Pringle's maneuver and selective inflow occlusion in living donor liver hepatectomy
Abstract
While inflow occlusion techniques such as Pringle's maneuver are accepted methods of reducing bleeding without inducing liver injury during liver surgery, donor hepatectomy for living donor liver transplantation is currently performed without inflow occlusion for fear that injury to the graft may result. We have performed donor hepatectomy for 12 years using selective intermittent inflow occlusion, a technique in which the portion used to form the graft is perfused during hepatectomy. Starting in November 2000, we applied intermittent Pringle's maneuver to donor hepatectomy in 81 cases of living donor liver transplantation. We reviewed our experience with Pringle's maneuver and selective inflow occlusion techniques in donor hepatectomy in living donor liver transplantation. The quality of the grafts was assessed and compared by determining maximum postoperative aspartate aminotransferase (AST) and alanine aminotransferase (ALT) values. Neither primary nonfunction nor dysfunction occurred. Maximum AST values in the recipients were the same whether the liver segments that formed the grafts were totally ischemic during dissection (total ischemia), partially ischemic (partial ischemia), perfused only with arterial blood flow (portal ischemia), or not ischemic at all (no ischemia). Maximum ALT values in the recipients of the total ischemia group was lower, albeit not significantly, than in other groups. Total inflow occlusion can be applied to living donor hepatectomy without causing graft injury. In conclusion, because the transection surface is blood-free, there is decreased risk to the donor during living donor liver transplantation surgery, and surgeons should not hesitate to apply this technique because it contributes to donor safety. (Liver Transpl 2004;10:771–778.)
The first successful living donor liver transplants were performed on children in 1989, in response to the severe shortage of pediatric organs.1 In the early 1990s, pediatric living donor liver transplant (LDLT) was developed extensively in Asia, due to the lack of cultural beliefs for the promotion of cadaveric organ donation. The expansion of LDLT to the adult population was first reported by our group, using left liver grafts.2, 3 The first successful series of adult-to-adult LDLT using right liver grafts was reported in Hong Kong in 1997.4 Since the late 1990s, adult right liver LDLT has also developed rapidly in the West due to severe shortages of cadaveric organs and the exponential increase in numbers of patients awaiting transplants.5 Currently, more than 20 European and more than 30 U.S. transplant centers have performed adult LDLT. The number of LDLTs in the United Sstates increased from 56 in 1996 to 509 in 2001.5
Donor safety is undoubtedly the highest priority for LDLT. However, to date, 9 deaths are known to have occurred worldwide, including 1 in Japan and 3 in the United States.6 In addition, 2 cases of LDLT donors who underwent liver transplantation due to postoperative liver failure have been reported formally or informally. Although intraoperative blood loss does not appear to be the reason for these donor deaths, and has not been reported to be a significant problem at most centers performing LDLT, bleeding should be recognized as a potential risk to the donor during hepatectomy. Inflow occlusion is a well-established technique for decreasing bleeding during the transection of liver parenchyma, without causing damage to the rest of the liver, and it has been widely used for conventional hepatectomy in various diseases.7 However, almost all liver transplant centers are carrying out donor hepatectomy in LDLT without applying inflow occlusion for fear of ischemia-induced graft damage.8-13
The authors' group (Makucchi et al.) began LDLT in children in 1990 at Shinshu University; LDLT began to be performed in adults in 1993.2 An LDLT program was begun in 1996 after relocation to Tokyo University. From the outset, in 1990, we have applied inflow occlusion14—initially selective occlusion, and more recently total inflow occlusion, i.e., Pringle's maneuver,15 with no adverse effects on either donors or recipients. The preliminary result of Pringle's maneuver in LDLT has been reported previously.16 In this study, we review our experience with donor inflow occlusion during LDLT at Tokyo University.
Abbreviations
LDLT, living donor liver transplantation; AST, aspartate aminotransferase; ALT, alanine aminotransferase; para., paramedian sector; MHV, middle hepatic vein.
Patients and Methods
Patients
A total of 205 living donor liver transplants were performed on 203 recipients between January 1996 and December 2002. The ages of the 205 donors ranged from 18 to 63 years (median, 34 years). The donors were parents (86 cases), children (56), spouses (21), siblings (26), nephews (8), grandmothers (2), uncles or aunts (2), cousins (1), brothers-in-law (2), and cousin of the spouse (1). In all but 1 case, blood types were identical to, or compatible with, the recipient's. Donor screening in the laboratory included complete blood counts, liver and renal biochemistry, and viral serology, as well as echocardiograms and chest and abdominal radiography. Doppler ultrasound examination, CT scans, and hepatic angiography were performed to evaluate hepatic arterial, portal venous, and hepatic venous anatomies and also to measure graft volume. Autologous blood was taken in all but 16 emergency cases. The LDLT recipients ranged in age from 8 months to 67 years (median, 37 years). Their underlying diseases were congenital biliary atresia (68 patients), primary biliary cirrhosis (38), liver cirrhosis from viral hepatitis (50), fulminant hepatic failure (17), cryptogenic liver cirrhosis (9), primary sclerosing cholangitis (7), Wilson's disease (4), autoimmune hepatitis (5), citrullinemia (3), Alagille syndrome (2), familial amyloid polyneuropathy (1), and hyperoxaluria (1).
The type of donor hepatectomy performed was determined by CT-based volumetric measurements.17 The ratio of graft volume to be obtained to standard liver volume of recipient was calculated as described previously.18 Right hepatectomy was not indicated when the estimated graft volume exceeded 70% of the entire liver volume of the donor.8 Accordingly, the hepatectomies performed for the 71 pediatric recipients were as follows (Table 1): left lobectomy (segments 2 and 3) in 35 donors, extended left lobectomy (segments 2 and 3 and part of 4) in 21, left hepatectomy without middle hepatic vein in 5, left hepatectomy with middle hepatic vein in 6, right lateral sectorectomy (segments 6, and 7) in 2, and right hepatectomy without middle hepatic vein in 2. Hepatectomies performed for the 134 adult recipients were extended left lobectomy (1 donor), left hepatectomy with middle hepatic vein (61), right lateral sectorectomy (12), right hepatectomy without middle hepatic vein (49), and right hepatectomy with middle hepatic vein (11).
| Type of Donor Hepatectomy | None | Selective | Pringle's | Total (Pediatric) |
|---|---|---|---|---|
| Left lobectomy* | 2 | 31 | 2 | 35 (35) |
| Extended left lobectomy | 1 | 16 | 17 (16) | |
| Left hepatectomy without MHV | 4 | 1 | 5 (5) | |
| Left hepatectomy with MHV | 1 | 48 | 23 | 72 (11) |
| Right lateral sectorectomy† | 9 | 5 | 14 (2) | |
| Right hepatectomy without MHV | 12 | 39 | 51 (2) | |
| Right hepatectomy with MHV | 11 | 11 (0) | ||
| Total number | 4 | 120 | 81 | 205 (71) |
- Abbreviation: MHV, middle hepatic vein.
- * Segments 2 and 3 graft.
- † Segments 6 and 7 graft.
Surgical Procedure
Donor hepatectomy and operations on recipients were performed as previously described.3, 14, 19 Parenchymal transection was performed with the intermittent inflow occlusion technique, using a clamp crushing method or an ultrasonic dissector, with application of electrocautery and ligatures. Details of the inflow occlusion technique are described in “Inflow Occlusion During Parenchymal Transection” in this section. The graft was collected by clamping and dividing the portal vein, hepatic artery, and hepatic vein(s), in that order. It was immediately flushed with heparinized (10,000 U/L) cold lactate Ringer's solution (1000 mL), followed by cold University of Wisconsin solution (1 L; Viaspan, Dupont, Wilmington, DE) through a 16-French catheter inserted into the portal vein branch. No systemic anticoagulation treatment with heparin was used in the donors. The graft was transported to the recipient room after the native hepatectomy had been completed. Graft implantation was carried out by anastomosing the hepatic and portal veins and hepatic artery, and reconstructing the bile duct, in that order. The graft was perfused with cold lactated Ringer's solution (1000 mL) via a portal venous catheter when the half circle of the hepatic vein reconstruction (dorsal well) was finished. The graft was reperfused immediately after the hepatic or portal venous reconstruction. Hepatic artery reconstruction was carried out with the aid of an operating microscope. Bile duct reconstruction was performed by duct-to-Roux-en-Y loop anastomosis of the jejunum, or by bile duct-to-duct anastomosis.20
Inflow Occlusion During Parenchymal Transection
The left portal vein and the middle hepatic artery, if present, were clamped in the majority of cases of left lobe or extended left lobe graft (Table 2). This technique produced complete ischemia in segment 4; segments 2 and 3 were fed by arterial flow during transection. In whole left hemiliver grafts, transection was carried out with occlusion of the right portal vein and hepatic artery, a maneuver resulting in right hemiliver ischemia. In right lateral sector grafts, portal venous and hepatic arterial branches to the right paramedian sector (segments 5 and 8) were occluded. This resulted in total ischemia in the right paramedian sector, while normal blood flow in the right lateral sector was maintained. For right hemiliver grafts, right portal vein and arterial branches to the right paramedian sector were occluded. In this procedure, the right paramedian sector became totally ischemic, and the right lateral sector was perfused by the arterial flow.
| Type of Donor Hepatectomy | Type of Inflow Occlusion | Number of Cases | Graft Ischemia | Classification* |
|---|---|---|---|---|
| Left lobectomy | Left P + Middle HA | 19 | Portal | Portal ischemia |
| Left P | 11 | Portal | Semiperfused | |
| Left P + Left HA | 1 | Total | Total ischemia | |
| Extended left lobectomy | Left P + Middle HA | 14 | Partial + Portal | Partial ischemia |
| Left P | 2 | Portal | Portal ischemia | |
| Left hepatectomy without MHV | Left P + Middle HA | 3 | Partial + Portal | Partial ischemia |
| Left P | 1 | Portal | Portal ischemia | |
| Left hepatectomy with MHV | Right P + Right HA | 42 | None | No ischemia |
| Right P + Right HA, then Left P | 2 | Portal | Portal ischemia | |
| Right and Left P + Right HA | 2 | Portal | Portal ischemia | |
| Right P + Right HA + Middle HA | 1 | Partial | Partial ischemia | |
| Right P + Right HA, then Left P + Middle HA | 1 | Partial + Portal | Partial ischemia | |
| Right lateral sectorectomy | Left P + Middle HA | 5 | None | No ischemia |
| Right P + Right para. HA | 1 | Portal | Semiperfused | |
| Right para. P + Right HA | 1 | Partial + Arterial | Partial ischemia | |
| Right P + Right HA | 2 | Total | Total ischemia | |
| Right hepatectomy without MHV | Right P + Right para. HA | 5 | Partial + Portal | Partial ischemia |
| Right P + Right HA | 3 | Total | Total ischemia | |
| Right para. P + Right para. HA | 2 | Partial | Partial ischemia | |
| Right para. P + Right HA | 1 | Partial + Arterial | Partial ischemia | |
| Right P | 1 | Portal | Portal ischemia |
- Abbreviations: P, portal; HA, hepatic arterial; MHV, middle hepatic vein; para., paramedian sector.
- * Portal ischemia denotes no portal flow but intact hepatic arterial flow throughout the graft.
The theoretical rationale for applying a selective occlusion technique is as follows: in all cases, parenchymal transection is carried out within the ischemic part of the liver; however, in left hemiliver and right lateral sector grafts, the liver segments that form the graft are perfused normally, while in left lobe grafts, portal venous flow alone is interrupted, with hepatic arterial flow left intact. In instances of right hemiliver and extended left lobe grafts, only a part of the graft is subjected to ischemia; the other part is perfused with hepatic arterial flow (Table 2).
Minor modifications of the above inflow occlusion technique that result in various degrees of graft ischemia have also been used, and hepatectomy without any inflow occlusion was performed sporadically. Primarily on the basis of experience with right hemiliver grafts in which a part of the graft becomes ischemic during parenchymal transection, we have performed donor hepatectomy under intermittent Pringle's maneuver since November 2000, mostly in right or left hemiliver grafts in which the transection planes were set in the main portal fissure. The Pringle's maneuver leads to complete ischemia in both the transection plane and the part of the liver that forms the graft. In contrast, selective inflow occlusion allows some arterial blood flow to the ischemic part of the liver through the arterial plexus at the hepatic hilum. All inflow occlusion techniques were applied intermittently, with 10 minutes of occlusion first, followed by 15 minutes of occlusion alternating with 5 minutes of reperfusion. In cases of selective occlusion in which intact graft portal and arterial flow was maintained (whole left hemiliver grafts and right lateral sector grafts), the duration of occlusion was set at 30 minutes.
Data Collection
Patient demographic, surgical, and postoperative data were collected by chart review and from surgical records. Type of donor hepatectomy, type of inflow occlusion maneuver, duration of inflow occlusion, and amount of blood loss during donor surgery were recorded. Donors' blood aspartate aminotransferase (AST), alanine aminotransferase (ALT), and total bilirubin levels were examined on postoperative days 1 to 10. Donor bile leakage and reoperations were also recorded. Graft cold preservation time (defined as the time between graft harvest and commencement of graft implantation) and graft rewarming time (defined as the time between commencement of graft implantation and portal venous reperfusion) were also recorded. The recipients' blood AST and ALT levels were determined on postoperative days 1 to 5. Blood tests were carried out on the recipients at the termination of surgery, on arrival in the intensive care unit, and every 12 hours thereafter. Furthermore, the incidence of biliary complications (bile leaks and bile duct strictures), the incidence of acute rejection, and the 1-year graft and patient survival rates were recorded.
Analyses
Graft quality was assessed on the basis of maximum postoperative AST and ALT values within 3 postoperative days. Its relationship to the following parameters was examined: cold ischemic time (<120 minutes vs. ≥120 minutes), rewarming time (<60 minutes vs. ≥60 minutes), and type of graft ischemia during hepatectomy (no ischemia, portal ischemia, partial ischemia, and total ischemia). Ischemia type (Table 2) was classified as no ischemia (the liver segment that formed the graft was perfused normally during hepatectomy); portal ischemia (the inflow occlusion technique interrupted graft portal flow while maintaining hepatic arterial flow); partial ischemia (part of the graft became ischemic during parenchymal transection); and total ischemia (the graft became totally ischemic, as during Pringle's maneuver). As described, the kind of graft ischemia induced by each selective occlusion technique was largely dependent on the type of liver graft that was required, and no patient-specific criteria existed. The effect of graft ischemic time during donor hepatectomy on maximum postoperative AST and ALT values were investigated separately in the portal ischemia, partial ischemia, and total ischemia groups. This was done by comparing AST and ALT values by stratifying graft ischemic times into <60 minutes versus ≥60 minutes. Additionally, the incidences of acute rejections and biliary complications and the 1-year graft and patient survival rates were compared between the Pringle's group and the non-Pringle's group (all other patients who received a graft that had been harvested without using the Pringle's maneuver).
With regard to donor data, the relationship between the amount of blood loss and the type of liver inflow occlusion technique (selective inflow occlusion vs. Pringle's maneuver) was determined. Hepatectomy without any inflow occlusion (n = 4) was not included in this assessment. Data were analyzed by unpaired t tests or 1-way ANOVA, as appropriate, after logarithmic transformation of the values. Fisher exact test was also used where appropriate.
Results
The inflow occlusion techniques used were (Tables 1 and 2) selective inflow occlusion (120 donors), total inflow occlusion by Pringle's maneuver (81), and no inflow occlusion (4). Donor hepatectomy parameters were median liver parenchymal transection time, 69 minutes (range, 16-185 minutes); median inflow occlusion time, 63 minutes (range, 0-181 minutes); and median blood loss during donor hepatectomy, 473 mL (range, 35-2000 mL). Homologous donor blood transfusion, including fresh-frozen plasma, was not required during any operation. The median (range) values of the postoperative maximum AST, ALT, and total bilirubin levels were 202 U/L (89-856 U/L); 61 U/L (13-315 U/L), and 0.9 mg/dL (0.5-5.9 mg/dL), respectively. Bile leakage was observed in 13 patients, 5 of whom underwent reoperations to close the leakage. Two other patients underwent reoperations for the drainage of intraabdominal abscesses. No donor deaths occurred.
With regard to graft implantation parameters, median graft cold preservation time was 71 minutes (range, 16-231 minutes), and median graft rewarming time was 55 minutes (range, 31-105 minutes). No primary lack of function or dysfunction of the graft was observed in any inflow occlusion group. One 9-year-old male patient, who underwent LDLT from his mother for fulminant hepatic failure of unknown etiology, developed subacute fulminant hepatic failure for a second time and underwent another LDLT, this time from his father, with a successful outcome. A 7-year-old female patient, who underwent LDLT for hyperoxaluria, developed concomitant portal venous and hepatic arterial thromboses on postoperative day 1. The graft could not be rescued in spite of an emergency thrombectomy, and the patient died on postoperative day 4. These 2 transplants are excluded from the analysis of the effect of inflow occlusion on graft quality. In the remaining 203 transplants, maximum values of AST were observed within 24 hours after graft reperfusion; this was followed by observation of maximum values of ALT, with a delay of 12-24 hours. The AST and ALT values declined gradually thereafter in all cases during the observation period. The median values of maximum AST and ALT were 326 IU/mL (range, 88-1205 IU/mL) and 243 IU/mL (range, 52-1111 IU/mL), respectively. The median follow-up period of the total 205 recipients was 1154 days (range, 6-2967 days). Acute rejection was observed in 84 patients (41%). Fifty-seven patients (28%) had biliary complications: bile leaks (34 patients), biliary strictures (20), and both bile leaks and biliary strictures (3). There were 10 in-hospital mortalities. The 1-year graft and patient survival rates were 91% and 92%, respectively.
The relationship between maximum postoperative AST and ALT values and various parameters of graft ischemia are set out in Tables 3, 4, and 5. Type of graft ischemia during hepatectomy did not affect AST or ALT values, although the values of the total ischemia group tended to be lower than those of other groups (Table 3). The duration of graft ischemia during donor hepatectomy (stratified by <60 minutes vs. ≥60 minutes, did not affect postoperative AST or ALT value in the partial ischemia or total ischemia group; however, longer ischemic time was associated with the higher AST and ALT values in the portal ischemia group (Table 4). Neither graft rewarming time (stratified by <60 minutes vs. ≥60 minutes) nor cold preservation time (stratified by <120 minutes vs. ≥120 minutes) is associated with higher postoperative AST or ALT value (Table 5). The incidences of complications and the survival data for the Pringle's and non-Pringle's groups (in 203 recipients) were compared as follows (Pringle's vs. non-Pringle's groups): acute rejection, 6.3% vs. 8.9% (P = .042); bile leaks, 22.5% vs. 15.4% (P = .26); bile duct strictures, 13.8% vs. 9.8% (P = .49); 1-year graft survival rate, 94% vs. 91% (P = .60); and 1-year patient survival rates, 94% vs. 92% (P = .79). Donor blood loss in the Pringle's group (n = 81) was 509 ± 27 mL (mean ± SE); in the selective inflow occlusion group (n = 120), it was 549 ± 30 mL. The difference is not statistically significant (P = .34).
| No ischemia (n = 51) | Portal ischemia (n = 39) | Partial ischemia (n = 27) | Total ischemia (n = 86) | P Value* | |
|---|---|---|---|---|---|
| AST (U/L) | 384 ± 28 | 394 ± 33 | 376 ± 32 | 335 ± 21 | .159 |
| ALT (U/L) | 324 ± 27 | 324 ± 30 | 352 ± 42 | 279 ± 23 | .058 |
- NOTE. Values are expressed as mean ± SE.
- * Comparisons were made using ANOVA.
| Type of Ischemia | Duration | P Value |
|---|---|---|
| <60 min vs. ≥60 min | ||
| AST (U/L) | ||
| Portal ischemia | 324 ± 29 (n = 21) vs. 475 ± 58 (n = 18) | .027 |
| Partial ischemia | 429 ± 62 (n = 9) vs. 350 ± 37 (n = 18) | .264 |
| Total ischemia | 339 ± 27 (n = 55) vs. 328 ± 31 (n = 31) | .924 |
| ALT (U/L) | ||
| Portal ischemia | 250 ± 33 (n = 21) vs. 410 ± 45 (n = 18) | .003 |
| Partial ischemia | 404 ± 80 (n = 9) vs. 325 ± 50 (n = 18) | .336 |
| Total ischemia | 281 ± 29 (n = 55) vs. 277 ± 37 (n = 31) | .800 |
- NOTE. Values are expressed as mean ± SE.
| AST (U/L) | P | |
|---|---|---|
| Cold preservation time | <120 min (n = 155) vs. ≥120 min (n = 38) | .634 |
| 373 ± 16 vs. 356 ± 30 | ||
| Rewarming time | <60 min (n = 109) vs. ≥60 min (n = 84) | .430 |
| 360 ± 18 vs. 382 ± 22 | ||
| ALT (U/L) | ||
| Cold preservation time | <120 min (n = 155) vs. ≥120 min (n = 38) | .300 |
| 322 ± 17 vs. 283 ± 30 | ||
| Rewarming time | <60 min (n = 109) vs. ≥60 min (n = 84) | .663 |
| 308 ± 20 vs. 321 ± 23 |
- NOTE. Values are expressed as mean ± SE. Ten cases were excluded from analysis due to the lack of precise warm and cold ischemic time data.
Discussion
For the last 14 years, we have applied inflow occlusion techniques during donor hepatectomy in LDLT, progressing from intermittent selective inflow occlusion to Pringle's maneuver. Our results demonstrate that intermittent inflow occlusion techniques, including Pringle's maneuver, can be safely applied to LDLT donors without inducing associated graft injury. At present, almost all liver transplant centers perform hepatectomy of the LDLT donor without inflow vascular occlusion, for fear that it will result in graft injury.7 Our findings are relevant because donor safety is the first priority in LDLT, and inflow occlusion is a well-established procedure for safely decreasing blood loss during parenchymal dissection in hepatectomy.
The description of Pringle's maneuver dates back to 190815; however, when M.M. tried to apply inflow occlusion in liver surgery in the early 1980's, the technique had not been widely used, and, when used, had not been applied in a consistent manner. In addition, it was almost dogma that the cirrhotic liver would not tolerate the ischemic injury induced by inflow occlusion. We therefore developed the selective inflow occlusion technique21 in order to be able to apply inflow occlusion to cirrhotic livers. This technique does not lead to splanchnic congestion and increased ischemia/reperfusion injury. Furthermore, to avoid ischemic injury induced by prolonged inflow occlusion, we applied the technique intermittently—i.e., 15 minutes of ischemia alternating with 5 minutes of reperfusion—with ischemia gradually prolonged up to 30 minutes. When the author first reported this technique, many people expressed concern that repeated cycles of occlusion/perfusion might increase ischemia/reperfusion injury and thus be more harmful than continuous occlusion. However, the superiority of intermittent to continuous inflow occlusion, from the viewpoint of liver injury, was confirmed in a controlled study based on transaminase values and total bilirubin level.7 The recent observation that “preconditioning”, i.e., a short period of ischemia followed by reperfusion, protects against subsequent substantial ischemic insult to the liver, may provide additional theoretical support for the intermittent clamping method.22, 23 In other words, intermittent inflow occlusion may act as preconditioning.
We have used intermittent selective inflow occlusion during LDLT donor hepatectomy since we began LDLT in 1990; this is based on our clinical experience of hepatectomy. Our rationale for applying selective inflow occlusion was as follows: usually the part of the liver that forms the graft is perfused, or only the portal flow is occluded, and interruption of portal flow, together with intact hepatic arterial supply, does not lead to an acute reaction as reflected in any marked increase in liver transaminase values.24 During the later period of our experience with intermittent inflow occlusion, we applied this technique to a number of cases in which a part of the graft became completely ischemic during liver parenchymal dissection, e.g., the right paramedian sector in the case of a right hemiliver graft. We were able to confirm that this did not lead to elevated postoperative transaminase values. On the strength of these observations, we finally beganto apply Pringle's maneuver to donor hepatectomy.
In this study, maximum AST and ALT values observed within 12 to 48 hours after graft implantation showed that subclinical graft injury exists in LDLT due to graft harvesting, cold preservation, and warm ischemia during graft implantation. The results of the present study also showed that the duration of clod ischemia did not affect the graft quality in an LDLT setting in which the cold ischemic time is within 4 hours at most. Notably, the magnitude of this mild injury was not increased by the application of intermittent inflow occlusion techniques, including Pringle's maneuver. Furthermore, the incidences of biliary complications and the 1-year survival rates were not influenced by whether or not the graft was harvested using Pringle's maneuver, although the incidence of acute rejections was paradoxically lower in the Pringle's group. The absence of any negative effect of inflow occlusion on the graft in the present study can be interpreted in two ways: (1) Either any negative effect of inflow occlusion is minimal and is compensated by advantages such as shorter transection time in LDLT donors that are healthy and have no underlying liver disease such as steatosis,7 or (2) inflow occlusion during donor hepatectomy may act as preconditioning for the subsequent cold and normothermic ischemia induced by graft preservation and implantation. In support of the latter explanation, the protective effect of donor liver ischemic preconditioning has recently been confirmed in a rat model of liver transplantation25 and in human liver resection.22
The efficacy of inflow occlusion in minimizing blood loss during hepatectomy without inducing liver injury to the remaining part of the liver has long been established.7 In the present series of donor hepatectomies, blood loss during surgery of 533 ± 292 mL (mean ± SD; range, 35-2000 mL) was considered to be within the range reported by another group that did not use an inflow occlusion technique,26 and no homologous blood transfusions were required. Since hepatectomies without inflow occlusion were only performed in 4 donors, the possible benefits of inflow occlusion, in terms of blood loss, could not be examined in the present analysis. The superiority of Pringle's maneuver for selective inflow occlusion, as assessed by blood loss during surgery, was not definitively proven by the present data. Although quantifiable advantages in terms of bleeding were not evident, we would emphasize another benefit of inflow occlusion during LDLT donor hepatectomy: the surgical field during parenchymal dissection is relatively bloodless under inflow occlusion. As a result, the surgeon can recognize the tributaries of the hepatic veins before they are damaged, and, if they are damaged, the surgeon can better manage the bleeding. The importance of this should not be underestimated, since hemorrhage during parenchymal transection with inflow occlusion arises from the hepatic veins, and it is always venous hemorrhage that leads to bleeding-related catastrophes during liver surgery. Donor hepatectomy is currently being performed safely and with an acceptable amount of blood loss without the use of inflow occlusion at most LDLT centers; however, when the surgeon considers hemorrhage from the transection plane to be disturbing, he or she should not hesitate to apply inflow occlusion by Pringle's maneuver.
Recently, liver parenchymal dissection by newly developed devices using an electrosurgical generator, such as the Monopolar Floating Ball (Tissue Link Medical, Dover, NH), have been introduced. By its heat conductance and resultant precoagulation of the liver parenchyma, sufficient hemostasis can be expected and may replace inflow occlusion techniques, including Pringle's maneuver. However, these devices damage the liver parenchyma along with the dissection plane on both sides and lead to a decrease in functioning liver parenchyma in both the graft and the donor's remaining liver.
In conclusion, inflow occlusion techniques, such as Pringle's maneuver, can be applied to donor hepatectomy in LDLT without inducing any ischemic injury attributable to the technique itself. Even though the effect on reducing blood loss was not significant, application of the Pringle's maneuver enabled the liver resection to be performed more precisely and meticulously while inflicting the least amount of surgical trauma and avoiding anatomical complications. Surgeons should not hesitate to apply this technique.
