The classic surgical technique for orthotopic liver

Transcription

1 Temporary Portocaval Shunt During Liver Transplantation With Vena Cava Preservation. Results of a Prospective Randomized Study Juan Figueras, * Laura Llado, * Emilio Ramos, * Eduardo Jaurrieta, * Antonio Rafecas, * Juan Fabregat, * Jaume Torras, * Antonio Sabate, and Antonia Dalmau This study aims to determine whether the use of a temporary portocaval shunt (PCS) improves hemodynamic and metabolic evolution during orthotopic liver transplantation (OLT). Preservation of the vena cava during OLT has gained wide acceptance. However, benefits of adding a temporary PCS to the piggyback technique during the anhepatic phase in patients with cirrhosis have not been shown. Eighty patients with cirrhosis were studied prospectively. They were randomly distributed into two groups: patients with a temporary PCS (n 40) and those without a PCS (n 40). In all cases, the piggyback technique was used. Hemodynamic profiles and biochemical data during OLT and clinical evolution after OLT were evaluated. Preoperative data were similar in both groups. Surgical time also was similar ( v minutes; P.3). Red blood cell requirements were lower in the PCS group ( v units), although differences were not significant. In the PCS group, 45% of patients did not need red blood cell transfusion, whereas in the other group, only 22% were not administered a transfusion (P.03). During the anhepatic phase, the decrease in cardiac output was lower in the PCS group ( 9.6% v 19%; P.05), whereas diuresis during the anhepatic phase was greater in the PCS group ( v ml/kg/h; P.005). There were no differences in liver biochemical parameters during the first 3 postoperative days. Nevertheless, creatinine levels increased significantly during this period only in the no- PCS group. The use of a temporary PCS during OLT improves hemodynamic status, reduces intraoperative transfusion requirements, and preserves renal function during and after OLT. (Liver Transpl 2001;7: ) From the Departments of *Surgery and Anesthesiology, Hospital Prínceps d Espanya, Ciudad Sanitaria y Universitaria (CSU) de Bellvitge, University of Barcelona, Spain. Supported by grants from the Fundació Agustí Pi i Sunyer, Barcelona; and grant no. FISS 98/0776 from the Fondo de Investigaciones Sanitarias, Ministerio de Sanidad y Consumo, Madrid, Spain. Address reprint requests to Juan Figueras, MD, Servicio de Cirugía General y Digestiva, Hospital Prínceps d Espanya, C/Feixa llarga s/n, Hospitalet de Llobregat, Barcelona, Spain. Telephone: ; FAX: ; jfigueras@csub.scs.es Copyright 2001 by the American Association for the Study of Liver Diseases /01/ $35.00/0 doi: /jlts The classic surgical technique for orthotopic liver transplantation (OLT) involves recipient hepatectomy with resection of the retrohepatic vena cava and cross-clamping of the portal vein. 1 Interruption of caval flow during the anhepatic phase results in a reduction in venous return to the heart 2 and a decrease in renal perfusion, 3 as well as splanchnic hyperemia secondary to portal clamping. Venovenous bypass improves hemodynamic stability and allows decompression of the occluded splanchnic venous system. 4 However, the use of venous bypass is associated with other complications, such as hypothermia and pulmonary thromboembolism. 5,6 In 1968, Calne and Williams 7 described preservation of the vena cava. This technique, known as piggyback, 8 was further developed in pediatric OLT, 9 but has gained wide acceptance in adults, mainly in Europe Nevertheless, it does not avoid splanchnic hyperemia secondary to portal clamping. The use of a temporary portocaval shunt (PCS) with the piggyback technique was first described by Tzakis et al, 14 then by Belghiti at al. 15 Previous studies have shown that preservation of both portal and caval blood flows throughout the procedure maintains hemodynamic stability and renal perfusion pressure This technique is particularly useful for patients with fulminant hepatitis who lack adequate portosystemic collaterals in the splachnic area. 17 However, benefits of adding a temporary PCS to the piggyback technique during the anhepatic phase in patients with cirrhosis have not been shown. 16,18,19 To verify the feasibility and real advantages of adding a temporary PCS to the piggyback technique, we began a prospective randomized study of patients with cirrhosis. A secondary objective is to elucidate whether benefits could be obtained in all patients or only in a selected group. Patients and Methods Case Material Between November 1998 and November 2000, a total of 116 OLTs were performed at our institution. Exclusion criteria 904 Liver Transplantation, Vol 7, No 10 (October), 2001: pp

2 Temporary Portocaval Shunt in Liver Transplantation 905 were as follows: patients without cirrhosis, re-olt, impossibility of preserving the inferior vena cava, portal vein thrombosis, pulmonary hypertension syndrome, and hepatorenal transplantation. Thirty-two patients were excluded from the study for the following reasons: 12 patients, noncirrhotic (familial amyloid neuropathy); 9 patients, re-olt; 2 patients, hepatorenal transplants; 4 patients, previous shunt or transjugular intrahepatic portosystemic shunt; 2 patients, impossibility of preserving the inferior vena cava (tumor was too close to it), and 2 patients, portal thrombosis. Four patients were excluded intraoperatively before randomization because of the presence of an enlarged caudate lobe posteriorly embracing the retrohepatic vena cava or inverted portal flow (Fig. 1). Patients were allocated to one of the groups by the closedenvelope method. This study was conducted in accordance with the Declaration of Helsinki and approved by the local ethics committee. All patients gave written informed consent before participating in the study. Eighty patients were included in the study and randomized into two groups: the PCS group (n 40), with a temporary PCS during the anhepatic phase; and the no-pcs group (n 40), without a PCS. Indications for the 80 OLTs included on the study were as follows: 18 patients (22.5%), alcoholic cirrhosis; 24 patients (30%), post hepatitis C cirrhosis; 2 patients (2.5%), post hepatitis B cirrhosis; 2 patients (2.5%), cryptogenic cirrhosis; 30 patients (37.5%), hepatocellular carcinoma with cirrhosis; and 4 patients (5%), primary biliary cirrhosis. Surgical Technique Anesthetic management in all cases followed the routine established in our institution. In brief, general anesthesia was induced and maintained with propofol, fentanyl, and atracurium. Mechanical ventilation was started at 10 ml/kg with a respiratory rate to obtain an end-expiratory PCO 2 of 30 to 35 mm Hg and an inspired oxygen fraction of 0.5 in air. Calcium was administered to maintain ionic calcium levels between 1 and 1.2 mmol/l, and sodium bicarbonate, to reach a ph greater than 7.3. All patients were placed on a warm blanket, lower limbs were covered with cotton and aluminum foil, and Figure 1. Scheme of randomization and exclusions. Four patients were excluded intraoperatively before randomization because of the presence of an enlarged caudate or inverted portal flow. serum warmed to 36 C was administered. Tranexamic acid, 10 mg/kg/h, was administered from the beginning of OLT until the vena porta and hepatic veins were unclamped as prophylaxis for fibrinolysis. At the beginning of the operation, a 14 G catheter (Venocath; Abbott Ireland, Sligo, Republic of Ireland) was placed in the superior mesenteric vein through an jejunal vein to measure portal pressure. The hilar structures were dissected. The hepatic artery and bile duct were ligated and divided. Portal flow was measured at the beginning of OLT with a transit-time ultrasound (Transonic Flowmeter, HT311; Transonic System Inc, Ithaca, NY). It should be noted that portal flow was not measured in 21 patients because the ultrasound device was not available. Next, the portal vein was cross-clamped and divided. In the PCS group, after exposure and lateral clamping of the infrahepatic vena cava, the proximal end of the portal vein was anastomosed end to side with the vena cava by using a running suture (Prolene 5/0; Ethicon, Somerville, NJ; Fig. 2). Hepatic devascularization allows the division of all ligamentous attachments and facilitates complete mobilization of the liver to the left. The small hepatic veins were ligated and divided. Major veins were cross-clamped together using a Klintman clamp (Pilling Weck Europe, Le Faget, France). Care was taken not to interrupt the vena cava lumen, thus preserving both portal and caval blood flows during the dissection and anhepatic phases. The three principal hepatic veins (right, middle, and left) were joined by dividing the intervening septa. The donor liver was placed orthotopically. An end-to-side caval anastomosis was performed between the donor suprahepatic vena cava and the cuff fashioned from the recipient s hepatic veins. In the PCS group, portacaval anastomosis was taken down, and an end-to-end anastomosis was performed. The inferior opening of the donor vena cava was left open to drain the effluent from the flush of the graft. Next, the distal end of the donor vena cava was closed using a vascular stapler (TA 30; Autosuture, Elancourt, France). The liver was perfused after portal anastomosis and before arterial anastomosis. In the no-pcs group, the surgical technique was identical except that portacaval anastomosis was not performed and consequently was not later dismantled. Hemodynamic Data and Other Measurements Intraoperative and postoperative hemodynamic monitoring included a Swan-Ganz catheter (Edwards Lifesciences Corp, Irvine, CA). Hemodynamic parameters evaluated throughout OLT were mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), mean pulmonary pressure (MPP), portal pressure, central venous pressure (CVP), pulmonary capillary pressure (PCP), systemic vascular resistance (SVR), and pulmonary vascular resistance (PVR). These parameters were recorded at the following periods: time 0 (T0), beginning of OLT; T1, after portal vein clamping; T2, after PCS unclamping (only the PCS group); T3, after hepatic vein clamping; T4, just before portal unclamping; T5, 5 minutes after reperfusion; and T6, 60 minutes after reperfusion. The anhepatic

3 906 Figueras et al Figure 2. (A) After end-to-side portacaval anastomosis, individual hepatic veins were ligated, exposing the major hepatic veins in a dry field without portal hypertension. (B) During the anhepatic stage, the three major hepatic veins are cross-clamped with a vascular clamp. Normal venous return to the heart is maintained because the lumen of the vena cava is only partially occluded and the portocaval anastomosis is still patent. phase was defined as the time between cross-clamping the portal vein and reperfusion of the graft. During the first 3 postoperative days (PODs), measurements of liver enzymes (alanine aminotransferase, aspartate aminotransferase), prothrombin time, bilirubin, platelet count, hemoglobin, urea, and creatinine were recorded. The incidence of primary nonfunction, days of ventilation, intensive care unit stay, hospital stay, vascular complications, re- OLT incidence, and survival also were evaluated. Immunosuppressive Regimen All patients were administered a fourfold immunosuppression regimen that included thymoglobulin, low-dose steroids, azathioprine, and tacrolimus. Tacrolimus was administered orally, starting with 2 mg, then continuing with 0.1 mg/kg of body weight if diuresis was normal and there was no renal insufficiency. Thereafter, tacrolimus dose was adjusted to clinical needs and controlled by whole levels. A methylprednisolone bolus of 500 mg was administered at the time of reperfusion. Prednisone was administered starting with 0.5 mg/kg POD 1 and decreasing to 20 mg POD 4, then individually tapered for each patient. In addition, all patients were administered thymoglobulin (Atgam; Pharmacia Upjohn, Barcelona, Spain; 10 mg/kg) as a continuous intravenous infusion over 4 hours PODs 0 to 5. From POD 5, all recipients were administered 50 mg of azathioprine orally. Statistical Analysis All groups were evaluated according to intent to treat. Data are expressed as mean SD, and groups were compared by means of Wilcoxon s rank-sum test or Mann-Whitney U test for continuous variables and P 2 test or Fisher s exact test for categorical variables. Differences are considered statistically significant at P less than.05. MAP during the anhepatic phase was used to calculate the needed sample size. Considering an error of 0.05 and error of 0.05 and trying to detect an MAP difference of 20 mm Hg, it was calculated that we required 40 patients in each group. To establish whether all patients or only a selected group derived benefits from a PCS, data were analyzed first by comparing all patients and next in several patient subgroups. To study the influence of portal pressure and portal flow on PCS, subgroups were formed according to portal flow and portacaval pressure gradient, with the median of these parameters

4 Temporary Portocaval Shunt in Liver Transplantation 907 considered the cutoff point. Follow-up finished 4 months after OLT. Results Both groups were similar for patient and donor characteristics and details of other parameters before OLT (Table 1). Mean duration of OLT was similar in both groups ( v minutes; P.3). Nevertheless, the anhepatic phase was longer in the PCS group than the no-pcs group ( v minutes; P.005). The temporary portocaval anastomosis was maintained for minutes. Hemodynamic Data Baseline values for HR, CO, MAP, MPP, portal pressure, CVP, SVR, and PVR did not differ for the two groups. Table 2 lists the hemodynamic evolution during OLT of the two groups. During the anhepatic phase (T3), CO decreased in both groups. However, the decrease in CO was significantly less in the PCS group than no-pcs group (P.05). MAP was maintained throughout the anhepatic phase because of increased HR and SVR. However, the proportional increase in Table 1. Characteristics of Patients and Donors: Comparison Between Groups PCS No-PCS P Clinical data Ascites 23/40 (57) 26/40 (65).65 Encephalopathy 10/40 (25) 15/40 (37).32 Previous GI hemorrhage 10/40 (25) 9/40 (22).9 HCV 25/40 (62) 24/40 (60).82 HBV 1/40 (2) 4/40 (10).36 Abdominal surgery* 14/40 (35) 11/40 (27).44 Child s class (A/B/C) 13/18/9 10/19/11.66 Age (yr) Sex (M/F) 25/15 28/12.64 Donor data Cross-match 1/40 (2) 1/40 (2).9 IgG CMV 31/40 (77) 33/40 (82).79 Sodium (mmol/l) Age (yr) ICU stay (d) NOTE. Numbers in parentheses are percentages. Abbreviations: GI, gastrointestinal; HCV, hepatitis C virus positive; HBV, hepatitis B virus positive; IgG CMV, immunoglobulin serological tests positive for cytomegalovirus; ICU, intensive care unit. *Frequency of previous right upper quadrant surgery. SVR was less in the PCS group than no-pcs group, although the difference was not significant (P.07). Portal pressure decreased significantly more during the anhepatic phase (T3) in the PCS group than the no-pcs group (P.001). At the end of the anhepatic phase (T4), MAP, CVP,and PCP increased significantly from the previous determination. This increase may have been related to fluid overload before unclamping (Table 2). After unclamping (T5), CO, MPP, CVP, and PCP increased significantly from the previous period, without differences between the two groups. MAP, SVR, and portal pressure decreased, showing the typical features of reperfusion syndrome, without differences between the two groups (Table 2). Sixty minutes after reperfusion (T6), all hemodynamic parameters returned to baseline values, except portal pressure, which was less than the T0 values because of the disappearance of portal hypertension (Table 2). Hemodynamic changes after unclamping the shunt in the PCS group are listed in Table 3. There was a significant decrease in portal pressure (P.0005). MAP, HR, SVR, and PVR increased significantly. However, there was only a 10% decrease in CO, and filling pressures (MPP, CVP, and PCP) were maintained. Reperfusion Syndrome, Blood Transfusion, and Renal Function After unclamping, 24 patients (30%) presented a decrease greater than 50% in MAP for more than 1 minute (reperfusion syndrome). There were no differences between the two groups (12 patients, PCS group; 12 patients, no-pcs group; P.8). Mean amount of blood component requirements of the two groups did not differ significantly (Table 4). Nevertheless, it should be noted that in the PCS group, more patients underwent OLT without red blood cell transfusion than in the no-pcs group (P.05). Urinary output during OLT was similar in both groups, but during the anhepatic phase, it was significantly greater in the PCS group (P.005; Fig. 3). Postoperative Evolution Liver, renal, and hematologic parameters during the first 3 PODs did not show differences between the two groups. Postoperative serum creatinine levels were stable in the PCS group, whereas in the no-pcs group, they increased until 48 hours after OLT (Fig. 4). Three patients required either hemodialysis or hemofiltration

5 908 Figueras et al Table 2. Hemodynamic Parameters During OLT T0 T1 T3 T4 T5 T6 MAP (mm Hg) PCS * * * No-PCS * * HR (beats/min) PCS * * * * No-PCS * * CO (L/min) PCS * * * No-PCS * * * MPP (mm Hg) PCS * * No-PCS * * * PP (mm Hg) PCS * * * * * No-PCS * * * CVP (mm Hg) PCS * * * No-PCS * * * * PCP (mm Hg) PCS * * * No-PCS * * * * SVR (dynes s cm 5 ) PCS * * * No-PCS * * * PVR (dynes s cm 5 ) PCS No-PCS * * * Abbreviations: T0, beginning of OLT; T1, after portal vein clamping; T3, after hepatic vein clamping; T4, just before portal declamping; T5, 5 minutes after reperfusion; T6, 60 minutes after reperfusion; PP, portal pressure. *Differences from previous period in same group, P.05. Differences between groups in same period, P.05. Table 3. Hemodynamic Parameters After Unclamping the PCS in the PCS Group T0 T2 P MAP (mm Hg) HR (beats/min) CO (L/min) MPP (mm Hg) PP (mm Hg) CVP (mm Hg) PCP (mm Hg) SVR (dynes s cm 5 ) PVR (dynes s cm 5 ) Abbreviations: T0, beginning of OLT; T2, after PCS shunt unclamping; PP, portal pressure. (PCS group, one patient; no-pcs group, two patients; P.55). Six patients presented primary graft dysfunction (PCS group, four patients; no-pcs group, two patients; P.64). One patient in the no-pcs group required urgent re-olt because of primary nonfunction. Table 4. Blood Component Requirements PCS Group No-PCS Group RBCs (units) Fresh frozen plasma (units) Platelets (units) No transfusion (RBCs) 18/40 (45) 9/40 (22).05 NOTE. Numbers in parentheses are percentages. Abbreviation: RBC, red blood cells. P

6 Temporary Portocaval Shunt in Liver Transplantation 909 Figure 4. Comparison of postoperative creatinine levels between groups. ( ), PCS group; (Œ), no-pcs group. *P <.05 versus previous period. Hospital stays were similar in both groups (PCS group, days v no-pcs group, days; P.45), and the intensive care unit stay was longer in the no-pcs group (PCS group, days v no-pcs group, days; P.04), as well as days of intubation (PCS group, days v no-pcs group, days; P.05). The postoperative mortality rate was 2.5% (one patient in the PCS group died of complications of primary graft dysfunction, and one patient in the no-pcs group died of sepsis). The overall survival rate after a 4-month follow-up was 97.5% in both groups. The incidence of vascular complications was similar in the two groups. There was no postoperative portal vein thrombosis, whereas there were three cases of arterial thrombosis (PCS group, one case; no-pcs group, two cases; P.55). There was one case of postoperative hepatovenous outflow obstruction syndrome at the site of the anastomosis in the PCS group that required surgical treatment. The incidence of biliary complications was 10%, with no differences between groups (PCS group, three cases; no-pcs group, five cases; P.45). Figure 3. Comparison of urinary output during OLT (global [ ] and anhepatic phase [ ]) between groups. Numbers within brackets are P values. Subgroups of Patients With Severe Portal Hypertension Thirty-five patients (59% of patients with portal flow measured) had a portal flow measurement at the beginning of the procedure of the median value (1,000 ml/min) or greater, whereas 45 patients (56%) had a pressure gradient between the vena porta and right atria of the median value (16 mm Hg) or greater. Nineteen patients (24%) presented both characteristics. During the anhepatic phase (T3), the decrease in CO was significantly less in the PCS group, mainly in those patients with greater preoperative portal flow (P.03). The increase in SVR needed to maintain CO

7 910 Figueras et al Table 5. Hemodynamic Changes During the Anhepatic Phase of Patients With Severe Portal Hypertension Depending on Portal Flow and Portocaval Gradient Hemodynamic Data PCS No-PCS P All patients CO decrease (%) SVR increase (%) Portal flow 1,000 ml/min* CO decrease (%) SVR increase (%) Portocaval gradient 16 mm Hg* CO decrease (%) SVR increase (%) Both characteristics CO decrease (%) SVR increase (%) *At the beginning of OLT. was less in the PCS group, but only reached statistical significance in those patients with a greater portocaval gradient (P.007; Table 5). The proportion of patients who underwent OLT without transfusion was significantly greater in all analyzed subgroups of the PCS group: patients with portal flow of 1,000 ml/h or greater (PCS group, 11 of 18 patients; 61% v the no-pcs group, 4 of 17 patients; 23%; P.04), patients with portocaval gradient of 16 mm Hg or greater (PCS group, 9 of 20 patients; 45% v the no-pcs group, 3 of 24 patients; 11%; P.01), and both characteristics (PCS group, 4 of 6 patients; 67% v the no-pcs group, 2 of 13 patients; 15%; P.04). Urinary output rates during all OLTs were greater in the PCS group, but only in those patients with both high portal flow and high portocaval gradient were the differences statistically significant (P.03; Fig. 3). Urinary output rates during the anhepatic phase were also greater in the PCS group, mainly in those patients with a high portocaval gradient (P.0005) or both characteristics (P.03; Fig. 3). Discussion The standard procedure for OLT remains transplantation of the whole organ together with resection of the vena cava and the use of venovenous bypass. 4,20 However, venovenous bypass is time consuming, and specific drawbacks have been reported. 5,6 The piggyback technique has many advantages over the traditional technique. First, it does not require dissection of the retrocaval compartment, thus reducing retroperitoneal blood loss. 11,12 Second, this technique makes it easy to solve problems posed by different recipient and donor vena cava sizes, particularly in pediatric, reducedsize, split, or living donor liver transplantation. 9,20-22 Finally, normal venous return to the heart is possible during the anhepatic stage, making venovenous bypass unnecessary. 2,3,10-13,15,16,18,19 However, the vena cava preservation technique does not avoid splanchnic congestion and portal hypertension secondary to vena porta clamping. This study shows that adding a temporary PCS to the piggyback technique can minimize hemodynamic instability and preserve renal function during the anhepatic stage. When results of the PCS were compared in the subgroup of patients with a preoperative high portal flow and high portocaval gradient, the superiority of the PCS was even more striking. In these subgroups, there was an even greater proportion of patients who did not require blood transfusion. The decrease in CO and compensatory increase in SVR was less than that in patients in the no-pcs group. Finally, urinary output during both the entire procedure and the anhepatic phase was maintained better than in patients in the no-pcs group. Completion of an end-to-side PCS required little time when hilar structures were dissected and did not affect the completion of the portal anastomosis. 17 In this series, one could argue that the completion of a temporary PCS is time consuming. However, we have shown that the duration of the entire surgical procedure was similar in both groups. Most of the time spent performing the PCS is regained afterward, disconnecting the recipient liver and retrohepatic vena cava in a dry surgical field without portal hypertension. This study and others 13 show that OLT can be performed with minimal blood product requirements or even without red blood cell transfusion. Recipients with less advanced disease, meticulous surgical technique, avoiding the need for temporary vena cava clamping, and greater experience of the transplant team are important factors. Using a temporary PCS is a further technical refinement that can avoid hemodynamic instability during the anhepatic phase and retroperitoneal bleeding secondary to portal hypertension, and thus the need for massive transfusion and fluid overload. This method also is safe from the nephrological point of view because serum creatinine levels are stable during the early postoperative period (Fig. 4). The lower transfusion rate, together with hemodynamic stability and absence of fluid overload during the anhepatic stage, may explain the significant reduction in artificial ventilation time and the shorter intensive care unit stay in the PCS group.

8 Temporary Portocaval Shunt in Liver Transplantation 911 In conclusion, the use of a temporary PCS during OLT improves hemodynamic status, reduces intraoperative transfusion requirements, and preserves renal function during and after OLT. However, clinical benefits of this technique are more evident in patients with a baseline portal flow of 1,000 ml/min or greater and those with severe portal hypertension and a portocaval gradient of 16 mm Hg or greater. Acknowledgment The authors thank the anesthetists on the liver transplant team for help with hemodynamic measurements, the surgical nurses for help and their patience during this study, and Eduardo Saiz for drawings. References 1. Starzl TE, Putnam CW. Experience in hepatic transplantation. Philadelphia: Saunders, Figueras J, Sabate A, Fabregat J, Torras J, Drudis R, Rafecas A, et al. Hemodynamics during the anhepatic phase in OLT with vena cava preservation. A comparative study. Transplant Proc 1993; 25: Taura P, Beltran J, Garcia-Valdecasas JC, Lopez A, Fontanals J, Zavalo E, et al. Renal function during liver transplantation with preservation of the recipient s inferior vena cava. Transplant Proc 1994;26: Shaw BW Jr, Martin DJ, Marquez JM, Kang YG, Bugbee AC Jr, Iwatsuki S, et al. Venous bypass in clinical liver transplantation. Ann Surg 1984;200: Khoury GF, Kaufman RD, Musich JA. Hypothermia related to the use of veno-venous bypass during liver transplantation in man. Eur J Anaesthesiol 1990;7: Khoury GF, Mann ME, Porot MJ, Abdul-Rasool IH, Busuttil RW. Air embolism associated with veno-venous bypass during orthotopic liver transplantation. Anesthesiology 1987;67: Calne RY, Williams R. Liver transplantation in man. Observations on technique and oragnization in five cases. BMJ 1968;4: Tzakis A, Todo S, Starzl TE. Orthotopic liver transplantation with preservation of the inferior vena cava. Ann Surg 1989;210: Ringe B, Pichlmayr R, Burdelski M. A new technique of hepatic vein reconstruction in partial liver transplantation. Transpl Int 1988;1: Parrilla P, Sánchez-Bueno F, Figueras J, Jaurrieta E, Mir J, Margarit C, et al. Analysis of the complications of the piggy-back technique in 1112 transplants. Transplantation 1999;67: Lerut JP, Molle G, Donataccio M, DeKock M. Cavocaval liver transplantation without bypass and without temporary portocaval shunting. The ideal technique for adult grafting? Transpl Int 1997;10: Jovine E, Mazzioti A, Grazi GL, Ercolani E, Masetti M, Morganti M, et al. Piggy-back versus conventional technique in liver transplantation: Report of a randomized trial. Transpl Int 1997; 10: Cacciarelli TV, Keefe EB, Moore DH, Burns W, Chuljian F, Busque S, et al. Primary liver transplantation without transfusion of red blood cells. Surgery 1996;12: Tzakis AG, Reyes J, Nour B, Marino IR, Todo S, Starzl TE. Temporary end-to-side portocaval shunt in orthotopic hepatic transplantation in humans. Surg Gynecol Obstet 1993;176: Belghiti J, Noun R, Sauvanet A. Temporary portocaval anastomosis with preservation of caval flow during orthotopic liver transplantation. Am J Surg 1995;169: Cherqui D, Lauzet J, Rotman N, Duvoux C, Dhumeaux D, Julien M, et al. Orthotopic liver transplantation with preservation of the caval and portal flow. Transplantation 1994;58: Belghiti J, Noun R, Sauvanet A, Durand F, Aschehoug J, Erlinger S, et al. Transplantation for fulminant and subfulminant hepatic failure with preservation of portal and caval flow. Br J Surg 1995;82: Steib A, Saada A, Clever B, Lehmann C, Freys G, Levy S, et al. Orthotopic liver transplantation with preservation of portocaval flow compared with venovenous bypass. Liver Transpl Surg 1997;35: Hesse UJ, Berrevoet F, Troisi R, Mortier E, Pattyn P, DeHemptinne B. Liver transplantation by preservation of the caval flow with temporary porto-caval shunt or veno-venous bypass. Transplant Proc 1997;29: Neuhaus P, Platz KP. Liver transplantation. Newer surgical approaches. Baillieres Clin Gastroenterol 1994;8: Lerut J, de Ville de Goyet J, Donataccio M, Reding R, Otte J. Piggyback transplantation with side-to-side cavocavostomy is an ideal technique for right liver allograft implantation. J Am Coll Surg 1994;179: Marcos A, Ham J, Fisher R, Olzinski AT, Posner HP. Surgical management of anatomical variations of the right lobe in living donor liver transplantation. Ann Surg 2000;231: