Introduction

Anatomic hepatic resection, i.e., removal of a liver segment confined by portal branches, is theoretically a logical procedure for eradication of intrahepatic metastasis of hepatocellular carcinoma (HCC), minimization of postoperative tumor recurrence and enhancement of overall survival of HCC patients [1, 2]. To achieve complete anatomic resection in sectionectomy, hemihepatectomy, or trisectionectomy, it is necessary to grasp the portal pedicles and ensure sufficient exposure of each hepatic vein from the confluence of veins. In such liver resection, transection of the liver parenchyma through an anterior approach without mobilization of the remnant liver is preferable because avoidance of liver rotation has the advantages of circumventing tumor dissemination and/or injury produced by compression of the remnant liver [3, 4]. However, it is sometimes difficult to achieve appropriate transection on both cut lines especially in the presence of a large liver tumor. On the other hand, a longer transection time may increase blood loss.

Belghiti et al. [5] proposed the liver hanging maneuver (LHM) for right hepatectomy without liver mobilization by using a tape inserted between the anterior surface of the vena cava and the liver. Lifting this tape allows easier parenchymal transection in the deeper site and a better control of bleeding. Since the publication of that report, LHM has gained worldwide popularity for major hepatectomy or hepatectomy for large-size liver tumor or invading surrounding tissues [5–9]. Recently, this technique has been applied for various anatomical resections [10]. To our knowledge, the usefulness of LHM assessed by clinical data, such as operating time and amount of blood loss has not been reported. In this study, we retrospectively examined various clinical and surgical parameters in patients who underwent anatomical resection with LHM and compared these parameters with those of patients who underwent liver resection without LHM.

Materials and methods

The study protocol of database access and review was approved by the medical board of Nagasaki University Graduate School of Biomedical Sciences. A signed consent from each patient for studying the clinical data was obtained before surgery. Between October 2000 and July 2007, we performed 71 liver resections (2- or 3-segmentectomies) among 297 liver resections. Since November 2001, we began to apply the method described by Belghiti et al. [5] for LHM. The 71 resections included 24 hepatectomy with LHM between 2001 and September 2007 (LHM group) compared with 47 without LHM between 2000 and 2003 (non-LHM group). The surgical team was the same between 2000 and 2007 period, and two experienced surgeons performed hepatectomy. Age, sex, background liver disease, liver disease, preoperative liver functions (alanine aminotransferase, total bilirubin, prothrombin activity, indocyanine green retention rate at 15 minutes, liver uptake ratio by technetium-99m-galactosyl, human serum albumin, liver scintigraphy, and serum hyaluronic acid) [11], surgical records (operative procedure, operating time, time of liver parenchymal transection, blood loss, blood transfusion), postoperative liver functions (alanine aminotransferase, total bilirubin, prothrombin activity), postoperative complications, clinical outcome, and the duration of hospitalization after operation were collected for analysis. The 24 hepatectomy with LHM included 11 right hepatectomy, 7 left hepatectomy, 1 right trisectionectomy, 1 left trisectionectomy, 1 right anterior sectionectomy, 1 hepatectomy of segment 4 + 5 + 8, and 2 right lateral sectionectomy with the combined resection of the right hepatic vein. The 47 non-LHM group included 24 right hepatectomy, 18 left hepatectomy, 2 left trisectionectomy, 1 right anterior sectionectomy, and 2 right lateral sectionectomy with the combined resection of the right hepatic vein.

Hospital death was defined as death from any cause during hospital stay. Biloma was defined as a bile discharge at the cut surface of the liver. Long-term ascites or pleural effusion was defined as intra-abdominal or intra-thoracic fluid collection under a use of diuretics during 2 weeks. Hepatic failure was defined as total bilirubin level >5 mg/dl at 7 days after operation.

The surgical procedure included a J-shaped incision laparotomy (upper median plus right-sided transverse incision to the 9th intercostal space). The falciform ligament was cut to expose the confluences of the right, middle, and left hepatic veins and the anterior surface of the vena cava. Mobilization of the remnant liver was not performed. LHM was basically conducted according to the method described by Belghiti et al. [5]. The space between the right and middle hepatic veins (RHV and MHV) was dissected by using a right-angled clamp. From this space, loose connective tissue between the anterior surface of the vena cava and the paracaval caudate lobe was dissected over a 3 cm length using a long right-angled clamp because short hepatic veins are absent in this space [12, 13]. Subsequently, the space between the vena cava and infra-hepatic caudate process was dissected and a few short hepatic veins were divided [6]. The loose tissue in this space was dissected using a long, light, and curved Kelly clamp. A 10-Fr size of nasogastric (NG) tube was inserted between the RHV and MHV and passed easily through the dissecting space. In our hands, the tube insertion was completed within approximately 20 min. After the insertion, the tube was lifted up for LHM as shown in Fig. 1. For transection along the right hepatic vein for left trisectionectomy and right anterior sectorectomy, the tube was repositioned in the space between Glisson’s pedicle of the right anterior and posterior sector at the hepatic hilum (Fig. 2). The cut plane along the right hepatic vein was then hung up by the tube. For right trisectionectomy, the tube was repositioned between the confluence of the middle and left hepatic veins and was replaced adjacent to the umbilical Glisson’s pedicle at the hepatic hilum (Fig. 3). Various anatomical resections are possible by the tube repositioning technique, as described by Kokudo et al. [14].

Fig. 1
figure 1

Route of tube used for the liver hanging maneuver for hemihepatectomy. (A) Right hepatectomy; (B) left hepatectomy. RHV right hepatic vein, MHV middle hepatic vein, LHV left hepatic vein, GP Glisson’s pedicle

Full size image
Fig. 2
figure 2

For transection along the right hepatic vein and tube repositioning between Glisson’s pedicle of the right anterior and posterior sectors. RHV right hepatic vein, GP Glisson’s pedicle

Full size image
Fig. 3
figure 3

Tube positioning in right trisectionectomy. MHV middle hepatic vein, LHV left hepatic vein, GP Glisson’s pedicle

Full size image

The transection method was similar between the groups during the study period (2000–2007). Coagulation instruments were not used in this series. Hepatic transection was performed mainly using the crush clamping method [15] and an ultrasonic dissector was used for dissection around the main vessels at hepatic hilum or inferior vena cava. The hepatic inflow was intermittently occluded during transection using the Pringle maneuver (15-min occlusion and 5-min de-clamping) [16]. The LHM-tube was always pulled up during transection and the direction of transection was always targeted toward the hanging tube. By maintaining the position of the tube, transection to the anterior aspect of the vena cava could be performed easily.

Statistical analysis

Continuous data were expressed as mean ± standard deviation. Data for different groups were compared by using one-way analysis of variance (ANOVA). The χ2 test was used for comparison of categorical variables. Differences between groups were analyzed by Fisher’s exact test or Scheffé’s multiple comparison test. A two-tailed P value < 0.05 was considered significant. All statistical analyses were performed by using the StatView Software for Windows, version 5.0 (SAS Institute, Inc., Cary, NC).

Results

Table 1 summarizes the demographic data and comparison of the LHM and non-LHM hemihepatectomy groups. Patients of the two groups were of similar age and sex. Comparison of background liver diseases and liver tumors showed higher frequencies of obstructive jaundice caused by hilar bile duct carcinomas in the non-LHM group than the LHM group, but the difference was not significant (P = 0.11). This was due to the resection of the entire caudate lobe and bile duct in the peripheral side. Child-Pugh classification was grade A in almost all patients of the two groups. Preoperative liver function tests, including ICGR15, LHL by 99mTc-GSA, alanine aminotransferase, total bilirubin, prothrombin activity, and serum hyaluronic acid level were not significantly different between the two groups.

Table 1 Patient demographics and preoperative liver function tests before hemihepatectomy
Full size table

Table 2 summarizes the surgical records of patients of both groups who underwent hemihepatectomy. Operative procedure, whether right or left hepatectomy, was not significantly different in both groups. The amount of intraoperative blood loss and blood transfusion were not significantly different between the two groups. The operating time tended to be shorter in the LHM group than in the non-LHM group (P = 0.1). On the other hand, the time required for liver parenchymal transection (almost equal to the time of clamping of hepatic blood inflow) in the LHM group was approximately 12 minutes shorter than in the non-LHM group (P = 0.004).

Table 2 Surgical details of hemihepatectomy
Full size table

Table 3 shows the results of postoperative liver function tests and complications. Postoperatively, hepatic function including the highest level of total bilirubin and alanine aminotransferase, and the lowest level prothrombin activity were not significantly different between the two groups. With regard to postoperative complications, the incidence of uncontrolled ascites under the use of diuretics tended to be greater in the non-LHM group (40%) than in the LHM group (13%) (P = 0.08), although the incidences of biloma and hepatic failure were not different between the two groups. The length of hospital stay was not significantly different between the groups. No hospital deaths were recorded in the present series.

Table 3 Postoperative liver function tests and complications in hemihepatectomy
Full size table

In the present series, the hanging maneuver could be applied in transection along the right hepatic vein in five cases. We compared the time of transection along the right hepatic vein in patients who underwent left trisectionectomy, right anterior sectionectomy, resection of segment 4 + 5 + 8, and right lateral sectionectomy with the combined resection of right hepatic vein in both groups (n = 5 in LHM group and n = 5 in non-LHM group). The transection time along the right hepatic vein in the LHM group was 38, 37, 35, 28, and 24 min (32.4 ± 6.1 min), whereas that in the non-LHM group was 56, 48, 45, 38, and 35 min (44.4 ± 8.3 min). Comparison of these data showed that the transection time tended to be shorter in the LHM group than in the non-LHM group, despite the small number of patients (P = 0.06). In one patient who underwent right trisectionectomy using LHM, the duration of liver parenchymal transection was only 18 min, although the large tumor compressed the umbilical Glisson’s pedicle and parenchyma of the left lateral sector.

Conclusion

The LHM proposed by Belghiti et al. is applied in many centers during liver resection [5–9, 17, 18] and living donor liver transplantation [14, 19]. In hepatic malignancy, mobilization or rotation of the resected liver with liver tumor could be avoided to prevent tumor dissemination during operation. In patients with a large tumor or a tumor that involves adjacent organs, anterior hepatic transection can be performed using LHM before the combined resection of the invasive parts [3, 4]. Unnecessary mobilization of the nonresected liver can be avoided in case of donation for liver transplantation as well. When the tape or tube is placed at the retrohepatic space on the vena cava, hepatic transection in the deeper part or near the confluence of hepatic veins is safe because the operator targets this tape or tube and, therefore, injury of short hepatic veins or vena cava can be avoided [13]. In addition, this could reduce the mental stress on the operator. In this series, we used a small nasogastric tube for this purpose, because it provides a better lifting force compared with silicon drain or vessel tape, and the insertion of such tube to the dissected space can be achieved safely because of the blunt tip. Kokudo et al. [14] proposed the gradual tape-repositioning technique in cases of living liver donation, in which the tape was inserted by passing it between Glisson’s pedicle and liver parenchyma. We prefer to use this repositioning technique for LHM. By applying this technique, LHM can be applied in various anatomical resections as reported by Kim et al. [10].

In the present study, we compared in a nonrandom fashion the clinical parameters and outcomes between the LHM and non-LHM groups. The decision to use LHM actually depended on the surgeon during the study period. As shown in Table 1, hepatic resection for hilar bile duct carcinoma was performed without LHM in our series. In such hepatectomy, total resection of the caudate lobe is necessary in most cases [20]. The bile duct is usually cut at the confluence of the branches in each segment to obtain enough surgical margin in hilar bile duct carcinoma and the cut line of hepatic transection is shifted to the peripheral side [20]. Therefore, LHM seems to be difficult in such conditions. The surgical records showed that LHM did not reduce blood loss and transfusion contrary to the previous report [21]. By lifting the tube during transection, bleeding form the hepatic vein could not be controlled completely. On the other hand, skillful surgeons performed hepatic resection in the non-LHM group, which could explain the limited intraoperative blood loss in our series. In the present study, the shorter time required for hepatic parenchymal transection in the LHM group compared with non-LHM group was probably the result of the following: 1) dissection of the liver parenchyma was more easily and rapidly performed by parenchymal compression using the hanging maneuver; 2) dissection of the parenchyma in the deeper part near the vena cava could be performed without hesitation of injury to cava or short hepatic veins by the shield provided by the covering tube; and 3) operator could always target the tube position during transection and, therefore, an adequate transection plane could be obtained without hesitation of direction for transection. The operating time also was shorter with shorter transection time in the present study. However, because extra time was required to prepare for LHM, our analysis showed no significant saving in total operating time.

Our results showed that postoperative liver function tests and incidences of biloma and hepatic failure were not different between the two groups. The LHM technique itself did not improve postoperative hepatic function. The incidence of long-term and uncontrolled ascites tended to be more frequent in the non-LHM group. Ascites might be a byproduct of dissection of the ligaments surrounding the liver for mobilization or after lymph node dissection. With respect to lymph node dissection, the technique used in this cohort was not different between the two groups, and some patients with uncontrolled ascites did not undergo lymph node dissection. We speculate that the limited liver mobilization through the use of LHM reduced injury of the lymphatic ducts in the hepatic ligament and hence the level of ascites. On the other hand, the long-term ascites noted after operation in our patients with injured liver seemed to contribute to longer hospitalization. Thus, the development of ascites might influence the duration of hospitalization in the non-LHM group, although the difference in this parameter between our groups was not significant.

We believe that the most important advantage of LHM is confirmation of the appropriate cut plane during transection in the present series. By placing a hanging tape or tube, the transected line can be targeted always to this tape. As the tape is placed adjacent to the hepatic vein, appropriate transection could be accomplished by LHM when the first cut-line is obtained appropriately. In the present study, we experienced four anatomical resections except hemihepatectomy. Kim et al. [10, 18] reported recently a modified LHM technique with “three Glisson’s pedicle and three hepatic veins” in a large number of patients who underwent anatomical hepatectomy. They performed right anterior or posterior sectionectomies using LHM in some cases. In the right anterior sectoriectomy, right posterior sectionectomy with right hepatic vein and left trisectionectomy, transection lines were the plane along the right hepatic vein. In LHM, the tube was placed between the right and middle hepatic veins first and could be placed between each Glisson’s pedicle by Kokudo’s repositioning technique [14]. To transect the plane along the right hepatic vein, the tube was placed between Glisson’s pedicle between the right anterior and posterior pedicles. Thus, LHM also can be performed in such a hepatectomy. Such hepatectomy often requires a longer time for transection or is associated with a larger amount of bleeding. Thus, it is necessary to reduce transection time or intraoperative bleeding by innovative techniques. We deduce that LHM is adequate to resolve this problem. As a result, transection time might be shorter in each patient in whom LHM is performed compared with cases without LHM. Because of the limited number of hepatectomy in the present study, we cannot make firm conclusions at this stage; however, LHM seems to be useful for hepatic transection along the right hepatic vein.

Conclusion

We examined the suitability of liver hanging maneuver for right or left hemihepatectomy in various liver diseases. The use of LHM tended to shorten the transection time and operating time. The use of LHM tends to reduce the development of postoperative long-term ascites and consequently reduce the associated hospital stay. LHM can be applied for transection at a plane along the right hepatic vein.