Simple Test on the Back Table for Justifying Single Hepatic-Arterial Reconstruction in Living Related Liver Transplantation 1

In living related liver transplantation (LRLT), a graft occasionally has multiple hepatic arterial branches (HABs), depending on where the HABs feeding a graft are divided (1–3). In this situation, however, there are no definite criteria for how many branches should be reconstructed before reconstruction is performed (2). We describe a simple test on the back table for justifying single hepatic arterial reconstruction (HAR) in LRLT. From January 1996 to July 1999, 64 LRLTs were performed in 63 patients at our department. The recipients were 27 males and 36 females, with a mean age of 16 years (range: 0.8−62 years). In donors, all HABs to the left hepatic lobe were divided at their roots (4). On the back table, the grafts were perfused through the portal vein with 1 L of cold lactate Ringer’s solution followed by 1 L of cold University of Wisconsin solution. At the same time, the HABs were irrigated with 10 ml of cold lactate Ringer’s solution and then University of Wisconsin solution. The connective tissue around an arterial stump was grasped carefully; a 24-gauge Teflon catheter was inserted into the lumen so as to not damage the intima; the arterial wall close to the stump was grasped gently with forceps or the fingers; and then each fluid was infused. When there were two or three branches, the presence of back flow from the other branches at the time of infusion was observed carefully. The same procedure was, if possible, repeated for the other branches. When back flow was confirmed, the thickest branch alone was reconstructed. HAR was performed under the optical field of a continuous zoom operating microscope at a magnification of approximately 10 times. After completing HAR, the presence of back flow from a nonanastomosed arterial stump was confirmed, and then arterial signals in the intrahepatic branches to each segment were examined by color Doppler ultrasound findings. Left hepatic lobectomy, extended left lateral segmentectomy, and left lateral segmentectomy were performed for 30, 9, and 25 donors, respectively. Thirty-nine (61%) of the 64 grafts had a single HAB, and the remaining 25 (39%) had multiple HABs (Table 1). Table 1: Details of the hepatic arterial branches of 64 grafts{tabft}a HA, hepatic artery; A2, A3 and A4, Hepatic arteries to the segments 2, 3, and 4, respectively; LHL, Left hepatic lobectomy graft; ext LS, extended lateral segmentectomy graft; LS, lateral segmentectomy graft; AHA, aberrant left hepatic artery from the left gastric artery; SLHA, Small branch of the left hepatic artery. b The arterial branches which were reconstructed.Back flow was confirmed on the back table for 24 of the 25 grafts, whereas no back flow was observed in the remaining graft. Back flow bleeding from the nonanastomosed branch was observed after single HAR for all 24 grafts, and intraoperative Doppler ultrasound showed pulsatile arterial wave forms in the respective segmental branches. However, in 2 of the 24 grafts, a 2nd HAR was performed for safety. These cases were two of our initial cases. In the last graft without back flow on the back table, there was neither back flow from the nonanastomosed branch after HAR, nor a pulsatile arterial wave form on Doppler ultrasound in the artery supplying segment 4. Subsequently, the 2nd branch or the middle hepatic artery was reconstructed, and a pulsatile arterial wave form was obtained in the branch to segment 4. Overall, in 22 grafts with multiple HABs, single HAR was performed (group I), whereas in the remaining 42 grafts with single or multiple HABs, all the branches were reconstructed (group II). Postoperatively, there were no differences between the two groups in the maximum levels of serum aspartate oxoglutarate aminotransferase, alanine oxoglutarate aminotransferase, or lactate dehydrogenase within the first 24 hr after transplantation. Hepatic arterial thrombosis (HAT) occurred in 3 patients in each group (9%) (median: 7th postoperative day, range: 4–23). Four patients underwent thrombectomy followed by re-anastomosis with sufficient arterial flow, and consequently the grafts were rescued. In one patient, HAT was treated conservatively. The one remaining patient died of graft failure as a result of simultaneous thrombosis of both the portal vein and hepatic artery. In the remaining 58 grafts, there were no problems with postoperative hepatic-arterial flow. Biliary stenosis occurred in 4 and 2 patients of groups I and II, respectively (median: 170th postoperative day, range: 87–349), but there was no difference in the incidence between the 2 groups (Fisher’s exact probability: 0.155). Furthermore, there were no statistically significant differences in the serum total bilirubin (median: 0.9 vs. 0.4 mg/dl) and alkaline phosphatase levels (median: 756 vs. 738 U/L) at 1 year after transplantation between the 2 groups. Finally, there was no significant difference in the 1-year graft survival rates (89.9% vs. 91.7%) between the two groups. To revascularize the graft satisfactorily, it is not always necessary to reconstruct all arteries that are considered to feed a graft from the viewpoint of anatomical arterial distribution. The necessity for a second HAR after a former HAR can be determined by intra-operative Doppler ultrasound findings and the presence of back flow bleeding from the stump (2). Our results indicate that the necessity for a second HAR can be determined readily on the back table. Even if back flow bleeding is weak and the flow speed determined by Doppler ultrasound findings is slow, the flow speed increases with an improved wave form at the end of surgery, consequently increasing the size of the reconstructed artery (2), and a 2nd HAR was not required in 22 of the 25 grafts with multiple HABs. Retrospectively, a second HAR might not have been necessary in two of the grafts given two HARs for safety. The graft with no back flow from the non-anastomosed branch may have been the only case that really required a second HAR. HAT is one of the most serious complications after LRLT and can result in graft loss (5). The rate of occurrence of HAT showed no relationship with the manner of arterial reconstruction. Finally, 22 patients with single HAR for multiple HABs showed no liver ischemia. Ikegami et al. (2) also explained that patients in whom only one of two or three HABs feeding the grafts was reconstructed had no episode of liver damage attributable to ischemic change in the graft during the postoperative course. Furthermore, the long-term results, including the incidence of biliary stenosis, the serum levels of total bilirubin and alkaline phosphatase at 1 year after LRLT, and the 1-year survival rates, support the use of single HAR for grafts with multiple HABs. In conclusion, our simple test on the back table correctly justifies single HAR, thus helping to avoid the need for a second HAR. K. Kubota2 M. Makuuchi T. Takayama Y. Harihara K. Hasegawa T. Aoki H. Asato H. Kawarasaki