Fig. 17.1
A 60-year-old man presented with a liver metastasis adjoining the right anterior branch of the portal triad. The indocyanine retention rate at 15 min was 8.5%, which indicated normal liver functional reserve. (a) A simulation 3D-image demonstrating the relationship between the tumor and the portal triad. (b) A simulation 3D-image of the right anterior sectionectomy supposed that 13.3% of the total liver parenchyma would be resected
When the main trunk of the portal vein or its first-order branch seems to abut the tumor, a hemi-hepatectomy, i.e., a right or left hepatectomy, should be considered, and the combined resection of the wall of the main portal vein may be required. In such cases, the combined resection and the following choledocho-jejunostomy may be also required, although much less frequently, and the indication should be limited from the oncological standpoint [5]. Indeed, the management of the portal vein, hepatic artery, and bile duct when a CLM tumor is located near the hepatic hilum, is almost the same as that in hilar bile duct carcinoma. When the tumor does not invade the whole circumference of the portal vein, a wedge resection of the wall is suitable (Fig. 17.2a). A primary closure may be enough to secure intrahepatic portal flow for a small defect of the venous wall (Fig. 17.2b); otherwise a patch repair is indicated (Figs. 17.2c and 17.3). The suture direction in a primary closure should be vertical, rather than longitudinal, to the portal axis in order to prevent stenosis of the portal vein (Fig. 17.2b). When the tumor invades the whole circumference or the area in contact with the tumor is relatively large, a segmental resection of the portal vein followed by either direct end-to-end anastomosis or interposed reconstruction using a tubular graft may be indicated (Fig. 17.2d). Compared with the caval flow, portal flow is more sensitive to the area of the anastomotic lumen, and meticulous care should be devoted to prevent stenosis during the suture. For example, the suturing thread should be pulled gently, and some surgeons prefer to make so-called “growth factor” finishing anastomosis (Fig. 17.2e).
Fig. 17.2
The schematic views indicating the reconstruction methods of portal vein. (a) A wedge resection is indicated when a tumor invades the portal vein on a narrow area. (b) Primary closure of the defect of the portal vein. The suture line should be vertical against the long axis of the portal vein. (c) A patch repair should be performed when the defect of the portal vein is large. (d) A circumferential segmental resection is indicated when a tumor invades the portal vein a large area. (e) An end-to-end anastomosis after segmental resection followed by “growth factor” to prevent stenosis
Fig. 17.3
A 63-year-old man with colorectal liver metastasis invading hepatic hilum underwent left hepatectomy. Combined portal venous and extrahepatic bile duct resections followed by reconstruction were accompanied. A patch graft using cryo-preserved allograft of portal vein was used
When a CLM tumor is located near the hepatic hilum, as in hilar bile duct carcinoma, the reconstruction of the hepatic artery as well as portal vein is sometimes required. It is usually accompanied with hemi-hepatectomy, and the hepatic arterial reconstruction is required when a tumor invades the hepatic arterial branch feeding the hepatic remnant. Radical surgery is contra-indicated when the hepatic artery which is subject to reconstruction cannot be encircled at the distal side of the tumor, which should first be confirmed after the laparotomy. Arterial reconstruction may be performed in direct end-to-end anastomosis between the stumps of the hepatic arteries; otherwise, an arterial pedicle graft must be prepared using, for example, the right gastroepiploic artery, the gastroduodenal artery, the middle colic artery, and the left gastric artery.
Hepatic Vein Resections
When the wall of the major hepatic vein abuts a CLM tumor, the surgeons must preoperatively decide whether to sacrifice or to reconstruct the vein taking the impact of hepatic congestion into consideration, as described in the following paragraph. When hepatic venous reconstruction is needed, the method of reconstruction should be chosen from among primary suture repair, patch reconstruction, and interposed graft reconstruction. Because of the small diameters of the hepatic veins, a wedge resection followed by simple primary closure is difficult or, if possible, may cause hepatic congestion. Therefore, a patch reconstruction following a wedge resection and a circumferential segmental resection, followed by a reconstruction using an interposed graft (Fig. 17.4) are frequently adopted. Few reports have assessed the long-term survival of CLM patients undergoing a combined hepatic venous resection. Aoki et al. reported a median survival time of 26 months, which is not favorable but not hopeless, after the combined resection of the hepatic venous confluence in the seven patients [6]. Saiura et al. analyzed 16 patients with hepatic venous reconstruction, in whom a total of 18 hepatic veins were reconstructed, and reported a much better prognosis, that is, a 5-year overall survival rate of 76% [7].
Fig. 17.4
A 68-year-old patient with colorectal liver metastasis underwent an extended left hepatectomy with combined resection and reconstruction of the middle hepatic vein. Because the middle hepatic vein was cut at its root and at the periphery, a tubal cryo-preserved allograft of the portal vein was interposed between the peripheral cut-end of middle hepatic vein and the vena cava. The graft has been worked for one and a half years
Resection of the major hepatic vein or its main tributary may cause congestion of the corresponding drained volume of the liver [8]. This congestive state had been merely a concern until graft necrosis and intractable massive ascites after liver transplantation was related to hepatic outflow block, namely in the late 1990s [9–11]. Recently developed computer software simulating hepatic resection by incorporating contrast-enhanced CT images into 3D-images has enabled the hepatic volume drained by a certain venous tributary to be calculated (Fig. 17.5). However, the proportion of the hepatic functional reserve that is impaired by venous congestion remains to be resolved. To secure patient safety, the hepatic functional reserve of the congested volume may be supposed to be zero; however, Sano et al. revealed that the portal vein may exhibit regurgitation after the occlusion of the hepatic vein, to partly substitute the drainage effects [8]. A clinical study comparing the uptake of liver-specific contrast agent during magnetic resonance imaging suggested an unimpaired hepatic functional reserve of approximately 65% in the congested hepatic area [12], while another study comparing the uptake of indocyanine green using fluorescence imaging suggested a reserve of approximately 40% [13]. As the results of the two reports differed and the validities of their methodologies have not been established, further investigation of this issue is needed. Nonetheless, it is important to note that the congestive area may have reduced, but not necessarily obliterated, hepatic functional reserve.
Fig. 17.5
A 49-year-old patient who had undergone a resection of sigmoid colon carcinoma presented with liver metastases, one of which adjoined the middle hepatic vein. A simulation using computer software was performed as the middle hepatic would be sacrificed, supposing that 47.9% of the total liver parenchyma was to be resected in right hepatectomy and an additional 18.7% of the total liver parenchyma was to be congested within the future hepatic remnant. Thus, 29.2% of the total liver parenchyma would be the future non-congested hepatic remnant, showing the procedure would be acceptable considering that the indocyanine retention rate at 15 min was 8.8%