. In these latter cases, for vascular outflow reconstruction, a cadaveric venous graft was interposed between the hepatic vein (HV) stump of the FAP liver and the recipient HV in 11 cases (28%). Since 2011, we have employed arterial grafts to be interposed between the vessels stumps: a tailored arterial graft in 5 patients and an aortic graft in 1 case. There was no postoperative mortality. Arterial and portal complications presented in 2 (5.1) and 4 patients (10.3), respectively. Postoperative outflow complications (post-LT subacute Budd-Chiari syndrome) occurred in 4 patients, and all of them had received a venous interposed graft for reconstruction. The incidence of outflow complications tended to be higher among patients with venous grafts than those with arterial graft interposition. Overall patient survival at 1, 3, 5, and 10 years was 97%, 79%, respectively. Arterial grafts constitute a feasible and safe option for vascular outflow reconstruction in DLT because they are associated with a relatively low incidence of complications. The recently proposed Bellvitge arterial graft technique should be added to the current range of available surgical modalities. C 2015 AASLD. Liver Transpl 21:1051-1055, 2015. V Received December 23, 2014; accepted March 30, 2015. In liver transplantation (LT), ensuring optimal venous outflow is critical to avoiding graft congestion and post-LT Budd-Chiari syndrome (BCS). BCS has a reported incidence in transplant patients of 1% to 7% and is associated with mortality rates up to 24%; therefore, it represents a significant and relatively common complication.1,2 Domino liver transplantation (DLT), also known as sequential LT, is a wellknown technique useful for expanding the number of
potential donors. Although its safety has already been established, special issues remain to be addressed. The main sources of DLT donors are patients with familial amyloid polyneuropathy (FAP). Traditionally, liver procurement in FAP donors has been performed with the classic technique of caval replacement. However, the increased hemodynamic lability of FAP patients has led many centers to progressively replace the classic technique with the double piggyback
Abbreviations: BCS, Budd-Chiari syndrome; DLT, domino liver transplantation; FAP, familial amyloid polyneuropathy; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HV, hepatic vein; IVC, inferior vena cava; LT, liver transplantation; MELD, Model for End-Stage Liver Disease. Potential conflict of interest: Nothing to report. Address reprint requests to Laura Llado, M.D., Ph.D., Liver Transplantation Unit, Hospital Universitari de Bellvitge, Feixa Llarga s/n, 08907 Barcelona, Spain. Telephone: 134-93-260-7940; FAX: 134-93-260-7603; E-mail: [email protected] DOI 10.1002/lt.24143 View this article online at wileyonlinelibrary.com. LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
C 2015 American Association for the Study of Liver Diseases. V
1052 DE LA SERNA ET AL.
technique in both the donor and the recipient.3,4 Nevertheless, venous reconstruction remains a major challenge of DLT, especially when the double piggyback technique is employed. The need to share the vascular stumps between the graft and the donor might compromise the length of the vessels and render the outflow reconstruction extremely laborious. Direct end-to-end anastomosis between the hepatic veins (HVs) of the graft and the recipient have also been associated with technical difficulties,5 and therefore, septoplasties or bridge venoplasties between the HV stumps in the FAP graft are advisable whenever possible.6-8 To overcome the problem of short cuffs, interposed grafts are often required to obtain a suitable operative field for the reconstruction, for which several venous graft options have been proposed in the literature.1,915 To this end, our group has recently proposed a novel technique using arterial grafts to create a tailored patch that can be anastomosed between the HV stumps of the FAP graft and the recipient.16 The aim of this study was to review the technical options for outflow reconstruction in DLT in a European single center and to analyze the short- and longterm results.
PATIENTS AND METHODS Study Design This was a retrospective study of 39 consecutive recipients of DLT performed at Hospital Universitari de Bellvitge (Barcelona, Spain) from January 1997 to May 2013. We performed LT 974 times at our institution during the study period. Thirty-nine patients (4%) required LT for FAP disease. Since 1999, when the DLT program was implemented at our institution, 33 patients with FAP have consented to donate their livers for transplantation. Among these 33 potential FAP donors, 2 were rejected preoperatively: one because of a history of breast cancer 7 years earlier (although the patient was free of tumor recurrence) and the other because of advanced age (67 years old) and the presence of liver fibrosis confirmed by elastography. In 1 patient, the donor liver was discarded intraoperatively because of a prolonged warm ischemia time. Another 9 FAP livers were shipped from other hospitals, and this produced a final case series of 39. The selection criteria for recipients of FAP livers were an age equal or greater than 60 years and hepatocellular carcinoma (HCC) within the Milan criteria.17 Patients less than 60 years old were considered if particular criteria were met.
Data Analysis Data were obtained from our prospectively collected LT database and retrospectively analyzed after data anonymization. Preoperative graft and recipient characteristics were evaluated along with the intraoperative details and postoperative outcomes of recipients.
LIVER TRANSPLANTATION, August 2015
TABLE 1. Preoperative Characteristics of the Recipients Recipient Characteristics Male/female, n Age, years Indication, n (%) HCC HCV cirrhosis Alcoholic cirrhosis HBV cirrhosis Primary biliary cirrhosis Insulinoma liver metastasis Re-LT Chronic rejection Ischemic cholangitis MELD
Values 27/12 61.8 6 4.3 18 (46.2) 9 (23.1) 5 (12.8) 1 (2.6) 1 (2.6) 1 (2.6) 4 (10.3) 3 (7.7) 1 (2.6) 16.8 6 5.7
Statistical Analysis Continuous variables were expressed as means and standard deviations. A chi-square test or Fisher’s exact test was employed to compare categorical variables. Patient actuarial survival was calculated with the Kaplan-Meier method. A P value < 0.05 was considered significant in all tests. All statistical analyses were performed with the PASW statistical package (version 18.0 for Windows; SPSS, Chicago, IL).
RESULTS Graft Characteristics In the context of DLT, we assessed the data for 39 FAP livers from 23 men and 16 women with a mean age of 42.9 6 14.7 years. Livers were retrieved from the FAP donor by the classic technique in 22 patients (56%); 17 were performed during the first 5 years of the program (before the first description of the double piggyback technique). Since 2005, the piggyback technique was considered the gold standard procedure for explanting FAP grafts at our institute.
Recipient Characteristics (Table 1) Recipients were predominantly male (27 males and 12 females) with a mean age of 61.8 6 4.3 years and a mean Model for End-Stage Liver Disease (MELD) score of 16.8 6 5.7. HCC within the Milan criteria16 was the most frequent indication for LT (46.2%), and it was followed by HCV cirrhosis (23.1%) and alcoholic cirrhosis (12.8%). Four patients underwent transplantation in the context of re-orthotopic liver transplantation (3 because of chronic rejection and 1 because of ischemic cholangitis). Eight patients had previous abdominal surgery, and nontumoral portal vein thrombosis was evident in 15.3%.
LIVER TRANSPLANTATION, Vol. 21, No. 8, 2015
DE LA SERNA ET AL. 1053
TABLE 2. Type of Outflow Reconstruction in DLT Recipients and Incidence of Outflow Complications Outflow Type of Venous Outflow Reconstruction (n 5 39)
n (%)
Complications, n (%)
0 (0) 6 (15) 16 (41)
0 (0) 0 (0) 0 (0)
11 (28)
4 (10.2)
IVC Auricle
4 2
1 0
Y-shaped IVC-iliac bifurcation
5
3
6 (15.3)
0 (0)
Aorta
1
0
“Bellvitge” arterial graft
5
0
Classic technique (donor) Classic anastomosis Cavocaval side-to-side anastomosis End-to-side anastomosis without graft Piggyback technique (donor) End-to-side anastomosis with venous graft interposition
End-to-side anastomosis with arterial graft interposition
Recipient Intraoperative Data The piggyback technique with caval preservation was performed in all recipients to remove the native liver.17,18 The total operating time was 363.5 6 87.9 minutes, and the total ischemia time was 558.5 6 137.2 minutes (the arterial ischemia time was 24.2 6 19.6 minutes). In total, 46.2% of the patients were transfused, and this required a median of 1 pack of red blood cells (range, 0-19 packs). For venous outflow reconstruction, among the FAP livers procured by the classic technique, we accomplished a cavocaval side-to-side anastomosis in 6 patients (15%) and an end-to-side anastomosis without graft interposition in 16 cases (41%). For the 17 FAP livers retrieved by the piggyback technique, we performed an end-to-side anastomosis with an interposed graft in all of the recipients. In 11 cases (28%), a cadaveric venous graft was interposed between the stump of the 3 HVs of the recipient and the graft HV to enlarge the end-to-side anastomosis. The types of venous grafts employed were as follows: inferior vena cava (IVC; n 5 4), auricles (n 5 2), and Yshaped IVC-iliac bifurcation (n 5 5; Table 2). Since 2011, we have employed arterial grafts for venous outflow reconstruction. In 1 case, we used a cadaveric aortic graft segment, and in 5 patients, we created a tailored patch with a cadaveric pulmonary or iliac artery. The technique has been described elsewhere16; briefly, the artery is longitudinally opened on the back table and sutured to the junction of the 3 HVs of the FAP graft before both ends are joined transversally. End-to-end portal vein anastomosis was performed in all cases, and it required an interposed venous graft in 2 patients. Arterial anastomosis was adapted
to the arterial anatomical variants of the hepatic graft, and this required arterial reconstruction during bench surgery in 25% of the patients. Biliary anastomosis was completed in an end-to-end fashion without a T-tube in all cases.
Postoperative Outcomes Among the 39 recipients, there was no primary nonfunction, and there was no in-hospital mortality. Three patients (7.7%) underwent relaparotomy because of hemoperitoneum. The mean intensive care unit stay was 3.9 6 4.6 days (range, 1-29 days), and the mean hospital stay was 16.6 6 9.1 days (range, 845 days). After a mean follow-up of 81.4 6 53.3 months, 6 patients (15.4%) developed biliary stenosis. As for arterial complications, 1 patient presented with arterial stenosis (2.6%) 2 months after re-LT, and 1 patient developed late thrombosis (2.5%) 12 months after LT. All of them were solved after endovascular and antiplatelet therapy. Additionally, portal thrombosis occurred in 2 patients (5.1%) 30 and 54 months after LT. One of them had a pre-LT portal thrombosis and had required an interposed graft for portal anastomosis. Two patients (5.1%) developed portal stenosis 51 and 14 months after LT. All portal complications were solved after endovascular and anticoagulant therapy. As for outflow complications, 4 patients (10.3 %) showed clinical features of subacute BCS (moderated ascites and mild perturbation of hepatic laboratory tests) after a median delay of 7 months (range, 4.9-84.4 months). The diagnosis was established by contrastenhanced imaging technique and the measurement of
1054 DE LA SERNA ET AL.
Figure 1. Overall survival. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
venous pressure gradients. All patients subsequently underwent percutaneous balloon dilatation and stent placement. It should be noted that all these patients who presented with BCS had received an interposed venous graft for outflow reconstruction, and none of the 6 patients who had received an interposed arterial graft developed any outflow complications at their last follow-up (Table 2). Thus, when we compared outcomes between venous and arterial graft interposition for outflow reconstruction, the incidence of outflow complications tended to be higher among the venous group versus the arterial group [venous graft, 4 (36%), versus arterial graft, 0 (0%); P 5 0.12]. At the end of follow-up (mean, 81.4 6 53.3 months), 15 recipients had died, mostly because of hepatitis C recurrence and cirrhosis complications (6 patients or 15.4%). HCC recurrence and ductopenic rejection each led to death in 2 patients (5.1%). At 1, 3, 5, and 10 years, overall patient survival was 97%, 79%, 76%, and 67%, respectively (Fig. 1).
DISCUSSION DLT is a validated technique that has been used to expand the donor pool with excellent results.19 The younger ages of donors with FAP, the shorter ischemia times, and the comparable morbidity in these recipients versus those with cadaveric donors20 are major advantages of DLT. The procedure notably takes advantage of the low risk of developing clinically relevant de novo systemic amyloidosis in the recipient. However, because this might occur earlier than thought, as already published by our group,21 it probably remains justifiable to restrict the use of these types of grafts to select, older “risky” recipients. Donors with FAP should be treated as living liver donors, and the safest surgical procedure should be
LIVER TRANSPLANTATION, August 2015
guaranteed. Although it has already been demonstrated that there are no differences in morbidity and survival rates between FAP donors and FAP nondonors,20 all efforts should be directed toward keeping this risk as low as possible. To this end, the double piggyback technique (in both donors and recipients) described by Pena et al.3 has progressively replaced the classic procedure of caval replacement in many centers. This avoids the use of venovenous bypass and attends to the particular problem of hemodynamic lability in this population. In DLT, vessels should be shared between the FAP donor liver and the FAP graft; thus, vascular stumps are usually short, and this renders the anastomosis more difficult, especially when the double piggyback technique is employed. This is particularly critical in venous anastomosis when one is trying to avoid BCS, which has a reported mortality rate up to 24% after LT in the literature.2 It occurred in 4 patients in our series, and fortunately, none of them died because of this complication. Venous outflow obstruction following LT has been related to technical factors. Although conflicting results have been published on the incidence after the piggyback technique versus the classic technique, it is clear that the side-to-side cavocaval approach diminishes the risk, and the number of HV stumps used for the end-to-end anastomosis seems to be highly related.1 To overcome the problem of short vascular stumps for outflow reconstruction in the special scenario of DLT, several technical options have been proposed in the literature. Direct end-to-end anastomosis between the HV stumps is technically difficult, and venoplasties are usually performed.6 The use of vascular grafts is also highly recommended, and several venous graft options have been proposed, including the use of a cadaveric Y-shaped IVC-iliac or portal vein bifurcation,9,10,13 a longitudinally opened iliac vein,11,12 or an autologous portal vein7,13 or umbilical vein.14 Cescon et al.15 have even described a cadaveric IVC patch that is sutured to each venous opening in the FAP graft to create a circular stump that can be anastomosed to the recipient’s IVC. Prosthetic materials have also been also employed for venous anastomosis but are associated with a higher incidence of thrombosis because of the lowpressure circulation system.22 In living donor LT, the use of arterial grafts is a common procedure used for venous reconstruction that overcomes the problem of venous tributaries. Excellent results have been published with the iliac bifurcation and aorta used to reconstruct the venous outflow.23-25 Nevertheless, using arterial grafts for venous reconstruction is an uncommon approach in DLT, and our group recently published the first description of the technique (Bellvitge arterial graft).16 The bench surgery technique does not differ from the conventional interposed venous graft reconstruction, and we believe that it provides superior strength to the anastomosis, which could have contributed to the lower incidence of BCS observed among our patients receiving an interposed arterial graft in this series.
LIVER TRANSPLANTATION, Vol. 21, No. 8, 2015
Some authors consider arterial grafts the best option for venous reconstruction in living donor LT.23 Although arteries usually shrink when employed in low-pressure systems, their patency mainly depends on the diameter of the artery.25 Therefore, the arterial graft that we propose is somewhat “tailored,” in that it is calibrated and adapted to the diameter of the HV stump in the recipient to ensure that it provides sufficient caliber to remain patent and avoid outflow complications. Indeed, in our experience, the incidence of postoperative BCS was comparable between patients with venous and arterial interposed grafts and tended to be lower among patients receiving arterial grafts. The small sample size and the retrospective nature of the review, which are important limitations of our study, could possibly explain the lack of statistical significance despite the clinically relevant difference. Larger series are needed to confirm our results. In conclusion, arterial grafts constitute a feasible and safe option for vascular outflow reconstruction in DLT because they are associated with a relatively low incidence of outflow complications. The recently proposed Bellvitge arterial graft technique should be added to the current range of available surgical modalities.
ACKNOWLEDGMENT We want to thank Michael Maudsley for reviewing the English in this article.
REFERENCES 1. Horton JD, San Miguel FL, Membreno F, Wright F, Paima J, Foster P, Ortiz JA. Budd-Chiari syndrome: illustrated review of current management. Liver Int 2008;28:455466. 2. Navarro F, Le Moine MC, Fabre JM, Belghiti J, Cherqui D, Adam R, et al. Specific vascular complications of orthotopic liver transplantation with preservation of the retrohepatic vena cava: review of 1361 cases. Transplantation 1999;68:646-650. 3. Pena JR, Barroso E, Martins A, Andrade JR, Pereira JP. Sequential whole liver transplant resected as piggy-back from FAP patients. Liver Transpl 2002;8:C24. 4. Escobar B, Taura P, Barreneche N, Beltran J, Balust J, Martinez-Palli G, et al. The influence of the explant technique on the hemodynamic profile during sequential domino liver transplantation in familial amyloid polyneuropathy patients. Liver Transpl 2009;15:869-875. 5. Inomata Y, Zeled on ME, Asonuma K, Okajima H, Takeichi T, Ishiko T, Ando Y. Whole-liver graft without the retrohepatic inferior vena cava for sequential (domino) living donor liver transplantation. Am J Transplant 2007;7:1629-1632. 6. Chan SC, Lo CM, Ng KK, Chok KS, Fan ST. Simplifying hepatic venous outflow reconstruction in sequential living donor liver transplantation. Liver Transpl 2009;15: 1514-1518. 7. Soin A, Kumaran V, Mohanka R, Mehta N, Mohan N, Nundy S. Bridge venoplasty: a new technique to simplify venous outflow reconstruction in living donor domino liver transplantation. Surgery 2010;148:155-157. 8. Lacerda CM, Miranda LE, Amorim AG, Sabat BD, de Melo PS, Neto OF, et al. Reconstruction of venous outflow of domino liver graft using deceased donor iliac/cava veins. Transplant Proc 2008;40:811-813.
DE LA SERNA ET AL. 1055
9. Pacheco-Moreira LF, de Oliveira ME, Balbi E, da Silva AC, Miecznikowski R, de Faria LJ, et al. A new technical option for domino liver transplantation. Liver Transpl 2003;9:632-633. 10. Jabbour N, Gagandeep S, Genyk Y, Selby R, Mateo R. Caval preservation with reconstruction of the hepatic veins using caval-common iliac bifurcation graft for domino liver transplantation. Liver Transpl 2006;12:342-345. 11. Liu C, Loong CC, Hsia CY, Tsou MY, Tsai HL, Wei CF. Venoplasty of hepatic venous outflow with a venous patch in domino liver transplantation. Liver Transpl 2008;14:1378-1379. 12. Pinheiro RS, Lai Q, Dahrenmoller C, Lerut J. Complex hepatic outflow reconstruction in domino liver transplantation. Hepatobiliary Pancreat Dis Int 2014;13:98-100. 13. Cerqueira A, Pacheco-Moreira L, Enne M, Alves J, Amil R, Balbi E, Martinho JM. Outflow reconstruction in domino liver transplantation with interposition of autologous portal vein graft. A new technical option in living donor domino liver transplant scenario. Liver Transpl 2006;12: 1298-1300. 14. Mergental H, Gouw AS, Slooff MJ, de Jong KP. Venous outflow reconstruction with surgically reopened obliterated umbilical vein in domino liver transplantation. Liver Transpl 2007;13:769-772. 15. Cescon M, Grazi GL, Ravaioli M, Cucchetti A, Ercolani G, Pinna AD. Modified outflow reconstruction with a venous patch in domino liver transplantation. Liver Transpl 2007;13:1756-1757. 16. Llado L, Ramos E, De LaSerna S, Fabregat J. Outflow reconstruction with arterial patch in domino liver transplantation: a new technical option. Hepatobiliary Pancreat Dis Int 2014;13:551-554. 17. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693-699. 18. Llad o L, Figueras J. Techniques of orthotopic liver transplantation. HPB (Oxford) 2004;6:69-75. 19. Department of Transplant Surgery, Karolinska University Hospital Huddinge. The familial amyloidotic polyneuropathy world transplant registry. www.fapwtr.org. Accessed May 1, 2015. 20. Tincani G, Hoti E, Andreani P, Ricca L, Pittau G, Vitale V, et al. Operative risks of domino liver transplantation for the familial amyloid polyneuropathy liver donor and recipient: a double analysis. Am J Transplant 2011;11: 759-766. 21. Llad o L, Baliellas C, Casasnovas C, Ferrer I, Fabregat J, Ramos E, et al. Risk of transmission of systemic transthyretin amyloidosis after domino liver transplantation. Liver Transpl 2010;16:1386-1392. 22. Hwang S, Jung DH, Ha TY, Ahn CS, Moon DB, Kim KH, et al. Usability of ringed polytetrafluoroethylene grafts for middle hepatic vein reconstruction during living donor liver transplantation. Liver Transpl 2012;18:955-965. 23. Hwang S, Lee SG, Ahn CS, Park KM, Kim KH, Moon DB, Ha TY. Cryopreserved iliac artery is indispensable interposition graft material for middle hepatic vein reconstruction of right liver grafts. Liver Transpl 2005;11:644649. 24. Dar FS, Faraj W, Heaton ND, Rela M. Variation in the venous drainage of left lateral segment liver graft requiring reconstruction of segment III vein with donor iliac artery. Liver Transpl 2008;14:576-579. 25. Kilic M, Aydin U, Sozbilen M, Ozer I, Tamsel S, Demirpolat G, et al. Comparison between allogenic and autologous vascular conduits in the drainage of anterior sector in right living donor liver transplantation. Transpl Int 2007;20:697-701.
