Living-donor liver transplantation (LDLT) has been developed as an important potential source of organs for treatment of children with acute and chronic liver disease. The feasibility of LDLT for adult patients has also been reported (1). However, liver grafts are small for adult-to-adult LDLT using a left lobe graft. Such small-for-size grafts in LDLT are sometimes insufficient to meet metabolic demands, resulting in a lower chance of survival. One solution to this problem is to use a right liver graft. Two surgical procedures for harvesting a right lobe have been reported: a right lobectomy with drainage of the right hepatic vein (RHV) alone (2–4), and an extended right lobectomy with both RHV and middle hepatic vein (MHV) drainage (5–7). In the case of the former procedure, which is widely applied because it is less invasive for donors, the need for drainage from the MHV tributaries has not yet been established. Venous outflow problems associated with deprivation of the MHV tributaries in LDLT using right lobes are uncommon (8–10) but are devastating when they occur (11,12). Even if a partial outflow disturbance does not result in graft dysfunction, venous congestion in LDLT is associated with disadvantages in postoperative liver volume regeneration (13). To prevent such venous outflow problems, a reliable method for evaluating the need for drainage from the MHV tributaries is currently desired.
Near-infrared spectroscopy (NIRS) enables nondestructive evaluation of hemoglobin (Hb) oxygenation and the redox state of cytochrome oxidase (Cyt.aa3) in living organ tissues (14–19). In the near-infrared (NIR) region (700–1,300 nm), light can penetrate deeply into biologic tissues over long distances, and the absorption intensity of NIR light depends on the level of Hb oxygenation, the redox state of Cyt.aa3, and the content of biologic pigments in the tissue (14,15). We applied NIRS for evaluation of venous outflow problems associated with the deprivation of the MHV tributaries in LDLT using right lobes.
PATIENTS AND METHODS
Fifteen consecutive patients who underwent adult-to-adult LDLT using a right lobe graft without the MHV from January 2001 to August 2002 at the Hiroshima University Hospital were enrolled in this study. The 15 patients included 11 men and 4 women, ranging in age from 27 to 66 years (mean±SD, 47±12 years). Graft weight and graft-to-recipient body weight ratio ranged from 480 to 790 g (mean, 621±81 g) and from 0.73% to 1.35% (mean, 0.93±0.19%), respectively. Original diseases of the patients are listed in Table 1. The graft donors were eight offspring, four siblings, and three spouses, with ages ranging from 19 to 54 years (mean, 32.2±9.01 years).
Donor hepatectomy and the recipient transplantation procedure were performed as described previously with minor modifications. In brief, the right lobe, without the middle hepatic vein, was harvested from the donor as follows. An intraoperative ultrasonographic examination was performed for final identification of the anatomy of the hepatic veins and portal veins in addition to an intraoperative cholangiographic examination to determine the branching pattern of the bile ducts. Before parenchymal transection, the right lobe was mobilized and the short hepatic veins were transected. Parenchymal transection was performed on the right side of the MHV. During parenchymal transection, major right tributaries of the MHV were clamped using a vascular clip and transected. After hepatectomy, ex vivo perfusion of the right lobe graft was performed through the portal vein (PV). The initial perfusate was saline solution (500 mL) followed by University of Wisconsin solution (1,000 mL). During initial perfusion, real-time and continuous NIRS measurement was performed on the anterior segment of the right lobe graft to determine the kinetics of Hb washout from the hepatic tissue as described later.
For the recipient, the implantation was performed after total hepatectomy. The middle and left hepatic veins were closed, and then the graft RHV was anastomosed to the RHV of the recipient in an end-to-end fashion. After end-to-end anastomosis of the graft right PV to the right PV of the recipient, the graft was reperfused before microsurgical reconstruction of the hepatic artery (HA) (end-to-end anastomosis of the HA of the graft to the right HA of the recipient). The bile duct of the graft liver was anastomosed in an end-to-end fashion to the common or right hepatic bile duct of the recipient. All recipients continuously received intravenous injection of prostaglandin E1 (0.01 μg/kg/min) during surgery and for 24 hr after surgery. Levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and serum bilirubin were measured as indexes of liver function. The initial immunosuppressive regimen consisted of tacrolimus and steroids. Computed tomographic (CT) scans and Doppler ultrasonography were routinely performed.
During the course of harvesting and implantation, in vivo NIRS measurements were performed on both the anterior and posterior segments of the liver grafts to determine Hb and Cyt.aa3 content in the hepatic tissues as indicators of congestion and redox state (Fig. 1A). In addition, during initial ex vivo perfusion, continuous NIRS measurement was performed on the anterior segment of the right lobe graft to determine the kinetics of Hb washout from the hepatic tissue (Fig. 1B). NIRS measurement was carried out as follows. NIR reflectance was measured with a multichannel photodetector (MCPD-2000; Otsuka Electronics, Osaka, Japan) connected to a personal computer (PC-9801FS; NEC, Tokyo, Japan). NIR light from a halogen lamp at 300 W was directed through a flexible bundle of quartz optical fibers to the liver, and the reflected light was conveyed through another bundle to a spectrophotometer. The tips of the two fiber bundles were fixed at a position approximately 1 cm above the liver, and the distance between the two probes was approximately 1 cm. The reflected light was measured sequentially in the range of 700 to 1,000 nm, with an interval of 2 nm. The sampling time for each scan was 4 sec. The difference between the spectrum from the liver immediately after laparotomy of the donor and that from the liver shortly before closure of the recipient’s abdomen (Fig. 1A) and the difference between the spectrum from the liver just before initial ex vivo perfusion and that from the liver during ex vivo perfusion (Fig. 1B) were calculated every 2 nm. The difference between the spectra was analyzed by a curve-fitting technique based on the least-squares method using standard spectra of purified oxyhemoglobin, deoxyhemoglobin, oxidized Cyt.aa3, and water. The five components were fitted with the following equation: MATH where OD(λ), L(λ), and e1–5(λ) are optical density, mean light path length, and extinction coefficient of each component at a wavelength of λ, respectively. The relative changes in each component were determined using this multicomponent analysis calculated on the basis of singular value decomposition (20). The changes in total Hb content were calculated by adding the changes in oxyhemoglobin and deoxyhemoglobin.
Statistical analysis was performed using Student t test. Differences at P <0.05 were considered significant.
Thirteen of the 15 patients did not incur fatal complications within 1 month after LDLT, but one patient (patient 2) died of multiple organ failure caused by perforation of the stomach and rupture of the HA at postoperative day (POD) 16, and one patient (patient 6) suffered from sepsis from POD 14 and died on POD 30.
Relationship between Caliber of MHV Tributaries and Graft Congestion
The 15 cases were divided into three groups according to the caliber of the major MHV tributaries in the right lobe grafts: the small group (<4 mm) (n=4), the intermediate group (4–7 mm) (n=5), and the large group (>7 mm) (n=6) (Table 1). Congestion in the graft livers was estimated as an increase in tissue Hb content (i.e., the difference between Hb content in the liver immediately after laparotomy of the donor and that in the liver shortly before closure of the recipient’s abdomen). After implantation, congestion was observed in both the anterior and posterior segments regardless of the caliber of the MHV tributaries (Fig. 2A). However, the congestion in the anterior segment was more severe than that in the posterior segment in the intermediate and large groups. The large group included two cases with extraordinarily severe congestion in the anterior segment (patients 4 and 15), resulting in a large standard deviation. The increase in Hb content reflected the increase in oxyhemoglobin rather than that in deoxyhemoglobin (Fig. 2B and C). No significant changes in oxidized Cyt.aa3 (Fig. 2D) were seen in either of the segments in the small and intermediate groups, indicating a well-preserved mitochondrial redox state. However, two cases with severe congestion in the large group showed reduction in oxidized Cyt.aa3 in the anterior segment.
If the right lobe graft is deprived of the MHV tributaries and the PV and HA inflow and the RHV outflow are adequate, anterior segment congestion after transplantation might depend on the venous collaterals between the MHV and the RHV in a right lobe graft. To estimate the significance of venous collaterals in a right lobe graft before implantation, NIRS measurement was performed on the anterior segment of the right lobe graft to quantify tissue content of Hb during initial ex vivo perfusion. Representative kinetics of Hb washout from the anterior segment calculated by NIRS is shown in (Figure 3A). In most cases, tissue content of Hb was decreased by ex vivo flushing and became constant within 180 sec. However, Hb content in the anterior segment remained high even after ex vivo flushing in some cases in the large group, demonstrating poor venous collaterals. (Figure 3B) shows the relationship between the remaining Hb content at 180 sec after initial flushing and the extent of postoperative congestion (ΔHb) in the anterior segment. They were significantly correlated, indicating that postoperative congestion can be predicted by NIRS during initial perfusion.
Relationship between Caliber of MHV Tributaries and Liver Faction
Serum levels of AST, ALT, and bilirubin after LDLT were measured as indexes of liver function. Those parameters were not significantly different among the groups, but two cases with severe congestion showed transient but marked elevation of AST and ALT levels in the first week after LDLT (Fig. 4).
Computed Tomographic Scan and Doppler Ultrasound Examinations
Computed Tomographic (CT) scan examination after LDLT generally revealed no remarkable graft congestion in the small and intermediate groups (Fig. 5A). In some of the cases in the intermediate and large groups, venous drainage from the right anterior segment lagged behind that from the posterior segment by 2 to 4 weeks after surgery (Fig. 5B). However, this time lag in MHV tributary drainage gradually improved (Fig. 5C). In such cases, reverse flow in the proximal-to-peripheral direction in the tributaries was observed by color Doppler ultrasound examination (Fig. 5E), suggesting growth of intrahepatic collaterals between the MHV tributaries and the RHV. In the case with the most severe congestion, as determined by intraoperative NIRS, a perfusion defect was seen in the anterior segment on a CT scan 2 weeks after LDLT (Fig. 5D). This circulatory impediment was persistent, causing atrophy of the anterior segment. However, compensatory hypertrophy of the posterior segment saved this patient (not shown).
In hepatectomy for LDLT, it is practically impossible to maintain complete venous drainage in both the right and left lobes, because the MHV can be preserved on only one side. Poor venous outflow can result in engorgement and graft failure (11,12). However, many authors have reported that the right liver graft showed good function without reconstruction of the MHV tributaries (3,4,9,21,22). One possible explanation for these good results is that drainage flow from the anterior segment to the posterior segment of a right lobe graft is established by means of intrahepatic PV and then into the RHV (23). Another possible explanation is that intrahepatic communication between the MHV tributaries and the RHV contributes to the prevention of congestion of the anterior segment (24). If such vein communications are significant in the right lobe graft, reconstruction of the MHV tributaries might not be required for adequate hepatic venous drainage. The necessity for reconstruction of the MHV tributaries or the inferior right hepatic vein is currently judged by the caliber of such vessels (3,11). Information on the presence or absence of hepatic segmental congestion after temporarily clamping the corresponding vein during donor surgery is also useful (9). Recent studies have demonstrated that concomitant temporary clamping of the HA is also useful for diagnosis of venous congestion of the liver, because the hepatofugal area is discolored under these conditions (10,25). Such methods would provide a gross extent but not a quantitative degree of hepatic engorgement. As a more objective method for predicting hepatic vein drainage problems, we have used NIRS, which enables nondestructive and continuous evaluation of Hb and Cyt.aa3 content in living organ tissues (16–18).
Using the NIRS system, we confirmed congestion in both the anterior and posterior segments after implantation. This finding is consistent with the results of our previous study showing transient congestion after rat liver transplantation (16). The congestion in the anterior segment tended to be more severe than that in the posterior segment in the grafted hepatic right lobe with significant MHV tributaries (>4 mm), reflecting the deprivation of the tributaries. In contrast, it was shown in a previous study that the anterior segment of the right loge graft was susceptible to ischemia rather than to congestion after LDLT (9). In that study, an NIRS system similar to that used in our study was used, and it was demonstrated that Hb content in the anterior segments after implantation was lower than before hepatic transection in donors. Ischemia of the grafted liver is thought to be associated with ischemia-reperfusion–induced disturbance of sinusoidal microcirculation, at least in the early phase after implantation, probably masking hepatic venous outflow insufficiency. To attenuate such microcirculatory disturbance, recipients routinely received prostaglandin E1 during implantation surgery in the present study. This might explain why graft congestion occurred even early after implantation in our patients. Thus, it seems difficult to evaluate only venous outflow problems in vivo, because hepatic hemodynamics is complexly influenced by inflow (portohepatic artery blood supply), sinusoidal microcirculation, and hepatic vein outflow. NIRS measurement of remaining tissue Hb during ex vivo flushing probably has the advantage of minimizing effects of factors other than venous outflow. Such NIRS measurement enabled accurate prediction of hepatic hemostasis caused by deprivation of the MHV tributaries, indicating that this method is useful for determining whether MHV tributaries should be reconstructed. In this study, two recipients of liver graft with high Hb residue even after ex vivo flushing (patients 4 and 15 in the large group), who showed impairment of mitochondrial respiration and significant increase in serum transaminase levels, would be appropriate candidates for reconstruction of MHV tributaries.
Congestion of a liver graft results not only from venous outflow insufficiency but might also be exacerbated by portal hypertension. Particularly in cases with marginal communicating veins, portal pressure after grafting might be an important factor for deciding whether to reconstruct MHV tributaries in a right liver graft. Because the present series, in which portal pressure after the LDLT (at the end of recipient surgery) was between 13 and 16 mm Hg in all cases (data not shown), did not allow us to analyze this issue, further studies are needed.
When in vivo spectroscopy is applied to living tissues, the field of view of scanning should be taken into account. It has been established that the “mean light pathlength,” labeled “field of view,” in in vivo NIRS is four- to sixfold greater than the distance between the transmitting and receiving optical bundles when those are vertically applied to the scanned tissues and the intensity of the light source is adequate (26). In our NIRS system, the distance between the transmitting and receiving optical bundles was 1 cm; thus, the mean light pathlength, spectrophotometric view, was theoretically 4 to 6 cm, suggesting that several-point scanning might be sufficient to quantitatively represent the state of venous congestion in a subsegment of the liver graft. However, this method did not enable accurate determination of the venocongestive areas in the liver grafts. To compensate for this defect, visualization of the venocongestive area by temporary clamping of the corresponding vein and the HA during donor surgery (25) would be helpful.
Intraoperative NIRS enabled quantification of the extent of congestion and the influence of the oxygenation state of right lobe liver grafts by determining changes in tissue content of oxyhemoglobin, deoxyhemoglobin, and oxidized and reduced Cyt.aa3. In addition, by determining the kinetics of Hb washout of the right hepatic lobe graft during ex vivo perfusion, NIRS enabled prediction of such a venous outflow problem before implantation.
The authors thank Yuka Tanaka for expert secretarial assistance.
1. Hashikura Y, Makuuchi M, Kawasaki S, et al. Successful living-related partial liver transplantation to an adult patient. Lancet 1994; 343 ( 8907): 1233.
2. Yamaoka Y, Washida M, Honda K, et al. Liver transplantation using a right lobe graft from a living related donor. Transplantation 1994; 57 ( 7): 1127.
3. Inomata Y, Uemoto S, Asonuma K, et al. Right lobe graft in living donor liver transplantation. Transplantation 2000; 69 ( 2): 258.
4. Grewal HP, Shokouh-Amiri MH, Vera S, et al. Surgical technique for right lobe adult living donor liver transplantation without venovenous bypass or portocaval shunting and with duct-to-duct biliary reconstruction. Ann Surg 2001; 233 ( 4): 502.
5. Lo CM, Fan ST, Liu CL, et al. Extending the limit on the size of adult recipient in living donor liver transplantation using extended right lobe graft. Transplantation 1997; 63 ( 10): 1524.
6. Fan ST, Lo CM, Liu CL, et al. Safety of donors in live donor liver transplantation using right lobe grafts. Arch Surg 2000; 135 ( 3): 336.
7. Fan ST, Lo CM, Liu CL. Technical refinement in adult-to-adult living donor liver transplantation using right lobe graft. Ann Surg 2000; 231 ( 1): 126.
8. Kaneko T, Kaneko K, Sugimoto H, et al. Intrahepatic anastomosis formation between the hepatic veins in the graft liver of the living related liver transplantation: Observation by Doppler ultrasonography. Transplantation 2000; 70 ( 6): 982.
9. Cui D, Kiuchi T, Egawa H, et al. Microcirculatory changes in right lobe grafts in living-donor liver transplantation: A near-infrared spectrometry study. Transplantation 2001; 72 ( 2): 291.
10. Cescon M, Sugawara Y, Sano K, et al. Right liver graft without middle hepatic vein reconstruction from a living donor. Transplantation 2002; 73 ( 7): 1164.
11. Ghobrial RM, Hsieh CB, Lerner S, et al. Technical challenges of hepatic venous outflow reconstruction in right lobe adult living donor liver transplantation. Liver Transpl 2001; 7 ( 6): 551.
12. Lee S, Park K, Hwang S, et al. Congestion of right liver graft in living donor liver transplantation. Transplantation 2001; 71 ( 6): 812.
13. Maema A, Imamura H, Takayama T, et al. Impaired volume regeneration of split livers with partial venous disruption: A latent problem in partial liver transplantation. Transplantation 2002; 73 ( 5): 765.
14. Jobsis FF. Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science 1977; 198 ( 4323): 1264.
15. Wyatt JS, Cope M, Delpy DT, et al. Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 1986; 2 ( 8515): 1063.
16. Tashiro H, Suzuki S, Kanashiro M, et al. A new method for determining graft function after liver transplantation by near-infrared spectroscopy. Transplantation 1993; 56 ( 5): 1261.
17. Ohdan H, Fukuda Y, Suzuki S, et al. Simultaneous evaluation of nitric oxide synthesis and tissue oxygenation in rat liver allograft rejection using near-infrared spectroscopy. Transplantation 1995; 60 ( 6): 530.
18. Noriyuki T, Ohdan H, Yoshioka S, et al. Near-infrared spectroscopic method for assessing the tissue oxygenation state of living lung. Am J Respir Crit Care Med 1997; 156 ( 5): 1656.
19. El-Desoky AE, Delpy DT, Davidson BR, et al. Assessment of hepatic ischaemia reperfusion injury by measuring intracellular tissue oxygenation using near infrared spectroscopy. Liver 2001; 21 ( 1): 37.
20. Press WH, Teukolsky SA, Vetterling WT, et al. Singular value decomposition. In: Numerical recipes in C: The art of scientific computing. New York: Cambridge University Press, 1992, p. 59.
21. Marcos A, Fisher RA, Ham JM, et al. Right lobe living donor liver transplantation. Transplantation 1999; 68 ( 6): 798.
22. Marcos A, Ham JM, Fisher RA, et al. Surgical management of anatomical variations of the right lobe in living donor liver transplantation. Ann Surg 2000; 231 ( 6): 824.
23. Deutsch V, Rosenthal T, Adar R, et al. Budd-Chiari syndrome: Study of angiographic findings and remarks on etiology. Am J Roentgenol Radium Ther Nucl Med 1972; 116 ( 2): 430.
24. Maguire R, Doppman JL. Angiographic abnormalities in partial Budd-Chiari syndrome. Radiology 1977; 122 ( 3): 629.
25. Sano K, Makuuchi M, Miki K, et al. Evaluation of hepatic venous congestion: Proposed indication criteria for hepatic vein reconstruction. Ann Surg 2002; 236 ( 2): 241.
© 2003 Lippincott Williams & Wilkins, Inc.
26. van der Zee P, Cope M, Arridge SR, et al. Experimentally measured optical pathlengths for the adult head, calf and forearm and the head of the newborn infant as a function of inter optode spacing. Adv Exp Med Biol 1992; 316: 143.