Effect of intermittent hepatic inflow occlusion with the Pringle maneuver during donor hepatectomy in adult living donor liver transplantation with right hemiliver grafts: A prospective, randomized controlled study†‡
Jae Berm Park,
Department of Surgery, Asan Medical Center, University of Ulsan College of Medicine, Ulsan, Korea
Intermittent hepatic inflow occlusion (IHIO) via clamping of the portal triad (also called the Pringle maneuver) is a safe and effective procedure for major hepatectomy in patients with liver disease. IHIO minimizes the blood loss and operation time during liver resection.1, 2 Reductions in blood loss and resultant decreases in transfusion requirements have been shown to improve the outcomes of patients undergoing liver resection for tumors.3, 4 In addition, ischemic preconditioning with IHIO has been reported to have protective effects on patients undergoing liver resection.5
In the setting of living donor liver transplantation (LDLT), one of the most important concerns is liver donor safety. IHIO (followed by short periods of reperfusion) has been introduced for living donor hepatectomy, and it is being used during donor hepatectomy for LDLT in a few transplant centers.6 Several studies have demonstrated the safety of IHIO in donors for liver transplantation (LT).7
However, the effects of preconditioning with IHIO during donor hepatectomy on LDLT recipients remain unclear. Several small series have assessed the effects of ischemic preconditioning during whole deceased donor LT on recipients.8, 9 The role of IHIO, however, has not been fully elucidated in liver donors and recipients during LDLT.2 In this randomized, prospective study, we evaluated the efficacy of IHIO in recipients and donors.
PATIENTS AND METHODS
This open-label, prospective, parallel-group, randomized controlled study was conducted between July 2008 and August 2009 in a single transplant center (Samsung Medical Center, Seoul, Korea). This randomized clinical trial was registered at ClinicalTrials.gov (NCT01171742). The donor-recipient pairs were randomized (1:1) into 2 groups (IHIO and control groups) at the time of anesthesia induction for donors via the extraction of a black or white (but otherwise identical) stone from an unseen box. Pairs of donors and LDLT recipients (all 18 years old or older) were eligible for this study: donors underwent right hemihepatectomy, and recipients received right hemiliver grafts. Donor-recipient pairs were excluded for one of the following reasons: the recipient had experienced fulminant hepatic failure, the graft-to-recipient body weight ratio (GRWR) was <0.9%, a frozen biopsy sample from the donor liver showed >30% macrovesicular steatosis before donor hemihepatectomy, the transplant was ABO-incompatible, the recipient had previously undergone organ transplantation, or the recipient had undergone or was scheduled to undergo multiorgan transplantation. According to randomization, IHIO was performed in the donors of the IHIO group during parenchymal resection; each IHIO procedure consisted of clamping of the hepatoduodenal ligament for 15 minutes and then reperfusion for 5 minutes. In the donors of the control group, parenchymal resection was performed without IHIO. The baseline and demographic data were recorded prospectively at the study entry point. Postoperatively, all LT recipients and donors were managed according to our center's routine protocol. The study protocol was approved by our institutional review board, and all recipients and donors provided written informed consent before surgery.
The primary endpoint of this study was the peak serum alanine aminotransferase (ALT) concentration, a surrogate marker for liver injury, in recipients within 5 days after transplantation. The sample size (n = 26 for each group) was determined for testing the equality of the peak serum ALT concentrations within 5 days after transplantation in the recipients of the 2 groups [clinically significant difference in recipient ALT levels > 200 IU/L (2-tailed), standard deviation = 250 IU/L, α = 0.05, power = 0.80].
Donors were chosen from a pool of individuals who were less than 65 years old on the basis of their liver anatomy [including the estimated GRWR and the remnant liver volume (>30%) according to preoperative computed tomography] and their serological status (hepatitis B or C virus). Histological examinations of frozen biopsy samples taken just before parenchymal resection showed that none of the grafts had >30% macrovesicular steatosis. The parenchymal dissection plane was determined according to a demarcation line that emerged during temporary clamping of the portal vein and hepatic artery and according to an assessment of the graft volume and the middle hepatic vein structure. During all donor hepatectomies, the middle hepatic vein trunk was retained in the remnant liver. The parenchyma was transected by a senior surgeon with a Cavitron ultrasonic aspirator and bipolar electrocautery with or without IHIO according to randomization. In the donors of the IHIO group, several IHIO procedures were performed during hepatectomy via the clamping of the hepatoduodenal ligament for 15 minutes with a vascular tourniquet and declamping for more than 5 minutes. Blood loss during parenchymal resection was determined by the measurement of the gauze weight and the aspiration volume.
Bench Work and Recipient Operation
Immediately after its removal from the donor, the graft liver was perfused through the portal vein by gravity flow with 2500 to 3000 mL of histidine tryptophan ketoglutarate (Custodiol, Dr. Franz Köhler, Chemie GmbH, Alsbach-Hähnlein, Germany) until the perfusate was clear. Interposition procedures for the drainage of the middle hepatic vein territory were performed with a cryopreserved iliac artery or vein for a middle hepatic vein tributary ≥ 5 mm in size (V5 or V8). Bile duct ductoplasty was performed whenever it was feasible. The discrete procedures of the recipient operation have been described previously.10, 11
All the recipients were treated with an immunosuppression regimen that included anti–interleukin-2 (anti–IL-2) receptor induction (Simulect from Novartis), tacrolimus [Prograf from Astellas (Dublin, Ireland) or TacroBell from Chon Keun Dang (Seoul, Korea)], mycophenolate mofetil (CellCept from Roche or Myfortic from Novartis), and a corticosteroid.
The primary endpoint was the peak serum ALT concentration within 5 days after transplantation in the recipients.
The secondary endpoints for the donors included liver function tests, hospital stays, and morbidity. The secondary endpoints for recipients included liver function tests, graft function, intensive care unit stays, and in-hospital mortality. For the measurement of IL-6, IL-8, tumor necrosis factor α (TNF-α), and hepatocyte growth factor (HGF) levels, blood samples were taken from donors immediately after anesthesia induction and 2 hours after graft removal and from recipients immediately after anesthesia induction, during the anhepatic phase, 2 hours after reperfusion, and on postoperative days 1 and 3.
Biopsy samples were taken from donors at the time of laparotomy, just before portal vein and hepatic artery clamping after parenchymal resection, and from recipients 2 hours after reperfusion. Hepatocyte injury was determined from the concentrations of caspase-3 (a proteolytic enzyme involved in apoptosis)12 and malondialdehyde (MDA; a surrogate for oxidative damage by reactive oxygen species), which were measured with enzyme-linked immunosorbent assay (ELISA) kits [human caspase-3 instant ELISA (Bender MedSystems, Vienna, Austria) and human MDA ELISA (USCN Life Science, Inc., Wuhan, China)] according to the manufacturers' directions.
Continuous variables were expressed as means and standard deviations or as medians and ranges and were compared with independent t tests or Mann-Whitney U tests (according to the normality of the distribution) and then Kolmogorov-Smirnov tests. Pearson's chi-square test or Fisher's exact test was used to compare noncontinuous variables. All statistical analyses were performed with SPSS for Windows 17.0 (SPSS, Inc., Chicago, IL). A P value less than 0.05 was considered statistically significant.
In all, 53 pairs of donors and recipients who underwent LDLT with right hemiliver grafts between July 2008 and August 2009 were eligible for this study and were randomized. Three pairs (1 in the control group and 2 in the IHIO group) were excluded after donor hemihepatectomy because the GRWR by measured graft weight was <0.9%; thus, 50 donor-recipient pairs (ie, 25 in each group) completed the study and were included in this analysis. The demographic and clinical findings for the donors and recipients are listed in Table 1. There were no significant differences between the groups with respect to recipient, donor, or procedure-related factors. The GRWRs and the remnant/total liver volume values were 1.15% ± 0.22% and 34.4% ± 4.35%, respectively, in the control group and 1.15% ± 0.20% and 36.9% ± 6.13%, respectively, in the IHIO group. In the IHIO group, IHIO during donor parenchymal resection was performed twice in 6 donors, 3 times in 15 donors, and 4 times in 4 donors (mean time per IHIO = 15.0 ± 1.47 minutes).
Table 1. Demographic and Clinical Characteristics of the Donor-Recipient Pairs (n = 50)
Control Group (n = 25)
IHIO Group (n = 25)
NOTE: Variables are expressed as means and standard deviations unless noted otherwise.
Measured on a scale.
Calculated by computed tomography volumetry estimation.
Postoperative ALT Levels and Clinical Course in the Recipients
The mean peak serum ALT level within the first 5 days after the operation was 344.8 ± 472.7 IU/mL in the control group and 477.3 ± 453.7 IU/mL in the IHIO group (P = 0.32; Fig. 1 and Table 2). In the control group, the serum ALT level peaked on posttransplant day 1 in 18 recipients, on day 2 in 6 recipients, and on day 3 in 1 recipient; in the IHIO group, the serum ALT level peaked on day 1 in 14 recipients, on day 2 in 10 recipients, and on day 3 in 1 recipient. The mean serum ALT level on postoperative day 5 was 134.2 ± 135.6 IU/mL in the control group and 176.1 ± 174.0 IU/mL in the IHIO group (P = 0.35). The mean peak prothrombin time (PT)/international normalized ratio (INR) within 5 days after the operation was 2.45 ± 0.51 in the control group and 2.57 ± 0.62 in the IHIO group (P = 0.438). The mean serum total bilirubin level on postoperative day 5 was 2.65 ± 3.68 mg/dL in the control group and 4.57 ± 6.11 mg/dL in the IHIO group (P = 0.19). Figure 2A,C,E shows postoperative changes in liver function tests. Neither group experienced in-hospital mortality within 1 month, primary nonfunction, or initially poor function. The mean length of stay in the intensive care unit was 9.6 ± 4.8 days (median = 8 days, range = 7-29 days) in the control group and 8.0 ± 3.6 days (median = 8 days, range = 4-22 days) in the IHIO group (P = 0.02 by the Mann-Whitney U test).
Table 2. Clinical Results for the Donor-Recipient Pairs (n = 50)
Control Group (n = 25)
IHIO Group (n = 25)
NOTE: Variables are expressed as means and standard deviations and as medians and ranges.
Mann-Whitney U test.
Peak serum ALT within 5 days after the operation (IU/mL)
Postoperative ALT Levels and Clinical Course in the Donors
The mean peak serum ALT level within 5 days after the operation was 199.0 ± 51.9 IU/mL in the control group and 289.3 ± 127.4 IU/mL in the IHIO group (P = 0.002; Fig. 1 and Table 2). The serum ALT level peaked on day 1 for 22 donors in the IHIO group and for all donors in the control group and on day 2 for 3 donors in the IHIO group. The mean serum ALT level on postoperative day 5 was 91.0 ± 25.2 IU/mL in the control group and 119.6 ± 40.3 IU/mL in the IHIO group (P = 0.004). The mean peak PT/INR within 5 days after the operation was 1.98 ± 0.77 in the control group and 1.76 ± 0.21 in the IHIO group (P = 0.19). The mean serum total bilirubin level on postoperative day 5 was 1.98 ± 0.94 mg/dL in the control group and 2.24 ± 1.42 mg/dL in the IHIO group (P = 0.46). Figure 2B,D,F shows postoperative changes in the liver function tests. There was no episode of hepatic function–related morbidity in either group. The mean donor operation time was 338.9 ± 39.3 minutes in the control group and 387.2 ± 44.0 minutes in the IHIO group (P = 0.78). The mean amount of blood loss during donor hepatectomy was 486.2 ± 289.5 mL (median = 400 mL, range = 180-1500 mL) in the control group and 324.4 ± 135.7 mL (median = 330 mL, range = 50-600 mL) in the IHIO group (P = 0.02 by the Mann-Whitney U test). In the control group, 1 donor experienced postoperative bleeding and required a red blood cell transfusion and relaparotomy 2 days later, and a second donor experienced biliary leakage, which was managed with percutaneous drainage. In the IHIO group, 1 donor experienced postoperative perihepatic fluid collection, which was managed with percutaneous drainage. The mean hospital stay was 19.3 ± 7.2 days (median = 18.0 days, range = 10-35 days) in the control group and 15.8 ± 4.6 days (median = 15.0 days, range = 10-26 days) in the IHIO group (P = 0.046).
IL-6, IL-8, TNF-α, and HGF
In donors, the mean serum concentrations of the cytokines IL-6, IL-8, TNF-α, and HGF did not differ significantly between the IHIO and control groups immediately after anesthesia induction or 2 hours after graft removal (Fig. 3). Moreover, in recipients, the mean serum concentrations of these cytokines did not differ between the IHIO and control groups immediately after anesthesia induction, during the anhepatic phase, 2 hours after reperfusion, or on postoperative days 1 and 3.
Caspase-3 and MDA in Liver Biopsy Samples
Caspase-3 and MDA were assayed in liver tissue from all patients 3 times: at the time of laparotomy, just before portal vein and hepatic artery clamping after parenchymal resection during the donor operation, and 2 hours after portal reperfusion during the recipient operation. The concentrations of caspase-3 were 2.96 ± 1.03, 3.12 ± 0.76, and 3.05± 0.84 ng/mL in the control group and 2.72 ± 1.03, 2.86 ± 0.99, and 3.28± 0.85 ng/mL in the IHIO group at the 3 time points (P = 0.40, P = 0.31, and P = 0.33; Fig. 3). The concentrations of MDA were 3.81 ± 1.34, 4.15 ± 1.48, and 4.79± 1.93 nmol/mL in the control group and 3.74 ± 1.24, 3.91 ± 1.30, and 4.68± 2.14 nmol/mL in the IHIO group at the 3 time points (P = 0.86, P = 0.55, and P = 0.86).
Donor safety is the highest priority in LDLT. For patients with hepatic tumors or diseases, the safe resection of the liver may include IHIO, intermittent hepatic inflow clamping of the portal triad, or the Pringle maneuver, which has been widely used without catastrophic deterioration of remnant liver function.13, 14 Although IHIO has been found to minimize bleeding during liver resection,14 the results after ischemia/reperfusion insults during IHIO in LDLT have not been determined. Because IHIO has been found to be a safe procedure for donors with normal healthy livers and to not negatively affect the quality of the graft,6, 7, 15 several centers have used IHIO at the time of donor liver resection. The studies by Imamura et al.6, 15 included left and right hemiliver grafts and adult and pediatric transplant recipients without randomization. A study by Miller et al.7 showed that intermittent inflow occlusion was associated in donors with reduced blood loss and less ischemic injury to the right or left hepatic remnant, but this was a report of results from small LDLT series with left or right lobe liver grafts. In our randomized study, all the recipients were adults and underwent LDLT with right hemiliver grafts; the mean remnant liver volume percentage for donors was 34.4% ± 4.35% in the control group and 36.9% ± 4.35% in the IHIO group. In this study, the postoperative peak ALT level in donors was significantly higher in the IHIO group versus the control group. The effect of IHIO on patients with cirrhosis may not be the same as the effect on patients with normal livers. Donors with IHIO may present with effects of synthetic dysfunction or other outcomes when larger numbers of patients are studied. However, the postoperative peak PT and total bilirubin values were not different in the 2 groups, and the amount of blood loss was lower in the IHIO group versus the control group. These findings indicate that IHIO can be a safe, tolerable, and effective procedure for healthy living donors of right hemiliver grafts and can reduce blood loss during hepatectomy and hospital stays without deterioration of the remnant liver function.
Another concern of many transplant surgeons is the safety of IHIO for recipients with sequential cold and warm ischemia in the setting of LDLT. According to the studies of Imamura et al. (which included pediatric recipients),6, 15 the peak aspartate aminotransferase and ALT levels of the recipients were not different in a 2-group comparison (Pringle and non-Pringle groups) or in a 4-group comparison (no-procedure, portal ischemia, partial ischemia, and total ischemia groups). In our series, which included only adult recipients of right hemiliver grafts with a GRWR > 0.9% and <30% macrovesicular steatosis, there was no difference in posttransplant peak ALT levels in the recipients between the IHIO and control groups without primary nonfunction or delayed function. In addition, reconstruction of the middle hepatic vein (V5 or V8) territory, GRWR (>1.0% versus 0.9%-1.0%), and steatosis (≤10% versus 10%-30%) were not associated with posttransplant peak ALT levels in the 2 groups (data not shown). The peak ALT levels in recipients may be related to the severity of liver disease at the time of transplantation or other conditions and not to IHIO-related ischemia injury. Therefore, the effects of IHIO during LDLT using grafts from old donors, small-for-size grafts, or grafts with moderate steatosis (>30%) may be related to recipient outcomes and should be further evaluated.
The benefits of IHIO in LDLT may be due to its protective role in preconditioning for ischemia/reperfusion injury. According to Clavien et al.,5 during liver resection, IHIO has a protective effect on the remnant liver and may, through preconditioning, overcome or reverse ischemia/reperfusion injuries in patients with liver disease. Preconditioning has been explained as a procedure that exposes the graft to limited stress and activates defense mechanisms against ischemic injury.16, 17 Short periods of ischemia and reperfusion may stimulate the generation and release of adenosine and its metabolites and thus prevent apoptosis. Many studies have attempted to elucidate the mechanisms underlying the protective effects of ischemic preconditioning.18-22 For example, in a mouse model of carbon tetrachloride–induced cirrhosis, intermittent vascular occlusion and ischemic preconditioning were associated with reduced activation of the apoptotic pathway, which was determined by the measurement of the p53 level and by staining with terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick-end labeling.21 Cirrhotic livers were significantly more susceptible to ischemia than normal livers, but the apoptosis level was higher in normal livers.
To date, the effects of IHIO during LT have not been clarified. Because ischemia/reperfusion is inevitable during LT, the impact of IHIO on the transplant recipient with sequential (subsequent) cold and warm ischemia/reperfusion injury may be different from the impact of IHIO during liver resection. Preconditioning during liver recovery from deceased donors has been associated with decreases in hepatocellular necrosis and moderate to severe hepatocyte swelling, but it has no clinical benefits.8, 23 In this study of LDLT, the levels of caspase-3 and MDA in graft liver tissue at 3 time points did not differ between the 2 groups.
We also measured the serum concentrations of the cytokines IL-6, IL-8, TNF-α, and HGF; all are factors related to pathways of liver regeneration or liver injury in donors and recipients.24-27
The serum concentrations of all 4 cytokines after anesthesia induction were higher in the recipients versus the donors, and this may have been due to liver diseases in the recipients. In the donors, the serum levels of IL-6 and HGF were distinctly elevated just after hepatectomy; in the recipients, the serum levels of IL-6 peaked during the anhepatic phase, but the levels of IL-8 peaked 2 hours after reperfusion. The levels of TNF-α and HGF, which are required for regeneration, showed a tendency to increase 2 hours after reperfusion and decrease thereafter. The recipient serum levels of IL-6, which are involved in hepatic acute-phase protein synthesis regulation, were higher in the IHIO group versus the control group, especially during the anhepatic phase and on postoperative day 3, but we did not find any significant differences in the levels of cytokines at the various time points between the 2 groups. During donor hepatectomy, the effects of IHIO on cytokine changes may be small in a normal, healthy liver. In LDLT patients who receive right hemiliver grafts with a GRWR > 0.9% and <30% steatosis, the effects of IHIO on cytokine changes may be less than the effects of ischemia and reperfusion.
This prospective, randomized controlled study has limitations. First, the primary endpoint was defined as the peak serum ALT level within 5 days after transplantation in recipients. Even if the peak serum ALT level in the early postoperative period is one of the most important variables for the clinical course of hepatic surgery, its relevance to the long-term clinical outcomes of LT recipients and donors is debatable. The ideal primary endpoint could be defined as graft and patient survival for recipients and morbidity or mortality for donors so that the effects of IHIO could be fully evaluated. However, it is very difficult and nearly impossible to elucidate the effects of IHIO on these clinical outcomes in recipients and donors because there are many other important prognostic factors, such as recipient, donor, and graft factors, surgical factors, infection episodes, primary disease recurrence, and rejection episodes during long-term follow-up. Therefore, we chose the peak serum ALT level within 5 days after transplantation as a surrogate marker for evaluating the direct effects of IHIO in the short-term postoperative period.
Second, the sample size was determined for the comparison of peak ALT levels within 5 days after transplantation (the primary endpoint) in the recipients of 2 groups. This sample size was not large enough for analyzing additional secondary outcomes. Therefore, the secondary outcomes in this study have limited generalization with respect to power.
Third, we enrolled donor-recipient pairs whose right hemiliver grafts had a GRWR > 0.9% and <30% steatosis. This may limit the applicability to donors who do not meet the study's inclusion criteria.
In conclusion, although the postoperative peak ALT levels in donors within 5 days after the operation were higher in the IHIO group, the postoperative peak ALT levels in the recipients and the peak PT and total bilirubin values of the recipients and donors on postoperative day 5 did not statistically differ between the IHIO and control groups; the amount of blood loss during donor hepatectomy was significantly lower in the IHIO group versus the control group. Donor liver resection with IHIO in the setting of LDLT using a right hemiliver graft with a GRWR > 0.9% and <30% steatosis can be a safe procedure for recipients as well as donors. Even in transplant centers in which IHIO during donor hepatectomy is not performed routinely, this study might be helpful for determining whether or not to use IHIO in the setting of LDLT using a right hemiliver graft with a GRWR > 0.9% and <30% steatosis in some select cases with troublesome bleeding during donor hepatectomy.