Orthotopic liver transplantation (OLT* ) is a routine therapeutic alternative in the management of irreversible hepatic diseases. The most commonly used surgical technique was described by Starzl et al. (1) in 1963, with the existence of an anhepatic phase with removal of the retrohepatic vena cava. In this phase there is a decrease in venous return to the heart that, in most cases, causes hemodynamic instability, metabolic alterations, and low renal flow. Occasionally it is necessary to fit a femoro-porto-axillary venovenous bypass (2) . In contrast to this classical surgical technique, Calne (3) in 1968, reported a hepatectomy with recipient retrohepatic vena cava preservation to maintain venous return to the heart, which avoids hemodynamic alterations. Since then other authors (4-8) have made several technical modifications and this technique, known as piggy-back OLT, has gained acceptance in adult OLT during the last few years, especially in European countries. At the moment, however, there is controversy over advantages or specific complications (suprahepatic thrombosis or narrowing, etc.) related to this surgical technique (9-17) .
The aim of this study is to know of the immediate per-and postoperative morbidity and mortality rates in 1112 OLTs performed with a vena cava preservation (VCP) technique.
PATIENTS AND METHODS
All 14 liver transplant units in Spain were sent a questionnaire on retrohepatic vena cava preservation during orthotopic liver transplantation. Only seven hospitals (50%) returned the questionnaire and were as follows: "Principes de España" Hospital in Bellvitge (Barcelona), La Fé (Valencia), Valle de Hebrón (Barcelona), Marqués de Valdecilla (Santander), Xeral (Santiago), Ramón y Cajal (Madrid), and "Virgen de la Arrixaca" (Murcia).
The postoperative VCP-related complications were divided into two groups: (1) early, if in the first week postoperatively; (2) late, if after this time. The complications not related to VCP were excluded in the description of results.
To compare the incidence of complications between the anastomosis performed with two or three suprahepatic veins the χ2 test and Fisher exact test (for small numbers) were used. Differences were considered statistically significant when P <0.05.
RESULTS
The number of OLTs that had been performed in the seven centers because the beginning of the program was 1674, with the VCP technique used in 1112. Having incorporated VCP into their therapeutic arsenal, six of the seven centers now use it routinely in more than 95% of all transplants done. Venovenous anastomosis was done on the patch obtained by joining the recipient's central and left suprahepatic veins in 440 cases (39.5%) and on the patch obtained by joining the three suprahepatic veins in the remaining 672 cases (60.5%). Except for one group that does so occasionally (six cases), none of the units performs a previous temporary portocaval anastomosis.
Intraoperative complications. Twenty-eight patients (2.5%) had intraoperative complications related to the VCP technique, which were treated during the operation. Two of the 28 patients had traumatic major tears of the recipient vena cava, which needed venous suture. The remaining 26 patients, after portal revascularization, presented with congestion of the graft due to mislocation and/or rotation of the same (22 of 26 had a recipient's central and left suprahepatic vein anastomosis). Five of the 26 received a "face-to-face" (laterolateral cavocaval) venous anastomosis, two had a resection and venous reanastomosis, and the remaining 19 were given a "neo-bed" due to a discordance between the size of the graft and the hepatic bed.
Early postoperative complications (within first week). Eleven patients (1%) had postoperative complications during the first week (Table 1) . In nine cases there was an acute Budd-Chiari syndrome (ABCS) in the first 48 hr, associated in one case with caval thrombosis (confirmed by cavography), seven of nine cases being a recipient's central and left suprahepatic vein anastomosis. Seven of the nine patients with ABCS underwent emergency retransplantation, after reoperation in one with a distal end-to-side cavocaval anastomosis. The patient with a caval thrombosis was treated with thrombolysis using streptokinase and the remaining patient did not undergo retransplantation, due to unavailability of a graft, and died while on the waiting list. The remaining two cases presented with intraabdominal bleeding. Both patients underwent reoperation for hemostasia of the collateral rami of the cava, and one of the patients also had severe graft dysfunction and required emergency retransplantation (4th day).
Table 1: Early and late postoperative complications
Late postoperative complications (after first week). Three patients, all with venovenous anastomosis done on the recipient's central and left suprahepatic veins, developed symptoms of massive ascites between 2 and 3 months, with patency of the retrohepatic cava (verified by cavography). A hemodynamic study revealed a hyperpressure at the suprahepatic veins. This chronic BCS was controlled in all patients with diuretics (Table 1) .
Comparative study between "two" and "three" suprahepatic vein anastomosis. As shown in Table 2 , both intraoperative venous congestion of the graft and acute or chronic BCS were more frequent when using the patch obtained by joining the central and left suprahepatic veins than when using the patch obtained with the three suprahepatic veins (P <0.05).
Table 2: Complications according to the use of two or three suprahepatic veins to perform the venous anastomosis
Postoperative mortality. Only six patients (0.5%) died as a result of the "piggy-back" technique related complications. The causes of death are shown in Table 1 .
DISCUSSION
Despite anesthetics and critical care progress, the anhepatic phase with the classical technique occasionally involves a hemodynamic drama which in the short and medium term affects postoperative morbidity and mortality. In 1984, the introduction of venovenous bypass by Shaw et al. (2) , in Pittsburgh, enabled orthotopic transplantation on patients who had hemodynamic instability during the anhepatic phase. In 1989, Tzakis et al. (4) published a hepatectomy with the VCP technique in pediatric OLT select cases, maintained venovenous bypass and reported a prolonged operating time due to the enormous technical difficulty involved in the dissection of the retrohepatic vena cava when ligating the numerous venous rami of the caudate lobe. In 1992, Jones et al. (5) published a study of 38 liver transplants divided into three groups (VCP with/without bypass versus the classical technique) and showed that operating time is similar with the three techniques, and that with VCP without bypass the transfusion requirements and hemodynamic repercussions are reduced, which leads them to claim that VCP without bypass can be performed in most patients undergoing liver transplantation. In 1992, Belghiti et al. (6) modified the VCP technique by reestablishing venous flow through a laterolateral cavocaval (face-to-face) anastomosis. In 1993, Tzakis et al. (7) introduced a new VCP modifícation in child patients using a temporary terminolateral portocaval shunt. They consider that the absence of established portosystemic collateral circulation causes technical difficulties with the external venovenous bypass, because the bypass produces a low blood flow, and there is secondary thrombosis and air embolism. Two years later, Belghiti et al. (8) proposed temporary portocaval shunt as a routine in all hepatectomies.
Over the last 2 years, several series have been published on VCP (9, 10, 13, 17) , without external or internal bypass, with minimal hemodynamic alteration in the anhepatic phase, a lower consumption of blood products during the operation and the low number of technique-related surgical complications. However, some authors (11) have reported VCP-related intraoperative complications. These were major venous tears in the recipient's cava vein and congestion of the graft due to mislocation of the same after anastomosis to the mid- and left suprahepatic veins. In our study, we found only 28 cases of intraoperative complications that could be attributed to the technique, which accounts for 2.5%. The most common intraoperative complications was venous congestion of the graft (26 cases) due to the piggy-back liver compressing the venous anastomosis rather like a ball valve, which occurs usually after performing anastomosis to the patch of the mid and left suprahepatic veins (22 of 26 patients). On such occasions it might be useful to create a "hepatic neo-bed" by suturing the peritoneum covering Gerota's fascia of the right kidney to the diaphragm, thus managing to reduce the size of the recipient hepatic fossa, because generally they are due to a discordance between the graft and the recipient hepatic bed, although others advocate a laterolateral cavocaval anastomosis in the same operating field (16) . Some authors (18, 19) consider that this mislocation of the graft can be avoided prophylactically if anastomosis is performed to the patch of the three suprahepatic veins, although it is necessary to partially clamp the anterior face of the retrohepatic vena cava, which causes certain hemodynamic alterations that are always well tolerated by the patient (20) . Accordingly, mislocation of the graft in our series was drastically reduced (from 5 to 0.59%) when the patch of the three suprahepatic veins was used. In no case was the use of venovenous bypass necessary.
The lower incidence of complications related to venous outflow of the graft when the patch of the three suprahepatic veins was used could be due to three factors: (1) the diameter of the anastomosis is at least 1 cm longer than when the patch of the mid and left suprahepatic veins is used; (2) the length of the venous segment between the graft and the wall of cava vein is shorter; and (3) with the patch of the mid and left suprahepatic veins the anastomosis is located on the left side of the anterior wall of the cava vein. The later two factors lead to mislocation of the graft when the liver occupies its place in the hepatic fosa, especially when there is discordance between the size of the liver and that of the hepatic fossa.
In the immediate postoperative period (within the first week), the patient could develop an acute BCS due to a poor venous drainage of the graft that might have gone unnoticed during operation. In our series, 1% of the patients had early postoperative complications related to mislocation of the graft not detected during operation, and to abdominal bleeding caused by collateral bleeding of the vena cava. Mislocation of the graft involves severe graft dysfunction that requires immediate surgical correction, using a laterolateral cavocaval anastomosis or emergency retransplantation, which always carries a high rate of morbidity and mortality. If there is thrombosis of the vena cava and/or the suprahepatic veins, the first step is emergency retransplantation, although in one patient we unsuccessfully used thrombolysis with streptokinase. As occurs with intraoperative congestion of the graft, the incidence of acute BCS is reduced significantly when using the three suprahepatic veins (from 1.6 to 0.28%).
Finally, some authors (9, 10, 12) reported VCP-related late complications as a hyperpressure at the exit of the suprahepatic veins, which conditions the presence of ascites, or edemas in the lower limbs due to compression of the recipient retrohepatic vena cava or thrombosis of the donor vena cava and, secondary to this, pulmonary embolism (9, 10, 12) . The presence of ascites (three patients) is a late postoperative complication, which means difficulties in the venous drainage of the graft, as has been seen hemodynamically by several authors (9, 14, 21) . If the ascites is not controlled with diuretics and the clinical symptoms become worse, it may be necessary to perform balloon-probe dilation of the venous anastomosis, or a laterolateral cavocaval anastomosis and, in very advanced cases, an elective retransplant (9, 13, 15) .
In conclusion, with the results of our series, the VCP technique can be used routinely in OLT because it is safe and efficient and involves few surgical complications, especially when for venous reconstruction we use the patch formed by the three suprahepatic veins.
REFERENCES
1. Starzl TE, Marchioro TL, Von Kaula KN, et al. Homotransplantation of the liver in humans. Surg Gynecol Obstet 1963; 117: 659-676.
2. Shaw BW Jr, Martin DJ, Marquez JM, et al. Venous bypass in clinical liver transplantation. Ann Surg 1984; 200: 524-534.
3. Calne RY, Williams R. Liver transplantation in man. I. Observations on technique and organization in five cases. Br Med J 1968; 4: 535-540.
4. Tzakis A, Todo S, Starzl TE. Orthotopic liver transplantation with preservation of the inferior vena cava. Ann Surg 1989; 210: 649-652.
5. Jones R, Hardy KJ, Fletcher DR, Michell I, McNicol PL, Angus PW. Preservation of the inferior vena cava in orthotopic liver transplantation with selective use of veno-venous bypass: the piggy back operation. Transplant Proc 1992; 24: 189-191.
6. Belghiti J, Panis Y, Sauvanet A, Gayet B, Fekete F. A new technique of side to side caval anastomosis during orthotopic hepatic transplantation without inferior vena cava occlusion. Surg Gynecol Obstet 1992; 175: 271-272.
7. Tzakis AG, Reyes J, Nour B, Marino IR, Todo S, Starzl TE. Temporary end to side portacaval shunt in orthotopic hepatic transplantation in humans. Surg Gynecol Obstet 1993; 176: 181-182.
8. Belghiti J, Noun R, Sauvanet A. Temporary portocaval anastomosis with preservation of caval flow during orthotopic liver transplantation. Am J Surg 1995; 169: 277-279.
9. Lerut J. Side to side cavo-cavotomy: a useful aid in "complicated" piggy-back liver transplantation. Transpl Int 1993; 6: 299-304.
10. Fleitas MG, Casanova D, Martino E, et al. Could the piggyback operation in liver transplantation be routinely used?. Arch Surg 1994; 129: 842-845.
11. Meunier B, Bardaxoglou E, Chareton B et al. La technique du "piggy back" en transplantation hépatique. Chirurgie 1993-1994; 119: 682-685.
12. Garcia-Valdecasas JC. Nuevos aspectos técnicos en el trasplante hepático. La conservación de la vena cava del receptor. Gastroenterol y Hepatol 1995; 18: 483-487.
13. Stieber AC. One surgeon's experience with the piggyback versus the standard technique in orthotopic liver transplantation: is one better than the other? Hepatogastroenterology 1995; 42: 403-405.
14. Ducerf C, Rode A, Adham M et al. Hepatic outflow study after piggyback liver transplantation. Surgery 1996; 120: 484-487.
15. Jovine E, Mazziotti A, Grazi GL, et al. Piggy-back versus conventional technique in liver transplantation: report of a randomized trial. Transpl Int 1997; 10: 109-112.
16. Lerut JP, Molle G, Donataccio M, De Kock M, Ciccarelli O, Laterre PF, et al. Cavocaval liver transplantation without venovenous bypass and without temporary portocaval shunting: the ideal technique for adult liver grafting? Transpl Int 1997; 10: 171-179.
17. Busque S, Esquivel CO, Concepcion W, So SKS. Experience with the piggyback technique without caval occlusion in adult orthotopic liver transplantation. Transplantation 1998; 65: 77-82.
18. Figueras J, Sabate A, Fabregat J, et al. Hemodynamic during the anhepatic phase in orthotopic liver transplantation with vena cava preservation: a comparative study. Transpl Proc 1993; 25: 2.588-2.589.
19. Acosta F. Hemodynamic stability during the anhepatic phase of liver transplantation. Influence of the surgical technique:with bypass, without bypass or with piggy-back. Transplantology 1996; 7: 54-56.
20. Lazaro JL, Charco, R, Revhaug A, et al. Hemodynamics in human liver transplantation with inferior vena cava preservation. Transpl Proc 1997; 29: 2851-2852.
21. Detry O, Honore P, Meurisse M, et al. Advantages of inferior vena caval flow preservation in combined transplantation of the liver and heart. Transpl Int 1997; 10: 150-151.
* OLT, orthotopic liver transplantation; VCP, vena cava preservation. Cited Here